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/W4385953193	https://www.frontiersin.org/articles/10.3389/fnut.2023.1147114/pdf	English	null	Elevated first-trimester hepcidin level is associated with reduced risk of iron deficiency anemia in late pregnancy: a prospective cohort study	Frontiers in nutrition	2,023	cc-by	6,870	Elevated first-trimester hepcidin level is associated with reduced risk of iron deficiency anemia in late pregnancy: a prospective cohort study OPEN ACCESS EDITED BY Chris Coe, University of Wisconsin-Madison, United States REVIEWED BY Jiafu Li, Soochow University, China Peng An, China Agricultural University, China CORRESPONDENCE Yanna Zhu zhuyn3@mail.sysu.edu.cn †These authors have contributed equally to this work and share first authorship RECEIVED 18 January 2023 ACCEPTED 02 August 2023 PUBLISHED 15 August 2023 CITATION Sun P, Zhou Y, Xu S, Wang X, Li X, Li H, Lin Z, Huang F, Zhu L and Zhu Y (2023) Elevated first- trimester hepcidin level is associated with reduced risk of iron deficiency anemia in late Peng Sun 1†, Yueqin Zhou 2†, Suhua Xu 2†, Xiaotong Wang 2, Xiuxiu Li 1, Hailin Li 2, Zongyu Lin 2, Fenglian Huang 2, Lewei Zhu 2 and Yanna Zhu 2 1 Shenzhen Nanshan Maternity and Child Healthcare Hospital, Shenzhen, China, 2 Department of Maternal and Child Health, School of Public Health, Sun Yat-sen University, Guangzhou, China PUBLISHED 15 August 2023 CITATION Sun P, Zhou Y, Xu S, Wang X, Li X, Li H, Lin Z, Huang F, Zhu L and Zhu Y (2023) Elevated first- trimester hepcidin level is associated with reduced risk of iron deficiency anemia in late pregnancy: a prospective cohort study. Front. Nutr. 10:1147114. Background: Iron deficiency (ID) and iron deficiency anemia (IDA) during pregnancy are highly prevalent worldwide. Hepcidin is considered an important biomarker of iron status. Currently, few longitudinal cohort studies have assessed the potential causal relationship between hepcidin and ID/IDA. Therefore, we aimed to investigate the association of first-trimester maternal serum hepcidin with third-trimester ID/IDA risk in a prospective cohort. © 2023 Sun, Zhou, Xu, Wang, Li, Li, Lin, Huang, Zhu and Zhu. 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. Methods: Total of 353 non-ID/IDA pregnant women at 11–13 weeks’ gestation were enrolled in Southern China and followed up to 38 weeks of gestation. Data on demography and anthropometry were obtained from a structured questionnaire at enrollment. Iron biomarkers including hepcidin were measured at enrollment and follow-up. TYPE Brief Research Report PUBLISHED 15 August 2023 DOI 10.3389/fnut.2023.1147114 TYPE Brief Research Report PUBLISHED 15 August 2023 DOI 10.3389/fnut.2023.1147114 hepcidin, iron deficiency, anemia, pregnancy, hemoglobin KEYWORDS 2.2. Laboratory tests Although there are some well-established methods to detect iron status, accurate identification of maternal ID and IDA with conventional markers still remains a challenge (8). Serum ferritin (SF) and hemoglobin (Hb) serve as common markers of ID and IDA. Hb is a sensitive indicator of ID, because IDA, as a type of Hb disorder, directly results from ID. However, Hb in the diagnosis of ID/IDA may be interfered with other Hb disorders such as thalassemia. Moreover, SF can be confounded by inflammation and infection (9). Therefore, a biomarker that enables accurate identification of the risk of iron status is needed. Each pregnant woman who met the eligible criteria had her venous blood drawn by a trained nurse in the morning after overnight fasting in the first trimester (12.14 ± 0.04 gestational weeks) and the third trimester (38.62 ± 0.13 gestational weeks), in accordance with the standard protocol. Serum iron biomarkers including serum hepcidin, SF, Hb and serum iron (SI), and inflammatory biomarkers like CRP were tested within 2 h by trained technicians in the hospital laboratory. Serum hepcidin levels were measured using a commercially available quantikine ELISA kit (DHP250; R&D system; USA) according to the manufacturer’s instructions, which is a validated and highly sensitive enzyme immunoassay for quantitative in vitro diagnostic determination of hepcidin in human cell culture supernate, serum, plasma or urine. SF concentrations were determined by the enzyme immunoassay method, using commercial kits (FERRITIN ELISA; Diametra, Boldon, UK). Hb levels were quantified on an automated hematological analyser (TC Hemaxa 1,000; Teco Diagnostic, Anaheim, CA, USA), using a hemiglobin-cyanide method. SI levels were assessed by a commercial test, using a colorimetric method (Ferentest, bioMérieux® SA, France). Serum concentrations of CRP were determined using an immunoturbidimetric assay. For these parameters, the intra-assay and inter-assay CV (%) were below 5.7 and 6.7%, respectively. The remaining blood samples were stored at˗80 C until assayed. Hepcidin, a cysteine-rich antimicrobial peptide hormone secreted by the liver, has been shown in previous studies that it exerts a crucial role in iron metabolism (10). Although both hepcidin and SF are correlated with inflammation, such as C-reactive protein (CRP), prior studies have shown that hepcidin is closely related to iron status during pregnancy (11–13). In detail, evidence from the studies have shown that the level of hepcidin in pregnant women with IDA was lower than that in controls (14, 15). Elevated first-trimester hepcidin level is associated with reduced risk of iron deficiency anemia in late pregnancy: a prospective cohort study Regression models were used to evaluate the association of first- trimester hepcidin with third-trimester ID/IDA risk. Results: Serum hepcidin levels substantially decreased from 19.39 ng/mL in the first trimester to 1.32 ng/mL in the third trimester. Incidences of third-trimester ID and IDA were 46.2 and 11.4%, respectively. Moreover, moderate and high levels of first-trimester hepcidin were positively related to third-trimester hepcidin (log-transformed β = 0.51; 95% CI = 0.01, 1.00 and log-transformed β = 0.66; 95% CI = 0.15, 1.17). Importantly, elevated first-trimester hepcidin was significantly associated with reduced risk of third-trimester IDA (OR = 0.38; 95% CI = 0.15, 0.99), but not with ID after adjustment with potential confounders. Conclusion: First-trimester hepcidin was negatively associated with IDA risk in late pregnancy, indicating higher first-trimester hepcidin level may predict reduced risk for developing IDA. Nonetheless, given the limited sample size, larger studies are still needed. hepcidin, iron deficiency, anemia, pregnancy, hemoglobin 01 01 Frontiers in Nutrition frontiersin.org 10.3389/fnut.2023.1147114 10.3389/fnut.2023.1147114 Sun et al. 2.3. Exposure assessment The exposure was serum hepcidin level in the first trimester, which was divided into tertiles based on its distribution among all participants. First-trimester serum hepcidin was classified into three categories: (1) low serum hepcidin, ≤11.85 ng/mL (reference); (2) moderate serum hepcidin, 11.86–27.43 ng/mL; and (3) high serum hepcidin, ≥27.44 ng/mL. Our objectives were to investigate the associations of serum hepcidin levels in the first trimester with iron status and ID/IDA risk in the third trimester through a prospective cohort study, thus providing scientific evidence for identification of ID/IDA risk at an early stage. 1. Introduction non-ID/IDA was defined as the absence of SF < 20 ng/mL or Hb < 110 g/L). Any subjects with (1) a history of iron-supplementation intake during the past 3 months; (2) a history of hypertension and diabetes; or (3) unable to complete questionnaire or refused to sign the informed consent, were excluded. Written informed consents were obtained from all participants at the time of recruitment. The participants were invited to complete a structured questionnaire, undergo physical examinations, and provide blood samples after enrollment and were recalled for blood samples collection at 38 weeks of gestation. This study was approved by the Ethics Committee of School of Public Health Sun Yat-sen University, and conducted corresponding to the Declaration of Helsinki. Iron deficiency (ID), and specifically iron deficiency anemia (IDA), has been well-recognized as one of the most common nutritional deficiencies during pregnancy globally (1). The prevalence of gestational anemia was reported to be 26% in developed regions such as the United States and Europe and reached 46%–48% in Southeast Asia and Africa (2). According to the WHO, 50% of anemia is attributed to ID (3). In China, the prevalence of maternal ID and IDA was reported to be 42.6% and 19.1%, respectively (4). Numerous studies have shown that antenatal ID and IDA are associated with increased risk of maternal mortality, preterm birth, small for gestational age, low birth weight, and long-term influence on cognitive function in the offspring (3, 5–7). Thus, it is essential to identify the risks of ID and IDA early. 2.2. Laboratory tests Some cohort studies also revealed the significant association between hepcidin and iron status, but mostly within a single time period (16, 17). Furthermore, only one sub-analysis of a pooled dataset among pregnant women before 28 weeks’ gestation found that baseline serum hepcidin >1.6 μg/L was associated with a reduced risk of IDA at delivery (11). Additionally, hepcidin is suppressed in the second and third trimester, especially the level of hepcidin was extremely low and hard to check out in the third trimester (18). Therefore, it is necessary to explore the hepcidin level at early pregnant stage with iron status in late pregnancy. Frontiers in Nutrition frontiersin.org 2.5. Assessment of covariates was 29.12 ± 4.45 years and the mean gestational age was 12.15 ± 0.64 weeks at enrollment. The median serum hepcidin concentration was 19.39 (9.69–33.59) ng/ml in the first trimester and 1.32 (0.52–6.67) ng/ml in the third trimester. In the first trimester, no significant differences were observed in demographic characteristics among pregnant women (p > 0.05), and significant differences were observed only in SF levels (p < 0.05). In the third trimester, significantly higher serum hepcidin, Hb and SI levels were observed among pregnant women with higher tertiles of first-trimester hepcidin compared to those with the lowest tertile (p < 0.05). In addition, 122 (46.2%) incident ID cases and 30 (11.4%) incident IDA cases were identified in the third trimester, respectively. Data on demographic information and pregnancy history were collected from a structured questionnaire at enrollment, including maternal age (years), educational level, monthly income, participant source, gestational age, gravidity and parity. Anthropometric data at enrollment were obtained by experienced clinicians and nurses. Barefoot height was measured to the nearest 0.1 cm using a stadiometer (Yilian TZG, Jiangsu, PRC), and body weight was measured to the nearest 0.1 kg with a self-zeroing scale (Hengxing TGT-140, Jiangsu, PRC). Pre-pregnancy weight was measured and recorded in health booklets by professional staffs during the pre-pregnancy checkups. Pre-pregnancy BMI (pre-BMI) was calculated by dividing one’s pre-pregnancy weight in kilograms by her height in meters and categorized as underweight (<18.5 kg/m2), normal (18.5–23.9 kg/m2), or overweight (≥24 kg/m2), according to Chinese criteria (20). Intake of iron supplementation during pregnancy was obtained from medical records. 2.6. Statistical analysis Normality was assessed using the Shapiro–Wilk test and Q-Q plot. Data were presented as mean ± standard deviation or median (inter- quartile range) for continuous variables and number (percentage) for categorical variables. The difference among three groups were compared by One-way ANOVA (continuous variables with normal distribution), or Kruskal-Wallis H tests (continuous variables with skewed distribution), and Chi-square tests (categorical variables). We used linear regression model to evaluate the associations of first- trimester serum hepcidin with third-trimester iron biomarkers (serum hepcidin, SF, Hb and SI) and inflammatory biomarker (CRP). Right-skewed biomarkers including serum hepcidin, SF, SI and CRP in the third trimester, as dependent variables, were log(e)-transformed to normalize distributions prior to linear regression analysis. Resulting regression coefficients (β) expressed the change in log-transformed biomarker levels that are associated with moderate or high tertiles of first-trimester hepcidin, compared to low tertile of first-trimester hepcidin. In addition, logistic regression models were built to examine the associations between first-trimester serum hepcidin and third- trimester ID/IDA risk. Maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy, and CRP or serum hepcidin in the third trimester were included as covariates in the regression models. All statistical analyses were performed using R 4.0; p < 0.05 was considered significant. 3. Results Table 3 shows the relationships between tertiles of first-trimester serum hepcidin and ID/IDA risk in the third trimester. In the unadjusted analyses, when compared to low first-trimester hepcidin, moderate and high first-trimester hepcidin was marginally associated with reduced risk of IDA in the third trimester (crude odds ratio [OR] = 0.40; 95% CI: 0.16, 1.03 and crude OR = 0.39; 95% CI: 0.15, 1.01, respectively). Furthermore, after additional adjustment with maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and CRP, first-trimester hepcidin was independently inversely associated with IDA risk in the third trimester (moderate vs. 2.1. Study design and participants The primary outcomes were the incidences of ID and IDA in the third trimester. ID and IDA were classified according to SF and Hb. ID was defined as SF < 20 ng/mL and IDA was defined as ID plus low Hb (Hb < 110 g/L), in accordance with the definition proposed by Chinese Medical Association, 2014 (19). The secondary outcomes were iron status in the third trimester, including serum hepcidin, SF, Hb and SI. This prospective cohort study was conducted among pregnant women at Shenzhen Nanshan Maternity and Child Healthcare Hospital in Shenzhen city, South China, from May 2019 to April 2020. Briefly, non-ID/IDA women with singleton pregnancy, aged 18 to 45 years, were recruited between 11 and 13 weeks of gestation (here 02 frontiersin.org Sun et al. 10.3389/fnut.2023.1147114 10.3389/fnut.2023.1147114 10.3389/fnut.2023.1147114 3.2. The relationships between tertiles of first-trimester serum hepcidin and third-trimester iron status We evaluated the relationships between tertiles of first-trimester serum hepcidin and third-trimester iron status (Table 2). Significantly higher serum hepcidin levels in the third trimester were more likely to be observed in pregnant women with moderate first-trimester hepcidin (log-transformed β = 0.51; 95% confidence interval [CI] = 0.01, 1.00), as well as those with high first-trimester hepcidin (log-transformed β = 0.66; 95% CI = 0.15, 1.17), compared to those with low first-trimester hepcidin, after adjusting for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and CRP. These β values, when calculated back to original scales of third-trimester serum hepcidin, mean that women with moderate and high first-trimester hepcidin, respectively, had 66.5% (i.e., [e0.51–1] × 100%) and 93.5% (i.e., [e0.66–1] × 100%) higher levels of serum hepcidin in the third trimester, compared to those with low first- trimester hepcidin. Moreover, in comparison to low first-trimester hepcidin, moderate first-trimester hepcidin was positively associated with SI (log-transformed β = 0.20; 95% CI = 0.06, 0.34) and Hb (β = 5.43; 95% CI = 1.17, 9.70) in the third trimester, after adjustment with the mentioned covariates. Similarly, these β values mean that women with moderate first-trimester hepcidin had 22.1% (i.e., [e0.20–1] × 100%) and 5.43 ng/mL higher levels of SI and Hb in the third trimester than those with low first-trimester hepcidin, respectively. However, no significant associations were observed between first-trimester hepcidin and SF or CRP in the third trimester. Frontiers in Nutrition frontiersin.org 3.1. Characteristics of the participants stratified by tertiles of first-trimester serum hepcidin The flow chart of the study participants is shown in Supplementary Figure S1. Of the total 353 pregnant women recruited, 264 participants with full data were included in the final analysis. The characteristics of the participants stratified by tertiles of first-trimester serum hepcidin are presented in Table 1. The mean age of participants 03 frontiersin.org Sun et al. 10.3389/fnut.2023.1147114 TABLE 1 Characteristics of individuals stratified by tertiles of first-trimester serum hepcidin levels. TABLE 1 Characteristics of individuals stratified by tertiles of first-trimester serum hepcidin levels. 3.1. Characteristics of the participants stratified by tertiles of first-trimester serum hepcidin Characteristics Total (N = 264) Low hepcidin Moderate hepcidin High hepcidin P ≤ 11.85 (N = 89) 11.86–27.43 (N = 87) ≥ 27.44 (N = 88) Age, years 29.12 ± 4.45 29.02 ± 4.61 29.07 ± 4.35 29.27 ± 4.43 0.924 Educational level College or lower 74 (41.3) 23 (39.7) 26 (44.1) 25 (40.3) 0.871 University or higher 105 (58.7) 35 (60.3) 33 (55.9) 37 (59.7) Monthly income ≤5,000 RMB 89 (50.0) 27 (46.6) 32 (55.2) 30 (48.4) 0.618 >5,000 RMB 89 (50.0) 31 (53.4) 26 (44.8) 32 (51.6) Participant source Rural 202 (76.5) 65 (73.0) 67 (77.0) 70 (79.5) 0.588 Urban 62 (23.5) 24 (27.0) 20 (23.0) 18 (20.5) Gestational age at enrolment, weeks 12.15 ± 0.64 12.14 ± 0.61 12.20 ± 0.66 12.11 ± 0.65 0.684 Gravidity 1 86 (32.6) 23 (25.8) 26 (29.9) 37 (42.0) 0.057 ≥2 178 (67.4) 66 (74.2) 61 (70.1) 51 (58.0) Parity Primiparous 112 (42.4) 33 (37.1) 33 (37.9) 46 (52.3) 0.072 Multiparous 152 (57.6) 56 (62.9) 54 (62.1) 42 (47.7) Height, cm 158.48 ± 5.18 158.84 ± 4.28 158.15 ± 6.48 158.42 ± 4.59 0.674 Weight at enrolment, kg 53.23 ± 8.23 53.62 ± 9.02 53.73 ± 7.34 52.48 ± 8.42 0.710 Pre-BMI, kg/m2 21.31 ± 2.76 21.28 ± 2.81 21.47 ± 2.92 21.18 ± 2.54 0.781 Pre-BMI category Normal 189 (71.6) 63 (70.8) 63 (72.4) 63 (71.6) 0.870 Underweight 35 (13.3) 13 (14.6) 9 (10.3) 13 (14.8) Overweight 40 (15.2) 13 (14.6) 15 (17.2) 12 (13.6) Iron supplementation during pregnancy Received 59 (22.3) 25 (28.1) 19 (21.8) 15 (17.0) 0.209 Not received 205 (77.7) 64 (71.9) 68 (78.2) 73 (83.0) First trimester SH, ng/mL 19.39 (9.69–33.59) 8.03 (5.01–9.76) 19.43 (16.14–23.84)a 40.74 (33.76–58.53)a <0.001* SF, ng/mL 85.00 (54.85–124.00) 52.00 (41.00–80.75) 86.00 (61.50–116.00)a 123.00 (93.75–175.25)a <0.001* Hb, g/L 126.21 ± 8.08 125.31 ± 7.26 126.61 ± 8.31 126.72 ± 8.63 0.473 SI, umol/L 21.07 (17.32–25.06) 20.30 (16.76–24.09) 22.43 (17.88–25.73) 20.41 (17.28–25.55) 0.128 Third trimester SH, ng/mL 1.32 (0.52–6.67) 0.85 (0.37–3.25) 1.47 (0.67–7.94) 2.31 (0.76–7.48)a 0.007* SF, ng/mL 20.00 (14.15–31.00) 20.00 (13.00–27.10) 20.00 (13.25–29.50) 20.85 (16.00–36.50) 0.201 Hb, g/L 118.17 ± 14.40 115.28 ± 14.44 120.79 ± 13.04a 118.50 ± 15.25 0.038* SI, umol/L 12.90 (9.80–17.56) 12.73 (8.60–16.48) 13.82 (11.32–18.60)a 12.90 (9.38–16.97) 0.041* CRP, mg/L 6.85 (3.77–15.47) 6.85 (3.67–11.83) 6.85 (3.50–14.30) 6.88 (4.73–26.91) 0.211 ID 122 (46.2) 43 (48.3) 41 (47.1) 38 (43.2) 0.774 IDA 30 (11.4) 16 (18.0) 7 (8.0) 7 (8.0) 0.054 RMB, Renminbi; pre-BMI: pre-pregnancy body mass index; SH, serum hepcidin; SF, serum ferritin; Hb, hemoglobin; SI, serum iron; CRP, C-reactive protein; ID, iron deficiency; IDA, iron deficiency anemia. resented as N (percentage) for categorical variables and mean ± standard deviation or median (inter-quartile range) for normal or skewed continuous variables. h d b h f l bl d d k l ll f bl h RMB, Renminbi; pre-BMI: pre-pregnancy body mass index; SH, serum hepcidin; SF, serum ferritin; Hb, hemoglobin; SI, serum iron; CRP, C-reactive protein; ID, iro eficiency anemia. i Data were presented as N (percentage) for categorical variables and mean ± standard deviation or median (inter-quartile range) for normal or skewed continuous var p < 0.05 among three groups, assessed by Chi-square test for categorical variables and one-way ANOVA and Kruskal-Wallis H test for continuous variables when ap with Bonferroni. pre-BMI: pre-pregnancy body mass index; SH, serum hepcidin; SF, serum ferritin; Hb, hemoglobin; SI, serum iron; CRP, C-reactive protein; ID, iron deficiency; I a dei c e cy a e a. Data were presented as N (percentage) for categorical variables and mean ± standard deviation or median (inter-quartile range) for normal or skewed continuous variables. p < 0.05 among three groups, assessed by Chi-square test for categorical variables and one-way ANOVA and Kruskal-Wallis H test for continuous variables when appropriate. Post-hoc tests with Bonferroni. ap<0 05 compared to low hepcidin group with Bonferroni. ap < 0.05, compared to low hepcidin group. p < 0.05 among three groups, assessed by Chi-square test for categorical variables and one-way ANOVA and Kruskal-Wallis H test for continuous variables when appropriate. Post-hoc tests with Bonferroni. ap < 0.05, compared to low hepcidin group. s, assessed by Chi-square test for categorical variables and one-way ANOVA and Kruskal-Wallis H test for continuous variables when appropriate. Post-hoc tests roni. mpared to low hepcidin group. p < 0.05, compared to low hepcidin group. deficiency anemia. Data were presented as N (percentage) for categorical variables and mean±standard deviation or median (inter quartile range) for normal or skewed continuous variables frontiersin.org roups, assessed by Chi-square test for categorical variables and one-way ANOVA and Kruskal-Wallis H test for continuous variables when appropriate. Post-hoc te ow hepcidin group. 3.1. Characteristics of the participants stratified by tertiles of first-trimester serum hepcidin Data were presented as N (percentage) for categorical variables and mean ± standard deviation or median (inter-quartile range) for normal or skewed continuous variables. h d b h f l bl d d k l ll f bl h h 04 frontiersin.org frontiersin.org Sun et al. 10.3389/fnut.2023.1147114 TABLE 2 The associations between first-trimester serum hepcidin levels and third-trimester iron status. TABLE 2 The associations between first-trimester serum hepcidin levels and third-trimester iron status. Outcomes Low hepcidin Moderate hepcidin High hepcidin ≤11.85 (N = 89) 11.86–27.43 (N = 87) P ≥27.44 (N = 88) P SH† Crude Reference 0.54 (0.03, 1.04) 0.038* 0.73 (0.22, 1.23) 0.005* Adjusted‡ Reference 0.51 (0.01, 1.00) 0.046* 0.66 (0.15, 1.17) 0.011* SF† Crude Reference 0.02 (−0.23, 0.27) 0.857 0.19 (−0.06, 0.44) 0.145 Adjusted‡ Reference 0.01 (−0.24, 0.26) 0.958 0.17 (−0.08, 0.42) 0.183 Hb Crude Reference 5.51 (1.27, 9.75) 0.011* 3.22 (−1.01, 7.45) 0.135 Adjusted‡ Reference 5.43 (1.17, 9.70) 0.013* 2.98 (−1.33, 7.30) 0.175 SI† Crude Reference 0.20 (0.06, 0.34) 0.006* 0.04 (−0.10, 0.19) 0.532 Adjusted‡ Reference 0.20 (0.06, 0.34) 0.005* 0.03 (−0.11, 0.17) 0.637 CRP† Crude Reference 0.15 (−0.28, 0.57) 0.489 0.34 (−0.08, 0.77) 0.110 Adjusted§ Reference 0.12 (−0.30, 0.54) 0.579 0.41 (−0.02, 0.84) 0.063 SH, serum hepcidin; SF, serum ferritin; Hb, hemoglobin; SI, serum iron; CRP, C-reactive protein. Data were presented as β (95% confidence interval). p < 0.05, estimated by liner regression model. †The dependent varibles were log(e)-transformed prior to linear regression analysis. Thus, the β values expressed the change in log-transformed biomarker levels that are associated with moderate or high tertiles of first-trimester hepcidin, compared to low tertile of first-trimester hepcidin. ‡Adjusted for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and CRP in the third trimester. §Adjusted for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and hepcidin in the third trimester. †The dependent varibles were log(e)-transformed prior to linear regression analysis. Thus, the β values expressed the change in log-transformed biomarker levels that are associated with moderate or high tertiles of first-trimester hepcidin, compared to low tertile of first-trimester hepcidin. ‡Adjusted for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and CRP in the third trimester. §Adjusted for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and hepcidin in the third trimester. TABLE 3 The associations between first-trimester serum hepcidin levels and third-trimester iron deficiency and iron deficiency anemia. 3.1. Characteristics of the participants stratified by tertiles of first-trimester serum hepcidin Outcomes Low hepcidin Moderate hepcidin High hepcidin ≤11.85 (N = 89) 11.86–27.43 (N = 87) P ≥27.44 (N = 88) P ID, n (%) 43 (48.3) 41 (47.1) 38 (43.2) Crude Reference 0.95 (0.53, 1.72) 0.875 0.81 (0.45, 1.47) 0.493 Adjusted† Reference 0.98 (0.54, 1.80) 0.950 0.82 (0.44, 1.51) 0.521 IDA, n (%) 16 (18.0) 7 (8.0) 7 (8.0) Crude Reference 0.40 (0.16, 1.03) 0.056 0.39 (0.15, 1.01) 0.053 Adjusted† Reference 0.38 (0.15, 0.99) 0.047 0.38 (0.14, 0.99) 0.049* ID, iron deficiency; IDA, iron deficiency anemia. Data were presented as odds ratio (95% confidence interval). *P < 0.05, estimated by logistic regression model. †Adjusted for maternal age, pre-pregnancy BMI, parity, iron supplementation during pregnancy and CRP in the third trimester. E 3 The associations between first-trimester serum hepcidin levels and third-trimester iron deficiency and iron deficiency anemia Our longitudinal cohort study showed that elevated first-trimester serum hepcidin was associated with increased serum hepcidin and higher iron status, as well as significantly lower risk of IDA in the third trimester, independent of potential confounders including CRP. low hepcidin: adjusted OR = 0.38; 95% CI: 0.15, 0.99 and high vs. low hepcidin: adjusted OR = 0.38; 95% CI: 0.14, 0.99). However, when extreme tertiles were compared, first-trimester hepcidin was not substantially related to ID risk in the third trimester regardless of whether the selected confounders were controlled for. ID and IDA are prevalent nutritional deficiency disorders during pregnancy worldwide. Therefore, early identification of risk factors for ID and IDA is essential to prevent the occurrence of ID and IDA. It has been shown that hepcidin is a key regulator of iron metabolism. As known in previous studies, hepcidin level would continue to fall throughout pregnancy to allow more iron to be released into plasma to meet maternal iron requirements and the needs of fetal growth and Frontiers in Nutrition frontiersin.org 4. Discussion The prospective associations between maternal serum hepcidin in early pregnancy and subsequent risks of ID and IDA are still unknown. 05 Sun et al. 10.3389/fnut.2023.1147114 10.3389/fnut.2023.1147114 ERFE is a glycoprotein hormone secreted by erythroblasts in response to erythropoietin (EPO) stimuli such as hemorrhage, hypoxia, EPO therapy, β-thalassemia, and anemia of inflammation, and it suppresses the hepatic production of hepcidin, thereby mobilizing iron for erythropoiesis (27). Accordingly, it can be speculated in our study that lower levels of first-trimester serum hepcidin may imply a state of EPO-stimulated ERFE production, as well as a possible condition of high anemia risk, although all the participating women were below the diagnosis threshold of anemia at enrollment. Taken together, further studies is warranted to collect more information on iron storage indicators and erythroid regulatory factors, thereby elucidating the mechanisms underlying the association between higher levels of first-trimester serum hepcidin and lower risk of third-trimester IDA. development (21). In the present study, we did observe a substantial decline of hepcidin from the first to the third trimester. Our finding is in accordance with previous findings that hepcidin undergoes apparent changes during pregnancy (18, 22). However, a prospective cohort study included 103 healthy Turkish women with normal pregnancies found no significant differences in hepcidin between early pregnancy and late pregnancy, probably because the effects of obstetrical complications such as anemia or ID on maternal hepcidin were not considered (23).h The association between hepcidin and iron status during pregnancy has been reported in previous researches (14, 17, 24). However, the relationships between hepcidin in early pregnancy and iron biomarkers in late pregnancy have rarely been analyzed. Of note, positive associations between first-trimester serum hepcidin and third-trimester serum hepcidin, Hb and SI were observed in our study. A plausible explanation is that although hepcidin decreases with the progress of pregnancy, hepcidin that is higher in the first trimester remains at a higher level in the third trimester, indicating a replete iron status, which partly counteracts suppression of hepcidin by pregnancy signals (9). i The relationship between hepcidin and ID was also investigated in the present study. No significant association was observed between first-trimester hepcidin and third-trimester ID risk, which is consistent with a Tanzanian study that found no relationship between baseline hepcidin (< 28 weeks) and ID at delivery (11). However, there are some studies with inconsistent findings (9, 11, 16). 4. Discussion For example, an analysis among pregnant women based on an incorporative dataset of clinical trials and a prospective cohort found that hepcidin in individuals with ID was significantly lower than iron-replete individuals (11). Additionally, in another cohort study in Gambia, the prevalence of maternal ID increased, while hepcidin gradually decreased with progressive gestation (16). Furthermore, a follow-up study among iron-deficient Tanzania pregnant women taking iron supplements showed that the prevalence of ID dropped from 93 to 12%, while hepcidin increased form 1.0 μg/L at baseline to 12.3 μg/L at delivery (28). The discrepancy may be partially due to the fact that criteria to determine ID differ and that serum hepcidin levels are often undetectable or low in ID. Therefore, those inconsistent results between hepcidin and ID still call for further investigation. Numerous prior studies measured hepcidin during pregnancy when ID/IDA was already diagnosed (14, 15), whereas we investigated hepcidin of non-ID/IDA pregnant women in the first trimester and ID/IDA risk in the third trimester. We found that elevated first- trimester serum hepcidin was associated with diminished risk of third-trimester IDA, similar to some studies reporting a negative relationship between hepcidin and IDA (14, 15, 25). Previous studies have demonstrated that hepcidin inhibits iron efflux into plasma by degrading its only receptor, ferroportin (FPN) in hepatocytes, intestinal enterocytes, and macrophages, thereby leading to iron sequestration in cells (26). During iron replete pregnancy, maternal iron and hepcidin metabolism keep homeostatic, the mothers maintain constant SI levels and relatively high but not overexpressed serum hepcidin levels, despite increased iron utilization in advanced pregnancy. However, in iron-overloaded mothers, hepcidin production is overstimulated, resulting in hypoferremia and limiting the iron availability for both the mother and the fetus (9). SF is a stable and valid indicator reflecting iron stores, and it can be found in our study that first-trimester SF levels were higher among those women with higher first-trimester serum hepcidin levels but did not reach the threshold of iron overload. Therefore, a possible reason for our findings is that elevated hepcidin levels in early pregnancy implies replete iron stores, indicating that sufficient iron is available for Hb synthesize as pregnancy progresses, thereby decreasing IDA risks in the third trimester. Frontiers in Nutrition frontiersin.org Conflict of interest The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. 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. Funding during consecutive trimesters of pregnancy while taking into account the effects of other inflammatory factors. during consecutive trimesters of pregnancy while taking into account the effects of other inflammatory factors. The study was funded by the Guangdong Provincial Natural Science Foundation (Grant No. 2021A1515010439), medical science and technology of Guangdong Province (No. A2019332), and the Sanming Project of Medicine in Shenzhen (Grant No. SZSM201803061). The funders had no role in study design, data collection and analysis, decision to publish, or manuscript preparation. 1. Camaschella C. Iron-deficiency Anemia. N Engl J Med. (2015) 372:1832–43. doi: 10.1056/NEJMra1401038 1. Camaschella C. Iron-deficiency Anemia. N Engl J Med. (2015) 372:1832–43. doi: 10.1056/NEJMra1401038 partum: a multilevel analysis. Lancet Glob Health. (2018) 6:e548–54. doi: 10.1016/ S2214-109X(18)30078-0 8. Cantor AG, Bougatsos C, Dana T, Blazina I, McDonagh M. Routine iron supplementation and screening for iron deficiency anemia in pregnancy: a systematic review for the U.S. preventive services task force. Ann Intern Med. (2015) 162:566–76. doi: 10.7326/M14-2932 8. Cantor AG, Bougatsos C, Dana T, Blazina I, McDonagh M. Routine iron supplementation and screening for iron deficiency anemia in pregnancy: a systematic review for the U.S. preventive services task force. Ann Intern Med. (2015) 162:566–76. doi: 10.7326/M14-2932 2. World Health Organization. Data: Prevalence of anaemia in pregnant women. Available at: https://www.who.int/data/gho/data/indicators/indicator-details/GHO/ prevalence of anaemia in pregnant women (Accessed October 24 2022) 3. Means RT. Iron deficiency and Iron deficiency anemia: implications and impact in pregnancy, fetal development, and early childhood parameters. Nutrients. (2020) 12:447. doi: 10.3390/nu12020447 3. Means RT. Iron deficiency and Iron deficiency anemia: implications and impact in pregnancy, fetal development, and early childhood parameters. Nutrients. (2020) 12:447. doi: 10.3390/nu12020447 9. Sangkhae V, Fisher AL, Wong S, Koenig MD, Tussing-Humphreys L, Chu A, et al. Effects of maternal iron status on placental and fetal iron homeostasis. J Clin Invest. (2020) 130:625–40. doi: 10.1172/JCI127341 9. Sangkhae V, Fisher AL, Wong S, Koenig MD, Tussing-Humphreys L, Chu A, et al. Effects of maternal iron status on placental and fetal iron homeostasis. J Clin Invest. (2020) 130:625–40. doi: 10.1172/JCI127341 4. Liao QK. Prevalence of iron deficiency in pregnant and premenopausal women in China: a nationwide epidemiological survey. Zhonghua Xue Ye Xue Za Zhi. (2004) 25:653–7. 10. Roth M-P, Meynard D, Coppin H. Regulators of hepcidin expression. Vitam Horm. (2019) 110:101–29. doi: 10.1016/bs.vh.2019.01.005 11. Abioye AI, Aboud S, Premji Z, Etheredge AJ, Gunaratna NS, Sudfeld CR, et al. Hemoglobin and hepcidin have good validity and utility for diagnosing iron deficiency anemia among pregnant women. Eur J Clin Nutr. (2020) 74:708–19. doi: 10.1038/ s41430-019-0512-z 11. Abioye AI, Aboud S, Premji Z, Etheredge AJ, Gunaratna NS, Sudfeld CR, et al. Hemoglobin and hepcidin have good validity and utility for diagnosing iron deficiency anemia among pregnant women. Eur J Clin Nutr. (2020) 74:708–19. doi: 10.1038/ s41430-019-0512-z 5. Janbek J, Sarki M, Specht IO, Heitmann BL. A systematic literature review of the relation between iron status/anemia in pregnancy and offspring neurodevelopment. Eur J Clin Nutr. (2019) 73:1561–78. doi: 10.1038/s41430-019-0400-6 6. Supplementary material The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnut.2023.1147114/ full#supplementary-material 5. Conclusion Our study demonstrated that elevated first-trimester serum hepcidin level was closely associated with decreased risk of IDA in the third trimester, which indicates that high hepcidin level in early pregnancy implies replete iron stores and therefore lower risk of IDA in late pregnancy. However, considering the relatively limited sample size, larger studies collecting more data on iron metabolism indicators and inflammatory factors during consecutive trimesters are still needed. Publisher’s note The studies involving human participants were reviewed and approved by the Ethics Committee of School of Public Health, Sun Yat-sen University. The patients/participants provided their written informed consent to participate in this study. 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. 4. Discussion Recently, Sangkhae et.al assessed maternal hepcidin suppression in different iron status using mouse models and found that compared with nonpregnant levels, hepcidin mRNA and protein levels in iron-replete pregnant mice were already almost suppressed at the earliest examined time point, whereas levels of liver iron were not yet largely decreased (9). The lowering of hepcidin is presumed to precede liver iron mobilization. This indicates that decline in hepcidin reflects a physiologic state of high iron requirement, although the onset of low iron stores may not yet occur at this time, when timely iron supplementation may effectively prevent the development of ID/IDA in late pregnancy (16). Except for iron stores, another important factor that affects hepcidin levels is anemia itself, through erythroferrone (ERFE). Studies have demonstrated that The strengths in our study include a prospective cohort design and detailed information on potential confounders. We confirmed a negative association between first-trimester hepcidin and third- trimester IDA risk in Chinese population. Additionally, this is the first study to examine the associations between first-trimester hepcidin and third-trimester iron biomarkers within the Chinese setting. However, limitations should be acknowledged as followed. First, a major limitation in our study is the relatively small number of IDA women in each category of first-trimester serum hepcidin levels, and thus the EPV (events per variable) criterion was marginally met when performing logistic regression analysis. Given that pregnant women are usually advised a range of measures preventing ID/IDA (e.g., iron, folic acid and vitamin C supplementation) during routine pregnancy care, and that the study population was derived from a single hospital, it is not surprising that few women developed IDA in late pregnancy. Hence, although we revealed the potential negative association of first-trimester serum hepcidin with third-trimester IDA, our findings were mainly explorative and caution should be exercised when extrapolating to other groups. Second, the relationships between serum hepcidin levels in the second trimester and iron status in the third trimester were not evaluated in the present study. Finally, only CRP was collected in our study, yet hepcidin is affected by other inflammatory factors. Thus, further multi-center study with larger sample size is warranted to provide more evidence on the association between serum hepcidin and IDA 06 frontiersin.org 10.3389/fnut.2023.1147114 Sun et al. 10.3389/fnut.2023.1147114 Acknowledgments We gratefully acknowledge the participants for their continuous and enthusiastic participation in the investigation. We also appreciate the doctors and nurses involved in this study for their clinical technical support. Author contributions YaZ and PS designed the study. PS, SX, YuZ, XW, XL, HL, ZL, LZ, and FH enrolled the participants. SX and YaZ analyzed the data. PS, SX, and YuZ wrote the article. SX, YuZ, and YaZ interpreted the results. PS, XL, LZ, ZL, and FH contributed intellectually to the manuscript. YaZ had primary responsibility for final content. All authors have read and approved the final manuscript. Frontiers in Nutrition 1. Camaschella C. Iron-deficiency Anemia. N Engl J Med. (2015) 372:1832–43. doi: 10.1056/NEJMra1401038 Abioye AI, McDonald EA, Park S, Ripp K, Bennett B, Wu HW, et al. Maternal anemia type during pregnancy is associated with anemia risk among offspring during infancy. Pediatr Res. (2019) 86:396–402. doi: 10.1038/s41390-019-0433-5 12. Zaman B, Rasool S, Jasim S, Abdulah D. Hepcidin as a diagnostic biomarker of iron deficiency anemia during pregnancy. J Matern Fetal Neonatal Med. (2021) 34:1288–96. doi: 10.1080/14767058.2019.1635112 12. Zaman B, Rasool S, Jasim S, Abdulah D. Hepcidin as a diagnostic biomarker of iron deficiency anemia during pregnancy. J Matern Fetal Neonatal Med. (2021) 34:1288–96. doi: 10.1080/14767058.2019.1635112 13. Baingana RK, Enyaru JK, Tjalsma H, Swinkels DW, Davidsson L. The aetiology of anaemia during pregnancy: a study to evaluate the contribution of iron deficiency and 7. Daru J, Zamora J, Fernández-Félix BM, Vogel J, Oladapo OT, Morisaki N, et al. Risk of maternal mortality in women with severe anaemia during pregnancy and post 07 Frontiers in Nutrition frontiersin.org 10.3389/fnut.2023.1147114 Sun et al. common infections in pregnant Ugandan women. Public Health Nutr. (2015) 18:1423–35. doi: 10.1017/S1368980014001888 common infections in pregnant Ugandan women. Public Health Nutr. (2015) 18:1423–35. doi: 10.1017/S1368980014001888 21. Koenig MD, Tussing-Humphreys L, Day J, Cadwell B, Nemeth E. Hepcidin and iron homeostasis during pregnancy. Nutrients. (2014) 6:3062–83. doi: 10.3390/ nu6083062 14. Basu S, Kumar N, Srivastava R, Kumar A. Maternal and cord blood Hepcidin concentrations in severe Iron deficiency anemia. Pediatr Neonatol. (2016) 57:413–9. doi: 10.1016/j.pedneo.2015.09.012 14. Basu S, Kumar N, Srivastava R, Kumar A. Maternal and cord blood Hepcidin concentrations in severe Iron deficiency anemia. Pediatr Neonatol. (2016) 57:413–9. doi: 10.1016/j.pedneo.2015.09.012 22. Hedengran KK, Nelson D, Andersen MR, Stender S, Szecsi PB. Hepcidin levels are low during pregnancy and increase around delivery in women without iron deficiency-a prospective cohort study. J Matern Fetal Neonatal Med. (2016) 29:1506–8. doi: 10.3109/14767058.2015.1052396 15. Qi L, Qiu H, Fu Y. Significance of serum Hepcidin-25 level in diagnosis and treatment of iron deficiency anemia during pregnancy and its relationship with perinatal outcome. Chin J Clin Res. (2019) 32:1471–5. doi: 10.13429/j.cnki.cjcr.2019.11.004 15. Qi L, Qiu H, Fu Y. Significance of serum Hepcidin-25 level in diagnosis and treatment of iron deficiency anemia during pregnancy and its relationship with perinatal outcome. Chin J Clin Res. (2019) 32:1471–5. doi: 10.13429/j.cnki.cjcr.2019.11.004 23. Simavli S, Derbent AU, Uysal S, Turhan NÖ. Hepcidin, iron status, and inflammation variables among healthy pregnant women in the Turkish population. J Matern Fetal Neonatal Med. (2014) 27:75–9. doi: 10.3109/14767058.2013.804054 16. 1. Camaschella C. Iron-deficiency Anemia. N Engl J Med. (2015) 372:1832–43. doi: 10.1056/NEJMra1401038 Bah A, Pasricha S-R, Jallow MW, Sise EA, Wegmuller R, Armitage AE, et al. Serum Hepcidin concentrations decline during pregnancy and may identify Iron deficiency: analysis of a longitudinal pregnancy cohort in the Gambia. J Nutr. (2017) 147:1131–7. doi: 10.3945/jn.116.245373 16. Bah A, Pasricha S-R, Jallow MW, Sise EA, Wegmuller R, Armitage AE, et al. Serum Hepcidin concentrations decline during pregnancy and may identify Iron deficiency: analysis of a longitudinal pregnancy cohort in the Gambia. J Nutr. (2017) 147:1131–7. doi: 10.3945/jn.116.245373 24. Ceulemans M, Van de Vel J, Swinkels DW, Laarakkers CMM, Billen J, Van Calsteren K, et al. Hepcidin status in cord blood: observational data from a tertiary institution in Belgium. Nutrients. (2023) 15:546. doi: 10.3390/nu15030546 17. Lee S, Guillet R, Cooper EM, Westerman M, Orlando M, Pressman E, et al. Maternal inflammation at delivery affects assessment of maternal iron status. J Nutr. (2014) 144:1524–32. doi: 10.3945/jn.114.191445 25. Huibers MHW, Calis JC, Allain TJ, Coupland SE, Phiri C, Phiri KS, et al. A possible role for hepcidin in the detection of iron deficiency in severely anaemic HIV-infected patients in Malawi. PLoS One. (2020) 15:e0218694. doi: 10.1371/journal.pone.0218694 18. van Santen S, Kroot JJC, Zijderveld G, Wiegerinck ET, Spaanderman MEA, Swinkels DW. The iron regulatory hormone hepcidin is decreased in pregnancy: a prospective longitudinal study. Clin Chem Lab Med. (2013) 51:1395–401. doi: 10.1515/ cclm-2012-0576 18. van Santen S, Kroot JJC, Zijderveld G, Wiegerinck ET, Spaanderman MEA, Swinkels DW. The iron regulatory hormone hepcidin is decreased in pregnancy: a prospective longitudinal study. Clin Chem Lab Med. (2013) 51:1395–401. doi: 10.1515/ cclm-2012-0576 26. Sangkhae V, Nemeth E. Regulation of the Iron homeostatic hormone Hepcidin. Adv Nutr. (2017) 8:126–36. doi: 10.3945/an.116.013961 27. Ganz T, Jung G, Naeim A, Ginzburg Y, Pakbaz Z, Walter PB, et al. Immunoassay for human serum erythroferrone. Blood. (2017) 130:1243–6. doi: 10.1182/ blood-2017-04-777987 19. Perinatal MCSO. Guideline for diagnosis and treatment of iron deficiency and iron deficiency anemia in pregnancy. Chin J Perinat Med. (2014) 17:451. doi: 10.3760/cma.j. issn.1007-9408.2014.07.006 19. Perinatal MCSO. Guideline for diagnosis and treatment of iron deficiency and iron deficiency anemia in pregnancy. Chin J Perinat Med. (2014) 17:451. doi: 10.3760/cma.j. issn.1007-9408.2014.07.006 28. Abioye AI, Aboud S, Premji Z, Etheredge AJ, Gunaratna NS, Sudfeld CR, et al. Iron supplementation affects hematologic biomarker concentrations and pregnancy outcomes among Iron-deficient Tanzanian women. J Nutr. (2016) 146:1162–71. doi: 10.3945/jn.115.225482 20. Pan X-F, Wang L, Pan A. 1. Camaschella C. Iron-deficiency Anemia. N Engl J Med. (2015) 372:1832–43. doi: 10.1056/NEJMra1401038 Epidemiology and determinants of obesity in China. Lancet Diabetes Endocrinol. (2021) 9:373–92. doi: 10.1016/S2213-8587(21)00045-0 20. Pan X-F, Wang L, Pan A. Epidemiology and determinants of obesity in China. Lancet Diabetes Endocrinol. (2021) 9:373–92. doi: 10.1016/S2213-8587(21)00045-0 08 Frontiers in Nutrition 08 Frontiers in Nutrition frontiersin.org
https://openalex.org/W4220993945	https://vestnik.sibadi.org/jour/article/download/1403/747	Russian	null	Classification of contaminants in diesel engine oils	Vestnik SibADI	2,022	cc-by	9,920	ТРАНСПОРТ РАЗДЕЛ II ТРАНСПОРТ РАЗДЕЛ II РАЗДЕЛ II РАЗДЕЛ II ТРАНСПОРТ КЛАССИФИКАЦИЯ ЗАГРЯЗНИТЕЛЕЙ МОТОРНЫХ МАСЕЛ ДЛЯ ДИЗЕЛЬНЫХ ДВИГАТЕЛЕЙ С.В. Пашукевич Омский государственный технический университет (ОмГТУ) г. Омск, Россия sofia96@bk.ru, https://orcid.org/0000-0002-8111-4725 CLASSIFICATION OF CONTAMINANTS IN DIESEL ENGINE OILS Sofia V. Pashukevich Omsk State Technical University, Omsk, Russia sofia96@bk.ru, https://orcid.org/0000-0002-8111-4725 АННОТАЦИЯ Введение. Ухудшение состояния моторного масла в двигателе внутреннего сгорания (ДВС) напрямую связано с попаданием в картер различного рода загрязнителей. В зависимости от типа загрязнения из- меняется вид отложений на поверхностях деталей двигателя. Нельзя не отметить тот факт, что на работоспособность моторного масла чрезвычайно влияет процесс окисления, органические кислоты, возникающие в течение него, способствуют появлению коррозии на деталях ДВС. Также невосполнимый ущерб наносят вода, дизельное топливо, охлаждающая жидкость, частицы сажи, асфальтены и т. д. Материалы и методы. В данной работе представлены результаты широкого литературного обзора, направленного на изучение основных типов загрязнителей моторных масел. Приведены классификации по агрегатному состоянию загрязнителей, а также по возможным путям их проникновения в смазочный материал. Наиболее узко в данной статье рассмотрены жидкостные загрязнители. Для демонстрации негативного влияния попадания в моторное масло загрязнений приведены фотографии деталей дви- гателя внутреннего сгорания с отложениями, находящимися на поверхностях составных частей ДВС. Результаты. Приведена классификация основных загрязнителей моторных масел, указаны послед- ствия, возникающие вследствие попадания инородных соединений в рассматриваемый смазочный ма- териал. р Заключение. Установлено воздействие загрязнителей на детали ДВС и смазочного материала. На ос- нове классификации можно судить о причинах попадания и возможных последствиях воздействия загряз- нений на работу двигателя. КЛЮЧЕВЫЕ СЛОВА: моторное масло, двигатель внутреннего сгорания, окисление моторного масла, металлические поверхности, сажа, шлам, охлаждающая жидкость, дизельное топливо, износ, трение Статья поступила в редакцию 22.11.2021; одобрена после рецензирования 10.02.2022; принят публикации 28.02.2022. Автор прочитал и одобрил окончательный вариант рукописи. П ф й д ф Автор прочитал и одобрил окончательный вариант рукописи. Прозрачность финансовой деятельности: автор не имеет финансовой заинтересованности в представленных материалах и методах. Конфликт интересов отсутствует. Прозрачность финансовой деятельности: автор не имеет финансовой заинтересованност представленных материалах и методах. Конфликт интересов отсутствует. Для цитирования: Пашукевич С.В. Классификация загрязнителей моторных масел для дизельных двигате- лей// Вестник СибАДИ. 2022. Т.19, № 1(83). С. 84-100. https://doi.org/10.26518/2071-7296- 2021-19-1-84-100 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 84 Контент доступен под лицензией Creative Commons Attribution 4.0 License. © Пашукевич С.В., 2022 Контент доступен под лицензией Creative Commons Attribution 4.0 License. Том 19, № 1. 2022 Vol. 19, No. 1. 2022 TRANSPORT PART II DOI: https://doi.org/10.26518/2071-7296-2022-19-1-84-100 Original article ABSTRACT Introduction. The deterioration of engine oil in an internal combustion engine (ICE) is directly related to the ingress of various pollutants into the crankcase. Depending on the type of contamination, the type of sediment on engine part surfaces varies. It should be noted that the functioning of the motor oil is extremely affected by the oxidation process, and organic acids produced during the process contribute to corrosion of ICE parts. Water, diesel, cooling fluid, soot particles, asphaltenes, etc. also cause irreparable damage. l p p p g Materials and methods. This paper presents the results of an extensive literature review aimed at studying the main types of motor oil pollutants. Classifications are given for the aggregate state of pollutants, as well as for possible ways of their penetration into the lubricant. Liquid pollutants are the most narrowly considered in this article. To demonstrate the negative effect of contaminants entering the engine oil, photographs of internal combustion engine parts with sediments on the surfaces of the internal combustion engine components are presented.i g p p esults. The classification of the main pollutants of motor oils is given; the consequences arising from the ingr foreign compounds into the lubricant in question are indicated.f Conclusion. The effect of contaminants on the parts of the internal combustion engine and the lubricant has been established. On the basis of the classification, it is possible to judge the causes of entry and the possible consequences of the impact of contamination on the operation of the engine. KEYWORDS: engine oil, internal combustion engine, engine oil oxidation, metal surfaces, soot, sludge, coolant, diesel fuel, wear, friction The article was submitted 22.11.2021; approved after reviewing 10.02.2022; accepted for publication 28.02.2022. i ВВЕДЕНИЕ свойств масел, включая защитные присадки: 1) присадки для защиты поверхности и ан- тикоррозионные присадки для уменьшения трения, коррозии и снижения контакта с ме- таллом; антикоррозионные присадки для пре- дотвращения ржавления внутренней части двигателя; присадки, очищающие поверхность от отложений и 2) присадки для улучшения ра- боты двигателя за счет стабилизации вязкости масла, чтобы уменьшить изменение вязкости при нагревании [3]. свойств масел, включая защитные присадки: 1) присадки для защиты поверхности и ан- тикоррозионные присадки для уменьшения трения, коррозии и снижения контакта с ме- таллом; антикоррозионные присадки для пре- дотвращения ржавления внутренней части двигателя; присадки, очищающие поверхность от отложений и 2) присадки для улучшения ра- боты двигателя за счет стабилизации вязкости масла, чтобы уменьшить изменение вязкости при нагревании [3]. Моторное масло способствует уменьше- нию трения, предотвращению коррозии де- талей двигателя, а также оно отводит тепло, особенно с днища поршня. Загрязнение дан- ного вида смазочного материала происходит и при нормальной работе двигателя внутренне- го сгорания. Среди наиболее распространен- ных загрязнителей – вода, топливо (несгорев- шее или частично сгоревшее), охлаждающая жидкость, которые смешиваются с моторным маслом и вызывают в совокупности с высо- кой температурой его окисление. Конечно, одним из основных загрязнителей смазочных масел, приводящим к ухудшению их состава, соответственно, и снижению качества работо- способности, является топливо. Разбавление масла топливом ускоряет износ компонентов двигателя и снижает рабочие характеристики моторного масла и срок его службы. Загряз- ненное моторное масло может вызвать отказ компонентов двигателя, поэтому в моторное масло добавляют присадки, чтобы умень- шить окислительное повреждение, вызванное несгоревшим топливом, следами воды, нали- чием гликолей и других загрязнителей в масле [1, 2]. р р [ ] Смазочное масло образует пленку между движущимися поверхностями, которая умень- шает трение. Однако моторное масло – кла- дезь примесей. Они бывают в виде твердых, жидких и газообразных загрязнителей. При неконтролируемом воздействии, особенно в условиях низких температур, в моторном масле эти загрязнители могут накапливаться до чрезмерного уровня. Высокий уровень за- грязнения смазочного материала вызывает износ механических компонентов, а также раз- рушение присадок и базовых масел. Результат – снижение производительности, сокращение срока службы двигателя и ограниченный ре- сурс моторного масла. Загрязнения смазочно- го материала уменьшают время достижения маслами своих предельных показателей и ухудшают параметры работы дизельных дви- гателей [4]. На рисунке 1 указаны основные типы загрязнений. Моторные масла состоят из двух частей: базового масла (различной природы) и хими- ческих присадок для улучшения качества и Том 19, № 1. 2022 Vol. 19, No. 1. The article was submitted 22.11.2021; approved after reviewing 10.02.2022; accepted for publication 28.02.2022. i The authors have read and approved the final manuscript. Financial transparency: the authors have no financial interest in the presented materials or methods. There is no conflict of interest. For citation: Pashukevich S.V. Classification of contaminants in diesel engine oils. The Russian Automobile and Highway Industry Journal. 2022; 19 (1): 84-100. https://doi.org/10.26518/2071-7296- 2022-19-1-84-100 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 85 Content is available under the license Creative Commons Attribution 4.0 License. © Pashukevich S.V., 2022 Content is available under the license Creative Commons Attribution 4.0 License. S.V., 2022 85 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 ТРАНСПОРТ РАЗДЕЛ II РАЗДЕЛ II © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal ВВЕДЕНИЕ 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 86 Рисунок 1 – Основные типы загрязнителей моторного масла Figure 1 – Main types of engine oil pollutants Рисунок 1 – Основные типы загрязнителей моторного масла Figure 1 – Main types of engine oil pollutants Рисунок 1 – Основные типы загрязнителей моторного масла Figure 1 – Main types of engine oil pollutants 86 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 TRANSPORT PART II Рисунок 2 – Пути попадания твёрдых загрязнителей в моторное масло Figure 2 – Ways for solid contaminants to enter the engine oil Рисунок 2 – Пути попадания твёрдых загрязнителей в моторное масло Figure 2 – Ways for solid contaminants to enter the engine oil Figure 2 – Ways for solid contaminants to enter the engine oil Возможные пути попадания загрязнений в систему смазки двигателя, а также примеры загрязнителей для каждого типа приведены на рисунке 2. Сильная ударная волна давления, созда- ваемая во время сгорания, выталкивает газы через зазоры поршневых колец в картер дви- гателя. Этот процесс, известный как прорыв, переносит загрязнения в моторное масло. Ча- стицы также могут задерживаться в масляной пленке. Затем они попадают в масляный под- дон при следующем ходе поршня. Выхлопные газы попадают в смазочное масло аналогич- но. Эти выхлопные газы включают несгорев- шее топливо, воду, окись азота, сажу и другие частично сгоревшие углеводороды. Более вы- сокие уровни рециркуляции выхлопных газов увеличивают загрязнение смазочного масла твердыми частицами [9, 10]. Производители дизельных двигателей проявляют большую осторожность во время процессов производства и сборки для обе- спечения высокого уровня качества. Однако материалы для изготовления отливок, меха- ническая стружка, абразивные материалы, по- лировальные пасты и даже ворсинки остаются после изготовления и капитального ремонта. Эти загрязнители могут быстро повредить подвижные части двигателя [5, 6, 7]. Внешнее проникновение является основ- ным источником загрязнения твердыми части- цами. Частицы в воздухе в виде песка, соли и других минералов попадают через систему воздухоподачи двигателя и смешиваются с распыленным топливом, которое сжимает- ся, а затем сгорает. Поскольку большинство этих частиц имеет температуры плавления значительно выше температур, достигаемых при сгорании дизельного топлива, в процессе горения они остаются твердыми абразивны- ми частицами. Загрязнение воздуха является основной причиной износа пары «кольцо–ци- линдр» [8]. Одним из путей для внешнего попадания загрязняющих веществ в моторное масло яв- ляется вентиляция картера, что за счет раз- режения может способствовать поступлению большого количества пыли и воды, последние попадают непосредственно в масляный под- дон. ВВЕДЕНИЕ Кроме того, вода в сочетании с антиф- ризными соединениями, такими как гликоль, может подаваться в масляную полость под давлением через дефектные прокладки голов- ки или иногда через трещину в блоке. Внутреннее образование загрязнений про- исходит в результате износа механических © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 87 ТРАНСПОР РАЗДЕЛ II РАЗДЕЛ II ТРАНСПОРТ деталей и разрушения смазки. Износ механи- ческих компонентов в результате истирания, усталости, адгезии и коррозии приводит к по- паданию вредных частиц в масло. Остатки из- носа находятся в виде частиц твердого метал- ла и оксидов абразивных металлов. Частицы износа, размеры которых не контролируются стандартной фильтрацией, могут накапливать- ся и сильно загрязнять смазочный материал [11, 12, 13]. деталей и разрушения смазки. Износ механи- ческих компонентов в результате истирания, усталости, адгезии и коррозии приводит к по- паданию вредных частиц в масло. Остатки из- носа находятся в виде частиц твердого метал- ла и оксидов абразивных металлов. Частицы износа, размеры которых не контролируются стандартной фильтрацией, могут накапливать- ся и сильно загрязнять смазочный материал [11, 12, 13]. один цилиндр не является идеально круглым, ни одна поверхность не является идеально гладкой, и нет идеального уплотнения. На каждом цилиндре двигателя имеется пере- крестная штриховка из микроскопических ка- навок, именно они помогают со смазкой, давая моторному маслу возможность задерживаться на поверхности зеркала цилиндра. В данном случае определенное количество топлива и выхлопных газов будет выходить через уплот- нение между поршневыми кольцами и цилин- дром. Ухудшение качества моторного масла – это потеря важных свойств масла плюс нако- пление вредных веществ, полученных в ходе эксплуатации автомобильной техники. Эти материалы включают кислоты, шламы, гели и осадки присадок. Эти загрязняющие веще- ства могут изнашивать движущиеся детали, а также засорять проточные каналы и поверхно- сти теплообмена. Если в масле скапливаются продукты износа и материалы, образующиеся в результате разрушения моторного масла, то в результате увеличивается износ и образует- ся лавинообразное поступление загрязнений. Процесс изнашивания поверхностей частица- ми и образование новых частиц, которые, в свою очередь, вызывают больший износ, из- вестен как цепная реакция износа [14]. Мокрая укладка – явление, которое про- исходит в двигателях с воспламенением от сжатия. ВВЕДЕНИЕ В то время как бензиновый двига- тель, например, воспламеняет свою топливо- воздушную смесь с помощью искры в подхо- дящее время для инициирования сгорания, двигатель с воспламенением от сжатия пола- гается исключительно на теплоту воздушного заряда в цилиндре во время процесса впры- ска, чтобы инициировать сгорание. Двигатель, который не достиг рабочей температуры, бу- дет иметь более низкую эффективность сго- рания, чем «горячий» двигатель. Это связано с природой процесса горения, заряд сжатого воздуха в двигателе при рабочей температуре (или близкой к ней) содержит больше тепла, чем заряд сжатого воздуха двигателя, который значительно ниже рабочей температуры. © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Попадание дизельного топлива в мотор- ное масло Это не означает, что дизельный двигатель следует нагружать сразу после запуска, однако нельзя допускать, чтобы он работал на холостом ходу в течение продолжительных периодов време- ни; от 3 до 5 мин (в зависимости от темпера- туры окружающей среды) работы на холостом ходу обычно достаточно, чтобы давление мас- ла стабилизировалось, а температура двига- теля достигла подходящего диапазона перед поездкой в холодных условиях. В теплую пого- ду вполне достаточно 1–2 мин работы двигате- ля на холостом ходу перед поездкой. ет повышенным износам деталей двигателей. Моторное масло – это единственный слой защиты между движущимися частями двига- теля. Тонкая пленка моторного масла подвер- гается экстремальному давлению и усилиям, поскольку она создает барьер между двумя движущимися / вращающимися поверхностя- ми. По мере разжижения масла этот барьер может выйти из строя, увеличивая скорость износа. По мере износа компонентов физиче- ские допуски между двумя поверхностями уве- личиваются [17]. Попадание воды в моторное масло д д р Вода может существовать в масле в трех состояниях или фазах. Первое состояние, из- вестное как растворенная вода, характеризу- ется отдельными молекулами воды, рассре- доточенными в масле. По этой причине масло может содержать значительную концентрацию растворенной воды без видимых признаков ее присутствия. Как только количество воды превысит максимальный уровень, чтобы она оставалась растворенной, масло насыщается. В этот момент вода переходит во взвешенное состояние и находится в виде микроскопиче- ских капель, известных как эмульсия. Попадание топлива в моторное масло ста- ло еще большей проблемой с введением са- жевого фильтра, который требует цикла реге- нерации для периодической очистки фильтра от твердых частиц. Во время цикла регене- рации в выхлопной системе требуется значи- тельное количество тепла, чтобы эти частицы сгорели. Следовательно, необходимо впры- скивать топливо в поток выхлопных газов, повышая температуру и создавая условия, в которых эти частицы могут полностью сгореть и выбрасываться из выхлопной трубы. Добавление большего количества воды к эмульгированной смеси масло/вода приведет к разделению двух фаз с образованием слоя свободной воды, а также свободного и/или эмульгированного масла. В системе смазки двумя наиболее вред- ными фазами являются свободная и эмуль- гированная вода. В подшипниках, например несжимаемости воды по отношению к нефти, может привести к потере гидродинамической масляной пленки, что, в свою очередь, приво- дит к чрезмерному износу. Даже один процент воды в масле может сократить срок службы подшипников скольжения на 90%. р ру Поскольку топливо впрыскивается в ци- линдр во время такта выпуска, оно следует по пути выходящего потока выхлопных газов и входит в выпускной коллектор в качестве рас- пыленного неочищенного топлива. Попадание дизельного топлива в мотор- ное масло Попадание дизельного топлива в мотор- ное масло Разбавление моторного масла дизель- ным топливом – явление, которое возникает естественным образом при каждом запуске двигателя. Данная ситуация наиболее часто встречается в дизельных двигателях, посколь- ку их специфика работы основана на цикле воспламенения от сжатия и, следовательно, эффективность сгорания полностью зависит от температуры двигателя, в частности от тем- пературы камеры сгорания. Теоретически чем холоднее камера сгорания, тем выше степень разбавления моторного масла топливом. Ос- новная проблема, связанная с разбавлением моторного масла топливом в дизельных дви- гателях, – это снижение вязкости моторного масла, которая потенциально оказывает зна- чительное влияние на ухудшение качества смазочного материала между различными движущимися частями двигателя в условиях высоких нагрузок, а также в условиях низких температур [15]. р р ур В условиях, когда температура в камере сгорания низкая, впрыскиваемое топливо име- ет тенденцию воспламеняться на более позд- нем этапе такта сжатия, чем когда температу- ра в камере сгорания выше. В таких случаях распыляемое форсункой топливо будет иметь тенденцию отлагаться на стенках цилиндра, смывая моторное масло и в конечном итоге проходя через неплотности поршневых колец и попадая в картер [16]. Мокрая укладка и, следовательно, разбав- ление моторного масла топливом обычно наи- более заметно в условиях низких температур сгорания. Эти события можно разделить на два случая: длительная работа двигателя на холостом ходу и период, в течение которо- го работающий двигатель еще не прогрелся до рабочей температуры. Первоначальный запуск дизельного двигателя, особенно в хо- лодную погоду, также представляет собой условия, при которых разжижение моторного масла топливом происходит с большей ско- ростью. Несмотря на то, что для обеспечения возможности сгорания вырабатывается доста- точно тепла, стенки цилиндра и общая темпе- ратура самого цилиндра относительно низкие. Обычной причиной разбавления моторно- го масла топливом является утечка (прорыв) газов между поршневыми кольцами и не- плотностями цилиндропоршневой группы. В идеальной системе поршень должен идеаль- но прилегать к стенке цилиндра. Однако ни Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 88 TRANSPORT PART II По этой причине обычно не рекомендуется оставлять работать дизельный двигатель на холостом ходу, чтобы довести его до полной рабочей температуры. Фактически дизельный двигатель редко достигает рабочей температу- ры в течение разумного промежутка времени без привода, двигатель нагревается за гораздо меньшее время, если приложить нагрузку. Попадание дизельного топлива в мотор- ное масло Когда продукты сгорания просачиваются мимо поршней, водяной пар поступает в об- ласть картера, следовательно, паровая часть вытекающих газов будет конденсироваться. При первом запуске холодного двигателя воз- дух в картере нагревается от горячего днища поршня. Из насыщенного паром воздуха будет конденсироваться влага, когда он соприкаса- ется с холодным маслом, холодными сторона- ми картера и стенками масляного поддона. Однако даже на этих уровнях все равно может быть нанесен значительный ущерб. Во- обще не бывает слишком мало воды и необ- ходимо приложить все разумные усилия для минимального загрязнения моторного масла водой. Самое главное, что значительная прямая конденсация воды будет происходить в рабо- чей части цилиндра до тех пор, пока темпе- ратура будет ниже 100 °С. Такое отложение, конечно, незначительно сказывается в период теплой погоды, но на него приходится основ- ная часть воды, образующейся при зимней эксплуатации автомобилей. Вода не только оказывает прямое вредное воздействие на компоненты машин, но также играет непосредственную роль в скорости ста- рения смазочных масел. Присутствие воды в смазочном масле может привести к десяти- кратному увеличению степени окисления, что приведет к преждевременному старению мас- ла, особенно в присутствии каталитических металлов, таких как медь, свинец и олово. Как только температура масла поднимет- ся выше 100 оС, при циркуляции масла вода, присутствующая в виде эмульгированной, ис- паряется. Таким образом, в летних условиях вождения при каждом запуске двигателя будет образовываться некоторое количество воды, и эффект утечки через поршни, конечно, во многом одинаков независимо от температуры атмосферы. Так как температура деталей ка- меры сгорания двигателя при нормальной ра- боте будет более 150 оС, а температура масла не менее 120 оС, смазка будет отторгать воду быстрее, чем она поступает [20]. Кроме того, известно, что некоторые типы синтетических масел, такие как сложные эфи- ры фосфорной кислоты и сложные эфиры двухосновной кислоты, вступают в реакцию с водой, что приводит к разрушению основного компонента и образованию кислот. Загрязнение влагой может повлиять не только на базовое масло. Некоторые добавки, такие как сернистые присадки и фенольные антиоксиданты, легко гидролизуются водой, что приводит как к гибели присадок, так и к об- разованию кислотных побочных продуктов. Эмульсия более вязкая, чем чистое масло, но не настолько, чтобы мешать нормальной работе, по крайней мере, если количество воды не превышает 6%. Эти кислотные побочные продукты могут затем вызвать коррозионный износ, особенно в компонентах, содержащих мягкие металлы, такие как баббит, используемый в подшипни- ках скольжения, а также в деталях из бронзы и латуни. Попадание дизельного топлива в мотор- ное масло Посколь- ку при впрыске этого топлива не происходит возгорания, возникает тенденция отложения на стенках цилиндра, и в процесс вступает ранее описанное условие «мокрой укладки». Ответом на проблемы, с которыми сталкива- ются циклы пост-впрыска для регенерации са- жевого фильтра, было введение 9-го или 7-го инжектора, в зависимости от числа цилиндров двигателя, чтобы топливо могло впрыскивать- ся в поток выхлопных газов, предотвращая попадание его в масло. Наиболее серьезные проблемы при разбавлении моторного масла топливом вызывают снижение вязкости и де- градация масла. Дизельное топливо в масле увеличивает скорость разложения моторного масла, а также снижает его вязкость. По мере того, как вязкость уменьшается, и масло стано- вится более жидким, а это приводит к потере несущей способности масел, что способству- С подшипниками качения дело обсто- ит еще хуже. Мало того, что вода разрушает прочность масляной пленки, но и свободная, и эмульгированная вода при экстремальных температурах и давлениях, возникающих в зоне нагрузки подшипника качения, может привести к мгновенному испарению, вызыва- ющему эрозионный износ. При определенных условиях молекулы воды могут быть разорваны на составляющие атомы кислорода и водорода в результате вы- сокого давления, создаваемого в зоне контак- та деталей подшипника качения. Из-за своего относительно небольшого размера ионы водо- рода, образующиеся в этом процессе, могут поглощаться поверхностью дорожки качения © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 89 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 РАЗДЕЛ II ТРАНСПОРТ Небольшая доля воды, которая присутству- ет в моторном масле, не окажет большого вли- яния ни на вязкость смазочного материала, ни на коэффициент трения. Когда масло не будет подвергаться эмульгированию, вода не причи- нит вреда при условии, что она попадет в под- дон таким образом, что не сможет попасть во всасывающее устройство насоса. подшипника, что приводит к явлению, извест- ному как водородное охрупчивание. Водородная хрупкость приводит к тому, что материал подшипника становится хрупким и склонным к растрескиванию под поверхно- стью дорожки качения. Когда эти подповерх- ностные трещины распространяются на по- верхность, это может привести к появлению точечной коррозии и сколов. Опасная точка достигается, когда либо смесь моторного масла и воды становится слишком вязкой, либо скопление свободной воды достигает впускного отверстия насоса [19]. Поскольку воздействие свободной и эмуль- гированной воды более вредно по сравнению с растворенной водой, общее практическое правило заключается в том, чтобы уровень влажности оставался значительно ниже точки насыщения. Для большинства используемых масел это означает от 100 до 300 частей на миллион или меньше в зависимости от типа масла и температуры. © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Попадание дизельного топлива в мотор- ное масло Другие добавки, например деэмуль- гирующие агенты, диспергаторы, детергенты и ингибиторы ржавчины, могут вымываться излишней влажностью. Это приводит к обра- зованию и накоплению шлама и отложений, засорению фильтров и плохой деэмульгируе- мости масла/воды [18]. Основная масса воды, которая попадает в масляный поддон, образуется в результате прямой конденсации на внутренних стенках цилиндров. Опускающийся поршень обнажает поверхность железа с масляным покрытием при температуре всего на несколько граду- сов выше температуры воды в рубашке. При температуре воды 65 оС внутренняя стенка © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 90 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 PART II TRANSPORT цилиндра может быть значительно горячее, чем сама вода, но при низких температу- рах разница очень незначительна. В течение большей части рабочего хода и всего такта выпуска цилиндр заполняется газами, значи- тельную часть которых составляет водяной пар. Во время тактов рабочего хода и выпуска времени достаточно для того, чтобы в цилин- дре сконденсировалось заметное количество воды. Образовавшаяся таким образом вода не выходит вместе с выхлопными газами, она остается в цилиндре и попадает в масляный поддон. В состав антифризов, используемых в ка- честве охлаждающих жидкостей, входит ши- рокий ассортимент металлоорганических и органических добавок. Они используются для защиты металлов в системе охлаждения от коррозии/кавитации, контроля накипи, пре- дотвращения пенообразования и поддержа- ния pH. Общие примеры добавок включают различные фосфаты, борат натрия, молибдат, силикат натрия, себацинат калия и нитрат на- трия. Как и присадки в моторном масле, эти присадки будут способствовать изменению концентрации элементов натрия, бора, калия, кремния и фосфора в охлаждающей жидкости. цилиндра может быть значительно горячее, чем сама вода, но при низких температу- рах разница очень незначительна. В течение большей части рабочего хода и всего такта выпуска цилиндр заполняется газами, значи- тельную часть которых составляет водяной пар. Во время тактов рабочего хода и выпуска времени достаточно для того, чтобы в цилин- дре сконденсировалось заметное количество воды. Образовавшаяся таким образом вода не выходит вместе с выхлопными газами, она остается в цилиндре и попадает в масляный поддон. Если бы смазка цилиндра или поршня была идеальной, то есть если бы смазка была настолько хорошей, насколько это возмож- но в идеальном подшипнике без какого-либо металлического контакта между поршневыми кольцами и стенкой цилиндра, вероятно, не имело бы большого значения, существовал ли небольшой процент воды в свободном состоя- нии или в виде эмульсии. Но быстрота износа поршневых колец является доказательством того, что смазка далека от идеала. Попадание дизельного топлива в мотор- ное масло Кроме того, в период низкотемпературных условий, когда осаждение воды происходит наиболее быстро, осаждение тяжелых остатков топлива также максимально, и они имеют мощное действие растворителя масляной пленки. Естественное несовершенство смазки плюс растворение масляной пленки означают, что части стенки цилиндра фактически обнажаются при рабо- чем ходе. При попадании в моторное масло вода будет иметь тенденцию накапливаться в этих открытых местах и смачивать их, что затруднит восстановление масляной пленки. Можно допустить, что эффект разбавителя в виде топлива при смывании масляной плен- ки со стенок оказывает большее влияние на ускорение износа, чем осаждение неболь- шого количества задействованной воды, но последним нельзя полностью пренебрегать. Хотя может быть трудно определить точную величину с точки зрения снижения износа по- глощающего воду свойства масла, тот факт, что оно должно быть полезным, – неоспорим [21, 22, 23]. Гликоль может попадать в моторные масла и другие смазочные масла различными спосо- бами, такими как дефектные или изношенные уплотнения, «выдутые» прокладки головки блока цилиндров, неправильно затянутые бол- ты с головкой, термически деформированные или треснувшие головки цилиндров (от низко- го уровня охлаждающей жидкости до заеда- ния термостата), треснувший блок или головка блока цилиндров от замерзшей охлаждающей жидкости, неправильно обработанные поверх- ности головки и блока цилиндров, коррози- онное повреждение гильз цилиндров, кави- тационная эрозия/коррозия гильз цилиндров, электрохимическая эрозия, повреждение или корродирование сердечников охладителя, от- каз уплотнения водяного насоса и засорение сливного отверстия. Фактически, по оценкам крупного произ- водителя дизельных двигателей, 53% всех серьезнейших отказов двигателей происхо- дят из-за утечек охлаждающей жидкости. Для многих дизельных двигателей и двигателей, работающих на природном газе, самый высо- кий риск загрязнения возникает, когда двига- тель не заведен. В таких случаях охлаждение двигателя при периодической работе может привести к внутренним утечкам, связанным с термической ползучестью, например в голов- ках цилиндров, где существует риск рецессии или смещения уплотнительных прокладок. Бо- лее высокое гидростатическое давление ох- лаждающей жидкости по отношению к систе- ме смазочного масла увеличивает риск, когда двигатель не работает. Это может привести к попаданию охлаждающей жидкости в смазку [25, 26]. Загрязнение моторного масла гликолем Загрязнение моторного масла гликолем Гликоль, основной ингредиент антифриза, обычно смешивается с водой в соотношении 50/50 с образованием жидкого «хладагента» для передачи тепла, повышения температуры кипения и понижения температуры замерза- ния. Когда в состав входят присадки, охлажда- ющая жидкость может эффективно защищать от коррозии и кавитации [24]. Другой распространенный источник утечки в двигателях с мокрыми гильзами цилиндров связан с химико-механической перфорацией гильз, вызванной паровой кавитацией. Это происходит, когда гильзы сильно вибрируют 91 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 ТРАНСПОРТ РАЗДЕЛ II РАЗДЕЛ II ТРАНСПОРТ (со стороны нагрузки) в ритме движения порш- ня, сжатия и сгорания. Это движение застав- ляет часть волн давления создавать области отрицательного давления, в которых образу- ются пузырьки пара (полости). При горении в камере сгорания пузырьки пара лопаются со скоростью звука, вызывая струи жидкости и высочайшее поверхностное давление. Такая локализованная энергия может буквально взорвать небольшие дыры в защитной оксид- ной пленке на стенке гильзы, как это проис- ходит в паровой кавитации в гидравлических насосах. также глиоксаля. Этиленгликоль, попав в кар- тер, может деградировать до любого из ряда низких органических альдегидов, кислот или диацидов. Кроме того, в дополнение к полиме- ризации в лаковые соединения эти продукты разложения очень агрессивны к сплавам, со- держащим медь и свинец. (со стороны нагрузки) в ритме движения порш- ня, сжатия и сгорания. Это движение застав- ляет часть волн давления создавать области отрицательного давления, в которых образу- ются пузырьки пара (полости). При горении в камере сгорания пузырьки пара лопаются со скоростью звука, вызывая струи жидкости и высочайшее поверхностное давление. Такая локализованная энергия может буквально взорвать небольшие дыры в защитной оксид- ной пленке на стенке гильзы, как это проис- ходит в паровой кавитации в гидравлических насосах. Нерастворимые в масле продукты разло- жения гликоля очень агрессивны к подшипни- кам даже при низкой концентрации [31]. Потеря дисперсности и засорение филь- тра Кислоты и вода, образующиеся в масле картера в результате загрязнения охлаждаю- щей жидкостью, часто нарушают диспергиру- емость сажи даже при низком её содержании. Загрязнение моторного масла гликолем Как только сажа начинает оседать на фильтре, может возникнуть цепная реакция, при этом возникают следующие отказы: потеря проти- воизносной защиты, липкий осадок на поверх- ностях клапанной площадки и углеродистые отложения на кольцевых канавках, посадоч- ных площадках днища поршня, компонентах клапанного механизма и масляных каналах к подшипникам и т. д. Затем цепная реакция обретает новую жизнь, поскольку детергенты и диспергаторы, поступающие со свежим мо- торным маслом, могут мобилизовать шлам и отложения. Затем, через несколько минут по- сле замены масла и фильтра, новый фильтр может снова засориться [32, 33]. Ниже (рису- нок 3) приводится краткое изложение этой ре- акции. Повреждение может быть дополнительно вызвано химическим воздействием на металл, обнаженный во время этой кавитации. Со временем это может привести к перфорации хвостовика и утечке охлаждающей жидкости. Режим разрушения распространяется за счет комбинации механического (локализованная кавитация) и химического (коррозия обнажен- ного основного металла) воздействия [27]. Было обнаружено, что некоторые добавки, такие как молибдат и нитрит натрия, резко за- медляют развитие кавитационной коррозии. Если защитная оксидная пленка расслаива- ется под действием энергии кавитации, до- бавка преобразует барьерную пленку, чтобы остановить дальнейшее её распространение. Однако для обеспечения качества важна кон- центрация этих добавок, вводимых в охлажда- ющую смесь. Недостаточное содержание мо- жет привести к ускоренной точечной коррозии, в то время как избыточная концентрация мо- жет вызвать гелеобразование охлаждающей жидкости, коррозию припоя на основе свинца и другие проблемы. Образование сажи в картере двигателя Образование сажи в картере двигателя Дизельные двигатели работают при более высоком соот- ношении воздуха и топлива, но в качестве то- плива используются более тяжёлые фракции углеводородов, что, как правило, и приводит к более высокому уровню образования сажи в двигателе. Большинство современных дизель- ных двигателей работает с использованием прямого впрыска топлива, которое закручива- ется в камере сгорания для содействия сме- шиванию топлива и воздуха с образованием однородной рабочей смеси. Возгорание на- чинается вблизи точки впрыска и происходит очень быстро в виде диффузионного пламени. В этот момент воздух и топливо хорошо пере- мешиваются, но особенно на переходных ре- жимах смесь очень богата топливом, что при- водит к образованию очень высокого уровня сажи. После диффузионного горения этот про- цесс проходит через остальную часть камеры сгорания путем пиролизного горения, в резуль- тате которого медленно сжигается большая часть оставшегося топлива. Это медленное сжигание приводит к образованию большего количества твердых частиц (сажи) и несгорев- ших углеводородов в конце процесса горения. В процессе горения образуются и разрушают- ся частицы сажи. Они создаются описанным выше процессом и разрушаются окислением. Окисление – это механизм, который возникает, когда сажа или предшественники сажи всту- пают в контакт с различными видами окисли- телей. Когда это происходит, углеводороды, попавшие в сажу, выгорают, и размер частиц уменьшается. Во время стадии процесса диф- фузионного горения частицы сажи, образую- щиеся на начальной стадии процесса горения, вступают в контакт с гораздо большим объе- мом воздуха по сравнению с топливом, и боль- шая часть частиц сажи окисляется. Требуется дальнейшее окисление, чтобы уменьшить количество сажи, окончательно израсходо- ванной. Когда выпускной клапан открывается, продукты сгорания выбрасываются в выхлоп- ную систему, которая содержит больше видов окислителей. Окислительные каталитические нейтрализаторы используются для дальней- шего уменьшения количества сажи, выбрасы- ваемой из выхлопной трубы. Большая часть образующейся сажи окисляется до выхлопа. Возможно, именно поэтому большинство ча- стиц сажи поглощается смазкой и относитель- но мало расходуется [37, 38]. К б Из сажи, образующейся в двигателе, только 29% попадает в атмосферу через выхлопную трубу, а остальная часть оседает на стенках цилиндра и головке поршня. Из сажи, которая остается в двигателе (в основном в смазке), 3% приходится на выбросы газов, остальное происходит в результате соскабливания порш- невыми кольцами отложений сажи в цилиндре, которые затем попадают в масляный поддон. Затем она перемещается по двигателю, где сажу можно ввести в контакты подвижных ком- понентов. Внутри клапанного механизма име- ется множество подвижных компонентов, все различной геометрии и движения. Существуют контакты скольжения, качения-скольжения и возвратно-поступательного движения, некото- рые из них конформны. Образование сажи в картере двигателя Это может быть частично связано с процессами окисления, че- рез которые проходят продукты сгорания. Как упоминалось выше, сажа, содержащаяся в смазочных материалах, имеет очень высокое содержание углерода и низкое содержание кислорода. Частицы сажи, как правило, счита- ются чрезвычайно твердыми по отдельности и гораздо более мягкими при агломерации. Сажа, взятая из двигателя, работающего с EGR, немного тяжелее, чем сажа из двигателя без EGR. Это увеличение твердости, возмож- но, связано с процессом вторичного нагрева и окисления, который испытывают частицы. различий в механизмах сгорания. Дизельные двигатели работают при более высоком соот- ношении воздуха и топлива, но в качестве то- плива используются более тяжёлые фракции углеводородов, что, как правило, и приводит к более высокому уровню образования сажи в двигателе. Большинство современных дизель- ных двигателей работает с использованием прямого впрыска топлива, которое закручива- ется в камере сгорания для содействия сме- шиванию топлива и воздуха с образованием однородной рабочей смеси. Возгорание на- чинается вблизи точки впрыска и происходит очень быстро в виде диффузионного пламени. В этот момент воздух и топливо хорошо пере- мешиваются, но особенно на переходных ре- жимах смесь очень богата топливом, что при- водит к образованию очень высокого уровня сажи. После диффузионного горения этот про- цесс проходит через остальную часть камеры сгорания путем пиролизного горения, в резуль- тате которого медленно сжигается большая часть оставшегося топлива. Это медленное сжигание приводит к образованию большего количества твердых частиц (сажи) и несгорев- ших углеводородов в конце процесса горения. В процессе горения образуются и разрушают- ся частицы сажи. Они создаются описанным выше процессом и разрушаются окислением. Окисление – это механизм, который возникает, когда сажа или предшественники сажи всту- пают в контакт с различными видами окисли- телей. Когда это происходит, углеводороды, попавшие в сажу, выгорают, и размер частиц уменьшается. Во время стадии процесса диф- фузионного горения частицы сажи, образую- щиеся на начальной стадии процесса горения, вступают в контакт с гораздо большим объе- мом воздуха по сравнению с топливом, и боль- шая часть частиц сажи окисляется. Требуется дальнейшее окисление, чтобы уменьшить количество сажи, окончательно израсходо- ванной. Когда выпускной клапан открывается, продукты сгорания выбрасываются в выхлоп- ную систему, которая содержит больше видов окислителей. Окислительные каталитические нейтрализаторы используются для дальней- шего уменьшения количества сажи, выбрасы- ваемой из выхлопной трубы. Большая часть образующейся сажи окисляется до выхлопа. Возможно, именно поэтому большинство ча- стиц сажи поглощается смазкой и относитель- но мало расходуется [37, 38]. Концентрация образующихся частиц сажи различий в механизмах сгорания. Образование сажи в картере двигателя р р р д Сажа – это микроскопическая углеродистая частица, являющаяся продуктом неполного сгорания углеводородов (в данном случае ди- зельного топлива). Сажа состоит из углерода, золы и ненасыщенных (несгоревших) углево- дородов. Ненасыщенные углеводороды по существу представляют собой ацетилен и по- лициклические ароматические углеводороды. Эти компоненты обладают особенно высоким уровнем кислотности и летучести [34 - 36]. Из- мерения показали, что сажа обычно содержит 90% углерода, 4% кислорода и 3% водорода, а остальное состоит из азота, серы и следов металлов. Было измерено, что отдельные или первичные частицы сажи от сжигания дизель- ного топлива составляют приблизительно 40 нм. Из-за коллоидных свойств сажи частицы агломерируются максимум примерно до 500 нм, при этом средний размер агломерата сажи составляет 200 нм. Частицы сажи, как прави- ло, более распространены в дизельных дви- гателях, чем в бензиновых двигателях, из-за ру р Негативные моменты, связанные с попа- данием охлаждающей жидкости в моторное масло, заключаются в следующих моментах: постоянная утечка охлаждающей жидкости постепенно приводит к высокой концентра- ции гликоля в смазочном материале; период интенсивной эксплуатации, во время которой моторное масло становится очень горячим и вызывает разложение гликоля, а затем поли- меризацию с образованием лаковых соеди- нений, которые остаются растворенными в горячем смазочном материале; при выключе- нии двигателя масло охлаждается, и лаковая пленка проявляется на поверхностях деталей двигателя [28 - 30]. Ключевым элементом являются продукты разложения или окисления гликоля, а не сам гликоль. Гликоль, попав в картер, разлагает- ся до гликолевой, глиоксалевой кислоты, а 92 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 TRANSPORT PART II Рисунок 3 – Последствия нарушения герметичности системы охла Figure 3 – Consequences of a leakage in cooling Том 19, № 1. 2022 Vol. 19, No. 1. 2022 93 ТРАНСПОРТ Л II ТРАНСПОРТ РАЗДЕЛ II ТРАНСПОРТ РАЗДЕЛ II ха к топливу приближается к стехиометриче- скому (14,5 для дизельного топлива), скорость образования сажи резко возрастает. Это свя- зано с тем, что вблизи стехиометрического соотношения в цикле недостаточно времени и кислорода для полного сжигания всего топли- ва; кроме того, будет низкая доля окисляющих веществ для окисления сажи. Как правило, при значениях 20-процентного обеднения топлива стехиометрически и выше, которые использу- ются в настоящее время, в процессе сгора- ния образуется чрезмерное количество сажи. Избыток воздуха необходим для повышения эффективности дизельного цикла и снижения выбросов углеводородов. Исследования пока- зали, что сажа, содержащаяся в смазке дви- гателя, и сажа, выбрасываемая из выхлопной трубы, сильно отличаются. © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Образование сажи в картере двигателя 2022 95 ТРАНСПОРТ РАЗДЕЛ II ТРАНСПОРТ РАЗДЕЛ II Рисунок 5 – Отложения в ловушке коленчатого вала двигателя Figure 5 – Deposits trapped in the engine crankshaft Рисунок 5 – Отложения в ловушке коленчатого вала двигателя Figure 5 – Deposits trapped in the engine crankshaft Образование сажи в картере двигателя Из-за различных дви- жений и нагрузок на каждой границе раздела будут очевидны разные режимы смазки. Это дополнительно усложняется механизмами на- несения смазки, которые варьируются от кон- тактов, где используется положительное сма- зывание, до тех, где смазка достигает контакта косвенно за счет смазки разбрызгиванием. В некоторых случаях контакты мало смазывают- Концентрация образующихся частиц сажи увеличивается с увеличением соотношения воздуха и топлива. Когда соотношение возду- 94 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal TRANSPORT PART II TRANSPORT PART II PART II ся из-за их расположения, и могут возникнуть проблемы с масляным голоданием; присут- ствие сажи еще больше усугубит это. статочно мягкая, но в виде отдельных частиц сажа считается достаточно твердой, чтобы из- нашивать металлические поверхности. у у у Было предложено три различных механиз- ма износа из-за загрязнения сажей. Раундс [39] постулировал, что химическая адсорбция про- тивоизносных компонентов смазки сажей сни- жает способность смазки защищать поверхно- сти. Другие исследователи предположили, что износ сажей мог произойти из-за недостатка смазки в контакте. Это когда сажа агломери- руется до размеров, превышающих толщи- ну масляной пленки, и блокирует попадание смазки в контакт. Последний предложенный механизм предполагает, что изнашивание по- верхностей происходит в результате частично- го истирания, при котором сажа действует как третье тело. В качестве агломератов сажа до- Контроль отложений в двигателе является фундаментальной необходимостью для обе- спечения длительного срока службы и эф- фективности работы двигателя. Образование отложений зависит от конструкции двигателя, условий эксплуатации, технического обслу- живания, типа топлива и сгорания, а также от характеристик масла. Отложения влияют на мощность двигателя и его производитель- ность, износ, шум, плавность, экономичность, срок службы и стоимость обслуживания [40, 41]. Наглядно усугубляющее воздействие за- грязненных моторных масел на детали двига- теля можно продемонстрировать с помощью рисунков 4 и 5. Рисунок 4 – Отложения на поверхностях головок блока цилиндров Figure 4 – Sediments on cylinder head surfaces Рисунок 4 – Отложения на поверхностях головок блока цилиндров Рисунок 4 – Отложения на поверхностях головок блока цилиндров Figure 4 – Sediments on cylinder head surfaces Figure 4 – Sediments on cylinder head surfaces Figure 4 – Sediments on cylinder head surfaces Том 19, № 1. 2022 Vol. 19, No. 1. © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal БИБЛИОГРАФИЧЕСКИЙ СПИСОК 1. Al S. O. A., Salehi F. M., Farooq U., Morina A., Neville A. Chemical and physical assessment of engine oils degradation and additive depletion by soot. Tribology International. 2021. 160. 107054. https://doi. org/10.1016/j.triboint.2021.107054 14. Vrcek A., Hultqvist T., Baubet Y., Björling M., Marklund P., Larsson R. Micro-pitting and wear assessment of engine oils operating under boundary lubrication conditions. Tribology International. 2019. 129: 338-346. https://doi.org/10.1016/j. triboint.2018.08.032. 2. Rostek E., Babiak M. The experimental analysis of engine oil degradation utilizing selected thermoanalytical methods. Transportation Research Procedia. vol. 40. 2019. pp. 82-89. https://doi. org/10.1016/j.trpro.2019.07.014 15. Laad M., Jatti V. K. S. Titanium oxide nanoparticles as additives in engine oil, Journal of King Saud University - Engineering Sciences. 2018. 30: 116-122. https://doi.org/10.1016/j.jksues.2016.01.008. 3. Heredia-Cancino J. A., Ramezani M., Álvarez- Ramos M. E. Effect of degradation on tribological performance of engine lubricants at elevated temperatures. Tribology International. 2018. 124: 230– 237. https://doi.org/10.1016/j.triboint.2018.04.015 p g j j 16. Li D., Kong N., Zhang B., Zhang Bo, Li R., Zhang Q. Comparative study on the effects of oil viscosity on typical coatings for automotive engine components under simulated lubrication conditions. Diamond and Related Materials. 2021. 112: 108226. https://doi.org/10.1016/j.diamond.2020.108226. 4. Mohanty S., Hazra S., Paul S. Intelligent prediction of engine failure through computational image analysis of wear particle. Engineering Failure Analysis. 2020. vol. 116. pp. 104731. https://doi. org/10.1016/j.engfailanal.2020.104731 g j 17. Zhang W., Zhang Z., Ma X., Awad O. I., Shuai Y.Li, S, Xu H. Impact of injector tip deposits on gasoline direct injection engine combustion, fuel economy and emissions. Applied Energy. 2020. 262 : 114538.https:// doi.org/10.1016/j.apenergy.2020.114538. 5. Deulgaonkar V. R., Ingolikar N., Borkar A., Ghute S., Awate N. Failure analysis of diesel engine piston in transport utility vehicles. Engineering Failure Analysis. 2021. 120: 105008, https://doi.org/10.1016/j. engfailanal.2020.105008.l 18. Tormos B., Pla B., Bastidas S., Ramírez L., Pérez T. Fuel economy optimization from the interaction between engine oil and driving conditions. Tribology International. 2019. 138: 263-270. https://doi. org/10.1016/j.triboint.2019.05.042. 6. Notay R. S., Priest M., Fox M. F. The influence of lubricant degradation on measured piston ring film thickness in a fired gasoline reciprocating engine. Tribology International. 2019. 129: 112–123. https://doi. org/10.1016/j.triboint.2018.07.002l g j 19. Slavchov R. I., Salamanca M., Russo D., Salama I., Mosbach S., Clarke S. M., Kraft M., Lapkin A. A., Filip S.V. The role of NO2 and NO in the mechanism of hydrocarbon degradation leading to carbonaceous deposits in engines. Fuel. 2020. 267:117218. https:// doi.org/10.1016/j.fuel.2020.117218. ОБСУЖДЕНИЕ И ЗАКЛЮЧЕНИЕ особенно характерны для дизельных двига- телей, где масло после эксплуатации содер- жит такие продукты глубокой окислительной конверсии, как нагар и смолистые вещества. Изменение температуры масла в картере в пределах всего 80–145 оС уже дает начальную степень окисления. Если объем системы смаз- ки двигателя уменьшается, то концентрация продуктов окисления в масле увеличивается. 1. В работе приведена классификация загрязнений моторного масла дизельных дви- гателей по агрегатному состоянию. р у 2. Рассмотрены пути попадания загряз- нений в моторные масла дизельных двигате- лей. 3. Приведены результаты воздействия загрязнений на состояние дизельных двига- телей и рассмотрены механизмы воздействия жидких загрязнений на детали двигателей. р у у 6. При окислении и старении смазочно- го материала образуются кислые побочные продукты в результате химического разложе- ния базовой основы и присадок в присутствии воздуха и тепла. Высокая концентрация кис- лотных соединений в смазочном материале может привести к коррозии деталей машин из-за загрязненного масла, нарушению рабо- ты фильтров из-за образования лака и шлама. Это приводит к образованию отложений на большинстве поверхностей двигателя, вклю- чая такие, которые могут вызвать залипание поршневых колец в их канавках. 6. При окислении и старении смазочно- го материала образуются кислые побочные продукты в результате химического разложе- ния базовой основы и присадок в присутствии воздуха и тепла. Высокая концентрация кис- лотных соединений в смазочном материале может привести к коррозии деталей машин из-за загрязненного масла, нарушению рабо- ты фильтров из-за образования лака и шлама. Это приводит к образованию отложений на большинстве поверхностей двигателя, вклю- чая такие, которые могут вызвать залипание поршневых колец в их канавках. 4. При возникновении шлама, вода и ох- лаждающая жидкость являются существен- ным фактором, в целом любое состояние, которое способствует проникновению и удер- жанию воды в моторном масле, способствует и образованию шлама. 5. Работавшее масло из двигателя вну- треннего сгорания поршневого типа пред- ставляет собой смесь продуктов сильного окисления углеводородов. Изменения в масле 5. Работавшее масло из двигателя вну- треннего сгорания поршневого типа пред- ставляет собой смесь продуктов сильного окисления углеводородов. Изменения в масле 96 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal TRANSPORT PART II TRANSPORT PART II Flame. 2021. 225: 48-56. https://doi.org/10.1016/j. combustflame.2020.10.047. БИБЛИОГРАФИЧЕСКИЙ СПИСОК 7. Sun Z., Wang Y., Yuan Ch. Influence of oil deposition on the measurement accuracy of a calorimetric flow sensor. Measurement. 2021. 185: 110052. https:// doi.org/10.1016/j.measurement.2021.110052. 20. Burke R.D., Madamedon M., Williams R. Newly identified effects of injector nozzle fouling in diesel engines. Fuel. 2020. 278:118336.https://doi. org/10.1016/j.fuel.2020.118336. 8. Bagi S., Sharma V., Aswath P. B. Role of dispersant on soot-induced wear in Cummins ISB engine test. Carbon. 2018. 136: 395-408.https://doi. org/10.1016/j.carbon.2018.04.066. g j 21. Olabi A.G., Maizak D., Wilberforce T. Review of the regulations and techniques to eliminate toxic emissions from diesel engine cars. Science of The Total Environment. 2020. 748: 141249. https://doi. org/10.1016/j.scitotenv.2020.141249. 9. Raposo H., Farinha J.T, Fonseca I., Galar D..Predicting condition based on oil analysis – A case study. Tribology International. 2019. 135: 65-74.https:// doi.org/10.1016/j.triboint.2019.01.041. 22. Qian Y., Li Z., Yu L., Wang X., Lu X. Review of the state-of-the-art of particulate matter emissions from modern gasoline fueled engines. Applied Energy. 2019. 238: 1269-1298. https://doi.org/10.1016/j. apenergy.2019.01.179. 10. Vaitkunaite G., Espejo C., Wang Ch., Thiébaut B., Charrin C., Neville A., Morina A.. MoS2 tribofilm distribution from low viscosity lubricants and its effect on friction. Tribology International. 2020. 151: 106531. https://doi.org/10.1016/j.triboint.2020.106531. 23. Sujesh G., Ramesh S. Modeling and control of diesel engines: A systematic review. Alexandria Engineering Journal. 2018. 57: 4033-4048.https://doi. org/10.1016/j.aej.2018.02.011. 11. Baskov V., Ignatov A., Polotnyanschikov V. Assessing the influence of operating factors on the properties of engine oil and the environmental safety of internal combustion engine. Transportation Research Procedia. 2020. 50: 37-43. https://doi.org/10.1016/j. trpro.2020.10.005. 24. Kozina A., Radica G., Nižetić S. Analysis of methods towards reduction of harmful pollutants from diesel engines. Journal of Cleaner Production. 2020. 262: 121105. https://doi.org/10.1016/j. jclepro.2020.121105. 12. Esfe M. H., Esfandeh S., Arani A. A. A. Proposing a modified engine oil to reduce cold engine start damages and increase safety in high temperature operating conditions. Powder Technology. 2019. 355: 251-263. https://doi.org/10.1016/j.powtec.2019.07.009. j 25. Ferraro G., Fratini E., Rausa R., Baglioni P. Impact of oil aging and composition on the morphology and structure of diesel soot. Journal of Colloid and Interface Science. 2018. 512: 291-299. https://doi. org/10.1016/j.jcis.2017.10.033. p g j p 13. Wang Y., Chen Yu, Liang X., Tan P., Deng S. Impacts of lubricating oil and its formulations on diesel engine particle characteristics. Combustion and p g j p 13. Wang Y., Chen Yu, Liang X., Tan P., Deng S. БИБЛИОГРАФИЧЕСКИЙ СПИСОК Impacts of lubricating oil and its formulations on diesel engine particle characteristics. Combustion and © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal 97 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 ТРАНСПОРТ РАЗДЕЛ II РАЗДЕЛ II ТРАНСПОРТ 26. Xu H.T., Luo Z.Q., Wang N., Qu Z.G., Chen J., An L. Experimental study of the selective catalytic reduction after-treatment for the exhaust emission of a diesel engine. Applied Thermal Engineering. 2019. 147: 198-204.https://doi.org/10.1016/j. applthermaleng.2018.10.067. из работающих моторных масел без их слива из картеров двигателей машин // Химия и технология топлив и масел. 2020. № 3. С. 18-21. 38. Утаев С.А. Закономерности накопления за- грязняющих примесей моторных масел в процессе эксплуатации двигателей // Современные материа- лы, техника и технологии. 2016. № 2. С. 207-214. pp g 27. Haas F. M., Won S. H., Dryer F. L., Pera C. Lube oil chemistry influences on autoignition as measured in an ignition quality tester. Proceedings of the Combustion Institute. 2019. 37:4645-4654. https:// doi.org/10.1016/j.proci.2018.06.165. 39. Зазуля А.Н., Белов С.А. Изменение свойств моторного масла во время эксплуатации в дизель- ном двигателе // Современные методы технической диагностики и неразрушающего контроля деталей и узлов. 2020. № 1. С. 5-8. g j 28. Kalghatgi G. Is it really the end of internal combustion engines and petroleum in transport? Applied Energy. 2018. 225: 965-974.https://doi. org/10.1016/j.apenergy.2018.05.076. 40. Остриков В.В., Сазонов С.Н., Афоничев Д.Н., Козлов В.Г. Изменение вязкости моторного масла как показатель технического состояния дви- гателя внутреннего сгорания и свойств смазочного материала // Вестник Воронежского государствен- ного аграрного университета. 2019. Т. 12. № 3. С. 54-61. 29. Abián M., Martín C., Nogueras P., Sánchez- Valdepeñas J., Rodríguez-Fernández J., Lapuerta M., Alzueta M. U. Interaction of diesel engine soot with NO2 and O2 at diesel exhaust conditions. Effect of fuel and engine operation mode. Fuel. 2018. 212: 455-461. https://doi.org/10.1016/j.fuel.2017.10.025. 41. Бусин И.В., Остриков В.В., Корнев А.Ю. Технология очистки работающего моторного масла от продуктов старения // Наука в центральной Рос- сии. 2015. № 3 (15). С. 82-87. 30. Deulgaonkar V. R., Pawar K., Kudle P., Raverkar A., Raut A. Failure analysis of fuel pumps used for diesel engines in transport utility vehicles. Engineering Failure Analysis. 2019. 105: 1262-1272. https://doi.org/10.1016/j.engfailanal.2019.07.048. 42. Корнеев С.В., Пашукевич С.В. Влияние воды на изменение показателей качества мотор- ного масла // Вестник Сибирского государственного автомобильно-дорожного университета. 2021. Т. 18. № 4 (80). С. 406-415. 31. БИБЛИОГРАФИЧЕСКИЙ СПИСОК Venkatachalam G., Kumaravel A. Experimental Investigations on the Failure of Diesel Engine Piston. Materials Today: Proceedings. 2019. vol. 16. pp. 1196-1203.https://doi.org/10.1016/j. matpr.2019.05.214. REFERENCES 1. Al S. O. A., Salehi F. M., Farooq U., Morina A., Neville A. Chemical and physical assessment of engine oils degradation and additive depletion by soot. Tribology International. 2021. 160. 107054. https://doi. org/10.1016/j.triboint.2021.107054 p 32. Hu C., You G., Liu J., Du Sh., Zhao X., Wu S. Study on the mechanisms of the lubricating oil antioxidants: Experimental and molecular simulation. Journal of Molecular Liquids. 2021. 324:115099. https://doi.org/10.1016/j.molliq.2020.115099. 2. Rostek E., Babiak M. The experimental analysis of engine oil degradation utilizing selected thermoanalytical methods. Transportation Research Procedia. 40. 2019: 82-89. https://doi.org/10.1016/j. trpro.2019.07.014 33. Agocs A., Nagy A. L., Tabakov Z., Perger J., Rohde-Brandenburger J., Schandl M., Besser Ch., Dörr N. Comprehensive assessment of oil degradation patterns in petrol and diesel engines observed in a field test with passenger cars – Conventional oil analysis and fuel dilution. Tribology International. 2021. 161:107079. https://doi.org/10.1016/j.triboint.2021.107079. 3. Heredia-Cancino J. A., Ramezani M., Álvarez- Ramos M. E. Effect of degradation on tribological performance of engine lubricants at elevated temperatures. Tribology International. 2018. 124: 230– 237. https://doi.org/10.1016/j.triboint.2018.04.015 34. Rossegger B., Eder M., Vareka M., Engelmayer M., Wimmer A. A novel method for lubrication oil consumption measurement for wholistic tribological assessments of internal combustion engines. Tribology International. 2021. 162: 107141. https://doi.org/10.1016/j.triboint.2021.107141. 4. Mohanty S., Hazra S., Paul S. Intelligent prediction of engine failure through computational image analysis of wear particle. Engineering Failure Analysis. 2020. vol. 116. pp. 104731. https://doi. org/10.1016/j.engfailanal.2020.104731 35. Temizer I., Cihan O. Experimental and numerical evaluation of combustion analysis of a DI diesel engine. Energy Reports. 2021. 7: 5549-5561. https://doi.org/10.1016/j.egyr.2021.08.192. 5. Deulgaonkar V. R., Ingolikar N., Borkar A., Ghute S., Awate N. Failure analysis of diesel engine piston in transport utility vehicles. Engineering Failure Analysis. 2021. 120: 105008, https://doi.org/10.1016/j. engfailanal.2020.105008.l g j gy 36. Korneev S.V., Permyakov V.B., Bakulina V.D., Yarmovich Y.V., Pashukevich S.V. Influence of high temperatures on changes in the performance characteristics of motor oils when diluted with fuel. AIP Conference Proceedings: “Oil and Gas Engineering, OGE 2020” 2020: 020010. https://doi. org/10.1063/5.0026994 6. Notay R. S., Priest M., Fox M. F. The influence of lubricant degradation on measured piston ring film thickness in a fired gasoline reciprocating engine. Tribology International. 2019. 129: 112–123. https://doi. org/10.1016/j.triboint.2018.07.002l 37. Остриков В.В., Афоничев Д.Н., Оробинский В.И., Балабанов В.И. Удаление продуктов старения 7. Sun Z., Wang Y., Yuan Ch. REFERENCES Influence of oil deposition on the measurement accuracy of a calorimetric 98 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 TRANSPORT PART II flow sensor. Measurement. 2021. 185: 110052. https:// doi.org/10.1016/j.measurement.2021.110052. flow sensor. Measurement. 2021. 185: 110052. https:// doi.org/10.1016/j.measurement.2021.110052. 20. Burke R.D., Madamedon M., Williams R. Newly identified effects of injector nozzle fouling in diesel engines. Fuel. 2020. 278:118336.https://doi. org/10.1016/j.fuel.2020.118336. 8. Bagi S., Sharma V., Aswath P. B. Role of dispersant on soot-induced wear in Cummins ISB engine test. Carbon. 2018. 136: 395-408.https://doi. org/10.1016/j.carbon.2018.04.066. 21. Olabi A.G., Maizak D., Wilberforce T. Review of the regulations and techniques to eliminate toxic emissions from diesel engine cars. Science of The Total Environment. 2020. 748: 141249. https://doi. org/10.1016/j.scitotenv.2020.141249. 9. Raposo H., Farinha J.T, Fonseca I., Galar D..Predicting condition based on oil analysis – A case study. Tribology International. 2019. 135: 65-74.https:// doi.org/10.1016/j.triboint.2019.01.041. 22. Qian Y., Li Z., Yu L., Wang X., Lu X. Review of the state-of-the-art of particulate matter emissions from modern gasoline fueled engines. Applied Energy. 2019. 238: 1269-1298. https://doi.org/10.1016/j. apenergy.2019.01.179. 10. Vaitkunaite G., Espejo C., Wang Ch., Thiébaut B., Charrin C., Neville A., Morina A.. MoS2 tribofilm distribution from low viscosity lubricants and its effect on friction. Tribology International. 2020. 151: 106531. https://doi.org/10.1016/j.triboint.2020.106531. 23. Sujesh G., Ramesh S. Modeling and control of diesel engines: A systematic review. Alexandria Engineering Journal. 2018. 57: 4033-4048.https://doi. org/10.1016/j.aej.2018.02.011. p g j 11. Baskov V., Ignatov A., Polotnyanschikov V. Assessing the influence of operating factors on the properties of engine oil and the environmental safety of internal combustion engine. Transportation Research Procedia. 2020. 50: 37-43. https://doi.org/10.1016/j. trpro.2020.10.005. g j j 24. Kozina A., Radica G., Nižetić S. Analysis of methods towards reduction of harmful pollutants from diesel engines. Journal of Cleaner Production. 2020. 262: 121105. https://doi.org/10.1016/j. jclepro.2020.121105. 12. Esfe M. H., Esfandeh S., Arani A. A. A. Proposing a modified engine oil to reduce cold engine start damages and increase safety in high temperature operating conditions. Powder Technology. 2019. 355: 251-263. https://doi.org/10.1016/j.powtec.2019.07.009. 25. Ferraro G., Fratini E., Rausa R., Baglioni P. Impact of oil aging and composition on the morphology and structure of diesel soot. Journal of Colloid and Interface Science. 2018. 512: 291-299. https://doi. org/10.1016/j.jcis.2017.10.033. 13. Wang Y., Chen Yu, Liang X., Tan P., Deng S. Impacts of lubricating oil and its formulations on diesel engine particle characteristics. Combustion and Flame. REFERENCES 2021. 225: 48-56. https://doi.org/10.1016/j. combustflame.2020.10.047. g j j 26. Xu H.T., Luo Z.Q., Wang N., Qu Z.G., Chen J., An L. Experimental study of the selective catalytic reduction after-treatment for the exhaust emission of a diesel engine. Applied Thermal Engineering. 2019. 147: 198-204.https://doi.org/10.1016/j. applthermaleng.2018.10.067. l 14. Vrcek A., Hultqvist T., Baubet Y., Björling M., Marklund P., Larsson R. Micro-pitting and wear assessment of engine oils operating under boundary lubrication conditions. Tribology International. 2019. 129: 338-346. https://doi.org/10.1016/j. triboint.2018.08.032. 27. Haas F. M., Won S. H., Dryer F. L., Pera C. Lube oil chemistry influences on autoignition as measured in an ignition quality tester. Proceedings of the Combustion Institute. 2019. 37:4645-4654. https:// doi.org/10.1016/j.proci.2018.06.165. 15. Laad M., Jatti V. K. S. Titanium oxide nanoparticles as additives in engine oil, Journal of King Saud University - Engineering Sciences. 2018. 30: 116-122. https://doi.org/10.1016/j.jksues.2016.01.008. 28. Kalghatgi G. Is it really the end of internal combustion engines and petroleum in transport? Applied Energy. 2018. 225: 965-974.https://doi. org/10.1016/j.apenergy.2018.05.076. 16. Li D., Kong N., Zhang B., Zhang Bo, Li R., Zhang Q. Comparative study on the effects of oil viscosity on typical coatings for automotive engine components under simulated lubrication conditions. Diamond and Related Materials. 2021. 112: 108226. https://doi.org/10.1016/j.diamond.2020.108226. g j gy 29. Abián M., Martín C., Nogueras P., Sánchez- Valdepeñas J., Rodríguez-Fernández J., Lapuerta M., Alzueta M. U. Interaction of diesel engine soot with NO2 and O2 at diesel exhaust conditions. Effect of fuel and engine operation mode. Fuel. 2018. 212: 455-461. https://doi.org/10.1016/j.fuel.2017.10.025. 17. Zhang W., Zhang Z., Ma X., Awad O. I., Shuai Y.Li, S, Xu H. Impact of injector tip deposits on gasoline direct injection engine combustion, fuel economy and emissions. Applied Energy. 2020. 262 : 114538.https:// doi.org/10.1016/j.apenergy.2020.114538. 30. Deulgaonkar V. R., Pawar K., Kudle P., Raverkar A., Raut A. Failure analysis of fuel pumps used for diesel engines in transport utility vehicles. Engineering Failure Analysis. 2019. 105: 1262-1272. https://doi.org/10.1016/j.engfailanal.2019.07.048. g j p gy 18. Tormos B., Pla B., Bastidas S., Ramírez L., Pérez T. Fuel economy optimization from the interaction between engine oil and driving conditions. Tribology International. 2019. 138: 263-270. https://doi. org/10.1016/j.triboint.2019.05.042. 31. Venkatachalam G., Kumaravel A. Experimental Investigations on the Failure of Diesel Engine Piston. Materials Today: Proceedings. 2019. vol. 16. pp. 1196-1203.https://doi.org/10.1016/j. matpr.2019.05.214. 19. Slavchov R. I., Salamanca M., Russo D., Salama I., Mosbach S., Clarke S. M., Kraft M., Lapkin A. A., Filip S.V. REFERENCES The role of NO2 and NO in the mechanism of hydrocarbon degradation leading to carbonaceous deposits in engines. Fuel. 2020. 267:117218. https:// doi.org/10.1016/j.fuel.2020.117218. 32. Hu C., You G., Liu J., Du Sh., Zhao X., Wu S. Study on the mechanisms of the lubricating oil antioxidants: Experimental and molecular simulation. 99 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal Том 19, № 1. 2022 Vol. 19, No. 1. 2022 РАЗДЕЛ II РАЗДЕЛ II ТРАНСПОРТ Journal of Molecular Liquids. 2021. 324:115099. https://doi.org/10.1016/j.molliq.2020.115099. accumulation of contaminants in engine oils during engine operation]. Sovremenny`e materialy`, texnika i texnologii, 2016, 2: 207-214. 33. Agocs A., Nagy A. L., Tabakov Z., Perger J., Rohde-Brandenburger J., Schandl M., Besser Ch., Dörr N. Comprehensive assessment of oil degradation patterns in petrol and diesel engines observed in a field test with passenger cars – Conventional oil analysis and fuel dilution. Tribology International. 2021. 161:107079. https://doi.org/10.1016/j.triboint.2021.107079. 39. Zazulya A.N., Belov S.A. Izmenenie svojstv motornogo masla vo vremya e`kspluatacii v dizel`nom dvigatele [Changes in the properties of engine oil during operation in a diesel engine]. Sovremenny`e metody` texnicheskoj diagnostiki i nerazrushayushhego kontrolya detalej i uzlov, 2020, 1: 5-8. 34. Rossegger B., Eder M., Vareka M., Engelmayer M., Wimmer A. A novel method for lubrication oil consumption measurement for wholistic tribological assessments of internal combustion engines. Tribology International. 2021. 162: 107141. https://doi.org/10.1016/j.triboint.2021.107141. 40. Ostrikov V.V., Sazonov S.N., Afonichev D.N., Kozlov V.G. Izmenenie vyazkosti motornogo masla kak pokazatel` texnicheskogo sostoyaniya dvigatelya vnutrennego sgoraniya i svojstv smazochnogo materiala [Changing the viscosity of engine oil as an indicator of the technical condition of the internal combustion engine and the properties of the lubricant]. Vestnik Voronezhskogo gosudarstvennogo agrarnogo universiteta, 2019, 3: 54-61. 35. Temizer I., Cihan O. Experimental and numerical evaluation of combustion analysis of a DI diesel engine. Energy Reports. 2021. 7: 5549-5561. https://doi.org/10.1016/j.egyr.2021.08.192. 41. Busin I.V., Ostrikov V.V., Kornev A.Yu. Texnologiya ochistki rabotayushhego motornogo masla ot produktov stareniya [Technology for cleaning running engine oil from aging products]. Nauka v central`noj Rossii, 2015, 3: 82-87. 36. Korneev S.V., Permyakov V.B., Bakulina V.D., Yarmovich Y.V., Pashukevich S.V. Influence of high temperatures on changes in the performance characteristics of motor oils when diluted with fuel. AIP Conference Proceedings: “Oil and Gas Engineering, OGE 2020” 2020: 020010. https://doi. org/10.1063/5.0026994 42. ИНФОРМАЦИЯ ОБ АВТОРЕ Пашукевич София Вячеславовна – аспирант. REFERENCES Korneev S.V., Pashukevich S.V. Vliyanie vody` na izmenenie pokazatelej kachestva motornogo masla [The influence of water on the change in engine oil quality indicators]. The Russian Automobile and Highway Industry Journal, 2021, 4: 406-415. g 37. Ostrikov V.V., Afonichev D.N., Orobinskij V.I., Balabanov V.I. Udalenie produktov stareniya iz rabotayushhix motorny`x masel bez ix sliva iz karterov dvigatelej mashin [Removal of aging products from working motor oils without draining them from crankcases of machine engines]. Ximiya i texnologiya topliv i masel, 2020, 3:18-21. INFORMATION ABOUT THE AUTHORS 38. Utaev S.A. Zakonomernosti nakopleniya zagryaznyayushhix primesej motorny`x masel v processe e`kspluatacii dvigatelej [Patterns of Sofia V. Pashukevich – Postgraduate student of the Chemistry and Chemical Technology Department. 100 Том 19, № 1. 2022 Vol. 19, No. 1. 2022 © 2004–2022 Вестник СибАДИ The Russian Automobile and Highway Industry Journal
https://openalex.org/W2160670827	https://biosignaling.biomedcentral.com/counter/pdf/10.1186/s12964-015-0106-x	English	null	Stimulus-dependent differences in signalling regulate epithelial-mesenchymal plasticity and change the effects of drugs in breast cancer cell lines	Cell communication and signaling	2,015	cc-by	17,171	* Correspondence: melissa.davis@unimelb.edu.au †Equal contributors 1Systems Biology Laboratory, Melbourne School of Engineering, University of Melbourne, Building 193, Parkville, VIC 3010, Australia Full list of author information is available at the end of the article Abstract Introduction: The normal process of epithelial mesenchymal transition (EMT) is subverted by carcinoma cells to facilitate metastatic spread. Cancer cells rarely undergo a full conversion to the mesenchymal phenotype, and instead adopt positions along the epithelial-mesenchymal axis, a propensity we refer to as epithelial mesenchymal plasticity (EMP). EMP is associated with increased risk of metastasis in breast cancer and consequent poor prognosis. Drivers towards the mesenchymal state in malignant cells include growth factor stimulation or exposure to hypoxic conditions. Methods: We have examined EMP in two cell line models of breast cancer: the PMC42 system (PMC42-ET and PMC42-LA sublines) and MDA-MB-468 cells. Transition to a mesenchymal phenotype was induced across all three cell lines using epidermal growth factor (EGF) stimulation, and in MDA-MB-468 cells by hypoxia. We used RNA sequencing to identify gene expression changes that occur as cells transition to a more-mesenchymal phenotype, and identified the cell signalling pathways regulated across these experimental systems. We then used inhibitors to modulate signalling through these pathways, verifying the conclusions of our transcriptomic analysis. Results: We found that EGF and hypoxia both drive MDA-MB-468 cells to phenotypically similar mesenchymal states. Comparing the transcriptional response to EGF and hypoxia, we have identified differences in the cellular signalling pathways that mediate, and are influenced by, EMT. Significant differences were observed for a number of important cellular signalling components previously implicated in EMT, such as HBEGF and VEGFA. cellular signalling components previously implicated in EMT, such as HBEGF and VEGFA. We have shown that EGF- and hypoxia-induced transitions respond differently to treatment with chemical inhibitors (presented individually and in combinations) in these breast cancer cells. Unexpectedly, MDA-MB-468 cells grown under hypoxic growth conditions became even more mesenchymal following exposure to certain kinase inhibitors that prevent growth-factor induced EMT, including the mTOR inhibitor everolimus and the AKT1/2/3 inhibitor AZD5363. Conclusions: While resulting in a common phenotype, EGF and hypoxia induced subtly different signalling systems in breast cancer cells. Our findings have important implications for the use of kinase inhibitor-based therapeutic interventions i b t h th h t i lli l d ill i fl th th ti We have shown that EGF- and hypoxia-induced transitions respond differently to treatment with chemical inhibitors (presented individually and in combinations) in these breast cancer cells. Cursons et al. Cell Communication and Signaling (2015) 13:26 DOI 10.1186/s12964-015-0106-x Cursons et al. Cell Communication and Signaling (2015) 13:26 DOI 10.1186/s12964-015-0106-x RESEARCH ARTICLE Open Access Stimulus-dependent differences in signalling regulate epithelial-mesenchymal plasticity and change the effects of drugs in breast cancer cell lines Joseph Cursons1,2†, Karl-Johan Leuchowius3,4†, Mark Waltham5, Eva Tomaskovic-Crook5, Momeneh Foroutan1, Cameron P Bracken6,7, Andrew Redfern8, Edmund J Crampin1,2,9,10, Ian Street3,4†, Melissa J Davis1† and Erik W Thompson5,11,12† © 2015 Cursons et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Introduction Epithelial mesenchymal transition (EMT) is the direc- tional process where sessile, polarised epithelial cells alter the expression of key adhesion and regulatory molecules and gain the ability to survive and migrate as single cells. EMT is a normal process that occurs early in development to generate the primary mesenchyme, and later in the ectoderm to form muscle, bone, nerve and connective tissues [1]. In development, EMT is transient, and the phenotypic shift is followed by the re- verse transition (MET) at the target site [1,2]. Metastasis is now recognized to have many elements in common with developmental EMT, such as single cell dispersal, increased migratory and invasive potential, and gene expression changes [2-5]. When these transitions occur in cancer, however, a hybrid/metastable phenotype is reached after the carcinoma undergoes a subtle EMT, rather than full mesenchymal conversion [6-8]. We use the term epithelial mesenchymal plasticity (EMP) for phenotypic flux of cancer cells along the EMT-MET axis, as they shift between organized, polarized, sessile epithe- lial cells and more individual and motile mesenchymal cells, facilitating metastatic spread [5,6,9,10]. Specific support for the importance of EMP in breast cancer (BrCa) pathogenesis comes from the observations that BrCa stem cells (BCSC) exhibit a mesenchymal phenotype [5,11-13]. BCSC exhibit dramatically en- hanced malignant/metastatic properties compared to their non-BCSC counterparts, and can regenerate a het- erogeneous tumour cell population [14,15]. They overex- press CD44, have low expression of the luminal marker CD24 (CD44hiCD24lo/-), and have a transcription profile resembling EMT-transformed cells [13,16]. Basal sub- types of BrCa, which have a poor prognosis, exhibit in- creased EMT marker expression [17]. The links between EMT, BCSC, and basal breast cancer therefore place EMP at the mechanistic core of the most malignant cells found in clinical BrCa. Further to this, in breast cancer patients EMT correlates with adverse prognosis. An EMT signature was found to predict delayed relapse using available on-line data in 4767 breast cancer tumour samples [18]. In multiple studies, poor patient outcomes have been shown to be correlated with the altered expres- sion of various protein markers of EMT development, including increased vimentin [19], loss of certain epithelial cytokeratins [20], loss of E-cadherin and gain of N-cadherin [21]. Additionally, EMT can be induced in patient breast cancers in response to standard chemotherapies [22] and hormonal therapies [23], suggesting a potential role for EMT in treatment resistance. Abstract Unexpectedly, MDA-MB-468 cells grown under hypoxic growth conditions became even more mesenchymal following exposure to certain kinase inhibitors that prevent growth-factor induced EMT, including the mTOR inhibitor everolimus and the AKT1/2/3 inhibitor AZD5363. Conclusions: While resulting in a common phenotype, EGF and hypoxia induced subtly different signalling systems in breast cancer cells. Our findings have important implications for the use of kinase inhibitor-based therapeutic interventions in breast cancers, where these heterogeneous signalling landscapes will influence the therapeutic response. Keywords: Epithelial mesenchymal plasticity, EMT, Metastasis, Breast cancer, EGF, Hypoxia, MEK, AKT, MDA-MB-468 Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 2 of 21 the importance of these pathways in treatment outcome, work by a number of groups has shown that over- expression of SNAI1/2, or TWIST1 in breast cancer cells results in both EMT and chemoresistance [26-28]. The activity of these transcription factors is controlled through a number of signalling pathways that sense changes to the cellular environment and initiate cascades of signalling that result in transcriptional activation or repression. The stim- uli that trigger these regulators to induce EMT vary. Signal- ling through EGFRs is a well-established driver of breast cancer progression [29,30], and EGF is also known to stimulate EMT in some cells [3,31-35]. Hypoxia has been shown to induce EMT through HIF1a activation of TWIST in a variety of cell lines [36,37], and through SNAI1 in hepatocellular carcinoma [38]. Furthermore, dysregulated signalling through pathways such p38 MAPK [39] and PI3K-Akt [28,40] has been implicated in EMP regulation. Because such signalling pathways are often druggable, they represent important targets for novel therapeutics. For ex- ample, considerable interest has been generated in recent years by classes of kinase inhibitors that are able to modu- late cellular signalling and interrupt oncogenic signalling. This motivates the question: if multiple stimuli are able to trigger the more aggressive mesenchymal phenotype in can- cer cells, do the responses to these stimuli converge upon common signalling elements, or do they achieve similar phenotypic outcomes through distinctly different molecular pathways? The answer to this question has clear implica- tions for the design of molecular targeted therapies, as well as for managing the fundamental heterogeneity of breast cancer. Introduction MDA-MB-468 xenografts exhibit distinct zones of mesen- chymal transition, one at the stromal periphery and the other at the interface with the central necrosis common in this xenograft model [48,49]. These models provided an opportunity to investigate both differential lineage-specific cellular responses to the same EMT-promoting stimulus as well as differential responses to varied EMT promoting stimuli in the same cell line. By observing the transcriptional changes in PMC42 cells and in MDA-MB-468 cells under different stimuli, we were able to identify patterns of disruption that are distinct to each stimulus as well as common to all. These observations have clear implications for the therapeutic benefit of pharmaceutical manipulation of these pathways during cancer progression. We tested this notion using drugs targeting these key pathways and demonstrated clear differences in the extent to which different drugs are able to block mesenchymal transition induced by different triggers. The divergence of signalling for EMP regulation between EGF and hypoxia that we characterise here is of therapeutic importance, particularly bearing in mind associations in breast cancer patients between EMT, poor prognosis and treatment resistance. The plates were imaged on a Perkin Elmer Operetta using a 10X/0.4 NA air objective and the appropriate excitation and emission filters. Excitation time and in- tensity were set to avoid overexposed pixels in the mea- sured images. Acquired images were analysed using Perkin Elmer Harmony 3.5 software. Cell nuclei were segmented using the Hoechst images, and the nuclear masks were expanded to cover the cytoplasm of the cells. The mean and total fluorescence intensities of vimentin and phospho-ERK were measured within the masked areas, using image data from the corresponding fluorescently labelled secondary antibodies. Cells were then classified as vimentin+or vimentin−using a decision tree classifier. The decision tree classifier used the mean and total vimentin intensities of the cells to determine thresholds that maximised the separation of the cell populations between unstimulated wells, and stimulated- cell control wells (16 positive and 16 negative control wells were included for each plate). The number of cells imaged, the percent vimentin+ cells and the average phospho-ERK intensity per cell were then calculated for each well. The percent inhibition was calculated as 100(1-(x-mean(xpos))/(mean(xneg) – mean(xpos)), where x indicates the measured variable and mean(x) indicates the mean of the measured variable x among the positive or negative control wells. Introduction We have employed two human BrCa cell line models of stable (PMC42) and dynamically induced (MDA MB 468) EMP. PMC42-LA is an epithelial subline derived from the vimentin+, E-Cadherin−parental PMC42-ET cells [41,42]. PMC42-LA cells demonstrate heterogeneity where ap- proximately 90% of the cells are E-cadherin+ while the remaining 10% lack E-cadherin and are vimentin+. PMC42-LA cells undergo a marked EGF-induced EMT [3,42,43]. PMC42-ET cells also undergo EGF-driven EMT, however the change in expression of mesenchymal markers is reduced due to higher basal expression of markers such as vimentin [3,42]. When compared to other BrCa cell lines [44], both PMC42-ET cells and the LA sub- line exhibit a “Basal B” (mesenchymal) transcriptome, clustering together despite their EMP differences, and irrespective of EGF treatment (E Tomaskovic-Crook & T Blick, unpublished observation). MDA-MB-468 cells have a “Basal A” transcriptome [44] indicating mixed lu- minal/basal attributes, and although predominantly epithe- lial and E-cadherin+, they lack α-catenin and tight junction protein 1 (TJP1, or ZO-1). About 5% of these cells are vimentin+ and they display intermediate invasiveness [45]. MDA-MB-468 cells also exhibit a marked EGF-regulated EMT is known to be controlled by a set of transcription factors including SNAI1/2, ZEB1/2, and other basic helix- loop-helix factors, which coordinate programs of gene ex- pression during EMT (reviewed in [24,25]). Demonstrating Page 3 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 incubated overnight at 4°C with mouse anti-vimentin antibody (V6630, Sigma-Aldrich), diluted 1:1600 in TBS with 1% BSA (Sigma-Aldrich) and 0.05% Tween-20 (Sigma-Aldrich). Where indicated, a rabbit anti-phospho- ERK1/2 antibody (#4370, Cell Signaling Technology) was also included at a dilution of 1:200. The cells were washed three times for 5 min with washing buffer (1xTBS with 0.05% Tween-20) then stained with an Alexa-594 conju- gated goat-anti mouse antibody (115-585-146, Jackson ImmunoResearch) diluted 1:200 in TBS with 1% BSA and 0.05% Tween-20. To stain the nuclei of the cells, Hoechst 33342 (Sigma-Aldrich) was included at a concentration of 3 μg.mL−1. When the pERK1/2 primary antibody was included, an Alexa-488 conjugated goat-anti rabbit antibody (#4412, Cell Signaling Technology) diluted 1:1000 was included in the secondary antibody mix. The cells were stained for 2 hours then washed 3 × 5 minutes with washing buffer. EMT, as well as hypoxia-driven EMT (2% O2) [42,46,47]. Immunofluorescence staining of cells g PMC42-LA and PMC42-ET cells were cultured in RPMI 1640 medium with 10% FBS. MDA-MB-468 cells were cultured in DMEM with 10% FBS. The cell lines had all tested negative for mycoplasma infection. Cells were seeded in 384 well flat bottom microtiter plates (#3712, Corning Life Sciences) and allowed to adhere overnight at 37°C/5% CO2. The next day, where indicated in the text, human EGF (#8916, Cell Signaling Technology) was added to the cells at a final concentration of 10 ng/mL. After 72 h the cells were fixed with 3.7% formaldehyde in PBS for 10 minutes, then washed with Tris-buffered saline (TBS). The cells were incubated with a blocking solution of 0.3% Triton X-100 (Sigma-Aldrich) and 5% sterile filtered goat serum (Sigma-Aldrich). Next, cells were Materials and methods Cell Lines PMC42-ET (ET) cells were derived from a breast cancer pleural effusion by Dr. Robert Whitehead, Ludwig Institute for Cancer Research, Melbourne, Australia, with appropri- ate institutional ethics clearance (Institutional Review Board of the Peter MacCallum Hospital, Melbourne) and informed consent of the patient. The PMC42-LA (LA) sub- line was derived further from the parental PMC42- ET cells by Dr. Leigh Ackland, Deakin University, Melbourne, Australia [50-53]. MDA-MB-468 cells originally from the ATCC were transferred from the Lombardi Cancer Center, Washington, DC, USA. Introduction The dose response curves of the inhibitors tested for the different cell lines and measured variables were fitted and plotted in GraphPad Prism 6. Cooperativity between inhibitors tested in combinations were calculated according to the Median effects method [54] using the CalcuSyn software (Biosoft). Pathway analysis Kyoto Encyclopedia for Genes and Genomes (KEGG) maps were downloaded and gene lists were extracted from asso- ciated KGML files. For the over-representation analysis (ORA), all maps annotated as signalling pathways or sys- tems were included. The expected and observed numbers of differentially expressed mRNA transcripts (q-value < 0.01) were compared using a χ2-test within each condition comparison as outlined above. A Bonferoni correction was applied, such that the estimated p-values were multiplied by 22 (the number of signalling pathways tested). Examining changes in transcript abundance that oc- curred with the phenotypic EMT (Figure 1a-h) consistent differences were observed for several transcripts that con- tribute to the canonical mesenchymal phenotype (Figure 1 & j). Transcripts for VIM were significantly increased in all models of induced EMT, including EGF-stimulated PMC42-ET cells, while several other regulatory/signalling components implicated in EMT [13,59-61] (further details in Additional file 1: Table S1) showed consistent changes with various degrees of significance (Figure 1i). A num- ber of transcripts encoding cellular signalling compo- nents implicated in EMT also showed large changes between some of the experimental models as detailed in the sections below. Kinase inhibitor treatment of cells Kinase inhibitors were purchased from Selleck Chemicals and were diluted in DMSO then added to the cells at the concentrations indicated within each figure. For de- termining IC50 values across the range of kinase inhibitors, Page 4 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 signalling pathways. The Bionet R package [55] was used to compute the top-scoring network in each experiment. First we downloaded the set of human protein interactions provided by the PINA2 website [56] and extracted the network corresponding to proteins encoded by transcripts measured in our MDA-MB-468 experiments (transcripts for which we have no data were excluded from the network). Both the network and the p values from the differential expression analysis were passed to the Bionet package, and we used the runfastheinz function to gener- ate the top scoring network for EGF and hypoxia-induced EMP. Networks were exported in the simple interaction format (.sif) for analysis in Cytoscape 3.1 [57]. compounds were serially diluted 1:2 to produce eleven con- centrations, with the highest concentration being 10 μM. The final concentration of DMSO in the wells was 0.5%. Positive and negative control wells were included for each plate where the cells were treated with 1 uM of Erlotinib (Selleck Chemicals) or 0.5% DMSO, respectively. For the hypoxia-treated cells, cells grown under normoxic conditions were used as negative controls. The cells were grown in a humidified atmosphere at 37°C/5% CO2 for 72 h. Hypoxia- treated cells were grown in a hypoxic chamber at 37°C/5% CO2 with 1% O2 for 72 h. Transcript abundance data A detailed description of the transcriptome analysis is given elsewhere (Tomaskovic-Crook E, Philip G, Blick T, van Denderen BJW, Haviv I, Thompson EW: RNA Se- quencing of Induced Epithelial-Mesenchymal Transition in Breast Cancer Cell Lines, in preparation). Briefly, an epithelial-to-mesenchymal transition was induced for PMC42-LA and -ETcell lines through EGF stimulation, and in MDA-MB-468 cells through EGF stimulation or growth under hypoxic conditions. All treatments were applied for 72 h, then RNA was collected from unstimulated and stimulated cells, and mRNA abundances were measured using RNASeq. The two resulting networks, based on differential ex- pression induced by EGF and Hypoxia, were merged using the Advanced Network Merge functions in Cytos- cape, which we also used to calculate node degree. Data on drugs or compounds and their known targets (and where available their mechanism of action) was down- loaded from the Drugbank Database (v3.0) [58] and mapped onto our network using the gene name attri- butes associated with both drug targets and proteins. These data were used to identify druggable proteins within each network. Nodes were then ranked by their degree within this resulting network and druggable nodes selected for further analysis. Transcript abundance data were compared between un- stimulated and stimulated conditions described above: PMC42- ET −/+EGF, PMC42-LA −/+EGF, MDA-MB-468 −/+EGF, MDA-MB-468 −/+HPX. Three ‘inter-model’ comparisons were also performed, between: PMC-42-ET versus PMC-42-LA with and without EGF, and MDA-MB-468 HPX versus EGF stimula- tion. These comparisons are arranged such that they are consistent with a ‘general EMT’ process, as classified by vimen- tin (VIM) up. Sequence alignment was performed using TopHat and differential analysis was performed using CuffDiff. Transcript abundances and test statistics were imported into the MATLAB scripting language (R2012b) for subsequent analysis and to generate heat map plots. Induced epithelial-to-mesenchymal transitions promote a similar cellular phenotype, but act through cell-line and stimulus-specific signalling mechanisms The stimulation of PMC42-LA and MDA-MB-468 cells with EGF, or growth of MDA-MB-468 cells under hypoxic conditions (HPX) each promoted EMT as indicated by an increased proportion of vimentin + cells (red fluorescence; Figure 1b & f, c & g, d & h). Unstimulated PMC42-ET cells express vimentin (Figure 1a), thus increases in the number of vimentin+ cells with EGF stimulation (Figure 1e) are relatively small, consistent with our previous reports on EMT within this cellular system [40,41]. Druggable target and protein-protein interaction networks Protein interaction networks provide a wider coverage of molecular interactions than are captured by canonical Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 5 of 21 Figure 1 Stimulation of PMC42-ET and PMC42-LA cells with EGF, or stimulation of MDA-MB-468 cells with EGF or growth under hypoxic conditions (HPX) promotes a mesenchymal phenotype. (a-h) Fluorescence images of stimulated and unstimulated cells labelled with DAPI (blue) and anti-vimentin (red). Scale bar represents 10 μm. Changes in mRNA transcript abundance between stimulated and unstimulated cells for (i) EMT markers and (j) EMT-implicated transcription factors. Note the use of alternative colour-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Grey squares indicate mRNA transcripts that were not reliably detected – normalised count data are shown in Additional file 2: Figure S1. Figure 1 Stimulation of PMC42-ET and PMC42-LA cells with EGF, or stimulation of MDA-MB-468 cells with EGF or growth under hypoxic conditions (HPX) promotes a mesenchymal phenotype. (a-h) Fluorescence images of stimulated and unstimulated cells labelled with DAPI (blue) and anti-vimentin (red). Scale bar represents 10 μm. Changes in mRNA transcript abundance between stimulated and unstimulated cells for (i) EMT markers and (j) EMT-implicated transcription factors. Note the use of alternative colour-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Grey squares indicate mRNA transcripts that were not reliably detected – normalised count data are shown in Additional file 2: Figure S1. of induced EMT, we first assessed the mRNA transcripts that responded within each model and then mapped these to KEGG pathways. EGF stimulation of PMC42- ET cells led to significant (q-value < 0.05) changes in abundance for 238 transcripts (Table 1). This was the lowest number across all of our in vitro models of EMT, consistent with PMC42-ET cells being relatively mesen- chymal in the unstimulated state (Figure 1a). The EGF- and HPX-stimulated MDA-MB-468 cells had significant changes in abundance for 3155 and 3716 transcripts, respectively, indicating a much greater response than the EGF- stimulated PMC42-ET or PMC42-LA cells (Table 1). The number of transcripts with differential abundance for the stimulated MDA-MB-468 cells was of a similar magni- tude to the inter-model comparisons between PMC42-ET and -LA sublines in the presence or absence of EGF (3261 and 2938, respectively; Table 1). Druggable target and protein-protein interaction networks These inter-model com- parisons also show that the number of transcripts with a significantly different abundance between the PMC42-ET and -LA sublines decreased with EGF stimulation, suggesting a potential convergence of phenotypes. The mRNAs of transcription factors (TFs) implicated in EMT was also examined and only FOSL1 (also known as FRA1) showed significant increases in transcript abundance across all models of induced EMT (Figure 1j). TFs known to play a role in EMT including ETS1, SOX9 and ZEB1 showed consistent increases in transcript abundance, while FOXO4, KLF8 and the epithelial TF GRHL2 were consistently reduced; however, not all of these changes were statistically significant (Figure 1j). Conversely, several well-studied TFs which drive EMT, such as SNAI1 and TWIST1, showed vastly different expression profiles between differing cell lines and differ- ing stimuli, while ZEB2 and SNAI2 were not reliably de- tected within the MDA-MB-468 cells, nor were FOXC2 and GSC detected across all cell lines tested (Figure 1j). Furthermore, normalised count data suggest that the mammary stem cell TF SOX9 was much more abun- dantly expressed in the MDA-MB-468 cells, while TWIST1 and ZEB1 had much higher transcript counts in the PMC42 sublines (Additional file 2: Figure S1). These results indicate that a phenotypically-similar EMT process was induced across these different cell lines and stimuli, with consistent changes in the tran- scripts which mediate these canonical changes, such as VIM, CD44, CDH1 and CDH2. However, variation in the differential abundance patterns observed for specific EMT-implicated TFs suggests that these similar pheno- typic behaviours are associated with different regulatory mechanisms. Next we examined the putative signalling effects of these altered transcript abundances, performing an over- representation analysis to identify intracellular signalling pathways that may have been perturbed (p-value < 0.05) by concerted changes to numerous signalling compo- nents during induced EMT. Eleven signalling pathways showed some evidence of dysregulation (p < 0.05) within at least one model of induced EMT (Table 1). The PI3K- Akt signalling pathway was the only pathway that showed perturbation of components across all models of induced EMT (Table 1); however, after further correcting for multiple hypothesis testing, the EGF-stimulated PMC42-LA cells remained as the only experimental Pathway analysis highlights alternative signalling mechanisms which contribute to EMT Strong dysregulation of Hippo, Hedgehog and TGF-beta signalling components was observed with EGF induced EMT within the PMC42-LA cells, and in the absence of EGF, components of the Wnt signalling path- way showed strong differences in transcript abundance between the PMC42-ET and PMC42-LA sublines (Table 1). Pathway analysis highlights alternative signalling mechanisms which contribute to EMT To identify signalling pathways likely to be affected by the transcriptional changes associated with each model Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 6 of 21 Table 1 Different signalling pathways are dysregulated between the models of in vitro induced EMT differential analysis between: Cell Lines PMC42- ET PMC42- LA MDA-MB- 468 MDA-MB- 468 PMC42-ET vs. PMC42-LA PMC42-ET vs. PMC42-LA MDA-MB- 468 Conditions +EGF vs. Unstim. +EGF vs. Unstim. +EGF vs. Unstim. +HPX vs. Unstim. Unstim. +EGF +HPX vs +EGF number of differentially expressed transcripts 238 591 3155 3716 3261 2938 1626 KEGG signalling pathway/ system: PI3K-Akt 0.015 0.000 0.030 0.005 0.153 0.559 0.943 HIF-1 0.952 0.244 0.002 0.002 0.982 0.732 0.000 Rap1 0.562 0.014 0.000 0.000 0.175 0.075 0.742 Hippo 0.173 0.000 0.022 0.085 0.004 0.297 0.996 Wnt 0.015 0.371 0.038 0.071 0.002 0.455 0.534 MAPK 0.214 0.044 0.312 0.680 0.360 0.634 0.768 Hedgehog 0.977 0.000 0.999 0.939 0.217 0.093 0.993 TGF-beta 0.029 0.000 0.215 0.328 0.053 0.632 0.694 Ras 1.000 0.169 0.259 0.022 0.120 0.065 0.783 Phosphatidyl- inositol 0.796 0.833 0.238 0.049 0.753 0.923 0.893 cAMP 0.992 0.639 0.242 0.017 0.983 0.756 0.201 The first row shows the number of mRNA transcripts with a significant (q-value < 0.05) difference in abundance between the specified cell lines or conditions. Subsequent rows show the estimated p-value for enrichment of elements within KEGG pathways without correction for multiple hypothesis testing. KEGG maps related to signal transduction with a significant (p-value < 0.05) enrichment in at least one comparison are shown (for a complete list please refer to Additional file 8: Table S2). Values in bold are statistically-significant following a Bonferroni correction for multiple hypothesis testing (adjusted p-value < 0.05; n = 22 ‘signalling pathway’ KEGG maps), p-value entries greater than 0.10 are in grey. ble 1 Different signalling pathways are dysregulated between the models of in vitro induced EMT ted between the models of in vitro induced EMT The first row shows the number of mRNA transcripts with a significant (q-value < 0.05) difference in abundance between the specified cell lines or conditions. Subsequent rows show the estimated p-value for enrichment of elements within KEGG pathways without correction for multiple hypothesis testing. KEGG maps related to signal transduction with a significant (p-value < 0.05) enrichment in at least one comparison are shown (for a complete list please refer to Additional file 8: Table S2). Pathway analysis highlights alternative signalling mechanisms which contribute to EMT Values in bold are statistically-significant following a Bonferroni correction for multiple hypothesis testing (adjusted p-value < 0.05; n = 22 ‘signalling pathway’ KEGG maps), p-value entries greater than 0.10 are in grey. MAP2K1, MAP2K2, PIK3CA, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R3, PIK3R5, PRKCB, PRKCG, and RAC1 (Figure 2a). The prevalence of MEK1/2-ERK1/2 and PI3K-Akt across these KEGG maps likely reflects the role of these signal transducers in the integration of numerous up- stream signals. MAP2K1, MAP2K2, PIK3CA, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R3, PIK3R5, PRKCB, PRKCG, and RAC1 (Figure 2a). The prevalence of MEK1/2-ERK1/2 and PI3K-Akt across these KEGG maps likely reflects the role of these signal transducers in the integration of numerous up- stream signals. system showing significant transcriptional dysregulation of PI3K-Akt signalling components. The results shown in Table 1 support the observation that although a phenotypically-similar EMT is induced (Figure 1e-h & 1i), as extensively characterised in previous reports by us and others [3,41-43,46-48], there are differences in the molecular mechanisms that drive these phenotypic changes (Figure 1j). p yp g g j Both the HIF-1 signaling pathway and Rap1 signaling pathway showed very strong transcriptional perturbations within EGF or HPX-stimulated MDA-MB-468 cells, and there was also evidence of HIF-1 signaling pathway dysreg- ulation between EGF and HPX-stimulated MDA-MB-468 cells (Table 1). Strong dysregulation of Hippo, Hedgehog and TGF-beta signalling components was observed with EGF induced EMT within the PMC42-LA cells, and in the absence of EGF, components of the Wnt signalling path- way showed strong differences in transcript abundance between the PMC42-ET and PMC42-LA sublines (Table 1). To identify common signalling elements across these different pathways we examined the frequency of compo- nents. Changes in mRNA transcript abundance of signal- ling proteins which were present within at least six of the 11 KEGG maps are shown in Figure 2a. Three proteins were found across seven pathways, encoded by: MAPK1, MAPK3 and PRKCA (Figure 2a; see membership matrix at right). Within six of the maps, the next most com- mon proteins were encoded by: AKT1, AKT2, AKT3, Both the HIF-1 signaling pathway and Rap1 signaling pathway showed very strong transcriptional perturbations within EGF or HPX-stimulated MDA-MB-468 cells, and there was also evidence of HIF-1 signaling pathway dysreg- ulation between EGF and HPX-stimulated MDA-MB-468 cells (Table 1). Systems-level computational analysis identified putative drug targets to alleviate signalling pathway dysregulation that occurs with induced EMT Note the use of alternative colour-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. signal transduction cascades also appear to be key inte- grators across all of the dysregulated signalling pathways (Table 1; Figure 2a). Together with further details below, and the results of our pathway analysis, this motivated our experimental screening to focus upon different classes of kinase inhibitors targeting PI3K/mTOR, AKT and MEK1/2 as indicated (Figure 3). abundance data were mapped onto an experimentally verified protein-protein interaction network and signalling components that could be targeted by drugs, inhibitors or antagonists were ranked by the relative level of dysregula- tion to their local interactome (Table 2). These results are discussed below with a schematic diagram showing the role of proteins and functional re- lationships between signalling components, within the context of a broader intracellular signalling network (Figure 3). These results were used to motivate pharma- cological targeting of several points within the dysregu- lated intracellular signalling network to examine the efficacy of blocking EMT, as indicated within Figure 3. Some of the strongest transcriptional changes with induced EMT were observed for integrin subunits and corresponding ECM components (Figure 3), and these changes would be expected to influence the formation and regulation of focal adhesion sites. Members of the Src kinase family play an important role in transducing signals from focal adhesion sites [66] to regulate Ras signalling, and the interactomes of both LYN and FYN are relatively enriched for disrupted binding partners, as is the homolog ABL1 (Table 2). Induction of FYN by PI3K/Akt signalling has previously been implicated as a key mediator of cell invasiveness [40]. LYN has also been identified as an important driver of phospho-tyrosine sig- nalling to induce invasiveness within basal subtype breast cancers, although that study reported a relatively low level of activated LYN within the MDA-MB-468 cell line [67]. Given the use of EGF within our experimental models of in vitro induced EMT (Figure 1), kinase inhibitors against EGFR/HER were included as a positive control. The local interaction neighbourhoods of EGFR and ERBB2 were amongst the most dysregulated (Table 2); however, this may reflect the numerous feedback mecha- nisms which have previously been elucidated for EGFR signalling [62-64]. Systems-level computational analysis identified putative drug targets to alleviate signalling pathway dysregulation that occurs with induced EMT As described earlier, EGF stimulation and hypoxic tumour environments are both thought to be clinically-relevant drivers of breast cancer progression in vivo. Thus, we focussed our analysis towards elucidating the convergent and divergent alterations to intracellular signalling which may encompass therapeutic targets for controlling EMT within MDA-MB- 468 cells as a model of triple-negative breast cancer (TNBC). To motivate drug target selection several analyses were per- formed and their results are described together below. First, transcripts showing similar or divergent patterns of differential expression between the EGF- and HPX-stimulated MDA- MB-468 cells were extracted (Additional file 3: Figure S2a & b, respectively). Components of the dysregulated signalling pathways (Table 1) are shown in Figure 2b & c. Transcripts that changed in the same direction across all models of induced EMT (including known EMT markers) are also shown in Additional file 4: Figure S3. Next, transcript To identify common signalling elements across these different pathways we examined the frequency of compo- nents. Changes in mRNA transcript abundance of signal- ling proteins which were present within at least six of the 11 KEGG maps are shown in Figure 2a. Three proteins were found across seven pathways, encoded by: MAPK1, MAPK3 and PRKCA (Figure 2a; see membership matrix at right). Within six of the maps, the next most com- mon proteins were encoded by: AKT1, AKT2, AKT3, Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 7 of 21 Figure 2 (See legend on next page.) Cursons et al. Cell Communication and Signaling (2015) 13:26 Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 8 of 21 (See figure on previous page.) Figure 2 Numerous signalling components showed significant differences between EGF and HPX mediated EMT. Heat maps for: (a) mRNA transcripts for signalling components which are present across at least six perturbed signalling pathway KEGG maps (Table 1); (b, c) mRNA transcripts with significant (q-value < 0.05) differences in mRNA transcript abundance within at least one PMC42 cell line condition comparison, and differences in mRNA transcript abundance going in (b) the same, or (c) different directions for EGF or HPX-stimulated MDA-MB-468 cells compared to unstimulated, with a significant difference in transcript abundance between the EGF- and HPX-stimulated MDA-MB-468 cells. Membership within KEGG maps that are listed in Table 1 is shown at right (black box). mRNA transcripts encoding proteins for which there are drugs, inhibitors or antagonists available (through DrugBank v3.0), ranked by degree within a protein-protein interaction network of differentially expressed transcripts. Degree reflects the number of interaction partners (for the encoded protein) which show significant changes in transcript abundance. mRNA transcripts encoding proteins for which there are drugs, inhibitors or antagonists available (through DrugBank v3.0), ranked by degree within a protein-protein interaction network of differentially expressed transcripts. Degree reflects the number of interaction partners (for the encoded protein) which show significant changes in transcript abundance. y differentially expressed transcripts. Degree reflects the number of interaction partners (for the encoded protein) which abundance Systems-level computational analysis identified putative drug targets to alleviate signalling pathway dysregulation that occurs with induced EMT Alternative ligands for EGFR (TGFA, AREG and HBEGF) show significant changes in tran- script abundance, suggesting that autocrine/paracrine signalling mechanisms may be activated, with HBEGF showing particularly strong differences between EGF and HPX stimulation (Figure 3). Pharmacological modulation of PI3K/mTOR was par- ticularly attractive for this model of in vitro induced EMT, as MDA-MB-468 cells are known to carry an inactivating mutation in the PIP3-phosphatase PTEN [68]. Regulatory Class IA PI3K subunits stabilise the Activation of EGFR is known to drive signalling through both PI3K-Akt and MEK-ERK [65], and these Table 2 Signalling pathway components showed variable levels of transcriptional disruption to their local interactome Rank HGNC symbol degree Rank HGNC symbol degree Rank HGNC symbol degree 1 HSP90AA1 164 11 PIK3R1 66 21 ERBB2 42 2 HSP90AB1 132 12 VCP 59 22 TGFBR1 41 3 EGFR 122 13 CALM1 58 23 HSPA1A 39 4 MYC 99 14 TUBB 57 24 RAC1 38 5 GSK3B 84 15 LYN 49 25 PIN1 37 6 FYN 76 16 JUN 48 26 NFKB1 37 7 ABL1 70 17 GAPDH 47 27 CDK6 35 8 PRKACA 68 18 FOS 45 28 MAPK3 35 9 PRKCA 67 19 CREBBP 45 29 TK1 34 10 CDK1 67 20 TUBA1A 42 30 PSMA7 34 mRNA transcripts encoding proteins for which there are drugs, inhibitors or antagonists available (through DrugBank v3.0), ranked by degree within a protein-protein interaction network of differentially expressed transcripts. Degree reflects the number of interaction partners (for the encoded protein) which show significant changes in transcript abundance omponents showed variable levels of transcriptional disruption to their local interactome Table 2 Signalling pathway components showed variable levels of transcriptional disruption to th Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 9 of 21 Figure 3 Differences in signalling component transcript changes between EGF and hypoxia induced EMT. Changes in transcript abundance (legend top right) for selected intracellular signalling components, within a schematic representation of the signalling network interactions between encoded proteins. Note the use of alternative colour-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Kinase inhibitors within the families selected for screening (described in text; shown in purple) are listed in Table 3. Figure 3 Differences in signalling component transcript changes between EGF and hypoxia induced EMT. EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling As detailed above, systems-level analysis of the mRNA transcript abundance changes that occurred with induc- tion of EMT identified several signalling molecules that were likely to have dysregulated activity, and may play a role in promoting the mesenchymal phenotype. To inves- tigate the potential for therapeutic intervention against these signalling components, a panel of kinase inhibitors (Table 3) was tested to determine the concentrations at which the fraction of vimentin+ cells or cell count was re- duced by 50% (IC50 concentrations). Although EGF stimulation further increased the mRNA transcript abundance of EMT markers (Figure 1i) the inability of EGFR inhibitors to reduce the fraction of vimentin+ PMC42-ET cells (Table 3a) suggests that the unstimulated mesenchymal phenotype of these cells is maintained through EGFR-independent signal- ling mechanisms. Inhibitors targeting the MEK1/2 (Table 3b) and Src- family kinases (Table 3c) showed a similar response profile to the EGFR inhibitors with potent blocking of vimentin expression within the EGF-stimulated cells. For MEK inhib- itors the IC50 values for inhibition of vimentin expression tended to be lower than the corresponding IC50 values for cell count, and within MDA-MB-468 cells the IC50 values were again higher with HPX stimulation than EGF stimula- tion (Table 3b). A similar effect was seen for inhibition of phospho-ERK1/2 (data not shown). A previously reported mRNA transcript signature for ‘compensatory resistance’ to AZD6244 (Additional file 5: Figure S4b) [72] shows some agreement with the observed efficacy of this MEK inhibitor (Table 3b). The EGF-stimulated PMC42-LA cells had the lowest IC50 for AZD6244 in reducing the fraction of vimentin+ cells by several orders of magnitude, and many of the AZD6244 resistance signature genes showed a decrease in transcript abundance relative to unstimu- lated PMC42-LA cells (Additional file 5: Figure S4b). Al- though the profile of this signature was very similar between EGF- and HPX-stimulated MDA-MB-468 cells, several of the transcripts showed a greater degree of up- regulation with hypoxia, in agreement with the reduced efficacy of AZD6244 within hypoxia-stimulated MDA- MB-468 cells (Table 3b). For most Src family inhibitors within EGF-stimulated MDA-MB-468 cells the values for IC50 of vimentin+ cells were lower than the IC50 values for cell count. Conversely, within EGF-stimulated PMC42-LA cells and HPX-stimulated MDA-MB-468 cells the IC50 for cell count is lower for most Src family inhibi- tors (Table 3c). Systems-level computational analysis identified putative drug targets to alleviate signalling pathway dysregulation that occurs with induced EMT small molecule inhibitor of HIF1α accumulation and gene transcriptional activity, CAY10585, to determine whether this could reduce the induction of EMT in these cells. At concentrations below 1 μM CAY10585 did not have a sig- nificant effect on the number of vimentin+ cells; however, the number of vimentin−cells was potently reduced, sug- gesting this may have a deleterious effect upon the cell population with an epithelial phenotype (Additional file 7: Figure S6). EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling Systems-level computational analysis identified putative drug targets to alleviate signalling pathway dysregulation that occurs with induced EMT Changes in transcript abundance (legend top right) for selected intracellular signalling components, within a schematic representation of the signalling network interactions between encoded proteins. Note the use of alternative colour-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Kinase inhibitors within the families selected for screening (described in text; shown in purple) are listed in Table 3. catalytic subunits to inhibit PI3K activity in the absence of upstream signals [69,70], and PIK3R1 (p85α) was sig- nificantly downregulated with EGF- or HPX-stimulation, although PIK3R2 (p85β) was increased, particularly with EGF stimulation (Figure 3). Furthermore, when consider- ing disruption to the local interactome PIK3R1 was highly ranked, suggesting a greatly reduced threshold for signal- ling through PI3K, particularly under conditions where HPX is driving EMT. transcript abundance increased significantly under hyp- oxic conditions (Figure 3). Increased signalling activity through MEK1/2-ERK-1/2 is the canonical downstream response to EGFR stimula- tion over many cell types [64,71], and activation of EGFR signalling induces a large number of feedback mecha- nisms to further modulate pathway activity [63]. This is consistent with the observation that MAPK3 (ERK1) showed some degree of disruption to its local interactome (Table 2), and with the notion that the MEK1/2-ERK-1/2 axis is a key integrator of dysregulated signalling pathways across the various models of induced EMT. It is possible that under conditions where key signalling proteins have been disrupted (e.g. an inactivating mutation in PTEN), some of these feedback mechanisms may lead to aberrant signalling. We examined differentially expressed genes with a previously identified transcriptional signature for MEK pathway activation [72] and found many of these Given the evidence for signalling through PI3K as described above, it was interesting to note changes in transcript abundance for the AKT scaffolding compo- nents Hsp90 and Cdc37 (Figure 3) with the HSP90AA1 and HSP90AB1 local networks showing the greatest de- gree of disruption (Table 2). Vivanco et al. demonstrated that GSK3B is an important downstream effector of AKT signalling [70], which also showed a high degree of disruption to the local interactome. Furthermore, AKT3 Page 10 of 21 Page 10 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 transcripts were significantly upregulated within the EGF or HPX-stimulated MDA-MB-468 cells (Additional file 5: Figure S4a). EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling The majority of inhibitors tested on EGF-stimulated PMC42-ET cells were efficacious at reducing cell count; however, nearly every inhibitor tested had an IC50 for re- ducing the number of vimentin+ cells well above pharmacologically relevant concentrations (Table 3a-e), thus off-target effects are likely. As expected, the panel of EGFR kinase inhibitors (Table 3a) were very effective at blocking EGF-induced EMT and cell growth in the PMC42-LA and MDA-MB- 468 cells, and with the exception of lapatinib, the IC50 values for inhibition of vimentin expression are 8–10 fold lower than the corresponding IC50 values for reduc- tion of cell count. Reduced levels of vimentin expression correlated with the ability of these compounds to inhibit the phosphorylation of ERK1/2 over a range of concentra- tions (Additional file 6: Figure S5), demonstrating the importance of the EGFR/MEK/ERK canonical pathway and its associated networks in promoting EMT-associated phenotypic changes. The EGFR kinase inhibitors also ap- peared to have an effect on HPX-stimulated MDA-MB- 468 cells, although IC50 values for HPX-stimulated cells were all higher than corresponding IC50 values for EGF- stimulated cells. In particular, inhibition of the HPX-in- duced vimentin response in MDA-MB-468 cells occurred at drug concentrations >10-fold higher than required for inhibition of ERK phosphorylation, indicating that the MEK/ERK pathway may be less important for EMP and the regulation of vimentin expression under hypoxic growth conditions. This effect may also be due to drug resistance mechanisms as discussed below. EGFR kinase inhibitor- mediated reductions in cell count for both EGF- and HPX- stimulation were generally observed at concentrations an order of magnitude greater than the effects on vimentinex- pression, indicating that our treatments are affecting EMT at relevant concentrations, while reduction in cell viability at higher concentrations may be caused by off target effects. Hypoxia-treated MDA-MB-468 cells were exposed to a Within EGF-stimulated PMC42-LA cells GDC-0941 and GSK2126458 were the only PI3K/mTOR inhibitors (Table 3d) with pharmacologically relevant IC50 values for reduction of vimentin+ cells, although most inhibi- tors were capable of reducing cell growth. EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling The PI3K/ mTOR inhibitors were much more efficacious within the MDA-MB-468 cells, and many of the tested compounds Table 3 Targeted Inhibition of signalling molecules show differential effects between EGF- and hypoxia-induced EMT Cell line: PMC42-ET PMC42-LA MDA-MB-468 Stimulus: EGF HPX Measured IC50 for: Biochemical assay (published data from compound vendors) % vimentin- positive cells (μM) Cell count (μM) % vimentin- positive cells (μM) Cell count (μM) % vimentin- positive cells (μM) Cell count (μM) % vimentin- positive cells (μM) Cell count (μM) Targets: Compound name Erlotinib (Tarceva) EGFR (2 nM) >25 15.04 0.18 16.42 0.14 16.42 5.71 >25 Lapatinib (GW572016) EGFR (10.2 nM), HER2 (9.8 nM) >25 2.46 0.57 1.65 4.32 2.99 1.96 1.44 a) HERs/EGFR Vandetanib (Zactima) VEGFR2 (40 nM), VEGFR3 (110 nM), EGFR (500 nM) >25 1.84 0.57 1.57 0.50 3.77 - - Gefitinib (Iressa) EGFR (33 nM) >25 5.78 0.26 1.48 0.25 2.31 6.62 6.27 TOVOK (Afatinib) Irreversible binder. EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling EGFR (0.5 nM), HER2 (14 nM) >25 1.24 0.02 0.96 0.03 0.26 2.38 1.01 AV-412 EGFR (43 nM), HER2 (282 nM) >25 0.33 0.05 0.32 0.05 0.54 >25 0.06 U0126 MEK1 (70 nM), MEK2 (60 nM) >25 >25 1.38 >25 0.38 8.74 4.99 >25 SL 327 MEK1 (180 nM), MEK2 (220 nM) >25 16.43 1.43 >25 >25 12.56 >25 0.03 b) MEK-1/2 PD 198306 MEK (8 nM) >25 1.36 0.34 2.50 0.46 0.97 1.7 2.94 AZD6244 (Selumetinib) MEK1 (14 nM) >25 >25 0.06 >25 1.38 13.71 >25 >25 CI-1040 (PD-184352) MEK (1–1.3 nM) >25 3.27 0.16 1.40 0.19 2.00 4.7 6.33 PD0325901 MEK (0.33 nM) >25 >25 <0.02 >25 <0.02 2.77 4.17 0.06 PD173955-Analogue 1 c-Src (9 nM) >25 5.94 >25 6.28 1.70 3.40 6.55 >25 Saracatinib (AZD0530) Src (2.7 nM) >25 0.75 0.94 0.50 >25 6.35 >25 0.72 c) Src family Bosutinib (SKI-606) Src (1.2 nM), Abl (1 nM) >25 1.40 0.23 1.17 0.25 1.05 1.19 0.54 Dasatinib (BMS-354825) Src (0.8 nM), Abl (0.6 nM) >25 0.04 0.76 <0.02 0.64 5.59 >25 0.04 PD173952 Src (8 nM), Lck (5 nM), FGFR1 (100 nM) >25 0.35 1.61 0.23 0.10 0.25 - - PIK-90 PI3K (α 11 nM, β 350 nM, γ 18 nM, δ 58 nM) >25 16.35 >25 16.36 3.26 >25 >25 <0.02 ZSTK474 PI3K (α 17 nM, β 53 nM, γ 6 nM) >25 0.83 >25 2.11 0.21 0.72 2.35 >25 GDC-0941 PI3K (α 3 nM, β 33 nM, γ 75 nM, δ 3 nM) >25 0.46 8.67 1.18 1.27 1.79 4.69 4.88 alling molecules show differential effects between EGF- and hypoxia-induced EMT Table 3 Targeted Inhibition of signalling molecules show differential effects between EGF- and hypoxia-induced EMT (Continued) BEZ-235 (NVP-BEZ235) p110 (α 4 nM, β 75 nM, γ 5 nM, δ 7 nM) >25 0.06 >25 >25 0.02 0.05 0.05 3.43 d) PI3K/mTOR PI103 DNA-PK (2 nM), mTORC1 (20 nM), PI3K-C2b (26 nM), p110 (α 8 nM, β 88 nM, γ 150 nM, δ 48 nM) 15.18 0.32 >25 1.82 0.38 0.88 1.22 1.04 GNE-493 PI3K (α 3.4 nM, β 12 nM, γ 16 nM, δ 16 nM) >25 1.11 >25 12.09 0.29 1.45 0.99 6.41 GSK2126458 (HYR-582) Ki: P110 (α 0.019 nM, β 0.13 nM, γ 0.06 nM, δ 0.024 nM), mTORC1 (0.18 nM), mTORC2 (0.3 nM) >25 0.02 6.69 0.62 <0.02 0.08 0.29 0.74 GNE-490 PI3K (α 3.5 nM, β 25 nM, γ 5.2 nM, δ 15 nM) >25 2.16 >25 2.23 0.93 1.25 12.68 >25 LY294002 PI3K (α 0.5 uM, β 0.97 uM, γ 0.57 uM) >25 14.86 >25 12.12 4.18 13.22 >25 >25 GSK690693 Akt1 (2 nM), Akt2 (13 nM), Akt3 (9 nM) >25 >25 >25 8.32 >25 3.31 >25 >25 A-674563 Ki: Akt1 (11 nM), PKA (16 nM), CDK2 (46 nM), ERK2 (260 nM) >25 0.48 0.17 0.76 0.65 0.25 2.83 0.60 Akt-i-1 Akt1 (4.6 μM) >25 >25 >25 >25 >25 6.46 >25 12.30 e) Akt Akt-i-1/2 Akt1 (58 nM), Akt2 (210 nM) >25 >25 >25 >25 >25 2.82 >25 4.79 AT7867 Akt1 (32 nM), Akt2 (17 nM), Akt3 (47 nM), PKA (20 nM) >25 2.63 >25 0.24 >25 2.95 >25 4.57 AZD5363 Akt1 (3 nM), Akt2 (8 nM), Akt3 (8 nM), ROCK2 (56 nM) >25 >25 >25 >25 0.63 >25 >25 >25 Merck-22-6 Akt1 (138 nM), Akt2 (212 nM) >25 4.27 >25 1.48 >25 0.46 >25 0.55 MK-2206 Akt1 (8 nM), Akt2 (12 nM), Akt3 (65 nM) >25 5.62 >25 3.16 >25 1.86 >25 9.77 Inhibition of vimentin expression and cell counts by a selection of kinase inhibitors. Inhibition of vimentin expression and cell counts by a selection of kinase inhibitors. Shown are the IC50 values (where the fraction of vimentin positive cells, or the cell count, was reduced by 50% compared to the controls). Concentrations are specified in μM units. Inhibitors have been grouped according to the kinases they target. The dose–response curves for selected kinase inhibitors are shown in Figure 4. For reference, the IC50 values of each compound measured in biochemical assays with purified enzymes are included. EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling Shown are the IC50 values (where the fraction of vimentin positive cells, or the cell count, was reduced by 50% compared to the Table 3 Targeted Inhibition of signalling molecules show differential effects between eted Inhibition of signalling molecules show differential effects between EGF- and hypoxia-induced gnalling molecules show differential effects between EGF- and hypoxia-induced EMT (Continued) Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 13 of 21 e 4 (See legend on next page.) (S l d t ) Cursons et al. Cell Communication and Signaling (2015) 13:26 Cursons et al. Cell Communication and Signaling (2015) 13:26 Page 14 of 21 (See figure on previous page.) Figure 4 Hypoxia- and EGF-induced metastatic MDA-MB-468 cells show markedly different responses to pharmacological inhibitors. Pharmacological dose–response curves showing the fraction of vimentin-positive cells (blue; left axes) and cell-count (red; right axes) in the presence of (a-c) the MEK inhibitor AZD6244, (d-f) the PI3K inhibitor GDC-0941, (g-i) the AKT1/2/3 inhibitor AZD5363 (j-l) and the mTOR inhibitor Everolimus. (m-o) pharmacological inhibition of vimentin with a combination of MEK-1/2 (AZD6244) and AKT1/2/3 (AZD5363) inhibitors at varying concentrations. (p) pharmacological inhibition of vimentin with a comination of MEK-1/2 (AZD6244) and AKT1 (Akt-i-1) inhibitors at varying concentrations. had a lower IC50 value for the reduction of vimentin+ cells compared to the reduction in cell count. assays compared to results obtained with purified en- zymes. This was expected due to effects such as competi- tion with high levels of intracellular ATP, binding to other proteins and limited cellular permeability [73]. In contrast to PI3K/mTOR inhibitors, the majority of compounds targeting Akt kinases (Table 3e) were only capable of reducing cell count, with A-674563 and AZD5363 the only inhibitors with a pharmacologically relevant IC50 value for vimentin+ cells across any of the cell lines and conditions. Unexpectedly, some Akt inhibi- tors and mTOR inhibitors were observed to increase the fraction of vimentin+cells and the relative cell density, particularly within HPX-stimulated MDA-MB-468 cells (Figure 4h & k). A range of factors regulate EMP through various signal- ling pathways [74], and “kinase switching” from the ErbB axis to FGFR and PDGFR has been associated with EMT in NSCLC models [75]. We focussed on differences in the signalling mechanisms associated with EGF- or HPX- induced EMT within MDA-MB-468 cells as a model of TNBC (Figure 4). EGF- and HPX-stimulated MDA-MB-468 cells show different responses to pharmacological inhibition of MEK-ERK and PI3K/Akt signalling Many drugs that induced a response within the EGF-stimulated MDA-MB-468 cells, such as the MEK inhibitor AZD6244 (Figure 4a & b), showed re- duced efficacy or even pro-proliferative effects within HPX-stimulated MDA-MB-468 cells. It has previously been reported that hypoxia can have varied effects across different kinase inhibitors, and this may be partially medi- ated by modulation of hypoxia-induced compensatory mechanisms, such as VEGF signalling [76]. Stark differ- ences were observed in the responses elicited by some in- hibitors, including the mTOR inhibitor Everolimus, and the AKT1/2/3 inhibitor AZD5363 (Figure 4 g & h; j & k). These divergent responses to pharmacological perturbation support the conclusion that subtly different signalling mechanisms are responsible for driving the phenotypically similar EMT processes that occurred with EGF or HPX stimulation of MDA-MB-468 cells (Figure 1). Intriguingly, synergistic effects for blocking EGF-induced EMT were observed when combining an AKT1/2/3 inhibitor with the MEK-1/2 inhibitor AZD6244, but not for an inhibitor which targeted AKT1 alone (Figure 4p), indicating that AKT1 is not solely responsible for the protective signalling seen in this system. The observation that several classes of inhibitors were efficacious within EGF-stimulated MDA-MB-468 cells, but had little effect under hypoxic growth conditions, supports the conclusion from the transcriptome analysis that the phenotypically similar EMT processes induced with EGF or hypoxia are driven by different signalling mechanisms. Furthermore, given differences observed between the EGF- and HPX-induced transcriptional pro- files, particularly for signalling ligands where the recep- tor also has strong increases in transcript abundance, such as HBEGF/EGFR and VEGFA/KDR, we hypothesised that pro-survival signalling through AKT may mediate the reduced efficacy of MEK-1/2 inhibitors under hypoxic con- ditions. Thus, we also applied the AKT1/2/3 inhibitors GSK690693 or AZD5363 in combination with the MEK1/2 inhibitor AZD6244. The pharmacological efficacy curves suggest that they provide a synergistic effect to block the relative fraction of vimentin+ cells (Figure 4 m). Further- more, this effect was not observed with the AKT1 or AKT1/2 inhibitors tested in combination with AZD6244 (Figure 4p). Differences in the transcriptional profile and pharmacological responsiveness between EGF- and hypoxia-induced EMT Al- though it is not included within the KEGG HIF-1 sig- naling pathway, DDIT4 is a known HIF-1-responsive transcript (HIF-1 responsive RTP801) which can modulate mTORC1 through the RHEB inhibitors TSC1/TSC2. Other HIF-1 target genes necessary for metabolic adapta- tion to hypoxic growth also showed large differences, in- cluding HK2, LDHA, PFKFB3, and SLC2A1 (Figure 2b). These differences likely reflect stabilisation of HIF-1α under hypoxic growth conditions, although many of the metabolism-associated HIF-1 targets also had increased transcript abundance with EGF stimulation (Figure 2b). This may be mediated by MEK1/2-ERK1/2 signalling through MKNK2 to eIF4E, influencing HIF-1α translation. Pre-clinical studies have observed such effects with hypoxia-induced EMT [39] and this may contribute to the deleterious effect of VEGFR inhibitors [87]. Assuming that increased abundance of SOX9 mRNA (Figure 1i) contributes to increased transcription factor activity, this driver of mammary stem cell behaviour likely promotes EMT (Figure 1a–h). The increase in SOX9 transcript abundance was only significant within the HPX-stimulated MDA-MB-468 cells (Figure 1i), al- though SOX9 also showed strong differences in tran- script abundance (q-value < 0.05) between the ET and LA sublines (Figure 2c). SOX9 has been linked to clin- ical chemoresistance in colorectal cancer [88], an affect which may be partially mediated by EMT changes. Fur- thermore, cancer stem cell markers are negative pre- dictive markers for the efficacy of everolimus in treating TNBC [89], and this may underpin the failure of everolimus to block EMT within hypoxia-stimulated MDA-MB-468 cells (Figure 4 k). Numerous transcripts showed differences in abundance between EGF- and HPX-mediated EMT (Figure 2 & 3), and some of the altered signalling components likely con- tributed to the variable drug efficacy. The multidrug resistance-promoting P-glycoprotein (ABCB1) had a large increase in transcript abundance within hypoxia-stimulated MDA-MB-468 cells (2.6-fold increase, q-value = 0.029; for EGF-stimulation 1.6 fold increase, q-value = 0.36), which may have reduced the efficacy of several kinase inhibitors [76,84,85]. The ability of EGFR inhibitors to block EMT in HPX- stimulated MDA-MB-468 s (albeit at higher concentra- tions than in EGF-stimulated MDA-MB-468 s; Table 3a) suggests that hypoxia-induced EMT may be partially me- diated by paracrine/autocrine EGFR signalling. This is supported by the observation that transcript abundances for several EGF ligands were significantly increased with HPX stimulation, as was EGFR itself (Figure 3). Differences in the transcriptional profile and pharmacological responsiveness between EGF- and hypoxia-induced EMT Transcriptional profiling of two human breast cancer models indicated that subtly different transcriptional re- sponses underpinned EMT induced with EGF or HPX (Figure 1a-h &i). This included variation in the relative abundance of EMT-implicated transcription factors (Figure 1j & k), and alternative signalling pathways dys- regulated by the transcriptional changes (Table 1 & 2; Figure 2 & 3). A panel of kinase inhibitors were selected from across the network of disrupted signalling compo- nents, within which PI3K-Akt and MEK1/2-ERK-1/2 ap- peared to act as signal integrators. In general, tested compounds had a much lower potency within our cellular We saw differences in the transcript abundance and/or regulation of several well-studied TFs previously associ- ated with EMT in breast cancer [77-79] (Figure 1j). Rela- tively large (although not statistically significant) changes in transcript abundance for TWIST1 under HPX condi- tions are consistent with its reported regulation by HIF1 [36,37,80], as are the increases for ZEB1 [80] (Figure 1j). Failure to detect SNAI2 transcripts within the MDA-MB- 468 cells data was consistent with one previous report [81]; however, increases in SNAI2 mRNA abundance Page 15 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 following EGF stimulation of MDA-MB-468 cells have been reported together with enrichment at sites of in vivo EMT [48]. This discrepancy may reflect the different prov- enance of MDA-MB-468 cells in Belgium and Australia, or alternatively, SNAI2 transcripts may be expressed at sufficiently low levels that they approach the signal-to- noise ratio of our RNA-Seq protocol. Our previous study showed that treatment of MDA-MB-468 cells with HPX caused a non-significant increased at different time points in SNAI2, TWIST1 and ZEB2, and a significant increase in ZEB1 [42]. The dominant role of ZEB1 in the current study is also consistent with our previous observations that PMC42-ET cells have significantly higher levels of ZEB1 and SNAI2 than PMC42–LA cells, and that ZEB1 and SNAI2 were both increased in PMC42-LA cells after EGF treatment [41,42]. ZEB1 appears to be a downstream integration point for EMT regulation [41], and is subject to complex regulation at multiple levels [82,83]. This dif- ferential control between EMT scenarios may have clinical utility in allowing selective inhibition of EMT mechanisms involved in tumour progression, whilst leaving critical physiological processes unperturbed to reduce toxicity. HIF-1, such as SERPINE1, VEGFA, and EDN1. Activation of MEK-ERK signalling and promotion of signalling through ERK2 may contribute to EMT development in hypoxia [102], and a number of MEK inhibitors are in early clinical trials across solid tumour types, although information on breast cancer responsiveness is still scarce [103]. A phase II clinical trial for the MEK inhibitor CI-1040 in chemo- therapy pre-treated metastatic breast cancer showed no major effects, although one patient developed stable dis- ease [104]. The lack of frequent mutations within the core Ras-Raf-MEK axis, but the potential for cross-talk with a plethora of pathways intrinsic to breast cancer progres- sion, may mean that the potential of MEK blockade lies in treatment combinations to overcome resistance. This is borne out by pre-clinical studies which have shown MEK inhibition has the potential to enhance sensitivity of breast cancer xenografts to HER2 blockade [105] and anti- estrogen treatment [106]. Furthermore, studies with breast cancer cell lines have shown that MEK inhibition also increases sensitivity to EGFR blockade [107], and reversed the effects of IGF-1R overexpression in promoting prolif- eration [108]. yp Recent evidence has indicated that ERK2 (MAPK1) sig- nalling is central to EMT, activating DEF-motif tran- scription factors such as FOSL1 (FRA1) and ZEB1/ZEB2 [92,93]. Although increases in ERK2 (MAPK1) transcript abundance were not significant with EGF or HPX stimu- lation, there were corresponding decreases in ERK1 (MAPK3) transcript abundance that were significant within hypoxia-stimulated MDA-MB-468 cells (Figure 3). The altered ratio of transcripts may have led to ERK2 becoming the dominant isoform, while activation of Ras under stimulated conditions drives signalling though MEK-1/2 (Additional file 5: Figure S4a) to phosphorylate ERK2. Kinome profiling of TNBC tumours suggests that ERK2 is activated compared to control tissue, while ERK1 activity remains unchanged [94], and it is tempting to speculate that the hypoxic tumour environment drives in vivo ERK2 activation. FOSL1 was one of the few tran- scripts significantly upregulated across all our models of induced EMT and ZEB2 was not expressed within MDA- MB-468 cells (Figure 1j), suggesting ZEB1 and FOSL1 may be sufficient to mediate this transformation. Combination therapies with PI3K inhibition have been shown to enhance the effect of MEK inhibition within basal subtype breast cancer cells by alleviating the com- pensatory activation of PI3K/AKT that occurs with MEK inhibition [109]. We observed that a combination of MEK1/2 and AKT1/2/3 inhibitors had synergistic effects in blocking vimentin induction within our in vitro model of EGF-induced EMT (Figure 4m). Activation of MEK-ERK signalling and promotion of signalling through ERK2 may contribute to EMT development in hypoxia It should be noted, however, that when EMT was induced with hypoxic growth conditions this combination of kinase inhibitors promoted an increase in the relative fraction of mesen- chymal cells (Figure 4n) demonstrating the importance of elucidating the detailed effects of pathway manipula- tion in this area before proceeding to clinical studies. Depletion of AKT3 has previously been reported to sensitize TNBC cell lines, including MDA-MB-468 cells, to the pan-Akt inhibitor GSK690693 [110] and The Cancer Genome Atlas Project (TCGA) data show that AKT3 is up- regulated in 28% of TNBCs [110,111]. The observation that inhibitors targeting AKT1 (Figure 4p) or AKT1/2 (data not shown) did not have a combinatorial effect with MEK1/2 inhibition suggests that AKT3 mediates sufficient signal transduction to provide functional compensation during inhibition of AKT1 and AKT2. This specific possibility re- mains to be tested. Clinical Implications Oncogenic mutations of Ras are important drivers of malignant behaviour within melanoma and pancreatic cancers, and although such activating mutations are relatively rare within breast cancer, overexpression of Ras mRNA and protein has been demonstrated [95]. Our data show a strong increase in transcript abundance for RRAS2 (Figure 3), consistent with reports that RRAS2 drives PI3K-dependent tumorigenesis and con- tributes to late stage metastasis in certain lung cancers [96]. Activation of Ras proteins by a range of growth fac- tor receptors [97] leads to activation of the Raf-MEK- ERK [94] and the PI3K-Akt signal transduction cascades, culminating in the regulation of cellular survival and proliferation genes [90,98]. Ras is difficult to target therapeutically [99], although up- and downstream path- way components may be inhibited [100]. Inhibition of Src upstream has proved disappointing with response rates below 10% in unselected TNBCs [101] whereas downstream B-Raf inhibition is currently unexplored. Differences in the transcriptional profile and pharmacological responsiveness between EGF- and hypoxia-induced EMT This could drive paracrine/autocrine EGFR signalling to promote EMT, although it should be noted that EGF was present within the culture media and this would also drive some EGFR signalling. HB-EGF mediated activation of EGFR is an important driver of MDA-MB-231 cell invasiveness, particularly for brain metastases [86]. A similar role has been described for autocrine TGFβ signalling in promot- ing EMT [82], and it is interesting that modulators of TGFB signalling showed significant changes in tran- script abundance, including: THBS1 (Figure 1i), INHBA, TGFBR2 (Figure 2b), INHBB and BMP4 (Figure 2c). Pathway convergence on PI3K-Akt and MEK1/2-ERK1/2 The presence of PI3K-Akt and MEK1/2-ERK1/2 com- ponents across multiple signalling pathways (Figure 2a) is consistent with the role of these evolutionarily- conserved modules in the integration of various extra- cellular stimuli [90]. As detailed above, PI3K-Akt and MEK1/2-ERK-1/2 are well-known effectors of EGFR signalling [65] and other receptor tyrosine kinases [69,90], and signalling through these pathways can in- duce a variety of feedback mechanisms [63]. Even within breast cancer cell lines that do not over-express HER2, such as triple negative MDA-MB-231 cells, EGFR-induced signalling through PI3K/Akt is thought to be involved in mediating EMT [31], while several other studies implicate MEK1/2-ERK1/2 signalling as an important driver [91]. The relatively high frequency of PI3K-Akt and MEK1/2-ERK1/2 may, simply reflect their relatively well-studied nature, however, leading to inclusion across numerous KEGG maps. Many of the ‘HIF-1 signalling’ transcripts with large dif- ferences in abundance between EGF- and HPX-stimulated MDA-MB-468 cells were known transcriptional targets of Page 16 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 Activation of MEK-ERK signalling and promotion of signalling through ERK2 may contribute to EMT development in hypoxia Activation of MEK-ERK signalling and promotion of signalling through ERK2 may contribute to EMT development in hypoxia Conclusions (d) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Neve data [44]. MDA-MB-468 cells are highlighted (cyan). Unstimulated MDA-MB-468 cells (Figure 1 & Additional file 2: Figure S1a) have a relatively high abundance of SOX9 and a relatively low abundance for FOXA2, FOSL1 SNAI2, TWIST1, ZEB1, ZEB2, in agreement with the both the Heiser et al. [112] (Additional file 2: Figure S1c) and the Neve et al. [44] (Additional file 2: Figure S1d) studies. Additional file 3: Figure S2. Changes in abundance for selected mRNA transcripts. Heat maps (legend at left) showing changes in transcript abundance between the specified conditions (at top) for specified transcripts where there is a significant difference in abundance between EGF and HPX stimulated MDA-MB-468 cells, with (a) divergent or (b) consistent changes between EGF and HPX stimulated MDA-MB-468 cells. Targets have been clustered by expression pattern and are ordered accordingly. Additional file 4: Figure S3. Changes in abundance for selected mRNA transcripts. Transcripts with consistent differential transcript abundance across all condition comparisons, and their membership across all KEGG maps (at right). We have demonstrated that hypoxic conditions funda- mentally change the way breast cancer cells respond to drugs and compounds in various stages of development for treatment of breast cancer. The role of HIF1 in promoting a mesenchymal phenotype has been well ex- plored [4,11,36,37], as has the role of hypoxia in promot- ing drug resistance components [76,84,85]. Perhaps most importantly, our data also indicate that under hypoxic conditions some therapeutic interventions may shift cells into an even more aggressive, mesenchymal phenotype. This highlights the vital importance of evaluating novel drug targets under a more-representative range of stimuli and conditions that mimic the heterogeneity of environ- mental conditions tumours are exposed to in situ. While attention has recently been drawn to the problem of genetic heterogeneity in breast cancer tumours, our study indicates that extra-cellular conditions, such as those we have explored here to stimulate EMT, can induce diver- gent molecular states even on a common genetic back- ground, resulting in altered drug sensitivity and response. Given the hypoxic conditions commonly prevalent in the core of solid breast tumours, these findings have clear clinical implications for both treatment, and the drug development process. Additional file 5: Figure S4. Components of transcriptional signatures for MEK signalling activation (at top) and compensatory resistance to AZD6244 (at bottom). Conclusions Pharmacological response curves for the MEK-1/2 inhibitor CI-1040, showing the %inhibition of vimentin (blue), %inhibition of phospho-ERK-1/2 (green) and reductions in cell count (red). (a) Note that there is good correlation between the inhibition of vimentin and phospho- ERK over a range of concentrations for EGF stimulated MDA-MB-468 cells. (b) For MDA-MB-468 cells grown under hypoxic conditions, inhibition of phospho-ERK shows a similar response; however, inhibition of vimentin expressing cells only occurs at relatively high concentrations of phospho-ERK, suggesting the activation of compensatory signalling mechanisms. Conclusions To examine the hypothesis that signalling through MEK is implicated in these models of induced EMT, we examined changes in the abundance of genes previously classified as ‘transcriptional signatures’ for MEK pathway activation and AZD6244 sensitivity [72]. A large number of transcripts within the MEK signalling activation signature showed increased transcript abundance in the EGF stimulated PMC42-ET and PMC42-LA cells, with most of the changes not being statistically significant. Although changes in transcript abundance for components of the MEK signalling signature are less consistent within the stimulated MDA-MB-468 s, there were more transcripts showing a significant (q-value < 0.05) increase in abundance suggesting that signalling through MEK1/2 is active. Examining mRNA transcripts which have been associated with compensatory resistance to AZD6244, the EGF stimulated PMC42-LA cells show a number of transcripts with decreased abundance. For the EGF and HPX stimulated MDA-MB-468 cells a number of these transcripts show increased abundance, with more than half showing a statistically significant (q-value < 0.05) increase in abundance within MDA-MB-468 cells grown under hypoxic conditions. It is interesting to note that the PMC42-ET and –LA subline comparison showed the weakest signature for MEK signalling activation, and the strongest signature for AZD6244 resistance, regardless of EGF stimulation. The AZD6244 resistance signature shows relatively good agreement with our inhibitor screen results (Table 3), such that EGF stimulated PMC42-LA cells are susceptible to AZD6244, the inhibitor is slightly less efficacious within EGF stimulated MDA-MB-468 cells, and MDA-MB-468 cells grown under hypoxic conditions are not affected by AZD6244. Additional file 6: Figure S5. Inhibition of ERK phosphorylation shows good correlation with the inhibition of vimentin-positive cells within EGF and HPX induced EMT. Pharmacological response curves for the MEK-1/2 inhibitor CI-1040, showing the %inhibition of vimentin (blue), %inhibition of phospho-ERK-1/2 (green) and reductions in cell count (red). (a) Note that there is good correlation between the inhibition of vimentin and phospho- ERK over a range of concentrations for EGF stimulated MDA-MB-468 cells. (b) For MDA-MB-468 cells grown under hypoxic conditions, inhibition of phospho-ERK shows a similar response; however, inhibition of vimentin expressing cells only occurs at relatively high concentrations of phospho-ERK, suggesting the activation of compensatory signalling mechanisms. Additional file 6: Figure S5. Inhibition of ERK phosphorylation shows good correlation with the inhibition of vimentin-positive cells within EGF and HPX induced EMT. Conclusions The effects of MEK inhibition within breast cancer are poorly studied in comparison to other cancers, particularly melanoma and lung cancer. Treatment of MDA-MB-231 and SUM159 cells with the MEK inhibitor AZD6244 has been shown to reduce c-Myc mRNA transcript and pro- tein abundance, leading to receptor tyrosine kinase repro- gramming which drives MEK inhibitor resistance [94]. Breast cancer cell lines sensitive to the MEK inhibitor selumetinib tend to be a basal subtype with Raf mutations In this report we have studied the mRNA transcript pro- file across several models of induced EMT to identify common dysregulated signalling components that may contribute to development and maintenance of the metastatic phenotype. Given the putative role for EGF signalling and hypoxia in mediating tumour progression in vivo, our analysis focussed on the differences between these stimuli in promoting EMT within MDA-MB-468 Page 17 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 cells as a model of triple negative breast cancer. A num- ber of kinase inhibitors were targeted at different points across the network of dysregulated signalling components, and the alternative stimuli were associated with variation in the efficacy of kinase inhibitors at blocking induction of EMT. A combination of AKT1/2/3 and MEK1/2 inhibitors was shown to have synergistic effects on block- ing the induction of EMT in vitro. The effects of simultan- eously blocking these important signalling pathways are likely to be deleterious to many different cell types; however, using novel targeted drug delivery mechanisms that are under development it may be possible to apply this combination therapy for the clinical treatment of EMT within TNBC. Furthermore, with further compara- tive study the differential control of EMT by alternative driver molecules we have identified may allow a selective effect to be exerted on pathogenic EMT processes whilst leaving physiological processes intact, thereby minimising toxicity to patients. indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. (b) Differential transcript abundance within and between the in vitro models of induced EMT for Blick Basal B discriminator transcripts [13]. Note the use of alternative color-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/ orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. Additional file 1: Table S1. Provenance of EMT-associated transcripts in Figure 1j. References which describe the role of transcripts listed Figure 1j in promoting EMT. Additional file 2: Figure S1. mRNA abundance data of selected targets with comparison to published data sources for selected transcription factors. (a) changes in mRNA transcript abundance (for Figure 1i & 1j transcripts) between specified cell lines and conditions (at top), with the mean count value (normalised to counts-per-million; CPM) of the compared conditions overlaid. Note the use of alternative color-bars to Conclusions (c) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Heiser data [112]. MDA-MB-468 cells are highlighted (cyan). (d) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Neve data [44]. MDA-MB-468 cells are highlighted (cyan). Unstimulated MDA-MB-468 cells (Figure 1 & Additional file 2: Figure S1a) have a relatively high abundance of SOX9 and a relatively low abundance for FOXA2, FOSL1 SNAI2, TWIST1, ZEB1, ZEB2, in agreement with the both the Heiser et al. [112] (Additional file 2: Figure S1c) and the Neve et al. [44] (Additional file 2: Figure S1d) studies. indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. (b) Differential transcript abundance within and between the in vitro models of induced EMT for Blick Basal B discriminator transcripts [13]. Note the use of alternative color-bars to indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/ orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. (c) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Heiser data [112]. MDA-MB-468 cells are highlighted (cyan). (d) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Neve data [44]. MDA-MB-468 cells are highlighted (cyan). Unstimulated MDA-MB-468 cells (Figure 1 & Additional file 2: Figure S1a) have a relatively high abundance of SOX9 and a relatively low abundance for FOXA2, FOSL1 SNAI2, TWIST1, ZEB1, ZEB2, in agreement with the both the Heiser et al. [112] (Additional file 2: Figure S1c) and the Neve et al. [44] (Additional file 2: Figure S1d) studies. indicate statistically significant (; q-value < 0.05; red-green) and non-significant (brown/orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. (b) Differential transcript abundance within and between the in vitro models of induced EMT for Blick Basal B discriminator transcripts [13]. Note the use of alternative color-bars to indicate statistically significant (**; q-value < 0.05; red-green) and non-significant (brown/ orange-teal) changes in abundance. Black squares indicate mRNA transcripts that were not reliably detected. (c) Relative transcripts abundances for EMT implicated TFs (from Additional file 2: Figure S1a) within the Heiser data [112]. MDA-MB-468 cells are highlighted (cyan). Abbreviations BCSC B 3. Hugo H, Ackland ML, Blick T, Lawrence MG, Clements JA, Williams ED, et al. Epithelial–mesenchymal and mesenchymal–epithelial transitions in carcinoma progression. J Cell Physiol. 2007;213:374–83. 3. Hugo H, Ackland ML, Blick T, Lawrence MG, Clements JA, Williams ED, et al. Epithelial–mesenchymal and mesenchymal–epithelial transitions in carcinoma progression. J Cell Physiol. 2007;213:374–83. BCSC: Breast cancer stem cell; BrCa: Breast cancer; EGF: Epidermal growth factor; EMP: Epithelial-mesenchymal plasticity; EMT: Epithelial-mesenchymal transition; ERK: Extracellular-signal regulated kinase; HPX: Hypoxia; MAPK: Mitogen activated protein kinase; MaSC: Mammary stem cell; MEK: MAPK/ERK kinase; MET: Mesenchymal-epithelial transition; 4. Yang J, Weinberg RA. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. 2008;14:818–29. 4. Yang J, Weinberg RA. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. 2008;14:818–29. 5. Polyak K, Weinberg RA. Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer. 2009;9:265–73. 5. Polyak K, Weinberg RA. Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer. 2009;9:265–73. PI3K: Phosphatidyl-inositol 3 kinase; TBS: Tris-buffered saline; TF: Transcription factor; TNBC: Triple-negative breast cancer. 6. Thiery JP, Sleeman JP. Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol. 2006;7:131–42. Authors’ contributions 8. Klymkowsky MW, Savagner P. Epithelial-mesenchymal transition: a cancer researcher’s conceptual friend and foe. Am J Pathol. 2009;174:1588–93. JC analysed the RNASeq data, performed the computational analysis, and prepared the manuscript. KJL designed and performed pharmacological assays, and analysed the pharmacological data. MW and ETC. participated in the study design and the RNASeq experiments. MF assisted with the computational analysis and analysis of the RNASeq data. CPB assisted with the computational analysis and analysed the RNASeq data. AF participated in study design and analysed the RNASeq data. EJC assisted with the computational analysis. IS designed the pharmacological assays and analysed the pharmacological data. MJD analysed the RNASeq data, performed the computational analysis, and prepared the manuscript. EWT participated in study design, analysed the RNASeq data, and prepared the manuscript. JC and KJL are joint first authors. All authors read and approved the final manuscript. 9. Thompson EW, Haviv I. The social aspects of EMT-MET plasticity. Nat Med. 2011;17:1048–9. 10. van Denderen BJ, Thompson EW. Cancer: The to and fro of tumour spread. Nature. 2013;493:487–8. 10. van Denderen BJ, Thompson EW. Cancer: The to and fro of tumour spread. Nature. 2013;493:487–8. 11. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;133:704–15. 12. Morel AP, Lievre M, Thomas C, Hinkal G, Ansieau S, Puisieux A. Generation of breast cancer stem cells through epithelial-mesenchymal transition. PLoS One. 2008;3:e2888. 13. Blick T, Hugo H, Widodo E, Waltham M, Pinto C, Mani SA, et al. Epithelial mesenchymal transition traits in human breast cancer cell lines parallel the CD44(hi/)CD24 (lo/-) stem cell phenotype in human breast cancer. J Mammary Gland Biol Neoplasia. 2010;15:235–52. Author details 1 l 1Systems Biology Laboratory, Melbourne School of Engineering, University of Melbourne, Building 193, Parkville, VIC 3010, Australia. 2ARC Centre of Excellence in Convergent Bio-Nano Science and Technology, Melbourne School of Engineering, University of Melbourne, Parkville, VIC 3010, Australia. 3The Walter and Eliza Hall Institute of Medical Research, 1G Royal Parade, Parkville, VIC 3052, Australia. 4Department of Medical Biology, The University of Melbourne, Parkville, VIC 3010, Australia. 5St. Vincent’s Institute, Melbourne, VIC, Australia. 6Centre for Cancer Biology, SA Pathology and University of South Australia, Adelaide, SA 5000, Australia. 7Discipline of Medicine, University of Adelaide, Adelaide, SA 5005, Australia. 8Royal Perth Hospital, Perth, Australia. 9School of Mathematics and Statistics, Faculty of Science, University of Melbourne, Parkville, VIC 3010, Australia. 10School of Medicine, Faculty of Medicine Dentistry and Health Sciences, University of Melbourne, Parkville, VIC 3010, Australia. 11Institute of Health and Biomedical Innovation and School of Biomedical Sciences, Queensland Institute of Technology, Brisbane, Australia. 12University of Melbourne Department of Surgery, St. Vincent’s Hospital, Melbourne, Australia. Received: 25 January 2015 Accepted: 22 April 2015 Additional file 8: Table S2. Estimated p-values for enrichment of elements within KEGG signal transduction maps without correction for multiple hypothesis testing. Signaling pathway KEGG maps which are over-represented (p-value < 0.05) for transcripts which show differential expression across at least one of the specified conditions are listed in Table 1. References 1. Duband JL, Monier F, Delannet M, Newgreen D. Epithelium-mesenchyme transition during neural crest development. Acta Anat (Basel). 1995;154:63–78. 1. Duband JL, Monier F, Delannet M, Newgreen D. Epithelium-mesenchyme transition during neural crest development. Acta Anat (Basel). 1995;154:63–78. 2. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial-mesenchymal transitions in development and disease. Cell. 2009;139:871–90. 2. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial-mesenchymal transitions in development and disease. Cell. 2009;139:871–90. Competing interests 7. Lee JM, Dedhar S, Kalluri R, Thompson EW. The epithelial-mesenchymal transition: new insights in signaling, development, and disease. J Cell Biol. 2006;172:973–81. The authors declare that they have no competing interests. Additional files EWT was supported in part by the National Breast Cancer Foundation (Australia) and Cancer Australia. MJD is funded by National Breast Cancer Foundation ECF-14-043. number of vimentin-negative cells. Hypoxia-treated MDA-MBA-468 cells were treated with the small molecule inhibitor of HIF1α activation and nuclear accumulation, CAY10585, or vector/negative-control (DMSO). Relative to the DMSO treated cells, CAY10585 caused no significant change in the cell count of vimentin+ cells; however, there was a profound reduction in the number of vimentin−cells. To ensure that the effect is target-related, an siRNA targeting HIF1α was tested, together with a scrambled siRNA/ negative-control. Relative to the scrambled siRNA treatment, the HIF1α siRNA caused a small increase in the number of vimentin+ cells; however, there was a much larger reduction in the number of vimentin−cells, in concordance with the inhibitor data. The transfection of siRNA into MDA-MB-468 cells was performed as a reverse transfection. Lipofectamine 2000 (Life technologies, 0.25 uL per well) and siRNA (final concentration 40 nM) were separately diluted into 25 uL of Opti-MEM (Life technologies) and incubated for 5 minutes. The solutions were then combined and incubated for 15 minutes at room temperature. To this mixture, 8,000 cells diluted into 100 uL of DMEM with 10% FBS, and the combined mixture was seeded in a 96 well assay plate (Corning, #3603). To other wells, 8,000 MDA-MB-468 cells were seeded without any siRNA or transfection reagent. The plates were incubated over night at 37C/5% CO2. The next day, media in the transfected wells were replaced with fresh media. HIF1a inhibitor (CAY10585, sc-205346, Santa Cruz biotechnology) was diluted in DMSO (0.5% final concentration) and added to the untrans- fected cells at the concentrations indicated in the figure. As a control, cells were treated with 0.5% DMSO. All siRNA and inhibitor reactions were performed in triplicate. The cells were incubated in a hypoxia chamber (1% O2) for 72 h, before being fixed with 3.7% formaldehyde for 15 minutes. The cells were then stained for Vimentin, imaged and analysed as previously described. The siRNAs used were HIF1a siRNA (sc-35561, Santa Cruz Biotechnology) and scrambled control siRNA (sc-37007, Santa Cruz Biotechnology). Additional files Additional file 1: Table S1. Provenance of EMT-associated transcripts in Figure 1j. References which describe the role of transcripts listed Figure 1j in promoting EMT. Additional file 2: Figure S1. mRNA abundance data of selected targets with comparison to published data sources for selected transcription factors. (a) changes in mRNA transcript abundance (for Figure 1i & 1j transcripts) between specified cell lines and conditions (at top), with the mean count value (normalised to counts-per-million; CPM) of the compared conditions overlaid. Note the use of alternative color-bars to Additional file 7: Figure S6. Within hypoxia-treated MDA-MB-468 cells, inhibition of HIF1α with the small molecule inhibitor CAY10585 or transfection with siRNA targeting HIF1α caused a large decrease in the Additional file 7: Figure S6. Within hypoxia-treated MDA-MB-468 cells, inhibition of HIF1α with the small molecule inhibitor CAY10585 or transfection with siRNA targeting HIF1α caused a large decrease in the Page 18 of 21 Page 18 of 21 Page 18 of 21 Page 18 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 Cursons et al. Cell Communication and Signaling (2015) 13:26 This work was supported in part by National Breast Cancer Foundation funding of the EMPathy Breast Cancer Network (CG-10-04), a National Collaborative Research Program. EWT was supported in part by the National Breast Cancer Foundation (Australia) and Cancer Australia. MJD is funded by National Breast Cancer Foundation ECF-14-043. This research was in part conducted and funded by the Australian Research Council Centre of Excellence in Convergent Bio-Nano Science and Technology (project number CE140100036). This study benefited from support by the Victorian Government’s Operational Infrastructure Support Program to St. Vincent’s Institute and The Walter and Eliza Hall Institute. This work was supported in part by National Breast Cancer Foundation funding of the EMPathy Breast Cancer Network (CG-10-04), a National Collaborative Research Program. EWT was supported in part by the National Breast Cancer Foundation (Australia) and Cancer Australia. MJD is funded by National Breast Cancer Foundation ECF-14-043. This research was in part conducted and funded by the Australian Research Council Centre of Excellence in Convergent Bio-Nano Science and Technology (project number CE140100036). This study benefited from support by the Victorian Government’s Operational Infrastructure Support Program to St. Vincent’s Institute and The Walter and Eliza Hall Institute. This work was supported in part by National Breast Cancer Foundation funding of the EMPathy Breast Cancer Network (CG-10-04), a National Collaborative Research Program. Acknowledgements Changes in cytoskeletal protein composition indicative of an epithelial-mesenchymal transition in human micrometastatic and primary breast carcinoma cells. Clin Cancer Res. 2005;11:8006–14. 45. Thompson EW, Paik S, Brunner N, Sommers CL, Zugmaier G, Clarke R, et al. Association of increased basement membrane invasiveness with absence of estrogen receptor and expression of vimentin in human breast cancer cell lines. J Cell Physiol. 1992;150:534–44. breast carcinoma cells. Clin Cancer Res. 2005;11:8006–14. 21. Kashiwagi S, Yashiro M, Takashima T, Nomura S, Noda S, Kawajiri H, et al. Significance of E-cadherin expression in triple-negative breast cancer. Br J Cancer. 2010;103:249–55. 46. Lundgren K, Nordenskjold B, Landberg G. Hypoxia, Snail and incomplete epithelial-mesenchymal transition in breast cancer. Br J Cancer. 2009;101:1769–81. 22. Qu C, Zhang W, Zheng G, Zhang Z, Yin J, He Z. Metformin reverses multidrug resistance and epithelial-mesenchymal transition (EMT) via activating AMP-activated protein kinase (AMPK) in human breast cancer cells. Mol Cell Biochem. 2014;386:63–71. 47. Jo M, Lester RD, Montel V, Eastman B, Takimoto S, Gonias SL. Reversibility of epithelial-mesenchymal transition (EMT) induced in breast cancer cells by activation of urokinase receptor-dependent cell signaling. J Biol Chem. 2009;284:22825–33. 23. Creighton CJ, Li X, Landis M, Dixon JM, Neumeister VM, Sjolund A, et al. Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci U S A. 2009;106:13820–5. 24. Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelial- mesenchymal transition. Nat Rev Mol Cell Biol. 2014;15:178–96. 48. Bonnomet A, Syne L, Brysse A, Feyereisen E, Thompson EW, Noel A, et al. A dynamic in vivo model of epithelial-to-mesenchymal transitions in circulating tumor cells and metastases of breast cancer. Oncogene. 2012;31:3741–53. 24. Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelial- mesenchymal transition. Nat Rev Mol Cell Biol. 2014;15:178–96. 25. Nieto MA, Cano A. The epithelial-mesenchymal transition under control: global programs to regulate epithelial plasticity. Semin Cancer Biol. 2012;22:361–8. 25. Nieto MA, Cano A. The epithelial-mesenchymal transition under control: global programs to regulate epithelial plasticity. Semin Cancer Biol. 2012;22:361–8. 49. Gunasinghe NP, Wells A, Thompson EW, Hugo HJ. Mesenchymal-epithelial transition (MET) as a mechanism for metastatic colonisation in breast cancer. Cancer Metastasis Rev. 2012;31:469–78. 26. Saxena M, Stephens MA, Pathak H, Rangarajan A. Transcription factors that mediate epithelial-mesenchymal transition lead to multidrug resistance by upregulating ABC transporters. Cell Death Dis. 2011;2:e179. 27. Acknowledgements 14. Visvader JE. Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes Dev. 2009;23:2563–77. The authors gratefully acknowledge Gayle Philip at Victorian Life Sciences Computation Initiative for assistance with processing RNASeq data; Bryce van Denderen at St. Vincent’s Institute for assistance with experimental procedures; Prof. Gregory J. Goodall at South Australia Health for advice and critical review of the manuscript. 15. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003;100:3983–8. Page 19 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 16. Chang CJ, Chao CH, Xia W, Yang JY, Xiong Y, Li CW, et al. p53 regulates epithelial-mesenchymal transition and stem cell properties through modulating miRNAs. Nat Cell Biol. 2011;13:317–23. 40. Yadav V, Denning MF. Fyn is induced by Ras/PI3K/Akt signaling and is required for enhanced invasion/migration. Mol Carcinog. 2011;50:346–52. 41. Hugo HJ, Kokkinos MI, Blick T, Ackland ML, Thompson EW, Newgreen DF. Defining the E-Cadherin Repressor Interactome in Epithelial-Mesenchymal Transition: The PMC42 Model as a Case Study. Cells Tissues Organs. 2011;193:23–40. 17. Sarrio D, Rodriguez-Pinilla SM, Hardisson D, Cano A, Moreno-Bueno G, Palacios J. Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype. Cancer Res. 2008;68:989–97. 18. Cheng Q, Chang JT, Gwin WR, Zhu J, Ambs S, Geradts J, et al. A signature of epithelial-mesenchymal plasticity and stromal activation in primary tumor modulates late recurrence in breast cancer independent of disease subtype. Breast Cancer Res. 2014;16:407. 42. Hugo HJ, Pereira L, Suryadinata R, Drabsch Y, Gonda TJ, Gunasinghe NP, et al. Direct repression of MYB by ZEB1 suppresses proliferation and epithelial gene expression during epithelial-to-mesenchymal transition of breast can- cer cells. Breast Cancer Res. 2013;15:R113. 43. Ackland ML, Newgreen DF, Fridman M, Waltham MC, Arvanitis A, Minichiello J, et al. Epidermal growth factor-induced epithelio-mesenchymal transition in human breast carcinoma cells. Lab Invest. 2003;83:435–48. 19. Fuchs IB, Lichtenegger W, Buehler H, Henrich W, Stein H, Kleine-Tebbe A, et al. The prognostic significance of epithelial-mesenchymal transition in breast cancer. Anticancer Res. 2002;22:3415–9. 44. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. A collection of breast cancer cell lines for the study of functionally distinct cancer subtypes. Cancer Cell. 2006;10:515–27. 20. Willipinski-Stapelfeldt B, Riethdorf S, Assmann V, Woelfle U, Rau T, Sauter G, et al. Acknowledgements Chen H, Zhu G, Li Y, Padia RN, Dong Z, Pan ZK, et al. Extracellular signal- regulated kinase signaling pathway regulates breast cancer cell migration by maintaining slug expression. Cancer Res. 2009;69:9228–35. 50. Ackland ML, Michalczyk A, Whitehead RH. PMC42, a novel model for the differentiated human breast. Exp Cell Res. 2001;263:14–22. 51. Whitehead RH, Bertoncello I, Webber LM, Pedersen JS. A new human breast carcinoma cell line (PMC42) with stem cell characteristics. I. Morphologic characterization. J Natl Cancer Inst. 1983;70:649–61. 28. Li J, Zhou BP. Activation of beta-catenin and Akt pathways by Twist are critical for the maintenance of EMT associated cancer stem cell-like characters. BMC Cancer. 2011;11:49. 52. Whitehead RH, Monaghan P, Webber LM, Bertoncello I, Vitali AA. A new human breast carcinoma cell line (PMC42) with stem cell characteristics. II. Characterization of cells growing as organoids. J Natl Cancer Inst. 1983;71:1193–203. 29. Masuda H, Zhang D, Bartholomeusz C, Doihara H, Hortobagyi GN, Ueno NT. Role of epidermal growth factor receptor in breast cancer. Breast Cancer Res Treat. 2012;136:331–45. 30. Foley J, Nickerson NK, Nam S, Allen KT, Gilmore JL, Nephew KP, et al. EGFR signaling in breast cancer: bad to the bone. Semin Cell Dev Biol. 2010;21:951–60. 53. Whitehead RH, Quirk SJ, Vitali AA, Funder JW, Sutherland RL, Murphy LC. A new human breast carcinoma cell line (PMC42) with stem cell characteristics. III. Hormone receptor status and responsiveness. J Natl Cancer Inst. 1984;73:643–8. 31. El Guerrab A, Zegrour R, Nemlin CC, Vigier F, Cayre A, Penault-Llorca F, et al. Differential impact of EGFR-targeted therapies on hypoxia responses: implications for treatment sensitivity in triple-negative metastatic breast cancer. PLoS One. 2011;6:e25080. 54. Ting-Chao C, Talalay P. Analysis of combined drug effects: a new look at a very old problem. Trends Pharmacol Sci. 1983;4:450–4. 55. Beisser D, Klau GW, Dandekar T, Muller T, Dittrich MT. BioNet: an R-Package for the functional analysis of biological networks. Bioinformatics. 2010;26:1129–30. 32. Wang SJ, Saadi W, Lin F, Minh-Canh Nguyen C, Li Jeon N. Differential effects of EGF gradient profiles on MDA-MB-231 breast cancer cell chemotaxis. Exp Cell Res. 2004;300:180–9. 56. Cowley MJ, Pinese M, Kassahn KS, Waddell N, Pearson JV, Grimmond SM, et al. PINA v2.0: mining interactome modules. Nucleic Acids Res. 2012;40:D862–865. 33. Kim H, Muller WJ. The role of the epidermal growth factor receptor family in mammary tumorigenesis and metastasis. Exp Cell Res. 1999;253:78–87. 34. Acknowledgements Wells A, Kassis J, Solava J, Turner T, Lauffenburger DA. Growth factor- induced cell motility in tumor invasion. Acta Oncol. 2002;41:124–30. 57. Cline MS, Smoot M, Cerami E, Kuchinsky A, Landys N, Workman C, et al. Integration of biological networks and gene expression data using Cytoscape. Nat Protoc. 2007;2:2366–82. 35. Ware MF, Wells A, Lauffenburger DA. Epidermal growth factor alters fibroblast migration speed and directional persistence reciprocally and in a matrix-dependent manner. J Cell Sci. 1998;111(Pt 16):2423–32. 58. Knox C, Law V, Jewison T, Liu P, Ly S, Frolkis A, et al. DrugBank 3.0: a comprehensive resource for ‘omics’ research on drugs. Nucleic Acids Res. 2011;39:D1035–1041. 36. Yang MH, Wu KJ. TWIST activation by hypoxia inducible factor-1 (HIF-1): implications in metastasis and development. Cell Cycle. 2008;7:2090–6. 59. Minafra L, Bravata V, Forte GI, Cammarata FP, Gilardi MC, Messa C. Gene expression profiling of epithelial-mesenchymal transition in primary breast cancer cell culture. Anticancer Res. 2014;34:2173–83. 37. Yang MH, Wu MZ, Chiou SH, Chen PM, Chang SY, Liu CJ, et al. Direct regulation of TWIST by HIF-1alpha promotes metastasis. Nat Cell Biol. 2008;10:295–305. 60. Katoh Y, Katoh M. FGFR2-related pathogenesis and FGFR2-targeted therapeutics (Review). Int J Mol Med. 2009;23:307–11. 38. Zhang L, Huang G, Li X, Zhang Y, Jiang Y, Shen J, et al. Hypoxia induces epithelial-mesenchymal transition via activation of SNAI1 by hypoxia- inducible factor -1alpha in hepatocellular carcinoma. BMC Cancer. 2013;13:108. 61. Jayachandran A, Anaka M, Prithviraj P, Hudson C, McKeown SJ, Lo PH, et al. Thrombospondin 1 promotes an aggressive phenotype through epithelial-to- mesenchymal transition in human melanoma. Oncotarget. 2014;5:5782–97. 39. Cooke VG, LeBleu VS, Keskin D, Khan Z, O’Connell JT, Teng Y, et al. Pericyte depletion results in hypoxia-associated epithelial-to-mesenchymal transition and metastasis mediated by met signaling pathway. Cancer Cell. 2012;21:66–81. 62. Avraham R, Yarden Y. Feedback regulation of EGFR signalling: decision making by early and delayed loops. Nat Rev Mol Cell Biol. 2011;12:104–17. Page 20 of 21 Page 20 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 63. Fujita Y, Komatsu N, Matsuda M, Aoki K. Fluorescence resonance energy transfer based quantitative analysis of feedforward and feedback loops in epidermal growth factor receptor signaling and the sensitivity to molecular targeting drugs. FEBS J. 2014;281:3177–92. development must consider the impact on metastasis. Clin Cancer Res. 2009;15:4529. development must consider the impact on metastasis. Clin Cancer Res. 2009;15:4529. 88. Acknowledgements Control of MAPK signalling: from complexity to what really matters. Oncogene. 2005;24:5533–42. 96. Larive RM, Moriggi G, Menacho-Marquez M, Canamero M, Alava E, Alarcon B, et al. Contribution of the R-Ras2 GTP-binding protein to primary breast tumorigenesis and late-stage metastatic disease. Nat Commun. 2014;5:3881. 72. Dry JR, Pavey S, Pratilas CA, Harbron C, Runswick S, Hodgson D, et al. Transcriptional pathway signatures predict MEK addiction and response to selumetinib (AZD6244). Cancer Res. 2010;70:2264–73. 97. Migliaccio A, Di Domenico M, Castoria G, de Falco A, Bontempo P, Nola E, et al. Tyrosine kinase/p21ras/MAP-kinase pathway activation by estradiol- receptor complex in MCF-7 cells. EMBO J. 1996;15:1292–300. 73. Smyth LA, Collins I. Measuring and interpreting the selectivity of protein kinase inhibitors. J Chem Biol. 2009;2:131–51. 74. Said NA, Williams ED. Growth factors in induction of epithelial-mesenchymal transition and metastasis. Cells Tissues Organs. 2011;193:85–97. 98. Burris 3rd HA. Overcoming acquired resistance to anticancer therapy: focus on the PI3K/AKT/mTOR pathway. Cancer Chemother Pharmacol. 2013;71:829–42. 75. Thomson S, Petti F, Sujka-Kwok I, Mercado P, Bean J, Monaghan M, et al. A systems view of epithelial-mesenchymal transition signaling states. Clin Exp Metastasis. 2011;28:137–55. 99. Lim SM, Westover KD, Ficarro SB, Harrison RA, Choi HG, Pacold ME, et al. Therapeutic targeting of oncogenic K-Ras by a covalent catalytic site inhibitor Angew Chem Int Ed Engl. 2014;53:199–204. 76. Strese S, Fryknas M, Larsson R, Gullbo J. Effects of hypoxia on human cancer cell line chemosensitivity. BMC Cancer. 2013;13:331. 100. Baker NM, Der CJ. Cancer: Drug for an ‘undruggable’ protein. Nature. 2013;497:577–8. 77. Moreno-Bueno G, Portillo F, Cano A. Transcriptional regulation of cell polarity in EMT and cancer. Oncogene. 2008;27:6958–69. 101. Finn RS, Bengala C, Ibrahim N, Roche H, Sparano J, Strauss LC, et al. Dasatinib as a single agent in triple-negative breast cancer: results of an open-label phase 2 study. Clin Cancer Res. 2011;17:6905–13. 78. de Herreros AG, Peiro S, Nassour M, Savagner P. Snail family regulation and epithelial mesenchymal transitions in breast cancer progression. J Mammary Gland Biol Neoplasia. 2010;15:135–47. 102. Garon EB, Finn RS, Hosmer W, Dering J, Ginther C, Adhami S, et al. Identification of common predictive markers of in vitro response to the Mek inhibitor selumetinib (AZD6244; ARRY-142886) in human breast cancer and non-small cell lung cancer cell lines. Mol Cancer Ther. 2010;9:1985–94. 79. Foubert E, De Craene B, Berx G. Key signalling nodes in mammary gland development and cancer. Acknowledgements Candy PA, Phillips MR, Redfern AD, Colley SM, Davidson JA, Stuart LM, et al. Notch-induced transcription factors are predictive of survival and 5-fluorouracil response in colorectal cancer patients. Br J Cancer. 2013;109:1023–30. 64. Shvartsman SY, Hagan MP, Yacoub A, Dent P, Wiley HS, Lauffenburger DA. Autocrine loops with positive feedback enable context-dependent cell signaling. Am J Physiol Cell Physiol. 2002;282:C545–559. 89. Yunokawa M, Koizumi F, Kitamura Y, Katanasaka Y, Okamoto N, Kodaira M, et al. Efficacy of everolimus, a novel mTOR inhibitor, against basal-like triple-negative breast cancer cells. Cancer Sci. 2012;103:1665–71. 65. Han W, Lo HW. Landscape of EGFR signaling network in human cancers: biology and therapeutic response in relation to receptor subcellular locations. Cancer Lett. 2012;318:124–34. 90. McKay MM, Morrison DK. Integrating signals from RTKs to ERK/MAPK. Oncogene. 2007;26:3113–21. 66. Avizienyte E, Frame MC. Src and FAK signalling controls adhesion fate and the epithelial-to-mesenchymal transition. Curr Opin Cell Biol. 2005;17:542–7. 91. Whyte J, Bergin O, Bianchi A, McNally S, Martin F. Key signalling nodes in mammary gland development and cancer. Mitogen-activated protein kinase signalling in experimental models of breast cancer progression and in mammary gland development. Breast Cancer Res. 2009;11:209. 67. Hochgrafe F, Zhang L, O’Toole SA, Browne BC, Pinese M, Porta Cubas A, et al. Tyrosine phosphorylation profiling reveals the signaling network characteristics of Basal breast cancer cells. Cancer Res. 2010;70:9391–401. y g p 92. Shin S, Blenis J. ERK2/Fra1/ZEB pathway induces epithelial-to-mesenchymal transition. Cell Cycle. 2010;9:2483–4. 93. Shin S, Dimitri CA, Yoon SO, Dowdle W, Blenis J. ERK2 but not ERK1 induces epithelial-to-mesenchymal transformation via DEF motif-dependent signaling events. Mol Cell. 2010;38:114–27. 68. Stemke-Hale K, Gonzalez-Angulo AM, Lluch A, Neve RM, Kuo WL, Davies M, et al. An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer. Cancer Res. 2008;68:6084–91. 94. Duncan JS, Whittle MC, Nakamura K, Abell AN, Midland AA, Zawistowski JS, et al. Dynamic reprogramming of the kinome in response to targeted MEK inhibition in triple-negative breast cancer. Cell. 2012;149:307–21. 69. Zhao L, Vogt PK. Class I PI3K in oncogenic cellular transformation. Oncogene. 2008;27:5486–96. 70. Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer. 2002;2:489–501. 95. Miyakis S, Sourvinos G, Spandidos DA. Differential expression and mutation of the ras family genes in human breast cancer. Biochem Biophys Res Commun. 1998;251:609–12. 71. Hornberg JJ, Binder B, Bruggeman FJ, Schoeberl B, Heinrich R, Westerhoff HV. Acknowledgements The Snail1-Twist1 conspiracy in malignant breast cancer progression. Breast Cancer Res. 2010;12:206. 80. Matsuoka J, Yashiro M, Doi Y, Fuyuhiro Y, Kato Y, Shinto O, et al. Hypoxia stimulates the EMT of gastric cancer cells through autocrine TGFbeta signaling. PLoS One. 2013;8, e62310. 103. Saini KS, Loi S, de Azambuja E, Metzger-Filho O, Saini ML, Ignatiadis M, et al. Targeting the PI3K/AKT/mTOR and Raf/MEK/ERK pathways in the treatment of breast cancer. Cancer Treat Rev. 2013;39:935–46. 81. Lo HW, Hsu SC, Xia W, Cao X, Shih JY, Wei Y, et al. Epidermal growth factor receptor cooperates with signal transducer and activator of transcription 3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene expression. Cancer Res. 2007;67:9066–76. 104. Rinehart J, Adjei AA, Lorusso PM, Waterhouse D, Hecht JR, Natale RB, et al. Multicenter phase II study of the oral MEK inhibitor, CI-1040, in patients with advanced non-small-cell lung, breast, colon, and pancreatic cancer. J Clin Oncol. 2004;22:4456–62. 82. Gregory PA, Bracken CP, Smith E, Bert AG, Wright JA, Roslan S, et al. An autocrine TGF-beta/ZEB/miR-200 signaling network regulates establishment and maintenance of epithelial-mesenchymal transition. Mol Biol Cell. 2011;22:1686–98. 105. Gayle SS, Castellino RC, Buss MC, Nahta R. MEK inhibition increases lapatinib sensitivity via modulation of FOXM1. Curr Med Chem. 2013;20:2486–99. 106. Sabnis GJ, Kazi A, Golubeva O, Shah P, Brodie A. Effect of selumetinib on the growth of anastrozole-resistant tumors. Breast Cancer Res Treat. 2013;138:699–708. 83. Chen A, Wong CS, Liu MC, House CM, Sceneay J, Bowtell DD, Thompson EW, Moller A: The ubiquitin ligase Siah is a novel regulator of Zeb1 in breast cancer. Oncotarget 2014; 6(2), 862-873. 107. Normanno N, Campiglio M, Maiello MR, De Luca A, Mancino M, Gallo M, et al. Breast cancer cells with acquired resistance to the EGFR tyrosine kinase inhibitor gefitinib show persistent activation of MAPK signaling. Breast Cancer Res Treat. 2008;112:25–33. 84. Roninson IB. The role of the MDR1 (P-glycoprotein) gene in multidrug resistance in vitro and in vivo. Biochem Pharmacol. 1992;43:95–102. 85. Wartenberg M, Ling FC, Muschen M, Klein F, Acker H, Gassmann M, et al. Regulation of the multidrug resistance transporter P-glycoprotein in multicellular tumor spheroids by hypoxia-inducible factor (HIF-1) and reactive oxygen species. FASEB J. 2003;17:503–5. 108. Yanochko GM, Eckhart W. Type I insulin-like growth factor receptor over-expression induces proliferation and anti-apoptotic signaling in a three-dimensional culture model of breast epithelial cells. Breast Cancer Res. 2006;8:R18. 86. Cursons et al. Cell Communication and Signaling (2015) 13:26 112. Heiser LM, Sadanandam A, Kuo WL, Benz SC, Goldstein TC, Ng S, et al. Subtype and pathway specific responses to anticancer compounds in breast cancer. Proc Natl Acad Sci U S A. 2012;109:2724–9. 111. Project TCGA. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61–70. 110. Chin YR, Yoshida T, Marusyk A, Beck AH, Polyak K, Toker A. Targeting Akt3 signaling in triple-negative breast cancer. Cancer Res. 2014;74:964–73. 111. Project TCGA. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61–70. 112. Heiser LM, Sadanandam A, Kuo WL, Benz SC, Goldstein TC, Ng S, et al. Subtype and pathway specific responses to anticancer compounds in breast cancer. Proc Natl Acad Sci U S A. 2012;109:2724–9. 110. Chin YR, Yoshida T, Marusyk A, Beck AH, Polyak K, Toker A. Targeting Akt3 signaling in triple-negative breast cancer. Cancer Res. 2014;74:964–73. Acknowledgements Nie F, Yang J, Wen S, An YL, Ding J, Ju SH, et al. Involvement of epidermal growth factor receptor overexpression in the promotion of breast cancer brain metastasis. Cancer. 2012;118:5198–209. 109. Mirzoeva OK, Das D, Heiser LM, Bhattacharya S, Siwak D, Gendelman R, et al. Basal subtype and MAPK/ERK kinase (MEK)-phosphoinositide 3-kinase feedback signaling determine susceptibility of breast cancer cells to MEK inhibition. Cancer Res. 2009;69:565–72. 87. Steeg PS, Anderson RL, Bar-Eli M, Chambers AF, Eccles SA, Hunter K, et al. An open letter to the FDA and other regulatory agencies: Preclinical drug Page 21 of 21 Cursons et al. Cell Communication and Signaling (2015) 13:26 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit
https://openalex.org/W3019250573	https://www.revistas.udesc.br/index.php/urdimento/article/download/1414573101372020085/11350	Italian	null	Insufflare luce, animare figure	Urdimento	2,020	cc-by	11,071	Insufflare luce, animare figure1 Insufflare luce, animare figure1 DOI: http:/dx.doi.org/10.5965/1414573101372020085 DOI: http:/dx.doi.org/10.5965/1414573101372020085 DOI: http:/dx.doi.org/10.5965/1414573101372020085 Parole-chiave: Luce; figure; animazione; atmosfera Palavras-chaves: Luz; figura; animação; atmosfera Parole-chiave: Luce; figure; animazione; atmosfera Palavras-chaves: Luz; figura; animação; atmosfera Parole-chiave: Luce; figure; animazione; atmosfera Palavras-chaves: Luz; figura; animação; atmosfera Insufflare luce, animare figure Inspirar luz, animar figuras Blowing Light, animating Puppets Cristina Grazioli 2 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 85 Abstract The article considers the assumption of the potential of “animation” by light, proposing a survey of some of the possible meanings of the relationship between light and puppet. Different examples of this landscape are approached, from the consonances between the two territories, to the more intimate relationships on a dramaturgical level, to the concept of a “performer” made of light. Keywords: Lighting; puppetry; animation; atmosphere E-ISSN: 2358.6958 E-ISSN: 2358.6958 E-ISSN: 2358.6958 1 Il presente articolo muove da un breve contributo pubblicato in Poétiques de l’illusion. Dialogues contemporains entre marionnette et magie, Alternatives Théâtrales: Liège 2018, pp. 86-89; viene qui integrato da altri materiali che sono stati oggetto di seminari e lezioni (tra gli altri, presso Sorbonne Nouvelle Paris3 nel 2009/2010; Unirio, Rio de Janeiro, agosto 2015; Université Paul Valéry di Montpellier, febbraio 2016). 2 Professore Associato all’Università di Padova (I). Le sue principali linee di ricerca si concentrano sulle relazioni tra Teatro e Arti Visive, su tedesca del primo Novecento, sull’Estetica della Marionetta, sull’Illuminazione scenica. Professora doutora associada, Università di Padova (I). As suas principais áreas de pesquisa se se focam nas relações entre Teatro e Artes Visuais, dramaturgia alemã do início do século XX, Estética da Marionete e Iluminação Cênica. Professor at the University of Padua (I). The lines of investigation of her researches focus on the relationship between Theatre and Visual Arts, German Drama at the beginning of the 20th century, Aesthetics of the Marionette, Lighting in the Theatre. 86 86 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Resumo O artigo segue o pressuposto do potencial de “animação” pela luz, ofere- cendo um reconhecimento de possíveis significados da relação luz / figura. Diferentes exemplos desta paisagem serão abordados: desde as consonâncias entre os dois territórios, passando pelas relações dramatúrgicas mais íntimas, até o conceito de “atores de luz”. L’articolo muove dal presupposto della potenzialità di “animazione” da parte del fattore luminoso, offrendo una ricognizione di alcune delle possibili accezioni della relazione luce/figure. Si offrono esempi diversi di questo paesagio: dalle consonanze tra i due territori, alle più intime relazioni a livello dramma-turgico, alle concezioni di “attori di luce”. Keywords: Lighting; puppetry; animation; atmosphere Les mains de lumière. Un attore irradiante luce L’interrogazione sulla natura e sulle manifestazioni della luce si pone inscindi- bilmente insieme alla questione stessa delle possibilità del vedere, della scelta di una angolazione del visibile, del nascondere, del rivelare o suggerire, dell’evocare e dello smascherare. Visione e sguardo non ne possono prescindere. Eppure, o forse proprio per la sua macroscopica evidenza, tra i codici dello spettacolo la luce è l’elemento meno studiato nella sua specificità. Solo negli ultimi anni si nota un’attenzione mirata in tal senso3. A maggior ragione rarissime sono le ricognizioni sull’illuminazione nel conte- sto dell’universo delle figure (marionette, burattini, ombre e tutti i generi specifici da questi derivati); pure questo un territorio che è oggetto di una rinnovata attenzione in tempi recenti, anche in virtù di uno spostamento ed ampliamento di statuto della “Fi- gura”4, del quale è necessario tenere conto per le riflessioni che qui proponiamo. Ad una attenta considerazione e accostamento di questi due temi, luce e figure, si rivela una complessa rete di legami e di complicità – e non solo in relazione alle estetiche del contemporaneo. È significativo che molti anni fa Didier Plassard intitolasse Les mains de lumière una preziosa antologia di testimonianze sull’arte delle marionette (Plassard, 1996). L’espressione era presa a prestito da Henri Gouhier. Ci sembra utile qui ritorna-re alle riflessioni del filosofo. Nel 1952 Gouhier pubblicava Le théâtre et l’existence (Gouhier, 2004), un saggio dove sviluppava gli interrogativi proposti in L’Essence du théâtre (1943; Gouhier, 1968), formulando importati considerazioni sulle relazioni tra la scena e il pensiero filosofico. Una parte del capitolo L’existence sur la scène è de-dicata alla marionetta (La Marionnette, Gouhier, 1968, p. 119-127). Il discorso poggia sulla convinzione che la marionetta non sopprima l’attore, bensì ne sia una forma particolare di esistenza, o meglio di “presenza”. Il marionettista è, a tutti gli effetti, attore (cosa oggi ovvia ma forse non così evidente nel 1952). Son rôle n’est nullement comparable à celui du manipulateur qui fait marcher la lanterne magique, mécanicien silencieux et en dehors du jeu. Que sa main tire les ficelles ou serve de corps à la poupée, elle est une main intelligente qui donne la vie par cette intelligence en même temps que le mouvement. Que sa voix parle au-dessus ou au-dessous des comédiens de bois et de leur comé- die, elle est une voix humaine qui donne une âme par cette humanité en même tant que l’animation. Insufflare luce, animare figure Insufflare luce, animare figure 4. Cfr. in merito l’importante lavoro di Julie Postel, Présences de la Marionnette contemporaine : figure, figuration, défiguration, Thèse de Doctorat dirigée par Amos Fergombé, Université Polytechnique Hauts-de-France, Valenciennes, 2019; anche la tesi di laurea magistrale di Lorenzo Diofili, Corpi, figure e spazio: la poetica di Gideon Obarzanek nel panorama delle arti performative dalle avanguardie storiche al XXI secolo, relatrice Cristina Grazioli, Università degli Studi di Padova. Insufflare luce, animare figure E più oltre: […] de l’acteur il ne reste qu’une présence sans contours et presque imper- sonnelle, chaleur rayonnant d’une source cachée, lumière tombée d’une étoile anonyme. L’imagerie visuelle attache le spectateur au spectacle et, pour lui, Guignol existe plus que la main sans laquelle Guignol n’existerait pas. (Gouhier, 1968, p. 123)6 . L’effetto che esercita la marionetta non sopprime la presenza dell’attore, ma provoca uno scarto del suo corpo rispetto alla dimensione dello spettacolo, così che le forme e il volto che hanno dotato di vita l’immaginario si fanno dimenticare. La marionetta è «fantôme» (Gouhier, 1968, p. 125), si divide tra esistenza fisica e immaginaria. Una diversa «etica professionale» emerge con questo attore nascosto dalle mani di luce («aux mains de lumière» (Gouhier, 1968, pp. 125-1267). «Azione che irraggia presenza», «calore irradiante da una fonte nascosta», «luce caduta da una stella anonima», «l’attore dalle mani di luce»: l’ambito semantico uti- lizzato da Gouhier emana luce. Certo Gouhier non intende parlare letteralmente dell’illuminazione scenica, ma la terminologia e l’immaginario messi all’opera sono significativamente quelli della luce. Il nostro sguardo di oggi ne ricava utili suggestioni e conferme. Les mains de lumière. Un attore irradiante luce C’est dire qu’une présence est intérieure à l’action et que l’action fait rayonner une présence; c’est dire que le théâtre de marionnettes exige, lui aussi, la présence active de l’acteur. (Gouhier, 1968, p. 120)5. 3 Ne sono il segno progetti dispiegati negli ultimi anni: Lumière de Spectacle, Université di Lille; Dire Luce. Le parole e le cose che illuminano la scena, Università di Padova. 4. Cfr. in merito l’importante lavoro di Julie Postel, Présences de la Marionnette contemporaine : figure, figuration, défiguration, Thèse de Doctorat dirigée par Amos Fergombé, Université Polytechnique Hauts-de-France, Valenciennes, 2019; anche la tesi di laurea magistrale di Lorenzo Diofili, Corpi, figure e spazio: la poetica di Gideon Obarzanek nel panorama delle arti performative dalle avanguardie storiche al XXI secolo, relatrice Cristina Grazioli, Università degli Studi di Padova. 5. «Il suo ruolo non è per nulla paragonabile a quello dell’operatore di lanterna magica, tecnico silenzioso ed esterno alla recitazione. Che la sua mano tiri i fili o serva come corpo del burattino, è una mano intelligente che dà la vita grazie alla sua intelligenza e allo stesso tempo il movimento. Che la sua voce parli al di sopra o al di sotto degli attori di legno e della loro commedia, è una voce umana, che dà un’anima attraverso questa umanità e contemporaneamente all’animazione. Come dire che una presenza è interiore all’azione e che l’azione fa irradiare una presenza; come dire che il teatro di marionette esige, anch’esso, la presenza attiva dell’attore» (traduzione nostra). 87 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 6 «dell’attore non resta che una presenza senza contorni e quasi impersonale, calore irradiante proveniente da una fonte nascosta, luce caduta da una stella ano- nima. L’immaginario visivo cattura lo spettatore dentro lo spettacolo, e per lui Guignol esiste più della mano senza la quale Guignol non esisterebbe» (traduzione nostra). 9 Scritto a Firenze nel 1907. 7 Inoltre considera essenziale «la metamorphose qui assure la presence», ivi, p. 126. 8 Cfr. Grazioli, 2013. Punti di riferimento: consonanze Rileggendo poetiche che sono state pietre miliari del Novecento come quelle di Maeterlinck o di Craig, oggi non possono sfuggire le assonanze tra universo della luce e dell’ombra e paradigma della Marionetta; i due autori segnano, diversamente tra loro, in modo decisivo la qualità di questa parentela tra materia della luce e figura8. Alla fine del XIX secolo Maeterlinck scrive drammi “per marionette” e nello stesso momento pubblica una delle più celebri dichiarazioni di appello ala sostituzione dell’attore in carne ed ossa con la marionetta (Menus propos, 1890). L’autore fiammingo fa riferimento a «une ombre, un reflet, une projection de formes symboliques» (Maeterlinck, 1890, p. 335) come equivalenti delle marionette. Craig come è noto ha consegnato al Novecento alcune tra le più fortunate riflessioni tanto sulla marionetta che sulla luce in scena. Nel suo celebre saggio The Actor and the Übermarionette (1908), sovrappone «shades», «shadows», «spirit», «glance», all’idea di Marionetta (Craig, 1908, p. 9)9. 6 «dell’attore non resta che una presenza senza contorni e quasi impersonale, calore irradiante proveniente da una fonte nascosta, luce caduta da una stella ano- nima. L’immaginario visivo cattura lo spettatore dentro lo spettacolo, e per lui Guignol esiste più della mano senza la quale Guignol non esisterebbe» (traduzione nostra). 88 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 88 Cristina Grazioli Cristina Grazioli 10 Cfr. Modernité de Maeterlinck. Denis Marleau, «Alternatives théâtrales», n˚ 73-74, Juillet 2002; se ne veda la recensione di Pierre Piret in https://journals.ope- nedition.org/textyles/829, ultima consultazione 1 dicembre 2019, tuttavia non si sottolinea la dimensione della luce ma della nuove tecnologie. 11 Visto al Festival di Charleville-Mézières 2019, http://lentrouvert.com/lenfant/, ultima consultazione 9 dicembre 2019. 12 Numerosi gli interventi in merito a tale ‘estensione’ della definizione ‘marionetta’, soprattutto in ambito francese. Rinviamo a titolo esemplificativo a: Marionnette, Corps-frontière, études réunis par Hélène Beauchamp, Joëlle Noguès et Élise Van Haesenbroeck, Artois Presse Université, Arras, 2016. Insufflare luce, animare figure Non può essere casuale il fatto che in epoca a noi più vicina nell’allestire i testi di Maeterlinck diversi registi abbiano sfruttato in primo luogo le potenzialità della luce: solo a titolo d’esempio, Claude Régy nella collaborazione con Dominique Bruguière, Denis Marleau con attori smaterializzati grazie alle nuove tecnologie10. Nel recente L’Enfant di Elise Vigneron (Théâtre de L’Entrouvert) (Figura 1) La mort de Tintagiles è tradotto in scena attraverso la chiave dell’instabilità, del fragile confine tra certo e incerto, reale e invisibile, dicibile e indicibile; l’espressione di questa fragilità è affidata alla concezione dello spazio, dove l’immersione nel buio e la creazione del percorso drammaturgico grazie a presenze in luce sono fondamentali e ‘generative’ di immagi-ne drammaturgica11. Figura 1 - L’enfant. Théâtre de l’Entrouvert, 2019. Foto: Christophe Loiseau Regia: Elise Vigneron. Creazione luce e dispositivi di scena: Benoît Fincker Figura 1 - L’enfant. Théâtre de l’Entrouvert, 2019. Foto: Christophe Loiseau Regia: Elise Vigneron. Creazione luce e dispositivi di scena: Benoît Fincker 11 Visto al Festival di Charleville-Mézières 2019, http://lentrouvert.com/lenfant/, ultima consultazione 9 dicembre 2019. 89 89 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure Insufflare luce, animare figure Un artista troppo poco studiato, Richard Teschner, sin dagli anni Venti coniuga ricerca luministica e innovazione nell’arte della marionetta, realizzando concreta- mente questa congiuntura: grazie a raffinatissimi di spositivi di fo nti lu minose e su - perfici riflettenti (specchi), ma anche costruendo le marionette in modo da renderle ricettive alla luce (lasciando gli occhi cavi, per esempio), raggiunge esiti di estrema originalità nel conferire alle figure una presenza misteriosa ed evocatrice di mondi altri (Jamain, 2018). Oggi diversi artisti contemporanei formatisi ‘nella marionetta’ tengono in gran- de considerazione il versante luministico. Ma colpisce anche che un artista ‘della luce’ come Bob Wilson condivida tanti momenti della riflessione su marionette e super- marionette. Se già in un articolo del 2000 se ne suggerivano le assonanze (Gaudé, 2000), è inevitabile a tal proposito citare una delle ultime sue creazioni: la felicissima congiuntura Robert Wilson/Isabelle Huppert in Mary said what she said (2019) fa pen-sare ad una declinazione dell’auspicio craighiano. Demiurgo, Bob Wilson; Superma-rionetta la Huppert; i fili: la luce. E su tutto una voce che abita il corpo della performer talmente in profondità da farsi voce dell’Altro (anche grazie all’’impiego della fonica, del tutto intessuta con la drammaturgia). Insufflare luce, animare figure “fare venire alla luce”; e ancora a “trasmettere movimento”. I legami semantici tra luce, vita, dinamismo, ma anche apparenza di vita (illusione) sono significativi13. Si tratta del punto d’origine, del concetto e della pratica costitutivi di ogni tecni- ca di manipolazione; d’altro canto, e pure per le implicazioni semantiche sopra viste, ‘animare’ richiama anche il momento fondativo della moderna riflessione sulla con- cezione della luce in scena. Punto d’avvio in questo caso è il pensiero imprescindibile di Adolphe Appia: per il regista che sia consapevole di un utilizzo appropriato della luce, scrive il grande teorico della scena nei primi anni del Novecento, «le tableau ne sera donc plus, à aucun stade de sa vision, un agencement de peinture inanimé, mais il sera toujours animé» (Appia, 1983-1992, II, p. 351). La dialettica animato/inanimato è centrale nelle definizioni del ruolo della luce e percorre tante pagine degli scritti dell’artista ginevrino. Corpo dell’attore, oggetti, spazio, scenografia vivono solo se animati dalla luce e dall’ombra. L’elemento architettonico della scala, che sarà di lì a poco declinato da Craig nel celebre progetto The Steps, è emblematico di tale pren- dere vita dello spazio e dell’architettura14, ora divenuti soggetti portatori di dram- maturgia. E allo stesso tempo, conferendo ad un soggetto ‘non umano’ qualità di presenza drammaturgica, il rapporto ‘attivo’ tra architettura e azione rimette in gioco il plesso di questioni legate alle figure. Forte e ricca di conseguenze per ogni pensiero sulla luce a venire, un germe di questa idea di una luce ‘vivificante’ si ritrova già in epoche precedenti, con le dovute distinzioni contestuali e come pronuncia d’eccezione. Pietro Gonzaga, scenografo italiano attivo a San Pietroburgo alla fine del Settecento, scrive in un prezioso saggio sulla inscindibilità tra visione e ascolto nell’opera dello scenografo (1807): […] comme l’air sonore n’opère rien sur le sentiment, s’il n’est pas agité arti- stement par l’adresse du musicien, ainsi la lumière n’est simplement que du jour, si elle n’est pas travaillé, pour ainsi dire, en fantôme par la surface des objets dans lesquels elle frappe, et en est renvoyé à l’œil avec la plus grande vitesse et exactitude […]. Animare cose Date queste coordinate di riferimento, per nulla esaustive degli esempi possibili che dimostrano le consonanze tra mondo delle figure e universo della luce, intra- prendiamo qui il tentativo di tracciare una possibile mappa, necessariamente gene- rica e incompleta, delle zone di sovrapposizione e intersecazione tra questi due ter- ritori, entrambi soggetti a slittamenti e mutamenti, evoluzioni e contaminazioni nel corso del Novecento e, sempre più frequentemente, nel paesaggio contemporaneo. Rileviamo in particolare la complessità del termine «Marionetta», soprattutto a parti- re dalle Avanguardie storiche, nella seconda metà del Novecento e più che mai oggi12, dove questa espressione si fa cerniera di un ventaglio di possibilità che implicano al- terazione della presenza e suo spostamento al di fuori del corpo, nello spazio e nelle relazioni che instaura. Partiremo dalla questione (intesa anche come “quête”) che ci sembra imprescin- dibile per entrambi i territori: l’animazione. Una parola complessa già nella sua radice etimologica. Un termine che sfugge alla presa, con connotazioni differenti a seconda dei contesti. Ci interessa il suo significato più semplice, quello del processo che dà vita a qualche cosa che, di per sé, vivo non è: in molte lingue “animare” significa “dare vita” ma anche “dare l’apparenza della vita”; così come equivale a “donare vita” e insieme 90 90 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli 13 Potremmo aggiungere anche la variante “animista”; si cfr. oltre il riferimento di Max Lébegue. 14 Decisive per il nostro discorso, le affermazioni di Craig circa lo statuto di ‘personaggio’ dell’architettura e di ‘azione’ dei movimenti dello spazio, creati da ombre. 15 «come l’aria sonora non agisce sul sentimento se non è vivificata artisticamente dal musicista, così la luce non è che chiarore se non è rielaborata, per così dire in fantasma dalla superficie degli oggetti che colpisce, e rinviata all’occhio con la più grande velocità ed esattezza […]. Ora, lo scenografo intelligente può trarre il maggior profitto dalla massa di luce che si trova tra lo spettacolo e lo spettatore, se sa preparare artisticamente le superfici delle quinte in modo da rinviarci i raggi luminosi trasformati in immagini illusorie» (traduzione nostra). Insufflare luce, animare figure Il richiamo a Rilke è ben più di una suggestione, dato che il poeta delle Elegie duinesi è uno degli artisti che profondamente hanno riflettuto sulle Figure. Il motivo delle “Cose” come presenze si dispiega, tra l’altro, entro la riflessione sullo spazio at- mosferico, melodia che le amalgama e che è sostanziata dalla luce. Scrive il poeta in un articolo del 1898: Come se, nell’elenco dei personaggi, vi fossero un armadio, uno specchio, un suono e tante cose più sottili e sommesse e ancora più discrete. Nella vita, tutto ha lo stesso valore: e una cosa non val meno di una parola, o di un pro- fumo o di un sogno. Questa pari dignità deve imporsi anche a teatro. (Rilke, 1995, p. 70). 16 Anche le Cose inanimate sono presenze, dotate di capacità di relazione e non basta che due o tre uomini si incontrino perché siano insieme. «Sono come mario- nette i cui fili sono manovrati da mani diverse. Solo quando un’unica mano le muove, le investe dall’alto una ‘comunanza’» che le fa agire (Rilke, 1995, p. 77)17. Or, le décorateur intelligent, pour tirer tout le parti qu’il veut de la masse de lumière qui se trouve entre le spectacle et le specta- teur, s’il sait artistement préparer les surfaces des coulisses de manière à nous renvoyer les rayons lumineux transformés en images illusoires. (Gonzaga, 1807, pp. 52-53) .15 Il soffio che dà la vita di cui parla Appia, invisibile eppure precisamente conce- pito e materialmente plasmabile, ci sembra apparentabile ad una atmosfera, o - per dirla con Rilke – ad una Melodia delle cose, che funge da amalgama tra le presenze sceniche ed è garanzia dell’evento di comunicazione, condivisione dello spazio e del movimento tra queste (corpi, oggetti, spazio) e tra esse e gli spettatori; una melodia che apparenta luce e marionetta. 91 91 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli 16 «Als ob im Personenverzeichnis stünde: ein Schrank, ein Glas, ein Klang und das viele Feinere und Leisere auch. Im Leben hat alles denselben Wert, und ein Ding ist nicht schlechter als ein Wort oder ein Duft oder ein Traum. Diese Gerechtigkeit muβ auch auf der Bühne nach und nach Gesetz werden» (Rilke, 1955- 1966, v. 10, p. 442). 17 «Sie sind wie Marionetten deren Drähte in verschiedenen Händen liegen. Erst wenn eine Hand alle lenkt, kommt eine Gemeinsamkeit über sie» (Rilke, 1955- 1966, v. 10, p. 416). Attori di luce Se i variegati esempi sopra evocati concernono poetiche e concezioni dell’arte scenica, la seconda zona d’intersezione si delinea sulla base di affinità formali, e trova parentele e numerose testimonianze in antecedenti storici, nell’ambito di quelle pre- senze che altrove ho chiamato “attori di luce”: una grande famiglia di figure che, pur nella diversità di significati, modalità, concezioni, prendono forma (e spazio) grazie alla luce o all’ombra. Per rimanere solo alle pratiche spettacolari18, una panoramica vede sfilare per prime le immagini di lanterna magica, utilizzate probabilmente già nel teatro barocco (Grazioli, 2008, p. 99-114; Grazioli 2008a, p. 11-22), insieme alle tante presenze fantasmatiche dei cosiddetti “generi ottici” - non dimentichiamo che la radice di fantasma è φῶς, luce; poi le fantasmagorie ottocentesche e tutti i truc- chi che ne derivano per mettere in scena gli spettri (Pepper’s Ghost, il brevetto più fortunato messo a punto da Henri Dircks negli anni Quaranta del XIX secolo); è noto come Giuseppe Verdi, artista attento alla complessità della messinscena, chiedesse allo scenografo Sanquirico di trovare una soluzione per l’apparizione del fantasma di Banquo in Macbeth, prendendo a modello la fantasmagoria (Conati, 1981); ma anche l’ombromania (giochi d’ombre con le mani); e l’universo sterminato di tante tradizioni dei Teatri d’Ombre, più o meno smaterializzati, più o meno ‘umbratili’, fino al teatro d’ombre contemporaneo. 18 Inesauribili gli esempi quando non limitati ai generi spettacolari ma ampliati a immagini e metafore letterarie o filosofiche, da Platone in su. 92 92 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Cristina Grazioli Insufflare luce, animare figure Si tratta di Marionette? Nelle epoche passate certamente no, tuttavia di una ‘specie’ di presenza scenica differente da quella dell’attore in carne ed ossa; di un’alterazione del corpo (o della sua visio-ne) che sarà una premessa fertile per le poetiche e le pratiche a venire19. Inoltrandosi nel XX secolo, si moltiplicano le figure che premono spingendo fuori dal palcosce-nico l’attore in carne ed ossa nelle scene sognate dalle avanguardie storiche. Non si tratta più di mera similitudine tecnica ma di concezioni che rinnovano lo statuto delle presenze: Prampolini in Scenografia e coreografia futurista (1915, in «La Balza futurista») propone guizzi luminosi e “attori-gas colorati”, For-tunato Depero in Colori anima suoni e rumori in una stanza-cubo azzurra, vuotissima. Quattro «in-dividualità astratte» sono mosse meccanicamente da fili invisibili e ogni «colore» ha una voce con ca-ratteristiche diverse (Depero, 1916). Nelle sintesi fu-turiste le luci sostituiscono i personaggi; Giacomo Balla fa agire solidi di luce colorata in Feux d’artifice e gli artisti al Bauhaus sperimentano i cromatismi della luce in movimento con materiali riflettenti che trasfigurano i corpi (e come non ricordare, anche se al di fuori delle scene, l’asso-ciazione innescata dagli oggetti di Man Ray (Figura 2), metamorfosati in umani dalle ombre, o dalla lampadina che ‘interpreta L’Americaine di Picabia?). Figura 2 - Man Ray, L’homme, 1918. Collezione privata (Vera et Arturo Schwarz, Milan) Figura 2 - Man Ray, L’homme, 1918. Collezione privata (Vera et Arturo Schwarz, Milan) 19 Non a caso i doppi virtuali della compagnia canadese Lemieux-Pilon vengono definiti «fantasmagoria tecnologica»; cfr. Maurin, 1996; nello stesso numero di Puck, utile nel suo insieme per i temi che stiamo affrontando, si veda anche Béatrice Picon-Vallin, La scène, l’acteur et ses doubles projetés, pp. 32-38. 93 93 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Insufflare luce, animare figure Vanno di pari passo le innovazioni nel contesto del Teatro di Regia, soprattutto in Germania: a partire dagli anni Venti le proiezioni (di immagini fisse o di immagini in movimento) cominciano ad assolvere la funzione di rendere una qualità diversa della presenza scenica; vengono dunque utilizzate con una motivazione drammaturgica. Piscator offre numerosi esempi in tal senso: basterà qui ricordare che per la messin-scena di Fahnen di Paquet (1924) alla Berliner Volksbühne, il prologo vede una figura ‘epica’ che presenta i personaggi: Piscator in un primo momento aveva pensato di farli sfilare nella forma di marionette, poi invece traduce quest’idea nella soluzione della proiezione cinematografica, poi ancora per ragioni tecniche risolve con diapo-sitive, così che il presentatore indica le loro fotografie sullo schermo (Mildenberger, 1961, p. 195) (un po’ come gli antichi contastorie). Il film, o comunque l’immagine di luce proiettata, serve qui evidentemente a rendere una qualità di presenza differente (ed è significativo che Piscator esiti tra proiezione e marionette20). Procedendo verso la metà del Novecento non possiamo prescindere da un ma- estro della luce che concepisce la proiezione filmica in scena come modalità di met- tere in discussione la presenza univoca del personaggio, di ‘alterarne’ l’identità tra- mite la moltiplicazione: nella sperimentazione sulla Laterna Magika (questo il nome del celebre dispositivo che poi designerà il suo teatro), Joseph Svoboda crea doppi luminosi ed eterei degli attori in scena. Da quel momento in poi, gli esempi non si contano, vuoi per avanzamenti delle tecnologie, vuoi per mutamenti delle poetiche e delle drammaturgie: impossibile elencare tutte le immagini novecentesche di pre- senze in assenza, in bilico sulla soglia tra visibile e invisibile: immagini video che in- teragiscono con gli attori (Giorgio Barberio Corsetti nella collaborazione con Studio Azzurro), raffinatissime presenze rese possibili in trame di luce dalle nuove tecnologie (un esempio di rara poesia: le creazioni di Shirazeh Houshiary e le luci di Lucy Car-ter per Amu (Figura 3), di Wayne McGregor, 200521. O ancora la scena multimediale di Robert Lepage, «corpo espanso» grazie a luce e a tecnologie22 o l’uso del Digital painting, per esempio nelle creazioni di Tam Teatromusica23 (Figura 4). 21 Si veda l’animazione Veil, utilizzata nello spettacolo con Wayne McGregor, http://www.shirazehhoushiary.com/animation/#project3, ultima consultazione 9 di-cembre 2019. 22 Cfr. http://www.annamonteverdi.it/digital/integrazione-cinematogra ica-nel-teatro-di-robert-lepage-le-polygraphe/ ultima consultazione 9 dicembre 2019. 23 22 Cfr. http://www.annamonteverdi.it/digital/integrazione-cinematogra ica-nel-teatro-di-robert-lepage-le-polygraphe/ ultima consultazione 9 dicembre 2019. 23 Tra i numerosi esempi, si veda Scritto dentro, https://nuovoteatromadeinitaly.sciami.com/tam-teatromusica-scritto-dentro-2013/, ultima consultazione 10 dicembre 2019. 23 Tra i numerosi esempi, si veda Scritto dentro, https://nuovoteatromadeinitaly.sciami.com/tam-teatromusica-scritto-dentro-2013/, ultima consultazione 10 dicembre 2019. 94 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 94 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 20 Rinvio a Grazioli 2017. Cristina Grazioli Cristina Grazioli Insufflare luce, animare figure Insufflare luce, animare figure Insufflare luce, animare figure Figura 3 - Amu, Company Wayne McGregor (previously Wayne McGregor \| Random Dance), 2005. Photo: Ravi Deepres; Wayne McGregor, Shirazeh Houshiary, Lucy Carter Figura 3 - Amu, Company Wayne McGregor (previously Wayne McGregor \| Random Dance), 2005. Photo: Ravi Deepres; Wayne McGregor, Shirazeh Houshiary, Lucy Carter Figura 4 - Scritto Dentro, Pierangela Allegro e Michele Sambin, TAM Teatromusica, 2013. Photo: Claudia Fabris Figura 4 - Scritto Dentro, Pierangela Allegro e Michele Sambin, TAM Teatromusica, 2013. Photo: Claudia Fabris 95 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Drammaturgie: la presenza di figure assenti A livello di concezione drammaturgica, riferimento imprescindibile è Beckett, uno degli autori a nostro avviso più profondamente consonanti con tante poetiche contemporanee della marionetta24; allo stesso tempo Beckett in molti suoi testi co-struisce azione drammatica grazie a personaggi che incarnano l’assenza, resa perce-pibile grazie alla luce. Pensiamo a Neither, Rockaby, oppure a racconti come Sans, Le Depepleur o l’emblematico Pas moi, testi dove magnifici ‘vuoti’ di personaggi ven-gono riempiti di sostanza luminosa, o la cui esistenza viene resa percepibile in modo indissociabile dal buio. Innumerevoli le compagnie di teatro di figura che si sono ispi-rate al drammaturgo irlandese, ma anche gli artisti che hanno esaltato tale dimen-sione incerta della presenza grazie a raffinate concezioni della luce. Nella “video-ambientazione” Neither (Figure 5-5B) di Studio Azzurro (2004) «a essere messa in scena è un’assenza» e “personaggio” è la luce stessa. Questa luce-dramma portatrice di azione nella prima parte crea lo spazio della «unspeakable home», nella seconda genera le immagini di cui cancellerà le tracce con abbagli. Le è consustanziale il mo-vimento: «un eterno pendolo da un bagliore all’altro, da un polo all’altro dell’ombra […], in direzione non di una meta ma di un arresto» (Di Marino, 2007, p. 178). Nelle note di regia Paolo Rosa sottolineava l’«essere captati proprio in quel Neither, in quel movimento incessante, assurdo che oscilla dall’una all’altra condizione, senza trovare dimora né dall’una né dall’altra» (Paolo Rosa, 2012, p. 223). Nel corso dell’ouverture orchestrale un fascio di luce scende dall’alto, “personaggio” nel senso sopra detto. Disegna un cerchio e poi si sposta, illumina il nulla (cfr. Di Marino, pp. 177-178; Pitta-luga-Valentini, 2012, pp. 222-241). Questa ricerca si confronta implicitamente con l’i-dea di “figura”, componendola secondo nuove strategie a partire dall’assenza. Appare un paesaggio fluido, indefinibile, quello dell’inconscio che può assumere mille forme o nessuna (Pittaluga-Valentini, 2012, pp. 222-241). 24 Di questo tasso ‘figurale’ dell’opera di Beckett è testimonianza il progetto Beckett & Puppet (Gorizia, 2016), cfr. Marchiori, 2007. 24 Di questo tasso ‘figurale’ dell’opera di Beckett è testimonianza il progetto Beckett & Puppet (Gorizia, 2016), cfr. Marchiori, 2007. 96 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 96 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure Figura 5 - Neither, Studio Azzurro, 2004; photo Studio Azzurro, Archivio Studio Azzurro Figura 5 - Neither, Studio Azzurro, 2004; photo Studio Azzurro, Archivio Studio Azzurro Figura 5B - Neither, Studio Azzurro, 2004; photo Studio Azzurro, Archivio Studio Azzurro Figura 5B - Neither, Studio Azzurro, 2004; photo Studio Azzurro, Archivio Studio Azzurro 97 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 97 Cristina Grazioli Curieuse figure vide et trouée, au nom improbable et sans fonction, mais qui vit intensément l’expérience de l’immobilité, bloquée sur le seuil de son épo-que. C’est comme s’il n’existait pas encore tout à fait et pourtant, comme on dit, il se pose là, trouble, dérange, fait obstacle26. Urdimento Florianópolis v 1 n 37 p 85 115 mar/abr 2020 Cristina Grazioli 27 Un trasparente evoca il colore della pittura di Rothko, di nuovo un pittore ‘della luce’. 28 http://www.getty.edu/art/collection/objects/34708/rudolph-binnemann-stabetanz-of-oskar-schlemmer-german-about-1927/ 29 Ne citiamo solo alcuni sui quali abbiamo spesso riflettuto in occasioni di seminari, in quanto la loro opera è legata alla dimensione performativa: Claudio Bernar-dini, Fabrizio Corneli, Olafur Eliasson, William Kentridge, Mario Martinelli, Maurizio Nannucci, Alfredo Pirri, Simon Touaux, James Turrell… 30 Performance recenti sono Pleine nuit, Parigi, Opéra Comique, 2016 (Christian Boltanski, Jean Kalman et Franck Krawczyk); gli stessi artisti presenteranno Fosse, programmata per gennaio 2020, sempre all’Opéra Comique. Insufflare luce, animare figure La luce ritaglia figure estratte da testi beckettiani, ma il pensiero corre anche ad immagini ricorrenti nella sperimentazione del Teatro Immagine (o dell’immaginario surrealista che lo ispirava). L’ombra della sedia a dondolo (unico oggetto reale) vuo- ta, in movimento «fa scoprire immagini sempre diverse» (Pittaluga-Valentini, 2012, p. 235). L’immagine di un topo illuminato da un fascio di luce scompare per lasciare spazio (o “lasciare il vuoto”) al soprano, unica voce dell’opera, che affiora lentamente dalla buca dell’orchestra, nel buio, imprigionata come tante figure beckettiane25. Sono temi che risuonano nell’ultimo spettacolo di Berangère Vantusso, Alors Carcasse (2019) (Figura 6). Motivo drammaturgico del bel testo di Mariette Navarro (2011) è un essere dalla presenza incerta, Carcasse, che è ovunque e in nessun luogo, strana figura caratterizzata dalla mancanza: ‘manque’ di contorno, di evoluzione, an- che di caduta nell’oblio al di là della soglia sulla quale tenta di percepirsi. Figura 6 - Alors Carcasse, Compagnie Trois-Six-Trente, 2019. Regia: Berangère Vantusso, Luci Jacob. Photo Ivan Boccara Figura 6 - Alors Carcasse, Compagnie Trois-Six-Trente, 2019. Regia: Berangère Vantusso, Luci Jacob. Photo Ivan Boccara 25 Vale la pena almeno ricordare come lo stesso testo sia il nucleo attorno a cui ruota un’altra creazione recente, quella di Romeo Castellucci presentata nell’am-bito della Ruhrtriennale del 2014, allestimento dalle dimensioni monumentali dove l’ampio spazio della Jahrhunderthalle di Bochum viene inondato di sostanza luminosa. Beckett costituisce una fonte anche per Kris Verdonck, che sembra risolvere la dicotomia umano/non-umano nell’impiego di proiezioni in strettissima interazione con le presenze sceniche. Ancora significativamente Le depepleur compare come ‘personaggio’ nel bel saggio di Georges Didi- Huberman dedicato a James Turrell, artista ‘imprescindibile’ per ogni discorso sulla luce nel contemporaneo (Didi-Huberman, 2001); cfr. Grazioli, 2012. 26 Berangère Vantusso, foglio di sala; «Curiosa figura vuota e bucata, dal nome improbabile e senza funzione, ma che vive intensamente l’esperienza dell’immo-bilità, bloccato sulla soglia del suo tempo. È come se non esistesse ancora del tutto e tuttavia, come si dice, si mette lì, disorienta, disturba, ostacola»; spettacolo visto al Festival di Charleville-Mézières 2019. 98 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli 27 Un trasparente evoca il colore della pittura di Rothko, di nuovo un pittore ‘della luce’. la morte; temi che incontrano naturalmente le dimensioni del buio, dell’ombra, del chiaroscuro. Dal 1985 le opere di Boltanski sono ‘autoilluminate’, l’artista rinuncia alla luce dall’alto e immerge lo spazio nella penombra; la luce fa parte integrante dei suoi lavori. Significativamente, nello stesso momento Boltanski espone i suoi primi Théâtres d’ombres (Figura 8) (Biennale di Parigi del 1985), sono l’inizio di una apertura del suo lavoro: il teatro d’ombre è le-gato «à l’idée de la mort, des fantômes, de la disparition. D’ailleurs, il suffit d’allumer la lumière et il n’y a plus rien» (Boltanski, 2019, p. 44)31.é Figura 8 - Théâtre d’ombres, Christian Boltanski, 1984-1997, opera tridimensionale, installazione: luce, proiezione di 20 figurine su parete. Fi- gurine: cartone, proiettori, piattoforma mobile, struttura in metallo, ventilatori, dimensioni variabili. Photo © André Morain; © Adagp, Paris, 2019 Figura 8 - Théâtre d’ombres, Christian Boltanski, 1984-1997, opera tridimensionale, installazione: luce, proiezione di 20 figurine su parete. Fi- gurine: cartone, proiettori, piattoforma mobile, struttura in metallo, ventilatori, dimensioni variabili. Photo © André Morain; © Adagp, Paris, 2019 Insufflare luce, animare figure Vantusso sceglie di incarnarlo grazie alla ‘dispersione’, tramite mezzi che ne dis- seminano la presenza nello spazio. Così allo spettatore viene mostrato un luogo dove la presenza invisibile si pone nella fluidità di una forma di presenza diffusa, senza mai poter essere collocata. Si evoca la Danza dei bastoni (Figura 7) di Oskar Schlemmer con il motivo delle direttrici delle linee di movimento che marcano l’interazione tra corpo del danzatore e leggi dello spazio: ma laddove per Schlemmer le forme che traducevano le linee del movimento del corpo tendevano a centralizzazione ed unità, qui Carcasse è involucro di una presenza che non può essere definita da coordinate certe. È come se la scena si configurasse rovesciata, in negativo (nel senso fotografi-co del termine), mostrandoci nel buio le tracce e il calco di questi percorsi27. Figura 7 - Stäbetanz (Danza dei Bastoni), Oskar Schlemmer, 1927 ca. Photo: Rudolph Binnemann28 Figura 7 - Stäbetanz (Danza dei Bastoni), Oskar Schlemmer, 1927 ca. Photo: Rudolph Binnemann28 Se il dialogo della marionetta con le arti figurative riceve un impulso decisivo con la sperimentazione delle Avanguardie Storiche, nell’ultimo mezzo secolo i nessi si stringono e anche le metamorfosi in corpi di luce e d’ombra affascinano sempre più gli artisti visivi29. In questo caso traiamo l’esempio dall’occasione della recentissima mostra di Christian Boltanski, un’installazione che consente di vivere la dimensione performati-va della sua opera (di lunga data la collaborazione con Jean Kalman30, maestro della creazione luci). Faire son temps accoglie significativamente il visitatore facendogli at-traversare uno spazio di teatro d’ombre. Luce e figure si alleano per dare forma – una forma che ha il flusso mutevole delle memoria – ai motivi centrali della sua poetica (Cfr. Boltanski 2019; Boltanski 1991): la traccia del vissuto, i suoi fantasmi veicolati dall’immagine di luce (fotografia), il confronto con 99 99 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure 32 Inteso come categoria estetica e procedimento di composizione; cfr. Grazioli, 1999. 31 «all’idea della morte, dei fantasmi, della scomparsa. Tuttavia, basta spegnere la luce e non c’è più nulla». Il montaggio: ritagliare il visibile Un’ulteriore prospettiva di lettura possibile delle sovrapposizioni tra marionette e luce è quella dispiegata dal principio del montaggio. Categoria antica, che sem- bra correre sotterranea parallelamente a forme e procedimenti di narrazione lineare, sovvertendoli, scompaginando la logica della compostezza e della composizione del discorso logico. Un procedimento intrinseco alla figura stessa di Marionetta, in questo senso esempio del Grottesco32, composta di pezzi assemblabili a piacere e le cui potenzia- lità si rivelano proprio quando la logica di ‘riproduzione’ viene alterata o sovvertita. Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 100 Cristina Grazioli Cristina Grazioli Insufflare luce, animare figure Parente stretto del collage nella sua declinazione novecentesca, a livello forma- le (e tecnico) il procedimento viene reinventato e fatto proprio dal cinema33, che lo declina nel mezzo ‘luministico’ che gli è consustanziale, e diventa a sua volta terreno al quale attinge un ampio versante teatrale a dominante visiva. Anche in questo caso gli esempi potrebbero essere molto variegati. Nell’ambito del Nuovo Teatro italiano tra anni Sessanta e Settanta, spesso la sperimentazione passa attraverso le nostre due categorie, la ricerca di mezzi alter- nativi alla presenza tradizionale dell’attore e l’indagine sul mezzo luminoso e proiet- tivo (strategie antitetiche a due colonne portanti del teatro della tradizione: il testo e l’attore). Pensiamo ad un artista come Mario Ricci (Figura 9), che inizia accanto al marionettista Michael Meschke; ammira i congegni meccanici di Harry Kramer e si inventa un originale teatro di figure; ma allo stesso tempo sperimenta sulle potenzia- lità della proiezione di luce di decostruire il personaggio, o viceversa di moltiplicarne l’immagine in modo vorticoso34. In Edgar Allan Poe (1967), il film come elemento sce- nico assume un’importanza decisiva. Proiettate su di un supporto fisso e frantumate, le immagini cinematografiche mostrano «ciò che gli spettatori avevano già visto e quello che avrebbero visto», nei particolari più intimi: dettagli di mani, di volti e de- gli oggetti utilizzati. Il movimento deriva «dalla sovrapposizione e frantumazione di immagini cinematografiche e immagini reali» (Ricci, 1977, pp. 225, 227-228)35 degli attori. La tridimensionalità dello spazio trasforma il mezzo filmico in codice teatrale. Figura 9 - I Viaggi di Gulliver, Mario Ricci, 1966. Photo: Pietro Galletti Figura 9 - I Viaggi di Gulliver, Mario Ricci, 1966. Photo: Pietro Galletti 33 I legami tra cinema e marionette sono fertili sin dalla prima metà del Novecento; in merito cfr. «Puck. La marionnette et les autres arts», Les marionnettes au cinéma, 15, 2008. 34 Rinviamo ai focus relativi presenti nel progetto Nuovo Teatro Made in Italy, all’interno del portale Sciami https://nuovoteatromadeinitaly.sciami.com/mario-ricci- -biografia-opere/. 35 cfr. anche Sinisi, 1983, p. 84. -biografia-opere/. 101 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 33 I legami tra cinema e marionette sono fertili sin dalla prima metà del Novecento; in merito cfr. «Puck. La marionnette et les autres arts», Les marionnettes au cinéma, 15, 2008. 34 Rinviamo ai focus relativi presenti nel progetto Nuovo Teatro Made in Italy, all’interno del portale Sciami https://nuovoteatromadeinitaly.sciami.com/mario-ricci- -biografia-opere/. 35 cfr. anche Sinisi, 1983, p. 84. Insufflare luce, animare figure In Salomé (1965) e ne I viaggi di Gulliver (1966) l’ombra è utilizzata come pre- senza attorica “alterata”. Il film scorre sull’attore e crea «un effetto di sovrimpressione distruggendo la consistenza piena del corpo umano, ridotto […] alla condizione fan- tasmatica di un’ombra»; si crea uno scarto tra la figura reale e la trasparenza incor- porea della proiezione: le immagini filmiche affiorano imperfette, debordando oltre i confini dei corpi e degli oggetti che investono, «con una frantumazione dinamica che invade lo spazio ambientale riplasmandolo sulla scia del flusso luminoso» (Sinisi, 1983, p. 84). La passione per scomposizione e ricomposizione, frantumazione e montaggio, va ricondotta all’interesse di Ricci per l’universo della Marionetta: una presenza sceni- ca che costitutivamente mette in gioco la possibilità di un diverso “montaggio” della figura rispetto all’attore in carne ed ossa. I suoi primi spettacoli: Movimento numero uno per marionetta sola (1962), Spettacolo di tre pezzi (1964), Movimento per mario- netta sola numero due (1964), Movimento uno e due (1965), recano titoli emblematici della sua ricerca intorno agli elementi “strutturali” dello spettacolo36. Anche per l’importanza che gli attribuisce lo stesso Ricci con uno sguardo “au- toretrospettivo”, l’incontro con il teatro di Harry Kramer37 durante il periodo trascorso presso Michael Meschke a Stoccolma assume i tratti di un “evento rivelatore”. Attivo nell’ambito dell’arte cinetica, a partire dal 1955 i suoi brevi spettacoli privi di elementi narrativi, vedono in scena piccole sculture antropomorfe dotate di movimento e sono costruiti su partiture sonore, luministiche e meccaniche di estrema precisione. Dello spettacolo visto a Stoccolma diversi elementi colpiscono Ricci: la composizione per frammenti; il fatto che il protagonista dell’azione sia il movimento stesso della mac- china-scultura cinetica. In uno spazio dalle dimensioni ridotte, i movimenti di queste figure dai colori accesi compongono la partitura spettacolare insieme a sonorità ete- rogenee, alternando momenti di stasi ad altri concitati. Germano Celant ricorda che Kramer in quegli anni lavora tramite «nebulose filiformi, di ferro e acciaio, che am- mettono la presenza di materiali eterogenei tali da dare una immagine fisico-ottica ai suoi lavori»38. La scena dell’artista tedesco è un congegno la cui sostanza dramma- turgica sta nel funzionamento dei materiali scenici, portatori di azione drammatica. Ma nella rievocazione di Ricci spicca un altro particolare: il pubblico partecipa con un’attenzione rara ed è affascinato dal ripetersi «all’infinito degli stessi movimenti, al- terati e alternati da sapientissime luci»39. 36 Spettacolo in tre pezzi viene presentato insieme ad una mostra di marionette alla Galleria Arco d’Alibert di Roma nel 1964, cfr. Mario Ricci, Teatro-rito e teatro-gioco, in Franco Quadri, L’avanguardia teatrale in Italia. Materiali (1960-1976), Torino, Einaudi, 1977, pp. 200-211: 203-204. 37 Cfr. Ricci,1970, pp. 50-55; Kramer crea le sue “marionette” dai primi anni Cinquanta. L’esclusa, vince il Leone d’oro a Venezia nella categoria cortometraggi nel 1961. Sul suo teatro meccanico cfr. Metken, 1989). 38 M. Ricci, Collage per una automitobiogra ia, cit., p. 213. 39 M. Ricci, Teatro-rito e teatro-gioco, cit., p. 201. Cristina Grazioli Cristina Grazioli Insufflare luce, animare figure questi «materiali scenici» il fattore luministico. Tant’è che nel passo immediatamente successivo cita Craig e la sua concezione di scena come dinamismo di luce e colore40. Nello stesso contesto di sperimentazione, Memé Perlini mutua dal cinema l’arte del ‘ritaglio’ al fine di ottenere uno statuto differente dell’immagine attorica: non im- piega la proiezione, ma ottiene l’esito di un montaggio visivo isolando tramite la luce ‘fotogrammi’ di spazio scenico, dettagli di corpi e di oggetti. La composizione della marionetta, regolata dal principio del frammento, dell’inversione e della scomposi- zione, viene a coincidere con il montaggio – anche cinematografico (cfr. Grazioli, 2015). A proposito delle consonanze tra creazione luce per il teatro di figura e proce- dimenti del montaggio, va ricordato il lavoro del brasiliano Renato Machado, A luz montagem (2015), dove il creatore luci racconta la sua esperienza con le compagnie Sobrevento e Pequod, che coniuga pensiero della luce e pratica di illuminazione per le marionette con forti rimandi al mondo del cinema (Machado, 2015). Dunque il movimento, ma attivato dalla luce. Ricci riconosce in questa sua esperienza di spettatore la spinta alla sperimentazione «di certi materiali scenici capaci di ‘movimento’, non esattamente meccanico come nel caso di Kramer». Non crediamo di forzare le parole di Ricci invitando a leggere in Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 102 Cristina Grazioli Cristina Grazioli 42 Rinvio all’intervento a due voci in questa stessa rivista, Plassard, Grazioli, 2018, pp. 64-65. 40 Ivi, pp. 201-202. 41 Per una ricognizione sul buio a teatro cfr. Perruchon, 2016. 42 Rinvio all’intervento a due voci in questa stessa rivista, Plassard, Grazioli, 2018, pp. 64-65. 43 Il già citato Poétiques de l’illusion è dedicato a queste relazioni – tuttavia il ruolo della luce e del buio viene centrato solo in pochi momenti, essendo motivo centrale dell’indagine la Marionetta; come riferimenti per le nostre riflessioni si segnalano in particolare gli interventi di Julie Postel e di Véronique Perruchon. Il nero buio La scelta dell’“inquadratura”, in quanto porzione del visibile da mostrare, nel te- atro di marionette si dà in modo complesso e ricco di variabili, dato che l’attore qui si articola nella doppia presenza di marionettista e figura. La questione della luce, e in modo anche più potente la dimensione del buio, inevitabilmente amplifica la scelta delle diverse possibilità dell’animatore di rendersi visibile o di rimanere nascosto, e in tal caso secondo quali modalità rimanere nel buio oppure varcare lo spazio della luce. Dopo essere uscito allo scoperto nello spazio visibile della scena sempre più nel corso del secondo Novecento, esibendo la fragilità dell’illusione e la permeabilità tra il suo mondo e quello dell’oggetto, il marionettista oggi spesso “rientra” nel buio. Ma questo buio ha da tempo cambiato statuto41: non più buco nero che inghiotte quel che va nascosto, ma condizione ‘altra’ della materia della luce, grembo generatore delle immagini e dei suoni. Il buio ha la sua voce; e in molti casi costituisce la dimora della voce del marionettista. Il costume, un elemento molto concreto, può essere fondamentale come au- silio del rivelare o del nascondere. In questa dinamica tra mostrato e celato, l’abito nero può avere un peso decisivo, corrispondente all’alternanza del ruolo del mario-nettista, tra manipolatore e personaggio42. Il costume diviene una presenza a sé, che si insinua nell’interstizio tra attore e figura. In Talita Kum (2012) (Figure 10, 10B) di Riserva Canini il nero del costume fonde e confonde le presenze dell’attrice e della figura; presenza, manipolazione, personaggi si fanno tutt’uno grazie ad un’unica at-trice. La presenza dell’Altro che si insinua nella sfera dell’identità è fortissima, tanto dal punto di vista fisico e materiale che drammaturgico. La sapienza tecnica unita alla cura drammaturgica pongono lo spettatore constanmente di fronte al dubbio sullo statuto delle presenze: in carne ed Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 103 Cristina Grazioli Cristina Grazioli Insufflare luce, animare figure ossa o in luce o nel ‘buio’. ossa o in luce o nel ‘buio’. Figura 10 - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Figura 10B - Talita Kum, Riserva Canini, 2012. Il nero buio Photo: Domenico Semeraro a tutti i nostri esempi è evidente come l’universo della luce sia permeato che un tempo ne erano comunemente considerati il rovescio: l’ombra o il b n territorio abitato tradizionalmente dal buio e luogo di sperimentazione d quello della magia, conquistato dalla marionetta negli ultimi anni sotto la d Figura 10 - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Figura 10 - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Figura 10 - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Figura 10B - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Figura 10B - Talita Kum, Riserva Canini, 2012. Photo: Domenico Semeraro Da tutti i nostri esempi è evidente come l’universo della luce sia permeato da quelli che un tempo ne erano comunemente considerati il rovescio: l’ombra o il buio. Un territorio abitato tradizionalmente dal buio e luogo di sperimentazione della i Da tutti i nostri esempi è evidente come l’universo della luce sia permeato da quelli che un tempo ne erano comunemente considerati il rovescio: l’ombra o il buio. Un territorio abitato tradizionalmente dal buio e luogo di sperimentazione della luce è quello della magia, conquistato dalla marionetta negli ultimi anni sotto la defi- nizione di Magie Nouvelle43. In Le grand oeuvre (2019) (Figura 11) della compagnia canadese La tortue noire il marionettista è un alchimista che trasmuta le presenze e a sua volta ne è assalito; 104 Cristina Grazioli Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 44 Compagnie Sans Souci, K (2019), se ne veda il teaser: https://vimeo.com/336188979 ultima consultazione il 10 dicembre 2019. Insufflare luce, animare figure campeggia a mezzobusto al centro dello spazio visibile, nel buio. Fenomeni lumini- stici accompagnano la trasmutazione, il buio rende visibile le presenze o le inghiotte, la luce sembra fecondare l’immagine. Figura 11 - Le Grand Oeuvre, La Tortue Noir, 2017. Photo: Patrick Simard Figura 11 - Le Grand Oeuvre, La Tortue Noir, 2017. Photo: Patrick Simard Non può sfuggire nel nostro contesto l’associazione tra “animazione” e “animi- smo”, evocata per esempio da Max Legoubé, artista formatosi in entrambi i territori, nella marionetta e nella Magie nouvelle (Legoubé, 2018, p. 43). Lo spettacolo K (Figure 12, 12B) rinvia per certi versi alle tematiche di Alors Carcasse e del nulla becket-tiano (ma anche della perdita kafkiana di riferimenti univoci), innestate sulla memoria del celebre saggio di Kleist. K è ovunque e in nessun luogo, tutto e nulla. L’incertezza si fa poesia del buio e della luce, sospende lo stato di presenza ad una tessitura di atmosfere.44 44 Compagnie Sans Souci, K (2019), se ne veda il teaser: https://vimeo.com/336188979 ultima consultazione il 10 dicembre 2019. 105 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Questo statuto di-vagante segna una condizione di incertezza. Questo statuto di-vagante segna una condizione di incertezza. Questo statuto di-vagante segna una condizione di incertezza. 47 Cfr. Grazioli, 2011. Insufflare luce, animare figure Figura 12 - K, Cie Sans Souci, 2019. Regia Max Legoubé. Photo Claude Boisnard Figura 12 - K, Cie Sans Souci, 2019. Regia Max Legoubé. Photo Claude Boisnard Figura 12B - K, Cie Sans Souci, 2019. Regia Max Legoubé. Photo Claude Boisnard Figura 12B - K, Cie Sans Souci, 2019. Regia Max Legoubé. Photo Claude Boisnard Secondo Legoubé marionettisti contemporanei e maghi della “magie nouvelle” hanno in comune «l’étrangeté et l’inévidence». Aggiungiamo che il buio è complice di tali aspetti. «L’animisme suppose la multiplicité des manières d’habiter le monde, mais at-tribue à tous les êtres le même genre d’intentionnalité, que nous dirions ‘humaine’». L’artista deve offrire allo spettatore lo spazio di una «divagation» (Legoubé, 2018, p. 43)45. 45 «L’animismo presuppone la molteplicità nelle modalità di abitare il mondo, ma attribuisce a tutti gli esseri lo stesso tipo di intenzionalità, che noi diciamo “uma-na”» (traduzione nostra). Su effetto di assenza, corpo in frammenti e corpo dissimulato, si veda il lucido saggio di Julie Postel (Postel, 2018). 45 «L’animismo presuppone la molteplicità nelle modalità di abitare il mondo, ma attribuisce a tutti gli esseri lo stesso tipo di intenzionalità, che noi diciamo “uma-na”» (traduzione nostra). Su effetto di assenza, corpo in frammenti e corpo dissimulato, si veda il lucido saggio di Julie Postel (Postel, 2018). 106 106 Cristina Grazioli Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 46 Il riferimento è all’“Estetica delle atmosfere”, su cui si segnalano gli studi di Tonino Griffero e di Gernot Böhme. Tanti gli artisti che avremmo voluto qui trattare ma che per motivi di spazio non compaiono; ne citiamo alcuni: Les Anges au plafond (in particolare per l’impiego dell’ombra in White dog), Ilka Schönbein (per la luce che sgretola matericamente l’immagine e il ‘ritaglio’ visivo), Plexus polaire (Yingvild Aspeli, per la potenza visionaria dell’interazione tra luce/buio e figura, n. 1,), Magali Chouinard (per il colore e la matericità del bianco e nero, per esempio in Ame nomade); l’elenco potrebbe continuare. Insufflare luce, animare figure Svaporare l’immagine: statuto incerto della presenza In questo territorio di dialogo continuo tra poetiche e procedimenti della luce e della marionetta contemporanea un’altra zona fertilissima è quella che possiamo definire degli stati di passaggio. Intendiamo le situazioni dove, grazie all’interazione con la luce e con l’ombra, la materia esibisce la propria costituzione frammentaria o in continuo divenire, così da offrire qualità di presenza ‘instabili’. Anche in questo caso, una qualità costitutiva della marionetta a livello di poetica, l’essere entre-deux, concretizzazione di una condizione di metamorfosi, di slittamenti semantici o di tra- sfigurazione, trova corrispondenza in uno degli ambiti più suggestivi e forieri di aper- ture oggi implicati dalla luce. Si tratta di un formidabile terreno dove si confrontano riflessione estetica46, pensiero teatrale, tecniche e pratiche dell’illuminazione scenica. Non è un caso che Didier Plassard parli di “chiaroscuro” come suggestione per defi- nire l’atmosfera dove abita il marionettista (cfr. Plassard, 2018). Numerosi sono gli artisti che interrogandosi sulla dimensione effimera – costi- tutiva dell’arte performativa - trovano risposta nel territorio fluido degli stati di transi- zione, in materiali lavorati attraverso l’impiego della luce e dell’ombra: vapore, fumo, acqua, ghiaccio, frammenti di vetro, tutti gli stati di transizione della materia sono impiegati a rendere l’impermanenza (Elise Vigneron, Impermanence) (Figura 13), la fragilità della coerenza (William Kentridge, Black Box o Return. Daccapo47), l’incertez- za della percezione (Alice Laloy, Sfu.ma.to / Sous ma peau.). 46 Il riferimento è all’“Estetica delle atmosfere”, su cui si segnalano gli studi di Tonino Griffero e di Gernot Böhme. Tanti gli artisti che avremmo voluto qui trattare ma che per motivi di spazio non compaiono; ne citiamo alcuni: Les Anges au plafond (in particolare per l’impiego dell’ombra in White dog), Ilka Schönbein (per la luce che sgretola matericamente l’immagine e il ‘ritaglio’ visivo), Plexus polaire (Yingvild Aspeli, per la potenza visionaria dell’interazione tra luce/buio e figura, n. 1,), Magali Chouinard (per il colore e la matericità del bianco e nero, per esempio in Ame nomade); l’elenco potrebbe continuare. Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 107 Cristina Grazioli 48 Non è forse incongruo mettere in relazione queste zone della marionetta con il pensiero del postumanesimo, un approccio al reale al di là dell’umano: si tratta di territori della cultura contemporanea che sarebbe interessante repportare alla marionetta, ma anche al potere di mutevolezza della prospettiva messo in gioco da luce e ombra; cfr., tra gli altri gli studi di Roberto Marchesini. Insufflare luce, animare figure Insufflare luce, animare figure Figura 13 - Impermanence, Théâtre de l’Entreouvert, 2013. Regia: Elise Vigneron. Photo: Eric Bourret. Figura 13 - Impermanence, Théâtre de l’Entreouvert, 2013. Regia: Elise Vigneron. Photo: Eric Bourret. Elise Vigneron eleva a potenza la sostanza ‘liquida’ dell’arte della marionetta nelle figure di ghiaccio, materiale che interagisce con la luce, destinato a sciogliersi facendosi il mezzo di trasmutazione dell’immagine e della presenza che essa incarna drammaturgicamente; la marionetta svapora e dichiara la sua natura di luce, il proprio statuto prossimo all’invisibile nella materialità di una nebulosa immateriale (Imper- manence, Anywhere). In Sfu.ma.to/Sous ma peau (Figura 14) (Alice Laloy, Cie S’appelle reviens) alla luce è affidato il compito di tradurre l’idea degli stati di passaggio (fisici, emotivi, percet- tivi) nella materialità dell’atmosfera come territorio dell’incertezza della percezione, dunque della nostra conoscenza del mondo, del fluire dei fenomeni, contro la messa a fuoco della linearità del pensiero e dell’univocità del reale. Partendo dall’idea di Sfu- mato in Leonardo e nella pittura, Alice Laloy lavora sulle ambiguità, visive e auditive, della nostra usuale comunicazione. Come anche nel caso di Elise Vigneron, siamo chiamati come spettatori a cogliere frammenti, a mutare prospettive, a leggere il re- ale secondo le infinite possibilità che da sempre ci insegna il corpo della Marionetta. 108 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 108 Cristina Grazioli Insufflare luce, animare figure Figura 14 - Sous ma peau / Sfu.ma.to, La Cie S’appelle Reviens, 2015. Regia: Alice Laloy. Photo: Elisabeth Carecchio. Figura 14 - Sous ma peau / Sfu.ma.to, La Cie S’appelle Reviens, 2015. Regia: Alice Laloy. Photo: Elisabeth Carecchio. “Contro la messa a fuoco” lavorano l’allusione, lo ‘sfumato’ e svaporato, lo smar- ginamento, il flou, che richiedono evocazione e completamento dell’immagine gra- zie alla mente e all’occhio dello spettatore (e anche alla sua memoria o emozione). Tuttavia la distinzione tradizionale Illusione / Allusione, sancita storicamente sin dalle pronunce di rinnovamento dell’arte teatrale in ambito simbolista (cioè contro il realismo e la sua evoluzione nel Naturalismo) appare oggi troppo generica: se è vero che la luce netta e il disegno preciso riducono il dubbio sulla veridicità di ciò che co- gliamo con lo sguardo, la materia lavora l’immagine nel senso di uno sgretolamento, riaprendo il varco ad un margine per l’immaginazione. Questa condizione è un prezioso esercizio alla mobilità e alla relativizzazione: la realtà non è univoca, la verità è molteplice, l’umano non è che uno dei tantissimi aspetti di un universo complesso.48 Si tratta di consapevolezze e dubbi da lungo tempo patrimonio delle arti della Marionetta, che le rivela più che mai nelle sue concezioni contemporanee, nell’uscire dal confine del proprio corpo, per sperimentare una presenza diffusa. In tale contesto il lavoro di Renaud Herbin, diversificato in quanto ad esiti formali ma profondamente coeso dal punto di vista della concezione, costituisce un osserva-torio interessante. Prendiamo ad esempio Milieu (2019) (Figure 15 e 16), ispirato an-ch’esso a Beckett (le Depepleur), dove la materia inghiotte la marionetta come sabbia mobile. La luce e il buio sembrano qui servire ad assorbire, con la marionetta, il suo movimento e lo stesso marionettista, a fare il vuoto nel ‘luogo’, un milieu senza cen-tro. Questo motivo è amplificato dall’installazione che ne deriva, Milieu et alentours, versione articolata nello spazio, che mantiene il suo centro nella struttura originaria ma lo mette in discussione tramite altri luoghi, connotati da materiali come sabbia, Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 109 Cristina Grazioli Insufflare luce, animare figure pietra, acqua, specchio: tutti materiali che interagiscono con luce e con l’ombra, spo- stando il centro dentro il fluire di quanto percepiamo, deambulando nello spazio49. pietra, acqua, specchio: tutti materiali che interagiscono con luce e con l’ombra, spo- stando il centro dentro il fluire di quanto percepiamo, deambulando nello spazio49. tro dentro il fluire di quanto percepiamo, deambulando nello s Figura 15 - Milieu, Renaud Herbin, 2017. Photo: Benít Schupp Figura 15 - Milieu, Renaud Herbin, 2017. Photo: Benít Schupp 49 Nella seconda installazione gli spettatori si muovono nello spazio. Cfr. http://www.renaudherbin.com/milieu, ultima consultazione 10 dicembre 2019. Altra crea- zione recente di Renaud Herbin, At the still point of the turning world (Figura 16) fa pensare ad una marionettizzazione dello spazio, suggerendo anche in questo caso lo spostamento da un presunto centro ad una presenza diffusa. 110 Cristina Grazioli Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Insufflare luce, animare figure Figura 17B - Symphonie Fantastique, Basil Twist, 1998-2019. Photo: Marie Tartar Figura 17B - Symphonie Fantastique, Basil Twist, 1998-2019. Photo: Marie Tartar Insufflare luce, animare figure Figura 16 - At the still point of the turning world, Renaud Herbin, 2019. Photo: Benoit Schupp Figura 16 - At the still point of the turning world, Renaud Herbin, 2019. Photo: Benoit Schupp Infine, la materia nel suo stato ‘di passaggio’ o fluido può anche fungere da mez- zo che conferisce alle figure il tempo musicale. Una declinazione giocosa e lievissima di queste tematiche è la Symphonie phan-tastique (Figure 17 e 17B) di Basil Twist. Qui l’acqua, ambiente che impone ai manipo-latori un difficile lavoro (i marionettisti agiscono le figure immerse in una grande vasca d’acqua, grazie a bastoni50), trasfigura grazie alla creazione di luce colorata la presenza materica, unisce figurativo e astratto, immanenza delle fi-gure ed evocazione nel bagliore della luce. Figura 17 - Symphonie Fantastique, Basil Twist, 1998-2019. Photo: Richard Termine Figura 17 - Symphonie Fantastique, Basil Twist, 1998-2019. Photo: Richard Termine 50 Un breve video che ne documenta la realizzazione: https://www.youtube.com/watch?v=fgkU2Vm3qOE; ultima consultazione 10 dicembre 201 111 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Cristina Grazioli Riferimenti APPIA, Adolphe. Œeuvres complètes. Éd. par Marie L. Bablet-Hahn, Genève: L’âge d’homme, 4 voll., 1983-1992, II, p. 351. BOLTANSKI, Christian. Faire son temps. Catalogo della mostra. Paris: Centre Pompi- dou, 2019. BOLTANSKI, Christian. Inventar, hrsg. von Uwe Schneede. Hamburg: Hamburger Kunsthalle, 1991. CONATI, Marcello. Aspetti della messinscena nel Macbeth di Verdi. Nuova Rivista Mu-sicale Italiana. 1981, n. 3, pp. 374-404. CRAIG, Edward Gordon. The Actor and the Übermarionette. The Mask. 2, April, 1908, pp. 4-18 DIDI-HUBERMAN, Georges. L’homme qui marchait dans la couleur. Paris: Les éditions de Minuit, 2001. DI MARINO, Bruno (a cura di). Tracce, sguardi e altri pensieri. In: Studio Azzurro. Vi- deoambienti, ambienti sensibili. Milano: Feltrinelli, 2007. GAUDÉ, Laurent. Robert Wilson et la “super-marionnette” ou la dépossession consen- 112 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure tie au comédien. In: Alternatives Théâtrales. 65-66, novembre 2000, pp. 62-63. tie au comédien. In: Alternatives Théâtrales. 65-66, novembre 2000, pp. 62-63. DEPERO, Fortunato. Colori, «Gli Avvenimenti», 1916. DEPERO, Fortunato. Colori, «Gli Avvenimenti», 1916. GONZAGA, Information à mon Chef ou éclaircissement con-venable du décorateur- théâtral Pierre Gothard Gonzague sur l’exercise de sa profes-sion, Saint Pétersburg: de l’Imprimerie d’Alex Pluchart, 1807; trad. it. in Pietro Gonza-ga, La musica degli occhi. Scritti di Pietro Gonzaga, a cura di Maria Ida Biggi, Firenze: Olschki, 2006, pp. 45-96. GOUHIER, Henri. Le théâtre et l’existence. Paris: Vrin, 2004 (prima edizione Paris, Au- bier, 1952). GOUHIER, Henri. L’essence du théâtre. Paris: Aubier-Montaigne, 1968 (Nouvelles édi- tion avec dédicace « à Gaston Baty »); prima edizione, 1943. GRAZIOLI, Cristina. Lo specchio grottesco. Marionette e automi nel teatro tedesco del primo Novecento. Padova, Esedra, 1999. GRAZIOLI, Cristina, Luce e ombra. Storia, teorie e pratiche dell’illuminazione teatrale, Roma-Bari: Laterza, 2008. GRAZIOLI, Cristina. Acteurs de lumière. In : Puck. La marionnette et les autres arts. Les marionnettes au cinéma, n. 15, 2008, pp. 11-22. GRAZIOLI, Cristina. Performing Black Box. William Kentridge o la luce dell’ombra, in «Mantichora», n. 1, dicembre 2011, pp. 379-404 GRAZIOLI, Cristina. Cavità di luce, riflessi d ’ombra: poetiche dell’assenza e dramma- turgie della luce. In: On presence, a cura di Enrico Pitozzi. Culture Teatrali, n. 21, 2012, pp. 41-58. GRAZIOLI, Cristina. Mensch, (Tod) und Kunstfigur: fi gures de la Mort et de l’altérité dans les réflexions de Rainer Maria Rilke et Oskar Schlemmer, Actes du Colloque, Sur- marionnettes et mannequins: Craig, Kantor et leurs héritages contemporains. Charle- ville-Mézières 15-17 mars 2012, Paris: L’Entretemps, 2013, pp. 153-172. GRAZIOLI, Cristina. Espaces partagés: les relations entre lumière, projection, espace. Réflexions et expérimentations (1910-1930) in Séances des années folles. Salles, tech- niques, pratiques, Paris, Kinétraces éditions (atti del seminario giugno 2017, in corso di pubblicazione). GRAZIOLI, Cristina, Proiezione, spazio, materia: comporre e scomporre con la luce in Valentina Valentini, Nuovo Teatro Made in Italy 1963-2013, con saggi di A. Barsotti, C. Grazioli, D. Orecchia. Roma: Bulzoni, 2015, pp. 325-357. 113 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure JAMAIN, Claude. Richard Teschner, le magicien de la marionnette. In : Poétiques de l’illusion. Dialogues contemporains entre marionnette et magie. Alternatives Théâ-trales: Liège 2018, pp. 84-85. LEGOUBÉ, Max. Un espace pour la divagation. In : Poétiques de l’illusion. Dialogues contemporains entre marionnette et magie. Alternatives Théâtrales: Liège 2018, pp. 42-44. MACHADO, Renato. A luz montagem. Curitiba: Prismas, 2015. MARCHIORI, Fernando, (a cura di). Beckett & Puppet. Studi e scene tra Samuel Be- ckett e il teatro di figura, Corazzano (PI): Titivillus, 2007. MAETERLINCK, Maurice. Menus Propos. Le théâtre. La jeune Belgique. Septembre 1890, pp. 331-336. METKEN, Gunther. Entre la marionnette et la machine. Le théâtre mécanique de Harry Kramer. Puck. La marionnette et les autres arts. Les plasticiens et la marionnette, n. 2, 1989, pp. 54-56. MILDENBERGER, Marianne. Film und Projektion auf der Bühne, Emsdetten: Lechte, 1961. MAURIN, Frédéric, Spectral Spectacle. In : Puck. La marionnette et les autres arts. Images virtuelles, 9, 1996, pp. 27-31. NAVARRO, Mariette. Alors Carcasse. Paris: Cheyne, 2011. PERRUCHON, Véronique. Noir. Lumière et théâtralité, Lille: Presses Universitaires du Septentrion, 2016. PITTALUGA, Noemi; VALENTINI Valentina (a cura di). Studio Azzurro. Teatro, Roma: Contrasto, 2012. PLASSARD, Didier (éd.). Les mains de lumière. Anthologie des écrits sur l’art de la ma- rionnette. Charleville-Mézières: Institut International la Marionnette, 1996. PLASSARD, Didier ; GRAZIOLI, Cristina, La marionnette, ou la mimésis complexe & La complexité des « figures » dans le théâtre en tant que « mimesis » (Convegno UNIMA, Targoviste, giugno 2017). Urdimento - Revista de Estudos em Artes Cênicas, Florianó-polis, v. 2, n. 32, 2018, p p. 56-72. http://www.revistas.udesc.br/index.php/urdimento/article/ view/1414573102322018056B/8919 PLASSARD, Didier. Illusion continue et illusion discontinue sur la scène des marion- 114 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 Cristina Grazioli Insufflare luce, animare figure nettes. In : Poétiques de l’illusion. Dialogues contemporains entre marionnette et ma- gie. Alternatives Théâtrales: Liège 2018, pp. 12-17. POSTEL, Julie. Dissimuler à vue le corps en scène: effets d’absence et morcellement des présences scèniques. In : Poétiques de l’illusion. Dialogues contemporains entre marionnette et magie. Alternatives Théâtrales: Liège 2018, pp. 56-59. POSTEL, Julie. Dissimuler à vue le corps en scène: effets d’absence et morcellement des présences scèniques. In : Poétiques de l’illusion. Dialogues contemporains entre marionnette et magie. Alternatives Théâtrales: Liège 2018, pp. 56-59. Puck. La marionnette et les autres arts. Les marionnettes au cinéma, 15, 2008. Puck. La marionnette et les autres arts. Les marionnettes au cinéma, 15, 2008. RICCI, Mario. Parentesi sull’uso del cinema e frantumazione dell’immagine, in Franco Quadri, L’avanguardia teatrale in Italia. Materiali (1960-1976). Einaudi: Torino 1977, 2 voll., I, p. 225-236, RICCI, Mario. Teatro-rito e teatro-gioco. In: Franco Quadri, L’avanguardia teatrale in Italia. Materiali (1960-1976), cit., pp. 200-211, RICCI, Mario. A partire da zero [autobiografia] di Mario Ricci, a cura di G. Celant, «Si- pario», n. 296, dicembre 1970, pp. 50-55 (poi Collage per una automitobiografia (a cura di G. Celant). In F. Quadri, L’avanguardia teatrale in Italia. Materiali (1960-1976), cit., pp. 212-221) RILKE, Rainer Maria. Ancora una parola sul valore del monologo (lettera aperta a Ru-dolf Steiner), in Rainer Maria Rilke, Scritti sul teatro, a cura di Umberto Artioli e Cristina Grazioli, Genova: Costa & Nolan, 1995, pp. 69-71. Ed. orig. Noch ein Wort über den Wert des Monologes (Offener Brief an Rudolf Steiner). In : «Dramaturgische Blätter», 40, 8. Oktober 1898, ora in Rainer Maria Rilke, Sämtliche Werke, Rilke-Archiv und R. Sieber-Rilke, éd. E. Zinn, Insel, 1975 (prima edizione in 6 volumi 1955-1966), 12 voll., v. 10, pp. 439-442. RILKE, Rainer Maria. Ancora una parola sul valore del monologo (lettera aperta a Ru-dolf Steiner), in Rainer Maria Rilke, Scritti sul teatro, a cura di Umberto Artioli e Cristina Grazioli, Genova: Costa & Nolan, 1995, pp. 69-71. Ed. orig. Noch ein Wort über den Wert des Monologes (Offener Brief an Rudolf Steiner). In : «Dramaturgische Blätter», 40, 8. Oktober 1898, ora in Rainer Maria Rilke, Sämtliche Werke, Rilke-Archiv und R. Sieber-Rilke, éd. E. Zinn, Insel, 1975 (prima edizione in 6 volumi 1955-1966), 12 voll., v. 10, pp. 439-442. RILKE, Rainer Maria. Appunti sulla melodia delle cose. In: Rainer Maria Rilke, Scritti sul teatro, cit., p. 75-84. Ed. orig. Notizen zur Melodie der Dinge [1898]. In: Rainer Maria Rilke, Sämtliche Werke, cit., pp. 412-425. RILKE, Rainer Maria. Appunti sulla melodia delle cose. In: Rainer Maria Rilke, Scritti sul teatro, cit., p. 75-84. Ed. orig. Notizen zur Melodie der Dinge [1898]. In: Rainer Maria Rilke, Sämtliche Werke, cit., pp. 412-425. RILKE, Rainer Maria. Appunti sulla melodia delle cose. In: Rainer Maria Rilke, Scritti sul teatro, cit., p. 75-84. Ed. orig. Notizen zur Melodie der Dinge [1898]. In: Rainer Maria Rilke, Sämtliche Werke, cit., pp. 412-425. Paolo Rosa. Note di regia. In: Noemi Pittaluga, Valentina Valentini (a cura di), Studio Azzurro. Teatro, Contrasto: Roma 2012. Paolo Rosa. Note di regia. Puck. La marionnette et les autres arts. Les marionnettes au cinéma, 15, 2008. In: Noemi Pittaluga, Valentina Valentini (a cura di), Studio Azzurro. Teatro, Contrasto: Roma 2012. Paolo Rosa. Note di regia. In: Noemi Pittaluga, Valentina Valentini (a cura di), Studio Azzurro. Teatro, Contrasto: Roma 2012. SINISI, Silvana. Dalla parte dell’occhio. Esperienze teatrali in Italia (1972-1982). Roma: Kappa, 1983. SINISI, Silvana. Dalla parte dell’occhio. Esperienze teatrali in Italia (1972-1982). Roma: Kappa, 1983. Ricevuto: 05/03/2020 Approvato: 06/03/2020 Urdimento, Florianópolis, v.1, n.37, p. 85-115, mar/abr 2020 115 Cristina Grazioli
https://openalex.org/W3175552426	https://www.e3s-conferences.org/10.1051/e3sconf/202127310048/pdf	English	null	The Dark Triad of personality in psychology students and eco-friendly behavior	E3S web of conferences	2,021	cc-by	2,979	* Corresponding author: buinna@mail.ru © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/). The Dark Triad of personality in psychology students and eco-friendly behavior Inna Budanova1,* 1Don state technical university, Rostov-on-Don 344000, Russia Abstract. Machiavellianism, narcissism, and non-clinical psychopathy are part of the dark triad and have recently been studied by scientists around the world. Psychologists, as representatives of helping professions, must have a pro-social personality orientation. It is known that the manifestation of individual personality traits occurs maximally in adolescence and early adulthood. Therefore, it is important to assess how the formation of not only professional, but also personal qualities of psychology students takes place. An empirical study examined the manifestation of the dark triad personality traits in psychology students enrolled in different courses of study. It was found that first-year students have a lower level of Machiavellianism than third-year students. Also, statistically significant differences were found in the manifestations of Machiavellianism in men and women. In the group of men, Machiavellianism manifested itself at a higher level than among women. The higher level of Machiavellianism among third-year students of psychology in comparison with first-year students probably reflects their mastery in the learning process of both theoretical and practical foundations of interaction, interpersonal communication, as well as the formation of practical communication skills. https://doi.org/10.1051/e3sconf/202127310048 https://doi.org/10.1051/e3sconf/202127310048 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 https://doi.org/10.1051/e3sconf/202127310048 attitude towards other people, ignoring social morality when it prevents it from achieving the desired result [5]. attitude towards other people, ignoring social morality when it prevents it from achieving the desired result [5]. Non-clinical psychopathy is characterized by the ability to make a good impression in superficial acquaintance (external charm), fearlessness, low anxiety, optional, dishonesty, craftiness, lack of regret and remorse, a tendency to antisocial behavior, an inability to learn from one's own mistakes, pathological egocentrism, poverty of the emotional sphere [5]. Students often participate in psychological, psychophysiological, psychogenetic research. [6-10]. Empirical studies from different countries have studied the relationship between the indicators of the Dark Triad and different professions studied by students. Comparing students of different humanitarian areas, namely students of psychology, economics, law and political science, it is known that psychology students had the lowest level of indicators of the dark triad [11]. In a study by Aslı, Doç & Göncü Köse, the results showed that the scores of Machiavellianism among students of economics, business, and engineering were significantly higher than those of psychology students. [12]. At the same time, comparing students of humanitarian and technical specialties, there is evidence that humanities students have a higher level of Machiavellianism than students of technical specialties. Perhaps this is a consequence of the fact that the content of their education implies the study of the psychological mechanisms of human behavior and methods of managing social systems [13]. Č Testing the professional interests of high school students with the Slovak study Čopková revealed that high rates of non-clinical psychopathy were shown by students oriented towards the production sphere, compared with students oriented towards art, science and education. Also, in psychopathy, business-oriented students scored higher than arts and education students. At the same time, Machiavellianism, in this study, is the most important feature of the Dark Triad in all professional spheres. [14]. When choosing a profession, it is important to take into account the role of the individual, since the profession affects not only individuals, but also society as a whole. The profession of a psychologist, as well as a teacher, assumes a pro-social orientation and poses certain requirements for personal and professional qualities, the severity of which becomes the key to future professional success [15, 16]. Thus, it is important for future psychologists to have such personal qualities as sympathy, tolerance, emotional responsiveness, altruism. 1 Introduction The study of negative personality traits in modern science is gaining popularity. The concept of the Dark Triad has become one of the established psychological terms relatively recently. Paulhus, Delroy & Williams, Kevin found that narcissism, Machiavellianism and non-clinical psychopathy correlate among themselves, and called the combination of these traits in personality the Dark Triad of Personality [1, 2, 3]. The phenomenon of the Dark Triad includes three psychological features - non-clinical narcissism, Machiavellianism and non-clinical psychopathy, each of which is an independent construct that is not reduced to the other two. The characteristics of narcissism include the lack of empathy and complete disinterest in the problems of others, confidence in superiority over others and the constant struggle to be in the spotlight, low self-control, ill-will, hostility and even aggressiveness when confronted with disagreement and criticism [4]. Machiavellianism in psychology is understood as a combination of at least three components: manipulative techniques in the process of interpersonal interaction, cynical E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 2 Participants and research methods On the basis of the faculty "Psychology, Pedagogy and Defectology" of the Don State Technical University in Rostov-on-Don, a study was conducted of the severity of the Dark Triad indicators among undergraduate students. The study involved 140 respondents. Of these, students of the 1st course of study - 76 and 3rd course of study - 64 people, 100 women and 40 men. All respondents participated in the study voluntarily and were informed that the data obtained in the study will only be used for scientific purposes. The study was conducted between September and December 2020 using paper and electronic testing. Assessment of severity indicators Dark Triad performed using diagnostic method personality traits questionnaire "Dark Dozen" [23]. Respondents were asked to express different degrees of agreement or disagreement with 12 statements. Statistical processing of data was carried out using the computer program STATISTICA 64. The results obtained were evaluated for compliance with the normal distribution. The application of the Kolmogorov-Smirnov test showed that the studied sample did not correspond to the normal distribution (p <0.05). Therefore, further analysis of the data was carried out using the nonparametric statistical tests of Spearman and Kruskell-Wallis. E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 Since the number of certified psychologists is increasing every year, and there is no special selection according to the severity of professionally important qualities when entering universities for training in psychological specialties, it is possible that people with a set of personal qualities that are significantly different from the professionally desirable ones may appear among students. This is facilitated by an increase in the number of psychology students in higher educational institutions, and an increasing interest in the study of human psychology, and the popularization of the profession in society as a whole. However, as some studies show, it is not uncommon for those who work in the field of assistance to have features of the Dark Triad [17-20]. The authors explain this by the fact that even a person who decides to help others may have a desire to abuse the power over clients that he has acquired in his profession. Since personality traits are maximally manifested in adolescence and early adulthood, and at the same time are hardly noticeable in childhood and adolescence, the conditioning of the manifestation of the Dark Triad traits becomes important. Egorova M.S., Sitnikova M.A. and others studied the phenomenon of the Dark Triad and its relationship with various indicators. Evaluating the relationship of the components of the Dark Triad with age, it is known that a negative correlation between age and the Dark Triad was observed in samples of a wide age range of 18-65 years. [21]. Sex differences in terms of the Dark Triad indicators, regardless of diagnostic methods, show that higher rates of Machiavellianism and non-clinical psychopathy are observed in 2 2 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 https://doi.org/10.1051/e3sconf/202127310048 men. [21]. Jonason Peter also writes about decreasing Dark Triad trait scores and increasing their consistency throughout life [22]. men. [21]. Jonason Peter also writes about decreasing Dark Triad trait scores and increasing their consistency throughout life [22]. The aim of the study was to study the Dark Triad of personality among students of the Faculty of Psychology at different stages of education. The work suggested that the formation of professional skills, the beginning of a career and an independent life can be a factor contributing to the manifestation of new personality traits, including the traits of the Dark Triad. 3 Analysis and results Table 1 presents the descriptive characteristics of the level of the Dark Triad by group. Table 1 presents the descriptive characteristics of the level of the Dark Triad by group Table 1. Descriptive characteristics of the level of the Dark Triad. Table 1. Descriptive characteristics of the level of the Dark Triad. Variables Narcissism M(SD) Psychopathy M(SD) Machiavellianism M(SD) Total by sample 12.67 (3.51) 7.36 (3.32) 10.39 (4.06) First year students 12.19 (3.31) 7.20 (3.30) 9.68 (4.09) Third year students 13.25 (3.69) 7.55 (3.37) 11.25 (3.89) Men 13.09 (5.39) 9.09 (4.28) 14.00 (3.29) Women 12.64 (3.33) 7.21 (3.21) 10.08 (3.98) Notes: M = Mean; SD = Standard deviation. Analyzing the data in Table 1 shows that mean values Machiavellianism among first-year students are lower than among third-year students. Also the mean values of Machiavellianism in men are higher than in women. Application of the Kruskell-Wallis test showed that these differences are statistically significant. The connection between Machiavellianism and the respondents' course of study is H = 6.22 at the level of significance p = 0.01. Indicators of narcissism and psychopathy are not statistically significantly associated with course of study. Evaluating the median values, it is shown that among the third-year students, the median values of the Dark Triad indicator of Machiavellianism are higher than among the group of first-year respondents. To determine the relationship between Dark Triad traits, correlations (Spearman r) were calculated between Machiavellianism, narcissism and psychopathy. The indicators of the 3 3 3 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 https://doi.org/10.1051/e3sconf/202127310048 INTERAGROMASH 2021 Dark Triad generally correlate with each other at the p <0.05 significance level. The correlation between narcissism and psychopathy was R = 0.20; between narcissism and Machiavellianism R = 0.44 and between psychopathy and Machiavellianism R = 0.46. Dark Triad generally correlate with each other at the p <0.05 significance level. The correlation between narcissism and psychopathy was R = 0.20; between narcissism and Machiavellianism R = 0.44 and between psychopathy and Machiavellianism R = 0.46. y y When evaluating the differences by sex, a statistically significant difference (H = 8.89, p = 0.01) was found only in the level of Machiavellianism too. Among the group of male respondents, the median values of the level of Machiavellianism are higher than in the group of women (see Fig. 1). Fig. 1. Median values of narcissism, psychopathy and Machiavellianism in men and women. 3 Analysis and results Notes: M = Mean; SD = Standard deviation. 14 (3.39) 7 (4.27) 13 (3.27) 13 (3.33) 6 (3.2) 10 (3.8) 0 2 4 6 8 10 12 14 16 Narcissism M(SD) Psychopathy M(SD) Machiavellianism M(SD) Men Women Fig. 1. Median values of narcissism, psychopathy and Machiavellianism in men and women. Notes: M = Mean; SD = Standard deviation. As seen in Fig. 1, the median values of narcissism and psychopathy in the group of male respondents are slightly higher than in the group of women, but these differences are not statistically significant (p> 0.05). 4 Discussion Differences between the groups of men and women manifested itself as a higher level of Machiavellianism in the group of men than among women. The result obtained agrees with the data of other authors [25]. This may be evidence that men are more typical of behavior characterized by the desire to achieve their goals at any cost, including manipulativeness and cynicism towards others. Thus, in the course of the analysis of the results obtained, the hypothesis about the different degrees of severity of the indicators of the Dark Triad in groups of students of different courses was confirmed. It was found that senior students are more likely to have Machiavellianism than younger students. The higher level of Machiavellianism among third-year psychology students in comparison with first-year students may also reflects their mastering in the course of training both the theoretical and practical foundations of interaction, interpersonal communication and also the formation of practical communication skills. If the level of Machiavellianism rises with study at the university and at the same time decreases at large time intervals in middle age, it becomes important to identify the predictors that contribute to the emergence of Machiavellianism. It is interesting to assess the relationship between the traits of the Dark Triad with emotional intelligence, the value system and a general level of life satisfaction of psychology students at different stages of education. Understanding the mechanisms of the formation of negative traits at a student age is an important condition for the choice of methods and techniques for teaching and assessing students. 4 Discussion Until recently, both in the choice of profession and in research in general, there was a tendency to concentrate primarily on the so-called bright side of the personality. We are currently seeing an increase in the popularity of studying the concept of the Dark Triad. However, these qualities cannot be unequivocally called negative. Obviously, people exhibiting Dark Triad traits are more likely to look for professions in which they can more easily demonstrate their strength. Success in certain types of professions definitely depends on the presence of such qualities as determination, insensitivity and work for the result. These include law, economics, sports, business and politics. As a psychologist, as a person working in the field of helping people, it is desirable to have pro-social personality traits. However, future students do not always correspond to this fact when choosing a profession. It is important to assess how studying at a university may be reflected not only in the formation of professional skills, but also in personal traits. Thus, as a result of the conducted empirical research, the following conclusions were formulated: 4 4 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 https://doi.org/10.1051/e3sconf/202127310048 1. Among students-psychologists of the 3rd year of undergraduate studies, the level of Machiavellianism is higher than among the group of students-psychologists of the first year. Perhaps the reason is that first-year students have recently graduated and have not yet adapted to the new conditions of study. It is possible that first-year students have a team spirit, like former schoolchildren, and senior students are already accustomed to learning and achieving their goals on their own and do not associate themselves with the team. It is noted that people brought up in a collectivist culture, in comparison with people brought up in a culture of individualism, tend to define themselves as part of a group and give priority to intragroup goals, pay more attention to context than content. [24]. 2. Differences between the groups of men and women manifested itself as a higher level of Machiavellianism in the group of men than among women. The result obtained agrees with the data of other authors [25]. This may be evidence that men are more typical of behavior characterized by the desire to achieve their goals at any cost, including manipulativeness and cynicism towards others. 2. References 1. D. Paulhus, K. Williams, Journal of Research in Personality 36, 556-563 (2002) 10.1016/S0092-6566 (02)00505-6 2. P. Jonason, J. Foster, A. Oshio et al., Personality and Individual Differences 113, 120- 124 (2017) 10.1016/j.paid.2017.02.053 3. P. Jonason, A. Oshio, T. Shimotsukasa et al., Personality and Individual Differences 120, 102-106 (2018) 10.1016/j.paid.2017.08.030 4. M.S. Egorova, The Dark Triad In Adult Sibling Dyads, 138-145 (2019) 10.15405/epsbs.2019.07.18 5. M.S. Egorova, O. Parshikova, M. Sitnikova, Personality and Individual Differences 101, 475-476 (2016) 10.1016/j.paid.2016.05.126 6. E.V. Vorobyeva, The technology of an objective psychophysiological assessment of the motivational functions of psychology students (Moscow, 2016) https://elibrary.ru/item.asp?id=27485610 7. E.V. Vorobyeva, P.N. Ermakov, O.S. Saakyan, Psychology in Russia: State of the Art 8(1), 32-42 (2015) 10.11621/pir.2015.0104 5 5 E3S Web of Conferences 273, 10048 (2021) INTERAGROMASH 2021 https://doi.org/10.1051/e3sconf/202127310048 INTERAGROMASH 2021 8. V.V. Kosonogov, E.V. Vorobyeva, E.M. Kovsh, P.N. Ermakov, International Journal of Cognitive Research in Science, Engineering and Education 7, 137-142 (2019) 10.5937/IJCRSEE1901137K 9. G. Kozhukhar, A. Belousova, E. Breus, E3S Web Conf. Innovative Technologies in Science and Education (ITSE-2020) 210 (2020) 10.1051/e3sconf/202021018009 10. A. Belousova, E. Belousova, International Journal of Cognitive Research in Science, Engineering and Education 8(2), 27-34 (2020) 10.5937/IJCRSEE2002027B 11. A. Vedel, D. Thomsen, Personality and Individual Differences 116, 86-91 (2017) 10.1016/j.paid.2017.04.030. 12. D. Aslı, K. Göncü Ü. Asli, F. Çankaya, B.P. Fen-Edebiyat, Kalem Uluslararasi Egitim ve Insan Bilimleri Dergisi. 10, 465-484 (2021) 13. S.V. Bykov, Bulletin of the Samara Humanitarian Academy. Series: Psychology 1(17) (2015) 14. R. Čopková, A. Jendrejáková, Social Education 8, 44 - 58 (2020) 10.7441/soced.2020.08.02.03 15. A. Belousova, G. Kozhukhar, INTED 2018 Proceedings. 12th International Technology, Education and Development Conference, 0291-0298 (2018) 0.21125/inted.2018 16. V.I. Pishchik, International Journal of Cognitive Research in Science, Engineering and Education 8, 69-81 (2020) 10.5937/IJCRSEE2002069P 17. P. Babiak, Applied Psychology 44(2), 171–188 (1995) 10.1111/j.1464- 0597.1995.tb01073.x 18. B. Bakir, M. Ozer, M. Uçar et al., Psychological Reports 92(3), 1169–1175 (2003) 10.2466/pr0.2003.92.3c.1169 19. V. Bucknall, S. Burwaiss, D. MacDonald et al., Pathologic Personality Traits in Health Care 187(18), 1359–1363 (2015) 10.1503/cmaj.151135 20. J. Pegrum, O. Pearce, The Bulletin of the Royal College of Surgeons of England 97(8), 331–334 (2015) 10.1308/rcsbull.2015.331 21. M.S. Egorova, M.A. Sitnikova, O.V. Parshikova, Yu.D. Chertkova, Psikhologicheskie Issledovaniya 8(43), 4 (2015) http://psystudy.ru/index.php/eng/2015v8n43e/1196-egor- sit43e .html 22. P.K. Jonason, Psikhologicheskie Issledovaniya 8(43), 3 (2015) http://psystudy.ru/index.php/eng/2015v8n43e/1195-jonason43e.html (2015) 23. T.V. Kornilova, S.A. Kornilov, M.A. Chumakova, M.S. Talmach, Psychological journal 36(2), 99-112 (2015) 24. H. References Triandis, Journal of personality 69, 907-24 (2002) 10.1111/1467-6494.696169. 25. M.S. Egorova, Dark Triad As A Predictor Of Bullying In High School Students, 222- 228 (2018) 10.15405/epsbs.2018.11.02.24 6
https://openalex.org/W2468985280	https://zenodo.org/records/5825310/files/731-732.pdf	English	null	Spectrophotometric determination of copper(II) and nickel(II) using thiosemicarbazone of citral	Zenodo (CERN European Organization for Nuclear Research)	2,006	cc-by	1,393	Keywords : SJlcL1rophotometry, copper(n), nickei(H), thiosemicarbazone. In continuation of our previous work on terpenoids and their derivatives 1 here in we report the spectrophotometric determination of the thiosemicarbazone of citra) (3,7- dimethyl-2,6-oc.:tm.Iien-1-al) with copper(u) and nickel(n). The reagent is applied for the determination of copper in synthetic alloys. Determination of copper and nickel : Copper(n) and nickel(n) reacts with CDOTSC in acidic pHs to give brown and grey coloured complexes. The order of addition of re· agent, metal ion, buffer has no effect on the absorbance of the complexes provided DMF is added prior to the addition of reagent solution. The absorbances of the coloured com- plexes remain constant for 6 h. A 5-fold molar excess of the reagent is adequate for full colour development. Addition of excess reagent has no adverse affect of the absorbance of the complexes. s II CH==N--NH--C--NH2 CH3 Tbioscmicarbatone of 3,7 -dimethyl-2,6-octadien-1-al (CDOTSC) The system obeys Beer's law in the concentration range 0.53-4.3, 0.46 j.tg ml-1 of copper and nickel respectively. The optimum range for the determination of copper( H) and nickel(n) from Ringbom's plot is found to be 0.27-().87, 0.32-().72 j.tg ml-1• The molar absorptivity and Sandell's sensitivity of the method for Cu and Ni found to be 8.85 x 10\ 1.29 x 104 L mor1 cm-1 and 0.073, 0.085 j.tg cm-2 re- spectively. The specific absorptivity of the system is 0.133, 0.127 mr1 g-1 cm-1 for Cu and Ni respectively. The relative standard deviation for ten replicate analyses Cu and Ni are 0.25 and 0.28% respectively. Job's and molar ratio methods gave the composition of the Cu and Ni complexes as I : 2, I : 1 (M : L) respectively. The stability constants of Cu and Ni complexes calculated by Job's method are found to be 8.1 x Jtf and 5.8 x 104 respectively. Tbioscmicarbatone of 3,7 -dimethyl-2,6-octadien-1-al (CDOTSC) NOTE NOTE J, Indian Chem. Soc., Vol. 83, July 2006, pp. 731-732 Anil Kr. Bansal und Meena Nagar * Department of Chemistry, University of Rajasthan, Jaipur-302 004, Rajasthan, India E-mail: anil_mandrail@yahoo.com Department of Chemistry, University of Rajasthan, Jaipur-302 004, Rajasthan, India E-mail: anil_mandrail@yahoo.com Manuscript received 17 March 2006, revised 6 April 2006, accepted /0 April 2006 Ahstruct : Tbioscmicurhuzone of citra!, a condensation product of industrially and ecofriendly terpenoid serve as very good analytical rea~ent for hiologicully important metal copper and nickel. The analytical properties of thiosemicarbazone of citra! (CDOTSC) are described for the lir~t time. The reugcnl gives brown colour complex with copper and grey colour complex with nickel in sodium ucetutc buffer medium of 5.6 and 2.6 respectively. The molar absorptivities of copper and nickel complexes respectively are 8.75 x 103 und 1.25 x 10~ L mor1 cm-1• Tllc.~e colour reactions have been investigated for the spectrophotometric determination of copper(n) und nickel(n) in aquL'Ous medium. Keywords : SJlcL1rophotometry, copper(n), nickei(H), thiosemicarbazone. J. Indian Chern. Soc., Vol. 83, July 2006 The reagent solution (0.01 m) was prepared by dissolv- ing 100 mg of the compound in 50 ml of dimethyl formam- ide and it is stable for 36 h. Hydrochloric acid (I ml), so- dium acetate (I ml) pH (0.7-3.7), 0.2 M NaOAc, 0.2 M AcOH (pH 4-6) and 2M NH4Cl, 2M NH40H buffer solu- tions were used in the present study. The standard copper(11) and nickel(n) solutions (I x 10-2 M) were prepared by using analytical reagent grade CuCI2.2H20 and NiCI2.6H20 re- spectively. The stock solutions of copper(n) and nickel(n) were standardised using titrimetric and gravimetric meth- ods3 respectively. adding Jifferent amounts of anions and cations to 0.74 and 2.46 (ppm) f.J.g mr1 of copper and nickel in solution. The colour is developed as described in the recommended pro- cedure an error of ±2% in the absorbance reading is consid- ered tolerable. (Larger amounts of iron(m) does not inter- fere in the presem;e of fluoride, masking reagent). The tol- erance limit of foreign ions in the determination of 0.74 and 2.46 Jlg mr 1 of copper and nickel are (fluoride 72-36, chloride 137-137, tartrate 897-596, bromide 315-315. car- bonate 235-235, cyanate 107-107, iodate 505-504, acetate 183-183, thiourea 0.7-6, Mov1 92-61, sulphate 981-381, nitrate 240-122, thiocyanate 0.3-26, Ce1v 3-5, Fe11 34-25, oxalate 0.3-23, Hg 11 0.5-l, Ag1 8-4, Th1v 42-43, Mn 22- 24. Ni 11 23-... , Co11 12-16. yv 2-2, Pd11 58-58, Sn11 4-0.3, Cd11 8-16, Cr1v 2-0.7, Zn 24-24, Cu 11 ... -0.8, Al111 12-0.02, zr'v 21-27. wVI 72-72, Ca11 14-14, Ba11 51-72) respectively. Schimadzu 160 A lN-visible spectrophotometric equipped with 1.0 em quartz cells Elico-model LI 120 pH meter were used in present study. The reactions of some important metal ions were tested at different pH values. The samples were prepared in 25 ml volumetric flask by mixing 10 ml of buffer, 1 ml of metal ion and I ml of 0.01 m CDOTSC solution. The reaction mixture was diluted to mark with distilled water. The absorbance was measured in 250-600 nm range against the reagent blank. Determination of copper in NTPC alloy sample : The amount of copper present in synthetic samples whose com- position corresponding to NTPC and aluminium alloys was determined by using the present method. The results are : NTPC allol :taken (0.448. 0.911, 1.831), foundb (0.428, 0.86 7, 1.719) mg/ml. J. Indian Chern. Soc., Vol. 83, July 2006 AI alloy' : taken ( 1.008, 2.0 II, 3.21 ), found (0.996, 1.999, 3.106) ("Composition of NTPC alloy Ni, 10; Cr, 15: Fe, 6.5: Cu, 4.5; Mn 2%. hAverage of 3 de- terminants. 'Composition of aluminium alloy Ni, 1.88; Cu, 4.12; Fe. 0.6: Mn, 0.2; Mg. 1.6 and rest AI). An aliquot metal ion in the Beer's law validity range (0.53-4.3 f.J.g/ml) Cu, 0.46-4.5 f.J.g/ml Ni), 10 ml of buffer solution (pH 5.6 for Cu and pH 2.6 for Ni) and I ml of I x w-2 m CDOTSC solution were taken in 25 ml standard flask and diluted to the mark with distilled water. The absorbance to the coloured solution was measured at corre- sponding wave length (373 and 376 nm for Cu11 and Ni11 re- spectively) against reagent blank. Calibration graphs were prepared. The measured absorbance values were used to compute the amount of copper and nickel present in the un- known solution. Conclusion : The reagent provides a simple rapid and accurate method for the spectrophotometric determination of Cu11 and Ni11• The reagent has the advantage of high sensitivity, selectivity and easy availability. Results and discussion The citra) chemically known as 3,7-dimethyl-2,6-oc- tadien-1-al, thiosemicarbazone derivative of this citra) (CDOTSC) is easily prepared under reflux conditions2. In basic medium the ligand presumably exists in enolic form and co-ordinates the divalent metal ion as mono anion. The reagent gives intense colour reaction only with copper and nickel and show maximum absorbance at 373 and 376 nm respectively. The reagent is considered as potential reagent for selective spectrophotometric determination of copper{n) and nickei(H). 8. Interferences : The effects of some of the ions which often accompany copper and nickel has been studied by 731 J. Indian Chern. Soc., Vol. 83, July 2006 Experimental Thiosemi<.:arbazone of citra! (3,7-dimethy-2,6-oxadiene-1- al) was prep;m:d by dissolving the thiosemicarbazide ( 10 mmol) i.n 5o/. aqueous AcOH (30 cm3) on a boiling water bath. The solution was slowly added with stirring to a freshly prepared solution of 3,7-dimethyl-2,6-octadien-1-al (10 mmol) in EtOH (20 cm3, 95%). After retluxing for I h the reaction mixture was cooled to room temperature and a yellow precipitated obtained. It was characterised by ele- mental analysis, IR and EPR spectral data. 3. J. Ba~sett, R. C. Danney and G. H. Jeffery, "Vogel's Text Book of Quantitative Inorganic Analysis", 4th ed., Longman, London, 1982. 321, 474. I. M. Chand, P. Lala and Meena Nagar. J. lndtwt Chern. Soc., 2003. 80, 861; Ani I Kr. Bansal. P. L. Sharma and Meena Nagar, J. Indian Chem. Soc., 2006, 83, 26. 2. Renu Sharml!. Anil Kr. Bansal and Meena Nagar, Indian J. Cltl'm .. Sect. A. 2005, 44, 2255. 3. J. Ba~sett, R. C. Danney and G. H. Jeffery, "Vogel's Text Book of Quantitative Inorganic Analysis", 4th ed., Longman, London, 1982. 321, 474. I. M. Chand, P. Lala and Meena Nagar. J. lndtwt Chern. Soc., 2003. 80, 861; Ani I Kr. Bansal. P. L. Sharma and Meena Nagar, J. Indian Chem. Soc., 2006, 83, 26. 2. Renu Sharml!. Anil Kr. Bansal and Meena Nagar, Indian J. Cltl'm .. Sect. A. 2005, 44, 2255. References 732
https://openalex.org/W2398152737	http://umu.diva-portal.org/smash/get/diva2:943194/FULLTEXT01	English	null	Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid-Adulthood: A Path Analysis Study	PloS one	2,016	cc-by	11,030	OPEN ACCESS OPEN ACCESS Citation: Jonsson F, San Sebastian M, Strömsten LMJ, Hammarström A, Gustafsson PE (2016) Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid-Adulthood: A Path Analysis Study. PLoS ONE 11(5): e0155963. doi:10.1371/journal.pone.0155963 Editor: Hajo Zeeb, Leibniz Institute for Prevention Research and Epidemiology (BIPS), GERMANY Received: January 14, 2016 Accepted: May 7, 2016 Published: May 23, 2016 Copyright: © 2016 Jonsson 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. While research examining the health impact of early socioeconomic conditions suggests that effects may exist independently of or jointly with adult socioeconomic position, studies exploring other potential pathways are few. Following a chain of risk life course model, this prospective study seeks to examine whether pathways of occupational class as well as material and social adversities across the life course link socioeconomic disadvantage in adolescent to functional somatic symptoms in mid-adulthood. Applying path analysis, a multiple mediator model was assessed using prospective data collected during 26 years through the Northern Swedish Cohort. The sample contained 987 individuals residing in the municipality of Luleå, Sweden, who participated in questionnaire surveys at age 16, 21, 30 and 42. Socioeconomic conditions (high/low) in adolescence (age 16) were operationalized using the occupation of the parents, while occupational class in adulthood (manual/non- manual) was measured using the participant’s own occupation at age 21 and 30. The adver- sity measurements were constructed as separate age specific parcels at age 21 and 30. Social adversity included items pertaining to stressful life events that could potentially harm salient relationships, while material adversity was operationalized using items concerning unfavorable financial and material circumstances. Functional somatic symptoms at age 42 was a summary measure of self-reported physical symptoms, palpitation and sleeping diffi- culties that had occurred during the last 12 months. An association between socioeconomic conditions at age 16 and functional somatic symptoms at age 42 (r = 0.068) which was par- tially explained by people’s own occupational class at age 21 and then material as well as social adversity at age 30 was revealed. http://www.diva-portal.org http://www.diva-portal.org This is the published version of a paper published in PLoS ONE. Citation for the original published paper (version of record): Jonsson, F., San Sebastian, M., Stromsten, L M., Hammarstrom, A., Gustafsson, P E. (2016) Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid-Adulthood: A Path Analysis Study. PLoS ONE, 11(5): e0155963 http://dx.doi.org/10.1371/journal.pone.0155963 Access to the published version may require subscription. N.B. When citing this work, cite the original published paper. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-123074 RESEARCH ARTICLE Frida Jonsson, Miguel San Sebastian, Lotta M. J. Strömsten, Anne Hammarström, Per E. Gustafsson a1111 Department of Public Health and Clinical Medicine, Unit of Epidemiology and Global Health, Umeå University, Umeå, Sweden frida.jonsson@umu.se * frida.jonsson@umu.se Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid- Adulthood: A Path Analysis Study Frida Jonsson, Miguel San Sebastian, Lotta M. J. Strömsten, Anne Hammarström, Per E. Gustafsson Introduction 259-2012-37) and by The County Council of Västerbotten (www.vll.se, grant# 355661 and 402131). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Not only a person’s current occupational class, education or income, but also the socioeco- nomic circumstances they encountered prior in life have been found to be important for self- rated health in adulthood [1–3]. While this research suggests that early socioeconomic condi- tions might affect later health independently of or jointly with adult socioeconomic position, studies examining other potential pathways are few. In accordance with a social chain of risk life course model, hypothesizing that adulthood health might be affected by a continuity of unfavorable life circumstances [4], this prospective study seeks examine if pathways of occupa- tional class as well as material and social adversities across the life course linked socioeconomic disadvantage in adolescence to functional somatic symptoms in mid-adulthood. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. The life course framework emphasizes the long-term effects of early exposures, and high- lights for example the importance of disadvantage during specific life course periods and chains of unfavorable conditions across life in relation to later health [4]. Following a sensitive (or crit- ical) period life course model, which hypothesize on enduring bodily damage and irreversible effects on adulthood health, a large body of research has examined whether socioeconomic conditions early in life may affect adult health independently of later exposures. Whether examined through parents’ education, income, wealth, some occupational based class measures or an index [5, 6], studies suggests that the socioeconomic conditions of the family during childhood/adolescence may be particularly influential for a variety of health outcomes in adult- hood [1–3, 7–17]. Besides the idea of a sensitive period, however, the impact of early socioeco- nomic conditions might also ripple across the life course due to a continuity of unfavorable life circumstances. Consistent with the social chain of risk (or pathway) life course model, health may be a result of pathways which unfold across life due to the fact that one hardship often leads to another [4], indicating that adverse life circumstances are temporally related and per- sists across the life span. Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Introduction While an extensive amount of research has investigated if the socio- economic conditions of the family has direct and independent effects on adulthood health, The life course framework emphasizes the long-term effects of early exposures, and high- lights for example the importance of disadvantage during specific life course periods and chains of unfavorable conditions across life in relation to later health [4]. Following a sensitive (or crit- ical) period life course model, which hypothesize on enduring bodily damage and irreversible effects on adulthood health, a large body of research has examined whether socioeconomic conditions early in life may affect adult health independently of later exposures. Whether examined through parents’ education, income, wealth, some occupational based class measures or an index [5, 6], studies suggests that the socioeconomic conditions of the family during childhood/adolescence may be particularly influential for a variety of health outcomes in adult- hood [1–3, 7–17]. Besides the idea of a sensitive period, however, the impact of early socioeco- nomic conditions might also ripple across the life course due to a continuity of unfavorable life circumstances. Consistent with the social chain of risk (or pathway) life course model, health may be a result of pathways which unfold across life due to the fact that one hardship often leads to another [4], indicating that adverse life circumstances are temporally related and per- sists across the life span. While an extensive amount of research has investigated if the socio- economic conditions of the family has direct and independent effects on adulthood health, only a few studies have examined whether they might be the foundation for a series of unfavor- able life circumstances. Tsenkova, Pudrovska, & Karlamangla [18] suggest that early socioeco- nomic disadvantage predict physical inactivity and depressive symptoms, which is in turn can lead to diabetes type 2 later in adulthood. Hagger-Johnson and colleagues [19] propose that the paternal class predicts people’s own occupational class, which in turn is linked to later health behaviour and BMI, ultimately affecting inflammatory markers later in life. Moreover, adoles- cents growing up in less affluent homes seem to experience more social and material stressors across the life course which has been linked to metabolic syndrome in adult women [20]. OPEN ACCESS Rather than proposing a direct and independent health effect of the socioeconomic conditions of the family, the present study suggests that growing up in an unfavorable socioeconomic environment might be a source for a chain of adverse material and social living situations, which in turn affects adult health. Citation: Jonsson F, San Sebastian M, Strömsten LMJ, Hammarström A, Gustafsson PE (2016) Life Course Pathways of Adversities Linking Adolescent Socioeconomic Circumstances and Functional Somatic Symptoms in Mid-Adulthood: A Path Analysis Study. PLoS ONE 11(5): e0155963. doi:10.1371/journal.pone.0155963 Editor: Hajo Zeeb, Leibniz Institute for Prevention Research and Epidemiology (BIPS), GERMANY Editor: Hajo Zeeb, Leibniz Institute for Prevention Research and Epidemiology (BIPS), GERMANY Copyright: © 2016 Jonsson et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: The dataset is part of the Northern Swedish Cohort (NSC) which is not freely accessible, this is because the Swedish Data Protection Act (1998:204) does not permit sensitive data on humans (like in the NSC questionnaires) to be openly shared. As such, the data are available upon request from the Principal Investigator Anne Hammarström (anne.hammarstrom@umu.se), pending ethical approval. Funding: The study was supported by The Swedish Research Council Formas (www.formas.se, grant# 1 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Conceptual framework Based on a large body of research suggesting that the socioeconomic conditions of the family is important for later health, either independently or by being the source of a series of unfavorable life circumstances, we developed a conceptual framework in which several plausible chains of risks are examined simultaneously (Fig 1). First (path A) we address the likelihood that people’s socioeconomic background, as indi- cated by their parents occupational class, has an impact on the positions in the social hierarchy they will occupy along their life course. Although some differences exist between nations and across time, research suggests that about 12–21 percent of variation in the son’s occupational class can be attributed to the class of the father [32]. To follow the sensitive period life course hypothesis, we also included path B to capture the possibility that an adverse socioeconomic background might predict functional somatic symptoms years later, irrespective of other cir- cumstances happening along the life course. The concept of class proposes that occupations can be hierarchically categorized based on differences in employment relations and conditions within the labour market [33]. As indi- cated by the social gradient, the health of the population differs between each vertical step along the occupational structure [34]. People higher up in the social hierarchy tend to be healthier and live longer, while a disproportionately large burden of disease is concentrated amongst the most socially disadvantaged groups. Occupational class is one of the most prominent determinants of health, and by influencing an array of material [35] and social aspects [36] of everyday life, it is often viewed as a factor distributing such resources [37]. Although this chain of risk is well established, theoretically, few have examined it empirically. Through paths E1 and F, rather than suggesting that low occupational class affects health directly (as in path A and B), we hypothesize that it predicts material and social adversities along the life course. Within the present study, material adversi- ties are viewed as circumstances including a possession of, or an access to financial and economic resources [38], while social adversities concern stressful relational incidents or con- ditions that may alter a person’s usual activities and impair important relationships [39]. Introduction By examining the independent effects of various determinants across life in relation to poorer self- rated health [21] and with regard the risk of heart attack in later life [22], additional pathways have been proposed. Studies of this kind do not consider the temporal relation among the exposures, however, and consequently they provide limited insight to whether such pathways may constitute a potential chain of risk. only a few studies have examined whether they might be the foundation for a series of unfavor- able life circumstances. Tsenkova, Pudrovska, & Karlamangla [18] suggest that early socioeco- nomic disadvantage predict physical inactivity and depressive symptoms, which is in turn can lead to diabetes type 2 later in adulthood. Hagger-Johnson and colleagues [19] propose that the paternal class predicts people’s own occupational class, which in turn is linked to later health behaviour and BMI, ultimately affecting inflammatory markers later in life. Moreover, adoles- cents growing up in less affluent homes seem to experience more social and material stressors across the life course which has been linked to metabolic syndrome in adult women [20]. By examining the independent effects of various determinants across life in relation to poorer self- rated health [21] and with regard the risk of heart attack in later life [22], additional pathways have been proposed. Studies of this kind do not consider the temporal relation among the exposures, however, and consequently they provide limited insight to whether such pathways may constitute a potential chain of risk. The health impact of socioeconomic circumstances is often considered to be the result of a long-term exposure to unmanageable and sustained stress [23, 24]. But since stress is a general mechanism not restricted to a certain disease or illness [25], the life course impact of socioeco- nomic conditions during childhood/adolescence has been linked to a variety of health prob- lems, albeit rarely with respect to functional somatic symptoms. Functional somatic symptoms is generally described as a clustering of physical complaints without a confirmed pathological origin [26]. It has been positively associated with increased levels of stress [27] and effects a person’s health status as well as quality of life [28]. Introduction Previous studies have found it relevant to examine stress-related health problems such as psychological health [29], allostatic load [30] PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 2 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms and even functional somatic symptoms [31] through a life course perspective. Thus, functional somatic symptoms may be a health outcome that is relevant to study as a consequence of adversity across the life course. Following the chain of risk life course model, and by using path analysis on prospective lon- gitudinal data, the aim of the present study is to examine whether pathways of occupational class as well as material and social adversities across the life course connects socioeconomic dis- advantage in adolescent with functional somatic symptoms in mid-adulthood. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Methods The analysis was based on prospective self-administrated questionnaire data collected during 26 years through the Northern Swedish Cohort [44]. The cohort was initiated in 1981 and included all ninth grade students (aged 16 at the time) attending compulsory school in the municipality of Luleå, Sweden. Follow-up surveys were carried out in 1983, 1986, 1995 and 2007, and out of the initial 1071 students, 94.3% participated across the entire period (n = 1010). The Northern Swedish Cohort has been granted ethical approval by the Regional Ethical Review Board in Umeå (dnr 07–057). Answering the survey was considered as consent to participate. For the present study, data from the surveys carried out in 1981, 1986, 1995 and 2007 were used (respondents were aged 16, 21, 30 and 42, respectively) and the analysis was based on a sample of 987 individuals. Conceptual framework In light of studies suggesting that people who are exposed to such adverse life events tend to expe- rience more stress-related health problems [40], we propose that these circumstances might be what connects class with functional somatic symptoms (paths E1 and F). In addition to people’s own occupational class in adulthood determining aspects of their social and material lives, the socioeconomic conditions of their family during adolescence might also be a source for such adversities. Research shows that people from unfavorable socio- economic backgrounds tend to experience more external material and interpersonal stressors [41]. Following this knowledge, and in line with previous studies [35], in path C we hypothesize that early socioeconomic disadvantage increases the risk of, for example, unemployment and financial hardship. Circumstances that may in turn predict higher levels of functional somatic symptoms in adulthood. Path D is based on a similar line of reasoning, but here we propose that growing up under unfavorable socioeconomic circumstances may also bring about social PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 3 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Fig 1. Conceptual model of adversity pathways in the association between socioeconomic circumstances early in life and functional somatic symptoms in adulthood. doi:10 1371/journal pone 0155963 g001 Fig 1. Conceptual model of adversity pathways in the association between socioeconomic circumstances early in life and functional somatic symptoms in adulthood. Fig 1. Conceptual model of adversity pathways in the association between socioeconomic circumstances early in life and functional somatic symptoms in adulthood. doi:10.1371/journal.pone.0155963.g001 doi:10.1371/journal.pone.0155963.g001 doi:10.1371/journal.pone.0155963.g001 stressors such as, for example, isolation and a limited ability to have influence over one’s life [42]. In path E2, we follow previous research suggesting that social stressors might be triggered more directly by a lack of material resources [43]. Thus, the model also hypothesizes that early socioeconomic disadvantage might contribute to materially strained living conditions, which then contributes to an adverse social environment, ultimately affecting adulthood health. Measures Occupational class. As an indicator of the socioeconomic conditions of the family in which people grew up, the occupation of the parent’s when the cohort participant was 16 years was used as a proxy. Occupational class across mid-life was assessed using the participant’s own occupation at age 21 and 30. The coding was done in accordance with the socioeconomic classification system of Statistics Sweden [45], differentiating between workers in manual labor 4 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms (blue-collar), non-manual employees (white-collar) and the self-employed. At age 16, the clas- sification was based on an older version of the classification scheme [46]; lower class (1) was defined as having two parents working in manual labor, while having at least one parent in non-manual or self-employed occupation indicated higher class (0). At age 21 and 30, partici- pants reporting that they were working in manual labor were classified as having low class (1) while non-manual employees and the self-employed were clustered as being of high class (0). For some participants at age 21 and 30, there was no current occupation (due to unemploy- ment, studies or military service), and in cases where the previous occupation was not accessi- ble, the highest educational attainment was used as a proxy (at age 21, n = 206 and at age 30, n = 41). High class (0) was indicated by university studies or a high school degree qualifying the person for university studies. Low class (1) reflected all other high school degrees or lower levels of education. Life course measures of social and material adversity. The operationalizations has been previously used elsewhere [47, 48], and were based on items from age specific questionnaires completed by the participants at age 21 and 30 making the set adversities differ slightly between the measures. Parcels were created by adding up the number of adversities for each concept and age while acknowledging that, as holistic constructs, the concepts are theoretically and empirically heterogeneous. Social adversity. Social adversity at age 21 included items pertaining to residential instabil- ity (the number of times the respondent had moved during the last 3 years) and exposure to ill- ness and death during the last 3 years (defined as someone close to the respondent suffering from serious or long-term illness and/or if someone close to them had died). PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 The empirical model As an operationalization of the conceptual framework (Fig 1), one multiple mediator model (Fig 2) was developed containing several hypotheses, which were combined and examined simultaneously [51]. First, to examine social and material adversities as plausible mediators, ten manifest vari- ables were connected through direct paths based on their temporal precedence [52]. Three var- iables were independent: socioeconomic conditions at age 16, and the control variables sex and functional somatic symptoms at 16. Seven were dependent; two binary measures—occupa- tional class at age 21 and 30 —and five continuous measures—material and social adversity at age 21 and 30 –as well as functional somatic symptoms at age 42. Second, as personal characteristics as well as social support may be potential sources of error [53], residuals of measures at age 21 and 30 were allowed to covary [51]. Within the model, all the paths were adjusted for sex and baseline functional somatic symptoms (age 16). Since social and psychological resources may vary according to gender [54], and given that men and women tend to experience and cope with adversity slightly differently [55], the analy- sis was also stratified by sex. Measures For eight 5 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms symptoms the following question was used: ‘Do you have (or have you during the last 12 months) had any of the following symptoms: headache or migraine; other stomachache; nau- sea; backache, hip pain or sciatica; fatigue; breathlessness; dizziness; overstrain. The response options were ‘No’ (0), ‘Yes, light’ (1) and ‘Yes, severe’ (2). Palpitation was covered using the question: ‘How often have you had nervous problems during the past 12 months’, with the type of symptom and frequency indicated as ‘never’ (0), ‘sometimes’ (1) and ‘always’ (2). For sleep- lessness, the following question was used: ‘Have you had sleeping difficulties during the past 12 months?’ with the response options coded as ‘Never’ (0), ‘Sometimes’ (1) and ´often’ or ‘always’ (2). When summarized, the variable ranged from 0–18 for women and 0–15 for men, with higher values corresponding to more somatic problems. Cronbach’s alpha revealed an internal consistency of 0.782. Control variables. Sex and functional somatic symptoms at age 16 were used as control vari- ables. The operationalization of functional somatic symptoms was identical to the one used at age 42 (see description above), but based on questionnaires completed by the participant at age 16. Measures At age 30, social adversity included items pertaining to exposure to illness and death during the last 12 months; experiencing a separation during the last 12 months (defined as breaking up from a long-term cohabitating relationship); social isolation (the total score on four items in the Availability of Social Integration scale of the Interview Schedule for Social Interaction [49]); having low deci- sion latitude (the summary response to six four-level Likert scale items about decision latitude, four on skill discretion and two on decision authority [50]; or being exposed to threat/violence during the last 12 months (due to low frequencies, defined as a positive response on either of four items was used: 1) personally being abused at work through mean words and actions from bosses or colleagues, 2) sexual harassment through unwelcome or degrading sexual insinua- tions, 3) physical violence or 4) threats of violence that were so serious that she or he was scared). Social adversity at age 21 ranged from 0–3 for both women and men, and at age 30 between 0–5 for men and 0–6 for women. Material adversity. Material adversity was operationalized by drawing on information pertaining to: unemployment, at both age 21 and 30, defined as currently being unemployed or on disability pension; also, low cash margin, not being able to raise 5,000 SEK and 13,000 SEK within a week, at age 21 and 30, respectively. At age 30, items pertaining to spousal unemploy- ment, defined by the participant’s partner being unemployed during the last 5 years, and finan- cial strain was also included. The latter was captured through a question regarding how often (answered ‘often’, ‘seldom’, ‘never’, or ‘not applicable’, where the number of ‘often’ responses were dichotomized at the 80th percentile) the respondent had abstained from any of 11 differ- ent material needs due to monetary problems (for example, eat a cooked meal, buy clothes, and pay the rent or other bills due to financial reasons). After being summarized, the measures ran- ged between 0–3 at age 21 and 0–4 at age 30 for both sexes. Functional somatic symptoms. The primary outcome, functional somatic symptoms at age 42, was a summary measure of ten different symptoms (cardiopulmonary/autonomic, gas- trointestinal, musculoskeletal and general symptoms) occurring during the last 12 months. Each symptom was coded 0–2, and collected through three survey questions. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Data analysis To analyze the empirical model (Fig 2), path analysis using Mplus 7 [56] was applied, enabling us to examine a multiple mediator model with a combination of binary and continuous vari- ables while estimating all parameters simultaneously [57]. By considering categorical measures to be indicators of a latent continuous variable, dichotomized at a threshold value, binary vari- ables could be estimated in combination with continuous variables. To attain path coefficients and model fit, WLSMV (a robust, mean- and variance-adjusted weighted least square method) was used with THETA parameterization [56]. The model-data correspondence was evaluated relative to a number of fit indices. The chi- square (x2) was used to examine the hypothesis of exact fit [58] while the value and confidence interval of the root mean square error of approximation (RMSEA) provided complementary information on the additional close fit hypothesis [59]. By also considering the comparative fit index (CFI) as well as the weighted root mean square residual (WRMR), our empirical model was examined against a baseline model that was based on the assumption that all the variables were uncorrelated. In accordance with Yu [60], RMSEA had to be around .06, the CFI close to .96 and a WRMR had to be smaller than 1.0 for model fit to be acceptable. 6 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Fig 2. The multiple mediator model examining the relation between socioeconomic conditions at age 16 (SC16) and functional somatic symptoms at age 42 (FSS42), via occupational class (OC21 and OC30), material adversity (MA21 and MA30) and social adversity (SA21 and SA30) at age 21 and 30. Paths are estimated direct effects between the variables, while controlling for sex and the baseline functional somatic symptoms for the respondents at age 16 (FSS16) (n = 987). Fig 2. The multiple mediator model examining the relation between socioeconomic conditions at age 16 (SC16) and functional somatic symptoms at age 42 (FSS42), via occupational class (OC21 and OC30), material adversity (MA21 and MA30) and social adversity (SA21 and SA30) at age 21 and 30. Paths are estimated direct effects between the variables, while controlling for sex and the baseline functional somatic symptoms for the respondents at age 16 (FSS16) (n = 987). Data analysis doi:10.1371/journal.pone.0155963.g002 doi:10.1371/journal.pone.0155963.g002 Within the model, the estimates of the direct effects are probit regression coefficients when the dependent variable is categorical, and linear regression coefficients when they are continu- ous. Mediation was examined using a percentile bootstrap, a nonparametric re-sampling tech- nique that provided 95% bootstrapped confidence intervals for indirect effects and bootstrapped standard errors (5,000 samples requested) [51]. Significant effects were displayed at the p < 0.05 level, provided that the confidence interval did not include 0 (zero). Selection bias regarding missing data was examined, analyzing whether the effective sample differed from respondents excluded because of incomplete data. The main variables were used as predictors of a binary missingness variable through simple logistic regression in SPSS 21. Two measures—socioeconomic position at age 30 (n = 21 excluded, p = 0.047) and social adver- sity at age 30 (n = 12 excluded, p = 0.030)–revealed significant estimates, indicating that the missing data is MAR, i.e. conditional only on covariates in the model. Since simulation studies have shown that weighted least square estimation using pairwise deletion provide accurate esti- mates under the MAR assumption [61] this method was used to handle our missing data. Preliminary data screening revealed no severe violations to the normality (skewness < 1.1, kurtosis < 2.0) or linearity assumptions, and as indicated by a low variance inflation factor (VIF) (values < 1.1) multicollinearity was not present. Since the direction and inclusion of paths were guided by theory, a sensitivity analysis was performed to examine whether we had been too restrictive in the operationalization of the conceptual framework. Rerunning a satu- rated model that allowed for all plausible pathways did however not change the results of the study, why the original model was retained and reported in the results section. doi:10.1371/journal.pone.0155963.g002 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms than half reported a low occupational class of their own (63%), while at age 30, this number had decreased to 43%. The sex stratified analysis (Table 1) revealed that social and material adversity as well as functional somatic symptoms at age 42 differed, where women presented more adversi- ties overall, as well as more functional somatic symptoms at age 42, compared to men. Table 2 presents zero-order correlation coefficients (Pearson’s r) between all the variables. Cross-validation using Spearman’s rho yielded similar results. Although only marginally, an unfavorable socioeconomic environment at age 16 was positively correlated with functional somatic symptoms at age 42 (r = 0.068), while both material and social adversities at age 30 Table 1. Descriptive statistics; mean and standard deviation, in the full sample (n = 987) and stratified by sex. All the variables and control variables in the model at four points in time—respondents aged 16, 21, 30 and 42. Predictor estimates are mean (SD). Variables Full sample Women Men Difference Range Estimate Range Estimate Range Estimate p-value Material adversity Age 21 0–3 0.54 (0.682) 0–3 0.57 (0.702) 0–3 0.51 (0.663) 0.156 a Age 30 0–4 0.77 (0.947) 0–4 0.88 (1.029) 0–4 0.66 (0.852) <0.0005 a Social adversity Age 21 0–3 0.76 (0.850) 0–3 0.91 (0.891) 0–3 0.62 (0.785) <0.0005 a Age 30 0–6 0.97 (0.989) 0–6 1.04 (1.002) 0–5 0.91 (0.974) 0.031 a Functional somatic symptoms Age 16 0–16 3.35 (2.540) 0–16 3.71 (2.510) 0–12 3.03 (2.526) <0.0005 a Age 42 0–18 4.24 (3.306) 0–18 4.67 (3.503) 0–15 3.75 (3.032) <0.0005 a Socioeconomic conditions Age 16 0–1 37.7% low Occupational class Age 21 0–1 62.9% low Age 30 0–1 43.2% low Sex Women 1 48.2% Men 2 51.8% a p-value based on t-test doi:10.1371/journal.pone.0155963.t001 Table 2. Pearson’s correlations for socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA), and functional somatic symptoms (FSS) at four points in time—respondents aged 16, 21, 30 and 42. Variables SC16 OC21 OC30 MA21 MA30 SA21 SA30 FSS16 FSS42 SC16 - 0.229* 0.205 0.055 0.192 0.033 0.123** 0.042 0.068* Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Table 1. Descriptive statistics; mean and standard deviation, in the full sample (n = 987) and stratified by sex. All the variables and control variables in the model at four points in time—respondents aged 16, 21, 30 and 42. Predictor estimates are mean (SD). Variables Full sample Women Men Difference Range Estimate Range Estimate Range Estimate p-value Material adversity Age 21 0–3 0.54 (0.682) 0–3 0.57 (0.702) 0–3 0.51 (0.663) 0.156 a Age 30 0–4 0.77 (0.947) 0–4 0.88 (1.029) 0–4 0.66 (0.852) <0.0005 a Social adversity Age 21 0–3 0.76 (0.850) 0–3 0.91 (0.891) 0–3 0.62 (0.785) <0.0005 a Age 30 0–6 0.97 (0.989) 0–6 1.04 (1.002) 0–5 0.91 (0.974) 0.031 a Functional somatic symptoms Age 16 0–16 3.35 (2.540) 0–16 3.71 (2.510) 0–12 3.03 (2.526) <0.0005 a Age 42 0–18 4.24 (3.306) 0–18 4.67 (3.503) 0–15 3.75 (3.032) <0.0005 a Socioeconomic conditions Age 16 0–1 37.7% low Occupational class Age 21 0–1 62.9% low Age 30 0–1 43.2% low Sex Women 1 48.2% Men 2 51.8% a p-value based on t-test doi:10.1371/journal.pone.0155963.t001 Table 1. Descriptive statistics; mean and standard deviation, in the full sample (n = 987) and stratified by sex. All the variables and control variables in the model at four points in time—respondents aged 16, 21, 30 and 42. Predictor estimates are mean (SD). Table 1. Descriptive statistics; mean and standard deviation, in the full sample (n = 987) and stratified by se in the model at four points in time—respondents aged 16, 21, 30 and 42. Predictor estimates are mean (SD). than half reported a low occupational class of their own (63%), while at age 30, this number had decreased to 43%. The sex stratified analysis (Table 1) revealed that social and material adversity as well as functional somatic symptoms at age 42 differed, where women presented more adversi- ties overall, as well as more functional somatic symptoms at age 42, compared to men. Table 2 presents zero-order correlation coefficients (Pearson’s r) between all the variables. Cross-validation using Spearman’s rho yielded similar results. Although only marginally, an unfavorable socioeconomic environment at age 16 was positively correlated with functional somatic symptoms at age 42 (r = 0.068), while both material and social adversities at age 30 Table 2 presents zero-order correlation coefficients (Pearson’s r) between all the variables. Cross-validation using Spearman’s rho yielded similar results. Results Table 1 presents descriptive statistics for the variables in the model, on the full sample and strati- fied by sex. Respondents with origins in a high socioeconomic setting were in the majority (62%), as indicated by the occupation of their parent’s. At age 21, the reverse was apparent and more 7 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Although only marginally, an unfavorable socioeconomic environment at age 16 was positively correlated with functional somatic symptoms at age 42 (r = 0.068), while both material and social adversities at age 30 socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA), and ) at four points in time—respondents aged 16, 21, 30 and 42. Table 2. Pearson’s correlations for socioeconomic conditions (SC), occupational class (OC), material adve functional somatic symptoms (FSS) at four points in time—respondents aged 16, 21, 30 and 42. Table 2. Pearson’s correlations for socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA), and functional somatic symptoms (FSS) at four points in time—respondents aged 16, 21, 30 and 42. Table 2. Pearson’s correlations for socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA), and functional somatic symptoms (FSS) at four points in time—respondents aged 16, 21, 30 and 42. Variables SC16 OC21 OC30 MA21 MA30 SA21 SA30 FSS16 FSS42 SC16 - 0.229 0.205 0.055 0.192 0.033 0.123 0.042 0.068* OC21 - 0.426 -0.009 0.188 -0.002 0.141** -0.006 0.060 OC30 - 0.063* 0.263 -0.005 0.218 -0.008 0.105 MA21 - 0.231 0.041 0.093 0.003 0.026 MA30 - 0.140 0.261** 0.079* 0.191** SA21 - 0.080* 0.086 0.094 SA30 - 0.064* 0.205 FSS16 - 0.230 FSS42 - functional somatic symptoms (FSS) at four points in time respondents aged 16, 21, 30 and 42. Variables SC16 OC21 OC30 MA21 MA30 SA21 SA30 FSS16 FSS42 SC16 - 0.229 0.205 0.055 0.192 0.033 0.123 0.042 0.068* OC21 - 0.426 -0.009 0.188 -0.002 0.141** -0.006 0.060 OC30 - 0.063* 0.263 -0.005 0.218 -0.008 0.105 MA21 - 0.231 0.041 0.093 0.003 0.026 MA30 - 0.140 0.261** 0.079* 0.191** SA21 - 0.080* 0.086 0.094 SA30 - 0.064* 0.205 FSS16 - 0.230 FSS42 - * p < 0.05 (2-tailed), ** p < 0.01 (2-tailed) doi:10.1371/journal.pone.0155963.t002 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 8 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms were significantly related to socioeconomic conditions in adolescence (r = 0.192 and 0.123, respectively) as well as functional somatic symptoms at age 42 (r = 0.191 and 0.205, respectively). The empirical model (Fig 2) was analyzed, assessing direct and indirect effects with boot- strapped confidence intervals and standard errors [62]. All fit indices, except a significant chi- square, indicated the model had a good fit (n = 987) to the data (x2 = 36.312 with p < 0.001 and df = 8, RMSEA = 0.060, CFI = 0.963 and WRMR = 0.726). Table 3 displays all the esti- mated direct and indirect effects, adjusted for sex and baseline functional somatic symptoms, in the full sample. In Fig 3, only significant (p < 0.001) path coefficients for direct effects are presented. The analysis revealed a significant association between socioeconomic conditions in adolescence and a person’s occupational class at age 21, and from age 21 to age 30, indicating a continuity of along the life course. With regard to the health impact, neither the socioeconomic conditions of the family at age 16 nor people’s occupational class at age 21 and 30 was directly related to functional somatic symptoms in adulthood. However, a person’s class at age 21 Table 3. Direct and indirect effects (5000 samples requested), adjusted for sex and baseline functional somatic symptoms, in the model for the full sample (n = 987). The variables are socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA) and functional somatic symptoms (FSS), at four points in time—respondents aged 16, 21, 30 and 42. Table 3. Direct and indirect effects (5000 samples requested), adjusted for sex and baseline functional somatic symptoms, in the model for the full sample (n = 987). The variables are socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA) and functional somatic symptoms (FSS), at four points in time—respondents aged 16, 21, 30 and 42. Table 3. Direct and indirect effects (5000 samples requested), adjusted for sex and baseline functional som sample (n = 987). The variables are socioeconomic conditions (SC), occupational class (OC), material adversity (M somatic symptoms (FSS), at four points in time—respondents aged 16, 21, 30 and 42. Table 3. Direct and indirect effects (5000 samples requested), adjusted for sex and baseline functional somatic symptoms, in the model for the full sample (n = 987). The variables are socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA) and functional somatic symptoms (FSS), at four points in time—respondents aged 16, 21, 30 and 42. Total effect a B (S.E.) CI β (S.E.) CI SC16 ! a Predictor estimates for the total, indirect, and speciﬁc indirect effects are unstandardized path coefﬁcients (B) with bootstrapped standard errors (S.E.) and bootstrapped 95% conﬁdence intervals (CI) and standardized path coefﬁcients (β) with bootstrapped standard errors (S.E.) and bootstrapped 95% conﬁdence intervals (CI). doi:10.1371/journal.pone.0155963.t003 ce te a s (C ) ctor estimates for the direct effects are unstandardized path coefﬁcients (S.E.) and 95% conﬁdence intervals (CI). and bootstrapped 95% conﬁdence intervals (CI) and standardized path coefﬁcients (β) with bootstrapped standard errors (S.E.) and bootstrapped 95% conﬁdence intervals (CI). b Predictor estimates for the direct effects are unstandardized path coefﬁcients (S E ) and 95% conﬁdence intervals (CI) doi:10.1371/journal.pone.0155963.t003 conﬁdence intervals (CI). b Predictor estimates for the direct effects are unstandardized path coefﬁcients (S.E.) and 95% conﬁdence intervals (CI). a Predictor estimates for the total, indirect, and speciﬁc indirect effects are unstandardized path coefﬁcients (B) with bootstrapped standard errors (S.E.) and bootstrapped 95% conﬁdence intervals (CI) and standardized path coefﬁcients (β) with bootstrapped standard errors (S.E.) and bootstrapped 95% doi:10.1371/journal.pone.0155963.t003 FSS42 0.425 (0.217) 0.001, 0.850 0.063 (0.032) 0.001, 0.125 Direct effects b Estimate CI SC16 ! FSS42 0.179 (0.246) -0.304, 0.662 SC16 ! OC21 0.773 (0.087) 0.606, 0.939 SC16 ! MA21 0.113 (0.044) 0.019, 0.206 SC16 ! SA21 0.070 (0.055) -0.038, 0.178 OC21 ! OC30 0.804 (0.078) 0.650, 0.957 OC21 ! MA30 0.250 (0.035) 0.182, 0.319 OC21 ! SA30 0.188 (0.038) 0.114, 0.263 OC21 ! FSS42 -0.019 (0.212) -0.434, 0.395 MA21 ! MA30 0.324 (0.050) 0.225, 0.422 MA21 ! SA30 0.115 (0.054) 0.010, 0.221 SA21 ! SA30 0.065 (0.037) -0.008, 0.139 OC30 ! FSS42 0.178 (0.158) -0.133, 0.489 MA30 ! FSS42 0.329 (0.141) 0.053, 0.606 SA30 ! FSS42 0.460 (0.123) 0.218, 0.701 Total indirect effect a B (S.E.) CI β (S.E.) CI SC16 ! FSS42 0.246 (0.109) 0.032, 0.461 0.036 (0.016) 0.005, 0.068 Speciﬁc indirect effect a SC16 ! OC21 ! OC30 ! FSS42 0.110 (0.102) -0.089, 0.310 0.016 (0.015) -0.013, 0.046 SC16 ! OC21 ! MA30 ! FSS42 0.064 (0.030) 0.004, 0.123 0.009 (0.004) 0.001, 0.018 SC16 ! OC21 ! SA30 ! FSS42 0.067 (0.024) 0.019, 0.114 0.010 (0.004) 0.003, 0.017 SC16 ! MA21 ! MA30 ! FSS42 0.012 (0.008) -0.004, 0.028 0.002 (0.001) -0.001, 0.004 SC16 ! MA21 ! SA30 ! FSS42 0.006 (0.005) -0.004, 0.016 0.001 (0.001) -0.001, 0.002 SC16 ! SA21 ! SA30 ! FSS42 0.002 (0.003) -0.003, 0.007 0.000 (0.000) 0.000, 0.001 a Predictor estimates for the total, indirect, and speciﬁc indirect effects are unstandardized path coefﬁcients (B) with bootstrapped standard errors (S.E.) and bootstrapped 95% conﬁdence intervals (CI) and standardized path coefﬁcients (β) with bootstrapped standard errors (S E ) and bootstrapped 95% PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 9 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms significantly predicted material as well as social adversity at age 30 (B = 0.250 and 0.188), which in turn related to functional somatic symptoms at age 42 (B = 0.329 and 0.460, respectively). With regard to the aim of the study, plausible chains of risks were examined assessing the indirect effects. First, in the full sample, the total effect which represent the overall impact of a person’s socioeconomic conditions at age 16 on functional somatic symptoms at age 42 (irre- spective of whether or not the effect runs through intervening variables) [62], was significant (B = 0.425, 95% CI = 0.001–0.850). The total indirect effect was also significant (B = 0.246, 95% CI = 0.032–0.461), suggesting a set of variables mediated the association (Table 3). Second, by assessing the specific indirect effects, the mediating role of a person’s occupa- tional class and their material and social adversity at age 21 and 30 in the direct association between socioeconomic conditions at age 16 and functional somatic symptoms at age 42, were obtained. The results suggested two plausible pathways from adolescent socioeconomic cir- cumstances: 1) via occupational class at age 21 and further through material adversity at age 30 (B = 0.064, 95% CI = 0.004–0.123), and 2) also through class at age 21 but then via social adver- sity at age 30 (B = 0.067, 95% CI = 0.019–0.114). Stratifying the model by sex (S1 Table) did not substantially alter the model fit (n = 473 and 514 for women and men respectively). Chi-square remained significant (x2 = 41.793 with p = 0.004 and df = 16), while all the other indices changed slightly (RMSEA = 0.057, CFI = 0.964 and WRMR = 0.829). When examined separately, the total indirect effect was significant for men (p = 0.008) but not for women (p = 0.507). Consequently, no path was evident for women, while for men an indirect pathway via occupational class at age 21 and social adversity at age 30 was significant (B = 0.093, 95% CI = 0.019–0.166). PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Discussion In light of all research pointing to the importance of early socioeconomic conditions for later health, to the best of our knowledge, only a few studies [18–21] have explicitly examined how growing up in a socioeconomically disadvantaged home may affect people’s health through chains of unfavorable life circumstances. Following the chain of risk hypothesis, our study adds to the field by demonstrating that adverse socioeconomic conditions of the family may contrib- ute to pathways of unfavorable material and social living situations, ultimately affecting health in adulthood. The socioeconomic circumstances under which one is brought up may structure the subse- quent life and affect health in many ways. Consistent with the sensitive period hypothesis, a large body of research suggests that health might be affected independently of or jointly with adult socioeconomic position [1–3, 7–17]. While our results follow these studies by proposing that the socioeconomic environment people grow up in might predict their occupational class as adults, we found no support for the idea that socioeconomic conditions of the family impact on health directly, or through a person’s own occupational class (Fig 3). Although this might be due to our fairly restrictive operationalization of early socioeconomic circumstances (parents’ occupation) or to our choice of outcome (functional somatic symptoms), it is possible that by only examining a person’s class in adulthood as a plausible pathway, previous studies have overlooked other explanatory circumstances along the life course [63]. Because as hypoth- esized (conceptual framework, Fig 1), this study suggests that the socioeconomic circumstances of the family can bring about a life of strained and stressful relational and financial situations, experiences that seem to stand for the more immediate effects on adult health. The sex stratified analyses (Table 1) revealed that women reported higher levels of func- tional somatic symptoms, but also more adversities than men. Although this is in line with PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 10 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Fig 3. Significant (p < 0.001) path coefficients in the model estimated with respect to the full sample (n = 987) while controlling for sex and baseline functional somatic symptoms. a Predictor estimates are probit regression coefficients. b Predictor estimates are unstandardized regression coefficients (B). Fig 3. Discussion Significant (p < 0.001) path coefficients in the model estimated with respect to the full sample (n = 987) while controlling for sex and baseline functional somatic symptoms. a Predictor estimates are probit regression coefficients. b Predictor estimates are unstandardized regression coefficients (B). doi:10.1371/journal.pone.0155963.g003 previous studies which suggest that health effects of social and material stressors may differ between the genders [54], the mechanisms are far from understood. Women’s structural disad- vantage with greater exposure to social and material stressors can explain some part, while the idea women would be more susceptible and vulnerable to these stressors provide limited insight as to why stress-related health problems differ between women and men [40]. In con- trast to these results, our sex stratified indirect effects suggested that by influencing the social environment in mid-adulthood, unfavorable socioeconomic circumstances early in life may increase the risk of later functional somatic symptoms, but only for men (S2 Table). Following the buffering hypothesis, notwithstanding it being fairly speculative, a reason for this finding might be that men generally have less social support and are consequently more affected by social adversity [64]. Even though poor health in adulthood may have its origins in, and be driven by the socio- economic environment of the family during adolescence, our results suggest this is not a sole and independent determinant. Instead of proposing a permanent and irreversible damage that may only be avoided through interventions early in life, the present study highlights the possi- bility that by breaking the subsequent links in the chain of risk, health effects along the life course might be altered. Thus, through interventions focused on improving the social and financial living conditions for people from impoverished backgrounds, we might have a chance to avert stress-related health problems later in life. Methodological considerations The over-all strengths of the study are the ability to examine a rather comprehensive theoretical model [52], assess mediation through prospective (spanning over 26 years) longitudinal data [62], a sample representativeness relative to the same age cohort in Sweden as a whole [44] and a reduced risk of selection bias due to the high response rate. However, our study has several limitations. First, while we developed a comprehensive model and adjusted for potential confounders, not all possible variables and paths have been PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 11 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms accounted for. We did not examine whether any measures at age 42 might have an impact on functional somatic symptoms, since this would have rendered the temporal sequence of the intermediate factors at age 21 and 30 years preceding the outcome at age 42 years less distinct. So while we tried to decrease the risk of omitted variable bias by estimating all the variables simultaneously [65] and acknowledged that they may share common omitted causes by corre- lating the residuals of contemporary measures [51] our estimates may still be biased as a result of variables not considered in the model [66]. Second, uncorrected measurement errors may introduce several problems in manifest path analysis [67]. Nevertheless, creating parcels enabled us to partially adjust for this potential problem, although operationalizing the adversity measures were limited to items available in the questionnaires. Thus, these variables have likely captured only a fraction of all possible hardship that may follow an unfavorable socioeconomic environment early in life. Also, by adopting a life course perspective we were required to operationalize class at age 21 which is a period of transition. Consequently, despite our class operationalization being guided by a stan- dard classification scheme [45], the measure at age 21 is debatable since we used education as a proxy for people that did not have a current occupation. In addition, we have made a differenti- ation between, and assume that class is temporally prior to our adversity measures by being a source for material and social resources [35, 37]. However, developing more comprehensive measures, including other items, approaching them as latent constructs, or their interconnec- tedness in another way, might have yielded quite different results [52]. Methodological considerations As methods to assess mediation are constantly evolving and improving, see for example VanderWeele [68], it is also possible that by using a causal mediation method to estimate our model (e.g. the one presented by Wang, Nelson and Albert [69]) the indirect effects might have been different. Third, in the process of specifying the model, alternative ideas were examined—one assess- ing autoregressive effects and another focused on health selection. When model fit was qualita- tively and subjectively compared between them, all proved to have a similarly acceptable fit. In the end, although all fit indices except the chi-square (a test that is sensitive to large sample sizes) indicated that the current model attained acceptable model-data correspondence, there are other versions that might fit the data equally well or better [70]. Lastly, growing up during adverse socioeconomic conditions was examined with regard to self-reported functional somatic symptoms in adulthood. The measure has been shown to have acceptable factorial invariance as well as internal consistency over time [71] but functional somatic symptoms is a complex concept and there is an ongoing debate about its nature, diag- nosis and impact [72]. Thus, while this health problem might be representative of other somatic problems linked to stress [73], our results may not be generalizable beyond this specific outcome. In addition, functional somatic symptoms were defined as a clustering of physical complaints with no or unknown pathology. Whether the measure is actually medically unex- plained is not certain, as the operationalization was based on items of self-reported symptoms, which have not been assessed in relation to the presence of an actual diagnosis. In addition, residual confounding may bias the results; for example, constitutional factors such as personal- ity, which has been shown to relate to functional somatic symptoms [74]. However, despite these shortcomings, our measure is comparable to those commonly used in population studies [75] and, to a fairly high degree, people reporting a multiple of these problems tend to be with- out a medical disorder that may explain their symptoms [26]. Conclusion The present study expands previous literature on the health effects of early socioeconomic dis- advantage by examining, and finding support for, the chain of risk life course hypothesis. 12 / 16 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms Specifically, our results indicate that growing up during unfavorable socioeconomic conditions might be a source for a chain of adverse material and social living situations along the life course, and these might be the circumstances that largely explain the effects of early disadvan- tage on health in adulthood. Instead of proposing a permanent and irreversible impact, the present study therefore highlights the possibility that interventions focused on improving the social and financial living conditions for people from impoverished backgrounds might help avert poor health in adulthood. Supporting Information S1 Table. Pearson’s correlations between all the variables for women (above diagonal) and men (below diagonal); socioeconomic conditions (SC), occupational class (OC), material adversity (MA), social adversity (SA) and functional somatic symptoms (FSS) at four points in time—respondents aged 16, 21, 30 and 42. (DOCX) S2 Table. Direct and indirect effects in the model (5000 samples requested), stratified by sex (n = 473 women, 514 men). The variables are socioeconomic conditions (SC), occupa- tional class (OC), material adversity (MA), social adversity (SA) and functional somatic symp- toms (FSS), at four points in time—respondents aged 16, 21, 30 and 42. (DOCX) Author Contributions Conceived and designed the experiments: FJ PEG MSS. Performed the experiments: FJ. Ana- lyzed the data: FJ. Wrote the paper: FJ PEG. Worked together with developing the analysis for the study: FJ LMJS. Worked together with developing the original idea of the ms and the inter- pretations of findings: FJ PEG MSS. Revising the manus critically for important intellectual content: FJ PEG MSS LMJS AH. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 References 1. Hyde M, Jakub H, Melchior M, Van Oort F, Weyers S. Comparison of the effects of low childhood socio- economic position and low adulthood socioeconomic position on self rated health in four European studies. J EPIDEMIOL COMMUN H. 2006; 60(10):882–6. 1. Hyde M, Jakub H, Melchior M, Van Oort F, Weyers S. Comparison of the effects of low childhood socio- economic position and low adulthood socioeconomic position on self rated health in four European studies. J EPIDEMIOL COMMUN H. 2006; 60(10):882–6. 2. Laaksonen M, Rahkonen O, Martikainen P, Lahelma E. Socioeconomic position and self-rated health: the contribution of childhood socioeconomic circumstances, adult socioeconomic status, and material resources. AM J PUBLIC HEALTH. 2005; 95(8):1403–09. PMID: 16006419 3. McKenzie SK, Carter KN, Blakely T, Ivory V. Effects of childhood socioeconomic position on subjective health and health behaviours in adulthood: how much is mediated by adult socioeconomic position? BMC PUBLIC HEALTH. 2011; 11(04):269–79. 4. Ben-Shlomo Y, Kuh D. A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. INT J EPIDEMIOL. 2002; 31(2):285–93. PMID: 11980781 5. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Davey Smith G. Indicators of socioeconomic position (part 1). J EPIDEMIOL COMMUN H. 2006; 60(1):7–12. 6. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Davey Smith G. Indicators of socioeconomic position (part 2). J EPIDEMIOL COMMUN H. 2006; 60(2):95–101. 7. Pudrovska T. Early-Life Socioeconomic Status and Mortality at Three Life Course Stages: An Increas- ing Within-Cohort Inequality. J HEALTH SOC BEHAV. 2014; 55(2):181–95. PMID: 24818953 8. Robertson T, Popham F, Benzeval M. Socioeconomic position across the lifecourse & allostatic load: data from the West of Scotland Twenty-07 cohort study. BMC PUBLIC HEALTH 2014; 14(02):184–93. 9. Harper S, Lynch JW, Hsu WL, Everson SA, Hillemeier MM, Raghunathan TE, et al. Life course socio- economic conditions and adult psychosocial functioning. INT J EPIDEMIOL. 2002; 31(2):395–403. PMID: 11980802 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 13 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms 10. Gustafsson PE, Janlert U, Theorell T, Westerlund H, Hammarström A. Socioeconomic status over the life course and allostatic load in adulthood: results from the Northern Swedish Cohort. J EPIDEMIOL COMMUN H. 2011; 65(11):986–92. 11. de Kok IM, van Lenthe FJ, Avendano M, Louwman M, Coebergh JW, Mackenbach JP. Childhood social class and cancer incidence: results of the globe study. SOC SCI MED. 2008; 66(5):1131–9. doi: 10. References 1016/j.socscimed.2007.11.035 PMID: 18164526 12. Otero-Rodriguez A, Leon-Munoz LM, Banegas JR, Guallar-Castillon P, Rodriguez-Artalejo F, Regidor E. Life-course socioeconomic position and change in quality of life among older adults: evidence for the role of a critical period, accumulation of exposure and social mobility. J EPIDEMIOL COMMUN H. 2011; 65(11):964–71. 13. Galobardes B, Lynch JW, Davey Smith G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. EPIDEMIOL REV. 2004; 26(1):7–21. 14. Galobardes B, Lynch JW, Davey Smith G. Is the association between childhood socioeconomic circum- stances and cause-specific mortality established? Update of a systematic review. J EPIDEMIOL COM- MUN H. 2008; 62(5):387–90. 15. Pollitt R, Rose K, Kaufman J. Evaluating the evidence for models of life course socioeconomic factors and cardiovascular outcomes: a systematic review. BMC PUBLIC HEALTH. 2005; 5(1):7–20. 16. Cohen S, Janicki-Deverts D, Chen E, Matthews KA. Childhood socioeconomic status and adult health. ANN NY ACAD SCI. 2010; 1186(1):37–55. 17. Galobardes B, Smith GD, Lynch JW. Systematic Review of the Influence of Childhood Socioeconomic Circumstances on Risk for Cardiovascular Disease in Adulthood. ANN EPIDEMIOL. 2006; 16(2):91– 104. PMID: 16257232 18. Tsenkova V, Pudrovska T, Karlamangla A. Childhood socioeconomic disadvantage and prediabetes and diabetes in later life: a study of biopsychosocial pathways. PSYCHOSOM MED. 2014; 76(8):622– 8. doi: 10.1097/PSY.0000000000000106 PMID: 25272201 19. Hagger-Johnson G, Mõttus R, Craig LCA, Starr JM, Deary IJ. Pathways from childhood intelligence and socioeconomic status to late-life cardiovascular disease risk. HEALTH PSYCHOL. 2012; 31 (4):403–12. doi: 10.1037/a0026775 PMID: 22309883 20. Gustafsson PE, Hammarstrom A. Socioeconomic disadvantage in adolescent women and metabolic syndrome in mid-adulthood: an examination of pathways of embodiment in the Northern Swedish Cohort. SOC SCI MED. 2012; 74(10):1630–8. doi: 10.1016/j.socscimed.2012.01.044 PMID: 22464906 21. Hertzman C, Power C, Matthews S, Manor O. Using an interactive framework of society and lifecourse to explain self-rated health in early adulthood. SOC SCI MED 2001; 53(12):1575–85. PMID: 11762884 22. Hamil-Luker J, Angela M. Gender differences in the link between childhood socioeconomic conditions and heart attack risk in adulthood. DEMOGRAPHY. 2007; 44(1):137–58. PMID: 17461340 23. Baum A, Garofalo J, Yali A. Socioeconomic status and chronic stress: does stress account for SES effects on health? ANN NY ACAD SCI. 1999; 896(1):131–44. 24. Kristenson M, Eriksen HR, Sluiter JK, Starke D, Ursin H. Psychobiological mechanisms of socioeco- nomic differences in health. SOC SCI MED 2004; 58(8):1511–22. PMID: 14759694 25. Cohen S, Janicki-Deverts D, Miller GE. Psychological stress and disease. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 References J AM MED ASSOC. 2007; 298(14):1685–87. 26. Kroenke K. Patients presenting with somatic complaints: epidemiology, psychiatric comorbidity and management. INT J METHOD PSYCH. 2003; 12(1):34–43. 27. van Gils A, Burton C, Bos EH, Janssens KA, Schoevers RA, Rosmalen JGM. Individual variation in temporal relationships between stress and functional somatic symptoms. J PSYCHOSOM RES 2014; 77(1):34–9. doi: 10.1016/j.jpsychores.2014.04.006 PMID: 24913339 28. Creed FH, Davies I, Jackson J, Littlewood A, Chew-Graham C, Tomenson B, et al. The epidemiology of multiple somatic symptoms. J PSYCHOSOM RES. 2012; 72(4):311–17. doi: 10.1016/j.jpsychores. 2012.01.009 PMID: 22405227 29. Lindström M, Fridh M, Rosvall M. Economic stress in childhood and adulthood, and poor psychological health: Three life course hypotheses. PSYCHIAT RES. 2014; 215(2):386–93. 30. Gruenewald TL, Karlamangla AS, Hu P, Stein-Merkin S, Crandall C, Koretz B, et al. History of socioeco- nomic disadvantage and allostatic load in later life. SOC SCI MED 2012; 74(1):75–83. doi: 10.1016/j. socscimed.2011.09.037 PMID: 22115943 31. Jonsson F, Hammarström A, Gustafsson PE. Social capital across the life course and functional somatic symptoms in mid-adulthood. SCAND J PUBLIC HEALTH. 2014; 42(7):581–8. doi: 10.1177/ 1403494814548749 PMID: 25204285 PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 14 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms 32. Beller E, Hout M. Intergenerational social mobility: The United States in comparative perspective. FUTURE CHILD. 2006; 16(2):19–36. PMID: 17036544 33. Chan TW, Goldthorpe JH. Class and Status: The Conceptual Distinction and its Empirical Relevance. AM SOCIOL REV. 2007; 72(4):512–32. 34. Wilkinson RG, Marmot MG. Social determinants of health: the solid facts. Copenhagen: World Health Organization; 2003. 35. Goldthorpe JH, McKnight A. The Economic Basis of Social Class. In: Morgan SL, Grusky DB, Fields GS, editors. Mobility and inequality: frontiers of research in sociology and economics Stanford, CA: Stanford University Press; 2006. p. 109–36. 36. Kristenson M. Socio-economic position and health: The role of coping. In: Siegrist J, Marmot M, editors. Social inequalities in health: new evidence and policy implications. Oxford: Oxford University Press; 2006. p. 127–52. 37. Krieger N, Williams DR, Moss NE. Measuring social class in US public health research: concepts, methodologies, and guidelines. ANNU REV PUBL HEALTH. 1997; 18(1):341–78. 38. Lynch JW, Kaplan GA. Socioeconomic position. In: Berkman LF, Kawachi I, editors. Social epidemiol- ogy. New York: Oxford University Press; 2000. p. 13–35. 39. Dohrenwend BP. Inventorying stressful life events as risk factors for psychopathology: Toward resolu- tion of the problem of intracategory variability. PSYCHOL BULL. References 2006; 132(3):477–95. PMID: 16719570 40. McDonough P, Walters V. Gender and health: reassessing patterns and explanations. SOC SCI MED. 2001; 52(4):547–59. PMID: 11206652 41. Amone-P'Olak K, Ormel J, Huisman M, Verhulst FC, Oldehinkel AJ, Burger H. Life stressors as media- tors of the relation between socioeconomic position and mental health problems in early adolescence: the TRAILS study. J AM ACAD CHILD PSY. 2009; 48(10):1031–8. 42. Lantz PM, House JS, Mero RP, Williams DR. Stress, life events, and socioeconomic disparities in health: results from the Americans' Changing Lives Study. J HEALTH SOC BEHAV 2005; 46(3):274– 88. PMID: 16259149 43. Kawachi I, Subramanian SV, Almeida-Filho N. A glossary for health inequalities. J EPIDEMIOL COM- MUN H. 2002; 56(9):647–52. 44. Hammarstrom A, Janlert U. Cohort profile: the northern Swedish cohort. INT J EPIDEMIOL. 2012; 41 (6):1545–52. doi: 10.1093/ije/dyr118 PMID: 21828110 45. Statistics Sweden. Swedish socioeconomic classification. Reports on statistical co-ordination 1982:4. Stockholm: Statistics Sweden; 1984. 46. Johansson S. The adult population’s state of health [in Swedish]. Stockholm: Fritzes; 1970. 47. Gustafsson PE, Janlert U, Theorell T, Westerlund H, Hammarström A. Social and material adversity from adolescence to adulthood and allostatic load in middle-aged women and men: results from the Northern Swedish Cohort. ANN BEHAV MED. 2012; 43(1):117–28. doi: 10.1007/s12160-011-9309-6 PMID: 22031214 48. San Sebastian M, Hammarstrom A, Gustafsson PE. Socioeconomic inequalities in functional somatic symptoms by social and material conditions at four life course periods in Sweden: a decomposition analysis. BMJ open. 2015; 5(8):e006581. doi: 10.1136/bmjopen-2014-006581 PMID: 26319773 49. Henderson S, Duncan-Jones P, Byrne DG, Scott R. Measuring social relationships. The Interview Schedule for Social Interaction. PSYCHOL MED. 1980; 10(4):723–34. PMID: 7208730 50. Karasek R, Theorell T. Healthy Work: Stress, productivity and the the reconstruction of working life. New York: Basic Books; 1990. 51. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. BEHAV RES METHODS. 2008; 40(3):879–91. PMID: 18697684 52. Kraemer HC, Stice E, Kazdin A, Offord D, Kupfer D. How do risk factors work together? Mediators, moderators, and independent, overlapping, and proxy risk factors. AM J PSYCHIATRY. 2001; 158 (6):848–56. PMID: 11384888 53. Rabkin JG, Struening EL. Life events, stress, and illness. SCIENCE. 1976; 194(4268):1013–20. 54. Denton M, Prus S, Walters V. Gender differences in health: a Canadian study of the psychosocial, structural and behavioural determinants of health. SOC SCI MED. 2004; 58(12):2585–600. PMID: 15081207 55. Matud MP. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 References Gender differences in stress and coping styles. PERS INDIV DIFFER. 2004; 37(7):1401– 15. 56. Muthén LK, Muthén BO. Mplus User’s Guide. Seventh Edition. Los Angeles, CA: Muthén & Muthén; 1998–2012. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 15 / 16 Adversity Pathways, Adolescent Socioeconomic Circumstances and Adult Functional Somatic Symptoms 57. MacKinnon DP, Cox MC. Commentary on "Mediation Analysis and Categorical Variables: The Final Frontier" by Dawn Iacobucci. J CONSUM PSYCHOL. 2012; 22(4):600–2. PMID: 23180961 58. Barrett P. Structural equation modelling: Adjudging model fit. PERS INDIV DIFFER. 2007; 42(5):815– 24. 59. Browne MW, Cudeck R. Alternative ways of assessing model fit. SOCIOL METHOD RES. 1992; 21 (2):230–58. 60. Yu C. Evaluating Cut-Off Criteria for Model Fit Indices for Latent Variable Models with Binary and Con- tinuous Outcomes. Los Angeles, University of California, 2002. Available: http://statmodel2.com/ download/Yudissertation.pdf. Unpublished doctoral dissertation. 2002. 61. Asparouhov T, Muthén B. Weighted least squares estimation with missing data. Mplus Technical Appendix. 2010:1–10. 62. Cole DA, Maxwell SE. Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling. J ABNORM PSYCHOL. 2003; 112(4):558–77. PMID: 14674869 63. Adler NE, Newman K. Socioeconomic disparities in health: pathways and policies. HEALTH AFFAIR. 2002; 21(2):60–76. 64. Dalgard OS, Dowrick C, Lehtinen V, Vazquez-Barquero JL, Casey P, Wilkinson G, et al. Negative life events, social support and gender difference in depression. SOC PSYCH PSYCH EPID. 2006; 41 (6):444–51. 65. MacKinnon DP, Lockwood CM, Williams J. Confidence Limits for the Indirect Effect: Distribution of the Product and Resampling Methods. MULTIVAR BEHAV RES. 2004; 39(1):99–128. 66. Tomarken AJ, Waller NG. Structural equation modeling: strengths, limitations, and misconceptions. ANNU REV CLIN PSYCHO. 2005; 1(04):31–65. 67. Cole DA, Preacher KJ. Manifest variable path analysis: potentially serious and misleading conse- quences due to uncorrected measurement error. PSYCHOL METHODS. 2014; 19(2):300–15. doi: 10. 1037/a0033805 PMID: 24079927 68. VanderWeele T. Explanation in causal inference: methods for mediation and interaction. New York: Oxford University Press; 2015. 69. Wang W, Nelson S, Albert JM. Estimation of causal mediation effects for a dichotomous outcome in multiple-mediator models using the mediation formula. STAT MED. 2013; 32(24):4211–28. doi: 10. 1002/sim.5830 PMID: 23650048 70. Kline RB. Principles and practice of structural equation modeling. New York: Guilford press; 2011. 71. Hammarström A, Westerlund H, Kirves K, Nygren K, Virtanen P, Hägglöf B. PLOS ONE \| DOI:10.1371/journal.pone.0155963 May 23, 2016 References Addressing challenges of validity and internal consistency of mental health measures in a 27- year longitudinal cohort study—the Northern Swedish Cohort study. BMC MED RES METHODOL. 2016; 16(1):1–9. 72. Voigt K, Nagel A, Meyer B, Langs G, Braukhaus C, Löwe B. Towards positive diagnostic criteria: A sys- tematic review of somatoform disorder diagnoses and suggestions for future classification. J PSYCHO- SOM RES. 2010; 68(5):403–14. doi: 10.1016/j.jpsychores.2010.01.015 PMID: 20403499 73. Tak LM, Rosmalen JGM. Dysfunction of stress responsive systems as a risk factor for functional somatic syndromes. J PSYCHOSOM RES. 2010; 68(5):461–8. doi: 10.1016/j.jpsychores.2009.12.004 PMID: 20403505 74. Katon W, Sullivan M, Walker E. Medical symptoms without identified pathology: relationship to psychi- atric disorders, childhood and adult trauma, and personality traits. ANN INTERN MED. 2001; 134(9 Pt 2):917–25. PMID: 11346329 75. Zijlema WL, Stolk RP, Lowe B, Rief W, White PD, Rosmalen JGM. How to assess common somatic symptoms in large-scale studies: a systematic review of questionnaires. J PSYCHOSOM RES. 2013; 74(6):459–68. doi: 10.1016/j.jpsychores.2013.03.093 PMID: 23731742 16 / 16 Copyright of PLoS ONE is the property of Public Library of Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
W2090851881.txt	null	en		A Scoring System to Predict the Risk of Postoperative Complications After Laparoscopic Gastrectomy for Gastric Cancer Based on a Large-Scale Retrospective Study	Medicine	2,015	cc-by	6,608	A Scoring System to Predict the Risk of Postoperative Complications After Laparoscopic Gastrectomy for Gastric Cancer Based on a Large-Scale Retrospective Study Chang-Ming Huang, MD, Ru-Hong Tu, MM, Jian-Xian Lin, MM, Chao-Hui Zheng, MM, Ping Li, MM, Jian-Wei Xie, MM, Jia-Bin Wang, MM, Jun Lu, MM, Qi-Yue Chen, MM, Long-Long Cao, MM, and Mi Lin, MM Abstract: To investigate the risk factors for postoperative complications following laparoscopic gastrectomy (LG) for gastric cancer and to use the risk factors to develop a predictive scoring system. Few studies have been designed to develop scoring systems to predict complications after LG for gastric cancer. We analyzed records of 2170 patients who underwent a LG for gastric cancer. A logistic regression model was used to identify the determinant variables and develop a predictive score. There were 2170 patients, of whom 299 (13.8%) developed overall complications and 78 (3.6%) developed major complications. A multivariate analysis showed the following adverse risk factors for overall complications: age 65 years, body mass index (BMI) 28 kg/m2, tumor withpyloricobstruction,tumorwithbleeding,andintraoperativebloodloss 75 mL; age 65 years, a Charlson comorbidity score 3, tumor with bleeding and intraoperative blood loss 75 mL were identified as independent risk factors for major complications. Based on these factors, the authors developed the following predictive score: low risk (no risk factors), intermediate risk (1 risk factor), and high risk (2 risk factors). The overall complication rates were 8.3%, 15.6%, and 29.9% for the low-, intermediate-, and high-risk categories, respectively (P < 0.001); the major complication rates in the 3 respective groups were 1.2%, 4.7%, and 10.0% (P < 0.001). This simple scoring system could accurately predict the risk of postoperative complications after LG for gastric cancer. The score might be helpful in the selection of risk-adapted interventions to improve surgical safety. (Medicine 94(17):e812) Abbreviations: ALB = albumin, AUC = area under the curve, BMI = body mass index, HB = hemoglobin, LG = laparoscopic gastrectomy, MVV = maximum ventilatory volume, ROC = receiver operating characteristic. Editor: Yong Zhang. Received: December 23, 2014; revised: April 2, 2015; accepted: April 3, 2015. From the Department of Gastric Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. Correspondence: Chang-Ming Huang, Department of Gastric Surgery, Fujian Medical University Union Hospital, No.29 Xinquan Road, Fuzhou 350001, Fujian Province, China (e-mail: hcmlr2002@163.com). The study was sponsored by the National Key Clinical Specialty Discipline Construction program of China (No. [2012]649) and Key Projects of Science and Technology Plan of Fujian Province (No. 2014Y0025). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors have no conflicts of interest to disclose. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. This is an open access article distributed under the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ISSN: 0025-7974 DOI: 10.1097/MD.0000000000000812 Medicine Volume 94, Number 17, May 2015 INTRODUCTION L aparoscopic techniques have been used to perform gastrectomies for gastric cancer since they were first reported for early gastric cancer in 1994.1 Surgeons have focused closely on the long-term results of laparoscopic gastrectomy (LG) for gastric cancer,2–4 and the safety of the procedure has been emphasized. Algorithms of standard treatments, such as the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology and Japanese Gastric Cancer Treatment Guidelines,5,6 have been published to clarify the operative indications. However, because of the differences in the epidemiology of gastric cancer and obesity between Caucasians and Asians, there is no consensus on the difficulty and safety of laparoscopic gastric surgery. Many studies have reported that morbidity rates for laparoscopic surgery range from 11.6% to 18.7%,7–9 although some centers have reported rates of 24.9% to 42.6%,3,4,10 which hampers the advancement and expanded use of the laparoscopic approach in the treatment of gastric cancer. The identification of patients at high risk for complications might allow the selection of a risk-adapted procedure, and intervening perioperative measures to reduce complications and increase the confidence of the surgeon; therefore, the development of a scoring system to predict the risk of complications is relevant. Few studies have been designed to develop scoring systems that accurately predict the risk of complications.11–15 Although the value of a scoring system in predicting complications in patients after an LG has been shown, these scoring systems frequently include many variables and might not be feasibly applicable in clinical practice. In addition, several risk factors have been associated with higher complication rates, such as the age, comorbidities, and body mass index (BMI) of the patient;9,16–22 to the best of our knowledge, there is no report of a simple scoring system to predict the risk when multiple risk factors are concurrent. The objective of this study was to identify the risk factors for postoperative complications after laparoscopic radical gastrectomy for gastric cancer in 2170 patients treated in our center. We aimed to use these risk factors to develop a scoring system to predict complications. MATERIALS AND METHODS Materials This study was a retrospective analysis of a prospectively collected database of 2170 primary gastric cancer patients treated with a laparoscopic radical gastrectomy in the Department of Gastric Surgery of Fujian Medical University Union Hospital, Fuzhou, China, between May 2007 and December 2013. The patient demographics, underlying diseases, clinicopathology, surgery data, and data on the preoperative and postoperative monitoring were recorded in a clinical data www.md-journal.com \| 1 Medicine Huang et al system for gastric cancer surgery.23 The staging was performed according to the 7th edition of the International Union against Cancer (UICC) tumor, lymph node, and distant metastasis (TNM) classification.24 The inclusion criteria were as follows: a histologically confirmed adenocarcinoma of the stomach; no evidence of tumors invading the adjacent organs (pancreas, spleen, liver, and transverse colon), paraaortic lymph node enlargement, or distant metastasis demonstrated by abdominal computed tomography and/or abdominal ultrasound and posteroanterior chest radiographs; and a D1 þ a/D1 þ b/D2 lymphadenectomy with curative R0 according to the pathological diagnosis after the operation. The exclusion criteria were as follows: intraoperative evidence of peritoneal dissemination, invasion of the adjacent organs, or a distant metastasis; conversion to an open laparotomy; and incomplete pathological data. The ethics committee of Fujian Union Hospital approved this retrospective study (Approval number: 20070428). All procedures were performed after obtaining written informed consent following an explanation of the surgical and oncological risks. The type of surgical resection (ie, a distal subtotal gastrectomy, proximal subtotal gastrectomy, or total gastrectomy) and the extent of lymph node dissection were selected according to the Japanese gastric cancer treatment guidelines,6 as reported in a detailed description in our previous study.25 Variables and Definitions The definition of each complication was based on the literature.26–34 Complications were classified according to the modified version of the Clavien–Dindo classification system reported by Dindo et al.35 A grade I complication was defined as any deviation from the normal postoperative course without requirement of pharmacological treatment or surgical, endoscopic, and radiological interventions (with the exceptions of drugs as antiemetics, antipyretics, analgetics, diuretics, electrolytes, or physiotherapy). A grade II complication was defined as any complication that requires pharmacological treatment with drugs other than such allowed for grade I complications (including blood transfusions and total parenteral nutrition). A grade III complication was defined as any complication requiring surgical, endoscopic, or radiological intervention, further subdivided into grades IIIa and IIIb depending on the need for general anesthesia. A grade IV complication was defined as any life-threatening complication (including central nervous system complications) requiring intermediate care/ intensive care unit management. Grade IV complications were subdivided into grades IVa and IVb, depending on whether the dysfunction was single- or multi-organ. A grade V complication indicated death of a patient due to a complication. The most severe complication was noted in the cases in which more than one complication occurred in a patient. Complications higher than grade III were defined as ‘‘major’’ complications that are potentially life threatening.10,35 The potential risk factors for postoperative complications were extracted from the database, including the sex, age, BMI, previous abdominal surgery, Charlson comorbidity score, hemoglobin (HB) level, albumin (ALB) level, maximum ventilatory volume (MVV), a tumor with pyloric obstruction (diagnosed by gastroscopy or computed tomography scan), tumor with bleeding (hematemesis, melena, or confirmation by gastroscopy), tumor location, tumor diameter, T stage, N stage, TNM stage, operative time (recorded from the skin incision to skin closure), intraoperative blood loss (estimated according to 2 \| www.md-journal.com Volume 94, Number 17, May 2015 the volume of blood absorbed by the gauze and suction pumped after subtracting the volume of fluids used for irrigation), type of surgical resection, type of reconstruction, D1þ/D2 lymphadenectomy, the number of resected lymph nodes, and the operative period (divided into 7 groups). Statistical Analysis The continuous data were reported as the mean SD, and the differences between the groups were analyzed using t tests. The categorical data were presented as the proportion and percentage and were analyzed with the chi-square test or Fisher’s exact test. The variables with P < 0.05 in the univariate analysis were subsequently included in a multivariate binary logistic regression model. The variables remaining significant (P < 0.05) in the multivariate analysis were used to construct a scoring system to classify the patients into groups according to their risk for complications. A P value < 0.05 was considered statistically significant. To assess how well the model could discriminate between patients with and without complications, a receiver operating characteristic (ROC) curve was calculated, and the area under the curve (AUC) was determined, shown as the absolute value and 95% confidence interval (95% CI). The AUC can be interpreted as the probability that a randomly chosen patient with complications will have a higher score than a randomly chosen patient without complications.36 The statistical analyses were performed with Statistical Program for Social Sciences (SPSS) version 18.0 (SPSS, Chicago, IL, USA). RESULTS Clinicopathological Characteristics of the Patients The clinicopathological characteristics of the 2170 patients are listed in Table 2. There were 1638 males and 532 females, with a mean age of 61.09 10.75 years. The average BMI of the patients was 22.19 3.07 kg/m2. There were 653 patients with a comorbidity (616 patients had a Charlson score of 1–2 points and 37 had a score of 3 points or higher). A total gastrectomy was performed in 1153 patients (53.1%), a distal gastrectomy in 963 patients (44.4%), and a proximal gastrectomy in 54 patients (2.5%); a D1þ lymphadenectomy or D2 lymphadenectomy was performed in 405 patients (18.7%) and 1765 patients (81.3%), respectively. The average surgery time was 180.70 51.54 minutes, including 191.03 50.19 minutes for a total gastrectomy, 169.17 50.95 minutes for a distal gastrectomy, and 153.78 32.80 minutes for a proximal gastrectomy. The blood loss was 73.67 106.95 mL, and the number of dissected lymph nodes per patient was 32.91 12.68. According to the UICC TNM Classification of Malignant Tumors, 7th Edition, 432 patients (19.9%) were in stage Ia, 199 (9.2%) were in stage Ib, 214 (9.9%) were in stage IIa, 247 (11.4%) were in stage IIb, 216 (10.0%) were in stage IIIa, 343 (15.8%) were in stage IIIb, and 519 (23.9%) were in stage IIIc. Postoperative Complications Table 1 shows the observed morbidities for all of the patients. Postoperative complications were observed in 299 patients (13.8%). Pneumonia (n ¼ 118, 5.4%), intra-abdominal abscess (n ¼ 43, 2.0%), and wound infection (n ¼ 38, 1.8%) were the most common problems among the overall complications. Major complications were observed in 78 patients (3.6%), among which local complications were present in 62.8% of the cases. Severe pneumonia (n ¼ 25, 1.1%), anastomotic leakage (n ¼ 14, Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. Medicine Volume 94, Number 17, May 2015 A Scoring System for Complications TABLE 1. Postoperative Morbidity After LG According to Clavien–Dindo Classification System Grades No. of Patients (%) Grade I Wound infection Chylous leak Grade II Anastomotic bleeding Abdominal bleeding Duodenal stump fistula Anastomotic leakage Pancreatic fistula Ileus Anastomotic stricture Remnant gastric stasis Wound infection Abdominal infection Chylous leak Pneumonia Arrhythmia Transient liver-enzyme abnormalities Urinary tract infection Sepsis Catheter-related infection DIC Grade IIIa Anastomotic bleeding Duodenal stump fistula Anastomotic leakage Ileus Wound infection Abdominal infection Chylous leak Pneumonia Grade IIIb Anastomotic bleeding Abdominal bleeding Adhesive intestinal obstruction Grade IVa Abdominal bleeding Anastomotic leakage Duodenal stump fistula Abdominal infection Ileus Cardiac failure Pneumonia Grade IVb Pneumonia, cardiac failurey DIC, cerebral infarctiony Grade V Infarct of spleen Abdominal infection Abdominal bleeding DIC Local complications System complications Overall complications 6 5 1 215 5 2 6 12 5 21 3 20 16 33 11 81 4 5 9 5 4 2 31 1 1 11 1 2 6 2 7 15 5 9 1 24 1 1 1 1 1 2 17 2 1 1 6 1 1 3 1 203 153 299 (0.3) (0.2) (0.0) (9.9) (0.2) (0.1) (0.3) (0.6) (0.2) (1.0) (0.1) (0.9) (0.7) (1.6) (0.5) (3.7) (0.2) (0.2) (0.4) (0.2) (0.2) (0.1) (1.4) (0.0) (0.0) (0.5) (0.0) (0.1) (0.3) (0.1) (0.3) (0.7) (0.2) (0.4) (0.0) (1.1) (0.0) (0.0) (0.0) (0.0) (0.0) (0.1) (0.8) (0.1) (0.0) (0.0) (0.3) (0.0) (0.0) (0.1) (0.0) (9.4%) (7.1%) (13.8%) LG ¼ laparoscopic gastrectomy. Disseminated intravascular coagulation (DIC). y Two most severe complications in the same grades occurred in a patient. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. 0.6%), and abdominal bleeding (n ¼ 13, 0.6%) requiring surgical, endoscopic, or radiological intervention were the major complications that occurred most frequently. A total of 21 patients required reoperation; the cause was abdominal bleeding in 12 cases, anastomotic bleeding in 5 cases, anastomotic leakage in 1 case, abdominal infection in 1 case, adhesive intestinal obstruction in 1 case, and splenic infarct in 1 case. Figure 1 shows the rates of local complications as well as the treatments for the complications. Six patients (0.3%) died following the surgery before the 30th postoperative day. The following causes of death were noted, anastomotic leakage and bleeding (2 patients); pancreatic fistula, anastomotic leakage, and bleeding (1 patient); severe pneumonia and abdominal infection (1 patient); splenic infarct (1 patient); and disseminated intravascular coagulation (1 patient). Univariable Analyses Associated with Complications Table 2 shows the results of the univariable analyses of the possible risk factors for the development of complications. Ten factors were associated with an increased risk of overall complications among 22 factors in total: age (P < 0.001), the Charlson comorbidity score (P ¼ 0.006), BMI (P ¼ 0.021), HB level (P ¼ 0.031), ALB level (P ¼ 0.026), tumor with pyloric obstruction (P ¼ 0.001), tumor with bleeding (P < 0.001), tumor diameter (P ¼ 0.031), intraoperative blood loss (P < 0.001), and operative period (P ¼ 0.011). Four factors were associated with major complications: age (P < 0.001), the Charlson comorbidity score (P < 0.001), tumor with bleeding (P ¼ 0.002), and intraoperative blood loss (P ¼ 0.005). Multivariate Analysis Associated with Overall Complications and the Scoring System The multivariate analysis revealed that age 65 years [odd ratio (OR) ¼ 2.016, P < 0.001], BMI 28 kg/m2 (OR ¼ 1.822, P ¼ 0.045), tumor with pyloric obstruction (OR ¼ 2.253, P ¼ 0.002), tumor with bleeding (OR ¼ 1.974, P < 0.001), and intraoperative blood loss 75 mL (OR ¼ 1.797, P < 0.001) were independent risk factors for overall complications (Table 3). Despite the differences in the regression coefficients, which ranged from 0.586 to 0.812, for simplicity, 1 point was assigned for each of the risk factors. Because fewer than 5% of the patients had 3 to 5 points, the following 3 risk groups were established low risk (0 points, ie, no risk factors), intermediate risk (1 point, ie, 1 risk factor), and high risk (2–5 points, ie, 2–5 risk factors). The distribution of the patients according to the scoring system was as follows low risk 47.3%, intermediate risk 38.3%, and high risk 14.4%. The incidence rates for overall complications among the patients in the low-, intermediate-, and high-risk categories were 8.3%, 15.6%, and 29.9%, respectively (P < 0.001). The relative risk of induction death in the intermediate- and high-risk groups compared with the low-risk category was 2.050 (95% CI, 1.533–2.741, P < 0.001) and 4.079 (95% CI, 2.919–5.699, P < 0.001), respectively (Table 4). Multivariate Analysis Associated with Major Complications and the Scoring System The multivariate analysis showed that age 65 years (OR ¼ 3.348, P < 0.001), the Charlson comorbidity score (OR ¼ 3.483, P ¼ 0.010), tumor with bleeding (OR ¼ 2.264, P ¼ 0.010), and intraoperative blood loss 75 mL (OR ¼ 1.882, P ¼ 0.015) were independent risk factors for www.md-journal.com \| 3 Medicine Huang et al Volume 94, Number 17, May 2015 FIGURE 1. The rates of the local complications and the treatments for the complications. TABLE 2. Univariable Analyses of Possible Risk Factors for the Development of Complications No. Patients Variables Age (year) <65 65 Sex Male Female BMI (kg/m2) <28 28 Previous abdominal surgery None Yes Charlson score 0 1–2 3 HB <60 60–90 90–120 120 ALB <28 28–35 35 MVV <60 60 Pyloric obstruction None Yes Tumor with bleeding None Yes Tumor diameter (mm) <60 60 4 \| www.md-journal.com Overall Complications (n ¼ 2170) (n ¼ 299) 1403 767 148 151 1638 532 236 63 2101 69 283 16 1853 317 254 45 1517 616 37 190 99 10 15 193 512 1450 2 33 83 181 52 362 1756 10 64 225 103 2067 17 282 2080 86 277 22 1988 182 257 42 1707 463 221 78 P Major Complications (n ¼ 78) <0.001 P <0.001 27 51 0.136 0.569 61 17 0.021 0.311 74 4 0.816 0.067 61 17 0.006 <0.001 46 26 6 0.031 0.171 0 9 24 45 0.026 0.072 3 18 57 0.749 0.213 6 72 0.001 0.234 73 5 0.002 <0.001 64 14 0.031 0.345 58 20 Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. Medicine Volume 94, Number 17, May 2015 A Scoring System for Complications No. Patients Variables Tumor location Upper Middle Lower 2 areas T stage T1 T2 T3 T4a N stage N0 N1 N2 N3 TNM stage IA IB IIA IIB IIIA IIIB IIIC Operative time (min) <120 120–180 180–240 240–300 300 IBL (mL) <75 75 Surgical resection Total Distal Proximate Reconstruction Roux-en-Y B-I B-II Esophagogastric Lymphadenectomy D1þ D2 No. of resected LNs <39 39 Operative period 2007 2008 2009 2010 2011 2012 2013 Overall Complications (n ¼ 2170) (n ¼ 299) 563 392 944 271 71 52 132 44 512 266 603 789 70 28 87 114 799 312 358 701 104 47 44 104 432 199 214 247 216 343 519 58 21 24 41 27 49 79 250 1253 471 124 72 27 177 58 21 16 1786 384 219 80 1153 963 54 163 132 4 1153 824 139 54 163 106 26 4 405 1765 55 244 1656 514 228 71 35 148 239 328 423 478 519 2 15 41 31 51 80 79 P Major Complications (n ¼ 78) 0.543 P 0.188 23 11 29 15 0.389 0.085 11 8 29 30 0.489 0.130 19 12 16 31 0.434 0.098 14 7 4 13 5 20 15 0.084 0.077 6 42 20 6 4 0.005 <0.001 55 23 0.373 0.171 48 28 2 0.145 0.178 48 25 3 2 0.898 0.869 14 64 0.979 0.689 61 17 0.011 0.222 2 3 6 12 15 18 22 ALB ¼ albumin level, BMI ¼ body mass index, HB ¼ hemoglobin level, IBL ¼ introoperative blood loss, MVV ¼ maximum ventilatory volume. Pyloric obstruction, tumor with pyloric obstruction. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. www.md-journal.com \| 5 Medicine Huang et al Volume 94, Number 17, May 2015 TABLE 3. Multivariate Analysis Associated with Complications Overall Complications b Variables Age 65 years Charlson score 3 BMI 28 kg/m2 Pyloric obstructiony Tumor with bleeding IBL 75 mL 0.701 / 0.600 0.812 0.680 0.586 Major Complications OR 95%CI P b OR 95%CI P 2.016 / 1.822 2.253 1.974 1.797 1.571–2.588 / 1.013–3.278 1.349–3.762 1.356–2.875 1.345–2.400 <0.001 / 0.045 0.002 <0.001 <0.001 0.817 1.284 / / 0.817 0.632 3.348 3.483 / / 2.264 1.882 2.070–5.415 1.349–8.992 / / 1.215–4.219 1.132–3.130 <0.001 0.010 / / 0.010 0.015 BMI ¼ body mass index, IBL ¼ introoperative blood loss. b, regression coefficients. y Pyloric obstruction, tumor with pyloric obstruction. TABLE 4. Scoring System for Overall Complications Risk score No. of Patients (n, %) No. of Risk Factors Low Intermediate High 1026 (47.3) 832 (38.3) 312 (14.4) 0 1 2 3 4 5 No. of Patients (n, %) 1026 832 281 30 1 0 (47.3) (38.3) (12.9) (1.4) (0.0) (0.0) major complications (Table 3). For simplicity, 1 point was assigned for each of these risk factors for which the regression coefficients ranged from 0.632 to 1.284. Because fewer than 5% of the patients had 3 to 4 points, the following 3 risk groups were established low-risk (0 points), intermediate-risk (1 point), and high-risk (2–4 points). The distribution of the patients according to the scoring system was as follows: low-risk 49.9%, intermediate-risk 38.2%, and high-risk 11.9%. The incidence rates of major complications among the patients in the low-, intermediate-, and high-risk categories were 1.2%, 4.7%, and 10.0%, respectively (P < 0.001). The relative risk of major complications in the intermediate-risk and high-risk groups compared with the low-risk group was 4.059 (95% CI, 2.153–7.656, P < 0.001) and 9.176 (95% CI, 4.646–18.125, P < 0.001), respectively (Table 5). Discrimination The score discriminated between patients with and without complications (overall and major complications) (Tables 4 and Complications (n, %) OR 95%CI P 85 (8.3) 130 (15.6) 84 (29.9) 1 2.050 4.079 / 1.533–2.741 2.919–5.699 / <0.001 <0.000 5). The area under the ROC curve was 0.641 (0.606–0.675) for the logistic regression model and 0.637 (0.602–0.671) for the simplified score for overall complications. In addition, the area under the ROC curve was 0.715 (0.658–0.772) for the logistic regression model and 0.707 (0.650–0.764) for the simplified score for major complications (Figure 2). DISCUSSION The development of laparoscopic devices and increased surgical experience has significantly increased the number of laparoscopic surgeries performed in gastric cancer patients. In the literature, reports of laparoscopic D2 lymph node dissections have shown the extent of lymph node dissection and demonstrated that the technical feasibility of the procedures is equivalent to those of open surgery, with no significant difference in the number of resected lymph nodes.37–39 Effectively improving LG safety is a global challenge. Surgery safety is subjective, and the incidence of postoperative complications is the most frequently used marker of surgery TABLE 5. Scoring System for Major Complications Risk Score No. of Patients (n, %) No. of Risk Factors Low Intermediate High 1082 (49.9) 829 (32.1) 259 (11.9) 0 1 2 3 4 6 \| www.md-journal.com No. of Patients (n, %) 1082 829 238 20 1 (49.9) (32.1) (11.0) (0.9) (0.0) Complications (n, %) OR 95%CI P 13 (1.2) 39 (4.7) 26 (10.0) 1 4.059 9.176 / 2.153–7.656 4.646–18.125 / <0.001 <0.000 Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. Medicine Volume 94, Number 17, May 2015 A Scoring System for Complications FIGURE 2. Receiver operating characteristic (ROC) curves for logistic regression model and scoring system predicting (A) overall complications, the area under the ROC curve was 0.641 (0.606–0.675) for the logistic regression model, and 0.637 (0.602–0.671) for the simplified score and (B) major complications, the area under the ROC curve was 0.715 (0.658–0.772) for the logistic regression model, and 0.707 (0.650–0.764) for the simplified score. safety.10 Significant differences in the definition and grading of complications have been reported for different surgeons and procedures as well as in surgeries within the same center. Recently, to overcome this problem, surgeons have used the Clavien–Dindo classification system for LG procedures. The system, which was revised and validated in a large cohort of patients who underwent general surgery, has been shown to be an objective and reliable tool for evaluating surgical safety and the severity of complications.35,40,41 In reports using this classification system, the rates of overall and major morbidity for laparoscopic surgery vary from 7.0% to 42.6% and 2.1% to 10.6%, respectively.10,17,42 –46 In this study, the overall and major morbidity rates were 13.8% and 3.6%, respectively. A method of predicting the risk of postoperative complications according to pre- and intra-operative risk factors and appropriate measures to reduce morbidity are needed. The risk factors associated with postoperative complications after an LG for gastric cancer are controversial. Ryu et al47 concluded that the degree of the lymph node dissection and surgical inexperience were risk factors for surgical complications after laparoscopy-assisted distal gastrectomy. Kunisaki et al48 reported that there is more surrounding tissue to separate and dissect in patients with a high BMI, particularly in patients with high visceral fat areas; obesity in these patients was associated with significantly higher rates of conversion to open surgery as well as postoperative complications, longer operation times, and greater blood loss. Kim et al49 showed that comorbidity, surgical inexperience, proximal reception, older age, and male sex were predictable risk factors for the occurrence of complications. From our data, we found that age 65 years, BMI 28 kg/m2, tumor with pyloric obstruction, tumor with bleeding, and intraoperative blood loss 75 mL were predictable risk factors for the occurrence of overall complications; age 65 years, a Charlson comorbidity score 3, tumor with bleeding, and intraoperative blood loss 75 mL were identified as independent risk factors for major complications. The patients with one or two comorbidities could Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. frequently tolerate a normal level of surgical stress with preoperative therapeutics and corrections in daily clinical practice. It is difficult to maintain a balance in physiological function with three or more comorbidities, and these patients frequently had major complications. Additionally, our study shows that more attention should be focused on elderly patients, particularly those with other risk factors, despite several recent studies on laparoscopic gastric surgery showing that gastrectomy in the elderly is safe and that older age alone should not be a contraindication to surgery.17,18 Our study showed a significantly higher risk of morbidity in patients with preoperative tumor complications, such as pyloric obstruction or tumor with bleeding. This population might be associated with a more advanced tumor stage and poorer nutritional status, which increases the surgical risks and rates of morbidity. In addition, intraoperative blood loss requires additional hemostasis by ligation and compression, and a massive hemorrhage might lead to hypovolemia; these conditions appeared to be associated with poor wound healing and increased infection rates from hypoxia.50–52 Although these risk factors were closely related to morbidity, few studies have been designed to create a simple scoring system to predict the risk of morbidity based on multiple risk factors. Previously reported scoring models, such as the Physiologic and Operative Severity Score for the Enumeration of Mortality, the National Surgical Quality Improvement Programme, and the Estimation of Physiologic Ability and Surgical Stress, have been reported to be useful for predicting complications. These scoring modes are not efficacious at the bedside because the models have many required parameters, with 66, 18, and 9 parameters in the National Surgical Quality Improvement Programme, Physiologic and Operative Severity Score for the Enumeration of Mortality, and Estimation of Physiologic Ability and Surgical Stress scoring models, respectively.11–14 The Surgical Apgar Score15 proposed by Miki uses the following intraoperative parameters: the estimated blood loss, lowest mean arterial pressure, and lowest heart rate. This scoring system is a useful predictor for the development of severe www.md-journal.com \| 7 Medicine Huang et al complications, whereas it is useless for selecting risk-adapted preoperative interventions. Our scoring system was based on the final logistic regression model. After giving the same weight to each predictor in the scoring system, the areas under the ROC curves for overall complications and major complications were 0.637 and 0.707, respectively. Both were similar to those in the logistic regression model, which had different weights (overall complications and major complications, 0.641 and 0.715, respectively). Concerning the risk stratification for morbidity, our scoring system classified the patients after LG into 3 groups and identified the highest risk group, which had a 4.1-fold higher risk of overall complications and a 9.2-fold higher risk of major complications than those of the lowest risk group. Patient and disease characteristics data are routinely available, which might have implications for selecting risk-adapted interventions to improve surgical safety. It is impossible to eliminate every risk factor for high-risk patients, such as age; correcting coincident risk factors that may be eliminated or improved by preoperative clinical therapeutics is useful for reducing the morbidity rates. For example, the aggressive treatment of comorbidities, including anemia and malnutrition caused by pyloric obstruction or tumor with bleeding, is required to improve the nutritional status of the patient. Surgical skill is required to identify the vasculature, nerves, and fascia as well as the specific fascial plane to minimize damage to the surrounding tissues and reduce blood loss. Furthermore, it is better to use different procedures for patients with different risks under the rule of complete resection. In addition, care as well as early diagnosis and treatments are necessary to decrease morbidity for high-risk patients. The score was raised in a large series of patients who underwent an LG for gastric cancer. There was a sufficient number of cases in each stage to apply the scoring model in early as well as advanced gastric cancer. Adopting the Clavien–Dindo classification system in our study demonstrated that the score could be easily validated and applied in other centers. The score could be helpful in training physicians in the selection of obvious candidates for laparoscopic surgery, predominantly those with low and intermediate risks of morbidity, which could increase the confidence of surgeons and facilitate progress on the surgery performance learning curve. The score could also facilitate the development of the LG technique for the method to become a universal surgical approach for patients with gastric cancer. The present study has some limitations. We evaluated patients by age, Charlson comorbidity score, HB level, ALB level, and MVV, but the performance status for some of our cases was not recorded, which might result in some biases. The model shows a good performance for major complications, and the area under the ROC curve for overall complications was approximately 0.65 with a 95% CI of less than 0.61. It might be important to develop a convincing prediction model for overall complications for ordinary patients. In conclusion, our scoring system allows for the easy risk stratification of morbidity in the clinical setting. This stratification might be helpful for selecting risk-adapted interventions to improve surgical safety. A prospective multiple-center study with a large series would provide valuable evidence for the validation of the score. ACKNOWLEDGEMENTS The authors are thankful to Fujian Medical University Union Hospital for her management of our gastric cancer patient database. 8 \| www.md-journal.com Volume 94, Number 17, May 2015 REFERENCES 1. Kitano S, Iso Y, Moriyama M, et al. Laparoscopy-assisted Billroth I gastrectomy. Surg Laparosc Endosc. 1994;4:146–148. 2. Kitano S, Shiraishi N, Uyama I, et al., Japanese Laparoscopic Surgery Study G. A multicenter study on oncologic outcome of laparoscopic gastrectomy for early cancer in Japan. Ann Surg. 2007;245:68–72. 3. Huscher CG, Mingoli A, Sgarzini G, et al. Laparoscopic versus open subtotal gastrectomy for distal gastric cancer: five-year results of a randomized prospective trial. Ann Surg. 2005;241:232–237. 4. Fujiwara M, Kodera Y, Misawa K, et al. Longterm outcomes of early-stage gastric carcinoma patients treated with laparoscopyassisted surgery. J Am Coll Surg. 2008;206:138–143. 5. National Comprehensive Cancer Network. Gastric Cancer version 02.2012. http://www.nccn.org/professionals/physician_gls/PDF/gastric.pdf. 6. Japanese Gastric Cancer, Association. Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer. 2011;14:113– 123. 7. Kim HH, Hyung WJ, Cho GS, et al. Morbidity and mortality of laparoscopic gastrectomy versus open gastrectomy for gastric cancer: an interim report – a phase III multicenter, prospective, randomized Trial (KLASS Trial). Ann Surg. 2010;251:417–420. 8. Yasunaga H, Horiguchi H, Kuwabara K, et al. Outcomes after laparoscopic or open distal gastrectomy for early-stage gastric cancer: a propensity-matched analysis. Ann Surg. 2013;257:640–646. 9. Jeong O, Ryu SY, Choi WY, et al. Risk factors and learning curve associated with postoperative morbidity of laparoscopic total gastrectomy for gastric carcinoma. Ann Surg Oncol. 2014;21:2994– 3001. 10. Lee JH, Park do J, Kim HH, et al. Comparison of complications after laparoscopy-assisted distal gastrectomy and open distal gastrectomy for gastric cancer using the Clavien-Dindo classification. Surg Endosc. 2012;26:1287–1295. 11. Copeland GP, Jones D, Walters M. POSSUM: a scoring system for surgical audit. Br J Surg. 1991;78:355–360. 12. Fink AS, Campbell DA Jr, Mentzer RM Jr et al. The National Surgical Quality Improvement Program in non-veterans administration hospitals: initial demonstration of feasibility. Ann Surg. 2002;236:344–353discussion 353–344.. 13. Khuri SF, Daley J, Henderson W, et al. The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, riskadjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA Surgical Quality Improvement Program. Ann Surg. 1998;228:491–507. 14. Haga Y, Ikei S, Ogawa M. Estimation of Physiologic Ability and Surgical Stress (E-PASS) as a new prediction scoring system for postoperative morbidity and mortality following elective gastrointestinal surgery. Surg Today. 1999;29:219–225. 15. Miki Y, Tokunaga M, Tanizawa Y, et al. Perioperative risk assessment for gastrectomy by surgical apgar score. Ann Surg Oncol Aug. 2014;21:2601–2607. 16. Kim MC, Kim W, Kim HH, et al. Risk factors associated with complication following laparoscopy-assisted gastrectomy for gastric cancer: a large-scale korean multicenter study. Ann Surg Oncol. 2008;15:2692–2700. 17. Kumagai K, Hiki N, Nunobe S, et al. Potentially fatal complications for elderly patients after laparoscopy-assisted distal gastrectomy. Gastric Cancer. 2014;17:548–555. 18. Hwang SH, Park do J, Jee YS, et al. Risk factors for operative complications in elderly patients during laparoscopy-assisted gastrectomy. J Am Coll Surg. 2009;208:186–192. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. Medicine Volume 94, Number 17, May 2015 19. Kodera Y, Sasako M, Yamamoto S, et al. Identification of risk factors for the development of complications following extended and superextended lymphadenectomies for gastric cancer. Br J Surg. 2005;92:1103–1109. 20. Kodera Y, Ito S, Yamamura Y, et al. Obesity and outcome of distal gastrectomy with D2 lymphadenectomy for carcinoma. Hepatogastroenterology. 2004;51:1225–1228. 21. Dhar DK, Kubota H, Tachibana M, et al. Body mass index determines the success of lymph node dissection and predicts the outcome of gastric carcinoma patients. Oncology. 2000;59:18–23. 22. Inagawa S, Adachi S, Oda T, et al. Effect of fat volume on postoperative complications and survival rate after D2 dissection for gastric cancer. Gastric Cancer. 2000;3:141–144. 23. Hu YF, Yu J, Zhang C, et al. Development and implementation of a clinical data mining system for gastric cancer surgery. Zhonghua Wei Chang Wai Ke Za Zhi. 2010;13:510–515. 24. Sobin LH, Gospodarowicz MK, Wittekind C. International Union Against Cancer (UICC) TNM Classification of Malignanttumours. 7th ed. New York: Wiley-Liss; 2010. A Scoring System for Complications 36. Gervaz P, Bandiera-Clerc C, Buchs NC. Scoring system to predict the risk of surgical-site infection after colorectal resection. Br J Surg. 2012;99:589–595. 37. Uyama I, Sugioka A, Fujita J, et al. Laparoscopic total gastrectomy with distal pancreatosplenectomy and D2 lymphadenectomy for advanced gastric cancer. Gastric Cancer. 1999;2:230–234. 38. Kawamura H, Homma S, Yokota R, et al. Inspection of safety and accuracy of D2 lymph node dissection in laparoscopy-assisted distal gastrectomy. World J Surg. 2008;32:2366–2370. 39. Kim MC, Jung GJ, Kim HH. Morbidity and mortality of laparoscopy-assisted gastrectomy with extraperigastric lymph node dissection for gastric cancer. Dig Dis Sci. 2007;52:543–548. 40. DeOliveira ML, Winter JM, Schafer M, et al. Assessment of complications after pancreatic surgery: a novel grading system applied to 633 patients undergoing pancreaticoduodenectomy. Ann Surg. 2006;244:931–937discussion 937–939. 41. Mazeh H, Samet Y, Abu-Wasel B, et al. Application of a novel severity grading system for surgical complications after colorectal resection. J Am Coll Surg. 2009;208:355–361. 25. Huang CM, Chen QY, Lin JX, et al. Laparoscopic spleen-preserving splenic hilar lymphadenectomy performed by following the perigastric fascias and the intrafascial space for advanced upper-third gastric cancer. PLoS One. 2014;9:e90345. 42. Lee JH, Ahn SH, Park do J, et al. Laparoscopic total gastrectomy with D2 lymphadenectomy for advanced gastric cancer. World J Surg. 2012;36:2394–2399. 26. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery. 2007;142:20–25. 43. Hayashi T, Yoshikawa T, Aoyama T, et al. Severity of complications after gastrectomy in elderly patients with gastric cancer. World J Surg. 2012;36:2139–2145. 27. Jung MR, Park YK, Seon JW, et al. Definition and classification of complications of gastrectomy for gastric cancer based on the accordion severity grading system. World J Surg. 2012;36:2400– 2411. 44. Li QG, Li P, Tang D, et al. Impact of postoperative complications on long-term survival after radical resection for gastric cancer. World J Gastroenterol. 2013;19:4060–4065. 28. Bruce J, Krukowski ZH, Al-Khairy G, et al. Systematic review of the definition and measurement of anastomotic leak after gastrointestinal surgery. Br J Surg. 2001;88:1157–1168. 45. Lee KG, Lee HJ, Yang JY, et al. Risk factors associated with complication following gastrectomy for gastric cancer: retrospective analysis of prospectively collected data based on the Clavien-Dindo system. J Gastrointest Surg. 2014;18:1269–1277. 29. Bassi C, Dervenis C, Butturini G, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery. 2005;138:8–13. 30. Orsenigo E, Bissolati M, Socci C, et al. Duodenal stump fistula after gastric surgery for malignancies: a retrospective analysis of risk factors in a single centre experience. Gastric Cancer. 2014;17:733– 744. 31. Horan TC, Gaynes RP, Martone WJ, et al. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol. 1992;13:606–608. 32. Assumpcao L, Cameron JL, Wolfgang CL, et al. Incidence and management of chyle leaks following pancreatic resection: a high volume single-center institutional experience. J Gastrointest Surg. 2008;12:1915–1923. 33. Holte K, Kehlet H. Postoperative ileus: a preventable event. Br J Surg. 2000;87:1480–1493. 34. Arozullah AM, Khuri SF, Henderson WG, et al. Participants in the National Veterans Affairs Surgical Quality Improvement P. Development and validation of a multifactorial risk index for predicting postoperative pneumonia after major noncardiac surgery. Ann intern Med. 2001;135:847–857. 35. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. 46. Tokunaga M, Kondo J, Tanizawa Y, et al. Postoperative intraabdominal complications assessed by the Clavien-Dindo classification following open and laparoscopy-assisted distal gastrectomy for early gastric cancer. J Gastrointest Surg. 2012;16:1854–1859. 47. Ryu KW, Kim YW, Lee JH, et al. Surgical complications and the risk factors of laparoscopy-assisted distal gastrectomy in early gastric cancer. Ann Surg Oncol. 2008;15:1625–1631. 48. Kunisaki C, Makino H, Takagawa R, et al. Predictive factors for surgical complications of laparoscopy-assisted distal gastrectomy for gastric cancer. Surg Endosc. 2009;23:2085–2093. 49. Kim W, Song KY, Lee HJ, et al. The impact of comorbidity on surgical outcomes in laparoscopy-assisted distal gastrectomy: a retrospective analysis of multicenter results. Ann Surg. 2008;248:793–799. 50. Hartmann M, Jonsson K, Zederfeldt B. Effect of tissue perfusion and oxygenation on accumulation of collagen in healing wounds. Randomized study in patients after major abdominal operations. Eur J Surg. 1992;158:521–526. 51. Esrig BC, Frazee L, Stephenson SF, et al. The predisposition to infection following hemorrhagic shock. Surg Gynecol Obstet. 1977;144:915–917. 52. Hunt TK, Zederfeldt BH, Goldstick TK, et al. Tissue oxygen tensions during controlled hemorrhage. Surg Forum. 1967;18:3–4. www.md-journal.com \| 9
https://openalex.org/W4226390578	https://www.researchsquare.com/article/rs-1015665/latest.pdf	English	null	Analysis of primary metabolites of Morchella fruit bodies and mycelium based on widely targeted metabolomics	Archives of microbiology	2,021	cc-by	6,729	Analysis of Primary Metabolites of Morchella Fruit Bodies And Mycelium Based On Widely Targeted Metabolomics Yuhong Yang Shenyang Agricultural University Jian Yang Shenyang Agricultural University Hongling Wang Shenyang Institute of Technology Yusong Jin Shenyang Agricultural University Jing Liu Shenyang Agricultural University Ranran Jia Shenyang Agricultural University Zhuo Wang Shenyang Institute of Physical Educa zongli kang (  401175060@qq.com Shenyang Agricultural University ht Yuhong Yang Shenyang Agricultural University Jian Yang Shenyang Agricultural University Hongling Wang Shenyang Institute of Technology Yusong Jin Shenyang Agricultural University Jing Liu Shenyang Agricultural University Ranran Jia Shenyang Agricultural University Zhuo Wang Shenyang Institute of Physical Education zongli kang (  401175060@qq.com ) Shenyang Agricultural University https://o Research Article Keywords: Morchella, widely targeted metabolomics, differential metabolism Posted Date: October 26th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-1015665/v1 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 Archives of Microbiology on December 29th, 2021. See the published version at https://doi.org/10.1007/s00203-021-02612-z. Page 1/19 Abstract Morchella is a kind of medicinal and edible homologous fungia that is rich in multiple metabolites. The metabolites from Morchella are a kind of essential substance because of their biological activities. In this study, Morchella fruit bodies and mycelium were selected to identify their metabolites. The primary metabolites of the two experimental group were analyzed using a method of widely targeted metabolome based on UPLC-ESI-MS/MS. A total of 354 different metabolites, including 188 upregulated metabolites and 166 downregulated metabolites, were characterized. Further, the main 20 metabolic pathways of the metabolites were analyzed. The first 9 ones are tyrosine metabolites, thyroid hormone biosynthetic pathway, phenylalanine metabolites, linoleic metabolites synthetic pathway, glycerophosphate metabolic pathway, choline in tumors, methyl butyl metabolites, arginine synthetic pathway, arginine, arginine and proline metabolites. This study provides theoretical basis for the analysis of metabolic pathway of Morchella fruit bodies and mycelium that serving for further research of their medicinal mechanism and effective components. Page 3/19 2.1. Materials Morchella fruit bodies: Collected from the planation base of Shenyang Agricultural University. Morchella mycelium: The Morchella strain was obtained from the Academy of Biological Science and Technology, Shenyang Agricultural University (Shenyang, China). Morchella mycelia were cultured on Potato Dextrose Agar medium in flasks at 28°C for 7 days in a rotary shaker at 120 rpm. The suspension was centrifuged and lyophilised to obtain the best fermented Morchella mycelium. 2.2 Sample Preparation Thaw the sample on ice. Take 50 mg of the sample and homogenize it with 1000 uL of ice-cold methanol/water (70%, v/v). Add cold steel balls to the mixture and homogenate for at 30Hz for 3 min. Whirl the mixture for 1 min, and then centrifuge it with 12,000 rpm at 4℃for 10 min. The supernatant collected will be used for LC-MS/MS analysis(Xian et al. 2008). 2.3 Analysis by LC-MS/MS The sample extracts were analyzed with an LC-ESI-MS/MS system (UPLC, Shim-pack UFLC SHIMADZU CBM A system, https://www.shimadzu.com/; MS, QTRAP® System, https://sciex.com/). The analytical conditions were as follows: UPLC: column, Waters ACQUITY UPLC HSS T3 C18 (1.8 µm, 2.1 mm100 Morchella fruit bodies: Collected from the planation base of Shenyang Agricultural University. Morchella mycelium: The Morchella strain was obtained from the Academy of Biological Science and Technology, Shenyang Agricultural University (Shenyang, China). Morchella mycelia were cultured on Potato Dextrose Agar medium in flasks at 28°C for 7 days in a rotary shaker at 120 rpm. The suspension was centrifuged and lyophilised to obtain the best fermented Morchella mycelium. 2.2 Sample Preparation Thaw the sample on ice. Take 50 mg of the sample and homogenize it with 1000 uL of ice-cold methanol/water (70%, v/v). Add cold steel balls to the mixture and homogenate for at 30Hz for 3 min. Whirl the mixture for 1 min, and then centrifuge it with 12,000 rpm at 4℃for 10 min. The supernatant collected will be used for LC-MS/MS analysis(Xian et al. 2008). 1. Introduction Morchella is well known for its high nutritional value and medicinal benefits(Kanwal and Reddy 2014). The fruit bodies and mycelium of Morchella are rich in nutrients, including protein, fat, carbohydrates, crude fiber, riboflavin, niacin, folic acid, vitamins and other components(Enkhjargal et al. 2011). Morchella is sweet in taste and can be used as medicine (Harpreet et al. 2011) which is famous for its antioxidant properties and shows extremely high medicinal value (B?Ckerman and Maliranta 2013). Morchella contains polysaccharides, enzymes, pyrone antibiotics, fatty acids and other chemical components(Gursoy et al. 2009). Modern medical research has shown that Morchella has various biology activities such as lowering blood lipids, regulating body immunity, anti-fatigue, protecting liver, anti-virus, inhibiting tumors, and reducing the side effects caused by radiotherapy and chemotherapy(Hai-Bin 2019). However, there was a lack of a thorough and dynamic evaluation of the identified metabolites in Morchella fruit bodies and mycelium. Morchella is well known for its high nutritional value and medicinal benefits(Kanwal and Reddy 2014). The fruit bodies and mycelium of Morchella are rich in nutrients, including protein, fat, carbohydrates, crude fiber, riboflavin, niacin, folic acid, vitamins and other components(Enkhjargal et al. 2011). Morchella is sweet in taste and can be used as medicine (Harpreet et al. 2011) which is famous for its antioxidant properties and shows extremely high medicinal value (B?Ckerman and Maliranta 2013). Morchella contains polysaccharides, enzymes, pyrone antibiotics, fatty acids and other chemical Metabolomics is a rapidly emerging discipline in bioomics and is an important part of systems biology(Widiarsih et al. 2021). Metabolomics research uses high-throughput chemical analysis technology to perform qualitative and quantitative analysis of small molecule metabolites in biological samples(Lin et al. 2011). Previous studies have shown that metabolomics is widely used in nutrition science, disease diagnosis, toxicology, plant metabolism and response mechanism and other aspects (Muazu et al. 2021). Wang et al.(Wang et al. 2018)identified the nutrients in black sesame seeds and related metabolites that play a role in traditional Chinese medicine based on extensively targeted metabolomics technology. Ho et al. (Ho et al. 2018) used metabolomics technology to identify 6 compounds with antibacterial activity from black walnut. Based on the quantitative analysis of metabolites, metabolomics can be used for the analysis of metabolic pathways or metabolic networks, the basic research of metabolism of the macroscopic phenotypic phenomena of different organisms, and the metabolites of different diseases, drugs and other physical, chemical or pathogenic organisms(Zhou et al. 1. Introduction 2020). the metabolites of different diseases, drugs and other physical, chemical or pathogenic organisms(Zhou et al. 2020). For the past few years, UPLC-ESI-MS/MS-based, widely targeted metabolome has become very popular in the field of analysis and identification of plant metabolites due to its advantages of high throughput, fast separation, high sensitivity, and wide coverage. UPLC-ESI-MS/MS-based widely targeted metabolome becomes an effective tool to deeply research secondary metabolites (Wang et al. 2019). Up to now, researchers have reported that the types of bioactive compounds in fungi are often different. However, there is a lack of research about the use of widely targeted metabolome to analyze the metabolic components of Morchella fruiting bodies and Morchella mycelium. In this study, we selected Morchella fruit bodies and mycelium as research materials, using ultra-high performance liquid chromatography tandem mass spectrometry technology to detect the types of metabolites of them. By means of comparing and analyzing their differences, it characterized the chemical substances in Morchella fruit bodies and mycelium from the perspective of different metabolites, providing reference for research on functional ingredients. Our study may be also help to understand the biological processes and mechanisms of the fruit bodies and mycelium more intuitively and effectively(Nadia et al. 2015) and might facilitate a deeply understanding of metabolites between Morchella fruit bodies and mycelium and provide a reference for their sufficient utilization in the future. 2.5 Data analysis Qualitative analysis of metabolites were based on MVDB V2.0 database of Maiwei Biotechnology Co., Ltd. and metabolite information in public databases. The primary and secondary mass spectrometry analysis were based on the existing MassBank, KNAPSAcK, HMDB and METLIN mass spectrometry databases(Isoherranen et al. 2009). A triple quadrupole mass spectrometry multiple reaction monitoring mode ( MRM ) was used for quantitative analysis of metabolites. After obtaining the metabolite data of different samples, use the software Analyst 1.6.3 to perform mass spectrometry qualitative and quantitative analysis, including baseline filtering, peak identification, integration, retention time correction, peak alignment, and mass spectrometry fragment attribution analysis(Bessa et al. 2013). The data was normalized and annotated based on the obtained retention time, mass-to-severity ratio and peak intensity. At the same time, most substances were further confirmed with standard products. Multi-dimensional statistical analysis was used to establish a reliable mathematical model to analyze Qualitative analysis of metabolites were based on MVDB V2.0 database of Maiwei Biotechnology Co., Ltd. and metabolite information in public databases. The primary and secondary mass spectrometry analysis were based on the existing MassBank, KNAPSAcK, HMDB and METLIN mass spectrometry databases(Isoherranen et al. 2009). A triple quadrupole mass spectrometry multiple reaction monitoring mode ( MRM ) was used for quantitative analysis of metabolites. After obtaining the metabolite data of different samples, use the software Analyst 1.6.3 to perform mass spectrometry qualitative and quantitative analysis, including baseline filtering, peak identification, integration, retention time correction, peak alignment, and mass spectrometry fragment attribution analysis(Bessa et al. 2013). The data was normalized and annotated based on the obtained retention time, mass-to-severity ratio and peak intensity. At the same time, most substances were further confirmed with standard products. Multi-dimensional statistical analysis was used to establish a reliable mathematical model to analyze metabolites(Markandan et al. 2021). First, use principal component analysis (PCA ) of unsupervised pattern recognition to analyze the detected metabolites to get a preliminary understanding of the overall metabolite differences between samples in each group and the degree of variability between samples within the group. Then use partial least squares-discriminant analysis ( PLS-DA ) with supervised pattern recognition to distinguish the overall differences in metabolites between groups to find different metabolites(Cozzolino et al. 2014). 2.3 Analysis by LC-MS/MS The sample extracts were analyzed with an LC-ESI-MS/MS system (UPLC, Shim-pack UFLC SHIMADZU CBM A system, https://www.shimadzu.com/; MS, QTRAP® System, https://sciex.com/). The analytical conditions were as follows: UPLC: column, Waters ACQUITY UPLC HSS T3 C18 (1.8 µm, 2.1 mm100 Page 3/19 mm); column temperature, 40 ℃; flow rate, 0.4 mL/min; injection volume, 2µL; solvent system, water (0.04% acetic acid): acetonitrile (0.04% acetic acid); gradient program, 95:5 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 14.0 min. mm); column temperature, 40 ℃; flow rate, 0.4 mL/min; injection volume, 2µL; solvent system, water (0.04% acetic acid): acetonitrile (0.04% acetic acid); gradient program, 95:5 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 14.0 min. LIT and triple quadrupole (QQQ) scans were acquired on a triple quadrupole-linear ion trap mass spectrometer (QTRAP), QTRAP® LC-MS/MS System, equipped with an ESI Turbo Ion-Spray interface, operating in positive and negative ion mode and controlled by Analyst 1.6.3 software (Sciex). The ESI source operation parameters were as follows: source temperature 500 ℃; ion spray voltage (IS) 5500 V (positive), -4500 V (negative); ion source gas I (GSI), gas II (GSII), curtain gas (CUR) were set at 55, 60, and 25.0 psi, respectively; the collision gas (CAD) was high. Instrument tuning and mass calibration were performed with 10 and 100 µmol/L polypropylene glycol solutions in QQQ and LIT modes, respectively. A specific set of MRM transitions were monitored for each period according to the metabolites eluted within this period(Oh et al. 2011) . 2.4 Quality control analysis Quality control samples ( QC ) were prepared by mixing sample extracts and were used to monitor the repeatability of analytical samples under the same processing method. In the process of instrumental analysis, a quality control sample was inserted into every 10 test analysis samples to monitor the repeatability of the analysis process(Bürger et al. 2012). 2.5 Data analysis Use orthogonal partial least squares discriminant analysis ( OPLS-DA ) to remove irrelevant differences to screen the difference variables to find the differences between the samples in each group, and the variable weight value (Variable importance in projection, VIP ) greater metabolites(Markandan et al. 2021). First, use principal component analysis (PCA ) of unsupervised pattern recognition to analyze the detected metabolites to get a preliminary understanding of the overall metabolite differences between samples in each group and the degree of variability between samples within the group. Then use partial least squares-discriminant analysis ( PLS-DA ) with supervised pattern recognition to distinguish the overall differences in metabolites between groups to find different metabolites(Cozzolino et al. 2014). Use orthogonal partial least squares discriminant analysis ( OPLS-DA ) to remove irrelevant differences to screen the difference variables to find the differences between the samples in each group, and the variable weight value (Variable importance in projection, VIP ) greater metabolites(Markandan et al. 2021). First, use principal component analysis (PCA ) of unsupervised pattern recognition to analyze the detected metabolites to get a preliminary understanding of the overall metabolite differences between samples in each group and the degree of variability between samples within the group. Then use partial least squares-discriminant analysis ( PLS-DA ) with supervised pattern recognition to distinguish the overall differences in metabolites between groups to find different metabolites(Cozzolino et al. 2014). Use orthogonal partial least squares discriminant analysis ( OPLS-DA ) to remove irrelevant differences to screen the difference variables to find the differences between the samples in each group, and the variable weight value (Variable importance in projection, VIP ) greater Page 4/19 Page 4/19 than 1 was considered to be a difference variable(Boccard and Rutledge 2013). Evaluation of PLA-DA and OPLS-DA predictive model parameters were R2 X, R2 Y and Q2 .The closer the 3 indicators were to 1, the more stable and reliable the model was. Use multi-dimensional statistics VIP value ( VIP>1 ), single- dimensional statistics ( P<0.05 ) and multiple of difference ( fold change ) to screen different metabolites. The multiple of difference was transformed by Log2, and select VIP>1, P<0.05, Log2FC ≥ 2 or metabolites with Log2FC ≤ 0.5 as differential metabolites(Notararigo et al. 2021) . 3.1 Overall analysis of metabolic components Based on Maiwei (Wuhan) self-built metabolite database and related mass spectrometry database, the multi-peak map of MRM metabolites was detected, the characteristic ions of each substance were screened through the triple quadrupole rod, and the signal intensity of the characteristic ions was obtained in the detector ( CPS ). Use MultiaQuant software to open the sample offline mass spectrometry file, and perform qualitative and quantitative analysis of the main metabolites. In Morchella fruit bodies and mycelium, 34 types of 610 metabolites were identified (Tab.1 ). Page 5/19 Table 1 Table 1 Primary metabolites identified in Morchella fruit bodies and mycelium Category of Metabolites Total Number of Metabolites (species) Proportion ( % ) Organic acids and their derivatives 122 20.0 Amino acids and their metabolites 103 16.8 Nucleotides and their metabolites 61 10.0 Benzene and its derivatives 61 10.0 Carbohydrates and their metabolites 42 6.9 Lipids and other fatty acids 32 5.2 Indole and its derivatives 18 2.9 Coenzymes and vitamins 18 2.9 Pyridine and its derivatives 18 2.8 Phenols and their derivatives 16 2.6 Lipids and other phospholipids 16 2.6 Fatty acyl 14 2.2 Oxidized lipid 12 1.9 Lipid 9 1.4 Heterocyclic compound 9 1.4 Ketones 6 0.9 Polyamine 5 0.8 Pteridine and its derivatives 5 0.8 Benzoic acid and its derivatives 5 0.8 Carnitines 4 0.6 Other 34 4.2 Metabonomics difference analysis of Morchella fruit ies and mycelium 1 PCA results PCA results can generally reflect the metabolite differences between the two groups of samples.Through the principal component analysis (PCA) of the samples, the degree of variability between the groups of Morchella fruit bodies and mycelium samples and between the samples within the group could be discriminated. The results contained three principal components(PCs), of which the contribution rate of PC1 (99.79%), PC2(0.15%) and PC3(0.03%) respectively.The two groups of samples showed a clear separation trend on the graph (Fig. 1). PCA results can generally reflect the metabolite differences between the two groups of samples.Through the principal component analysis (PCA) of the samples, the degree of variability between the groups of Morchella fruit bodies and mycelium samples and between the samples within the group could be discriminated. The results contained three principal components(PCs), of which the contribution rate of PC1 (99.79%), PC2(0.15%) and PC3(0.03%) respectively.The two groups of samples showed a clear separation trend on the graph (Fig. 1). 3.2.2 OPLS-DA results Although the PCA analysis method can effectively extract the main information, it is not sensitive to less correlated variables.The PLS-DA can maximize the distinction between groups, which is conducive to finding different metabolites(Peng et al. 2018). Orthogonal Partial Least Square Discriminant Analysis (OPLS- DA) combined orthogonal signal correction (OSC) and PLS-DA methods to screen difference variables by removing irrelevant differences. The OPLS-DA model (Fig. 2 ) of 610 metabolites set of data analysis showed that the sample distribution of Morchella fruit bodies was on the right side of the confidence interval, while mycelium samples was on the left side.The difference effect of the two samples was often obvious. Two principal components were analysed by OPLS-DA, of which the contribution rate were 98.6% and 0.623%. R2X=0.992, R2Y=1, Q2=1 (Fig. 3), showing that the grouping model had a strong interpretation and prediction ability, and the clustering results were reliable. The effect was better than the PCA model. Perform permutation verification to OPLS-DA( n=200, that is, conduct 200 permutation experiments). 3.2.3 Identification of Differential Metabolites In the obtained multivariate analysis of the variable importance in project ( VIP ) of the OPLS-DA model, VIP value represents the difference between the groups of the corresponding metabolites in the classification of each group of samples in the model. The intensity of the impact is considered to be significant generally for metabolites with VIP ≥ 1. The metabolites with differences between the two types of samples were initially screened, and then select the metabolites with VIP ≥ 1. Morchella fruit bodies and mycelium were screened to obtain a difference metabolite 32 class 354 species. Overall, the different metabolic components (354 kinds) of Morchella fruit bodies and mycelium accounted for 58.03% of total metabolic components (610 kinds), indicating the Morchella fruit bodies and the Morchella mycelial metabolites were significantly different. Among them, there were five categories of organic acids and derivatives, amino acids and their metabolites, nucleotides and their metabolites, benzene and their derivatives, and carbohydrates and their metabolites, accounting for 20.0%, 16.8%, 10.0%, 10.0% and 6.9% respectively . 3.3 Differential metabolite analysis 3.3.1 Analysis of main differences in metabolic components Page 7/19 Comparing the difference fold change of the quantitative information of the metabolic components in the fruit bodies and mycelium of Morchella and processing it by log2, the top 20 differentially expressed metabolic components are shown in Fig. 4.The relative content of Trans-4-Hydroxy-L-Proline, N-α-Acetyl-L- asparagine, 3-Iodo-L-Tyrosine,L-phenylalanyl-L-proline,N-Acetyl-L-alanine,N-amidino-L-aspart-ic acid, 3- Ketone-sphingosine, N-Phenylacetylglycine, 3-Methoxytyramine, and Ellagic acid were significantly high in mycelium.The relative content of N-methyle -phedrine,2-Furoylglycine,Herniarin,Caffeic Acid, Indolelactic acid, 3,5-Dimethoxy -4-Hydroxycinnamic Acid, 3-Hydroxy-DL-kynurenine, Xanthosine, Melatonin, and Lysopc 14:0 were significantly high in the fruit bodies. N-Methylephedrine is one of the main components of Chinese herbal medicine Ephedra, similar to ephedrine hydrochloride, with an effect of expanding the bronchus(Wei et al. 2009). Without the excitatory center of ephedrine co-acid salt, it has stronger anti- allergic and antitussive effects, and can accelerate heart rate and raise blood pressure(Holzgrabe et al. 2015), which is suitable for the treatment of influenza, bronchial wheezing, allergic reactions and other pathogens(Zhang et al. 2013). The mycelium and fruit bodies are rich in different active ingredients, mainly because of the difference between their development in liquid fermentation and solid cultivation, leading to different metabolites . 3.3.2 Differential metabolite Volcanic maps analysis In order to facilitate the observation of the changes of metabolites, normalize the metabolites with significant differences and draw a volcanic map. Differential metabolites (188 upregulated and 166 downregulated) were detected (Fig. 5 ), representing 53.11% and 46.89% (Tab. 2 ). Table 2 Statistical table of the number of differential metabolites groupName totalSigMetabolites Down-regulated Up-regulated Fruit_bodies_of_morchella_vs_ Fermented_mycelia_of_morchella 354 166 188 f ff Table 2 4. Discussion Morchella is an important kind of fungi that exhibits various bioactivity due to its variety of metabolites. However, its metabolites and main pathways have not been previously investigated. The utilization rate of Morchella is extremely inadequate. Thus, this study aimed to provide a theoretical basis for the utilization of Morchella. In this study, the primary metabolites of Morchella were comprehensively analyzed. The analysis of the experimental data showed that 354 different compounds were reported for the first time. The first 10 richest compounds in the fruit bodies and mycelium of Morchella were different and exhibited diverse efficacy. Comparison groups showed that the anti-4-hydroxyL-proline was higher in mycelium, which mainly used as flavor agent, nutritional intensifier, flavor material and biochemical reagents in present study. Aromatic amino acids, including phenylalanine, tyrosine and tryptophan, were important essential amino acids in organisms for their important biological functions. 3-iodide tyrosine and N-acetyl-L-alanine were richer in mycelium which could be widely used in fields of medicine, food and feed(Chen et al. 2019). N-methyl ephedrine rich in Morchella fruit bodies was one of the main Chinese medicinal herbs, that exhibited the effects of dilating bronchial lines and a stronger anti-allergy and antitussive. They were suitable for the treatment of influenza, bronchial wheezing, allergic reaction and other bacteria(A et al. 2021). 7- Comparison groups showed that the anti-4-hydroxyL-proline was higher in mycelium, which mainly used as flavor agent, nutritional intensifier, flavor material and biochemical reagents in present study. Aromatic amino acids, including phenylalanine, tyrosine and tryptophan, were important essential amino acids in organisms for their important biological functions. 3-iodide tyrosine and N-acetyl-L-alanine were richer in mycelium which could be widely used in fields of medicine, food and feed(Chen et al. 2019). N-methyl ephedrine rich in Morchella fruit bodies was one of the main Chinese medicinal herbs, that exhibited the effects of dilating bronchial lines and a stronger anti-allergy and antitussive. They were suitable for the treatment of influenza, bronchial wheezing, allergic reaction and other bacteria(A et al. 2021). 7- Methoxycoumarin, which could be used as an organic synthesis intermediate, was also a drug with the effect of flat asthma, expectorant, cough suppression(Han et al. 2021).The basic function of melatonin was to participate in the antioxidant system and prevent cells from causing oxidative damage ,which was the strongest endogenous free radical scavenger ever found. In this respect, it outperforms all known substances in vivo(Cruz et al. 2014). 3.4 Pathway analysis of differential metabolites The results of pathway enrichment analysis of differential metabolites through the KEGG(Kyoto Encyclopedia of Genes and Genomes)database (Zhang et al. 2017) ( Fig. 6 ) showed that the identified 354 significantly different metabolites were mainly distributed in 20 metabolic pathways .The first nine pathways with the largest number of differential metabolites were tyrosine metabolism ;thyroid hormone synthesis ; benzene Alanine metabolite synthesis pathway ( phenylalanine metabolis m ); linoleic acid metabolism synthesis pathway (linoleic acid metabolis m)glycerophospholipid metabolism (glycerophospholipid metabolism); choline metabolism in cancer (choline metabolis m in cancer); methyl butyrate metabolite synthesis pathway (butanoate metabolism); arginine synthetic pathway (arginine biosynthesis ); arginine and proline metabolites biosynthetic pathway ( arginine and proline metabolism ) Page 8/19 Acknowledgments This work was supported by the National Key Research and Development Program of China (Grant No. 2018YFC1801200). 4. Discussion The biosynthesis pathways of arginine and proline metabolites could be a significant increase in anti-4- hydroxyl-L-proline and L-amphetamine-L-proline in the mycelium(He et al. 2019). and provided new ideas and methods for the research and development of traditional Chinese medicine. The biosynthesis pathways of arginine and proline metabolites could be a significant increase in anti-4- hydroxyl-L-proline and L-amphetamine-L-proline in the mycelium(He et al. 2019). These findings have improved to our understanding of the molecular mechanisms accounting for the biological activities of Morchella fruit bodies and mycelium. It provides a basis for the drug research of Morchella to provide novel insight into further applications in drug industry. Conclusions In this paper, the primary metabolites in fruit bodies and mycelium of Morchella based on extensive target metabolomics were analyzed. 354 significantly different metabolites were identified in the study. The first 9 of which can be clarified. The relationship between the detected various compounds and the pharmacological activity of Morchella may be as an orientation of scientific research for the development and utilization of Morchella in the future. Metabolic group information can provide references for the separation and purification identification of many active components in Morchella. Associated metabolic pathways also provide a basis for subsequent studies of the function of key genes in the metabolic pathways and the biosynthesis of the major metabolites. Conflicts 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. Author contribution Conceptualization, YY and ZW; software, JY; formal analysis, HW; investigation, YJ; resources,RJ; data curation, JL; writing—original draft preparation, YY; writing—review and editing, YY; visualization, ZK; supervision, ZW. All authors have read and agreed to the published version of the manuscript. All authors have read and agreed to the published version of the manuscript. 4. Discussion Recent research proves that melatonin was the commander in chief of endocrine and it controlled the activity of various endocrine glands in the body, thus to indirectly control the function of our whole body(Cezary et al. 2021). The above analysis results revealed the differential metabolites between Morchella fruit bodies and mycelium that serving for the development and utilization of Morchella resources. Methoxycoumarin, which could be used as an organic synthesis intermediate, was also a drug with the effect of flat asthma, expectorant, cough suppression(Han et al. 2021).The basic function of melatonin was to participate in the antioxidant system and prevent cells from causing oxidative damage ,which was the strongest endogenous free radical scavenger ever found. In this respect, it outperforms all known substances in vivo(Cruz et al. 2014). Recent research proves that melatonin was the commander in chief of endocrine and it controlled the activity of various endocrine glands in the body, thus to indirectly control the function of our whole body(Cezary et al. 2021). The above analysis results revealed the differential metabolites between Morchella fruit bodies and mycelium that serving for the development and utilization of Morchella resources. The enrichment analysis of different metabolites showed that different metabolites mainly participated in the synthesis of tyrosine, biosynthesis of thyroid hormones, synthesis of phenylalanine metabolites synthesis, linoleic acid metabolites synthesis, glycerol phosphate metabolites synthesis, metabolic pathway of choline in tumors, methyl butyl metabolites synthesis, arginine synthesis pathway, biosynthesis of arginine and proline metabolites(Dobson et al. 2012) that were the main signal ways of Morchella metabolism. The ways directly affected the accumulation of the key substances. The phenylalanine and tyrosine metabolic pathways mainly determined phenolic difference metabolites. The changes of these material content were related to the accumulation of phenylalanine and tyrosine, and indirectly affected the umami flavor of Morchella. Thyroid hormone signaling pathway was associated with improving the efficacy of myocardial ischemia and provided a theoretical basis for its clinical application(Zeng et al. 2019). Therefore, it was speculated that the active components in Morchella may achieve the purpose of improving myocardial ischemia through the key factors acting in the above signaling pathway. It provided a basis for the drug research of Morchella to improve myocardial ischemia, Page 9/19 Page 9/19 and provided new ideas and methods for the research and development of traditional Chinese medicine. References Page 10/19 1. A JMA, E BGBCD, H CBAFG (2021) The role of virtual interactive simulators in medical education: Exploring their integration as an assessment methodology in clinical years. Educación Médica. DOI: 10.1016/j.edumed.2021.06.011 2. Petri Böckerman, Maliranta M (2013) Outsourcing, Occupational Restructuring, and Employee Well- Being: Is There a Silver Lining? Industrial Relations A Journal of Economy Society 52:878–914. DOI:10.1111/irel.12039 2. Petri Böckerman, Maliranta M (2013) Outsourcing, Occupational Restructuring, and Employee Well- Being: Is There a Silver Lining? Industrial Relations A Journal of Economy Society 52:878–914. DOI:10.1111/irel.12039 3. Bessa N et al (2013) Prospecção fitoquímica preliminar de plantas nativas do cerrado de uso popular medicinal pela comunidade rural do assentamento vale verde - Tocantins. Revbrasplantas Med 15:692–707. doi:10.1590/S1516-05722013000500010 DOI 4. Boccard J, Rutledge DN (2013) A consensus orthogonal partial least squares discriminant analysis (OPLS-DA) strategy for multiblock Omics data fusion. Anal Chim Acta 769:30–39. DOI:10.1016/j.aca.2013.01.022 5. Bürger S et al (2012) Uranium and plutonium analysis of nuclear material samples by multi-collector thermal ionisation mass spectrometry: Quality control, measurement uncertainty, and metrological traceability. Int J Mass Spectrom 311:40–50. DOI:10.1016/j.ijms.2011.11.016 6. Cezary C, Marta MS, Paulina K, Jan C, Aleksandra B (2021) The value of melatonin supplementation in postmenopausal women with Helicobacter pylori-associated dyspepsia. BMC women's health262- 267. DOI: 10.21203/rs.3.rs-31148/v3 7. Chen X et al (2019) Mechanism of Ginseng in Treatment of Myocardial Ischemia Based on Network Pharmacology. Evaluation and Analysis of Drug-Use in Hospitals of China19(01):9–13. DOI: 10.14009/j.issn.1672-2124.2019.01.003 7. Chen X et al (2019) Mechanism of Ginseng in Treatment of Myocardial Ischemia Based on Network Pharmacology. Evaluation and Analysis of Drug-Use in Hospitals of China19(01):9–13. DOI: 10.14009/j.issn.1672-2124.2019.01.003 8. Cozzolino D, Roumeliotis S, Eglinton J (2014) Combining Partial Least Squares (PLS) Discriminant Analysis and Rapid Visco Analyser (RVA) to Classify Barley Samples According to Year of Harvest and Locality. Food Anal Methods 7:887–892. DOI:10.1007/s12161-013-9696-3 9. Cruz MHC, Leal CLV, Cruz JFD, Tan DX, Reiter RJ (2014) Essential actions of melatonin in protecting the ovary from oxidative damage. Theriogenology 82(7):925–932. DOI:10.1016/j.theriogenology.2014.07.011 10. Dobson V, Alexander CJ (2012) Evaluation of different protocols for the analysis of lipophilic plant metabolites by gas chromatography-mass spectrometry using potato as a model. METABOLOMICS 2012,8(5):880-893. DOI: 10.1007/s11306-011-0383-3 10. Dobson V, Alexander CJ (2012) Evaluation of different protocols for the analysis of lipophilic plant metabolites by gas chromatography-mass spectrometry using potato as a model. METABOLOMICS 2012,8(5):880-893. DOI: 10.1007/s11306-011-0383-3 11. References Enkhjargal U, Jeng-Leun M (2011) Nutrient compositions of culinary-medicinal mushroom fruiting bodies and mycelia. International journal of medicinal mushrooms 13:343–349. DOI:10.1615/IntJMedMushr.v13.i4.40 11. Enkhjargal U, Jeng-Leun M (2011) Nutrient compositions of culinary-medicinal mushroom fruiting bodies and mycelia. International journal of medicinal mushrooms 13:343–349. DOI:10.1615/IntJMedMushr.v13.i4.40 12. Gursoy N, Sarikurkcu C, Cengiz M, Solak MH (2009) Antioxidant activities, metal contents, total phenolics and flavonoids of seven Morchella species. Food Chemical Toxicology 47:2381–2388 12. Gursoy N, Sarikurkcu C, Cengiz M, Solak MH (2009) Antioxidant activities, metal contents, total phenolics and flavonoids of seven Morchella species. Food Chemical Toxicology 47:2381–2388 13. Hai-Bin LI (2019) Effect of Morel on Free Radical Metabolism of Skeletal Muscle in Athletes. Edible Fungi of China 38(02):34–36. DOI:10.13629/j.cnki.53-1054.2019.02.010 Page 11/19 14. Han S, Yang L, Wang Y, Ran Y, Ding W (2021) Preliminary Studies on the Antibacterial Mechanism of a New Plant-Derived Compound, 7-Methoxycoumarin, Against Ralstonia solanacearum. Front Microbiol 12:697911. DOI:10.3389/fmicb.2021.697911 15. Harpreet et al (2011) Molecular Characterization of Morchella Species from the Western Himalayan Region of India. Curr Microbiol 62(4):1245–1252. DOI:10.1007/s00284-010-9849-1 15. Harpreet et al (2011) Molecular Characterization of Morchella Species from the Western Himalayan Region of India. Curr Microbiol 62(4):1245–1252. DOI:10.1007/s00284-010-9849-1 16. He H, Henderson AC, Du YL, Ryan KS (2019) Two-Enzyme Pathway Links L-Arginine to Nitric Oxide in N-Nitroso Biosynthesis. J Am Chem Soc 141:4026–4033. DOI:org/10.1021/jacs.8b13049 16. He H, Henderson AC, Du YL, Ryan KS (2019) Two-Enzyme Pathway Links L-Arginine to Nitric Oxide in N-Nitroso Biosynthesis. J Am Chem Soc 141:4026–4033. DOI:org/10.1021/jacs.8b13049 17. Ho KV et al (2018) Identifying Antibacterial Compounds in Black Walnuts (Juglans nigra) Using a Metabolomics Approach. Metabolites 8(4):58–69. DOI:10.3390/metabo8040058 17. Ho KV et al (2018) Identifying Antibacterial Compounds in Black Walnuts (Juglans nigra) Using a Metabolomics Approach. Metabolites 8(4):58–69. DOI:10.3390/metabo8040058 18. Holzgrabe U, Mallwitz H, Branch SK, Jefferies TM, Wiese M (2015) Chiral discrimination by NMR spectroscopy of ephedrine and N-methylephedrine induced by β‐cyclodextrin, heptakis(2,3‐di‐O‐ acetyl)β‐cyclodextrin, and heptakis (6‐O‐acetyl)β‐cyclodextrin. Chirality 9:56-61. DOI: 10.1002/(SICI)1520-636X(1997)9:3<211::AID-CHIR2>3.0.CO;2-I 18. Holzgrabe U, Mallwitz H, Branch SK, Jefferies TM, Wiese M (2015) Chiral discrimination by NMR spectroscopy of ephedrine and N-methylephedrine induced by β‐cyclodextrin, heptakis(2,3‐di‐O‐ acetyl)β‐cyclodextrin, and heptakis (6‐O‐acetyl)β‐cyclodextrin. Chirality 9:56-61. DOI: 10.1002/(SICI)1520-636X(1997)9:3<211::AID-CHIR2>3.0.CO;2-I 19. Isoherranen N, Ac Had H, Yeung H, Levy CK RH (2009) Qualitative analysis of the role of metabolites in inhibitory drug-drug interactions: literature evaluation based on the metabolism and transport drug interaction database. Chem Res Toxicol 22:294–298. DOI:10.1021/tx800491e 19. References Isoherranen N, Ac Had H, Yeung H, Levy CK RH (2009) Qualitative analysis of the role of metabolites in inhibitory drug-drug interactions: literature evaluation based on the metabolism and transport drug interaction database. Chem Res Toxicol 22:294–298. DOI:10.1021/tx800491e 20. Kanwal HK, Reddy MS (2014) Influence of sclerotia formation on ligninolytic enzyme production in Morchella crassipes. J Basic Microbiol 54:S63–S69. DOI:10.1002/jobm.201200802 20. Kanwal HK, Reddy MS (2014) Influence of sclerotia formation on ligninolytic enzyme production in Morchella crassipes. J Basic Microbiol 54:S63–S69. DOI:10.1002/jobm.201200802 21. Lin X et al (2011) A method for handling metabonomics data from liquid chromatography/mass spectrometry: combinational use of support vector machine recursive feature elimination, genetic algorithm and random forest for feature selection. Metabolomics 7:549–558. DOI:10.1007/s11306- 011-0274-7 22. Muazu DM, Toyin MO, Abdulhamid AK (2021) Chemosystematic Evaluation of Some Nigerian Gossypium hirsutum L. Using Qualitative and Quantitative Phytochemical Analysis. Borneo Journal of Resource Science Technology 11:43–50 DOI: org/10.33736/bjrst.2728.2021 23. Nadia T, Tai H, Da-Woon J, Williams DR (2015) Fishing for Nature's Hits: Establishment of the Zebrafish as a Model for Screening Antidiabetic Natural Products. Evidence-Based Complementray and Alternative Medicine, 2015:1-16. DOI: 10.1155/2015/287847 23. Nadia T, Tai H, Da-Woon J, Williams DR (2015) Fishing for Nature's Hits: Establishment of the Zebrafish as a Model for Screening Antidiabetic Natural Products. Evidence-Based Complementray and Alternative Medicine, 2015:1-16. DOI: 10.1155/2015/287847 24. Oh E, Lu M, Park C, Han BO, Lee J (2011) Dynamic modeling of lactic acid fermentation metabolism with Lactococcus lactis. Journal of Microbiology Biotechnology 21:162–169. DOI:10.4014/jmb.1007.07066 25. Peng Y, Zhao F, Li L, Xing Y, Fang X (2018) Discrimination of heat-damaged tomato seeds based on near infrared spectroscopy and PCA-SVM method. Transactions of the Chinese Society of Agricultural Engineering 05(34):167–173. DOI:10.11975/j.issn.1002-6819.2018.05.021 25. Peng Y, Zhao F, Li L, Xing Y, Fang X (2018) Discrimination of heat-damaged tomato seeds based on near infrared spectroscopy and PCA-SVM method. Transactions of the Chinese Society of Agricultural Engineering 05(34):167–173. DOI:10.11975/j.issn.1002-6819.2018.05.021 26. Wang D et al (2018) Identification of Nutritional Components in Black Sesame Determined by Widely Targeted Metabolomics and Traditional Chinese Medicines. Molecules 23:1180–1189. DOI:10.3390/molecules23051180 26. Wang D et al (2018) Identification of Nutritional Components in Black Sesame Determined by Widely Targeted Metabolomics and Traditional Chinese Medicines. Molecules 23:1180–1189. DOI:10.3390/molecules23051180 Page 12/19 Page 12/19 27. Wang F, Chen L, Chen H, Chen S, Liu Y (2019) Analysis of Flavonoid Metabolites in Citrus Peels (Citrus reticulata "Dahongpao") Using UPLC-ESI-MS/MS. Molecules 24:2680–2692. DOI:10.3390/molecules24152680 28. References Wei Z, Deng AJ, Du GH, Zhang JL, Qin HL (2009) Chemical constituents of the stems of Ephedra sinica. J Asian Nat Prod Res 11:168–171. DOI:10.1080/10286020802573552 29. Widiarsih S, Nagel M, Brner A, Feussner K, Feussner I, Mllers C (2021) Inheritance of seed quality and seed germination in two doubled haploid populations of oilseed rape segregating for acid detergent lignin (ADL) content. Euphytica 217:1–16 DOI: org/10.1007/s10681-021-02891-z 30. Xian Z et al (2008) Induction of hepatic enzymes and oxidative stress in Chinese rare minnow (Gobiocypris rarus) exposed to waterborne hexabromocyclododecane (HBCDD). Aquat Toxicol 86:4– 11. DOI:10.1016/j.aquatox.2007.07.002 31. Zeng B, Liu L, Liao X, Zhang C, Ruan H (2019) Thyroid hormone protects cardiomyocytes from H2O2- induced oxidative stress via the PI3K-AKT signaling pathway. Exp Cell Res 380:205–215. DOI:10.1016/j.yexcr.2019.05.003 31. Zeng B, Liu L, Liao X, Zhang C, Ruan H (2019) Thyroid hormone protects cardiomyocytes from H2O2- induced oxidative stress via the PI3K-AKT signaling pathway. Exp Cell Res 380:205–215. DOI:10.1016/j.yexcr.2019.05.003 32. Zhang H, Zhang X, Huang J, Fan X (2017) Identification of key genes and pathways for peri- implantitis through the analysis of gene expression data. Experimental Therapeutic Medicine 13(5):1832–1840. DOI:10.3892/etm.2017.4176 32. Zhang H, Zhang X, Huang J, Fan X (2017) Identification of key genes and pathways for peri- implantitis through the analysis of gene expression data. Experimental Therapeutic Medicine 13(5):1832–1840. DOI:10.3892/etm.2017.4176 33. Zhang L, Zhang F, Wang Z, Jiang Y, Meng XU, Hong LI (2013) Simultaneous determination of ephedrine and N-methylephedrine in urine by solid phase extraction-ultra performance liquid chromatography-electrospray ionization tandem mass spectrometry. Chin J Chromatogr 31:898– 902. DOI:10.3724/sp.j.1123.2013.02028 34. Zhou Y, Liu H, Zhang M (2020) Analysis of the metabolic pathways affected by hot-Humid or dry climate based on fecal metabolomics coupled with serum metabolic changes in broiler chickens. Poult Sci 99–106. DOI:10.1016/j.psj.2020.07.039 34. Zhou Y, Liu H, Zhang M (2020) Analysis of the metabolic pathways affected by hot-Humid or dry climate based on fecal metabolomics coupled with serum metabolic changes in broiler chickens. Poult Sci 99–106. DOI:10.1016/j.psj.2020.07.039 Figures Figures Page 13/19 Figure 1 PCA score chart of the total sample Figure 1 PCA score chart of the total sample Page 14/19 Figure 1 PCA score chart of the total sample Figure 1 PCA score chart of the total sample PCA score chart of the total sample Page 14/19 Figure 2 Page 15/19 Page 15/19 Page 16/19 Figure 3 OPLS-DA verification diagram Figure 3 OPLS-DA verification diagram Figure 3 OPLS-DA verification diagram Figure 3 OPLS-DA verification diagram OPLS-DA verification diagram Page 16/19 Figure 4 Figure 4 The 20 metabolites with the largest differences in content between the two samples Note: The top 10 metabolic components with the largest (red) and smallest (green) multiples of difference between morel fruit body and mycelium . MEDP083 : trans- 4 -hydroxy -L- proline; MEDP083 : N- α - acetyl- L- asparagine; MEDP033 : 3- iodotyrosine, MEDP083 : L- phenylpropan -L- proline acid ; MEDP525 : N- acetyl- -L- alanine; MEDP886 : N- amidino- L- aspartic acid; MEDP592 : 3- keto - sphingosine; MEDP073 : phenylacetyl glycine;MEDP894 : 3 -methoxytyramine; MEDN627 : ellagic acid; MEDP290 : sinapinic acid ; MEDN301 ; caffeic acid; MEDN523 : 3 -indole - lactic acid ; MEDN093 : 3- hydroxy -2- aminobenzoic acid ; MEDP788 : two Aniline; MEDP556 : 7 -methoxycoumarin; MEDP551 : N -methylephedrine; MEDP892 : Catechol; MEDP680 : hexylamine; MEDP617 : N-(2- furoyl ) glycine Figure 4 The 20 metabolites with the largest differences in content between the two samples Note: The top 10 metabolic components with the largest (red) and smallest (green) multiples of difference between morel fruit body and mycelium . MEDP083 : trans- 4 -hydroxy -L- proline; MEDP083 : N- α - acetyl- L- asparagine; MEDP033 : 3- iodotyrosine, MEDP083 : L- phenylpropan -L- proline acid ; MEDP525 : N- acetyl- -L- alanine; MEDP886 : N- amidino- L- aspartic acid; MEDP592 : 3- keto - sphingosine; MEDP073 : phenylacetyl glycine;MEDP894 : 3 -methoxytyramine; MEDN627 : ellagic acid; MEDP290 : sinapinic acid ; MEDN301 ; caffeic acid; MEDN523 : 3 -indole - lactic acid ; MEDN093 : 3- hydroxy -2- aminobenzoic acid ; MEDP788 : two Aniline; MEDP556 : 7 -methoxycoumarin; MEDP551 : N -methylephedrine; MEDP892 : Catechol; MEDP680 : hexylamine; MEDP617 : N-(2- furoyl ) glycine Page 17/19 Figure 5 Volcanic maps of differential metabolites Note: Green dots represent downregulateddifferentially expressed metabolites; red spots represented upregulated differentially expressed metabolites; and black represented non-differentially expressed metabolites. Page 18/19 Page 18/19 Figure 6 Differential metabolite KEGG enrichment map Note: The abscissa represents the rich factor corresponding to each channel , the ordinate is the channel name, and the color of the point is pvalue . The more red, the more significant the enrichment. The size of the dot represents the number of different metabolites enriched. Figure 6 Differential metabolite KEGG enrichment map Note: The abscissa represents the rich factor corresponding to each channel , the ordinate is the channel name, and the color of the point is pvalue . The more red, the more significant the enrichment. The size of the dot represents the number of different metabolites enriched. Differential metabolite KEGG enrichment map Note: The abscissa represents the rich factor corresponding to each channel , the ordinate is the channel name, and the color of the point is pvalue . The more red, the more significant the enrichment. The size of the dot represents the number of different metabolites enriched. Differential metabolite KEGG enrichment map Note: The abscissa represents the rich factor corresponding to each channel , the ordinate is the channel name, and the color of the point is pvalue . The more red, the more significant the enrichment. The size of the dot represents the number of different metabolites enriched. Page 19/19
https://openalex.org/W2051959783	https://europepmc.org/articles/pmc3739749?pdf=render	English	null	xopAC-triggered Immunity against Xanthomonas Depends on Arabidopsis Receptor-Like Cytoplasmic Kinase Genes PBL2 and RIPK	PloS one	2,013	cc-by	11,663	Abstract * E-mail: laurent.noel@toulouse.inra.fr ¤ Current address: CIRAD, UMR BGPI, Montpellier, France ¤ Current address: CIRAD, UMR BGPI, Montpellier, France Endrick Guy1,2, Martine Lautier1,2,3, Matthieu Chabannes1,2¤, Brice Roux1,2, Emmanuelle Lauber1,2, Matthieu Arlat1,2,3, Laurent D. Noël1,2* 1 INRA, Laboratoire des Interactions Plantes Micro-organismes (LIPM), UMR 441, Castanet-Tolosan, France, 2 CNRS, Laboratoire des Interactions Plantes Micro-organismes (LIPM), UMR 2594, Castanet-Tolosan, France, 3 Université de Toulouse, Université Paul Sabatier, Toulouse, France 1 INRA, Laboratoire des Interactions Plantes Micro-organismes (LIPM), UMR 441, Castanet-Tolosan, France, 2 CNRS, La Micro-organismes (LIPM), UMR 2594, Castanet-Tolosan, France, 3 Université de Toulouse, Université Paul Sabatier, Toulous Abstract Xanthomonas campestris pv. campestris (Xcc) colonizes the vascular system of Brassicaceae and ultimately causes black rot. In susceptible Arabidopsis plants, XopAC type III effector inhibits by uridylylation positive regulators of the PAMP-triggered immunity such as the receptor-like cytoplasmic kinases (RLCK) BIK1 and PBL1. In the resistant ecotype Col-0, xopAC is a major avirulence gene of Xcc. In this study, we show that both the RLCK interaction domain and the uridylyl transferase domain of XopAC are required for avirulence. Furthermore, xopAC can also confer avirulence to both the vascular pathogen Ralstonia solanacearum and the mesophyll-colonizing pathogen Pseudomonas syringae indicating that xopAC-specified effector-triggered immunity is not specific to the vascular system. In planta, XopAC-YFP fusions are localized at the plasma membrane suggesting that XopAC might interact with membrane-localized proteins. Eight RLCK of subfamily VII predicted to be localized at the plasma membrane and interacting with XopAC in yeast two-hybrid assays have been isolated. Within this subfamily, PBL2 and RIPK RLCK genes but not BIK1 are important for xopAC-specified effector-triggered immunity and Arabidopsis resistance to Xcc. Citation: Guy E, Lautier M, Chabannes M, Roux B, Lauber E, et al. (2013) xopAC-triggered Immunity against Xanthomonas Depends on Arabidopsis Receptor-Like Cytoplasmic Kinase Genes PBL2 and RIPK. PLoS ONE 8(8): e73469. doi:10.1371/journal.pone.0073469 Editor: Ching-Hong Yang, University of Wisconsin-Milwaukee, United States of America Editor: Ching-Hong Yang, University of Wisconsin-Milwaukee, United States of America Received March 28, 2013; Accepted July 23, 2013; Published August 9, 2013 Received March 28, 2013; Accepted July 23, 2013; Published August 9, 2013 Copyright: © 2013 Guy 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 a PhD grant from the French Ministry of National Education and 16 Research and French Guyana to EG, the LABEX TULIP (ANR-10-LABX-41), an INRA-SPE 17 post-doctoral fellowship to MC and a Jeunes Chercheurs grant from the Agence Nationale de 18 la Recherche (Xopaque ANR-10-JCJC-1703-01) to LDN. 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. xopAC-triggered Immunity against Xanthomonas Depends on Arabidopsis Receptor-Like Cytoplasmic Kinase Genes PBL2 and RIPK Endrick Guy1,2, Martine Lautier1,2,3, Matthieu Chabannes1,2¤, Brice Roux1,2, Emmanuelle Lauber1,2, Matthieu Arlat1,2,3, Laurent D. Noël1,2* xopAC-Triggered Immunity in Arabidopsis xopAC-Triggered Immunity in Arabidopsis and translocated directly into the host cell by the type III secretion (T3S) system [12]. Phytopathogenic bacteria of the genus Ralstonia, Erwinia, Pseudomonas and Xanthomonas express Hrp (Hypersensitive response and pathogenicity) T3S systems which are essential virulence determinants and secrete ca. 20-70 effectors inside the plant cell [13–15]. To date, most T3E for which molecular functions have been elucidated target PTI components resulting in effector-triggered susceptibility (ETS) [1]. For instance, the Pseudomonas syringae cysteine protease AvrPphB cleaves the RLCK (Receptor-like cytoplasmic kinases) PBS1 and several PBL (PBS1-like such as BIK1) proteins and inhibits FLS2- dependent PTI [7,16,17]. RLCK are kinases of RLK lacking extracellular and transmembrane domains compared to RLK. RLCK are often located at the plasma membrane due to palmitoylation/myristoylation or association with membrane proteins [18,19]. Two other P. syringae effectors AvrB and AvrRpm1 enhance phosphorylation of RIN4 (RPM1- interacting), a negative regulator of PTI [20–22]. RIN4 promotes stomatal opening via its interaction with two plasma membrane H+-ATPases [23]. Targeting RIN4 allows P. syringae to enhance plant stomatal opening, facilitating its entry inside the leaf. plant RLCK RIPK (RPM1-induced protein kinase) and BIK1, the XopAC fic domain uridylylates (transfer of uridine 5'- monophosphate; UMP) and inhibits the kinases [30]. BIK1 and RIPK are positive and negative regulators of PTI, respectively [7,17,31]. Importantly, xopAC deletion mutants in Xcc 8004 show reduced growth when infiltrated in leaves of cabbages and Arabidopsis ecotype Col-0 [30]. Because the fitness gain conferred by xopAC is BIK1-dependent in Arabidopsis, BIK1 inhibition seems to prevail over RIPK inhibition in planta [30]. In addition, RIPK is a positive regulator of the AvrB/RPM1- dependent ETI in Arabidopsis and mediates AvrB-induced RIN4 phosphorylation [31]. XopAC-mediated inhibition of RIPK by uridylylation [30] limits RIN4 phosphorylation monitored by RPM1 [31] and could explain the observed suppression of the AvrB/RPM1-dependent ETI in Arabidopsis [30]. To date, it remains unclear whether XopAC interacts with few or many members of the RLCK family. The plant genes required for resistance against Xcc and xopAC-mediated ETI are also as yet unknown. In this study, we show that xopAC can confer avirulence to vascular and non-vascular bacterial pathogens. Several RLCK interacting with XopAC were identified by yeast two-hybrid assay. At least two of these RLCK are required to mount full resistance to Xcc and are distinct from the virulence-promoting RLCK inhibited by xopAC during a compatible interaction. The LRR and fic domains of XopAC are required for Xcc avirulence in Arabidopsis ecotype Col-0 xopAC was previously reported to be essential for Xcc 8004 avirulence on Col-0 [28]. In order to determine which domains might be critical for the XopAC avirulence function on Arabidopsis ecotype Col-0, chromosomal deletions of the sequences coding for the LRR or fic domains were engineered in Xcc strain 8004. In parallel, the H469A mutation abolishing XopAC uridylylation activity [30] was also introduced in xopAC. Importantly, those mutations do not modify the XopAC N terminus so their secretion-translocation by the T3S system should not be affected [28]. The wild-type strain and the xopAC mutant derivatives were inoculated by piercing in the central vein of Col-0 (resistant to Xcc 8004) and Kas (susceptible to Xcc 8004) leaves. Seven days post-inoculation, 8004∆xopAC, 8004xopAC∆LRR, 8004xopAC-H469A and 8004xopAC∆fic were able to cause disease on Col-0 in contrast to the wild-type strain (Figure 1). These mutations were complemented by xopAC. On Kas, all strains were fully virulent (Supporting Figure S1). In order to test the expression level and the stability of these xopAC mutant alleles, chromosomal versions of the hrpG* (E44) [32] gain of function mutation were introduced in each strain. HrpG positively regulates a large regulon which includes the T3S system and most of its substrates in X. campestris pv. vesicatoria (Xcv) [33]. The HrpG* mutation allows the constitutive expression in rich medium of the T3S system and its substrates such as XopAC in Xcc (LDN, unpublished results). Immunoblot analysis using affinity purified anti-XopAC anti-serum demonstrated that these XopAC mutant proteins accumulated to detectable levels in total cell extracts The Gram-negative bacterium Xanthomonas campestris pv. campestris (Xcc) is a vascular pathogen and the causal agent of black rot disease in Brassicaceae [25]. Xcc can infect economically important crops like cabbage, mustard, radish, turnip, as well as the model plant Arabidopsis thaliana. Xcc colonizes and multiplies inside the xylem vessels after its entry via the hydathodes or wounds. The Xcc T3S system and the ca. 20-30 T3E called Xanthomonas outer proteins (Xop) are essential for bacterial pathogenicity [25,26]. XopAC (also called AvrAC) is a Xanthomonas campestris-specific type III effector. xopAC confers avirulence to Xcc only when bacteria are inoculated into the vascular system of the Columbia-0 (Col-0) ecotype of A. thaliana by piercing the main leaf vein, but not by mesophyll infiltration [27–29]. xopAC encodes a protein with an N-terminal LRR domain and a C-terminal fic (filamentation- induced by cAMP, consensus HPFxxG/ANGR) domain [28]. xopAC-Triggered Immunity in Arabidopsis Though important for ETS, T3E can also betray pathogens since their specific recognition by plant resistance (R) proteins causes effector-triggered immunity (ETI) [1]. ETI generally results in a hypersensitive response (HR) which is rapid and localized cell death at the infection site which limits pathogen growth and spread. R proteins essentially encode nucleotide- binding-site leucine-rich-repeat (NBS-LRR) proteins which usually recognize T3E indirectly by the modifications that the T3E induce on host proteins. For instance, the RPS5 R protein perceives the cleavage of PBS1 by AvrPphB [16]. Similarly, the RPM1 monitors the AvrB/AvrRpm1-dependent phosphorylation of RIN4 [20,22]. Conserved signalling modules downstream of R proteins usually require signalling hubs such as EDS1 (enhanced disease susceptibility) and PAD4 (phytoalexin- defficient) or NDR1 (non-race-specific disease resistance) [24]. While most results have been obtained on the mesophyll- colonizing P. syringae, relatively little is known about the mechanisms of plant immunity towards vascular pathogens. Introduction specifically perceive the presence of the PAMPs have been identified at the plasma membrane [3]. The best-studied PRR are the receptor-like kinases (RLK) FLS2 (Flagellin sensing 2) and EFR (EF-Tu receptor) of A. thaliana which are required for flagellin/AX21 and EF-Tu perception, respectively [4–6]. Upon perception of flagellin and EF-Tu, both FLS2 and EFR dissociate from BIK1 (Botrytis-induced kinase) and associate with the co-receptor BAK1 (BRI1-associated kinase) to initiate a highly complex phosphorylation cascade leading to the establishment of PTI [7,8]. Plant innate immunity is a multilayer system which limits pathogen entry and multiplication in leaf tissues. These defence mechanisms are initially triggered by generic elicitors called PAMPs (Pathogen-associated molecular patterns) [1]. PAMPs are highly conserved, present in multiple organisms and are usually important for microbial fitness or viability [2]. For example, in bacteria, lipopolysaccharides, peptidoglycan, flagellin, AX21 secreted protein or the elongation factor-Tu (EF- Tu) rapidly elicit various defences such as plant cell wall reinforcement, ion fluxes, MAP kinase signalling cascades, transcriptional reprogramming or production of reactive oxygen species (ROS) and anti-bacterial compounds. These responses are globally referred to as PAMP-triggered immunity (PTI). Several pattern-recognition receptors (PRR) which Pathogens have evolved a number of strategies to evade PTI such as the production of non-recognizable PAMPs [9], the scavenging of active PAMPs [10] or the interference with the PTI perception/signalling cascade. Inhibition of PTI can be achieved by bacterial toxins [11] or type III effector (T3E) proteins. In Gram-negative bacteria, T3E proteins are secreted PLOS ONE \| www.plosone.org 1 1 August 2013 \| Volume 8 \| Issue 8 \| e73469 xopAC does not confer avirulence to Xcc when infiltrated in Arabidopsis ecotype Col-0 of 8004::hrpG* (8004, Supporting Figure S2A). These results indicate that the LRR domain, the fic domain and the uridylylation activity of XopAC are required for XopAC avirulence function on Col-0. These results contrast with previous studies performed with Xcc [28]: xopAC-mediated avirulence in the mesophyll was not observed when high bacterial titres (108 cfu/ml) were used for infiltration of Col-0 [28]. . Thus, WT strain 8004 and the ∆xopAC mutant were infiltrated in Col-0 and Kas plants at 108 and 105 cfu/ml. No differences of symptoms caused by the two strains were observed (Figure 2E, Supporting Figure S3A). Yet, growth of the ∆xopAC mutant was significantly increased at 5dpi compared to the WT strain when infiltrated at 105 cfu/ml in Col-0 (Figure 2F, Supporting Figure S3B). Thus, xopAC is unable to confer macroscopic avirulence to Xcc in the mesophyll whatever the bacterial inoculum used, in contrast to our previous observations with Pst. Yet, Xcc multiplication in the mesophyll is reduced at 5 dpi by xopAC indicating that Xcc faces plant xopAC-triggered ETI in both the vasculature and the mesophyll during infection. xopAC confers fic-dependent avirulence to the mesophyll pathogen P. syringae on Arabidopsis ecotype Col-0 To learn whether xopAC-triggered immunity can also be observed in the mesophyllic tissues, xopAC was introduced in the mesophyll-colonizing pathogen P. syringae pv. tomato (Pst) strain DC3000. The pEDV6 vector was used to express fusion of the N-terminus of the Pst effector AvrRPS4 with XopAC to ensure its efficient delivery by the Pst T3S system [37]. Three days after infiltration of Col-0 leaves, Pst expressing the xopAC fusion caused reduced chlorotic symptoms compared to WT only when low inocula were employed (5x105 colony-forming units (cfu)/ml; Figure 2C). Furthermore, this phenotype was dependent on H469 similarly to Xcc 8004 avirulence on Col-0. Importantly, the reduced chlorosis was correlated with reduced growth of the Pst strain expressing WT xopAC compared to a WT Pst strain or a Pst expressing xopAC-H469A (Figure 2D). XopAC and XopAC-H469A fusion proteins accumulated to comparable levels in total bacterial extracts (Supporting Figure S2C). These results demonstrate that xopAC can function as a bona fide avirulence protein in the mesophyll against a mesophyll-colonizing pathogen and that xopAC-induced ETI can be effective both in vascular and mesophyllic tissues. Figure 1. The LRR and fic domains of XopAC are required for XopAC-triggered immunity in Arabidopsis ecotype Col-0. A boxplot representation of pathogenicity of wild-type Xcc strain 8004 and xopAC mutants (∆xopAC, ∆LRR, ∆fic, xopAC-H469A) complemented or not with pCZ917-xopACA (+xopAC) is shown: middle bar = median; box limit = upper and lower quartile; extremes = Min and Max values. Bacteria were inoculated by piercing the leaf central vein and infection symptoms were scored 7 days post-inoculation. Disease index indicates: 0-1 no symptoms; 1-2 weak chlorosis, 2-3 strong chlorosis; 3-4 necrosis. N=3. Each time, at least 4 plants were inoculated on at least 3 leaves. Statistical groups were determined using a Tukey HSD test (P<0.001) and are indicated by a letter. doi: 10.1371/journal.pone.0073469.g001 xopAC-Triggered Immunity in Arabidopsis xopAC-Triggered Immunity in Arabidopsis Figure 1. The LRR and fic domains of XopAC are required for XopAC-triggered immunity in Arabidopsis ecotype Col-0. A boxplot representation of pathogenicity of wild-type Xcc strain 8004 and xopAC mutants (∆xopAC, ∆LRR, ∆fic, xopAC-H469A) complemented or not with pCZ917-xopACA (+xopAC) is shown: middle bar = median; box limit = upper and lower quartile; extremes = Min and Max values. Bacteria were inoculated by piercing the leaf central vein and infection symptoms were scored 7 days post-inoculation. Disease index indicates: 0-1 no symptoms; 1-2 weak chlorosis, 2-3 strong chlorosis; 3-4 necrosis. N=3. Each time, at least 4 plants were inoculated on at least 3 leaves. Statistical groups were determined using a Tukey HSD test (P<0.001) and are indicated by a letter. Col-0 roots. Wilting symptoms were then monitored (Figure 2A and B). While the WT strain wilted all plants by 11 days post- inoculation (dpi), the hrcV mutant and the strain expressing wild-type xopAC did not cause any visible wilting symptoms (Figure 2A). This xopAC-mediated avirulence was dependent on the H469 residue since the Rs strain expressing xopAC- H469A was fully virulent (Figure 2A). Finally, the xopAC- mediated avirulence of Rs was correlated with a significant decrease in bacterial growth (20-fold) but not as strong as the decrease observed with the hrcV mutant (Figure 2B). In conclusion, xopAC-mediated vascular ETI is fully efficient against another vascular bacterial pathogen and recruits the same functional domains as for Xcc 8004 avirulence on Col-0. The LRR and fic domains of XopAC are required for Xcc avirulence in Arabidopsis ecotype Col-0 While the XopAC N terminus is sufficient to interact with the 2 August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org xopAC-Triggered Immunity in Arabidopsis xopAC Confers fic-dependent Avirulence to the Vascular Pathogen Ralstonia solanacearum in Arabidopsis Ecotype Col-0 In order to determine whether xopAC can confer avirulence to other vascular pathogens, XopAC and its mutant derivatives were expressed in Ralstonia solanacearum (Rs) strain GMI1000, the causal agent of bacterial wilt in a wide range of plants. xopAC was expressed under the control of the promoter of the GALA7 effector [34]. Xcc and Rs T3S systems are closely related so proteins from one genus can be secreted by the T3S system of the other [35,36]. Rs strains expressing a wild-type or mutated xopAC gene were inoculated on wounded August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org 3 xopAC-Triggered Immunity in Arabidopsis Figure 2. xopAC can confer avirulence to Pst strain DC3000 and Rs strain GMI100 on Arabidopsis ecotype Col-0. Four- week-old Col-0 plants were inoculated. (A, B) Wild-type and hrcV (hrp-) mutant of Rs GMI1000 or strain derivatives carrying xopAC (+xopAC) or xopAC-H469A (+H469A) were inoculated by root dipping. (A) Pictures were taken at 11 days post-inoculation. (B) Bacterial populations in the aerial parts of the plants were determined at 5 dpi and expressed as log of colony-forming units per gram of fresh weight (cfu/gfw). For each strain, three samples of three plants each were analysed. Two independent experiments were performed. Statistical groups were determined using a Wilcoxon test (P<0.003) and indicated by different letters. (C, D) Leaves were infiltrated with wild-type Pst DC3000 or derivatives carrying pEDV6-xopAC (+xopAC) or pEDV6-xopAC-H469A (+H469A). (C) Bacterial suspensions of Pst at 2x107 cfu/ml or 5x105 cfu/ml were used and pictures were taken 3 days post- inoculation. (D) Bacterial suspensions at 5x105 cfu/ml were infiltrated in leaves. In planta bacterial populations in the inoculated areas were determined 0 and 3 days post-inoculation and expressed as log (cfu/cm2). Standard deviations were calculated on two independent experiments with three samples of two leaf discs from different plants for each strain. Statistical groups were determined using a Wilcoxon test (P<0.012) and indicated by different letters. (E, F) Leaves were inoculated by hand infiltration with wild-type Xcc strain 8004 and 8004∆xopAC. (E) Bacterial suspensions of Xcc at 108 cfu/ml or 105 cfu/ml were used and pictures were taken 4 days post-inoculation. (F) Xcc strains were infiltrated at a bacterial density of 105 cfu/ml. In planta bacterial populations in the inoculated areas were determined 0, 3 and 5 days post-inoculation and expressed as log (cfu/cm2). xopAC Confers fic-dependent Avirulence to the Vascular Pathogen Ralstonia solanacearum in Arabidopsis Ecotype Col-0 doi: 10.1371/journal.pone.0073469.g002 xopAC Confers fic-dependent Avirulence to the Vascular Pathogen Ralstonia solanacearum in Arabidopsis Ecotype Col-0 One representative experiment out of three is shown. Standard deviations were calculated on at least 4 biological samples. For each experiment, three samples of two leaf discs from different plants were collected for each strain. Statistical groups were determined using a Wilcoxon test (P<0 021) and indicated by different letters can confer avirulence to Pst strain DC3000 and Rs strain GMI100 on Arabidopsis ecotype Col-0 Figure 2. xopAC can confer avirulence to Pst strain DC3000 and Rs strain GMI100 on Arabidopsis ecotype Col-0. Four- week-old Col-0 plants were inoculated. (A, B) Wild-type and hrcV (hrp-) mutant of Rs GMI1000 or strain derivatives carrying xopAC (+xopAC) or xopAC-H469A (+H469A) were inoculated by root dipping. (A) Pictures were taken at 11 days post-inoculation. (B) Bacterial populations in the aerial parts of the plants were determined at 5 dpi and expressed as log of colony-forming units per gram of fresh weight (cfu/gfw). For each strain, three samples of three plants each were analysed. Two independent experiments were performed. Statistical groups were determined using a Wilcoxon test (P<0.003) and indicated by different letters. (C, D) Leaves were infiltrated with wild-type Pst DC3000 or derivatives carrying pEDV6-xopAC (+xopAC) or pEDV6-xopAC-H469A (+H469A). (C) Bacterial suspensions of Pst at 2x107 cfu/ml or 5x105 cfu/ml were used and pictures were taken 3 days post- inoculation. (D) Bacterial suspensions at 5x105 cfu/ml were infiltrated in leaves. In planta bacterial populations in the inoculated areas were determined 0 and 3 days post-inoculation and expressed as log (cfu/cm2). Standard deviations were calculated on two independent experiments with three samples of two leaf discs from different plants for each strain. Statistical groups were determined using a Wilcoxon test (P<0.012) and indicated by different letters. (E, F) Leaves were inoculated by hand infiltration with wild-type Xcc strain 8004 and 8004∆xopAC. (E) Bacterial suspensions of Xcc at 108 cfu/ml or 105 cfu/ml were used and pictures were taken 4 days post-inoculation. (F) Xcc strains were infiltrated at a bacterial density of 105 cfu/ml. In planta bacterial populations in the inoculated areas were determined 0, 3 and 5 days post-inoculation and expressed as log (cfu/cm2). One representative experiment out of three is shown. Standard deviations were calculated on at least 4 biological samples. For each experiment, three samples of two leaf discs from different plants were collected for each strain. Statistical groups were determined using a Wilcoxon test (P<0.021) and indicated by different letters. XopAC localization at the plant plasma membrane is LRR-dependent to XopAC, XopAC-H469A and XopAC∆fic were localized at the plasma membrane while XopAC∆LRR was observed in nuclei, cytosol and cytoplasmic strands (Figure 3). XopAC plasma membrane localization was confirmed by co-localization with the RLK (At4g23740)-cyan-fluorescent protein (CFP) fusion known to be addressed to the plasma membrane (Figure 3E) [31]. In addition, YFPv-XopAC did not co-localize with the nucleocytoplasmic marker MIEL1-CFP (Figure 3F) [38]. Since no membrane-targeting signal or transmembrane domain was XopAC subcellular localization was determined by Agrobacterium-mediated expression of Yellow Fluorescent Protein venus (YFPv) fusions to XopAC variants (XopAC- H469A, XopAC∆LRR and XopAC∆fic) in Nicotiana benthamiana leaf epidermis cells. Accumulation and stability of the different YFPv-XopAC proteins was verified by Immunoblot analysis (Supporting Figure S2B). Interestingly, YFPv fusions PLOS ONE \| www.plosone.org August 2013 \| Volume 8 \| Issue 8 \| e73469 4 xopAC-Triggered Immunity in Arabidopsis Figure 3. The LRR domain is required to target XopAC to the plasma membrane of N. benthamiana epidermal cells. YFPv- XopAC (A, E and F) and mutant variants (B, YFPv-XopAC-H469A; C, YFPv-XopAC∆fic and D, G, YFPv-XopAC∆LRR) were expressed using Agrobacterium-mediated transient transformation and imaged in epidermal cells by confocal laser microscopy 48 hours after inoculation. (A) The plasma membrane localized RLK-CFP fusion (At4g23740) and the nucleo-scytoplasmic marker MIEL1 (At5g18650) were co-expressed with YFPv-XopAC (E, F) or YFPv-XopAC∆LRR (G) and used as controls. The merged pictures are shown (E, F and G). Scale bars = 25 µm. White arrowheads indicate nuclei (N), cytosol (Cyto) and cytoplasmic strands (CS). doi: 10.1371/journal.pone.0073469.g003 Figure 3. The LRR domain is required to target XopAC to the plasma membrane of N. benthamiana epidermal cells. YFPv- XopAC (A, E and F) and mutant variants (B, YFPv-XopAC-H469A; C, YFPv-XopAC∆fic and D, G, YFPv-XopAC∆LRR) were expressed using Agrobacterium-mediated transient transformation and imaged in epidermal cells by confocal laser microscopy 48 hours after inoculation. (A) The plasma membrane localized RLK-CFP fusion (At4g23740) and the nucleo-scytoplasmic marker MIEL1 (At5g18650) were co-expressed with YFPv-XopAC (E, F) or YFPv-XopAC∆LRR (G) and used as controls. The merged pictures are shown (E, F and G). Scale bars = 25 µm. White arrowheads indicate nuclei (N), cytosol (Cyto) and cytoplasmic strands (CS). doi: 10.1371/journal.pone.0073469.g003 stabilize the interaction with its substrate in pull-down assays [40], full-length XopAC-H469A was used as bait. From 2 million primary transformants, 68 bait-dependent interactors were isolated and corresponded to 50 putative interactors of XopAC (PIX) genes in the Arabidopsis genome. XopAC localization at the plant plasma membrane is LRR-dependent Remarkably, eight PIX genes (PIX 1, 7, 8, 13, 14, 15, 16 and 17) represented by eleven independent cDNA fragments (Table 1) encode RLCK belonging to the subfamily VII of the RLK [18]. To date, kinase activity was only described for two of these PIX-RLCK proteins (PIX8 and 15) [31,41]. RIPK/ACIK1A/PIX8 is the only interactor with a reported biological function in plant innate immunity [31] and a known interactor/substrate of XopAC [30]. More than 600 RLK have been identified in the Col-0 genome [18]. Although the functions of RLK are mostly unknown, RLK are proposed to predicted in XopAC, its subcellular localization might be the result of an LRR-mediated interaction to a plasma membrane- associated protein. XopAC can trigger ETI against both vascular and mesophyll-colonizing bacterial pathogens Previous studies have shown that xopAC can restrict Xcc growth in the vasculature of Col-0 but not in Sf-2 ecotypes of Arabidopsis [27,28]. In this study, we provide evidence that xopAC is able to confer avirulence to other phytopathogenic bacteria. In particular, the vascular pathogen Rs expressing XopAC did not cause any wilting symptoms on Col-0 plants similar to a T3S mutant. Bacterial populations in plant shoots were also significantly reduced. This indicates that the immune responses triggered by XopAC can be efficient against other vascular pathogens and that the genetic dissection of this particular ETI could help to identify novel targets to manage/ control other vascular pathogens such as Rs. XopAC also XopAC interacts with a subfamily of the Arabidopsis receptor-like cytosolic kinases (RLCK) in yeast two- hybrid assay In order to try and identify XopAC interactors, a LexA-based yeast two-hybrid screen [39] was performed against a normalized Col-0 cDNA library (Arabidopsis thaliana universal, 5.6 million clones, Dualsystems Biotech). Because the H348A mutation (equivalent of the H469A in XopAC) in the fic domain of VopS T3E from Vibrio parahaemolyticus was shown to PLOS ONE \| www.plosone.org August 2013 \| Volume 8 \| Issue 8 \| e73469 5 xopAC-Triggered Immunity in Arabidopsis Table 1. List and properties of RLCK proteins identified by yeast two-hybrid assay as putative interactors with XopAC- H469A. Table 1. List and properties of RLCK proteins identified by yeast two-hybrid assay as putative interactors with XopAC- H469A. Table 1. List and properties of RLCK proteins identified by yeast two-hybrid assay as putative interactors with XopAC- H469A. PIX Names Gene N° of cDNA Protein domain PIX1 PBL15 AT1G61590 2 83-329 and 83-326 PIX7 - AT5G15080 2 192-347 and 192-337 PIX8 RIPK/ACIK1A/PBL14 AT2G05940 2 85-240 and 81-240 PIX13 - AT2G17220 1 58-241 PIX14 APK2B/PBL3 AT2G02800 1 44-250 PIX15 APK1A/PBL9 AT1G07570 1 15-217 PIX16 APK1B/PBL10 AT2G28930 1 30-245 PIX17 PBL8 AT5G01020 1 169-299 be involved in nearly every aspect of plant life including growth, development and immunity [42]. Based on protein similarities, we have further subdivided the RLCK subfamily VII into VIIa and VIIb. Interestingly, all PIX-RLCK proteins belonged to subfamily VIIa (Figure 4A) in which the two known substrates of XopAC BIK1 (Botrytis-induced kinase) [43] and RIPK can be found [30] as well as other players of plant innate immunity such as ACIK1B (Avr9/Cf-9 induced kinase) [44] or PBL1 (PBS1-like) [7]. Overlap between the different partial cDNA fragments identified in this yeast two-hybrid screen delimits a highly conserved minimal interaction domain of 45-46 residues (Supporting Figure S4). This region encompasses the kinase catalytic domain (domain VI) of the RLCK and is situated directly upstream of the uridylylation site of RIPK and BIK1 [30]. cst) were inoculated with Xcc strain 8004 by piercing. The CAST AWAY (CST) gene encodes an RLCK of subfamily VIIa which is important for organ abscission [19]. cst mutant was included as a negative control. Eight days post-inoculation, the ripk and pbl2 mutants developed significant disease symptoms (Figure 5A and B). Yet, the ripk mutant was less susceptible than the ecotype Kas or the pbl2 mutant. The ripk mutant was complemented by overexpression of RIPK-HA indicating that RIPK is genetically important for xopAC-triggered immunity. XopAC interacts with a subfamily of the Arabidopsis receptor-like cytosolic kinases (RLCK) in yeast two- hybrid assay Interestingly, RIPK positively regulates RPM1-/RIN4-dependent avrB-triggered immunity [31]. Because XopAC and AvrB are both members of the fido family (fic, doc and AvrB) [45] and might share other signalling components, the rpm1 and the rps2/rin4 mutants were inoculated with strain 8004 by piercing (Figure 5A). The rps2/rin4 mutant was used because a rin4 mutant is not viable [20]. These two mutants were as resistant as the wild-type Col-0 plants. In agreement with these observations, in planta growth of strain 8004 was significantly higher in the pbl2 mutant (Figure 5C). Bacterial populations in the ripk, pbs1 and pbl1 mutants and the wild-type plants were comparable. These Arabidopsis mutants showed wild-type susceptibility to the Xcc 8004∆xopAC mutant strain (Supporting Figure S6). 8004∆xopAC growth was not affected in the different Arabidopsis mutants tested (Supporting Figure S6). These results suggest that PBL2, and to a lesser extent RIPK, positively regulate xopAC-triggered immunity. Interaction of full-length PIX8-RIPK with XopAC variants was tested in the yeast two-hybrid system. While RIPK/XopAC- H469A interaction was observed, no interaction with wild-type XopAC, XopAC∆LRR or XopAC∆fic was detected though proteins accumulated to similar levels by Immunoblot analysis (Figure 4B, Supporting Figure S5). The central kinase domain of RIPK (residues 87-367) interacted specifically with XopAC- H469A but not with wild-type XopAC (Supporting Figure S5A). Four other full-length RLCK PIX1, PIX7, BIK1 and PBL2 were tested for their interaction with XopAC. PIX1 interacted with wild-type XopAC but not with XopAC-H469A while PIX7 interacted with both the XopAC forms (Figure 4B and C). Finally, PBL2 was able to interact with XopAC-H469A but not wild-type XopAC while BIK1 did not interact to wild-type XopAC nor XopAC-H469A. Among all RLCK-XopAC interactions tested here, PBL2-XopAC-H469A interaction was the only one detectable on the selective medium lacking both histine and adenine (-WLHA) suggesting that PBL2-XopAC interaction is stronger than RIPK-XopAC. These results suggest that, besides RIPK and BIK1, XopAC might have the potential to interact with other members of the RLCK VIIa subfamily such as PBL2. XopAC avirulence functions likely depend on host target uridylylation No rapid HR-like symptoms were observed when Xcc or Pst expressing xopAC were infiltrated into Col-0 leaf tissues. Furthermore, in Pst, xopAC avirulence was symptomless (Figure 2C). These observations suggests that xopAC- triggered immunity might not rely on rapid cell death and that it might differ mechanistically from other ETIs: Interestingly, resistance against bacterial pathogens can be achieved without plant cell death [46] and xopAC-triggered ETI might be too weak to induce HR as exemplified by TAO1-dependent recognition of avrB [47]. Importantly, the XopAC fic domain and its conserved H469 residue which are important for UMP transferase activity in vitro [30] are also essential for the xopAC avirulence function: XopAC uridylylation activity cannot be uncoupled from XopAC biological functions. Thus, xopAC- triggered immunity likely results from host target uridylylation. PBL2 and RIPK are needed for xopAC-triggered immunity in Col-0 P53 was used as specificity control for the prey. Ten-fold serial dilutions of yeast transformants were spotted from left to right on minimal medium (-WL) and minimal medium without histidine (-WLH) or Figure 4. XopAC interacts with several members of the RLCK VIIa subfamily. (A) Neighbour-joining consensus tree PBL2 and RIPK are needed for xopAC-triggered immunity in Col-0 To determine whether these RLCK might be important to mount resistance against Xcc strains expressing xopAC, several RLCK mutants (ripk, pbs1, pbl1, pbl2, bik1/pbl1 and PLOS ONE \| www.plosone.org August 2013 \| Volume 8 \| Issue 8 \| e73469 6 xopAC-Triggered Immunity in Arabidopsis conferred avirulence to Pst DC3000 in Col-0 ecotype but not in Kas when infiltrated at low bacterial densities. This result indicates that xopAC has the potential to be recognized in the mesophyll and to trigger potent ETI in this tissue. Yet, such ETI was not observed macroscopically when Xcc was infiltrated even at low bacterial densities in leaf mesophyll (Figure 4E) [27,28]. Interestingly, an increased growth of the ∆xopAC mutant was reproducibly observed on Col-0. This observation suggests that Xcc encounters xopAC-triggered immunity not only in Arabidopsis vasculature, but also in the leaf mesophyll (Figure 4F). The number of inoculated bacteria (a few hundred by piercing versus millions by infiltration) might explain why a strong avirulence is observed by piercing but not by infiltration of Xcc. In conclusion, xopAC cannot be considered as a vasculature-specific avirulence determinant which indicates that its pathogenicity and/or avirulence targets should be found both in mesophyll and vascular tissues. Figure 4. XopAC interacts with several members of t RLCK VIIa subfamily. (A) Neighbour-joining consensus tr of the 45 full-length Arabidopsis RLCK proteins aligned us Geneious alignment with default settings. AT1G24030 prot kinase was used to root the tree. Shaded areas define the t subfamilies VIIa and VIIb of RLK. PIX-RLCK proteins identif in the yeast two-hybrid screen with XopAC-H469A a indicated in red. Published protein names are indicated wh available. (B, C, D) Yeast two-hybrid interaction tests betwe XopAC or its mutant allele H469A as bait and full-length PIX PIX7 PIX8-RIPK PBL2 or BIK1 as prey P53 was used Figure 4. XopAC interacts with several members of the RLCK VIIa subfamily. (A) Neighbour-joining consensus tree of the 45 full-length Arabidopsis RLCK proteins aligned using Geneious alignment with default settings. AT1G24030 protein kinase was used to root the tree. Shaded areas define the two subfamilies VIIa and VIIb of RLK. PIX-RLCK proteins identified in the yeast two-hybrid screen with XopAC-H469A are indicated in red. Published protein names are indicated when available. (B, C, D) Yeast two-hybrid interaction tests between XopAC or its mutant allele H469A as bait and full-length PIX1, PIX7, PIX8-RIPK, PBL2 or BIK1 as prey. Plasma membrane-localized XopAC interacts with members of the VIIa RLCK subfamily In our search for XopAC targets, we identified eight Arabidopsis RLCK proteins as the major class of XopAC- H469A interactors by yeast two-hybrid assay. The H469A mutation was used since the orthologous mutation in VopS was shown to stabilize the interaction with Rho-GTPases [40]. Yet, validation of the interactions with full length RLCK cDNAs showed that the interaction could be reconstituted with wild- type or mutant XopAC or both depending on the RLCK studied. The biological relevance of some of these interactions was recently demonstrated by the identification of xopAC- dependent uridylylation of three RLCK (BIK1, PBL1 and RIPK- PIX8) resulting in the inhibition of their kinase activity [30]. Several RLCK like BIK1 and to a lesser extent PBL1 and PBL2 are positive regulators of PTI downstream of EFR, FLS2 and CERK1 while RIPK is a negative regulator of PTI [7,31]. Several RLCK like RIPK, PBL2 or CST are localized at the plasma membrane and many possess predicted myristoylation or palmitoylation sequences in their N-terminal domain [7,19,31]. For instance, PBL2-YFP can be myristoylated in vitro and its plasma membrane localization in tobacco depends on Figure 4. XopAC interacts with several members of the RLCK VIIa subfamily. (A) Neighbour-joining consensus tree of the 45 full-length Arabidopsis RLCK proteins aligned using Geneious alignment with default settings. AT1G24030 protein kinase was used to root the tree. Shaded areas define the two subfamilies VIIa and VIIb of RLK. PIX-RLCK proteins identified in the yeast two-hybrid screen with XopAC-H469A are indicated in red. Published protein names are indicated when available. (B, C, D) Yeast two-hybrid interaction tests between XopAC or its mutant allele H469A as bait and full-length PIX1, PIX7, PIX8-RIPK, PBL2 or BIK1 as prey. P53 was used as specificity control for the prey. Ten-fold serial dilutions of yeast transformants were spotted from left to right on minimal medium (-WL) and minimal medium without histidine (-WLH) or histidine and adenine (-WLHA) which were used to visualize prey/bait interaction. Pictures were taken 4 days after spotting. doi: 10.1371/journal.pone.0073469.g004 Figure 4. XopAC interacts with several members of the 7 August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org xopAC-Triggered Immunity in Arabidopsis this myristoylation site [48]. In Arabidopsis, proteomic studies of plasma membrane-enriched fractions identified PBL2 [49] and PBL2 can be co-immunoprecipitated with the plasma membrane receptor FLS2 [7]. Therefore, there is compelling evidence for PBL2 localization at the plasma membrane. Plant material and growth conditions Wild-type Col-0 and Kas ecotypes were used as well as mutants in the Col-0 genetic background: ripk (GT22343 backcrossed into Col-0) and ripk 35S–RIPK-HA (line 16-5) [31], rps2-101C [55], rps2-101C/rin4 [56], rpm1-3 [57], pbs1-1 [58], pbl1 (SAIL_1236_D07), pbl2 (SALK_149140) [7] and cst-1 [19]. Arabidopsis plants were grown on Jiffy pots in a growth chamber at 22° C, with a 9-h light period and a light intensity of 192 µmol.m-2.s-1. After 2 days at 4° C, plates were transferred to a growth chamber at 20° C with 16-h light period. N. benthamiana plants were grown in temperate greenhouses. Bacterial strains, plasmids and growth conditions Strains and plasmids used in this study are listed in Supporting Table S1. Xcc, A. tumefasciens, Rs and Pst strains were grown at 28° C in MOKA [52], in yeast extract-beef medium (YEB), complete medium BG [53] and King’s B medium [54], respectively. Escherichia coli cells were grown on Luria-Bertani medium at 37° C. For solid media, agar was added at a final concentration of 1.5% (w/v). Antibiotics were used at the following concentrations: for all bacteria, 50 µg/ml kanamycin, 50 µg/ml rifampicin, 50 µg/ml ampicillin; for Xcc, 5 µg/ml tetracycline; for A. tumefaciens, 15 µg/ml gentamycin, 100 µg/ml spectinomycin and 5 µg/ml tetracycline; for Pst, 50 µg/ml gentamycin; for Rs, 40 µg/ml spectinomycin, 5 µg/ml gentamycin; for E. coli, 40 µg/ml spectinomycin and 10 µg/ml tetracycline. Plasma membrane-localized XopAC interacts with members of the VIIa RLCK subfamily Interestingly, XopAC also localizes at the plasma membrane in N. benthamiana leaves though lacking membrane-anchoring motifs. This association to the plasma membrane is dependent on the XopAC highly conserved LRR domain which is needed to interact with BIK1 and RIPK [30]. Thus, it is tempting to propose that XopAC subcellular localization is achieved by interaction of its N-terminal-LRR domains with membrane- associated RLCK. In our yeast-two hybrid screen, only cDNAs of RLCK subfamily VIIa were identified. Its substrates BIK1, PBL1 and RIPK also belong to that group. These results suggest that XopAC associate preferentially to RLCK subfamily VIIa though the ST or TT uridylylated residues are highly conserved throughout subfamily VII. In planta interaction studies as well as uridylylation tests will be needed to address this question. In the PTI context, BIK1 probably remains the most relevant XopAC target. Because RLCK are highly conserved, interaction with them does not have to be highly specific to a few biologically important RLCK: a global inhibition of all RLCK subfamily VIIa members including BIK1, PBL1, PBL2 and RIPK would be sufficient to achieve PTI suppression. Such inhibition of RLCK subfamily VIIa by XopAC is very reminiscent of the proteolytic cleavage of many RLCK subfamily VII members by AvrPphB [7]. This observation further defines the RLCK family as a prime pathogenicity target for bacterial pathogens to suppress PAMP-triggered immunity or other RLCK-dependent physiological responses. close homologues (such as ACIK1B/PBL13) is an obvious option. Also, RIPK acts as a negative regulator of PTI so its inhibition might mask ETI breakdown. Whether PBL2- and RIPK-dependent ETI act independently or additively is not known. Interestingly, resistance to Xcc did not genetically require RIN4 nor RPM1. On the one hand, RIPK could signal through RIN4/RPM1-independent signalling cascades yet to be identified. On the other hand, genetic redundancy in the RIN4 and RIN4-like proteins might blur our analyses [50]. Further experimental evidence will be needed to establish the recruitment of the RIN4/RPM1 signalling module in xopAC- triggered immunity. To conclude, our results are very reminiscent of PBS1 being a decoy guarded by RPS5 to detect AvrPphB-dependent degradation of other RLCK important for PTI [7,51]. During the xopAC-mediated ETI against Xcc in Arabidopsis, PBL2 and/or RIPK could well be decoys detecting uridylylation and inhibition of RLCK important for plant innate immunity such as BIK1. xopAC-dependent ETI in Arabidopsis depends on RIPK and PBL2 While xopAC-mediated suppression of PTI is rather well understood, the mechanisms required for xopAC-triggered immunity against Xcc are lacking. In this study, PBL2 was the strongest interactor of XopAC among all RLCK tested in yeast- two hybrid (Figure 4D). Interestingly, we identified the PBL2 RLCK gene as critical components of this ETI and RIPK to a lesser extent. ETI suppression in the pbl2 mutant is unexpected since this kinase was so far only known as a positive PTI regulator partially needed to respond to flg22 and efl18 but not chitin [7]. Yet, pbl2 was fully susceptible to Xcc strain 8004 expressing xopAC both in terms of disease symptoms and bacterial growth. Importantly, pbs1, pbl1, bik1/ pbl1 or cst mutants were fully resistant suggesting that PBL2 is specifically recruited to mount this ETI. The ripk mutant showed an intermediate susceptibility to Xcc strain 8004 for disease symptoms, but sustained wild-type bacterial growth. RIPK is needed for the RPM1-specified ETI, interacts with AvrB and RIN4 and can phosphorylate both AvrB and RIN4 [31]. Phosphorylation of RIN4 then activates RPM1 and the subsequent ETI response. In a compatible interaction between Xcc and its host, XopAC can uridylylate RIPK and consequently suppress the RPM1-specified ETI [30]. The partial susceptibility of the ripk mutant could be explained in several ways: Functional redundancy of RIPK with one of its xopAC-Triggered Immunity in Arabidopsis Figure 5. The RLCK genes RIPK-PIX8 and PBL2 are required for xopAC-mediated avirulence of Xcc strain 8004. (A,B) Boxplot representation of pathogenicity of strain 8004 on Col-0 mutants and transgenics inoculated by piercing the central vein of the leaves is shown: middle bar = median; box limit = upper and lower quartile; extremes = Min and Max values. Kas was used as a susceptible control. Mutants in genes coding for the RIPK-RIN4/RPM1 complex (A) and various RLCK (B) were tested. Disease indices were scored 8 days post-inoculation: 0-1 no symptoms; 1-2 weak chlorosis, 2-3 strong chlorosis; 3-4 necrosis. N=3. Each time, at least 4 plants were inoculated on at least 3 leaves. Statistical groups were determined using a Tukey HSD test (P<0.001) and are indicated by different letters. (C) A bacterial suspension (105 cfu/ml) of Xcc strain 8004 was inoculated by piercing leaves of Col-0 mutants and transgenics. In planta bacterial populations in the inoculated areas were determined 0 and 4 days post- inoculation and expressed as log of colony forming units per Pst pathogenicity was assayed on A. thaliana, by infiltration of bacterial suspensions at 2x107 cfu/ml. Each strain was tested on 6 plants with 4 leaves per plant. After inoculation, plants were kept at almost 100% relative humidity. Rs pathogenicity was assayed on A. thaliana Col-0 by root inoculations essentially as described [59]: Approximately 2 cm was cut from the bottom of the Jiffy pot and the exposed roots of the plants were immersed for 15 min in a bacterial suspension containing 107 bacteria/ml. The plants were then transferred to a growth chamber at 27° C (16 h light): 26° C (8 h dark) until symptom appearance. Disease development was scored daily, using a macroscopic scale describing the observed wilting: 1 for 25% of the leaves wilted; 2 for 50%; 3 for 75% and 4 for complete wilting. The plants were grown at 25° C. At least three independent repetitions were performed for all pathogenicity tests. Statistical analysis was performed using a fixed effect model and a Tukey HSD test with correction for multiple testing. All analyses were performed using R software version 2.14.1 (www.r-project.org). Pathogenicity tests Xcc pathogenicity was assayed on the A. thaliana accessions Columbia (Col-0) and Kashmir (Kas) by piercing inoculation of bacterial suspension at 108 cfu/ml as described [27]. Each strain was tested on 4 plants with 4 leaves per plant. Xcc pathogenicity was also assayed on A. thaliana by infiltration of bacterial suspensions at 105 and 108 cfu/ml as described [27]. August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org 8 xopAC-Triggered Immunity in Arabidopsis Determination of in planta bacterial populations For Xcc and Pst, plant leaves were infiltrated with a bacterial suspension of 105 cfu/ml and 5x105 cfu/ml, respectively. Two leaf discs from distinct infiltrated leaves of different inoculated plants were sampled using a cork borer (area = 0.33 cm2) at 0, 3, 4 or 5 days after inoculation and were homogenized in 200 µl sterile water. Serial dilutions of the homogenates were performed and 5 µl drops were spotted in triplicate for each dilution on plates supplemented with appropriate antibiotics. The plates were incubated at 28° C for 48 h and colonies were counted in spots containing 1 to 30 colonies [28]. For R. solanacearum, the in planta bacterial populations were determined as described [59] with the following modifications: plants were root-inoculated with a bacterial suspension containing 107 cfu/ml. The aerial parts of three whole inoculated plants were weighed, surface sterilized three times in 250 ml of 70% ethanol, rinsed twice in sterile water and ground in sterile water. Bacterial densities in leaves were determined by plating serial dilutions on complete BG medium. At least two independent experiments were performed. Figure 5. The RLCK genes RIPK-PIX8 and PBL2 are required for xopAC-mediated avirulence of Xcc strain 8004. (A,B) Boxplot representation of pathogenicity of strain 8004 on Col-0 mutants and transgenics inoculated by piercing the central vein of the leaves is shown: middle bar = median; box limit = upper and lower quartile; extremes = Min and Max values. Kas was used as a susceptible control. Mutants in genes coding for the RIPK-RIN4/RPM1 complex (A) and various RLCK (B) were tested. Disease indices were scored 8 days post-inoculation: 0-1 no symptoms; 1-2 weak chlorosis, 2-3 strong chlorosis; 3-4 necrosis. N=3. Each time, at least 4 plants were inoculated on at least 3 leaves. Statistical groups were determined using a Tukey HSD test (P<0.001) and are indicated by different letters. (C) A bacterial suspension (105 cfu/ml) of Xcc strain 8004 was inoculated by piercing leaves of Col-0 mutants and transgenics. In planta bacterial populations in the inoculated areas were determined 0 and 4 days post- inoculation and expressed as log of colony-forming units per square cm (cfu/cm2). Standard deviations were calculated on two independent experiments. For each experiment, three samples of two leaf discs from different plants were collected for each strain. Statistical groups identified using a Wilcoxon test (P<0.05) are indicated by different letters. doi: 10 1371/journal pone 0073469 g005 XopAC expression vectors for Pst and Rs For expression and efficient typeIII-dependent translocation XopAC in Pst, xopAC and xopAC-H469A ENTRY clones were recombined by LR into pEDV6 [37] giving pEDV6-xopAC and pEDV6-xopAC-H469A, respectively. Plasmids were introduced in Pst by triparental mating. For expression in Rs, ENTRY clones of wild-type and mutant xopAC were recombined by LR into pNP269 (N. Peeters, unpublished) destination vector. Plasmids were introduced in Rs by natural transformation as described [64]. Pairwise interactions between XopAC and derivatives vs. PIX-RLCK were tested in the GAL4 yeast two-hybrid system (Clontech). To this end, pENTRY clones of wild-type XopAC and H469A mutant derivative were recombined by LR in pGBKT7-GW (Clontech, Gateway-compatible derivative, L. Deslandes) to generate pGBKT7-xopAC and pGBKT7-xopAC- H469A. pENTRY-PIX1, pENTRY-PIX7, pENTRY-PIX8, pENTRY-PIX887-367, pENTRY-BIK1 and pENTRY-PBL2 were recombined by LR into a Gateway-compatible pGADT7 (Clontech, L. Deslandes). Saccharomyces cerevisiae strain AH109 were co-transformed with bait and prey vectors as described (Clontech). Auto-activation of the His auxotrophy reporter was tested on SD-WLH (Supporting Figure S4B). Interactions were tested on at least 3 different clones. Construction of xopAC and hrpG mutant derivatives in Xcc 8004 Deletion mutants or gene replacements in Xcc 8004 were obtained using the SacB system with p∆13 suicide vector (L. Noël, unpublished), a pK18 derivative [60] designed for BsaI- based Goldengate cloning [61]. Plasmids were introduced into E. coli by electroporation and into Xcc by triparental mating as described [62,63]. Deletion events were verified by PCR (details available upon request). Standard methods were used unless stated differently. Fluorescence microscopy YFPv fluorescence in N. benthamiana leaves was analysed with a confocal laser scanning microscope (TCS SP2-SE, Leica) using a x40 water immersion objective lens (numerical aperture 1.20; PL APO) or the x63 oil immersion objective lens (numerical aperture 1.40; PL APO). Fluorescence of YFPv and CFP was excited with the 514 nm and 458 nm rays of the argon laser and detected between 520 and 575 nm and 465 and 520 nm, respectively. Images were acquired in the sequential mode (20 Z planes per stack of images; 0.5 mm per Z plane) using Leica LCS software (version 2.61). Identification of XopAC interactors by yeast two-hybrid approach A LexA-based yeast two-hybrid screen using the full-length XopAC-H469A (Amplicon LN608-609 cloned PstI/XmaI in the bait vector pLexA-DIR) was carried out by Dualsystems Biotech AG, Zurich, Switzerland. The bait construct was transformed into the strain NMY32 using standard procedures [66]. Correct expression of the bait was verified by immunoblotting of cell extracts using a mouse monoclonal antibody directed against the LexA domain (Dualsystems Biotech, Switzerland). The absence of self-activation was verified by co-transformation of the bait together with a control prey and selection on minimal medium lacking the amino acids tryptophan, leucine and histidine (selective medium). For the yeast two-hybrid screen, the bait was co-transformed together with a normalized Arabidopsis universal cDNA library (Col-0; Dualsystems P02403) into NMY32. 2x106 transformants were screened yielding 85 transformants that grew on selective medium. Positive transformants were tested for β- galactosidase activity using a PXG β-galactosidase assay (Dualsystems Biotech). 68 of the 85 initial positives showed β- galactosidase activity and were considered to be true positives. Library plasmids were isolated from positive clones. The identity of positive interactors was determined by sequencing and comparison to the Arabidopsis databases. In order to introduce the HrpG E44K mutation in Xcc 8004 genome, XC_3077 gene fragment was amplified with LN251/252 from 8004 genomic DNA and cloned into p∆13 giving p∆13-hrpG. Site-directed mutagenesis was performed using LN202/203 giving p∆13-hrpG* and introduced in wild- type strain 8004 and 8004∆hrcV (L. Noël, unpublished). To engineer xopAC∆LRR genomic deletion, site-directed mutagenesis of one internal BsaI restriction site in pENTRY- xopAC∆LRR was performed using EG3/4. The EG5/6 amplicon was cloned by Goldengate into p∆13 giving p∆13-xopAC∆LRR. For xopAC∆fic deletion, site-directed mutagenesis of one internal BsaI restriction site in pENTRY-xopAC was performed using EG3/4. The EG7/MC8 and EG8/MC9 amplicons were assembled by Goldengate into p∆13 giving p∆13-xopAC∆fic. For H469 mutagenesis, amplicons LN436/9 and LN442/443 were cloned in p∆13 and subjected to site-directed mutagenesis with LN433/434, giving p∆13-xopAC-H469A. Construction of xopAC and RLCK ENTRY clones Construction of xopAC and RLCK ENTRY clones All oligonucleotides mentioned in this study are described in Supporting Table S2. XopAC open reading frame (with stop codon) was amplified from genomic DNA of wild-type Xcc 8004 with Phusion DNA polymerase (Finnzymes, Vantaa, Finland) using MC12/13 primers. xopAC∆LRR and xopAC∆fic amplicons were produced by fusion-PCR between amplicons MC4/12+MC5/13 and MC8/12+MC9/13, respectively. attB1 and attB2 sites were added with a third PCR using MC18/MC19 oligonucleotides. Wild-type and xopAC deletion mutant amplicons were recombined by BP reaction into pDONR207 (Invitrogen) giving pENTRY-xopAC, pENTRY-xopAC∆LRR and pENTRY-xopAC∆fic. pENTRY-xopAC-H469A was produced by site-directed mutagenesis of pENTRY-xopAC with primers LN433/434 using the QuikChange mutagenesis kit (Stratagene). PIX1, PIX7, PIX8 (full length and amino acids 9 August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org xopAC-Triggered Immunity in Arabidopsis 87-367), BIK1 and PBL2 ORFs (with stop codon) were amplified from cDNA of Col-0 seedlings grown in vitro using primers EG45/46, EG58/59, EG62/63, EG167/168, LN714/715 and LN716-717, respectively and recombined by BP into pDNOR207 giving pENTRY-PIX1, pENTRY-PIX7, pENTRY- PIX8 and pENTRY-PIX887-367, pENTRY-BIK1 and pENTRY PBL2. chamber at 21° C, with a 16-h light period and 70% hygrometry. Transient expression of YFPv-XopAC fusions in N benthamiana N-terminal fusions of YFPv to wild-type and mutant XopAC were generated by LR recombination of the corresponding ENTRY vectors in pBin19-35S-YFPv-GW (S. Rivas, unpublished) giving pBin19-YFPv-xopAC, pBin19-YFPv- xopAC∆LRR, pBin19-YFPv-xopAC∆fic and pBin19-YFPv- xopAC-H469A. Plasmids were transformed in Chemo- competent C58C1 Agrobacterium cells. Agrobacterium- mediated transient expression in N. benthamiana leaves was performed as described [65]. Plants were grown in a growth (PDF) pLysS. A culture in exponential phase was treated with 2mM IPTG for 2 hrs and cells were harvested by centrifugation. Inclusion bodies of His6-tagged XopAC were solubilized in urea-containing buffer, loaded on TALON resin as recommended (Clontech) and eluted with 200mM imidazole. One mg of His-XopAC was used for the immunisation of two rabbits (GenCust, Dudelange, Luxemburg). Anti-serum was purified on nitrocellulose-immobilized His-XopAC and used at dilutions of 1/1000 (Plant extracts) or 1/5000 (other extracts) for XopAC detection by immunoblotting on plant and bacterial protein samples. A specific signal at the expected size was detected in 8004* strain grown in MOKA but not 8004∆xopAC (Supporting Figure S2A). Figure S4. Protein alignment of the eight full-length PIX proteins belonging to the RLCK family identified by yeast two- hybrid assay as putative interactors with XopAC-H469A. (PDF) Figure S5. XopAC-H469A interacts with the PIX8 kinase domain in a yeast two-hybrid assay. (PDF) Figure S6. Pathogenicity and in planta growth of strain 8004∆xopAC on Col-0 mutants and transgenics inoculated by piercing the central leaf vein. (PDF) Supporting Information Figure S1. The LRR and fic domains of XopAC are not required for pathogenicity on Arabidopsis ecotype Kas. (PDF) Table S2. Oligonucleotides used in this study. (PDF) Table S2. Oligonucleotides used in this study. (PDF) Table S2. Oligonucleotides used in this study. (PDF) Acknowledgements We wish to thank colleagues from the LIPM/FR3450/UPS: Catherine Zanchetta for technical assistance, Céline Remblière for plant transformation, Susana Rivas, Laurent Deslandes, Benoît Lefebvre and Nemo Peeters for providing unpublished vectors or anti-sera, Aurélie Le Ru from the FR3450 microscopy platform, Anne-Claire Cazalé for advice on Ralstonia inoculation procedures and Peter Winterton for proofreading the manuscript. We are grateful to Gitta Coaker, Roger Innes, Sara Liljdren, David Mackey, Brian Staskawicz, Jian-Min Zhou and Cyril Zipfel for contributing several Arabidopsis mutants and transgenic lines used in this manuscript. Author Contributions Conceived and designed the experiments: EG MC MA LDN. Performed the experiments: EG ML MC BR EL LDN. Analyzed the data: EG BR EL MA LDN. Wrote the manuscript: EG LDN. Figure S2. Accumulation and stability of XopAC variants expressed in Xcc 8004, Pst DC3000 and N. benthamiana. (PDF) Figure S2. Accumulation and stability of XopAC variants expressed in Xcc 8004*, Pst DC3000 and N. benthamiana. (PDF) Figure S3. Symptom development and in planta growth of Xcc strain 8004 infiltrated into leaves of Arabidopsis ecotype Kas are not affected by XopAC. Figure S3. Symptom development and in planta growth of Xcc strain 8004 infiltrated into leaves of Arabidopsis ecotype Kas are not affected by XopAC. XopAC antibody preparation pENTRY-xopAC was recombined by LR in pDEST17 giving pDEST17-xopAC and transformed in E. coli BL21 (DE3) August 2013 \| Volume 8 \| Issue 8 \| e73469 10 PLOS ONE \| www.plosone.org xopAC-Triggered Immunity in Arabidopsis August 2013 \| Volume 8 \| Issue 8 \| e73469 5. Zipfel C, Kunze G, Chinchilla D, Caniard A, Jones JD et al. (2006) Perception of the bacterial PAMP EF-Tu by the receptor EFR restricts Agrobacterium-mediated transformation. Cell 125: 749-760. doi: 10.1016/j.cell.2006.03.037. PubMed: 16713565. 6. Danna CH, Millet YA, Koller T, Han SW, Bent AF et al. (2011) The Arabidopsis flagellin receptor FLS2 mediates the perception of Xanthomonas Ax21 secreted peptides. Proc Natl Acad Sci U S A 108: 9286-9291. doi:10.1073/pnas.1106366108. PubMed: 21576467. 7. Zhang J, Li W, Xiang T, Liu Z, Laluk K et al. (2010) Receptor-like cytoplasmic kinases integrate signaling from multiple plant immune receptors and are targeted by a Pseudomonas syringae effector. Cell Preparation of protein extracts, SDS-PAGE and immunoblotting Table S1. Strains and plasmids used in this study. (PDF) Yeast protein extracts were prepared under denaturing conditions as described [67] with the following modifications: after washing in buffer A, the cells were frozen twice in liquid nitrogen. After the last centrifugation step, the cells are washed with 100% ethanol, centrifuged at 11000 g and resuspended in 1X Laemmli buffer (0.5 M Tris pH 6.8, SDS 10%, glycerol 20% and 0.2 M DTT). Total soluble protein extracts were prepared from N. benthamiana leaf tissues as described [68]. Bacterial protein extracts were prepared by resuspending bacterial pellets directly in 1X Laemmli buffer. Samples were separated by SDS-PAGE (8 or 10% polyacrylamide) and transferred to nitrocellulose. The following antibodies were used: rat anti-HA (1867423; Roche), rabbit anti-GFP (B. Lefebvre, unpublished), mouse anti-c-Myc (Santa Cruz). Secondary antibodies were purchased from Sigma (AP conjugates) and detected as described [68]. 6. Danna CH, Millet YA, Koller T, Han SW, Bent AF et al. (2011) The Arabidopsis flagellin receptor FLS2 mediates the perception of Xanthomonas Ax21 secreted peptides. Proc Natl Acad Sci U S A 108: 9286-9291. doi:10.1073/pnas.1106366108. PubMed: 21576467. xopAC-Triggered Immunity in Arabidopsis (2013) Arabidopsis ubiquitin ligase MIEL1 mediates degradation of the transcription factor MYB30 weakening plant defence. Nat Communications 4: 1476. doi:10.1038/ncomms2479. PubMed: 23403577. 20. Mackey D, Holt BF, Wiig A, Dangl JL (2002) RIN4 interacts with Pseudomonas syringae type III effector molecules and is required for RPM1-mediated resistance in Arabidopsis. Cell 108: 743-754. doi: 10.1016/S0092-8674(02)00661-X. PubMed: 11955429. 39. Gyuris J, Golemis E, Chertkov H, Brent R (1993) Cdi1, a human G1 and S phase protein phosphatase that associates with Cdk2. Cell 75: 791-803. doi:10.1016/0092-8674(93)90498-F. PubMed: 8242750. 21. Kim MG, da Cunha L, McFall AJ, Belkhadir Y, DebRoy S et al. (2005) Two Pseudomonas syringae type III effectors inhibit RIN4-regulated basal defense in Arabidopsis. Cell 121: 749-759. doi:10.1016/j.cell. 2005.03.025. PubMed: 15935761. 40. Yarbrough ML, Li Y, Kinch LN, Grishin NV, Ball HL et al. (2009) AMPylation of Rho GTPases by Vibrio VopS disrupts effector binding and downstream signaling. Science 323: 269-272. doi:10.1126/science. 1166382. PubMed: 19039103. 22. Chung EH, da Cunha L, Wu AJ, Gao Z, Cherkis K et al. (2011) Specific threonine phosphorylation of a host target by two unrelated type III effectors activates a host innate immune receptor in plants. Cell Host Microbe 9: 125-136. doi:10.1016/j.chom.2011.01.009. PubMed: 21320695. 41. Hirayama T, Oka A (1992) Novel protein kinase of Arabidopsis thaliana (APK1) that phosphorylates tyrosine, serine and threonine. Plant Mol Biol 20: 653-662. doi:10.1007/BF00046450. PubMed: 1450380. 23. Liu J, Elmore JM, Fuglsang AT, Palmgren MG, Staskawicz BJ et al. (2009) RIN4 functions with plasma membrane H+-ATPases to regulate stomatal apertures during pathogen attack. PLOS Biol 7: e1000139. PubMed: 19564897. 42. Afzal AJ, Wood AJ, Lightfoot DA (2008) Plant receptor-like serine threonine kinases: roles in signaling and plant defense. Mol Plant Microbe Interact MPMI 21: 507-517. doi:10.1094/MPMI-21-5-0507. PubMed: 18393610. 24. Aarts N, Metz M, Holub E, Staskawicz BJ, Daniels MJ et al. (1998) Different requirements for EDS1 and NDR1 by disease resistance genes define at least two R gene-mediated signaling pathways in Arabidopsis. Proc Natl Acad Sci U S A 95: 10306-10311. doi:10.1073/ pnas.95.17.10306. PubMed: 9707643. 43. Veronese P, Nakagami H, Bluhm B, Abuqamar S, Chen X et al. (2006) The membrane-anchored Botrytis-INDUCED KINASE1 plays distinct roles in Arabidopsis resistance to necrotrophic and biotrophic pathogens. Plant Cell 18: 257-273. doi:10.1105/tpc.105.035576. PubMed: 16339855. 44. Rowland O, Ludwig AA, Merrick CJ, Baillieul F, Tracy FE et al. xopAC-Triggered Immunity in Arabidopsis Host Microbe 7: 290-301. doi:10.1016/j.chom.2010.03.007. PubMed: 20413097. campestris pv. campestris pathosystem. Mol Plant Pathol 6: 327-333. doi:10.1111/j.1364-3703.2005.00287.x. PubMed: 20565661. campestris pv. campestris pathosystem. Mol Plant Pathol 6: 327-333. doi:10.1111/j.1364-3703.2005.00287.x. PubMed: 20565661. 8. Chinchilla D, Zipfel C, Robatzek S, Kemmerling B, Nürnberger T et al. (2007) A flagellin-induced complex of the receptor FLS2 and BAK1 initiates plant defence. Nature 448: 497-500. doi:10.1038/nature05999. PubMed: 17625569. 28. Xu RQ, Blanvillain S, Feng JX, Jiang BL, Li XZ et al. (2008) AvrAC(Xcc8004), a type III effector with a leucine-rich repeat domain from Xanthomonas campestris pathovar campestris confers avirulence in vascular tissues of Arabidopsis thaliana ecotype Col-0. J Bacteriol 190: 343-355. doi:10.1128/JB.00978-07. PubMed: 17951377. 9. Sun W, Dunning FM, Pfund C, Weingarten R, Bent AF (2006) Within- species flagellin polymorphism in Xanthomonas campestris pv campestris and its impact on elicitation of Arabidopsis FLAGELLIN SENSING2-dependent defenses. Plant Cell 18: 764-779. doi:10.1105/ tpc.105.037648. PubMed: 16461584. 29. Guy E, Genissel A, Hajri A, Chabannes M, David P et al. (2013) Natural Genetic Variation of Xanthomonas campestris pv. campestris Pathogenicity on Arabidopsis Revealed by Association and Reverse Genetics. mBio. p. 4. 10. de Jonge R, van Esse HP, Kombrink A, Shinya T, Desaki Y et al. (2010) Conserved fungal LysM effector Ecp6 prevents chitin-triggered immunity in plants. Science 329: 953-955. doi:10.1126/science. 1190859. PubMed: 20724636. 30. Feng F, Yang F, Rong W, Wu X, Zhang J et al. (2012) A Xanthomonas uridine 5'-monophosphate transferase inhibits plant immune kinases. Nature 485: 114-118. doi:10.1038/nature10962. PubMed: 22504181. 31. Liu J, Elmore JM, Lin ZJ, Coaker G (2011) A receptor-like cytoplasmic kinase phosphorylates the host target RIN4, leading to the activation of a plant innate immune receptor. Cell Host Microbe 9: 137-146. doi: 10.1016/j.chom.2011.01.010. PubMed: 21320696. 11. Melotto M, Underwood W, Koczan J, Nomura K, He SY (2006) Plant stomata function in innate immunity against bacterial invasion. Cell 126: 969-980. doi:10.1016/j.cell.2006.06.054. PubMed: 16959575. j 12. Galán JE, Wolf-Watz H (2006) Protein delivery into eukaryotic cells by type III secretion machines. Nature 444: 567-573. doi:10.1038/ nature05272. PubMed: 17136086. 32. Wengelnik K, Rossier O, Bonas U (1999) Mutations in the regulatory gene hrpG of Xanthomonas campestris pv. vesicatoria result in constitutive expression of all hrp genes. J Bacteriol 181: 6828-6831. PubMed: 10542187. 13. Lindeberg M, Cunnac S, Collmer A (2012) Pseudomonas syringae type III effector repertoires: last words in endless arguments. Trends Microbiol 20: 199-208. doi:10.1016/j.tim.2012.01.003. PubMed: 22341410. 33. xopAC-Triggered Immunity in Arabidopsis Noël L, Thieme F, Nennstiel D, Bonas U (2001) cDNA-AFLP analysis unravels a genome-wide hrpG-regulon in the plant pathogen Xanthomonas campestris pv. vesicatoria. Mol Microbiol 41: 1271-1281. doi:10.1046/j.1365-2958.2001.02567.x. PubMed: 11580833. 14. White FF, Potnis N, Jones JB, Koebnik R (2009) The type III effectors of Xanthomonas. Mol Plant Pathol 10: 749-766. doi:10.1111/j. 1364-3703.2009.00590.x. PubMed: 19849782. j 34. Remigi P, Anisimova M, Guidot A, Genin S, Peeters N (2011) Functional diversification of the GALA type III effector family contributes to Ralstonia solanacearum adaptation on different plant hosts. New Phytol 192: 976-987. doi:10.1111/j.1469-8137.2011.03854.x. PubMed: 21902695. 15. Poueymiro M, Genin S (2009) Secreted proteins from Ralstonia solanacearum: a hundred tricks to kill a plant. Curr Opin Microbiol 12: 44-52. doi:10.1016/j.mib.2008.11.008. PubMed: 19144559. 35. Meyer D, Cunnac S, Guéneron M, Declercq C, Van Gijsegem F et al. (2006) PopF1 and PopF2, two proteins secreted by the type III protein secretion system of Ralstonia solanacearum, are translocators belonging to the HrpF/NopX family. J Bacteriol 188: 4903-4917. doi: 10.1128/JB.00180-06. PubMed: 16788199. 16. Shao F, Golstein C, Ade J, Stoutemyer M, Dixon JE et al. (2003) Cleavage of Arabidopsis PBS1 by a bacterial type III effector. Science 301: 1230-1233. doi:10.1126/science.1085671. PubMed: 12947197. 17. Lu D, Wu S, Gao X, Zhang Y, Shan L et al. (2010) A receptor-like cytoplasmic kinase, BIK1, associates with a flagellin receptor complex to initiate plant innate immunity. Proc Natl Acad Sci U S A 107: 496-501. doi:10.1073/pnas.0909705107. PubMed: 20018686. 36. Rossier O, Wengelnik K, Hahn K, Bonas U (1999) The Xanthomonas Hrp type III system secretes proteins from plant and mammalian bacterial pathogens. Proc Natl Acad Sci U S A 96: 9368-9373. doi: 10.1073/pnas.96.16.9368. PubMed: 10430949. p 18. Shiu SH, Bleecker AB (2001) Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases. Proc Natl Acad Sci U S A 98: 10763-10768. doi:10.1073/pnas. 181141598. PubMed: 11526204. 37. Sohn KH, Lei R, Nemri A, Jones JD (2007) The downy mildew effector proteins ATR1 and ATR13 promote disease susceptibility in Arabidopsis thaliana. Plant Cell 19: 4077-4090. doi:10.1105/tpc. 107.054262. PubMed: 18165328. 19. Burr CA, Leslie ME, Orlowski SK, Chen I, Wright CE et al. (2011) CAST AWAY, a membrane-associated receptor-like kinase, inhibits organ abscission in Arabidopsis. Plant Physiol 156: 1837-1850. doi: 10.1104/pp.111.175224. PubMed: 21628627. 38. Marino D, Froidure S, Canonne J, Ben Khaled S, Khafif M et al. References 1. Jones JD, Dangl JL (2006) The plant immune system. Nature 444: 323-329. doi:10.1038/nature05286. PubMed: 17108957. 5. Zipfel C, Kunze G, Chinchilla D, Caniard A, Jones JD et al. (2006) Perception of the bacterial PAMP EF-Tu by the receptor EFR restricts Agrobacterium-mediated transformation. Cell 125: 749-760. doi: 10.1016/j.cell.2006.03.037. PubMed: 16713565. 2. Zipfel C, Felix G (2005) Plants and animals: a different taste for microbes? Curr Opin Plant Biol 8: 353-360. doi:10.1016/j.pbi. 2005.05.004. PubMed: 15922649. 2. Zipfel C, Felix G (2005) Plants and animals: a different taste for microbes? Curr Opin Plant Biol 8: 353-360. doi:10.1016/j.pbi. 2005.05.004. PubMed: 15922649. 6. Danna CH, Millet YA, Koller T, Han SW, Bent AF et al. (2011) The Arabidopsis flagellin receptor FLS2 mediates the perception of Xanthomonas Ax21 secreted peptides. Proc Natl Acad Sci U S A 108: 9286-9291. doi:10.1073/pnas.1106366108. PubMed: 21576467. 6. Danna CH, Millet YA, Koller T, Han SW, Bent AF et al. (2011) The Arabidopsis flagellin receptor FLS2 mediates the perception of Xanthomonas Ax21 secreted peptides. Proc Natl Acad Sci U S A 108: 9286-9291. doi:10.1073/pnas.1106366108. PubMed: 21576467. 3. Ronald PC, Beutler B (2010) Plant and animal sensors of conserved microbial signatures. Science 330: 1061-1064. doi:10.1126/science. 1189468. PubMed: 21097929. 3. Ronald PC, Beutler B (2010) Plant and animal sensors of conserved microbial signatures. Science 330: 1061-1064. doi:10.1126/science. 1189468. PubMed: 21097929. 4. Gómez-Gómez L, Boller T (2000) FLS2: an LRR receptor-like kinase involved in the perception of the bacterial elicitor flagellin in Arabidopsis. Mol Cell 5: 1003-1011. doi:10.1016/ S1097-2765(00)80265-8. PubMed: 10911994. 7. Zhang J, Li W, Xiang T, Liu Z, Laluk K et al. (2010) Receptor-like cytoplasmic kinases integrate signaling from multiple plant immune receptors and are targeted by a Pseudomonas syringae effector. Cell 11 August 2013 \| Volume 8 \| Issue 8 \| e73469 PLOS ONE \| www.plosone.org xopAC-Triggered Immunity in Arabidopsis diseases resistance in Arabidopsis. Genetics 152: 1-12. PubMed: 10224239. 47. Eitas TK, Nimchuk ZL, Dangl JL (2008) Arabidopsis TAO1 is a TIR-NB- LRR protein that contributes to disease resistance induced by the Pseudomonas syringae effector AvrB. Proc Natl Acad Sci U S A 105: 6475-6480. doi:10.1073/pnas.0802157105. PubMed: 18424557. 59. Deslandes L, Pileur F, Liaubet L, Camut S, Can C et al. (1998) Genetic characterization of RRS1, a recessive locus in Arabidopsis thaliana that confers resistance to the bacterial soilborne pathogen Ralstonia solanacearum. Mol Plant Microbe Interact MPMI 11: 659-667. doi: 10.1094/MPMI.1998.11.7.659. PubMed: 9650298. p 48. Stael S, Bayer RG, Mehlmer N, Teige M (2011) Protein N-acylation overrides differing targeting signals. FEBS Lett 585: 517-522. doi: 10.1016/j.febslet.2011.01.001. PubMed: 21219905. 49. Elmore JM, Liu J, Smith B, Phinney B, Coaker G (2012) Quantitative proteomics reveals dynamic changes in the plasma membrane during Arabidopsis immune signaling. Molecular Cell Proteomics MCP 11: 014555. PubMed: 22215637. 60. Schäfer A, Tauch A, Jäger W, Kalinowski J, Thierbach G et al. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145: 69-73. doi:10.1016/0378-1119(94)90324-7. PubMed: 8045426. 50. Afzal AJ, da Cunha L, Mackey D (2011) Separable fragments and membrane tethering of Arabidopsis RIN4 regulate its suppression of PAMP-triggered immunity. Plant Cell 23: 3798-3811. doi:10.1105/tpc. 111.088708. PubMed: 21984695. ( ) 61. Engler C, Kandzia R, Marillonnet S (2008) A one pot, one step, precision cloning method with high throughput capability. PLOS ONE 3: e3647. doi:10.1371/journal.pone.0003647. PubMed: 18985154. 51. van der Hoorn RA, Kamoun S (2008) From Guard to Decoy: a new model for perception of plant pathogen effectors. Plant Cell 20: 2009-2017. doi:10.1105/tpc.108.060194. PubMed: 18723576. 62. Ditta G, Stanfield S, Corbin D, Helinski DR (1980) Broad host range DNA cloning system for Gram-negative bacteria: Construction of a gene bank of Rhizobium meliloti. Proc Natl Acad Sci U S A 77: 7347-7351. doi:10.1073/pnas.77.12.7347. PubMed: 7012838. 52. Blanvillain S, Meyer D, Boulanger A, Lautier M, Guynet C et al. (2007) Plant carbohydrate scavenging through TonB-dependent receptors: a feature shared by phytopathogenic and aquatic bacteria. PLOS ONE 2: e224. doi:10.1371/journal.pone.0000224. PubMed: 17311090. 63. Figurski DH, Helinski DR (1979) Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc Natl Acad Sci U S A 76: 1648-1652. doi:10.1073/pnas. 76.4.1648. PubMed: 377280. 53. xopAC-Triggered Immunity in Arabidopsis (2005) Functional analysis of Avr9/Cf-9 rapidly elicited genes identifies a protein kinase, ACIK1, that is essential for full Cf-9-dependent disease resistance in tomato. Plant Cell 17: 295-310. doi:10.1105/tpc. 104.026013. PubMed: 15598806. 25. Vicente JG, Holub EB (2013) Xanthomonas campestris pv. campestris (cause of black rot of crucifers) in the genomic era is still a worldwide threat to Brassica crops. Mol Plant Pathol 14: 2-18. doi:10.1111/j. 1364-3703.2012.00833.x. PubMed: 23051837. 26. Arlat M, Gough CL, Barber CE, Boucher C, Daniels MJ (1991) Xanthomonas campestris contains a cluster of hrp genes related to the larger hrp cluster of Pseudomonas solanacearum. Mol Plant Microbe Interact MPMI 4: 593-601. doi:10.1094/MPMI-4-593. PubMed: 1666525. 45. Kinch LN, Yarbrough ML, Orth K, Grishin NV (2009) Fido, a novel AMPylation domain common to fic, doc, and AvrB. PLOS ONE 4: e5818. doi:10.1371/journal.pone.0005818. PubMed: 19503829. 46. Al-Daoude A, de Torres Zabala M, Ko JH, Grant M (2005) RIN13 is a positive regulator of the plant disease resistance protein RPM1. Plant Cell 17: 1016-1028. doi:10.1105/tpc.104.028720. PubMed: 15722472. 27. Meyer D, Lauber E, Roby D, Arlat M, Kroj T (2005) Optimization of pathogenicity assays to study the Arabidopsis thaliana-Xanthomonas PLOS ONE \| www.plosone.org August 2013 \| Volume 8 \| Issue 8 \| e73469 12 xopAC-Triggered Immunity in Arabidopsis xopAC-Triggered Immunity in Arabidopsis Boucher CA, Barberis P, Trigalet AP, Demery DA (1985) Transposon mutagenesis of Pseudomonas solanacearum: isolation of Tn5-induced avirulent mutants. J Gen Microbiol 131: 2449-2457. 64. Cunnac S, Occhialini A, Barberis P, Boucher C, Genin S (2004) Inventory and functional analysis of the large Hrp regulon in Ralstonia solanacearum: identification of novel effector proteins translocated to plant host cells through the type III secretion system. Mol Microbiol 53: 115-128. doi:10.1111/j.1365-2958.2004.04118.x. PubMed: 15225308. 54. King EO, Ward MK, Raney DE (1954) Two simple media for the demonstration of pyocyanin and fluorescin. J Lab Clin Med 44: 301-307. PubMed: 13184240. 55. Yu GL, Katagiri F, Ausubel FM (1993) Arabidopsis mutations at the RPS2 locus result in loss of resistance to Pseudomonas syringae strains expressing the avirulence gene avrRpt2. MPMI 6: 434-443. doi: 10.1094/MPMI-6-434. PubMed: 8400373. 65. Rivas S, Rougon-Cardoso A, Smoker M, Schauser L, Yoshioka H et al. (2004) CITRX thioredoxin interacts with the tomato Cf-9 resistance protein and negatively regulates defence. EMBO J 23: 2156-2165. doi: 10.1038/sj.emboj.7600224. PubMed: 15131698. 66. Gietz RD, Woods RA (2001) Genetic transformation of yeast. BioTechniques 30: 816-820, 822-816, 828 passim 56. Mackey D, Belkhadir Y, Alonso JM, Ecker JR, Dangl JL (2003) Arabidopsis RIN4 is a target of the type III virulence effector AvrRpt2 and modulates RPS2-mediated resistance. Cell 112: 379-389. doi: 10.1016/S0092-8674(03)00040-0. PubMed: 12581527. 67. Caspari T, Dahlen M, Kanter-Smoler G, Lindsay HD, Hofmann K et al. (2000) Characterization of Schizosaccharomyces pombe Hus1: a PCNA-related protein that associates with Rad1 and Rad9. Mol Cell Biol 20: 1254-1262. doi:10.1128/MCB.20.4.1254-1262.2000. PubMed: 10648611. 57. Grant MR, Godiard L, Straube E, Ashfield T, Lewald J et al. (1995) Structure of the Arabidopsis RPM1 gene enabling dual specificity disease resistance. Science 269: 843-846. doi:10.1126/science. 7638602. PubMed: 7638602. 68. Witte CP, Noël LD, Gielbert J, Parker JE, Romeis T (2004) Rapid one- step protein purification from plant material using the eight-amino acid StrepII epitope. Plant Mol Biol 55: 135-147. doi:10.1007/ s11103-004-0501-y. PubMed: 15604670. 58. Warren RF, Merritt PM, Holub E, Innes RW (1999) Identification of three putative signal transduction genes involved in R gene-specified PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 13 August 2013 \| Volume 8 \| Issue 8 \| e73469
https://openalex.org/W3005630125	https://figshare.com/articles/journal_contribution/A_Cost_and_Cost-Benefit_Analysis_of_the_Stand_More_AT_Work_SMArT_Work_Intervention/11902338/1/files/21827661.pdf	English	null	A Cost and Cost-Benefit Analysis of the Stand More AT Work (SMArT Work) Intervention	International journal of environmental research and public health/International journal of environmental research and public health	2,020	cc-by	5,420	Received: 18 December 2019; Accepted: 6 February 2020; Published: 13 February 2020 Abstract: This study conducted a cost and cost-beneﬁt analysis of the Stand More AT (SMArT) Work workplace intervention, designed to reduce sitting time. The study was a cluster two-armed randomised controlled trial involving 37 oﬃce clusters (146 desk-based workers) in a National Health Service Trust. The intervention group received a height-adjustable workstation with supporting behaviour change strategies. The control group continued with usual practice. Self-report absenteeism, presenteeism and work productivity were assessed at baseline, 3, 6 and 12 months; and organisational sickness absence records 12 months prior to, and 12 months of the intervention. Mean per employee costs associated with SMArT Work were calculated. Absenteeism, presenteeism and work productivity were estimated, and employer-recorded absence data and employee wage-banding were used to provide a human-capital-based estimate of costs to the organisation. The return-on-investment (ROI) and incremental cost-eﬃcacy ratios (ICER) were calculated. Intervention cost was £692.40 per employee. Cost-beneﬁt estimates show a net saving of £1770.32 (95%CI £-354.40, £3895.04) per employee as a result of productivity increase. There were no signiﬁcant diﬀerences in absence data compared to the control group. SMArT Work provides supporting evidence for policy-makers and employers on the cost beneﬁts of reducing sitting time at work. Keywords: cost-beneﬁt analysis; sitting; standing; sit-stand; presenteeism; sick leave; sickness absence; workplace health promotion A Cost and Cost-Beneﬁt Analysis of the Stand More AT Work (SMArT Work) Intervention Fehmidah Munir 1,* , Paul Miller 2, Stuart J.H. Biddle 3 , Melanie J. Davies 4,5,6, David W. Dunstan 7,8 , Dale W. Esliger 1, Laura J. Gray 9, Sophie E. O’Connell 6, Ghazala Waheed 4 , Thomas Yates 4,5 and Charlotte L. Edwardson 4,5,,† 1 School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough LE11 3TU, UK D.Esliger@lboro.ac.uk 2 Miller Economics Ltd., Biohub Alderley Park, Alderley Edge SK10 4TG, UK; drpsjmiller@gmail.com 3 Institute for Resilient Regions, University of Southern Queensland, Education City, Springﬁeld Central, QLD 4300, Australia; Stuart.Biddle@usq.edu.au 2 Miller Economics Ltd., Biohub Alderley Park, Alderley Edge SK10 4TG, UK; drpsjmiller@gmail.com 3 Institute for Resilient Regions, University of Southern Queensland, Education City, Springﬁeld Central, QLD 4300, Australia; Stuart.Biddle@usq.edu.au 4 Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester LE5 4PW, UK; Melanie.davies@uhl-tr.nhs.uk (M.J.D.); Gw136@leicester.ac.uk (G.W.); ty20@le.ac.uk (T.Y.) 5 4 Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester LE5 4PW, UK; Melanie.davies@uhl-tr.nhs.uk (M.J.D.); Gw136@leicester.ac.uk (G.W.); ty20@le.ac.uk (T.Y.) 5 NIHR Leicester Biomedical Research Centre, Leicester General Hospital, Leicester LE5 4PW, UK 6 L i t Di b t C t U i it H it l f L i t L i t G l H it l 4 Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester LE5 4PW, UK; Melanie.davies@uhl-tr.nhs.uk (M.J.D.); Gw136@leicester.ac.uk (G.W.); ty20@le.ac.uk (T.Y.) 5 NIHR Leicester Biomedical Research Centre, Leicester General Hospital, Leicester LE5 4PW, UK 6 Leicester Diabetes Centre, University Hospitals of Leicester, Leicester General Hospital, Leicester LE5 4PW, UK; Sophie.OConnell@uhl-tr.nhs.uk 7 Baker Heart and Diabetes Institute, Melbourne, VIC, Australia; David.Dunstan@bakeridi.edu.au 8 Mary MacKillop Institute for Health Research, The Australian Catholic University, Melbourne, VIC 3000, Australia 9 Department of Health Sciences, University of Leicester, Leicester LE1 7RH, UK; lg48@leicester.ac.uk 9 Department of Health Sciences, University of Leicester, Leicester LE1 7RH, UK; lg48@leicester.ac.uk Correspondence: f.munir@lboro.ac.uk (F.M.); ce95@le.ac.uk (C.L.E.) † Joint ﬁrst author. † Joint ﬁrst author. International Journal of Environmental Research and Public Health 2.2. Ethical Considerations All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki. This study was approved by Loughborough University, and Research and Innovation approval was obtained from the University Hospitals of Leicester NHS Trust (EDGE ID 34571). 1. Introduction High levels of sedentary behaviour (sitting time) have been identiﬁed as an important modiﬁable behavioural risk factor for multiple chronic diseases [1–4], poor mental health [5,6] and premature www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2020, 17, 1214; doi:10.3390/ijerph17041214 2 of 9 Int. J. Environ. Res. Public Health 2020, 17, 1214 mortality [4]. In oﬃce workers, workplace sitting accounts for the largest proportion of daily sitting time [7]; with 70%-85% of time spent sitting at work, and over a third of that total sitting time undertaken in bouts of over 30 min at a time—prolonged sitting [8,9]. With problems such as neck and shoulder pain, high presenteeism and low work engagement associated with sitting time at work [10–12], the oﬃce workplace is a priority setting to reduce total occupational sitting time [13]. Whilst workplace interventions have shown that signiﬁcant short and medium-term reductions in workplace sitting time can be achieved [8,14–17], few have reported on the cost-beneﬁts of their interventions [18], and this lack of evidence may act as a barrier for organisations in adopting and promoting strategies to reduce sitting time at work. Evaluating the cost-beneﬁts of reducing occupational sitting time are important due to the estimated costs of £30bn ($39bn; €34bn) for sickness presenteeism (i.e., working despite being unwell to work) [19], and sickness absenteeism in the United Kingdom, with presenteeism costing twice as much as absenteeism [20]. The Stand More AT Work (SMArT Work) intervention is a multicomponent intervention designed to reduce occupational sitting time. It was tested within a cluster randomised controlled trial over 12 months in a sample of oﬃce workers working within the English National Health Service. The intervention successfully reduced sitting time over the short, medium and longer term, and led to positive changes in work related and psychological health [21]. This study reports the cost-beneﬁt analysis of the SMArT Work intervention by providing an assessment of the costs associated with implementing the intervention over a 12- month period. The cost-analysis was conducted from the perspective of the employer—who generally pays the costs of implementing workplace interventions—and the intervention costs were measured against a control group. The aim of the study was to provide accurate cost data and the cost-beneﬁt of the intervention. This will support both employers and public health policy makers in drawing evidence-based decisions concerning the economic feasibility and scalability of reducing workplace sitting time. 2.1. Study Design A detailed description of the intervention development, randomised controlled trial design and main results, have been published previously [21,22]. Brieﬂy, SMArT Work was a two-arm cluster randomised controlled trial, consisting of an intervention group and a control group. Randomization conducted at oﬃce level at a ratio of 1:1. Participants in the intervention group received a height-adjustable workstation (a choice of workstation—a full, electrical desk or VARIDESK desk platform) as well as supporting behaviour change strategies including education, behavioural feedback, self-monitoring and prompt tool (DARMA cushion), quarterly coaching sessions and information leaﬂets and motivational posters aimed at reducing sitting time in staﬀwho were predominantly desk-based. The control group continued in their usual working environment. 2.4. Intervention Costs The total and per employee cost for providing this intervention package for the 12-month study period was calculated. The costs of the intervention were estimated based on material costs and research facilitator costs to implement the intervention. Material costs included the expense of the height adjustable workstation, the DARMA cushion (self-monitoring and prompt tool) and printing costs of supporting documents, i.e., action plan and goal setting diary, feedback sheets on behaviour, posters and education material. Facilitator costs, estimated using the facilitator’s hourly wage (£18.20), included removal of the existing desk for those choosing the electric height adjustable desk, installation of the height adjustable workstation, including demonstration on how to use it, production of behaviour feedback, coaching sessions and delivery of an educational seminar. Indirect work loss costs associated with participant engagement with the intervention (e.g., attendance at seminar) was estimated using the mean hourly wage rate for the intervention cohort (£16.01) based on the mid-point of each wage band. 2.3. Analyses Approach A simple cost-beneﬁt analysis was constructed by synthesizing: a) cost analyses—the mean per employee costs associated with implementing the SMArT Work intervention; and b) organizational eﬀect measures—the change in absenteeism, presenteeism and overall work productivity gain/loss expressed in terms of time and estimated costs to the organization. Where both costs and organizational eﬀect measures were estimated in money terms, the net eﬀect in terms of cost-beneﬁt Int. J. Environ. Res. Public Health 2020, 17, 1214 3 of 9 was calculated (i.e., to what extent are beneﬁts/eﬀects expected to outweigh the costs, and the estimated return-on-investment (ROI)). was calculated (i.e., to what extent are beneﬁts/eﬀects expected to outweigh the costs, and the estimated return-on-investment (ROI)). 2.5. Measures of Eﬀects Absenteeism (work time missed), presenteeism (impairment at work/reduced on-the-job eﬀectiveness), work productivity (overall change in productivity/absenteeism plus presenteeism) and activity impairment were estimated via self-report using the Work Productivity and Activity Impairment Questionnaire: General Health V2.0 (WPAI:GH) [23]. Absolute estimates of absenteeism, presenteeism and work productivity from the WPAI measure were multiplied by individual employee wage-banding information to provide a human-capital-based estimate of costs to the employing organization. WPAI-based time and cost data for the intervention and control groups were compared at the four time points (baseline, 3 months, 6 months and 12 months). Analyses were based on the mean of these time points. Productivity and sickness absence were analysed using Generalized Estimating Equation (GEE) model with an exchangeable correlation structure and accounting for clustering. The model included a binary indicator for randomization group and adjusted for cluster size (<= 4 participants or > 4 participants). We also adjusted for sex and ethnicity as these diﬀered between the intervention and control group at baseline, with more males (27.3 vs. 13.0%) and South Asians (20.8 vs. 13.0%) represented in the intervention group Analyses were conducted using Stata version 14. Objectively measured absenteeism data were also collected through employer-recorded absence for the intervention and control participants who consented and were compared for two time periods: 12 months prior to the study initiation (period 1) and for the 12 months of the intervention and follow-up (period 2). The change from time period 1 to 2 was compared across the two groups. Absolute measures of absenteeism based on employer-recorded data were multiplied by individual employee wage-banding information to provide a human-capital-based estimate of costs to the employing organization. 2.6. Cost Beneﬁt Analysis All the economic calculations presented are based on the full cohort data for control and intervention. Two separate economic analyses were conducted based on the two distinct sources of data to estimate measures of organizational eﬀects (self-report vs. archival data). In both analyses, measures of eﬀects (change in work loss time) were monetized using the human-capital-based method applying the employee’s wage rate to value their productive time at work. The net impact in terms of cost-beneﬁt was also calculated, only where beneﬁts are found to outweigh costs this net impact will be a positive number. The ratio of costs to monetized beneﬁts were used to calculate the return-on-investment (ROI). Net impact cost-beneﬁt and ROI were calculated for Int. J. Environ. Res. Public Health 2020, 17, 1214 4 of 9 the whole intervention group compared to controls deﬁned by key parameters including observed degree of exposure [21]. Finally, incremental cost-eﬃcacy ratios (ICER) were calculated as the diﬀerence in costs of the intervention and control group divided by the diﬀerence in their eﬀect. The ICER was expressed as the cost per unit (minutes per workday) reduction in workplace sitting time at 12 months. Both complete case and Intention to treat data imputed using multiple imputation [21], were used for ICER calculations. Means of normal distributions and standard distributions were used (the latter expressed as the mean of non-zero) for both the intervention and control group. 3.1. Intervention Costs Table 1 shows the breakdown of the intervention costs. Overall intervention costs were calculated at £595 per participating employee, with ~90% of costs attributed to equipment (workstation and cushion). Table 2 shows the indirect work loss costs associated with participant work time spent engaging with the intervention, estimated at £97.40 per participant over the 12 months. The aggregate costs of the intervention (direct + indirect) over the 12-month period was calculated as £595 + £97.40 = £692.40 per participant. Table 1. Intervention cost breakdown: Materials and facilitator time. Table 1. Intervention cost breakdown: Materials and facilitator time. Item Unit Cost Quantity Total Cost Desks a Varidesk 36 Pro Plus £301.50 38 £11,457 Varidesk 40 Plus £352.50 7 £2467.50 Varidesk installation £12.50 45 £562.50 Electric desk £453.71 30 £13,611.30 Removal of old desk £18.50 30 £555 Electric desk installation £7.50 30 £225 Sub-Total £28,878.30 ActivPAL feedback b Report production £4.55 255 £1160.25 DARMA a £169 75 £12,675 Information support b Design and delivery of Seminars £22.75 c 8 £182 Printing of posters/leaﬂets/diaries £6.99 75 £524.25 Researcher demonstration of desk use £0.91 75 £68 Coaching £15.16 75 £1,138 Sub-Total £1912.25 TOTAL Per participant £44,625.80 75 £595 a RRP; b based on researcher time £18.20 per hour; c includes organisation time. Table 2. Participant time spent on intervention and estimated workloss costs. Intervention Element Time/cost to participant Workloss Costing over 12m (@ mean hourly rate £16.02) Desks set-up and demonstration Varidesk: 10 min to set up (n = 45) Electric desk: 30 min to remove old desk and set up new desk (n = 30) Assume 15 mins one-oﬀworkloss = £4.00 Seminar Approx. 45 min of their workday to attend a one oﬀseminar on site Assume 45mins one-oﬀworkloss = £12.02 Table 2. Participant time spent on intervention and estimated workloss costs. Int. J. Environ. Res. Public Health 2020, 17, 1214 5 of 9 Table 2. Cont. Table 2. Cont. 3.1. Intervention Costs Intervention Element Time/cost to participant Workloss Costing over 12m (@ mean hourly rate £16.02) activPAL feedback reading and reading initial leaﬂet Self-monitoring sitting via DARMA cushion (per unit) Gola setting diaries activPAL and leaﬂet total time per participant is 15 min Utilising DARMA feedback and diaries a total time of 5 min per week per participant Workloss (505mins + 15min = 265mins) = £70.71 Coaching sessions Brief 10 min with researcher to discuss progress, motivations, goals and plans every 3 months = a total of 40 min over a 12-month period per participant Assume 40mins one-oﬀworkloss = £10.67 TOTAL Intervention cohort (n = 75) Per participant £7305 £97.40 a Adjusted diﬀerence in the productivity and sickness absence between treatment groups (intervention group compared to control group) with 95% conﬁdence interval, p-value; adjusted for cluster eﬀect, sex, ethnicity and stratiﬁcation categories (oﬃce size <4 and oﬃce size>=4). 4. Discussion This analysis reports the costs of a multi-component intervention to reduce sitting time in desk-based workers and the cost-beneﬁt from an employer perspective. When considering all intervention components, regardless of whether the participants engaged with them, the intervention was estimated to cost £692.40 per person, with approximately 90% of these costs attributed to the height-adjustable workstation and Darma cushion. The intervention is estimated to provide a net beneﬁt (saving) per employee, calculated over the course of the 12 month randomised controlled trial, of £1770.32 (mean of all follow up time points) as a result of reduction in productivity loss due to health problems. The return on investment is 256%. In other words, £2.56 would be returned for every £1 spent on the intervention. The results of the randomised controlled trial [21], showed no diﬀerence in the eﬀectiveness of the intervention between the two diﬀerent types of height-adjustable workstations used. However, there is a substantial price diﬀerence between these workstations in addition to installation and removal costs, resulting in the electric workstation being approximately £150 more expensive than the platform workstation. Furthermore, our process evaluation found the majority of participants did not use the self-monitoring device (Darma cushion) and nearly 25% of the total intervention costs were attributed to this device. Some participants chose not to self-monitor or use formal prompts, whereas others sought their own free methods such as electronic timers and computer software for tracking their sitting time and providing reminders to break up their sitting. If the intervention was costed based on the platform workstation and the use of free self-monitoring and prompt methods, the cost of the intervention would reduce from £692.40 to £430.87 per person (intervention and work loss costs included) and the return on investment would increase to 472% (£4.72 returned for each £1 spent). Furthermore, it can be reasonably expected that as height-adjustable workstations become more common, prices will continue to fall, which will increase the return on investment. The primary outcome of the study was change in occupational sitting time at 12-month follow up [21]. A signiﬁcant diﬀerence between groups (in favour of the intervention group) was found in occupational sitting time at 12 months (−83.28 min/workday, 95% conﬁdence interval −116.57 to − 49.98, p = 0.001). 3.2. Measures of Eﬀects Table 3 shows self-reported absence and productivity. Results show employee productivity improved in the intervention group (mean 1.75hrs better per week) and worsened in the control group (mean 0.44hrs worse) from baseline to 12 months. Signiﬁcant diﬀerences between the groups were found with respect to the cost of self-reported productivity: adjusted diﬀerence mean £47.36 (95%CI £6.50, £88.22) savings for past 7 days. There were no signiﬁcant diﬀerences in self-reported absenteeism due to health reasons between the intervention and control groups nor when archival absence data were used (data not shown). Table 3. Self-reported productivity change and absenteeism due to health problems. Metrics Change from Baseline hrs £’s Productivity change Intervention n 65 Mean non-zero 2.41 £62.83 Mean (95% CI) 1.75 (gain) (0.09 to 3.42) £33.72 (gain) (£60.87 to -£6.58) Control n 52 Mean non-zero -1.07 -£35.21 Mean (95% CI) -0.44 (loss) (-2.09 to 1.21) -£17.68 (loss) (-£50.44 to £15.08) Delta (I-C) Mean (95%CI) 2.19 (-0.16 to 4.55) £51.41 (£9.68 to £93.13) Adjusted diﬀerence a Mean (95% CI) 1.58 (-0.50 to 3.67) £47.36 (£6.50 to £88.22) p-value 0.137 0.023 Sickness Absence Intervention n 66 Mean non-zero 3.65 £46.52 Mean (95%CI) 0.28 (-0.43 to 0.99) £5.42 (-£3.76 to £14.61) Control n 52 Mean non-zero 7.67 £64.86 Mean (95%CI) 0.23 (-0.77 to 1.22) £3.33 (-£8.46 to £15.13) Delta (I-C) Mean (95% CI) 0.05 (-1.12 to 1.23) £2.09 (-£12.47 to £16.64) Adjusted diﬀerence a Mean (95% CI) 0.07 (-1.27 to 1.41) £3.82 (-£13.41 to £21.04) p-value 0.918 0.664 a Adjusted diﬀerence in the productivity and sickness absence between treatment groups (intervention group compared to control group) with 95% conﬁdence interval, p-value; adjusted for cluster eﬀect, sex, ethnicity and stratiﬁcation categories (oﬃce size <4 and oﬃce size>=4). Table 3. Self-reported productivity change and absenteeism due to health problems. a Adjusted diﬀerence in the productivity and sickness absence between treatment groups (intervention group compared to control group) with 95% conﬁdence interval, p-value; adjusted for cluster eﬀect, sex, ethnicity and stratiﬁcation categories (oﬃce size <4 and oﬃce size>=4). Int. J. Environ. Res. Public Health 2020, 17, 1214 6 of 9 3.3. Cost Beneﬁt In terms of cost-beneﬁt analyses, the delta costs and eﬀects observed between the intervention and control groups are as follows: 1. ∆cost = + £595 (intervention costs) + £97.40 (indirect work loss costs) = + £692.40 2. ∆eﬀect (1) = WPAI productivity gain £47.36 per week x 52 weeks = £2462.72 3. (95%CI∆£6.50, £88.22 per week x 52 weeks = £338; £4587.44) 4. ∆eﬀect (2) = Archival sickness absence = no diﬀerence 5. ∆eﬀect (3) = WPAI sickness absence = no diﬀerence Therefore, the SMArT Work intervention was estimated to provide a net beneﬁt in productivity per intervention group employee over 52 weeks of £1770.32 (£2462.72 - £692.40) (95%CI∆£-354.40, £3895.04). This equates to a return on investment of 256% ((£2462.72-692.40) / £692.40 100). The productivity gains associated with SMArT Work (Table 3 - £47.36 savings for past 7 days), equates to mean wage rates of £29.97. Where productivity gains associated with the intervention are circa 2 hours per week and hourly wage rates vary between £10 and £30, savings will be between £20 and £60 per week per employee. p p y In terms of cost per unit (minutes per workday) reduction in workplace sitting time at 12 months, the ICER for the SMArT Work intervention is £8.31 (using complete case data), £8.48 (using intention to treat data) and £16.77 (standardised to 8h workday). 4. Discussion In terms of cost per unit (minutes per workday) reduction in workplace sitting time at 12 months, the ICER for the SMArT Work intervention is £8.31 (complete case), £8.48 (ITT population) and £16.77 (standardised to 8h workday). There are a lack of studies reporting intervention costs and 7 of 9 Int. J. Environ. Res. Public Health 2020, 17, 1214 beneﬁts, however, one recently published study reported the economic evaluation of the Stand Up Victoria sitting reduction intervention in Australia (Gao et al 2018). Intervention costs were reported at $431AUD (~£235) per participant and the ICER was AU$9.94 (~£5.49) cost per minute reduction in workplace sitting time, which is less than the SMArT Work intervention costs. There are some explanations for this. Although, both interventions were multi-component interventions including a height-adjustable desk per participant, diﬀerent types of desks were used across these studies, which varied in price. Furthermore, other intervention components, although similar (e.g., education) were implemented diﬀerently. The SMArT Work intervention also included expensive self-monitoring and prompt equipment. These factors will have led to a diﬀerence in cost. Moreover, in the Stand Up Victoria study, the cost of the height-adjustable workstations was annuitized over 5 years, this was not done in the current analysis. y This cost analysis was conducted from the perspective of the employer, hence the focus on productivity gains, presenteeism and absenteeism. In a recent paper, Gao and colleagues [24] translated the outcome observed in Stand Up Victoria into costs per life year (LY) gained and costs per health-adjusted life years (HALY) gained and reported higher LY and HALY gains and lower long-term health costs, suggesting that the intervention was cost-eﬀective over the lifetime of the cohort. Subsequently, Gao and colleagues [18], modelled an economic evaluation to simulate the long-term health beneﬁts of workplace interventions targeting a reduction in sitting time in preventing cardiovascular disease (CVD). They showed a resultant ICER of $43,825 per QALY gained, making the intervention eﬀective for the primary prevention of CVD. Despite this recent progress on the cost-beneﬁt, return on investment and cost-eﬀectiveness of workplace sitting reduction interventions, the data are still limited and more evidence is needed to continue to build a stronger business case for organisations to adopt a focus on sitting less in the workplace. 4. Discussion The strengths of this study include the randomised controlled trial, with short and medium term follow up time points, and the inclusion of organization absenteeism records. Presenteeism and work productivity however were assessed via self-report which may introduce bias with employees either over-estimating or under-estimating their productivity. Furthermore, productivity gains were calculated from a questionnaire assessing the past 7 days, completed four times across the 12-month period. These gains were then estimated over 52 weeks and therefore may not extrapolate exactly but the magnitude of net beneﬁt per employee would appear to leave some margin for variance in these estimates. The costs of the intervention are based on the actual facilitator’s and participants’ pay grades, therefore the costs in the intervention may ﬂuctuate depending on who facilitates the implementation and takes part. Author Contributions: C.L.E., S.J.H.B., M.J.D., D.W.D., D.W.E., L.J.G., T.Y., F.M. obtained funding for the research. All authors have contributed to the design of the study. S.E.O.C. was involved in data collection and study co-ordination throughout. C.L.E. and F.M. supervised S.E.O.C. P.M. and G.W. conducted the analysis and the ﬁrst draft the manuscript was produced by C.L.E., P.M. S.E.O.C. and F.M. All authors have read and agreed to the published version of the manuscript. Funding: This project was funded by the Department of Health Policy Research Programme (project number PR-R5-0213-25004). Acknowledgments: The research was supported by the National Institute for Health Research (NIHR) Leicester Biomedical Research Centre which is a partnership between University Hospitals of Leicester NHS Trust, Loughborough University and the University of Leicester, the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care – East Midlands (NIHR CLAHRC – EM) and the Leicester References 1. Wilmot, E.G.; Edwardson, C.L.; Achana, F.A.; Davies, M.J.; Gorely, T.; Gray, L.J.; Khunti, K.; Yates, T.; Biddle, S.J.H. Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis. Diabetologia 2012, 55, 2895–2905. [CrossRef] [PubMed] 2. De Rezende, L.F.M.; Rey-López, J.P.; Matsudo, V.K.R.; Luiz, O.D.C. Sedentary behavior and health outcomes among older adults: A systematic review. BMC Public Health. 2014, 14, 333. [CrossRef] [PubMed] 2. De Rezende, L.F.M.; Rey-López, J.P.; Matsudo, V.K.R.; Luiz, O.D.C. Sedentary behavior and health outcomes among older adults: A systematic review. BMC Public Health. 2014, 14, 333. [CrossRef] [PubMed] 3. Shen, D.; Mao, W.; Liu, T.; Lin, Q.; Lu, X.; Wang, Q.; Lin, F.; Ekelund, U.; Wijndaele, K. Sedentary Behavior and Incident Cancer: A Meta-Analysis of Prospective Studies. PLoS ONE 2014, 9, e105709. [CrossRef] [PubMed] 3. Shen, D.; Mao, W.; Liu, T.; Lin, Q.; Lu, X.; Wang, Q.; Lin, F.; Ekelund, U.; Wijndaele, K. Sedentary Behavior and Incident Cancer: A Meta-Analysis of Prospective Studies. PLoS ONE 2014, 9, e105709. [CrossRef] [PubMed] 4. Biswas, A.; Oh, P.I.; Faulkner, G.E.; Bajaj, R.R.; Silver, M.A.; Mitchell, M.S.; Alter, D.A. Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: A systematic review and meta-analysis. Ann. Intern. Med. 2015, 162, 123–132. [CrossRef] [PubMed] 4. Biswas, A.; Oh, P.I.; Faulkner, G.E.; Bajaj, R.R.; Silver, M.A.; Mitchell, M.S.; Alter, D.A. Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: A systematic review and meta-analysis. Ann. Intern. Med. 2015, 162, 123–132. [CrossRef] [PubMed] 5. Teychenne, M.; A Costigan, S.; Parker, K. The association between sedentary behaviour and risk of anxiety: A systematic review. BMC Public Health. 2015, 15, 513. [CrossRef] [PubMed] 5. Teychenne, M.; A Costigan, S.; Parker, K. The association between sedentary behaviour and risk of anxiety: A systematic review. BMC Public Health. 2015, 15, 513. [CrossRef] [PubMed] 6. Zhai, L.; Zhang, Y.; Zhang, D. Sedentary behaviour and the risk of depression: A meta-analysis. Br. J. Sports Med. 2015, 49, 705–709. [CrossRef] 7. Parry, S.; Straker, L. The contribution of oﬃce work to sedentary behaviour associated risk. BMC Public Health. 2013, 13, 296. [CrossRef] 8. Healy, G.N.; Eakin, E.G.; Lamontagne, A.D.; Owen, N.; Winkler, E.A.; Wiesner, G.; Gunning, L.; Neuhaus, M.; Lawler, S.; Fjeldsoe, B.S.; et al. Reducing sitting time in oﬃce workers: Short-term eﬃcacy of a multicomponent intervention. Prev. Med. 2013, 57, 43–48. Int. J. Environ. Res. Public Health 2020, 17, 1214 Int. J. Environ. Res. Public Health 2020, 17, 1214 Clinical Trials Unit. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. D.D is supported by a NHMRC Senior Research Fellowship (NHMRC 1078360) and the Victorian Government’s Operational Infrastructure Support Program. 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. Trial Registration: ISRCTN10967042. Registered on 2 February 2015. 5. Conclusions In summary, based on material costs, facilitator costs to implement the intervention and indirect work loss costs the SMArT Work intervention cost £692.40 per person. Despite its high cost, the intervention is estimated to provide a net saving per employee of £1770.32, over an initial 12-month time period, as a result of an increase in productivity. This equates to a £2.56 return on investment for every £1 spent on the intervention. Author Contributions: C.L.E., S.J.H.B., M.J.D., D.W.D., D.W.E., L.J.G., T.Y., F.M. obtained funding for the research. All authors have contributed to the design of the study. S.E.O.C. was involved in data collection and study co-ordination throughout. C.L.E. and F.M. supervised S.E.O.C. P.M. and G.W. conducted the analysis and the ﬁrst draft the manuscript was produced by C.L.E., P.M. S.E.O.C. and F.M. All authors have read and agreed to the published version of the manuscript. Funding: This project was funded by the Department of Health Policy Research Programme (project number PR-R5-0213-25004). Acknowledgments: The research was supported by the National Institute for Health Research (NIHR) Leicester Biomedical Research Centre which is a partnership between University Hospitals of Leicester NHS Trust, Loughborough University and the University of Leicester, the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care – East Midlands (NIHR CLAHRC – EM) and the Leicester 8 of 9 References [CrossRef] 9. Clemes, S.A.; O’connell, S.E.; Edwardson, C.L. Oﬃce Worker’s Objectively Measured Sedentary Behavior and Physical Activity During and Outside Working Hours. J. Occup. Environ. Med. 2014, 56, 298–303. [CrossRef] 10. Hallman, D.M.; Gupta, N.; Mathiassen, S.E.; Holtermann, A. Association between objectively measured sitting time and neck–shoulder pain among blue-collar workers. Int. Arch. Occup. Environ. Health 2015, 88, 1031–1042. [CrossRef] 11. Munir, F.; Houdmont, J.; Clemes, S.; Wilson, K.; Kerr, R.; Addley, K. Work engagement and its association with occupational sitting time: Results from the Stormont study. BMC Public Health. 2015, 15, 30. [CrossRef] [PubMed] 12. Brown, H.E.; Ryde, G.C.; Gilson, N.D.; Burton, N.W.; Brown, W.J. Objectively measured sedentary behavior and physical activity in oﬃce employees: Relationships with presenteeism. J. Occup. Environ. Med. 2013, 55, 945–953. [CrossRef] [PubMed] 13. Neuhaus, M.; Healy, G.N.; Fjeldsoe, B.S.; Lawler, S.; Owen, N.; Dunstan, D.W.; Lamontagne, A.D.; Eakin, E.G. Iterative development of Stand Up Australia: A multi-component intervention to reduce workplace sitting. Int. J. Behav. Nutr. Phys. Act. 2014, 11, 21. [CrossRef] [PubMed] 14. Healy, G.N.; Winkler, E.A.; Eakin, E.G.; Owen, N.; Lamontagne, A.D.; Moodie, M.; Dunstan, D.W. A cluster RCT to reduce workers’ sitting time: Impact on cardio-metabolic biomarkers. Med. Sci. Sports Exerc. 2017, 49, 2032–2039. [CrossRef] 15. Karakolis, T.; Callaghan, J.P. The impact of sit–stand oﬃce workstations on worker discomfort and productivity: A review. Appl. Ergon. 2014, 45, 799–806. [CrossRef] 16. Danquah, I.H.; Kloster, S.; Holtermann, A.; Aadahl, M.; Bauman, A.; Ersbøll, A.K.; Tolstrup, A.S. Take a Stand!-a multi-component intervention aimed at reducing sitting time among oﬃce workers-a cluster randomized trial. Int. J. Epidemiol. 2017, 46, 128–140. [CrossRef] 9 of 9 Int. J. Environ. Res. Public Health 2020, 17, 1214 17. Dunstan, D.W.; Wiesner, G.; Eakin, E.G.; Neuhaus, M.; Owen, N.; Lamontagne, A.D.; Moodie, M.; Winkler, E.A.; Fjeldsoe, B.S.; Lawler, S.; et al. Reducing oﬃce workers’ sitting time: Rationale and study design for the Stand Up Victoria cluster randomized trial. BMC Public Health. 2013, 13, 1057. [CrossRef] 18. Gao, L.; Nguyen, P.; Dunstan, D.; Moodie, M. Are Oﬃce-Based Workplace Interventions Designed to Reduce Sitting Time Cost-Eﬀective Primary Prevention Measures for Cardiovascular Disease? A Systematic Review and Modelled Economic Evaluation. Int. J. Environ. Res. Public Health 2019, 16, 834. [CrossRef] Sitting Time Cost Eﬀective Primary Prevention Measures for Cardiovascular Disease? A Systematic Review and Modelled Economic Evaluation. Int. J. Environ. Res. Public Health 2019, 16, 834. [CrossRef] 19. Aronsson, G.; Gustafsson, K. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References Sickness presenteeism: Prevalence, attendance-pressure factors, and an outline of a model for research. J. Occup. Environ. Med. 2005, 47, 958–966. [CrossRef] 20. ERS Research and Consultancy. Health at Work Economic Evidence Report. 2016. Available online: https://www.bhf.org.uk/informationsupport/publications/health-at-work/health-at-work- --economic-evidence-report (accessed on 18 November 2017). 21. Edwardson, C.L.; Yates, T.; Biddle, S.J.H.; Davies, M.J.; Dunstan, D.W.; Esliger, D.W.; Gray, L.J.; Jackson, B.; E O’Connell, S.; Waheed, G.; et al. Eﬀectiveness of the Stand More AT (SMArT) Work intervention: Cluster randomised controlled trial. BMJ 2018, 363, k3870. [CrossRef] 22. O’Connell, S.E.; Jackson, B.R.; Edwardson, C.L.; Yates, T.; Biddle, S.J.H.; Davies, M.J.; Dunstan, D.; Esliger, D.; Gray, L.; Miller, P.; et al. Providing NHS staﬀwith height-adjustable workstationsand behaviour change strategies to reduce workplace sitting time: Protocol for the Stand More AT (SMArT) Work cluster randomised controlled trial. BMC Public Health 2015, 15, 1219. [CrossRef] [PubMed] 23. Reilly, M.C.; Zbrozek, A.S.; Dukes, E.M. The Validity and Reproducibility of a Work Productivity and Activity Impairment Instrument. Pharmacoeconomics 1993, 4, 353–365. [CrossRef] [PubMed] 24. Gao, L.; Flego, A.; Dunstan, D.W.; Winkler, E.A.; Healy, G.N.; Eakin, E.G.; Willenberg, L.; Owen, N.; Lamontagne, A.D.; Lal, A.; et al. Economic evaluation of a randomized controlled trial of an intervention to reduce oﬃce workers’ sitting time: The "Stand Up Victoria" trial. Scand. J. Work. Environ. Health. 2018, 44, 503–511. [CrossRef] [PubMed] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4307502945	https://publikationen.sulb.uni-saarland.de/bitstream/20.500.11880/34938/1/s00417-022-05881-6.pdf	English	null	Reliability analysis of successive Corvis ST® measurements in keratoconus 2 years after accelerated corneal crosslinking compared to untreated keratoconus corneas	Graefe's archive for clinical and experimental ophthalmology	2,022	cc-by	5,119	Extended author information available on the last page of the article Kassandra Xanthopoulou1 · Berthold Seitz1 · Michael W. Belin2 · Elias Flockerzi1 Kassandra Xanthopoulou1 · Berthold Seitz1 · Michael W. Belin2 · Elias Flockerzi1 Received: 21 March 2022 / Revised: 12 October 2022 / Accepted: 20 October 2022 © The Author(s) 2022 CORNEA CORNEA Graefe's Archive for Clinical and Experimental Ophthalmology https://doi.org/10.1007/s00417-022-05881-6 Graefe's Archive for Clinical and Experimental Ophthalmology https://doi.org/10.1007/s00417-022-05881-6 Abstract Purpose To assess the reliability of successive Corvis ST® measurements (CST, Oculus, Wetzlar, Germany) in keratoconus (KC) ≥ 2 years after accelerated corneal crosslinking (9 mW/cm2, 10 min, 5.4 J/cm2) compared to untreated KC corneas. Methods Three successive CST measurements per eye were performed in ≥ 2 years after CXL (CXLG, n = 20 corneas of 16 patients) and a control group consisting of non-operated, ABC-stage-matched KC corneas according to Belin’s ABCD KC grading (controls, n = 20 corneas, 20 patients). Main outcome measures included maximal keratometry (Kmax), the Belin/ Ambrósio-Enhanced-Ectasia-Deviation-Index BAD-D; the biomechanical parameters A1 velocity, deformation amplitude (DA) ratio 2 mm, Ambrósio relational thickness to the horizontal profile (ARTh), integrated radius, stiffness parameter A1 (SP-A1), and the Corvis Biomechanical Factor (CBiF, the linearized term of the Corvis Biomechanical Index). Mean values, standard deviations, and Cronbach’s alpha (CA) were calculated. Results Both groups were tomographically comparable (BAD: 11.5 ± 4.7\|11.2 ± 3.6, p = 0.682, Kmax: 60.5 ± 7.2\|60.7 ± 7.7, p = 0.868 for controls\|CXLG, paired t-test). A1 velocity (mean ± SD: 0.176 ± 0.02\|0.183 ± 0.02, p = 0.090, CA: 0.960\|0.960), DA ratio 2 mm (6.04 ± 1.13\|6.14 ± 1.03, p = 0.490, CA: 0.967\|0.967), integrated radius (12.08 ± 2.5\|12.42 ± 1.9, p = 0.450, CA: 0.976\|0.976), and CBiF (4.62 ± 0.6\|4.62 ± 0.4, p = 0.830, CA: 0.965\|0.965) were also comparable (controls\|CXLG). ARTh was significantly higher in controls (177.1 ± 59, CA: 0.993) than after CXL (155.21 ± 65, p = 0.0062, CA: 0.993) and SP-A1 was significantly higher after CXL (59.2 ± 13, CA: 0.912) than in controls (52.2 ± 16, p = 0.0018, CA: 0.912). Conclusion ARTh and SP-A1 differed significantly between controls and CXLG. Biomechanical measurements were gener- ally of excellent reliability in both groups. CXL seems to affect biomechanical measurements of human corneas over more than 2 years. Introduction This retrospective cross-sectional cohort study was per- formed at the Department of Ophthalmology at Saarland University Medical Center in Homburg, Germany. The patient charts of KC patients ≥ 18 years were enrolled from the Homburg Keratoconus Center (HKC) which was estab- lished in 2010 [15]. All patients in the current study signed an informed consent for the use of their data for analysis and therefore participated in the HKC observational clinical study, which was approved by the local ethics committee (Ethikkommission bei der Ärztekammer des Saarlandes, reference number 121/20, trial number NCT03923101, US National Institutes of Health, ClinicalTrials.gov) and respects the principles of the Declaration of Helsinki. Keratoconus (KC) is an ectatic corneal disease that can be diagnosed based on the abnormal corneal biomechanical response to mechanical stress. Several studies investigated corneal deformation and found KC to show higher deforma- tion amplitudes compared to healthy corneas [1, 2].l The Corneal Visualization Scheimpflug Technology Cor- vis ST® (CST, Oculus, Wetzlar, Germany) is a non-contact pneumotonometer that analyzes the corneal deformation response to a standardized air puff using an ultra-high- speed Scheimpflug camera that captures over 4300 frames per second [3, 4]. The Corvis Biomechanical Index (CBI) was designed to differentiate between ectatic and normal corneas [5]. It is reported to consist of a combination of dynamic corneal response (DCR) parameters: (1) the speed of the corneal apex at inward applanation A1 (A1 velocity), (2) the maximum value of the ratio between the deforma- tion amplitude at the apex and 2 mm from the center (DA Ratio 2 mm), (3) Ambrósio’s relational thickness through the horizontal profile (ARTh), (4) the radius of curvature during the concave phase of the deformation (integrated radius), and (5) the stiffness parameter at inward applanation A1 (SP-A1) [1, 5]. The Corvis Biomechanical Factor (CBiF) is the linearized term of the CBI [6] and a biomechanical E-staging for KC and ectatic corneal diseases with stages E0 to E4 was developed by dividing the CBiF value range into five groups to augment the existing tomographic ABCD KC grading system [7]. The ABCD KC grading system itself was introduced in 2016 and is based on anterior (“A”) and posterior (“B”) radius of curvature measured over a 3.0 mm zone centered on the thinnest corneal pachymetry (“C”) and it also includes best-spectacle-corrected visual acuity (“D”) [8]. Introduction The crosslinking group (CXLG) consisted of 20 KC cor- neas of 16 KC patients who underwent epithelium-off accel- erated CXL (9 mW/cm2, 10 min, 5.4 J/cm2) two or more years ago with the Avedro KXL® system (Avedro, Waltham, MA, USA) using riboflavin VibeX Rapid™ 0.1% solution (Avedro, Waltham, MA, USA). Diagnosis of KC was based on clinical features such as corneal hemosiderin deposition known as a Fleischer ring, Vogt striae, corneal thinning on the slit lamp examination, and/or tomographic abnormali- ties as detected within the Belin/Ambrósio-enhanced ectasia screening display within the Pentacam® HR software (high resolution (HR), Oculus, Wetzlar, Germany) [16, 17]. KC progression was defined as an increase in corneal astigma- tism ≥ 1 diopter (D), and/or an increase of maximal kerato- metry (Kmax) ≥ 1 D and/or a decrease of corneal thickness of 30 µm within 1 year. These patients underwent 3 suc- cessive CST measurements at 2 or more years after CXL (crosslinking group, CXLG) during one regular follow-up examination. A control group for biomechanical measurements was formed consisting of 20 non-operated KC corneas of 20 KC patients (controls) that underwent the same examina- tions. Those corneas were chosen as ABC-stage-matched controls according to Belin’s tomographic ABCD KC grad- ing. Both groups had to stop wearing contact lenses at least 3 days prior to the measurements. First, the ABC param- eters were collected for the CXLG from the “topometric/KC staging” display of the Pentacam software. Second, ABC- stage-matched controls were enrolled from the HKC. Con- sequently, each cornea treated with CXL was paired with an untreated cornea of the same ABC stage as a control. Analysis of corneal biomechanics has shown that corneal cross-linking (CXL) leads to an increase in corneal stiffness and a reduced maximal deformation [9, 10]. The ultraviolet (UV) corneal crosslinking (CXL) with riboflavin instilla- tion results in an increased corneal rigidity and stiffening in a minimally invasive way, thus reducing the steepness and halting progression [11, 12]. Accelerated epithelium- off (epi-off) CXL (9 mW/cm2, 10 min, fluence 5.4 J/cm2) has been proven to have the same long-term efficacy as the standard CXL procedure (3 mW/cm2, 30 min, fluence 5.4 J/ cm2) [13, 14]. This study aimed to assess the reliability of biomechani- cal corneal analysis based on three successive CST measure- ments in two groups of KC patients. Key messages The reliability of successive Corvis ST® measurements in keratoconus (KC) after accelerated crosslinking (CXL) has not been evaluated yet. In the current study the A1 velocity, integrated radius and Corvis Biomechanical Factor (CBiF) were comparable for the KC controls and for the KC group 2 years after CXL. The Ambrósio relational thickness to the horizontal profile (ARTh) was significantly higher in the KC controls in comparison to the group after CXL, while the stiffness-parameter A1 (SP-A1) was significantly higher after CXL. Biomechanical measurements were generally of excellent reliability in both groups. CXL seems to affect biomechanical measurements of human corneas over more than two yea Keywords Corneal biomechanics · Corvis · Keratoconus · Corneal crosslinking · Biomechanical E-staging :(0123 1 23456789) 3 Graefe's Archive for Clinical and Experimental Ophthalmology Introduction Group 1 consisted of patients who underwent epithelium-off accelerated corneal CXL with a minimum 2-year follow-up (crosslinking group, CXLG). Group 2 consisted of non-operated, ABC-stage- matched KC patients serving as controls (controls). The ABC stage was derived from Pentacam measure- ments, which always proceeded the CST measurements to avoid tomographical changes caused by the CST air puff indentation. Both the Pentacam and CST measurements were only included with an “OK” score (in advanced stages, “model deviations” were also accepted) and the 3 3 Graefe's Archive for Clinical and Experimental Ophthalmology measurements were independently reviewed by two physi- cians (KX, EF). Table 1 Comparison of the control group and the crosslinking group (CXLG). ABC-stage-matched non-operated KC controls were paired with KC corneas ≥ 2 years after accelerated corneal crosslinking (9 mW/cm2, 10 min, 5.4 J/cm2). Kmax (diopters), maximal keratometry; ARC (mm), anterior radius of curvature; PRC (mm), posterior radius of curvature; TCT (µm), thinnest corneal thickness. BAD-D, Belin/ Ambrósio-Enhanced-Ectasia-Deviation-Index. P-values calculated by paired t-testT, if normally distributed The maximal keratometry (Kmax) and the Belin/Ambró- sio-Enhanced-Ectasia-Deviation-Index BAD-D were ana- lyzed to determine tomographic KC severity in addition to the ABC severity stage. The main outcome DCR param- eters included A1 velocity, deformation amplitude (DA) ratio 2 mm, Ambrósio relational thickness to the horizontal profile (ARTh), integrated radius, stiffness parameter A1 (SP-A1), and the Corvis Biomechanical Factor (CBiF, the linearized term of the Corvis Biomechanical Index, CBI). The outcome measures were first analyzed for normal distri- bution using the Shapiro–Wilk test and assuming a normal distribution with p ≥ 0.05. The parameters resulting from three measurements per eye per patient were subsequently compared between the control group and the CXLG using the two-tailed paired t-test (if normally distributed) or the Wilcoxon matched-pairs test (if not normally distributed) assuming significant differences with p < 0.05. The paired tests were used to obtain the most accurate comparison between the respective crosslinked and stage-matched non- treated control corneas.fi Controls CXLG P Gender 13 male, 7 female 13 male, 3 female Age 39 ± 14 31 ± 11 0.065 T Kmax 60.5 ± 7 60.7 ± 8 0.868 T ARC 6.2 ± 0.6 6.3 ± 0.6 0.344 T PRC 4.6 ± 0.6 4.5 ± 0.4 0.605 T TCT 456 ± 34 452 ± 27 0.401 T BAD-D 11.5 ± 5 11.3 ± 4 0.682 T comparable between controls and the CXLG. Introduction Mean ARTh was significantly higher in controls (177.1 ± 59) than after CXL (155.21 ± 65, p = 0.0062) and mean SP-A1 was sig- nificantly higher after CXL (59.2 ± 13) than in controls (52.2 ± 16, p = 0.0018, Table 2). Bland–Altman plots were created for the two parameters that differed signifi- cantly between controls and the CXLG (ARTh and SP-A1, Fig. 1) showing the mean difference and the 95% limits of agreement. The coefficients of repeatability were calculated as the within-subject standard deviation Sw × √2 × 1.96. The intra- class correlation coefficients (ICC) that correlate successive measurements carried out on the same subject or a collective of patients with the same underlying disease and Cronbach’s alpha (CA) were calculated to determine the reliability of the biomechanical measurements. The coefficients of repeatability were lower in controls than the CXLG for A1 velocity, SP-A1, and the CBiF and lower in the CXLG than in controls for DA ratio 2 mm, inte- grated radius, and ARTh (Table 2).fi Results The intraclass correlation coefficients and Cronbach’s alpha values were identical in both the CXLG and controls and indicated an excellent reliability of the biomechanical measurements (CA ≥ 0.912, Table 2). The CXLG consisted of 12 right eyes and 8 left eyes. Accel- erated CXL was performed on average 48 ± 19 months ago in these corneas. The mean age of the patients in the CXLG was 31 ± 11 years. Eleven right eyes and 9 left eyes were chosen ABC-stage-matched (Table 1) as controls and the mean age of the control patients was 39 ± 14 years. The age of the patients was normally distributed (controls: p = 0.805 and CXLG: p = 0.111; Shapiro–Wilk test) and did not dif- fer significantly between the control group and the CXLG (p = 0.065, paired t-test). Discussion Mean ± SD, standard deviation resulting out of three measurements per eye; ICC, intra- class correlation coefficient; CA, Cronbach’s alpha; DA ratio 2 mm, deformation amplitude (DA) ratio 2 mm; ARTh, Ambrósio relational thickness to the horizontal profile; SP-A1, stiffness parameter A1; CBiF, Corvis Biomechanical Factor (the linearized term of the Cor- vis Biomechanical Index, CBI). Comparable values for mean ± SD in Parameter Controls Mean ± SD CXLG Mean ± SD P Controls Coefficient of repeatability CXLG Coefficient of repeatability Controls and CXLG ICC Controls and CXLG CA A1 velocity 0.176 ± 0.02 0.183 ± 0.02 0.090 T 0.01 0.02 0.956 0.960 DA ratio 2 mm 6.04 ± 1.13 6.14 ± 1.03 0.490 W 0.97 0.82 0.952 0.967 Integrated radius 12.08 ± 2.5 12.42 ± 1.90 0.450 W 2.07 1.28 0.969 0.976 ARTh 177.1 ± 59 155.21 ± 65 0.0062 W 33.12 25.00 0.991 0.993 SP-A1 52.2 ± 16 59.2 ± 13 0.0018 T 12.31 16.46 0.894 0.912 CBiF 4.62 ± 0.6 4.62 ± 0.4 0.830 W 0.27 0.39 0.955 0.965 CXL changes the viscoelastic corneal properties and leads to corneal stiffening [18]. In contrast, an in vivo study found no significant differences between pre- and 4 years post- operative biomechanical parameters obtained with the CST system after CXL with exception of the integrated radius [19]. The current study found no significant differences between controls and the CXLG at 2 years postoperatively for the majority of the biomechanical CST parameters that were analyzed: A1 velocity, the speed of the corneal apex at inward applanation, the ratio between the deformation amplitude at the apex and 2 mm away from the center (DA ratio 2 mm), the radius of curvature during the concave phase of the deformation (integrated radius), and CBiF, which is the recently introduced linearized term of the CBI and serves as a basis for the Homburg biomechanical E-staging [7]. All the patients of this study presented during regular follow-up examinations and their tomographic values remained stable after CXL without requiring repeated CXL in the CXLG. Consequently, comparable results of the CBiF between controls and CXLG also indicate a stabilization of the corneas in the CXLG at the level of non-progressive KC corneas within the control group. study found significantly lower ARTh values in the CXLG (Table 2), although the controls were selected ABC-stage- matched and thus were comparable with respect to the thin- nest corneal thickness. Discussion This study analyzed the reliability of biomechanical CST measurements in KC corneas ≥ 2 years after accelerated corneal cross-linking compared to untreated KC controls of the same ABC stage. Besides other studies, the recent introduction of a biomechanical E-staging for ectatic corneal diseases based on the CBiF [7, 15] raised the question of whether CXL has a detectable long-term effect on corneal biomechanics and on the reliability of biomechanical meas- urements of the human cornea. Both the controls and CXLG group were tomographically comparable: mean Kmax amounted to 60.5 ± 7.2\|60.7 ± 7.7 (controls\|CXLG, p = 0.868) and mean BAD-D to 11.5 ± 4.7\|11.2 ± 3.6 (controls\|CXLG, p = 0.682, paired t-test). The anterior and posterior radii of curvature (ARC, PRC, Table 1) and thinnest corneal thickness (TCT, Table 1) were also without statistical differences and, thus, compara- ble in both groups. The CST assesses the biomechanical properties of the cornea [1] and its measurements depend on corneal rigid- ity. An in vitro study reported that corneal indentation and inward versus outward deformation after air puff indentation were reduced significantly after CXL, which indicates that The mean values of three measurements for A1 veloc- ity, DA ratio 2 mm, integrated radius, and CBiF were also 1 1 3 3 Graefe's Archive for Clinical and Experimental Ophthalmology controls and CXLG except for ARTh and SP-A1, p-values calculated by (1) paired two-tailed t-testT, if normally distributed or by (2) Wil- coxon matched-pairs testW if not normally distributed—as determined by Shapiro–Wilk test. Coefficients of repeatability for each param- eter in controls and CXLG calculated as the within-subject standard deviation Sw × √2 × 1.96. Identical intraclass correlation coefficients and Cronbach’s alpha values in the CXLG and in untreated controls. CA ≥ 0.912 indicating excellent reliability Table 2 Main outcome measures in the control group (controls) and ≥ 2 years after crosslinking group (CXLG). Mean ± SD, standard deviation resulting out of three measurements per eye; ICC, intra- class correlation coefficient; CA, Cronbach’s alpha; DA ratio 2 mm, deformation amplitude (DA) ratio 2 mm; ARTh, Ambrósio relational thickness to the horizontal profile; SP-A1, stiffness parameter A1; CBiF, Corvis Biomechanical Factor (the linearized term of the Cor- vis Biomechanical Index, CBI). Comparable values for mean ± SD in Table 2 Main outcome measures in the control group (controls) and ≥ 2 years after crosslinking group (CXLG). Discussion It is known that the thinnest corneal thickness may decrease after CXL [20] which might result in lower ARTh values postoperatively. The lower ARTh value in our CXLG could be due to a coinciding effect of a post- operative decrease in corneal thickness and flattening of the corneal apex as a CXL result [20]. SP-A1 is a parameter that has been reported to increase markedly after CXL as a result of the increasing corneal rigidity [21]. Comparing controls and CXLG, the cur- rent study established the largest difference between both groups for SP-A1 with significantly higher values after CXL (Table 2). This could be interpreted as a surrogate meas- ure for CXL efficiency and this indicates that despite (1) the small sample size of this study and (2) a tomographic ABC-matching, biomechanical differences remain measur- able even more than 2 years (on average 48 ± 19 months) after CXL.i Although not reaching statistical significance, the result of postoperatively higher SP-A1 values in the long term has also been found by Sedaghat et al. [19] in a long-term fol- low-up of 18 eyes of 18 KC patients at 4 years after standard CXL according to the Dresden protocol. Interestingly—and in contrast to our results—they found slightly, yet not sig- nificantly higher ARTh values 4 years after CXL than pre- operatively, which would indicate a centrally thicker cornea with a lower thickness increase towards the periphery. It has still to be determined (1) to what extent and (2) how long postoperatively thickness-related tomographic and bio- mechanical Scheimpflug-measured parameters are prone to measurement artifacts, and, thus, may lead to seemingly contradictory results. Strikingly, there were highly significant differences between controls and CXLG for the biomechanical parame- ters (1) Ambrósio’s relational thickness through the horizon- tal profile (ARTh) and (2) the stiffness parameter at inward apö8planation A1 (SP-A1). In contrast, the aforementioned study found no significant differences when comparing those parameters pre- and postoperatively four years after CXL [19]. ARTh results out of the division of the thinnest cor- neal thickness through the pachymetric progression index and lower values indicate a centrally thinner cornea with a fast thickness increase towards the periphery [5]. This 3 3 Graefe's Archive for Clinical and Experimental Ophthalmology 1 Several studies confirmed a good to excellent reliability of biomechanical CST measurements in normal and KC eyes [22–24]. Yang et al. Declarations Ethics approval Ethical Committee that approved the study: local eth- ics committee (Ethikkommission bei der Ärztekammer des Saarlandes, reference number 121/20, trial number NCT03923101, U.S. National Institutes of Health, ClinicalTrials.gov) with respect to the principles of the Declaration of Helsinki. Conflict of interest M. W. Belin is a consultant for Oculus GmbH (Wetzlar, Germany). E. Flockerzi has received a travel grant to the Second and Third Ophthalmology Cystinosis Forum (Orphan Europe, Ulm, Germany) and an invitation to a seminar on presentation training organized by the Santen GmbH (Munich, Germany). The remaining authors have no conflicts of interest to disclose. Limitations of this study are the small sample size with the majority being moderate to advanced KC stages (Table 1) and the choice of a control group. Ideally, patients would have been followed up with three succes- sive measurements prior to CXL and with 3 successive measurements more than 2 years after CXL. Since these measurements were not available, an ABC-stage-matched control group was created and these controls were tomo- graphically comparable to the CXLG (Table 1) at the time of comparison—which does not take into account differences in duration of KC disease. The primary aim of this study was to analyze the reliability of biomechan- ical CST measurements in KC corneas ≥ 2 years after accelerated CXL compared to untreated KC corneas— which is why the CXLG consisted of initially progres- sive KC corneas and the control group of stable KC cor- neas. The KC progression within the CXLG should have been halted by the CXL effect which should, in turn, mitigate the aspect of different progression rates in both groups. Another limitation is the inclusion of two eyes per patient in eight cases in the CXLG. This could lead to bias, as two eyes of one patient are not considered to be independent. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. Discussion [24] found high-reliability values for A1 velocity and DA ratio 2 mm after three sequen- tial measurements in 77 healthy corneas and 77 mild to moderate KC eyes. A recent study evaluated the reliability of CST parameters in untreated KC based on 5 successive CST measurements and found good to excellent reliabil- ity independent of the KC stage [25]. The current study examined the reliability of the biomechanical parameter Fig. 1 Bland–Altman plots showing the mean difference (solid black line) between controls and CXLG for ARTh and SP-A1 and the 95% limits of agreement (dotted lines) Fig. 1 Bland–Altman plots showing the mean difference (solid black line) between controls and CXLG for ARTh and SP-A1 and the 95% limits of agreement (dotted lines) Several studies confirmed a good to excellent reliability of biomechanical CST measurements in normal and KC eyes [22–24]. Yang et al. [24] found high-reliability values for A1 velocity and DA ratio 2 mm after three sequen- tial measurements in 77 healthy corneas and 77 mild to Several studies confirmed a good to excellent reliability of biomechanical CST measurements in normal and KC eyes [22–24]. Yang et al. [24] found high-reliability values for A1 velocity and DA ratio 2 mm after three sequen- tial measurements in 77 healthy corneas and 77 mild to moderate KC eyes. A recent study evaluated the reliability of CST parameters in untreated KC based on 5 successive CST measurements and found good to excellent reliabil- ity independent of the KC stage [25]. The current study examined the reliability of the biomechanical parameter 1 3 1 3 Graefe's Archive for Clinical and Experimental Ophthalmology Author contribution All authors contributed to this work by providing their data, revising the manuscript, approving it for publication, and agreeing to take full responsibility for all aspects of this work. Author contribution All authors contributed to this work by providing their data, revising the manuscript, approving it for publication, and agreeing to take full responsibility for all aspects of this work. measurements based on three successive CST measure- ments and (1) the calculation of the intraclass correla- tion coefficients and (2) the Cronbach’s alpha. The ICC (≥ 0.894) and CA values were identical in both the con- trols and the CXLG. Declarations 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/. Discussion The CA values ranged from 0.912 (SP-A1) to 0.993 (ARTh) which indicates an excellent reli- ability [26] of the biomechanical measurements not only in the stabilized CXLG, but also in the untreated controls. Although it was not developed to measure the effect of CXL, the excellent reliability of the CBiF (CA: 0.965 in both controls and CXLG) indicates that this parameter can also be used after CXL to assess KC severity, and therefore biomechanical stability. Funding Open Access funding enabled and organized by Projekt DEAL. References 1. Ambrósio R Jr, Ramos I, Luz A et al (2013) Dynamic ultra high speed Scheimpflug imaging for assessing corneal biomechanical properties. Rev Bras Oftalmol 72:99–102. https://doi.org/10.1590/ S0034-72802013000200005 p In summary, we found an excellent reliability of the biomechanical measurements in KC corneas ≥ 2 years after CXL and in untreated KC corneas of the same ABC stage. Significant differences between both groups were found for (1) ARTh (controls > CXLG, p = 0.0062, Table 2) and (2) SP-A1 (CXLG > controls, p = 0.0018, Table 2). Together with a CXL-induced stiffening effect, this may be attribut- able to a postoperative decrease in corneal thickness and flattening of the corneal apex. This study thus indicates that the biomechanical effects of CXL remain measurable far beyond 2 years after surgery (48 ± 19 months on aver- age). Larger scale studies are required to define when the biomechanical stabilization after CXL begins and how long it lasts postoperatively. 2. Ye C, Yu M, Lai G, Jhanji V (2015) Variability of corneal defor- mation response in normal and keratoconic eyes. Optom Vis Sci 92:e149-153. https://doi.org/10.1097/OPX.0000000000000628 2. Ye C, Yu M, Lai G, Jhanji V (2015) Variability of corneal defor- mation response in normal and keratoconic eyes. Optom Vis Sci 92:e149-153. https://doi.org/10.1097/OPX.0000000000000628 p g 3. Ambrósio R, Correia FF, Lopes B et al (2017) Corneal biome- chanics in ectatic diseases: refractive surgery implications. Open Ophthalmol J 11:176–193. https://doi.org/10.2174/1874364101 711010176 3. Ambrósio R, Correia FF, Lopes B et al (2017) Corneal biome- chanics in ectatic diseases: refractive surgery implications. Open Ophthalmol J 11:176–193. https://doi.org/10.2174/1874364101 711010176 4. Piñero DP, Alcón N (2014) In vivo characterization of corneal biomechanics. J Cataract Refract Surg 40:870–887. https://doi. org/10.1016/j.jcrs.2014.03.021 g j j 5. Vinciguerra R, Ambrósio R, Elsheikh A et al (2016) Detection of keratoconus with a new biomechanical index. J Refract Surg 32:803–810. https://doi.org/10.3928/1081597X-20160629-01 p g 6. Flockerzi E, Vinciguerra R, Belin MW et al (2022) Correlation of the Corvis Biomechanical Factor CBiF with tomographic param- eters in keratoconus. J Cataract Refract Surg 48(2):215–221. https://doi.org/10.1097/j.jcrs.0000000000000740 6. Flockerzi E, Vinciguerra R, Belin MW et al (2022) Correlation of the Corvis Biomechanical Factor CBiF with tomographic param- eters in keratoconus. J Cataract Refract Surg 48(2):215–221. https://doi.org/10.1097/j.jcrs.0000000000000740 7. Flockerzi E, Vinciguerra R, Belin MW et al (2021) Combined bio- mechanical and tomographic keratoconus staging: adding a bio- mechanical parameter to the ABCD keratoconus staging system. Acknowledgements We would like to thank Mrs. Acknowledgements We would like to thank Mrs. Christina Turner for her linguistic corrections in this manuscript. References Christina Turner for her linguistic corrections in this manuscript. 1 3 3 Graefe's Archive for Clinical and Experimental Ophthalmology Acta Ophthalmol 100(5):e1135–e1142. https://doi.org/10.1111/ aos.15044 18. Kling S, Marcos S (2013) Contributing factors to corneal defor- mation in air puff measurements. Invest Ophthalmol Vis Sci 54:5078–5085. https://doi.org/10.1167/iovs.13-12509 8. Belin MW, Duncan JK (2016) Keratoconus: the ABCD Grading System. Klin Monatsbl Augenheilkd 233:701–707. https://doi.org/ 10.1055/s-0042-100626 p g 19. Sedaghat M-R, Momeni-Moghaddam H, Ambrósio R et al (2018) Long-term evaluation of corneal biomechanical properties after corneal cross-linking for keratoconus: a 4-year longitudinal study. J Refract Surg 34:849–856. https://doi.org/10.3928/1081597X-20181012-02 9. Fuchsluger TA, Brettl S, Geerling G et al (2019) Biomechani- cal assessment of healthy and keratoconic corneas (with/with- out crosslinking) using dynamic ultrahigh-speed Scheimpflug technology and the relevance of the parameter (A1L–A2L). Br J Ophthalmol 103:558–564. https://doi.org/10.1136/bjophthalm ol-2017-311627 20. Vounotrypidis E, Athanasiou A, Kortüm K et al (2018) Long-term database analysis of conventional and accelerated crosslinked keratoconic mid-European eyes. Graefes Arch Clin Exp Ophthal- mol 256:1165–1172. https://doi.org/10.1007/s00417-018-3955-3 10. Jabbarvand M, Moravvej Z, Shahraki K et al (2021) Corneal biomechanical outcome of collagen cross-linking in keratoconic patients evaluated by Corvis ST. Eur J Ophthalmol 31(4):1577– 1583. https://doi.org/10.1177/1120672120944798fi 21. Pedrotti, Caldarella, Fasolo et al (2019) Topographic and bio- mechanical changes after application of corneal cross-linking in recurrent keratoconus. Int J Environ Res Public Health 16:3872. https://doi.org/10.3390/ijerph16203872 p g j p 22. Ali NQ, Patel DV, McGhee CNJ (2014) Biomechanical responses of healthy and keratoconic corneas measured using a noncon- tact scheimpflug-based tonometer. Invest Ophthalmol Vis Sci 55:3651–3659. https://doi.org/10.1167/iovs.13-13715 p g 11. Wernli J, Schumacher S, Spoerl E, Mrochen M (2013) The effi- cacy of corneal cross-linking shows a sudden decrease with very high intensity UV light and short treatment time. Invest Opthalmol Vis Sci 54:1176. https://doi.org/10.1167/iovs.12-11409l p g 12. Wollensak G, Spoerl E, Seiler T (2003) Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthal- mol 135:620–627. https://doi.org/10.1016/s0002-9394(02)02220-1 23. Lopes BT, Roberts CJ, Elsheikh A et al (2017) Repeatability and reproducibility of intraocular pressure and dynamic corneal response parameters assessed by the Corvis ST. J Ophthalmol 2017:8515742. https://doi.org/10.1155/2017/8515742 g 13. Nicula CA, Nicula D, Rednik AM, Bulboacă AE (2020) Com- parative results of “epi-off” conventional versus “epi-off” accel- erated cross-linking procedure at 5-year follow-up. J Ophthalmol 2020:1–13. https://doi.org/10.1155/2020/4745101 24. Yang K, Xu L, Fan Q et al (2019) Repeatability and comparison of new Corvis ST parameters in normal and keratoconus eyes. Sci Rep 9:15379. References https://doi.org/10.1038/s41598-019-51502-4 p g 14. Shetty R, Pahuja NK, Nuijts RMMA et al (2015) Current pro- tocols of corneal collagen cross-linking: visual, refractive, and tomographic outcomes. Am J Ophthalmol 160:243–249. https:// doi.org/10.1016/j.ajo.2015.05.019 25. Flockerzi E, Häfner L, Xanthopoulou K et al (2021) Reliability analysis of successive Corneal Visualization Scheimpflug Tech- nology measurements in different keratoconus stages. Acta Oph- thalmol 100(1):e83–e90. https://doi.org/10.1111/aos.14857 p g 26. Koo TK, Li MY (2016) A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 15:155–163. https://doi.org/10.1016/j.jcm.2016.02.012 15. Flockerzi E, Xanthopoulou K, Goebels SC et al (2020) Kera- toconus staging by decades: a baseline ABCD classification of 1000 patients in the Homburg Keratoconus Center. Br J Oph- thalmol 105(8):1069–1075. https://doi.org/10.1136/bjophthalm ol-2020-316789 Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 16. Belin MW (2020) Keratoconus and ectatic disease: evolving cri- teria for diagnosis. Klin Monatsbl Augenheilkd 237:740–744. https://doi.org/10.1055/a-1077-8105 17. Duncan J, Gomes JA (2015) A new tomographic method of staging/ classifying keratoconus: the ABCD Grading System. Int J Kerat Ect Cor Dis 4:85–93. https://doi.org/10.5005/jp-journals-10025-1105 1 Department of Ophthalmology, Saarland University Medical Center, Homburg, Germany 2 Department of Ophthalmology & Vision Science, University of Arizona, Tucson, AZ, USA Authors and Affiliations Kassandra Xanthopoulou1 · Berthold Seitz1 · Michael W. Belin2 · Elias Flockerzi1 Kassandra Xanthopoulou1 · Berthold Seitz1 · Michael W. Belin2 · Elias Flockerzi1 * Kassandra Xanthopoulou kassixanthop@gmail.com 1 3 1 3 1 3 3
https://openalex.org/W2902066176	http://old.scielo.br/pdf/rgo/v66n3/1981-8637-rgo-66-03-00199.pdf	English	null	Association between Periodontitis and Hyperglycemia	RGO - Revista Gaúcha de Odontologia	2,018	cc-by	4,265	http://dx.doi.org/10.1590/1981-863720180003000013217 http://dx.doi.org/10.1590/1981-863720180003000013217 Conclusion i Conclusion It is suggested, therefore, that the request of laboratory tests to check glucose levels becomes a part of the dental treatment protocol of individuals with periodontitis. Indexing terms: Complications. Diabetes Mellitus. Periodontitis. Results Results In the sample studied a higher percentage of individuals with hyperglycemia among those with periodontitis than those without periodontitis was identified. l Results In the sample studied a higher percentage of individuals with hyperglycemia among those with periodontitis than those without periodontitis was identified. Métodos Métodos A amostra foi composta por indivíduos selecionados aleatoriamente que se inscreveram para tratamento no Curso de Odontologia do Centro Universitário Newton Paiva, Belo Horizonte, Minas Gerais. A amostra foi dividida em dois grupos, o dos indivíduos sem periodontite (n=30) e o dos indivíduos com periodontite (n=20). Todos responderam a um questionário sobre hábitos e condição de saúde geral. Também tiveram o fluxo salivar quantificado e passaram por um exame periodontal para avaliação do sangramento gengival, da profundidade de sondagem e do nível de inserção clínica. Através de exames laboratoriais, quantificou-se a glicemia em jejum e os níveis de hemoglobina glicada. Associação entre Periodontite e Hiperglicemia Associação entre Periodontite e Hiperglicemia Marília Terezinha Gonçalves OLIVEIRA1 ORCID iD 0000-0001-9787-852X Paulo Guilherme Santos FURTADO1 ORCID iD 0000-0002-9461-3953 Rafaella Ferreira Cordeiro e CARDOSO1 ORCID iD 0000-0001-6950-5179 Ana Carolina Dupim SOUZA1 ORCID iD 0000-0002-5798-0711 Rafael Paschoal Esteves LIMA2 ORCID iD 0000-0003-4343-3845 Marília Terezinha Gonçalves OLIVEIRA1 ORCID iD 0000-0001-9787-852X Paulo Guilherme Santos FURTADO1 ORCID iD 0000-0002-9461-3953 Rafaella Ferreira Cordeiro e CARDOSO1 ORCID iD 0000-0001-6950-5179 Ana Carolina Dupim SOUZA1 ORCID iD 0000-0002-5798-0711 Rafael Paschoal Esteves LIMA2 ORCID iD 0000-0003-4343-3845 Santuza Maria Souza de MENDONÇA1 ORCID iD 0000-0002-3033-6301 1 Centro Universitário Newton Paiva. Av. Silva Lobo, 1730, 30431-262, Nova Granada, Belo Horizonte, MG. Brasil. Correspondência para / Correspon dence to: SMS MENDONÇA. E-mail: <santuzam@yahoo.com.br>. 2 Universidade Federal de Minas Gerais, Faculdade de Odontologia. Belo Horizonte, MG, Brasil. Como citar este artigo / How to cite this article Oliveira MTG, Furtado PGS, Cardoso RFC, Souza ACD, Lima RPE, Mendonça SMS. Association between periodontitis and hyperglycemia. RGO, Rev Gaúch Odontol. 2018;66(3):199-204. http://dx.doi.org/10.1590/1981-863720180003000013217 ▼ ▼ ▼ ▼ ▼ Resultados Resultados Na amostra estudada foi identificado maior percentual de indivíduos com hiperglicemia no grupo com periodontite do que no grupo sem periodontite. Conclusão Sugere-se, portanto, que a solicitação de exames laboratoriais para avaliação dos níveis glicêmicos faça parte do protocolo odontológico de atendimento dos indivíduos com periodontite. RESUMO Objetivo O objetivo desse estudo foi analisar a prevalência de hiperglicemia não diagnosticada em indivíduos com periodontite. Objetivo O objetivo desse estudo foi analisar a prevalência de hiperglicemia não diagnosticada em indivíduos com periodont ABSTRACT Objective The objective of this study was to analyze the prevalence of undiagnosed hyperglycemia in individuals with periodontitis. bjective e objective of this study was to analyze the prevalence of undiagnosed hyperglycemia in individuals with periodontit Methods h l Methods The sample was composed of randomly selected individuals that had signed-up for treatment at the Newton Paiva University dental clinic. The volunteers were divided in two groups, the first composed by individuals without periodontitis (n = 30) and the second by individuals with periodontitis (n = 20). All participants answered a questionnaire about habits and overall health condition. Salivary flow was quantified and periodontal examination evaluated bleeding when probed, probing depth and clinical attachment level. Through laboratory tests, the fasting glucose and the glycated hemoglobin levels were quantified. Como citar este artigo / How to cite this article Oliveira MTG, Furtado PGS, Cardoso RFC, Souza ACD, Lima RPE, Mendonça SMS. Association between periodontitis and hyperglycemia. RGO, Rev Gaúch Odontol. 2018;66(3):199-204. http://dx.doi.org/10.1590/1981-863720180003000013217 ▼ ▼ ▼ ▼ ▼ METHODS Sialometry was employed for quantitative analysis of resting and stimulated saliva. In order to collect the saliva the participants remained 30 minutes without eating or drinking and 2 minutes without performing any chewing, swallowing or talking movements. For resting and stimulated sialometry, the individuals collected, Behavioral and biological social characteristics With a structured questionnaire, the following data were collected: gender, age, tobacco use, hypertension and family history of DM. The participants were also questioned about the presence of the following symptoms that are characteristic to DM: polydipsia, polyuria, polyphagia, unexplained wait loss and visual alteration. Immune-inflammatory alterations present in diabetic individuals increase the susceptibility of developing periodontitis, inflammatory disease that affects the supporting tissues of the teeth, being considered the sixth classic complication of DM. Epidemiological studies have emphasized that DM is a risk factor for periodontitis, increasing its probability of occurring by three times [9,10]. DM is also associated with greater severity of periodontitis [11]. The immune-inflammatory process resulting from the bacterial aggression, is the main responsible for the damage observed on the periodontal tissues, clinically characterized by bleeding to probing (BP), increase in probing depth (PD) and insertion loss, followed by bone tissue destruction. The evolution of periodontitis may eventually lead to loss of the dental element [12,13]. Conclusão Conclusão Sugere-se, portanto, que a solicitação de exames laboratoriais para avaliação dos níveis glicêmicos faça parte do protocolo odontológico de atendimento dos indivíduos com periodontite. Conclusão Sugere-se, portanto, que a solicitação de exames laboratoriais para avaliação dos níveis glicêmicos faça part atendimento dos indivíduos com periodontite. Termos de indexação: Complicações. Diabetes Mellitus. Periodontite. indexação: Complicações. Diabetes Mellitus. Periodon 1 Centro Universitário Newton Paiva. Av. Silva Lobo, 1730, 30431-262, Nova Granada, Belo Horizonte, MG. Brasil. Correspondência para / Correspon dence to: SMS MENDONÇA. E-mail: <santuzam@yahoo.com.br>. 2 U i id d F d l d Mi G i F ld d d Od t l i B l H i t MG B il va. Av. Silva Lobo, 1730, 30431-262, Nova Granada, Belo Horizonte, MG. Brasil. Correspondência para / Correspon mail: <santuzam@yahoo.com.br>. Como citar este artigo / How to cite this article Oliveira MTG, Furtado PGS, Cardoso RFC, Souza ACD, Lima RPE, Mendonça SMS. Association between periodontitis and hyperglycemia. RGO, Rev Gaúch Odontol. 2018;66(3):199-204. http://dx.doi.org/10.1590/1981-863720180003000013217 RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 MTG OLIVEIRA et al. INTRODUCTION up for dental treatment at the Newton Paiva University Center dentistry course, between the months of August and November 2015. The individuals were subdivided into two groups: without periodontitis (n=30) and with periodontitis (n=20). All the participants were informed about the objectives of the study and were only included as part of the study after manifesting interests by signing an informed consent form (ICF). Diabetes Mellitus (DM) constitutes of a group of alterations characterized by hyperglycemia, resulting from the deficiency of production and/or resistance to the role of insulin [1]. It represents one of the main public health problems, being the fourth cause of death in Brazil [2]. DM is classified, according to its etiopathogeny, in type 1 DM, type 2 DM, gestational DM and other specific types [3]. The inclusion factors were the presence of a minimum of 12 teeth, 18 years of age or older and absence of any contraindication for periodontal examination. Any individuals previously diagnosed with DM, submitted to antibiotic therapy or periodontal therapy during the last three months and carriers of the acquired immune deficiency syndrome (AIDS), were excluded. According to the World Health Organization, the average prevalence of DM worldwide is of 10% [4]. In Brazil, nearly five million individuals are diabetic, having an expectancy of a 50% increase to that number by 2025 [5]. According to studies made by the Ministry of Health, the total number of individuals with the disease increased 40% between 2006 and 2012 [6]. Around the world, nearly 4 million deaths per year are directly related to DM, since it also favors the worsening of systemic diseases [7,8]. The study was previously submitted and approved by the Ethics Committee under the number CAAE 37644314.6.0000.5097. Hyperglycemia contributes for the development of classic complications associated to DM, such as visual changes, nephropathies, cardiovascular diseases, neuropathies, susceptibility to infections and periodontitis [3]. Medical data Weight and height of the participating individuals were registered to calculate the body mass index (BMI). According the BMI the individuals were classified as underweight (BMI below 18.5), normal weight (BMI between 18.5 and 24.9), overweight (BMI between 25.0 and 29.9) and obesity (BMI above 30.0) [4]. All the participants performed, at the same laboratory, the fasting glycaemia and glycated hemoglobin (A1c) exams. Glycaemia was considered altered for fasting glycaemia and glycated hemoglobin values above 99mg/dL and 5.6%, respectively [3]. The present study aims to evaluate the prevalence of hyperglycemia not yet diagnosed in individuals with periodontitis. RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 Periodontal clinical examination The periodontal exam was performed to analyze the clinical parameters of BP, PD and clinical attachment level (CAL) of all teeth present. Mesial, distal, buccal and lingual measurements were registered for each tooth. The periodontal exam was performed by two researchers (MTGO and PGSF), who were previously calibrated, using a millimeter probe, model UNC-15 North Carolina (Hu- Friedy, Chicago, USA). Table 1. Characterization of the sample groups with and without periodontitis. Group without periodontitis Group with periodontitis p value Gender (%) Male 3 (10%) 7 (35%) p = 0.067 Female 27 (90%) 13 (65%) Average age (±) 41.0 ± 13.6 45.8 ± 10.9 p = 0.187 Family history of DM (%) 10 (33.3%) 5 (25%) p = 0.529 Polydipsia (%) 8 (26.7%) 4 (20%) p = 0.589 Polyuria (%) 5 (16.7%) 1 (5%) p = 0.214 Polyphagia (%) 6 (20%) 2 (10%) p = 0.345 Weight loss (%) 5 (16.7%) 3 (15%) p = 0.875 Visual alteration (%) 17 (56.7%) 4 (20%) p = 0.010 Hypertension (%) 6 (20%) 3 (15%) p = 0.652 BMI (±) 24.7 ± 4.9 25.3 ± 4.0 p = 0.566 BMI (%) underweight 1 (3.3%) 1 (5%) normal weight 18 (60%) 10 (50%) p = 0.633 Overweight 6 (20%) 7 (35%) Obesity 5 (16.7%) 2 (10%) Resting sialometry ml/ min. (±) 0.3 ± 0.2 0.4 ± 0.4 p = 0.461 Stimulated sialometry ml/ min 1.2 ± 0.7 1.5 ± 0.8 p = 0.204 Table 1. Characterization of the sample groups with and without periodontitis. The inability to determine the cementoenamel limit, presence of gingival morphology alteration, extensive carious lesion, iatrogenic restorative procedures or excess calculus that could difficult the periodontal exam were considered exclusion criteria for the teeth [15]. The bleeding exam was conducted while probing, by gently and carefully introducing the probe into the gingival sulcus until reaching its base. The bleeding was analyzed 30 to 60 seconds after probing. The occurrence of BP was registered for each surface in a dichotomy manner, indicating its presence or absence. The probing depth was obtained by measuring the distance from the gingival margin to the sulcus base or periodontal pocket. The CAL was determined by the distance between the cementoenamel junction and the periodontal sulcus or pocket base. Sample The convenience sample was composed by 50 randomly selected individuals from those that had signed 200 RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 Periodontitis and Hyperglycemia Periodontal clinical examination The criteria used to diagnose periodontitis was the presence of four or more teeth with at least one site PD ≥ 4 mm and CAL ≥ 3 mm associated to BP at the same site [16]. RESULTS in a millimeter tube, all the saliva produced during 5 minutes. During the stimulated sialometry the individuals chewed on sialogogo. The quantity of saliva and foam produced was verified directly on the collector during the five minutes. The total volume collected was divided by 5, obtaining a result in mL/min. The participants were grouped, according to the obtained salivary flow, in normal (flow from 1.5 to 3.0 mL/min), hypossalivation (flow from 0.05 to 1.45 mL/min) and sialorrhea (flow above 3.0 mL/min) [14]. Table 1 presents the characterization of the sample in the groups with and without periodontitis. The groups were considered similar according to the majority of the evaluated criteria. Among the DM symptoms investigated, visual alteration was more frequently reported by the individuals with periodontitis. The individuals from the group with periodontitis had significantly higher average fasting glycaemia than the individuals without periodontitis. However, when fasting glycaemia was categorically measured the difference was not maintained. The same occurred with glycated hemoglobin, when evaluated quantitatively and categorically. The participants with periodontitis presented a 97.7 mg/dL average fasting glycaemia, while for the participants without periodontitis the average was 85.5 mg/dL. Statistical analysis Group with normal glycated hemoglobin Group with altered glycated hemoglobin p value Periodontitis 11 (35.15%) 9 (47.4%) p = 0.405 Teeth present (±) 25.5 (4.7) 22.3 (5.1) p = 0.022 Sites with PD = 4 mm (%) 6.7 (7.0) 7.0 (5.6) p = 0.508 Sites with PD = 5 e 6 mm (%) 4.5 (8.3) 4.3 (7.9) p = 0.609 Sites with PD ≥ 7 mm (%) 0.3 (1.0) 1.0 (2.6) p = 0.112 Sites with BP (%) 16.6 (15.3) 23.0 (21.4) p = 0.294 Table 2. Characterization of the periodontal condition in the groups with and without periodontitis. Group without periodontitis Group with periodontitis p value Teeth present (±) 25.3 (4.5) 22.8 (5.5) p = 0.091 Sites with PD = 4 mm (%) 3.4 (4.3) 11.9 (5.9) p< 0.001 Sites with PD = 5 and 6 mm (%) 0.4 (0.9) 10.4 (10.2) p< 0.001 Sites with PD ≥ 7 mm (%) 0.1 (0.3) 1.4 (2.6) p< 0.001 Sites with BP (%) 8.9 (9.1) 34.3 (17.2) p< 0.001 Table 2. Characterization of the periodontal condition in the groups with and without periodontitis. Statistical analysis The comparison between the groups for the variables of interest was performed by the Chi-square, Mann-Whitney and Fishers’ Exact tests when appropriate. The results were considered significant with a probability below 5% (p < 0.05). All the statistical analysis was made with SPSS program (version 17.0). RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 201 MTG OLIVEIRA et al. Sialometry (%) Normal 9 (30%) 9 (45%) p = 0.432 Hyposalivation 20 (66.7%) 11 (55%) Sialorrhea 1 (3.3%) 0 (0%) Glycaemia (±) 85.5 (11.8) 97.7 (20.9) p = 0.005 Altered glycaemia (%) 4 (13.3%) 4 (20%) p = 0.529 Glycated hemoglobin (±) 5.6 (0.4) 6.0 (1.3) p = 0.493 Altered glycated hemoglobin (%) 10 (33.3%) 9 (45%) p = 0.405 Table 3. Characterization of the periodontal condition of the sample for the groups with normal and altered fasting glycaemia. Table 3. Characterization of the periodontal condition of the sample for the groups with normal and altered fasting glycaemia. Table 3. Characterization of the periodontal condition of the sample for the groups with normal and altered fasting glycaemia. Group with normal fasting glycaemia Group with altered fasting glycaemia p value Periodontitis 16 (38.1%) 4 (50%) p = 0.529 Teeth present (±) 24.7 (5.0) 22.4 (4.9) p = 0.230 Sites with PD = 4 mm (%) 7.4 (6.8) 4.0 (3.1) p = 0.259 Sites with PD = 5 and 6 mm (%) 4.3 (7.7) 5.0 (10.5) p = 0.803 Sites with PD ≥ 7 mm (%) 0.4 (0.9) 1.7 (4.0) p = 0.724 Sites with BP (%) 16.1 (14.2) 34.5 (27.1) p = 0.050 The characterization of the periodontal condition of the sample in the groups with and without periodontitis is presented in table 2. As expected, individuals with periodontitis presented greater percent of locations with altered PD and with BP, although the total number of teeth were similar between groups. Table 4. Characterization of the periodontal condition of the sample for the groups with normal and altered glycated hemoglobin. for the groups with normal and altered glycated hemoglobin. RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 1. Chávarry NGM, Vettore MV, Sansone C, Sheiham A. The relationship between diabetes mellitus and destructive disease: CONCLUSION Periodontitis is a classic complication of DM and its presence can help identify glycemic alterations not yet diagnosed. In the sample studied a greater prevalence of individuals with undiagnosed hyperglycemia in the group with periodontitis was identified. Therefor, dental appointments can be opportune moments to monitor periodontal patients’ glycemic levels. It is believed that periodontitis diagnosis can be an important indicator of the presence of hyperglycemia and a tool capable of identifying future candidate to develop DM. It is suggested, therefore, that the request of exams to evaluate glycemic levels becomes part of the dental appointment service protocol of individuals with periodontitis. Due to the multifactorial character of DM, scientific studies must consider the presence of confounding variables that could compromise the results. Increased age and high BMI, for example, are considered risk factors for DM [19]. In this study, however, the individuals with and without periodontitis were considered to be similar with regards to age and BMI, which increases the credibility in the obtained results. In contrast, smoking is a risk factor for periodontitis [20]. The habit of smoking interferes on the pathogenesis of periodontitis by altering vascularization, immune- inflammatory response and homeostasis of the tissues, increasing the susceptibility to the disease and causing greater destruction of periodontal tissues [21]. In the present study there was a greater prevalence of smokers in the group with periodontitis, although this difference was not statistically significant. DISCUSSION When the sample was characterized considering the groups of patients with normal and altered fasting glycaemia (table 3), the individuals with altered fasting glycaemia presented greater PD, prevalence of periodontitis and lower amount of teeth present, although these differences were not statistically significant. The percent of sites with BP was statistically higher for the group with altered fasting glycaemia (p=0.007). For the analysis of the sample considering the groups of individuals with normal and altered glycated hemoglobin there was no statistically significant difference between the groups with regards to the evaluated periodontal parameters, with exception to the number of teeth which was significantly higher in the group of individuals with normal levels of glycated hemoglobin (table 4). DM is a prevalent condition with high rates of morbidity and mortality. Its early diagnosis can contribute to the reduction of associated diseases and an improvement in prognosis. The fact that DM presents, in the majority of cases, a chronic evolution makes it remain undiagnosed for a long period of time, until symptoms and classic complications manifest themselves [9]. Periodontitis is a known complication to DM [17,18] and can aid in the detection of undiagnosed hyperglycemia. The results from the present study evidence that individuals with periodontitis present increased levels of blood glucose than individuals without periodontitis. Average fasting glycaemia values were higher for individuals with periodontitis when compared to individuals without periodontitis. Although there was no statistical significance, RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 202 Periodontitis and Hyperglycemia individuals. In the present study there was not salivary flow difference between groups with or without periodontitis. glycated hemoglobin levels were more elevated in individuals with periodontitis. The reduced sample size is a limitation to the study that can explain the difference among the statistical results for the fasting glycaemia and glycated hemoglobin exams. However, the two exams present high level of agreement and are indicated for diagnosis of hyperglycemia [3]. Collaborators RPE LIMA, study design, bibliographical gathering, creation of forms for data collection, calibration of grant students and volunteers, supervision of examining researchers, statistical analysis of the data, writing of the final article and article submission. SMS MENDONÇA, study design, bibliographical gathering, creation of forms for data collection, calibration of grant students and volunteers, supervision of examining researchers, statistical analysis of the data, writing of the final article and article submission. Even though periodontitis is the most prevalent DM oral complication, some studies have also shown a reduction to salivary flow in individuals with hyperglycemia [17]. However, the elevated age that is a risk factor for diabetes must be considered because it may be associated to the reduction of salivary flow, which could difficult result interpretation [19]. On the other hand, other studies have not identified salivary flow reduction among diabetic 2. Pace AE, Nunes PD, Ochoa-vigo K. O conhecimento dos familiares acerca da problemática do portador de diabetes mellitus. Rev Lat-am Enf. 2003;11(3):312-9. a meta-analysis. Oral Health Prev Dent. 2009;7:107-127. a meta-analysis. Oral Health Prev Dent. 2009;7:107-127. 2. Pace AE, Nunes PD, Ochoa-vigo K. O conhecimento dos familiares acerca da problemática do portador de diabetes mellitus. Rev Lat-am Enf. 2003;11(3):312-9. Collaborators MTG OLIVEIRA, bibliographic gathering, creation of forms for data collection, data collection and periodontal examination, creation of the database, writing of the final article. PGS FURTADO, bibliographic gathering, creation of forms for data collection, data collection and periodontal examination, creation of the database, writing of the final article. RFC CARDOSO, bibliographic gathering, data collection and periodontal examination, creation of the database, writing of the final article. ACD SOUZA, study design, bibliographical gathering, supervision of examining researchers, writing of the final article, review of the article. RPE LIMA, study design, bibliographical gathering, creation of forms for data collection, calibration of grant students and volunteers, supervision of examining researchers, statistical analysis of the data, writing of the final article and article submission. SMS MENDONÇA, study design, bibliographical gathering, creation of forms for data collection, calibration of grant students and volunteers, supervision of examining researchers, statistical analysis of the data, writing of the final article and article submission. not statistically significant. Additionally, DM is proved to be a risk factor for periodontitis. Immune and inflammatory system alterations are associated to the prevalence and increased severity of periodontitis in individuals with hyperglycemia. The neutrophils have reduced functioning, while monocytes and macrophages are found to be hyper reactive with increased production of proinflammatory cytocines [22]. The results evidenced that the prevalence of periodontitis was bigger for individuals with increased glycemic levels, although not statistically significant. Again, the reduced sample size could have affected the identification of the statistical significance. Even though periodontitis is the most prevalent DM oral complication, some studies have also shown a reduction to salivary flow in individuals with hyperglycemia [17]. However, the elevated age that is a risk factor for diabetes must be considered because it may be associated to the reduction of salivary flow, which could difficult result interpretation [19]. On the other hand, other studies have not identified salivary flow reduction among diabetic Additionally, DM is proved to be a risk factor for periodontitis. Immune and inflammatory system alterations are associated to the prevalence and increased severity of periodontitis in individuals with hyperglycemia. The neutrophils have reduced functioning, while monocytes and macrophages are found to be hyper reactive with increased production of proinflammatory cytocines [22]. The results evidenced that the prevalence of periodontitis was bigger for individuals with increased glycemic levels, although not statistically significant. Again, the reduced sample size could have affected the identification of the statistical significance. REFERENCES 203 RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204 MTG OLIVEIRA et al. 3. American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2014; 37(1):81-90. doi: 10.2337/dc10-S062. 14. Conceição MD, Marocchio LS, Fagundes RL. Técnica de sialometria para uso na prática clínica diária. Rev Assoc Paul Cir Dent. 2006;60(5):350-354. 4. World Health Organization. Diabetes. Geneva: World Health Organization; 2014. 15. Costa FO, Cota LOM, Costa JE, Pordeus IA. Periodontal disease progression among young subjects with no preventive dental care: A 52-month follow-up study. J Periodontol. 2007;78(1):198-203. doi: 10.1590/S1806- 83242010000200017 5. Moreira PL, Dupas G. Vivendo com diabetes: a experiência contada pela criança. Rev Lat-am Enf. 2006;14(1):25-32. 6. Brasil. Ministério da Saúde. Vigitel Brasil 2012: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico. 2013 [citado 2017 Set 15]. Disponível em: <http:// bvsms.saude.gov.br/bvs/publicacoes/vigitel_brasil_2012_ vigilancia_risco.pdf>. 16. Gomes-Filho IS, Cruz SS, Rezende EJC, dos Santos CAST, Soledade KR, Magalhaes MA, et al. Exposure measurement in the association between periodontal disease and prematurity/ low birth weight. J Clin Periodontal. 2007;34(1):957-963. doi: 10.1111/j.1600-051X.2007.01141.x 7. Barnes VM, Kennedy AD, Panagakos F, Devizio W, Trivedi HM, Jönsson T, et al. Global Metabolomic analysis of human saliva and plasma from healthy and diabetic subjects, with and without periodontal disease. PLoS One. 2014;9(8):1051-81. doi: 10.1371/journal.pone.0105181 17. Matthews DC. The relationship between diabetes and periodontal disease. J Can Dent Assoc. 2002 Mar;68(3):161-4. 18. Esteves Lima RP, Miranda Cota LO, Costa FO. Association between periodontitis and gestational diabetes mellitus: a case-control study. J Periodontol. 2013;84(1):1257-1265. doi: 10.1902/jop.2012.120350 8. Weidlich P. Association between periodontal diseases and systemic diseases. Braz Oral Res. 2008;22(11):32-43. doi: 10.1590/S1806-83242008000500006 19. Kahn SE, Cooper ME, Del Prato S. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present, and future. Lancet. 2014;383(1):1068-1083. doi: 10.1016/ S0140-6736(13)62154-6 9. Graves DT, Liu R, Oates TW. Diabetes-enhanced inflammation and apoptosis - impact on periodontal pathosis. Periodontol 2000. 2007;45(1):128-137. doi: 10.1111/j.1600- 0757.2007.00219.x 20. Warmakulasuriya S, Dietrich T, Bornstein MM, Casals Peidro E, Preshaw PM, Walter C, et al. Oral health risks of tobacco use and effects of cessation. Int Dent J. 2010;60(1):7-30. doi: 10.1922/IDJ_2532Warnakulasuriya24 10. Preshaw PM, Alba AL, Herrea D, Jepsen S, Konstantinidis A, Makrilakis K, et al. Periodontitis and diabetes: a two-way relationsheep. Diabetologia. 2012;55(1):21-31. doi: 10.1007/ s00125-011-2342-y 21. Palmer RM, Wilson RF, Hansan AS, Scott DA. Mechanisms of action of environmental factors-tobacco smoking. J Clin Periodontol. 2005;32(1)180-195. doi: 10.1111/j.1600- 051X.2005.00786.x 11. Received on: 10/8/2017 Final version resubmitted on: 12/9/2017 Approved on: 7/11/2017 REFERENCES Lalla E, Lamster IB, Drury S, Fu C, Schmidt AM. Hyperglycemia, glycoxidation and receptor for advanced glycation endproducts: potential mechanisms underlying diabetic complications, including diabetes-associated periodontitis. Periodontol 2000. 2000;23(1):50-62. doi: 10.1034/j.1600-0757.2000.2230104.x 22. Mealey BL, Oates TW. Diabetes mellitus and periodontal diseases. J Periodont. 2006;77:1289-1303. doi: 10.1902/ jop.2006.050459 22. Mealey BL, Oates TW. Diabetes mellitus and periodontal diseases. J Periodont. 2006;77:1289-1303. doi: 10.1902/ jop.2006.050459 12. Page RC, Offenbacher S, Schroeder HE, Seymour GJ, Kornman KS. Advances in the pathogenesis of periodontitis: summary of developments, clinical implication and future directions. Periodontol 2000. 1997;14(1):216-248. doi: 10.1111/j.1600- 0757.1997.tb00199.x 13. Kinane DF. Causation and pathogenesis of periodontal disease. Periodontol 2000. 2001;25:8-20. doi.: 10.1034/j.1600- 0757.2001.22250102.x 204 RGO, Rev Gaúch Odontol. 2018 Jul-Set; 66(3):199-204
https://openalex.org/W2934701388	https://europepmc.org/articles/pmc6445408?pdf=render	English	null	Disruption of KPC-producing Klebsiella pneumoniae membrane via induction of oxidative stress by cinnamon bark (Cinnamomum verum J. Presl) essential oil	PloS one	2,019	cc-by	11,086	RESEARCH ARTICLE Shun-Kai YangID1, Khatijah Yusoff2, Mokrish Ajat3, Warren Thomas4, Aisha Abushelaibi5, Riaz Akseer5, Swee-Hua Erin Lim4,5, Kok-Song Lai1* 1 Department of Cell and Molecular Biology, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, Serdang, Selangor, Malaysia, 2 Department of Microbiology, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, Serdang, Selangor, Malaysia, 3 Department of Veterinary Preclinical Sciences, Faculty of Veterinary Medicine, Universiti Putra Malaysia, Serdang, Selangor, Malaysia, 4 Perdana University-Royal College of Surgeons in Ireland, School of Medicine, Perdana University, Serdang, Selangor, Malaysia, 5 Health Sciences Division, Abu Dhabi Women’s College, Higher Colleges of Technology, Abu Dhabi, United Arab Emirates * laikoksong@upm.edu.my Abstract Citation: Yang S-K, Yusoff K, Ajat M, Thomas W, Abushelaibi A, Akseer R, et al. (2019) Disruption of KPC-producing Klebsiella pneumoniae membrane via induction of oxidative stress by cinnamon bark (Cinnamomum verum J. Presl) essential oil. PLoS ONE 14(4): e0214326. https://doi.org/10.1371/ journal.pone.0214326 Klebsiella pneumoniae (KP) remains the most prevalent nosocomial pathogen and car- ries the carbapenemase (KPC) gene which confers resistance towards carbapenem. Thus, it is necessary to discover novel antimicrobials to address the issue of antimicrobial resistance in such pathogens. Natural products such as essential oils are a promising source due to their complex composition. Essential oils have been shown to be effective against pathogens, but the overall mechanisms have yet to be fully explained. Under- standing the molecular mechanisms of essential oil towards KPC-KP cells would provide a deeper understanding of their potential use in clinical settings. Therefore, we aimed to investigate the mode of action of essential oil against KPC-KP cells from a proteomic per- spective by comparing the overall proteome profile of KPC-KP cells treated with cinna- mon bark (Cinnamomum verum J. Presl) essential oil (CBO) at their sub-inhibitory concentration of 0.08% (v/v). A total of 384 proteins were successfully identified from the non-treated cells, whereas only 242 proteins were identified from the CBO-treated cells. Proteins were then categorized based on their biological processes, cellular components and molecular function prior to pathway analysis. Pathway analysis showed that CBO induced oxidative stress in the KPC-KP cells as indicated by the abundance of oxidative stress regulator proteins such as glycyl radical cofactor, catalase peroxidase and DNA mismatch repair protein. Oxidative stress is likely to oxidize and disrupt the bacterial membrane as shown by the loss of major membrane proteins. Several genes selected for qRT-PCR analysis validated the proteomic profile and were congruent with the proteomic abundance profiles. In conclusion, KPC-KP cells exposed to CBO undergo oxidative stress that eventually disrupts the bacterial membrane possibly via interaction with the phospholipid bilayer. Interestingly, several pathways involved in the bacterial Editor: Marie-Joelle Virolle, Universite Paris-Sud, FRANCE Editor: Marie-Joelle Virolle, Universite Paris-Sud, FRANCE Disruption of KPC-producing Klebsiella pneumoniae membrane via induction of oxidative stress by cinnamon bark (Cinnamomum verum J. Presl) essential oil Shun-Kai YangID1, Khatijah Yusoff2, Mokrish Ajat3, Warren Thomas4, Aisha Abushelaibi5, Riaz Akseer5, Swee-Hua Erin Lim4,5, Kok-Song Lai1* * laikoksong@upm.edu.my Introduction Klebsiella spp. are Gram-negative rod shaped bacteria that cause bacterial pneumonia with a high fatality rate if infection remains untreated in the clinical setting [1]. Globally, the vast majority of Klebsiella infections are hospital-acquired. Nosocomial Klebsiella infections are mainly caused by Klebsiella pneumoniae, the medically most important species of the genus which primarily attacks immune-compromised individuals who are hospitalized and suffer from severe underlying diseases such as diabetes mellitus, chronic pulmonary obstruction or even cancer. It is estimated that Klebsiella spp. cause 8% of all nosocomial bacterial infections in the United States and Europe, with 50.1% of these cases being caused by Klebsiella pneumo- niae placing Klebsiella spp. among the eight most important infectious pathogens in hospitals [1]. In 1983, the first report of a plasmid-mediated extended spectrum beta-lactamases (ESBLs) capable of hydrolyzing extended-spectrum cephalosporins was discovered [2,3]. Car- bapenems are one of the last lines of antibiotic treatment for severe drug-resistant bacterial infections, and are now the treatment of choice for serious infections caused by pathogens car- rying the ESBL gene. This has led to an increased reliance on carbapenems in clinical practice [4]. In tandem with this, the first carbapenemase producing K. pneumoniae isolate was reported in North Carolina in 2001. This enzyme was termed K. pneumoniae carbapenemase (KPC) and conferred resistance to carbapenem antibiotics [5]. KPCs are encoded by the gene blaKPC, whose potential for inter-species and geographic dissemination is largely explained by its location within a Tn3-type transposon, Tn4401 that is capable of inserting itself into diverse plasmids of Gram-negative bacteria. Although K. pneumoniae remains the most preva- lent bacterial species carrying KPCs, the enzyme has been identified in several other Gram- negative bacilli such as Escherichia coli, Pseudomonas aeruginosa and Salmonella enterica due to horizontal gene transfer [6]. To worsen this issue, KPC-producing K. pneumoniae (KPC-KP) possesses innate antibiotic resistance in the form of an efflux pump, which generally removes the antibiotics that have penetrated the bacterial membrane, from the cytoplasm into the extracellular environment. Membrane permeability can also be altered in the presence of antibiotics; preventing the access of antibiotics into the cells, which when coupled to the other mechanisms, enables resistance against higher concentrations of antibiotics [7]. In order to address to this particular issue, there had been constant efforts to discover novel antimicrobials for clinical use. Cinnamon bark essential oil and its role in bacterial membrane disruption membrane repair system were also affected by oxidative stress, contributing to the loss of cells viability. membrane repair system were also affected by oxidative stress, contributing to the loss of cells viability. Competing interests: The authors have declared that no competing interests exist. Editor: Marie-Joelle Virolle, Universite Paris-Sud, FRANCE Received: November 27, 2018 Accepted: March 10, 2019 Published: April 2, 2019 Copyright: © 2019 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: © 2019 Yang 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 files. Funding: The research received funding UPM Internal Grant (GP-IPS/2016/9505800), Malaysian Medical Association (MMA), and Fundamental Research Grant Scheme (FRGS/1/2018/SKK11/ PERDANA/02/1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 1 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Introduction (2015) identified the mode of action of paclitaxel as chemotherapeutic drugs in HeLa cells by tampering with the abundance of tumor suppressor PDCD4 via LC-MS/MS proteomic profiling [18]. Similarly, Kawatani et al. (2016) also revealed the role of collismycin A as an iron chelator antagonizing cancer cells, using a proteomic approach [19]. Similarly, Silva et al. (2011) hypothesized that coriander essential oil exerts its bactericidal activity towards both Gram-positive and Gram–negative bacteria via membrane damage by measuring their efflux activity, respiratory activity and membrane potential [16]. To further support and understand the antimicrobial activity of essential oils, mass spectrometry-based proteomics analysis has become the tool of choice offering the identification and quantifica- tion of the proteome of an organism. There has been a tremendous improvement in instru- ment performance and the computational tools used in proteomic studies in recent years, which facilitates the understanding of the mechanisms of action of potential antimicrobial agents in the clinical setting. In the most widely used “bottom-up” approach to proteomics, liq- uid chromatography coupled with mass spectrometry (LC-MS/MS), enables a complex mix- ture of proteins to be first subjected to enzymatic cleavage; the resulting peptide products are separated based on chemical or physical properties and analyzed using a mass spectrometer. The proteome can then be analyzed, quantified and compared by using third party analytical software such as Progenesis QI (Progenesis Group Sdn. Bhd.) or Perseus (Max Planck Institute of Biochemistry). For instance, Xu et al. (2015) identified the mode of action of paclitaxel as chemotherapeutic drugs in HeLa cells by tampering with the abundance of tumor suppressor PDCD4 via LC-MS/MS proteomic profiling [18]. Similarly, Kawatani et al. (2016) also revealed the role of collismycin A as an iron chelator antagonizing cancer cells, using a proteomic approach [19]. In our previous study, we have successfully shown that cinnamon bark (Cinnamomon verum J. Presl) essential oil (CBO) is, in fact, an effective antimicrobial when used against KPC-KP with an extremely low minimum inhibitory concentration (MIC) of 0.16% (v/v) [11]. Furthermore, combinations of CBO and antibiotic meropenem further reduced the MIC of CBO to 0.08% (v/v) which makes it a putative candidate to be used as an antimicrobial in the clinical setting [11]. Other studies showed that CBO with a low MIC value is effective against a variety of Gram-positive and Gram–negative pathogens. Introduction Zamirah and colleagues (2013) dem- onstrated that CBO is effective against oral pathogens such as Aggregatibacter actinomycetem- comitans, Fusobacterium nucleatum, Porphyromonas gingivalis, Streptococcus salivarius, S. mitis and S. mutans with a low MIC ranging from 0.02 to 0.06% (v/v) [20]. Kaskatepe et al (2016) also showed that CBO is highly effective against other opportunistic pathogens, includ- ing Escherichia coli, Pseudomonas aeruginosa and multidrug-resistant Staphylococcus aureus (MRSA), with a MIC ranging from 0.009% to 0.078% (v/v) [21]. As already mentioned above, preliminary studies had postulated that essential oils affect bacterial membrane and/or their efflux system via numerous assays. Nevertheless, none had explained or gave an overview on the overall changes that the bacterial cells undergo when exposed to essential oils, especially in the perspective of proteomics. Thus, this study was carried out to understand and possibly bridge the missing links among previous studies regarding the mode of action of essential oils as an antimicrobial agent from the proteomic perspective, using CBO and KPC-KP as a model of study. Introduction Natural products such as essential oil consisting a plethora of chemical compounds, are becoming a popular mainstream platform for researchers in drug discovery [8]. Numerous studies have also demonstrated the efficacy of essential oils from curry plant (Helichrysum italicum (Roth) G. Don fil.), peppermint (Mentha x piperita L.), tea tree (Melaleuca alternifolia (Maiden & Betche) Cheel.) and marjoram (Origanum majorana L.) as promising antimicrobials. Multiple studies have shown the synergistic effects between vari- ous essential oils and antibiotics, potentially solving the antibiotic resistance issue in the clini- cal setting [9–14]. Despite this, only a few studies have been carried out to elucidate the mode of action of several essential oils on different bacteria; most of these studies have postulated that essential oils exert their antimicrobial activities by disrupting bacterial cell membrane and/or their efflux systems through various assays [15–17]. For example, de Souza et al. (2009) postulated that Origanum vulgare L. essential oil affects the membrane permeability of Staphy- lococcus aureus by studying potassium ion efflux and scanning electron microscopy [15]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 2 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Similarly, Silva et al. (2011) hypothesized that coriander essential oil exerts its bactericidal activity towards both Gram-positive and Gram–negative bacteria via membrane damage by measuring their efflux activity, respiratory activity and membrane potential [16]. To further support and understand the antimicrobial activity of essential oils, mass spectrometry-based proteomics analysis has become the tool of choice offering the identification and quantifica- tion of the proteome of an organism. There has been a tremendous improvement in instru- ment performance and the computational tools used in proteomic studies in recent years, which facilitates the understanding of the mechanisms of action of potential antimicrobial agents in the clinical setting. In the most widely used “bottom-up” approach to proteomics, liq- uid chromatography coupled with mass spectrometry (LC-MS/MS), enables a complex mix- ture of proteins to be first subjected to enzymatic cleavage; the resulting peptide products are separated based on chemical or physical properties and analyzed using a mass spectrometer. The proteome can then be analyzed, quantified and compared by using third party analytical software such as Progenesis QI (Progenesis Group Sdn. Bhd.) or Perseus (Max Planck Institute of Biochemistry). For instance, Xu et al. CBO and KPC-producing K. pneumoniae CBO (Cinnamon bark Sri Lanka, lot number: 6488) used throughout the study was purchased from Aroma Trading Ltd. (Milton Keynes, UK). The composition of CBO had been deter- mined in our previous study via GC-MS [10]. The CBO was filter-sterilized using a 0.22 μm PES syringe membrane filter (Bioflow, Malaysia). KPC-producing K. pneumoniae, K. pneumo- niae BAA-1705 (KPC-KP) was purchased from American Type Culture Collection (ATCC, 3 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Cinnamon bark essential oil and its role in bacterial membrane disruption Manassas, VA, USA) and was cultured on Mueller-Hinton broth (MHB) and agar (MHA) both from Sigma Aldrich, USA. Peptide digestion Approximately 100 μg of total protein was resuspended in 100 μL of 50mM ammonium bicar- bonate (pH 8.0). RapiGest (Waters Corporation, USA) at final concentration of 0.05% was added to the protein in equal parts. Protein from each sample was then concentrated to 100μL using Vivaspin column (GE Healthcare, USA) with a molecular weight cut-off (MWCO) of 3000 and incubated at 80˚C for 15 min. The proteins were reduced in the presence of 5 mM dithiothreitol (DTT) at 60˚C for 30 min, and then alkylated in the dark using 10 mM iodoace- tamide at room temperature for 45 min. Proteolytic digestion was performed using Trypsin Gold (Promega, USA) at a ratio of 1:200 parts of protein, followed by incubation at 37˚C over- night. Tryptic digestion and RapiGest activity were terminated by the addition of 1 μL concen- trated trifluoroacetic acid (TFA) followed by the incubation of samples at 37˚C for 20 min. The tryptic peptide solution of each sample was centrifuged at 14000 rpm for 20 min, and the resulting supernatants were collected in clean microcentrifuges tube kept at -80˚C until subse- quent analysis. CBO treatment and protein extraction KPC-KP cell cultures were divided into two treatment groups, namely control (no treatment) and CBO treated prior to protein extraction. The CBO final concentration used was 0.08% (v/ v), as determined by Yang et al. (2017) [11]. Both treatment groups had a final volume of 50 mL, supplemented with Tween 80 at final concentration of 10% to enhance the solubility of CBO and a standard inoculum of 1 × 105 cfu/mL KPC-KP cells. Samples were incubated at 37˚C with shaking at 200 rpm for 16 h to obtain sufficient cells for protein extraction. The cells from both treatment groups were pelleted by centrifugation at 9000 rpm for 10 min, washed for at least three times and resuspended in 500 μL cold protein extraction buffer (50 mM ammonium bicarbonate, 10 mM phenylmethylsulfonyl fluoride). Samples were then sonicated on ice at 20 amplitude for 10 cycles to lyse the cells; each cycle consisted of 10 seconds of soni- cation followed by 20 s of cooling, with a Qsonica Sonicator Q55 (Fischer Scientific, USA). Sonicated samples were then pelleted at 4˚C and 10000 rpm for 1 h; supernatants were then collected and quantified via Bradford assay. The protein concentration of each sample was standardized to 1 mg/mL for the subsequent proteomic analysis. Treatment and analysis was standardized in three distinct biological replicates to ensure reproducibility of the experiment. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Protein identification Raw data were processed using Thermo Scientific Proteome Discoverer Software v2.1 with the SEQUEST HT search engine. The MS ion intensities were calculated based on the accurate mass and time tag strategy. The accurate alignment of the detected LC retention time and m/z value across different analyses, together with the area under chromatographic elution profiles of the identified peptides could be compared between different samples. For protein identifica- tion, data was searched against the Uniprot K. pneumoniae database with 1% strict FDR and 5% relaxed FDR criteria using Percolator. Search parameters were set to 2 miscleavages with fixed modification of carbamidomethylation and variable modification through methionine oxidation or asparagine and glutamine deamidation. A fragment tolerance of 0.6 Da and a pre- cursor tolerance of 10 ppm were used with trypsin as a digestion enzyme. Proteins with at least 2 unique peptides implied a greater confidence of protein identity. Peptide separation and MS analysis The nanoLC-MS/MS analysis was performed on an Orbitrap Fusion Tribrid mass spectrome- ter (Thermo Scientific, USA). The samples (2 μL containing 2 μg peptides) were injected and separated on an EASY-nLC 1000 (Dionex, Thermo Scientific, USA) equipped with an Easy- Spray Column Acclaim PepMap C18 100Å (2 μm, 50 μm × 15 cm, Thermo Scientific, USA). Samples were separated at a flow rate of 250 nL/min and using a gradient of 5% to 50% aceto- nitrile (ACN) in 0.1% formic acid (FA) for 45 min followed by a further gradient to 85% ACN in 0.1% FA for 2 min. The column was equilibrated back to 5% ACN with 0.1% FA over 1 min and maintained at 5% CAN until the next sample injection. Mass spectrometry was conducted in a positive ion mode with a nanospray voltage of 1.5 kV and a source temperature of 250˚C. The instrument was operated in a data-dependent acquisition (DDA) mode with an Orbitrap MS (OTMS) survey scan using the following parameters: mass range of m/z 310–1800 with resolving power of 120000, automatic gain control (AGC) of 400000 and a maximum injection PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 4 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption time of 50 ms. The Top Speed Mode of 3 seconds was used in the selection of precursors with the monoisotopic charge state of 2 to 7. These precursors were further analysed by MS/MS scanning. All precursors were filtered using a 20 s dynamic exclusion window and intensity threshold of 5000. The MS/MS spectra were analysed using ion trap MS (ITMS) with the fol- lowing parameters: rapid scan rate with resolving power of 60000, AGC of 100, isolation win- dow of 1.6 m/z and maximum injection time of 250 ms. Precursors were then fragmented by collision-induced dissociation (CID) and high-energy collision dissociation (HCD) at normal- ized collision energy of 30% to 28%. Protein quantification and data analysis Protein quantification and statistical analyses were performed using Perseus Software v1.6.0.7 (Max Planck Institute of Biochemistry). Each control and treated samples analysis consisted of three biological replicates with three technical replicates, independently injected into the LCMS/MS. The protein file with three technical replicates in txt. format from Proteome Dis- coverer software were uploaded to Perseus for further comparative analysis between the sam- ples. The data were log2-transformed to stabilise the variance and scale-normalised to the same mean intensity across the technical replicates. The mean values for all three technical rep- licates of the same biological samples were grouped together in the same matrix and valid val- ues were obtained by filtering with ‘at least 2’, eliminating proteins which only existed in one of the technical replicates. Finally, all biological replicates of the same treatment group were consolidated into the same matrix, with the missing values imputated with the random num- ber derived from a normal distribution. The histograms were plotted to compare the ratio distributions between all samples. Differentially expressed proteins between control and treat- ment were detected using a T-test, the p-values were also adjusted for multiple-testing using the permutation-based false discovery rate, with a number of randomization of 250. Proteins were considered to be significantly differentially expressed between treatment groups with adjusted p-values of <0.05 and a fold changes of -1 or +1. CBO treatment, RNA extraction and cDNA synthesis KPC-KP cell culture were treated with CBO or vehicle buffer prior to RNA extraction. The CBO final concentration used was 0.08% (v/v), as determined in our previous work [11]. Both treatment groups had a final volume of 50 mL, and contained Tween 80 at final concentration of 10% to enhance the solubility of CBO. The standard KPC-KP cell inoculum was 1 × 105 cfu/ mL. Samples were incubated at 37˚C with shaking at 200 rpm for 4 h followed by RNA extrac- tion using the TransZol RNA purification kit (Transgen Biotech, China). RNA (0.5 ng) was 5 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Cinnamon bark essential oil and its role in bacterial membrane disruption subjected to reverse transcription with QuantiNova Reverse Transcription Kit (QIAGEN, Ger- many) in a 20-μl reaction volume. The synthesized complementary DNA (cDNA) was stored at −20 ˚C until further use. Proteomic expression validation through qRT-PCR analysis The RNA abundance for several upregulated proteins from CBO-treated K. pneumoniae BAA- 1705 was determined by qRT-PCR using QuantiNova SYBR Green PCR (QIAGEN, Germany) on CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Inc, USA). The Livak method was employed to assess the relative expression of three upregulated genes, namely cytidine deaminase (cdd), thiamine phosphate synthase (thiE) and uridine phosphory- lase (udp), one down regulated gene, namely, 3-hydroxydecanoyl-[acyl-carrier-protein] dehy- dratase (fabA) and two housekeeping genes, namely 16S rRNA and OmpK36 porin. Reactions were performed in triplicate and data were analysed by using the CFX Manage Software (Bio- Rad). The thermal cycling conditions were as follows: 95 ˚C for 2 min, followed by 40 cycles of 95 ˚C for 5 s, and 60 ˚C for 10 s. In all experiments, no template reactions were used as nega- tive controls. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Comparative proteome profiling of KPC-KP treated with CBO Comparative proteomic analysis was carried out between non-treated and CBO-treated KPC-KP cells in four independent experiments using Perseus Software v1.6.0.7 (Max Planck Institute of Biochemistry). The protein profiles of treated and non-treated cells varied signifi- cantly with no outliers from the replicates, as shown in the principle component analysis (PCA) in Fig 1A. In addition, the Pearson correlation values between non-treated and CBO- treated KPC-KP were of high confidence (0.788 to 0.822), indicating that both the compared groups are of the same organism with no contamination within the samples (Fig 1B). The volcano plot (Fig 1C) of the comparative proteome between non-treated and CBO-treated KPC-KP cells showed a total of 46 proteins with significantly different abundance; 25 proteins of which were upregulated, whereas the other 21 proteins were downregulated in response to CBO treatment. A total of 384 proteins were identified from the non-treated cells, whereas only 242 proteins were identified from the CBO-treated cells, A total of 152 proteins were only expressed in the non-treated cells while 10 proteins were only expressed in the CBO-treated cells, the other 232 proteins were present in both groups of cells (Fig 2A). Proteins that were present in both sam- ples were compared within the Perseus software to measure up- and downregulation of pro- teins while the rest of the proteins are referred as absent or present from each treatment group (Table 1). Proteins that were differentially present or absent, and up- or downregulated between the two groups of cells were subjected to gene ontology analysis with regards to bio- logical process involvement, cellular component and molecular function (Fig 2). The majority of the proteins identified were linked to cellular and metabolic processes (37.5% and 35.7%), a number were also involved with cellular component organization and response to stimulus (8.2% and 7.9%) (Fig 2B). Cellular component wise, the majority of the proteins identified were categorized under cytosol and cytoplasm (41.5% and 30.3%) followed by macromolecular complex and plasma membrane (11.9% and 10.5%; Fig 2d). The categorization by molecular function identified proteins that were involved in catalytic activity and binding (46% and 45%) in addition to proteins that were structural molecules and or had transcription regulator activ- ity (2.8% and 2.1%) (Fig 2F). Most proteins involved in biological processes, cellular compo- nent and molecular function were downregulated (Fig 2C, 2E and 2G). PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Comparative proteome profiling of KPC-KP treated with CBO PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 6 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Fig 1. Exploratory analysis output of non-treated and CBO-treated KPC-KP cells using Perseus v1.6.0.7 software. (A) Principal component analysis (PCA) identifies differences between the non-treated (designated with green square) and CBO-treated (designated with red square) KPC-KP cells. (B) Multi-scatter plot with the Pearson correlation value of one of the profiles from the non-treated and CBO-treated KPC-KP cells. (C) Volcano plot showing up- (designated with green square) and downregulated (designated with red square) proteins from the CBO-treated KPC-KP cells. https://doi.org/10.1371/journal.pone.0214326.g001 Fig 1. Exploratory analysis output of non-treated and CBO-treated KPC-KP cells using Perseus v1.6.0.7 software. (A) Principal component analysis (PCA) identifies differences between the non-treated (designated with green square) and CBO-treated (designated with red square) KPC-KP cells. (B) Multi-scatter plot with the Pearson correlation value of one of the profiles from the non-treated and CBO-treated KPC-KP cells. (C) Volcano plot showing up- (designated with green square) and downregulated (designated with red square) proteins from the CBO-treated KPC-KP cells. https://doi.org/10.1371/journal.pone.0214326.g001 https://doi.org/10.1371/journal.pone.0214326.g001 https://doi.org/10.1371/journal.pone.0214326.g001 Of the overlapping 232 proteins identified in both treatment groups, only 41 proteins showed significant differences, in terms of fold change between the groups. The majority of the proteins which showed significant abundance difference were downregulated (51.2%) when compared to the non-treated KPC-KP cells, whereas the other 48.8% of the proteins were upregulated as shown in Table 1. Proteins that were absent or present in each treatment group were also listed in Table 1, with proteins that were only present in the CBO-treated KPC-KP cells listed under the upregulated proteins whereas proteins which were only present in the non-treated KPC-KP cells listed under the downregulated protein section. These pro- teins were then subjected to KEGG pathway analysis to elucidate mechanism involved in the action of CBO on KPC-KP cells. 7 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Cinnamon bark essential oil and its role in bacterial membrane disruption Fig 2. Comparative proteomic analysis of protein from CBO-treated KPC-KP cells. (A) Venn diagram of the total protein obtained from non-treated KPC-KP and CBO-treated KPC-KP cells, a total of 394 proteins were identified in non-treated KPC-KPs (384 proteins) and CBO-treated KPC-KPs (242 proteins) together. These total proteins were categorized based on the related biological process (B), cellular component (D) and molecular function (F). Comparative proteome profiling of KPC-KP treated with CBO The numbers of upregulated and downregulated genes were also compared with respect to the relevant biological process (C), cellular component (E) and molecular function (G). Gene ontology analysis in terms of biological process of protein with significant abundance from CBO-treated KPC-KP cells. Fig 2. Comparative proteomic analysis of protein from CBO-treated KPC-KP cells. (A) Venn diagram of the total pro Fig 2. Comparative proteomic analysis of protein from CBO-treated KPC-KP cells. (A) Venn diagram of the total protein obtained from non-treated KPC-KP and CBO-treated KPC-KP cells, a total of 394 proteins were identified in non-treated KPC-KPs (384 proteins) and CBO-treated KPC-KPs (242 proteins) together. These total proteins were categorized based on the related biological process (B), cellular component (D) and molecular function (F). The numbers of upregulated and downregulated genes were also compared with respect to the relevant biological process (C), cellular component (E) and molecular function (G). Gene ontology analysis in terms of biological process of protein with significant abundance from CBO-treated KPC-KP cells. Fig 2. Comparative proteomic analysis of protein from CBO-treated KPC-KP cells. (A) Venn diagram of the total protein obtained from non-treated KPC-KP and CBO-treated KPC-KP cells, a total of 394 proteins were identified in non-treated KPC-KPs (384 proteins) and CBO-treated KPC-KPs (242 proteins) together. These total proteins were categorized based on the related biological process (B), cellular component (D) and molecular function (F). The numbers of upregulated and downregulated genes were also compared with respect to the relevant biological process (C), cellular component (E) and molecular function (G). Gene ontology analysis in terms of biological process of protein with significant abundance from CBO-treated KPC-KP cells. https://doi.org/10.1371/journal.pone.0214326.g002 https://doi.org/10.1371/journal.pone.0214326.g002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Cinnamon bark essential oil and its role in bacterial membrane disruption Table 1. Top 100 proteins showing significant abundance difference (together with their accession numbers) in KPC-KP cells treated with CBO. The list is sorted according to descending positive or negative fold changes of each protein. No Proteins Gene Name Uniprot Accession No. Bacterial membrane disruption The Gram negative bacterial cell outer barrier consists of three separate component, the outer membrane, the peptidoglycan and the plasma membrane [22]. Both the outer membrane and the plasma membrane are made up of a phospholipid bilayer embedded with membrane pro- teins. The biochemical feature which differentiates the layers is the presence of lipopolysaccha- rides uniquely in the outer membrane. Of the identified proteins from the KPC-KP cells, 10.5% were located at the bacterial membrane (Fig 2D and 2E). Following exposure to CBO, our proteomic profiling showed 5 outer membrane exclusive proteins and 26 plasma mem- brane exclusive proteins were completely lost after the exposure to CBO (Table 2). Similarly, Wu and colleagues (2016) found that 3-p-trans-coumaroyl-2-hydroxyquinic acid, a phenolic compound isolated from Himalayan cedar essential oil caused bacterial membrane damage and the loss of membrane proteins due to specific interactions between the compound and the lipid and proteins within the bacterial membrane [23]. This disrupted the membrane integrity of the bacteria which caused the loss of plasma membrane protein. The downregulation of outer membrane integrity regulators such as the TolB porin-interacting protein, the large-con- ductance mechanosensitive channel and outer membrane protein assembly factor BamA in KPC-KP cells exposed to CBO also indicated the loss of membrane integrity in KPC-KP cells [24]. In addition, proteins involved in energy generation, such as the ATP synthase, the elec- tron transport complex and the NADH-quinone oxidoreductases which are often embedded PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 8 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Table 1. (Continued) No Proteins Gene Name Uniprot Accession No. General Function Fold Change Upregulated Proteins 1 Autonomous glycyl radical cofactor grcA A6TCJ1 Stress response 7.07 2 Beta-galactosidase 2 lacZ2 A6TI29 Carbohydrate metabolism 4.65 3 50S ribosomal protein L33 rpmG B5XTG7 Protein biosynthesis 3.97 4 Sucrose porin scrY P27218 Transport 3.05 5 ATP-dependent protease ATPase subunit HslU hslU B5XZ37 Stress response 2.68 6 Uridine phosphorylase udp P52671 Pyrimidine biosynthesis 2.57 7 Agmatinase speB B5XUB2 Amine and polyamine biosynthesis 2.55 8 Phosphoribosylaminoimidazole-succinocarboxamide synthase purC A6TC99 Purine biosynthesis 2.52 9 Phosphomethylpyrimidine synthase thiC A6TGQ1 Thiamine biosynthesis 2.33 10 Phosphoribosylformylglycinamidine cyclo-ligase purM A6TCB3 Purine biosynthesis 2.21 11 NAD-dependent malic enzyme maeA A6T9K7 Carbohydrate metabolism 2.19 12 Probable Fe(2+)-trafficking protein yggX A6TDX0 Stress response 2.17 13 Thiamine-phosphate synthase thiE A6TGQ0 Thiamine biosynthesis 2.08 14 Chaperone protein DnaK dnaK A6T4F4 Stress response 1.73 15 Iron-sulfur cluster insertion protein ErpA erpA A6T4W0 Stress response 1.70 16 Cytidine deaminase cdd A6TBN1 Purine biosynthesis 1.62 17 Elongation factor Ts tsf A6T4X2 Protein biosynthesis 1.58 18 Ribosomal protein S12 methylthiotransferase RimO rimO A6T6T1 RNA processing and modification 1.50 19 Dual-specificity RNA methyltransferase RlmN rlmN A6TCD6 RNA processing and modification 1.26 20 Ribosomal RNA small subunit methyltransferase C rsmC A6THY6 RNA processing and modification 1.16 CBO-treated KPC-KP Exclusive Proteins 21 Bifunctional protein GlmU glmU B5XZM7 Peptidoglycan biosynthesis CBO-treated 22 Carbon storage regulator homolog csrA B5XVB9 Protein biosynthesis CBO-treated 23 Cobalamin biosynthesis protein CobD cobD A6TDC6 Cobalamin biosynthesis CBO-treated 24 DNA ligase ligA A6TC47 Stress response CBO-treated 25 Ferrochelatase hemH B5Y0N2 Porphyrin biosynthesis CBO-treated 26 Hydroxyacylglutathione hydrolase gloB A6T510 Secondary metabolite metabolism CBO-treated 27 Integration host factor subunit beta ihfB A6T704 DNA processing CBO-treated 28 Probable malate:quinone oxidoreductase (Fragment) mqo O32719 Energy synthesis CBO-treated 29 tRNA 2-thiocytidine biosynthesis protein TtcA ttcA B5XRN3 RNA processing and modification CBO-treated Downregulated Proteins 30 3-hydroxydecanoyl-[acyl-carrier-protein] dehydratase fabA A6T748 Fatty acid biosynthesis -3.46 31 ATP-dependent protease subunit HslV hslV B5XZ36 Stress response -3.25 32 HTH-type transcriptional regulator IscR iscR A6TCF2 Stress response -3.05 33 Ribosome maturation factor RimP rimP A6TEI9 Ribosome biogenesis -2.80 34 50S ribosomal protein L32 rpmF A6T7E9 Translation -2.78 35 Recombination-associated protein RdgC rdgC B5Y109 DNA processing -2.48 36 Peptide methionine sulfoxide reductase MsrB msrB A6T7R0 Stress response -2.27 37 6-phosphogluconate dehydrogenase, decarboxylating gnd P41576 Carbohydrate metabolism -2.24 38 50S ribosomal protein L17 rplQ A6TEU7 Protein biosynthesis -2.13 39 Cysteine desulfurase IscS iscS A6TCF1 Stress response -1.95 40 50S ribosomal protein L5 rplE A6TEW0 Protein biosynthesis -1.94 41 Uridine kinase udk A6TBG6 Pyrimidine biosynthesis -1.89 42 Probable septum site-determining protein MinC minC A6TAX3 Cell division -1.67 43 UDP-4-amino-4-deoxy-L-arabinose—oxoglutarate aminotransferase arnB A6TFA0 Lipopolysaccharide biosynthesis -1.62 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 9 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Table 1. (Continued) No Proteins Gene Name Uniprot Accession No. General Function Fold Change 90 Histidine biosynthesis bifunctional protein HisB hisB A6TBC5 Protein biosynthesis Control 91 Holliday junction ATP-dependent DNA helicase RuvA ruvA B5XQ04 Stress response Control 92 HTH-type transcriptional regulator CysB cysB P45600 Protein biosynthesis Control 93 HTH-type transcriptional repressor PurR purR A6TA06 Purine metabolism Control 94 Hydroxyethylthiazole kinase thiM A6TBJ8 Thiamine biosynthesis Control 95 Imidazole glycerol phosphate synthase subunit HisF hisF A6TBC8 Protein biosynthesis Control 96 Iron-binding protein IscA iscA A6TCE9 Stress response Control 97 Ketol-acid reductoisomerase (NADP(+)) ilvC A6TGG1 Protein biosynthesis Control 98 Large-conductance mechanosensitive channel mscL B5XNC0 Transport Control 99 L-lactate dehydrogenase lldD A6TFK0 Energy synthesis Control 100 L-threonine 3-dehydrogenase tdh A6TFL2 Protein biosynthesis Control https://doi.org/10.1371/journal.pone.0214326.t001 within the bacterial membrane were lost completely. This is yet another indicator for disrupted bacterial membrane integrity that may contribute to bacterial cell killing through the shut- down ofenergy production in CBO-treated KPC-KP cells. With compromised plasma mem- brane integrity, intracellular proteins easily escape into the extracellular environment as suggested by Ukuku et al. (2007) [25]. within the bacterial membrane were lost completely. This is yet another indicator for disrupted bacterial membrane integrity that may contribute to bacterial cell killing through the shut- down ofenergy production in CBO-treated KPC-KP cells. With compromised plasma mem- brane integrity, intracellular proteins easily escape into the extracellular environment as suggested by Ukuku et al. (2007) [25]. General Function Fold Change 44 S-adenosylmethionine synthase metK A6TDV1 Protein biosynthesis -1.58 45 Glutamate—tRNA ligase gltX A6TC43 Protein biosynthesis -1.49 46 Arginine—tRNA ligase argS A6TB43 Protein biosynthesis -1.39 47 HTH-type transcriptional regulator MalT malT A6TF41 Carbohydrate metabolism -1.28 48 Protein TolB tolB B5XZC1 Cell division -1.10 49 Acetyl-coenzyme A carboxylase carboxyl transferase subunit alpha accA A6T4Y7 Lipid metabolism -1.08 50 30S ribosomal protein S3 rpsC B5XNA0 Protein biosynthesis -1.04 Control KPC-KP Exclusive Proteins 51 10 kDa chaperonin grosA B5Y369 Stress response Control 52 1-deoxy-D-xylulose-5-phosphate synthase dxs A6T5F3 Thiamine biosynthesis Control 53 3-dehydroquinate synthase aroB A6TF13 Chorismate biosynthesis Control 54 3-methyl-2-oxobutanoate hydroxymethyltransferase panB A6T4T0 Pantothenate biosynthesis Control 55 3-octaprenyl-4-hydroxybenzoate carboxy-lyase ubiD A6TGM1 Ubiquinone biosynthesis Control 56 3-phosphoshikimate 1-carboxyvinyltransferase aroA B5XY87 Chorismate biosynthesis Control 57 4-hydroxy-3-methylbut-2-enyl diphosphate reductase ispH A6T4G3 Isoprenoid biosynthesis Control 58 50S ribosomal protein L34 rpmH A6TG05 Protein biosynthesis Control 59 6-phosphogluconolactonase pgl A6T6J6 Carbohydrate metabolism Control 60 Adenine phosphoribosyltransferase apt A6T5N2 Purine metabolism Control 61 Aminomethyltransferase gcvT A6TDR7 Protein biosynthesis Control 62 Argininosuccinate synthase argG A6TEJ0 Protein biosynthesis Control 63 ATP synthase epsilon chain atpC A6TG35 Energy synthesis Control 64 ATP-dependent RNA helicase RhlB rhlB A6TGG9 RNA processing and modification Control 65 Biosynthetic arginine decarboxylase speA B5XUB1 Amine and polyamine biosynthesis Control 66 Catalase-peroxidase katG A6T9H9 Stress response Control 67 Cell division protein ZipA zipA A6TC48 Cell division Control 68 Cell division topological specificity factor minE A6TAX5 Cell division Control 69 Chorismate synthase aroC A6TC15 Chorismate biosynthesis Control 70 Chromosomal replication initiator protein DnaA dnaA B5XT51 DNA processing Control 71 D-amino acid dehydrogenase dadA A6TAW4 Protein biosynthesis Control 72 Dihydroorotase pyrC A6T7D6 Pyrimidine biosynthesis Control 73 Dihydroorotate dehydrogenase (quinone) pyrD A6T739 Pyrimidine biosynthesis Control 74 Dihydroxy-acid dehydratase ilvD A6TGF8 Protein biosynthesis Control 75 DNA gyrase inhibitor YacG yacG A6T4P3 DNA processing Control 76 DNA mismatch repair protein MutS mutS A6TD24 Stress response Control 77 DnaA initiator-associating protein DiaA diaA A6TEG8 DNA processing Control 78 Electron transport complex subunit C rnfC B5XWP9 Energy synthesis Control 79 Elongation factor P—(R)-beta-lysine ligase epmA A6TH74 Protein biosynthesis Control 80 Endonuclease V nfi A6TGQ5 Stress response Control 81 Exodeoxyribonuclease 7 small subunit xseB B5Y0W9 DNA processing Control 82 Formate-dependent phosphoribosylglycinamide formyltransferase purT B5XQ21 Purine metabolism Control 83 Fructose-6-phosphate aldolase fsa A6TGD7 Carbohydrate metabolism Control 84 Gamma-glutamyl phosphate reductase proA A6T561 Protein biosynthesis Control 85 Glucans biosynthesis protein D mdoD B5XWS2 Glycan metabolism Control 86 Glucokinase glk A6TC33 Energy synthesis Control 87 Glucosamine-6-phosphate deaminase nagB A6T6C1 Carbohydrate metabolism Control 88 Glutamate 5-kinase proB A6T562 Protein biosynthesis Control 89 Glycerol-3-phosphate acyltransferase plsB A6TGV0 Phospholipid biosynthesis Control (Continued) PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 10 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Oxidative stress CBO contains a plethora of different chemical compounds which are dominated by a large group of oxygenated terpenes and terpenoids [10]. These compounds are hypothesized to be responsible for CBO’s action on the bacterial membrane. From our proteomic data, upregula- tion of proteins such as autonomous glycyl radical cofactor and catalase peroxidase in the CBO-treated KPC-KP cells suggest the presence of significant oxidative stress. Autonomous glycyl radical cofactor is upregulated by 7.07 fold, and acts as a radical domain which protects pyruvate formate lyase from oxidative stress. Wagner et al. (2001) and Shisler et al. (2014) found that upregulation of glycyl radical cofactor indicates oxidative stress which affects the pyruvate formate lyase that is involved in glucose metabolism [26, 27]. In addition, catalase peroxidase, an enzyme which alleviates oxidative stress from reactive oxygen species (ROS), was also induced when KPC-KP cells were exposed to CBO. Under condition of oxidative stress, high abundance ROS cause oxidative damage to nucleic acids [28]. The detection of the DNA mismatch repair protein MutS and the DNA ligase following the exposure to CBO showed that genetic materials of KPC-KP had been damaged, and that elevated expression of these proteins could alleviate some of the oxidative DNA damage. The work of Vogel et al. (2011) showed that oxidative stress leads to the degradation of ribosomal protein [29] and this appears to be in line with our findings in CBO-treated KPC-KP cells where 14 ribosomal pro- teins and ribosome related proteins showed decreased abundance. Specifically, 30S ribosomal protein S3; 50S ribosomal proteins L17, L32, L34 and L5; ribosomal protein L11 methyltrans- ferase; ribosomal RNA large subunit methyltransferase E, F and I; ribosomal RNA small sub- unit methyltransferase B and G; ribosome maturation factor M and P, and ribosome binding factor A were all reduced in abundance. As one of the key proteins in the maturation of 30S ribosomal subunit, ribosome maturation factor RimP protein is crucial in the process of pro- tein translation by allowing correct pairing of the mRNA and the corresponding anticodon of the tRNA [30, 31]. In agreement with this finding is the decrease in abundance of the other three 50S ribosomal subunit fragments namely, L5, L17 and L32. This further supports the PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 11 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Table 2. List of proteins belonging to the outer membrane and plasma membrane of KPC-KP cells and their relative status in proteomic profiling. Uniprot Accession No. Protein Status Outer membrane protein A6TGU6 Maltoporin 2 Lost P40786 Nucleoside-specific channel-forming protein tsx Lost P24017 Outer membrane protein A Upregulated by 2.04 folds A6T4X9 Outer membrane protein assembly factor BamA Lost A6T7G4 Penicillin-binding protein activator LpoB Lost P27218 Sucrose porin Lost Plasma membrane protein A6TGM1 3-octaprenyl-4-hydroxybenzoate carboxy-lyase Lost A6TG35 ATP synthase epsilon chain Lost B5XZ37 ATP-dependent protease ATPase subunit HslU Lost B5XUB1 Biosynthetic arginine decarboxylase Lost A6TC48 Cell division protein ZipA Lost A6TAX5 Cell division topological specificity factor Lost A6T4F4 Chaperone protein DnaK Upregulated by 1.73 folds B5XT51 Chromosomal replication initiator protein DnaA Lost A6TDC6 Cobalamin biosynthesis protein CobD Upregulated A6TAW4 D-amino acid dehydrogenase Lost A6T739 Dihydroorotate dehydrogenase (quinone) Lost B5XWP9 Electron transport complex subunit C Lost B5XWS2 Glucans biosynthesis protein D Lost A6TGV0 Glycerol-3-phosphate acyltransferase Lost A6TDR6 Glycine cleavage system H protein Lost B5XNC0 Large-conductance mechanosensitive channel Lost A6TFK0 L-lactate dehydrogenase Lost A6TEQ3 Malate dehydrogenase Lost B5XXP0 NAD(P)H dehydrogenase (quinone) Lost A6TBX3 NADH-quinone oxidoreductase subunit B Lost A6TBX2 NADH-quinone oxidoreductase subunit C/D Lost A6TC99 Phosphoribosylaminoimidazole-succinocarboxamide synthase Upregulated by 2.52 folds Q07411 Polyphosphate kinase Lost O32719 Probable malate:quinone oxidoreductase (Fragment) Upregulated A6TDG4 Prolipoprotein diacylglyceryl transferase Lost P27219 PTS system sucrose-specific EIIBC component Lost A6TC94 Succinyl-diaminopimelate desuccinylase Lost A6THZ7 Thymidine phosphorylase Lost A6TGL3 Ubiquinone/menaquinone biosynthesis C-methyltransferase UbiE Lost A6T4N3 UDP-N-acetylglucosamine—N-acetylmuramyl-(pentapeptide) pyrophosphoryl-undecaprenol N-acetylglucosamine transferase Lost refers to the proteins that were only present in CBO-treated KPC-KP cells. e outer membrane and plasma membrane of KPC-KP cells and their relative status in proteomic profiling. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 12 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption idea that induction of oxidative stress by CBO affects the abundance of ribosomal subunit frag- ments, as a result of the denaturation of protein fragments or the interruption of protein syn- thesis in KPC-KP cells. Cosentino et al. (2014) found that bergamot essential oil induced the production of ROS in polymorphonuclear leukocytes, contributing to enhanced eradication of infection [32]. Another study by Yoo et al (2005) also suggest that eugenol, one of the main constituents of essential oils such as from nutmeg and cinnamon bark induces the production of ROS in leukemia cells, eventually killing the cells by initiating apoptosis [33]. The induction of oxidative stress by essential oils seems to be in conflict with the perception that essential oils contain high concentration of antioxidants. We previously found that the CBO used in this study comprised 13 compounds, of which nine were antioxidants whereas the other four were not [10, 34–38]. These non-antioxidant compounds may be responsible for inducing the oxi- dative stress observed, by generating ROS or inhibiting anti-oxidizing systems [39]. Mimica- Dukić et al. (2016) suggested that essential oils may act both as antioxidants and also as pro- oxidant due to their complex chemistry [40]. Additionally, a single compound might exhibit dual antioxidant and prooxidant effects. For instance, Bezerra et al. (2017) found that eugenol had such dual effects, acting as an antioxidant in free-radical scavenging while inducing DNA damage via ROS generation [41]. So, the prooxidant activity within CBO may contribute to the oxidative stress, leading to lipid peroxidation in the cell membrane, while inhibiting the antioxidant activity, as indicated in the proteomic profile. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 KEGG pathway analysis When analysing significant pathways via the Panther classification system we found that the main pathways associated with the proteins identified included the lipid, cell wall and lipopoly- saccharide biosynthesis pathways. Lipid biosynthesis. Lipids have critical functions in the bacterial cytoplasmic and outer membranes in separating the bacterial cytoplasm from the external environment [42]. KEGG pathway analysis identified 4 proteins which are involved in the lipid biosynthesis pathway that were affected by CBO treatment. Two of these proteins were downregulated in CBO- treated KPC-KP cells: 3-hydroxydecanoyl-[acyl-carrier-protein] dehydratase (-3.46 fold) and acetyl-coenzyme A carboxylase carboxyl transferase subunit alpha (-1.08 fold). While two other proteins were undetectable following CBO treatment: glycerol-3-phosphate acyltransfer- ase and large-conductance mechanosensitive channel protein. The former two proteins are key components in the fatty acid biosynthesis pathway of Gram-negative bacteria. Acetyl- coenzyme A carboxylase carboxyl transferase is one of the major enzymes in the synthesis of malonyl-CoA, a substrate required in the synthesis of fatty acids [43]. 3-hydroxydecanoyl- [acyl-carrier-protein] dehydratase functions as an essential mediator in the synthesis of phos- pholipids in bacteria [44, 45]. As shown in Table 1, glycerol-3-phosphate acyltransferase (G3PAT) was undetectable in the CBO-treated KPC-KP cells. G3PAT, a rate determining enzyme belonging to the glycerolphospholipid metabolism pathway catalyzes the synthesis of phosphatidic acid from glycerol-3-phosphate and long-chain acyl-CoA, is an essential precur- sor in the synthesis of the bacterial phospholipid bilayer [46, 47]. The absence of G3PAT indi- cates a perturbed bacterial membrane repair system. This further indicates a disruption in the integrity of the phospholipid membrane in CBO-treated KPC-KP cells. The absence of the large-conductance mechanosensitive channel protein in the treated group also indicated a dis- rupted membrane structure as this channel regulates membrane stretching and stability under osmotic stress [48, 49]. Upon exposure to CBO, oxidative stress may promote protein denatur- ation and so affect bacterial cell membrane stretch capacity under osmotic pressure, eventually leading to increased CBO influx and bacterial cell killing. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 13 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Cell wall biosynthesis. Peptidoglycan in the cell wall is a major structural component in prokaryotic cells, forming a stable layer which protects the bacteria from lysis under osmotic stress [50]. The proteomic profile showed that three essential proteins in bacterial cell wall syn- thesis were lost upon exposure to CBO. KEGG pathway analysis This may indicate that CBO inhibited the expression of these proteins or that CBO-induced oxidative stress resulted in degradation of these proteins, so preventing cell wall synthesis and repair, and eventually causing cell death. These proteins included UDP-N-acetylglucosamine—N-acetylmuramyl-(pentapeptide) pyrophosphoryl-unde- caprenol N-acetylglucosamine transferase (murG); penicillin-binding protein activator LpoB and succinyl-diaminopimelate desuccinylase. The MurG protein is involved in the biosynthesis of the N-acetylmuramic acid-N-acetylglucosamine intermediate, which form a single unit of peptidoglycan cell wall. In the absence of murG, no N-acetylmuramic acid would not be linked to N-acetylglucosamine, preventing the formation of peptidoglycan bilayer linkage and eventu- ally killing the cell due to osmotic pressure and oxidative stress [50]. Additionally, succinyl-dia- minopimelate desuccinylase also play a major role in maintaining the structure and integrity of the peptidoglycan, as this protein is essential in the synthesis of meso-diaminopimelic acid which is one of the penta-peptides found linked to N-acetylmuramic acid, and is crucial for the cross-linkage with the opposite layer of N-acetylglucosamine. Interestingly, penicillin-binding protein activator LpoB which stimulates the biosynthesis of peptidoglycan molecules was not detected in CBO treated KPC-KP cells. These findings may indicate that the peptidoglycan bio- synthesis and repairing had been completely shut down [51]. Lipopolysaccharide biosynthesis. Lipopolysaccharide (LPS) is a highly acylated saccharo- lipid located on the outer layer of the outer membrane of Gram-negative bacteria. LPS is cru- cial in the maintenance of membrane integrity and has a barrier function which prevents the passive diffusion of hydrophobic solutes, such as antibiotics and detergents into the cell [52]. Many studies have postulated that CBO exerts its bactericidal activity through damage to the outer membrane which allows other bactericidal molecules to enter the cell, eventually killing the cells [11, 15, 16]. LPS consist of a few components including lipid A; core oligosaccharide and O-antigen [53]. We identified three proteins that are involved in LPS biosynthesis that became undetectable upon CBO exposure. These were O-antigen export system ATP-binding protein (RfbB); UDP-4-amino-4-deoxy-L-arabinose—oxoglutarate aminotransferase (arnB) and UDP-3-O-acyl-N-acetylglucosamine deacetylase (lpxC). In order to achieve resistance towards antibiotics such as polymyxin, the 4-amino-4-deoxy-L-arabinose moiety must be added to the lipid A [54]. This reaction is catalyzed by the arnB protein which is not detected in the KPC-KP cells exposed to CBO. Furthermore, lpxC protein, a critical enzyme which cata- lyzes the synthesis of lipid A was also undetectable in CBO-treated KPC-KP cells. KEGG pathway analysis 90% of the bacterial outer membrane consists of lipid A, which is crucial for the attachment of core oligo- saccharide as well as o-antigen, granting antibiotic resistance to bacteria while maintaining the membrane integrity. In the absence of the lpxC protein, the outer membrane integrity cannot be maintained, compromising cellular resistance against antibiotics. Due to the importance of this protein in lipid A biosynthesis, lpxC has been a target for the development of novel anti- microbial drugs [55, 56]. The O-antigen export system ATP-binding protein, which facilitates the export of O-antigen into the outer membrane was also not detected following exposure to CBO. Thus, CBO may contain compounds that inhibit the expression of these important membrane proteins and sensitize bacteria to antibiotics and changes in osmotic pressure. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 qRT-PCR analysis of differentially expressed proteins PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 15 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption CBO exerts antimicrobial activity on KPC-KP cells through membrane disruption, and proteomic profiling shows that the membrane damage induced was due to oxidative stress. This is supported by the increased in abundance of oxidative stress regulators when KPC-KP cells were exposed to CBO. A review by Itri et al. (2014) consolidated information from stud- ies involving oxidative stress and membrane damage. The study concluded that oxidative stress would reduce the permeability and integrity of the plasma membrane, leading to leak- age of intracellular contents and eventually killing the bacterial cells [57]. The membrane dis- ruptive effects of CBO could also be deduced from our proteomic profiles, and numerous proteins showed decreased abundance or were lost following CBO exposure. Interestingly, CBO treatment also interfered with the biosynthesis of the plasma membrane, cell wall and outer membrane, disabling the structural repair system. The proteomic profiles were vali- dated using qRT-PCR analysis, where proteomic changes were mirrored by changes in corre- sponding mRNA abundance. Together with the evidence from our previous study on the antimicrobial potential and the mode of action of CBO against KPC-KP cells [11], we showed that the antibacterial activity of CBO derives from its ability to induce oxidative stress in bacterial cells. The resulting oxidation disrupts the bacterial membrane, eventually enabling the influx of ROS into the cells and, at the same time, leads to intracellular content leakage. ROS induce genetic damage and impair DNA and membrane repair systems. Our results demonstrate that CBO causes oxidative stress, damage to bacterial membranes, cellu- lar leakage and cell killing. Supporting information S1 Attachment. Raw data of non-treated and CBO-treated KPC-KP proteome profile. (ZIP) S1 Dataset. Comparative proteomic analysis spreadsheet. (XLSX) S1 Dataset. Comparative proteomic analysis spreadsheet. (XLSX) S1 Dataset. Comparative proteomic analysis spreadsheet. (XLSX) S1 Table. List of primers used in this study. (DOCX) S1 Table. List of primers used in this study. (DOCX) qRT-PCR analysis of differentially expressed proteins Standard curve for qRT-PCR. We selected four genes for further validation of our prote- omic data using qRT-PCR together with two housekeeping genes, namely 3-hydroxydecanoyl- PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 14 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Fig 3. Expression patterns of fabA, cdd, thiE and udp genes in KPC-KP cells subjected to CBO treatment. Results are presented as differential relative transcript abundance. https://doi org/10 1371/journal pone 0214326 g003 Fig 3. Expression patterns of fabA, cdd, thiE and udp genes in KPC-KP cells subjected to CBO treatment. Results are presented as differential relative transcript abundance. Fig 3. Expression patterns of fabA, cdd, thiE and udp genes in KPC-KP cells subjected to CBO treatment. Results are presented as differential relative transcript abundance. https://doi.org/10.1371/journal.pone.0214326.g003 https://doi.org/10.1371/journal.pone.0214326.g003 https://doi.org/10.1371/journal.pone.0214326.g003 [acyl-carrier-protein] dehydratase (fabA), cytidine deaminase (cdd), thiamine phosphate synthase (thiE), uridine phosphorylase (udp), 16s rRNA and OmpK36 porin. Out of these, three were significantly higher in abundance and one was less abundant in the proteomic profile fol- lowing exposure to CBO. The designed primers for each selected genes were listed in S1 Table. Their efficiency ranged within 90 and 100%. y g Relative expression level of selected genes. The relative abundance of mRNA for the differentially expressed proteins investigated was measured by qRT-PCR, comparing the untreated cells with CBO-treated KPC-KP cells (Fig 3). Proteomic profiling found that fabA was downregulated by -3.46 fold, whereas cdd, thiC and udp were upregulated by 1.62, 2.08 and 2.57 fold respectively (Table 1). The qRT-PCR analysis of mRNA abundance found that fabA was downregulated by 0.8 fold whereas cdd, thiC and udp were upregulated by 0.8, 11 and 1.4 fold respectively. The pattern in mRNA abundance change following CBO treatment mirrors that of the protein abundance for these four gene products, indicating that at least part of the change in the abundance for these four proteins is due to changes in gene transcription. The differences in the magnitude of the fold changes between the protein abundance and mRNA transcript abundance may be a consequence of oxidative stress-induced protein degra- dation or different stabilities for the mRNA species. Nevertheless, the trends in expression changes shown in the protein profiles and mRNA transcription levels are consistent with each other and validate our results obtained from proteomic profiling. Acknowledgments The authors like to thank all members of the Floral Biotechnology Laboratory, Universiti Putra Malaysia. All authors read and approved the final manuscript. Additionally, the authors would also like to thank Malaysia Genome Institute (MGI) for providing the proteomic facili- ties throughout this study. References 1. Durdu B, Hakyemez IN, Bolukcu S, Okay G, Gultepe B, Aslan T. Mortality markers in nosocomial Kleb- siella pneumoniae bloodstream infection. Springerplus. 2016; 5(1): 1892. https://doi.org/10.1186/ s40064-016-3580-8 PMID: 27843749 2. Paterson DL, Bonomo RA. Extended-spectrum beta-lactamases: a clinical update. Clin Microbiol Rev. 2005; 18(4): 657–686. https://doi.org/10.1128/CMR.18.4.657-686.2005 PMID: 16223952 3. Ur Rahman S, Ali T, Ali I, Khan NA, Han B, Gao J. The Growing genetic and functional diversity of extended spectrum beta-lactamases. Biomed Res Int. 2018; 2018:9519718. https://doi.org/10.1155/ 2018/9519718 PMID: 29780833 4. Rhomberg PR, Jones RN. Summary trends for the Meropenem Yearly Susceptibility Test Information Collection Program: a 10-year experience in the United States (1999–2008). Diagn Microbiol Infect Dis. 2009; 65(4): 414–426. https://doi.org/10.1016/j.diagmicrobio.2009.08.020 PMID: 19833471 5. Yigit H, Queenan AM, Anderson GJ, Domenech-Sanchez A, Biddle JW, Steward CD, et al. Novel car- bapenem-hydrolyzing beta-lactamase, KPC-1, from a carbapenem-resistant strain of Klebsiella pneu- moniae. Antimicrob Agents Chemother. 2001; 45(4): 1151–1161. https://doi.org/10.1128/AAC.45.4. 1151-1161.2001 PMID: 11257029 6. Kitchel B, Rasheed JK, Patel JB, Srinivasan A, Navon-Venezia S, Carmeli Y, et al. Molecular epidemiol- ogy of KPC-producing Klebsiella pneumoniae isolates in the United States: clonal expansion of multilo- cus sequence type 258. Antimicrob Agents Chemother. 2009; 53(8): 3365–3370. https://doi.org/10. 1128/AAC.00126-09 PMID: 19506063 7. Yang SK, Low LY, Yap PSX, Yusoff, Mai CW, Lai KS, et al. Plant-derived antimicrobials: insights into mitigation of antimicrobial resistance. Rec Nat Prod. 2018; 12(4): 295–316. 8. Yap PSX, Yiap BC, Ping HC, Lim SHE. Essential oils, a new horizon in combating bacterial antibiotic resistance. Open Microbiol J. 2014; 8: 6–14. https://doi.org/10.2174/1874285801408010006 PMID: 24627729 9. Lorenzi V, Muselli A, Bernardini AF, Berti L, Pages JM, Amaral L, et al. Geraniol restores antibiotic activ- ities against multidrug-resistant isolates from gram-negative species. Antimicrob Agents Chemother. 2009; 53(5): 2209–2211. https://doi.org/10.1128/AAC.00919-08 PMID: 19258278 10. Yap PSX, Krishnan T, Chan KG, Lim SHE. Antibacterial mode of action of Cinnamomum verum bark essential oil, alone and in combination with piperacillin, against a multidrug-resistant Escherichia coli strain. J Microbiol Biotechnol. 2015; 25(8): 1299–1306. https://doi.org/10.4014/jmb.1407.07054 PMID: 25381741 11. Yang SK, Yusoff K, Mai CW, Lim WM, Yap WS, Lim SHE, et al. Additivity vs synergism: investigation of the additive interaction of cinnamon bark oil and meropenem in combinatory therapy. Molecules. 2017; 22(11). 12. Yap PSX, Lim SHE, Hu CP, Yiap BC. Combination of essential oils and antibiotics reduce antibiotic resistance in plasmid-conferred multidrug resistant bacteria. Phytomedicine. 2013; 20(8–9): 710–713. https://doi.org/10.1016/j.phymed.2013.02.013 PMID: 23537749 13. Resources: Warren Thomas, Aisha Abushelaibi, Riaz Akseer, Swee-Hua Erin Lim, Kok-Song Lai. Supervision: Khatijah Yusoff, Mokrish Ajat, Swee-Hua Erin Lim, Kok-Song Lai. Visualization: Shun-Kai Yang. Visualization: Shun-Kai Yang. Writing – original draft: Shun-Kai Yang. Writing – review & editing: Khatijah Yusoff, Mokrish Ajat, Warren Thomas, Aisha Abushe- laibi, Riaz Akseer, Swee-Hua Erin Lim, Kok-Song Lai. Author Contributions Conceptualization: Swee-Hua Erin Lim, Kok-Song Lai. Conceptualization: Swee-Hua Erin Lim, Kok-Song Lai. C p S , S g Data curation: Shun-Kai Yang, Kok-Song Lai. Formal analysis: Shun-Kai Yang. Funding acquisition: Aisha Abushelaibi, Kok-Song Lai. Investigation: Shun-Kai Yang, Swee-Hua Erin Lim, Kok-Song Lai. Methodology: Shun-Kai Yang, Swee-Hua Erin Lim, Kok-Song Lai. Project administration: Swee-Hua Erin Lim, Kok-Song Lai. Data curation: Shun-Kai Yang, Kok-Song Lai. Data curation: Shun-Kai Yang, Kok-Song Lai. Formal analysis: Shun-Kai Yang. Funding acquisition: Aisha Abushelaibi, Kok-Song Lai. Investigation: Shun-Kai Yang, Swee-Hua Erin Lim, Kok-Song Lai. Methodology: Shun-Kai Yang, Swee-Hua Erin Lim, Kok-Song Lai. Project administration: Swee Hua Erin Lim Kok Song Lai Project administration: Swee-Hua Erin Lim, Kok-Song Lai. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 16 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption Resources: Warren Thomas, Aisha Abushelaibi, Riaz Akseer, Swee-Hua Erin Lim, Kok-Song Lai. Software: Shun-Kai Yang. Supervision: Khatijah Yusoff, Mokrish Ajat, Swee-Hua Erin Lim, Kok-Song Lai. Visualization: Shun-Kai Yang. Writing – original draft: Shun-Kai Yang. Writing – review & editing: Khatijah Yusoff, Mokrish Ajat, Warren Thomas, Aisha Abushe- laibi, Riaz Akseer, Swee-Hua Erin Lim, Kok-Song Lai. Resources: Warren Thomas, Aisha Abushelaibi, Riaz Akseer, Swee-Hua Erin Lim, Kok-Song Lai. References Yap PSX, Yang SK, Lai KS, Lim SHE. Essential oils: the ultimate solution to antimicrobial resistance in Escherichia coli. 2017. 14. Yang SK, Yap PSX, Krishnan T, Yusoff K, Chan KG, Yap WS, et al. Mode of action: synergistic interac- tion of peppermint (Mentha x piperita L. Carl) essential oil and meropenem against plasmid-mediated resistant E. coli. Rec Nat Prod. 2018; 12(6): 13. 15. de Souza EL, de Barros JC, de Oliveira CE, da Conceicao ML. Influence of Origanum vulgare L. essen- tial oil on enterotoxin production, membrane permeability and surface characteristics of Staphylococcus 17 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 Cinnamon bark essential oil and its role in bacterial membrane disruption aureus. Int J Food Microbiol. 2010; 137(2–3): 308–311. https://doi.org/10.1016/j.ijfoodmicro.2009.11. 025 PMID: 20015563 aureus. Int J Food Microbiol. 2010; 137(2–3): 308–311. https://doi.org/10.1016/j.ijfoodmicro.2009.11. 025 PMID: 20015563 16. Silva F, Ferreira S, Queiroz JA, Domingues FC. Coriander (Coriandrum sativum L.) essential oil: its anti- bacterial activity and mode of action evaluated by flow cytometry. J Med Microbiol. 2011; 60: 1479– 1486. https://doi.org/10.1099/jmm.0.034157-0 PMID: 21862758 17. Turgis M, Han J, Caillet S, Lacroix M. Antimicrobial activity of mustard essential oil against Escherichia coli O157:H7 and Salmonella typhi. Food Control. 2009; 20(12): 1073–1079. 18. Xu H, Dephoure N, Sun H, Zhang H, Fan F, Liu J, et al. Proteomic profiling of paclitaxel treated cells identifies a novel mechanism of drug resistance mediated by PDCD4. J Proteome Res. 2015; 14(6): 2480–2491. https://doi.org/10.1021/acs.jproteome.5b00004 PMID: 25928036 19. Kawatani M, Muroi M, Wada A, Inoue G, Futamura Y, Aono H, et al. Proteomic profiling reveals that col- lismycin A is an iron chelator. Scientific Reports. 2016; 6: 38385. https://doi.org/10.1038/srep38385 PMID: 27922079 20. Zainal-Abidin Z, Mohd-Said S, Adibah F, Majid A, Mustapha WAW, Jantan I. Anti-bacterial activity of cinnamon oil on oral pathogens. The Open Conference Proceedings Journal. 2013; 4: 5. 21. Kaskatepe B, Kiymaci ME, Simsek D, Erol HB, Erdem SA. Comparison of the contents and antimicro- bial activities of commercial and natural cinnamon oils. Indian J Pharm Sci. 2016; 78(4): 8. 22. Silhavy TJ, Kahne D, Walker S. The bacterial cell envelope. Cold Spring Harb Perspect Biol. 2010; 2(5): a000414. https://doi.org/10.1101/cshperspect.a000414 PMID: 20452953 23. Wu Y, Bai J, Zhong K, Huang Y, Qi H, Jiang Y, et al. Antibacterial activity and membrane-disruptive mechanism of 3-p-trans-coumaroyl-2-hydroxyquinic acid, a novel phenolic compound from pine nee- dles of Cedrus deodara, against Staphylococcus aureus. Molecules. 2016; 21(8). References 24. Gerding MA, Ogata Y, Pecora ND, Niki H, de Boer PA. The trans-envelope Tol-Pal complex is part of the cell division machinery and required for proper outer-membrane invagination during cell constriction in E. coli. Mol Microbiol. 2007; 63(4): 1008–1025. https://doi.org/10.1111/j.1365-2958.2006.05571.x PMID: 17233825 25. Ukuku DO, Geveke DJ, Cooke P, Zhang HQ. Membrane damage and viability loss of Escherichia coli K-12 in apple juice treated with radio frequency electric field. J Food Prot. 2008; 71(4): 684–690. PMID: 18468020 26. Shisler KA, Broderick JB. Glycyl radical activating enzymes: structure, mechanism, and substrate inter- actions. Arch Biochem Biophys. 2014; 546: 64–71. https://doi.org/10.1016/j.abb.2014.01.020 PMID: 24486374 27. Wagner AF, Schultz S, Bomke J, Pils T, Lehmann WD, Knappe J. YfiD of Escherichia coli and Y06I of bacteriophage T4 as autonomous glycyl radical cofactors reconstituting the catalytic center of oxygen- fragmented pyruvate formate-lyase. Biochem Biophys Res Commun. 2001; 285(2): 456–462. https:// doi.org/10.1006/bbrc.2001.5186 PMID: 11444864 28. Willi J, Kupfer P, Evequoz D, Fernandez G, Katz A, Leumann C, et al. Oxidative stress damages rRNA inside the ribosome and differentially affects the catalytic center. Nucleic Acids Res. 2018; 46(4): 1945– 1957. https://doi.org/10.1093/nar/gkx1308 PMID: 29309687 29. Vogel C, Silva GM, Marcotte EM. Protein expression regulation under oxidative stress. Mol Cell Proteo- mics. 2011; 10(12): M111 009217. https://doi.org/10.1074/mcp.M111.009217 PMID: 21933953 30. Bunner AE, Nord S, Wikstrom PM, Williamson JR. The effect of ribosome assembly cofactors on in vitro 30S subunit reconstitution. J Mol Biol. 2010; 398(1): 1–7. https://doi.org/10.1016/j.jmb.2010.02.036 PMID: 20188109 31. Nord S, Bylund GO, Lovgren JM, Wikstrom PM. The RimP protein is important for maturation of the 30S ribosomal subunit. J Mol Biol. 2009; 386(3): 742–753. https://doi.org/10.1016/j.jmb.2008.12.076 PMID: 19150615 32. Cosentino M, Luini A, Bombelli R, Corasaniti MT, Bagetta G, Marino F. The essential oil of bergamot stimulates reactive oxygen species production in human polymorphonuclear leukocytes. Phytother Res. 2014; 28(8): 1232–1239. https://doi.org/10.1002/ptr.5121 PMID: 24458921 33. Yoo CB, Han KT, Cho KS, Ha J, Park HJ, Nam JH, et al. Eugenol isolated from the essential oil of Euge- nia caryophyllata induces a reactive oxygen species-mediated apoptosis in HL-60 human promyelocytic leukemia cells. Cancer Lett. 2005; 225(1): 41–52. https://doi.org/10.1016/j.canlet.2004.11.018 PMID: 15922856 34. Carvalho RL, Cabral MF, Germano TA, de Carvalho WM, Brasil IM, Gallão MI, et al. Chitosan coating with trans-cinnamaldehyde improves structural integrity and antioxidant metabolism of fresh-cut melon. Postharvest Biol Tech. 2016; 113: 29–39. 35. Ciftci O, Ozdemir I, Tanyildizi S, Yildiz S, Oguzturk H. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 References Antioxidative effects of curcumin, beta-myr- cene and 1,8-cineole against 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced oxidative stress in rats PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 18 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption liver. Toxicol Ind Health. 2011; 27(5): 447–453. https://doi.org/10.1177/0748233710388452 PMID: 21245202 liver. Toxicol Ind Health. 2011; 27(5): 447–453. https://doi.org/10.1177/0748233710388452 PMID: 21245202 36. Bubols GB, Vianna Dda R, Medina-Remon A, von Poser G, Lamuela-Raventos RM, Eifler-Lima VL, et al. The antioxidant activity of coumarins and flavonoids. Mini Rev Med Chem. 2013; 13(3): 318–334. PMID: 22876957 37. Dahham SS, Tabana YM, Iqbal MA, Ahamed MB, Ezzat MO, Majid AS, et al. The anticancer, antioxi- dant and antimicrobial properties of the sesquiterpene beta-caryophyllene from the essential oil of Aqui- laria crassna. Molecules. 2015; 20(7): 11808–11829. https://doi.org/10.3390/molecules200711808 PMID: 26132906 38. Pavan B, Dalpiaz A, Marani L, Beggiato S, Ferraro L, Canistro D, et al. Geraniol pharmacokinetics, bio- availability and its multiple effects on the liver antioxidant and xenobiotic-metabolizing enzymes. Front Pharmacol. 2018; 9: 18. https://doi.org/10.3389/fphar.2018.00018 PMID: 29422862 39. Rahal A, Kumar A, Singh V, Yadav B, Tiwari R, Chakraborty S, et al. Oxidative stress, prooxidants, and antioxidants: the interplay. Biomed Res Int. 2014; 2014: 761264. https://doi.org/10.1155/2014/761264 PMID: 24587990 40. Mimica-Dukić N, Orčić D, Lesjak M, Sˇ ibul F. Essential oils as powerful antioxidants: misconception or scientific fact? Medicinal and aromatic crops: production, phytochemistry, and utilization. ACS Sympo- sium Series. 1218: American Chemical Society; 2016. p. 187–208. 41. Bezerra DP, Militao GCG, de Morais MC, de Sousa DP. The dual antioxidant/prooxidant effect of euge- nol and its action in cancer development and treatment. Nutrients. 2017; 9(12). 42. Cronan JE, Thomas J. Bacterial fatty acid synthesis and its relationships with polyketide synthetic path- ways. Methods Enzymol. 2009; 459: 395–433. https://doi.org/10.1016/S0076-6879(09)04617-5 PMID: 19362649 43. Broussard TC, Price AE, Laborde SM, Waldrop GL. Complex formation and regulation of Escherichia coli acetyl-CoA carboxylase. Biochemistry. 2013; 52(19): 3346–3357. https://doi.org/10.1021/ bi4000707 PMID: 23594205 44. Emiola A, Andrews SS, Heller C, George J. Crosstalk between the lipopolysaccharide and phospholipid pathways during outer membrane biogenesis in Escherichia coli. Proc Natl Acad Sci USA. 2016; 113 (11): 3108–3113. https://doi.org/10.1073/pnas.1521168113 PMID: 26929331 45. Moynie´ L, Leckie SM, McMahon SA, Duthie FG, Koehnke A, Taylor JW, et al. Structural insights into the mechanism and inhibition of the β-hydroxydecanoyl-acyl carrier protein dehydratase from Pseudo- monas aeruginosa. J Mol Biol. 2013; 425(2): 365–377. https://doi.org/10.1016/j.jmb.2012.11.017 PMID: 23174186 46. Wendel AA, Lewin TM, Coleman RA. PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 55. Lee CJ, Liang X, Gopalaswamy R, Najeeb J, Ark ED, Toone EJ, et al. Structural basis of the promiscu- ous inhibitor susceptibility of Escherichia coli LpxC. ACS Chem Biol. 2014; 9(1): 237–246. https://doi. org/10.1021/cb400067g PMID: 24117400 56. Barb AW, Zhou P. Mechanism and inhibition of LpxC: an essential zinc-dependent deacetylase of bac- terial lipid A synthesis. Curr Pharm Biotechnol. 2008; 9(1): 9–15. PMID: 18289052 57. Itri R, Junqueira HC, Mertins O, Baptista MS. Membrane changes under oxidative stress: the impact of oxidized lipids. Biophys Rev. 2014; 6(1): 47–61. https://doi.org/10.1007/s12551-013-0128-9 PMID: 28509959 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 References Glycerol-3-phosphate acyltransferases: rate limiting enzymes of triacylglycerol biosynthesis. Biochim Biophys Acta. 2009; 1791(6): 501–506. https://doi.org/10.1016/j. bbalip.2008.10.010 PMID: 19038363 47. Yao J, Rock CO. Phosphatidic acid synthesis in bacteria. Biochim Biophys Acta. 2013; 1831(3): 495– 502. https://doi.org/10.1016/j.bbalip.2012.08.018 PMID: 22981714 48. Birkner JP, Poolman B, Kocer A. Hydrophobic gating of mechanosensitive channel of large conduc- tance evidenced by single-subunit resolution. Proc Natl Acad Sci USA. 2012; 109(32): 12944–12949. https://doi.org/10.1073/pnas.1205270109 PMID: 22826215 49. Blount P, Iscla I, Moe PC, Li Y. MscL: The bacterial mechanosensitive channel of large conductance. Curr Top Membr. 58: Academic Press; 2007. p. 201–33. 50. Lovering AL, Safadi SS, Strynadka NC. Structural perspective of peptidoglycan biosynthesis and assembly. Annu Rev Biochem. 2012; 81: 451–478. https://doi.org/10.1146/annurev-biochem-061809- 112742 PMID: 22663080 51. Egan AJ, Jean NL, Koumoutsi A, Bougault CM, Biboy J, Sassine J, et al. Outer-membrane lipoprotein LpoB spans the periplasm to stimulate the peptidoglycan synthase PBP1B. Proc Natl Acad Sci USA. 2014; 111(22): 8197–8202. https://doi.org/10.1073/pnas.1400376111 PMID: 24821816 52. Zhang G, Meredith TC, Kahne D. On the essentiality of lipopolysaccharide to Gram-negative bacteria. Curr Opin Microbiol. 2013; 16(6): 779–785. https://doi.org/10.1016/j.mib.2013.09.007 PMID: 24148302 53. Rosenfeld Y, Shai Y. Lipopolysaccharide (endotoxin)-host defense antibacterial peptides interactions: role in bacterial resistance and prevention of sepsis. Biochim Biophys Acta. 2006; 1758(9): 1513–1522. https://doi.org/10.1016/j.bbamem.2006.05.017 PMID: 16854372 54. Breazeale SD, Ribeiro AA, Raetz CR. Origin of lipid A species modified with 4-amino-4-deoxy-L-arabi- nose in polymyxin-resistant mutants of Escherichia coli. An aminotransferase (ArnB) that generates UDP-4-deoxyl-L-arabinose. J Biol Chem. 2003; 278(27): 24731–24739. https://doi.org/10.1074/jbc. M304043200 PMID: 12704196 PLOS ONE \| https://doi.org/10.1371/journal.pone.0214326 April 2, 2019 19 / 20 Cinnamon bark essential oil and its role in bacterial membrane disruption 55. Lee CJ, Liang X, Gopalaswamy R, Najeeb J, Ark ED, Toone EJ, et al. Structural basis of the promiscu- ous inhibitor susceptibility of Escherichia coli LpxC. ACS Chem Biol. 2014; 9(1): 237–246. https://doi. org/10.1021/cb400067g PMID: 24117400 56. Barb AW, Zhou P. Mechanism and inhibition of LpxC: an essential zinc-dependent deacetylase of bac- terial lipid A synthesis. Curr Pharm Biotechnol. 2008; 9(1): 9–15. PMID: 18289052 57. Itri R, Junqueira HC, Mertins O, Baptista MS. Membrane changes under oxidative stress: the impact of oxidized lipids. Biophys Rev. 2014; 6(1): 47–61. https://doi.org/10.1007/s12551-013-0128-9 PMID: 28509959 20 / 20
https://openalex.org/W4362627830	https://figshare.com/articles/journal_contribution/Figure_S4_from_Cell_of_Origin_Influences_Pancreatic_Cancer_Subtype/22535627/1/files/39998903.pdf	English	null	Figure S1 from Cell of Origin Influences Pancreatic Cancer Subtype	null	2,023	cc-by	842	C. tary Figure 4: Transcriptome and histological analysis of mouse acinar and ductal cell-derived tumors mmary of the tumors analyzed by RNA-Seq. Days Post-Tamoxifen refers to the time after the first day of mouse tamoxifen treatment, when mice succumbed to pancreatic mary tumor grade (comprising >50% of the tumor) was called as moderately/well-differentiated adenocarcinoma or poorly-differentiated adenocarcinoma. (B) Representative s of acinar cell-derived and ductal cell-derived tumors used for RNA-Seq classified as poorly and moderately/well-differentiated adenocarcinomas (C) Comparison of normal cinar and ductal cell signature (38) enrichment scores in acinar and ductal cell-derived tumors based on a two-tailed Student’s t-test (Not Significant). gin Genotype Days Post- Tamoxifen Primary Tumor Grade Differentiation KT;Ptf1aCreER;Trp53fl/fl 119 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 120 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 128 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 134 Poor KT;Sox9CreER;Trp53fl/fl 195 Poor KT;Sox9CreER;Trp53fl/fl 215 Moderate/Well KT;Sox9CreER;Trp53fl/fl 218 Moderate/Well Poorly- Differentiated Moderately/Well- Differentiated B. -0.4 -0.2 0.0 0.2 0.4 Normal Ductal Cell Signature Normal Acinar Cell Signature Signature Enrichment Score Normal Pancreas Signature Acinar cell-derived tumor Ductal cell-derived tumor Acinar cell- derived tumor Ductal cell- derived tumor C. tary Figure 4: Transcriptome and histological analysis of mouse acinar and ductal cell-derived tumors mmary of the tumors analyzed by RNA-Seq. Days Post-Tamoxifen refers to the time after the first day of mouse tamoxifen treatment, when mice succumbed to pancreatic mary tumor grade (comprising >50% of the tumor) was called as moderately/well-differentiated adenocarcinoma or poorly-differentiated adenocarcinoma. (B) Representative s of acinar cell-derived and ductal cell-derived tumors used for RNA-Seq classified as poorly and moderately/well-differentiated adenocarcinomas (C) Comparison of normal cinar and ductal cell signature (38) enrichment scores in acinar and ductal cell-derived tumors based on a two-tailed Student’s t-test (Not Significant). gin Genotype Days Post- Tamoxifen Primary Tumor Grade Differentiation KT;Ptf1aCreER;Trp53fl/fl 119 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 120 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 128 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 134 Poor KT;Sox9CreER;Trp53fl/fl 195 Poor KT;Sox9CreER;Trp53fl/fl 215 Moderate/Well KT;Sox9CreER;Trp53fl/fl 218 Moderate/Well Poorly- Differentiated Moderately/Well- Differentiated B. -0.4 -0.2 0.0 0.2 0.4 Normal Ductal Cell Signature Normal Acinar Cell Signature Signature Enrichment Score Normal Pancreas Signature Acinar cell-derived tumor Ductal cell-derived tumor Acinar cell- derived tumor Ductal cell- derived tumor C. Poorly- Differentiated Moderately/Well- Differentiated B. -0.4 - Normal Ductal Cell Signature Normal Acinar Cell Signature Signat Normal A Acinar cell- derived tumor Ductal cell- derived tumor C. -0.4 -0.2 0.0 0.2 0.4 Normal Ductal Cell Signature Normal Acinar Cell Signature Signature Enrichment Score Normal Pancreas Signature Acinar cell-derived tumor Ductal cell-derived tumor C Poorly- Differentiated Moderately/Well- Differentiated B. Acinar cell- derived tumor Ductal cell- derived tumor A. B. C. Moderately/Well- Differentiated -0.4 -0.2 0.0 0.2 0.4 Normal Ductal Cell Signature Normal Acinar Cell Signature Normal Pancreas Signature Normal Pancreas Signature Cell-of-Origin Genotype Days Post- Tamoxifen Primary Tumor Grade Differentiation Acinar KT;Ptf1aCreER;Trp53fl/fl 119 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 120 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 128 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 134 Poor Ductal KT;Sox9CreER;Trp53fl/fl 195 Poor KT;Sox9CreER;Trp53fl/fl 215 Moderate/Well KT;Sox9CreER;Trp53fl/fl 218 Moderate/Well Normal Ductal Cell Signature Signature Enrichment Score gure 4: Transcriptome and histological analysis of mouse acinar and ductal cell-derived tumors y of the tumors analyzed by RNA-Seq. Days Post-Tamoxifen refers to the time after the first day of mouse tamoxifen treatment, when mice succumbed to pancreatic mor grade (comprising >50% of the tumor) was called as moderately/well-differentiated adenocarcinoma or poorly-differentiated adenocarcinoma. (B) Representative nar cell-derived and ductal cell-derived tumors used for RNA-Seq classified as poorly and moderately/well-differentiated adenocarcinomas (C) Comparison of normal and ductal cell signature (38) enrichment scores in acinar and ductal cell-derived tumors based on a two-tailed Student’s t-test (Not Significant). C. tary Figure 4: Transcriptome and histological analysis of mouse acinar and ductal cell-derived tumors mmary of the tumors analyzed by RNA-Seq. Days Post-Tamoxifen refers to the time after the first day of mouse tamoxifen treatment, when mice succumbed to pancreatic mary tumor grade (comprising >50% of the tumor) was called as moderately/well-differentiated adenocarcinoma or poorly-differentiated adenocarcinoma. (B) Representative s of acinar cell-derived and ductal cell-derived tumors used for RNA-Seq classified as poorly and moderately/well-differentiated adenocarcinomas (C) Comparison of normal cinar and ductal cell signature (38) enrichment scores in acinar and ductal cell-derived tumors based on a two-tailed Student’s t-test (Not Significant). gin Genotype Days Post- Tamoxifen Primary Tumor Grade Differentiation KT;Ptf1aCreER;Trp53fl/fl 119 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 120 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 128 Moderate/Well KT;Ptf1aCreER;Trp53fl/fl 134 Poor KT;Sox9CreER;Trp53fl/fl 195 Poor KT;Sox9CreER;Trp53fl/fl 215 Moderate/Well KT;Sox9CreER;Trp53fl/fl 218 Moderate/Well Poorly- Differentiated Moderately/Well- Differentiated B. -0.4 -0.2 0.0 0.2 0.4 Normal Ductal Cell Signature Normal Acinar Cell Signature Signature Enrichment Score Normal Pancreas Signature Acinar cell-derived tumor Ductal cell-derived tumor Acinar cell- derived tumor Ductal cell- derived tumor Supplementary Figure 4: Transcriptome and histological analysis of mouse acinar and ductal cell-derived tumors (A) Table summary of the tumors analyzed by RNA-Seq. Days Post-Tamoxifen refers to the time after the first day of mouse tamoxifen treatment, when mice succumbed to pancreati tumors. Primary tumor grade (comprising >50% of the tumor) was called as moderately/well-differentiated adenocarcinoma or poorly-differentiated adenocarcinoma. (B) Representative H&E images of acinar cell-derived and ductal cell-derived tumors used for RNA-Seq classified as poorly and moderately/well-differentiated adenocarcinomas (C) Comparison of norma pancreatic acinar and ductal cell signature (38) enrichment scores in acinar and ductal cell-derived tumors based on a two-tailed Student’s t-test (Not Significant).
https://openalex.org/W4240771670	https://peerj.com/articles/5549v0.2/submission	English	null	Peer Review #2 of "Influence of mutation bias and hydrophobicity on the substitution rates and sequence entropies of protein evolution (v0.2)"	null	2,018	cc-by	14,608	Influence of mutation bias and hydrophobicity on the substitution rates and sequence entropies of protein l ti María José Jiménez-Santos 1 , Miguel Arenas 2 , Ugo Bastolla Corresp. 1 1 Bioinformatics Unit, Center for Molecular Biology Severo Ochoa, CSIC-UAM, Madrid, Spain 2 Department of Biochemistry, Genetics and Immunology, University of Vigo, Vigo, Spain Corresponding Author: Ugo Bastolla Email address: ubastolla@cbm.csic.es 1 Bioinformatics Unit, Center for Molecular Biology Severo Ochoa, CSIC-UAM, Madrid, Spain 2 Department of Biochemistry, Genetics and Immunology, University of Vigo, Vigo, Spain Corresponding Author: Ugo Bastolla Email address: ubastolla@cbm.csic.es The number of amino acids that occupy a given protein site during evolution reflects the selective constraints operating on the site. This evolutionary variability is strongly influenced by the structural properties of the site in the native structure, and it is quantified either through sequence entropy or through substitution rates. However, while the sequence entropy only depends on the equilibrium frequencies of the amino acids, the substitution rate also depends on the exchangeability matrix that describes mutations in the mathematical model of the substitution process. Here we apply two variants of a mathematical model of protein evolution with selection for protein stability, both against unfolding and against misfolding. Exploiting the approximation of independent sites, these models allow computing site-specific substitution processes that satisfy global constraints on folding stability. We find that site-specific substitution rates do not depend only on the selective constraints acting on the site, quantified through its sequence entropy. In fact, polar sites evolve faster than hydrophobic sites even for equal sequence entropy, as a consequence of the fact that polar amino acids are characterized by higher mutational exchangeability than hydrophobic ones. Accordingly, the model predicts that more polar proteins tend to evolve faster. Nevertheless, these results change if we compare proteins that evolve under different mutation biases, such as orthologous proteins in different bacterial genomes. In this case, the substitution rates are faster in genomes that evolve under mutational bias that favor hydrophobic amino acids by preferentially incorporating the nucleotide Thymine that is more frequent in hydrophobic codons. This appearingly contradictory result arises because buried sites occupied by hydrophobic amino acids are characterized by larger selective factors that largely amplify the substitution rate between hydrophobic amino acids, while the selective factors of exposed sites have a weaker effect. [p]Thus, changes of the mutational bias produce deep effects on the biophysical properties of the protein (hydrophobicity) and on its evolutionary properties (sequence entropy and substitution rate) at the same time. Manuscript to be reviewed Inﬂuence of mutation bias and hydrophobicity on the 1 substitution rates and sequence entropies of protein 2 evolution 3 Mar´ıa Jos´e Jim´enez-Santos(1), Miguel Arenas(2), and Ugo Bastolla(1,3) (1) Centro de Biologia Molecular ”Severo Ochoa” (CSIC-UAM) Madrid, Spain Mar´ıa Jos´e Jim´enez-Santos(1), Miguel Arenas(2), and Ugo Bastolla(1,3) (1) Centro de Biologia Molecular ”Severo Ochoa” (CSIC-UAM) Madrid, Spain (2) Department of Biochemistry, Genetics and Immunology, University of Vigo, Vigo, Spain (3) E-mail: ubastolla@cbm.csic.es (2) Department of Biochemistry, Genetics and Immunology, University of Vigo, Vigo, Spain (3) E-mail: ubastolla@cbm.csic.es Influence of mutation bias and hydrophobicity on the substitution rates and sequence entropies of protein l ti The program Prot_evol that implements the two site-specific substitution processes is freely available at https://ub.cbm.uam.es/prot_fold_evol/prot_fold_evol_soft_main.php#Prot_Evol.[p] PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Introduction 10 The evolutionary variability of an amino acid site in a protein family is an important 11 indicator of the selective constraints that the site experiences. This variability is usu- 12 ally quantiﬁed either through the sequence entropy (e.g. Goldstein & Pollock 2017) or 13 through the substitution rate (e.g. Grishin, Wolf and Koonin 2000). These two measures 14 of evolutionary variability are considered to be essentially equivalent, see for instance the 15 arguments presented in the seminal paper by Halpern and Bruno (1998). Here we adopt 16 a model of protein evolution with global selective constraints for the maintenance of the 17 thermodynamic stability of the native state both against unfolding and against misfold- 18 ing, and we show that these two measures of evolutionary variability are not in general 19 equivalent since they are diﬀerently inﬂuenced by the mutational process, which in general 20 favors exchanges between polar amino acids, so that for sites with equal sequence entropy 21 the site-speciﬁc substitution rate tends to be higher at exposed sites occupied by polar 22 amino acids. Because of the same reason, we ﬁnd that substitution rates averaged across 23 sites of the same protein are higher for more polar proteins. However, when we compare 24 diﬀerent mutational processes, we ﬁnd the counterintuitive result that mutational pro- 25 cesses that favor hydrophobic residues, such as those taking place in the genomes of AT 26 rich intracellular bacteria, tend to favor higher substitution rates. This is a result that 27 we argue is due to the diﬀerential constraints imposed by natural selection on buried and 28 exposed sites. 29 The evolutionary variability of a protein site is strongly inﬂuenced by the structural 30 properties of the site in the native state of the protein (Echave, Spielman and Wilke, 31 2016). In particular, the substitution rate changes dramatically between exposed and 32 buried sites in such a way that buried sites tend to evolve more slowly than exposed sites. 33 This is generally attributed to the fact that natural selection imposes stronger constraints 34 on buried sites (Franzosa & Xia, 2009). It was later shown that the number of native 35 inter-residue contacts formed by a protein site, which is negatively correlated with the 36 solvent accessibility, is a stronger predictor of the substitution rate (Yeh et al. 2014). Manuscript to be reviewed nucleotide Thymine that is more frequent in hydrophobic codons. This appearingly contradictory result arises because buried sites occupied by hydrophobic amino acids are characterized by larger selective factors that largely amplify the substitution rate between hydrophobic amino acids, while the selective factors of exposed sites have a weaker eﬀect. Thus, changes of the mutational bias produce deep eﬀects on the biophysical properties of the protein (hydrophobicity) and on its evolutionary properties (sequence entropy and substitution rate) at the same time. The program Prot evol that implements the two site-speciﬁc substitution processes is freely avail- able at https://ub.cbm.uam.es/prot fold evol/prot fold evol soft main.php#Prot Evo Abstract The number of amino acids that occupy a given protein site during evolution reﬂects the selective constraints operating on the site. This evolutionary variability is strongly inﬂuenced by the structural properties of the site in the native structure, and it is quantiﬁed either through sequence entropy or through substitution rates. However, while the sequence entropy only depends on the equilibrium frequencies of the amino acids, the substitution rate also depends on the exchangeability matrix that describes mutations in the mathematical model of the substitution process. Here we apply two variants of a mathematical model of protein evolution with selection for protein stability, both against unfolding and against misfolding. Ex- ploiting the approximation of independent sites, these models allow computing site- speciﬁc substitution processes that satisfy global constraints on folding stability. We ﬁnd that site-speciﬁc substitution rates do not depend only on the selective con- straints acting on the site, quantiﬁed through its sequence entropy. In fact, polar sites evolve faster than hydrophobic sites even for equal sequence entropy, as a con- sequence of the fact that polar amino acids are characterized by higher mutational exchangeability than hydrophobic ones. Accordingly, the model predicts that more polar proteins tend to evolve faster. Nevertheless, these results change if we compare proteins that evolve under dif- ferent mutation biases, such as orthologous proteins in diﬀerent bacterial genomes. In this case, the substitution rates are faster in genomes that evolve under muta- tional bias that favor hydrophobic amino acids by preferentially incorporating the 1 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Of course mutations modify both the stability and the precise structure 14 of the native state, but current models of ﬁtness cannot compute both eﬀects. 15 In a recent work, we have shown that stability-constrained models that take into ac- 16 count negative design for destabilizing misfolded conformations (Berezovsky, Zeldovich 17 & Shakhnovich 2007; Noivirt-Brik, Horovitz & Unger 2009; Minning, Porto & Bastolla 18 2013) predict that both the substitution rate and the entropy are maximal not at exposed 19 sites with few contacts, as observed, but at sites where the number of contacts is interme- 20 diate, which can accomodate both hydrophobic and polar amino acids and are predicted 21 to be extremaly tolerant to mutations (Jimenez, Arenas & Bastolla, 2018). On the other 22 hand, when stability with respect to misfolding is neglected, stability-constrained models 23 predict that the variability is maximal at exposed sites with few contacts (Scherrer, Meyer 24 & Wilke 2012; Echave, Jackson & Wilke, 2015), but these kinds of models overestimate 25 both the tolerance to mutations and the average hydrophobicity at almost all positions 26 (Jimenez, Arenas & Bastolla 2018) and they score much worse than models that consider 27 misfolding in likelihood calculations (Arenas Sanchez-Cobos & Bastolla 2015), so that 28 models that consider misfolding have to be preferred. In contrast, structure-constrained 29 models correctly predict that the variability is inversely related with the number of native 30 contacts (Huang et al. 2014). These results support the view that the structural eﬀect 31 of mutations cannot be neglected, in particular at sites with intermediate numbers of 32 contacts that are extremely tolerant to mutations under the point of view of the stability. 33 Here we adopt the stability-constrained mean-ﬁeld (MF, Arenas & Bastolla 2015; Bas- 34 tolla et al. 2006) and wild-type (WT, Jimenez, Arenas & Bastolla 2018) models of protein 35 Here we adopt the stability-constrained mean-ﬁeld (MF, Arenas & Bastolla 2015; Bas- 34 tolla et al. 2006) and wild-type (WT, Jimenez, Arenas & Bastolla 2018) models of protein 35 evolution that we used in the above-mentioned study. These models assume that sites 36 in the protein evolve independently in a site-speciﬁc manner, and determine their site- 37 speciﬁc properties by imposing a global constraint on the thermodynamic stability of the 38 known native state against both unfolding and misfolding. Manuscript to be reviewed Manuscript to be reviewed is a sigmoidal function of the folding free energy ∆G, i.e. f = 1/(1 + exp(−∆G/kT)) 1 (see Goldstein 2011; Serohijos & Shakhnovich 2014; Bastolla Dehouck & Echave 2017). 2 The second kind of model is the structurally-constrained model of protein evolution, 3 which estimates how mutations aﬀect the structure of the native state and computes the 4 ﬁtness from this predicted structural change (Echave 2008). In the literature stability- 5 constrained models are sometimes called structurally-constrained models, but we think 6 that this wording is misleading, since the ﬁtness function that they assume depends only 7 on stability and not on structural changes. On the other hand, structurally-constrained 8 models model the mutation as a perturbation that changes the wild-type as predicted 9 through the Elastic Network Model (ENM, Tirion 1996) and linear response theory and 10 assume that the stability does not change. Thus, stability-constrained models predict the 11 eﬀect of mutations through the predicted stability change but neglect the eﬀect of the 12 corresponding structure change, and structure-constrained models adopt the complemen- 13 tary perspective. Of course mutations modify both the stability and the precise structure 14 of the native state, but current models of ﬁtness cannot compute both eﬀects. 15 is a sigmoidal function of the folding free energy ∆G, i.e. f = 1/(1 + exp(−∆G/kT)) 1 (see Goldstein 2011; Serohijos & Shakhnovich 2014; Bastolla Dehouck & Echave 2017). 2 The second kind of model is the structurally-constrained model of protein evolution, 3 which estimates how mutations aﬀect the structure of the native state and computes the 4 ﬁtness from this predicted structural change (Echave 2008). In the literature stability- 5 constrained models are sometimes called structurally-constrained models, but we think 6 that this wording is misleading, since the ﬁtness function that they assume depends only 7 on stability and not on structural changes. On the other hand, structurally-constrained 8 models model the mutation as a perturbation that changes the wild-type as predicted 9 through the Elastic Network Model (ENM, Tirion 1996) and linear response theory and 10 assume that the stability does not change. Thus, stability-constrained models predict the 11 eﬀect of mutations through the predicted stability change but neglect the eﬀect of the 12 corresponding structure change, and structure-constrained models adopt the complemen- 13 tary perspective. Introduction 10 37 Two diﬀerent models rationalize why sites that form many contacts are subject to 38 stronger selective constraints. The ﬁrst kind of model, which we call stability-constrained 39 ﬁtness model, models the ﬁtness as the fraction of protein found in the native state, which 40 2 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Nevertheless, we ﬁnd that these two measures are not equivalent, since the 22 sequence entropy is only inﬂuenced by the equilibrium distribution of amino acids while 23 the substitution rate is also inﬂuenced by the mutation process that acts in evolution. 24 I i l f l i h f i ll i d b i Here we address the question whether the sequence entropy and the substitution rate 19 are equivalent measures of the evolutionary variability of a position, as suggested by 20 Halpern and Bruno (1998) who argue that these measures should be positively related 21 in general. Nevertheless, we ﬁnd that these two measures are not equivalent, since the 22 sequence entropy is only inﬂuenced by the equilibrium distribution of amino acids while 23 the substitution rate is also inﬂuenced by the mutation process that acts in evolution. 24 In particular, for equal sequence entropy, sites that are preferentially occupied by amino 25 acids with higher exchangeability have higher substitution rate, so that the substitution 26 rate is not a monotonic function of the sequence entropy in general. We also found that 27 polar amino acids are characterized by higher exchangeability so that, for equal sequence 28 entropy, exposed sites occupied by polar amino acids tend to substitute faster. However, 29 a bit counterintuitively but expected on the basis of the present model, if we simulate 30 mutational processes that favor hydrophobic amino acids the substitution rate increases 31 and the maximum across sites of the substitution rate moves towards more hydrophobic 32 sites, so that which sites are substituted faster ultimately depends on the mutation bias. 33 Manuscript to be reviewed The MF model signiﬁcantly 39 improves the likelihood of inferred evolutionary events with respect to empirical models 40 that do not take into account the structural properties of each site (Arenas & Bastolla 41 2015), and it improves the reconstruction of the stability properties of ancestral sequences 42 3 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed (Arenas et al. 2017). The WT model shows even better performances on several data sets 1 (Arenas & Bastolla, in preparation). Both models exploit the formal analogy between the 2 Boltzmann distribution in statistical physics, in which the probability of each conforma- 3 tion depends on the energy changed of sign and on the inverse of the temperature, and 4 the stationary distribution of a protein family in which the probability of each sequence 5 depends on its ﬁtness and on the eﬀective population size (Sella & Hirsh 2005, Mustonen 6 & Lassig 2005). 7 (Arenas et al. 2017). The WT model shows even better performances on sever 1 (Arenas & Bastolla, in preparation). Both models exploit the formal analogy b 2 tion depends on the energy changed of sign and on the inverse of the temperature, and 4 the stationary distribution of a protein family in which the probability of each sequence 5 depends on its ﬁtness and on the eﬀective population size (Sella & Hirsh 2005, Mustonen 6 & Lassig 2005). 7 In the MF model, the eﬀect on stability of amino-acid a at site i is predicted self- 8 consistently against the MF distribution at all other sites, in the spirit of mean-ﬁeld 9 models in statistical mechanics. In turn, the WT model predicts the eﬀect on stability 10 and ﬁtness of mutations of the wild-type sequence towards amino acid a at site i. Thus, in 11 theory the WT model is more suited for short evolutionary divergences and the MF model 12 is more suited for long evolutionary divergences (Arenas & Bastolla, in preparation). 13 After the site-speciﬁc amino-acid frequencies have been determined, the site-speciﬁc 14 exchangeability matrices that allow constructing the full site-speciﬁc substitution process 15 are computed applying the Halpern and Bruno formulas (Halpern & Bruno 1998), which 16 impose that the ﬁxation probabilities agree with Kimura’s formulas (Kimura 1962). Both 17 formulas are reproduced below for completeness. 18 Here we address the question whether the sequence entropy and the substitution rate 19 are equivalent measures of the evolutionary variability of a position, as suggested by 20 Halpern and Bruno (1998) who argue that these measures should be positively related 21 in general. 2014) 1 f(A) = e−∆G(A)/kT/ 1 + e−∆G(A)/kT . (1) (1) The computation is performed assuming that the native contact matrix Cnat does not 2 change in evolution. Upon single mutation, the free energy change ∆Gmut = ∆Gwt+∆∆G 3 is predicted adopting some models of protein stability (see below). 4 Equilibrium distribution 5 15 Importantly, the above ﬁxation probability deﬁnes a Monte Carlo process in sequence 16 space that fulﬁls detailed balance, so that its stationary distribution can be computed 17 exactly (Sella & Hirsh 2005; Mustonen & Lassig 2005), except for the normalization 18 constant, which would require a sum over 20L possible sequences A = A1 · · · AL: 19 P(A1 · · · AL) ∝exp ((N −1)ϕ(A1 · · · AL)) (3) (3) Note the analogy between this formula and the Boltzmann distribution with energy equal 20 to −ϕ and temperature equal to 1/(N −1). This explicit formula holds when the mutation 21 process is unbiased, so that all sequences are equally probable under the mutation model. 22 Note the analogy between this formula and the Boltzmann distribution with energy equal 20 to −ϕ and temperature equal to 1/(N −1). This explicit formula holds when the mutation 21 process is unbiased, so that all sequences are equally probable under the mutation model. 22 In the presence of mutation bias, the stationary distribution can be determined as the 23 distribution with minimal Kullback-Leibler divergence from the mutational distribution, 24 dKL = P A P mut(A) [log(P mut(A)) −log(P(A))], with a constraint on the average ﬁtness 25 P A P(A)ϕ(A). This condition generalizes the Boltzmann principle, and it was adopted 26 for developing the mean-ﬁeld model of protein evolution (Arenas & Bastolla, 2015). 27 gy gy to −ϕ and temperature equal to 1/(N −1). This explicit formula holds when the mutation 21 process is unbiased, so that all sequences are equally probable under the mutation model. 22 In the presence of mutation bias, the stationary distribution can be determined as the 23 distribution with minimal Kullback-Leibler divergence from the mutational distribution, 24 dKL = P A P mut(A) [log(P mut(A)) −log(P(A))], with a constraint on the average ﬁtness 25 P A P(A)ϕ(A). This condition generalizes the Boltzmann principle, and it was adopted 26 for developing the mean-ﬁeld model of protein evolution (Arenas & Bastolla, 2015). 27 Equilibrium distribution 5 Another approximation that is often used in these models is that the mutation rate 6 is extremely slow (Nµ ≪1) so that at every time there is only one mutant gene that 7 “competes” with the wild-type gene for ﬁxation in the population with eﬀective population 8 size of N individuals. Under this scenario, the probability that the mutation gets ﬁxed 9 in the population can be computed with Kimura’s formula (Kimura 1962) as 10 Pﬁx Awt →Amut = e−(ϕ(Amut)−ϕ(Awt)) −1 e−N(ϕ(Amut)−ϕ(Awt)) −1 (2) (2) where ϕ(A) = log(f(A)) is the logarithmic ﬁtness associated with the amino acid sequence 11 A, Eq.(1). As it is well known, the ﬁxation probability tends to the neutral limit Pﬁx = 12 1/N when ∆ϕ tends to zero, it tends exponentially to zero when ∆ϕ is negative and large, 13 and it tends to 1 −e−∆ϕ when ∆ϕ is positive. Nearly neutral mutations with selective 14 eﬀect \|∆ϕ\| ≈1/N are likely to be ﬁxed even when their eﬀect is deleterious (Ohta 1976). 15 Importantly, the above ﬁxation probability deﬁnes a Monte Carlo process in sequence 16 space that fulﬁls detailed balance, so that its stationary distribution can be computed 17 exactly (Sella & Hirsh 2005; Mustonen & Lassig 2005), except for the normalization 18 constant, which would require a sum over 20L possible sequences A = A1 · · · AL: 19 where ϕ(A) = log(f(A)) is the logarithmic ﬁtness associated with the amino acid sequence 11 A, Eq.(1). As it is well known, the ﬁxation probability tends to the neutral limit Pﬁx = 12 where ϕ(A) = log(f(A)) is the logarithmic ﬁtness associated with the amino acid sequence 11 A, Eq.(1). As it is well known, the ﬁxation probability tends to the neutral limit Pﬁx = 12 1/N when ∆ϕ tends to zero, it tends exponentially to zero when ∆ϕ is negative and large, 13 and it tends to 1 −e−∆ϕ when ∆ϕ is positive. Nearly neutral mutations with selective 14 eﬀect \|∆ϕ\| ≈1/N are likely to be ﬁxed even when their eﬀect is deleterious (Ohta 1976). Stability constrained ﬁtness model 35 Stability constrained models of protein evolution assume that the ﬁtness of a protein with 36 sequence A is proportional to the fraction of protein that is in the native state, which can 37 be computed from the folding free energy as (Goldstein 2011; Serohijos & Shakhnovich 38 4 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed 2014) 1 Manuscript to be reviewed Of course this assumption is not realistic, since diﬀerent sites determine protein stability 1 through their interactions, but it is needed for performing likelihood computations in an 2 eﬃcient way. Our strategy consists in determining the eﬀect of a mutation at site i self- 3 consistently, with respect to the MF distribution at all other sites. For simplicity, we shall 4 sometimes use the vectorial notation P i a for indicating P i(a), where a denotes one of the 5 twenty amino acid types. 6 The mean-ﬁeld distribution is determined by minimizing the Kullback-Leibler diver- 7 gence (distance between distribution) with respect to a global mutational distribution 8 P mut a , i.e. P ia P i a log (P i a/P mut a ). We impose a constraint on the average ﬁtness, which is 9 transformed into a constraint on the folding free energy ∆G. This condition on stability is 10 imposed through the Lagrange multiplier Λ that represents the strength of selection and 11 is related with the eﬀective population size. Furthermore, we impose the normalization 12 constraints P a P i a = 1 at all sites. 13 Since the parameters that determine the folding free energy are ﬁxed for all proteins 14 (see below), the only free parameters of the model are Λ and P mut a . The frequencies 15 are generally determined from the observed sequences in the protein of known structures 16 and the other sequences of the protein family, while Λ is determined by maximizing the 17 log-likelihood of the PDB sequence, P i log P i(APDB i ) , which yields a well-deﬁned single 18 maximum. The pre-computation of the moments of the contacts makes the computation 19 very fast, it runs in a few minutes even for proteins of several hundreds of amino acids. 20 For further computational details see (Arenas & Bastolla 2015). 21 Wild-type model of protein evolution 22 In the wild-type model (Jimenez, Arenas & Bastolla 2018), we also assume that sites 23 evolve independently. We further assume that the site-speciﬁc distribution P i a of amino 24 acid a at position i is proportional to the background distribution P mut a multiplied by 25 the exponential of the logarithmic ﬁtness of the corresponding mutation in which the 26 wild-type amino acid in the PDB AWT i is substituted by the new amino acid a: 27 P WT,i a ∝P mut a exp Λϕ mut(AWT i →a , (5) (5) The ﬁtness of a sequence is computed as in Eq.(1). The parameter Λ is again determined 28 by maximizing the likelihood of the wild-type sequence, P i log P WT,i(AWT i ) . 29 The ﬁtness of a sequence is computed as in Eq.(1). The parameter Λ is again determined 28 by maximizing the likelihood of the wild-type sequence, P i log P WT,i(AWT i ) . 29 Mean-ﬁeld model of protein evolution 28 The mean-ﬁeld (MF) model assumes that the equilibrium amino acid distribution is the 29 product of independent distributions at each protein site, 30 P(A1, · · · AL) = L Y i=1 P i(Ai) . (4) (4) 5 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Halpern-Bruno exchangeability matrices 4 To fully specify the site-speciﬁc substitution processes, besides the site-speciﬁc frequen- 5 cies P i a we need to compute consistent exchangeability matrices with the Halpern-Bruno 6 formulas (Halpern & Bruno 1998). 7 Given a site-speciﬁc amino acid distribution that reﬂects selective constraints, the 8 Halpern-Bruno method allows computing the rate matrices of the associated site-speciﬁc 9 substitution processes Qi ab = Ei abP i b that are produced by a global (not site-speciﬁc) 10 mutation process consistent together with Kimura’s ﬁxation probability, Eq.(2). 11 Without loss of generality, we parametrize the rate matrix of the global mutation 12 process as Qmut ab = Emut ab P mut b , where P mut a is the stationary matrix of the mutation process 13 and Emut ab is its exchangeability matrix. To simplify formulas, here we assume detailed 14 balance, i.e. we assume that Emut ab is a symmetric matrix (this condition can be easily 15 relaxed). We write the rate matrices as Qi ab = Qmut ab Pﬁx(f i a, f i b), where f i a is the “ﬁtness” of 16 amino acid a at site i. We impose that Pﬁx is the ﬁxation probability Eq.(2). Halpern and 17 Bruno showed that the site-speciﬁc ﬁtness can be inferred from the stationary distribution 18 from P i a = P mut a (f i a) N, yielding the following site-speciﬁc substitution process 19 Qi ab = Ei abP i b (7) Ei ab = Emut ab ln(F sel,i b ) −ln(F sel,i a ) F sel,i b −F sel,i a ! (8) with F sel,i a = P i a P mut a (9) (7) (8) (9) The selective factors F sel,i a quantify how much the site-speciﬁc distribution P i a deviates 20 from the background distribution P mut a induced by mutation alone. 21 The selective factors F , a quantify how much the site-speciﬁc distribution Pa deviates 20 from the background distribution P mut a induced by mutation alone. 21 It can be immediately seen that the exchangeability matrices Ei ab are symmetric, which 22 implies that detailed balance holds and P i a is the stationary distribution. 23 Manuscript to be reviewed where P i a is obtained either from the evolutionary model (mean-ﬁeld or wild-type) or 1 from a MSA or from pooled amino acids at equivalent structural positions with the same 2 number of contacts. 3 where P i a is obtained either from the evolutionary model (mean-ﬁeld or wild-type) or 1 from a MSA or from pooled amino acids at equivalent structural positions with the same 2 number of contacts. 3 number of contacts. 3 Sequence entropy 30 The sequence entropy at position i measures the variability of this position as 1 Si = − 20 X a=1 P i a log(P i a) , (6) (6) 6 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed l, Ri ab = P i aP i bEi ab = Ri ba, with are equal, Ri ab = P i aP i bEi ab = Ri ba, with 1 are equal, Ri ab = P i aP i bEi ab = Ri ba, with 1 Ri ab = P mut a P mut b Emut ab F sel,i a F sel,i b ln(F sel,i b ) −ln(F sel,i a ) F sel,i b −F sel,i a (10) (10) (10) In the above equation, the ﬂux is partitioned into a global component that is attributed 2 to the mutation process (superscript mut) and a site-speciﬁc component that is attributed 3 to selection (superscript sel), which allows analysing the contributions of mutation and 4 selection separately. The ﬂux is maximal for substitutions ab that have large and almost 5 equal selective factors F sel,i a ≈F sel,i b and have large mutational ﬂux P mut a P mut b Emut ab . The 6 site-speciﬁc substitution rates are computed as the weighted average of the substitution 7 rate matrix Qab = Ei abP i b, 8 Ri = X a̸=b P i aEi abP i b = X a̸=b Ri ab (11) (11) Since the ﬂux between any pair of amino acids a and b decreases when their diﬀerence 9 of ﬁtness increases, Halpern and Bruno argued that the substitution rate Ri is higher at 10 position with higher sequence entropy (Halpern & Bruno 1998). However, this argument 11 is not rigorously valid, since it neglects the fact that the substitution rate is enhanced 12 at sites where the site-speciﬁc selective favtors F sel,i a F sel,i b are correlated with the global 13 mutational ﬂux (P mut a P mut b Emut ab ). Consistent with this argument, we observed that the 14 substitution rate is not a strictly increasing function of sequence entropy but, for the same 15 sequence entropy, it tends to increase at sites that favor polar amino acids (see Results), 16 which are characterized by higher mutational ﬂuxes than hydrophobic amino acids. 17 Evolutionary rates 24 For neutral substitutions with F sel,i a = F sel,i b , in particular synonymous substitutions a = b, 25 applying l’Hopital’s rule we ﬁnd Ei ab = Emut ab /F sel,i b and Qi ab = Qmut ab , i.e. the rate of 26 synonymous substitutions equals the mutation rate, in agreement with Kimura’s theory. 27 If the amino acid b is favored by selection with respect to amino acid a, F sel,i b > F sel,i a , 28 then the substitution rate is enhanced with respect to the neutral rate, and it is decreased 29 in the opposite case. Because of detailed balance, the ﬂux in one direction and the other 30 7 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed which is the observational data from which empirical models are deduced: P mut a P mut b Eﬂux ab 1 L X i F sel,i a F sel,i b ln(F sel,i b ) −ln(F sel,i a ) F sel,i b −F sel,i a = P emp a P emp b Eemp ab (12) (12) where we use more compact matricial notation. We call the corresponding exchange- bility matrix Eﬂux ab the ﬂux matrix (ﬂux).This mutation model yields optimal results in phylogenetic inference (Arenas & Bastolla, 2015). where we use more compact matricial notation. We call the corresponding exchange- 2 bility matrix Eﬂux ab the ﬂux matrix (ﬂux).This mutation model yields optimal results 3 in phylogenetic inference (Arenas & Bastolla, 2015). 4 where we use more compact matricial notation. We call the corresponding exchange- 2 bility matrix Eﬂux ab the ﬂux matrix (ﬂux).This mutation model yields optimal results 3 in phylogenetic inference (Arenas & Bastolla, 2015). 4 3. Thirdly, we model the mutational process at the nucleotide level, using the genetic code and parameterizing the process through the nucleotide frequencies and the transition-transversion ratio κ. The four free parameters are ﬁxed by imposing that the resulting background distribution P mut a yields amino acid frequencies as close as possible to those observed in the data, P obs a (Arenas & Bastolla, 2015), as detailed below. We call the corresponding exchangeability matrix the optimized nucleotide (nuc opt) matrix. 4. The last model is identical to the nuc opt model, except that the nucleotide fre- 12 quencies are not optimized but they are input parameters. In this way, we can vary 13 the average hydrophobicity of the complete model by varying the Thymine content, 14 since hydrophobic amino acids are enriched in the T base at second codon position. 15 We call this model the nuc var model. 16 In the nuc models, for any set of nucleotide frequencies and transition-tranversion rate 17 we combine the substitution process at the nucleotide level with a selection process that 18 assigns ﬁtness one to sense codons and ﬁtness zero to stop codons. Mutation process 18 Finally, we have to deﬁne the global exchangeability matrix Emut ab that characterizes the 19 mutation process. For this, we consider four types of mutational models. To compare 20 the resulting substitution rates, in all cases we ﬁx the scale of the exchangeability matrix 21 equating the substitution rate under mutation alone, P a̸=b P mut a P mut b Emut ab = 1. 22 1. In the ﬁrst model, the global exchangeability matrix is equal to the empirical ex- 23 changeability matrix (WAG, Whelan & Goldman 2001; or JTT, Jones Taylor and 24 Thornton 1992), i.e. Emut ab ≡Eemp ab . We call this model the empirical (emp) ex- 25 changeability matrix. Since empirical substitution processes include information 26 both on mutation and selection, we expect that they strongly correlate with the 27 selection process. 28 2. In the second model, we remove the eﬀect of selection from the empirical substitution 29 model by imposing that for each pair of amino acids, the ﬂux predicted by the global 30 model and averaged over all positions is equal to the empirical ﬂux P emp a P emp b Eemp ab , 31 8 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Data and observed substitution rates 1 We performed our computations on 213 proteins that were examined in a previous study 2 (Echave, Jackson & Wilke 2015). The results were qualitatively identical from one protein 3 to the other. 4 We performed our computations on 213 proteins that were examined in a previous study 2 (Echave, Jackson & Wilke 2015). The results were qualitatively identical from one protein 3 to the other. 4 p p p p y (Echave, Jackson & Wilke 2015). The results were qualitatively identical from one protein 3 to the other. 4 The observed substitution rates of 213 proteins that we show for comparison were esti- 5 mated in (Echave, Jackson & Wilke 2015) from the MSA of homologous sequences through 6 the program Rate4Site (Pupko et al. 2002), which builds the phylogenetic tree using a 7 neighbour-joining algorithm (Saitou & Nei 1987) and estimates rates with an empirical 8 Bayesian approach adopting the JTT model of sequence evolution (Jones, Taylor & Thorn- 9 ton 1992). The multiple sequence alignments were generously provided by Juli´an Echave 10 and are publicly available at the url https://github.com/wilkelab/therm constraints rate variation 11 Manuscript to be reviewed Detailed balance is 19 fulﬁlled at the nucleotide level, but it is only approximated at the codon level because 20 of this selection against stop codons, therefore the transition to transversion rate can 21 inﬂuence the stationary frequencies and we have to compute the stationary distribution 22 of the 61 sense codons numerically. 23 In the nuc models, for any set of nucleotide frequencies and transition-tranversion rate 17 we combine the substitution process at the nucleotide level with a selection process that 18 assigns ﬁtness one to sense codons and ﬁtness zero to stop codons. Detailed balance is 19 fulﬁlled at the nucleotide level, but it is only approximated at the codon level because 20 of this selection against stop codons, therefore the transition to transversion rate can 21 inﬂuence the stationary frequencies and we have to compute the stationary distribution 22 of the 61 sense codons numerically. 23 More precisely, we model the mutation rate between two codons diﬀering at one po- 24 sition, say the third one (n1n2n3 and n1n2n′ 3) as µκ(n3, n′ 3)f nuc(n′ 3)S(n1n2n′ 3), where µ is 25 a global rate parameter, κ(n3, n′ 3) is one if n3, n′ 3 are related through a transversion and 26 is the transition-tranversion rate otherwise, f nuc(n′ 3) is the stationary frequency of the 27 new nucleotide and S(n1n2n′ 3) is zero if n1n2n′ 3 is a stop codon, one otherwise. After the 28 frequencies of the 61 sense codons evolve to their equilibrium state, the stationary fre- 29 quencies of amino acids P mut a are computed summing over codons and the exchangeability 30 matrix is computed from the equilibrium ﬂuxes between pairs of codons that code for any 31 pair of amino acids. In the nuc opt model, the score of each set of mutation parameters is 32 computed as the likelihood of the observed number of amino acids, P a nobs(a) log (P mut a ), 33 and the parameters that maximize the likelihood are chosen. 34 9 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed (Minning, Porto & Bastolla 2013): 1 ! to & Bastolla 2013): (Minning, Porto & Bastolla 2013): 1 Gmisf ≡ −T log X C e−P i<j CijU(Ai,Aj)/T+S(C) ! (13) ≈ ⟨E⟩− (E −⟨E⟩)2 2T + (E −⟨E⟩)3 6T 2 −LSCT (13) 2T where LSC is the logarithm of the number of compact contact matrices, ⟨.⟩represents the 2 average over the set of alternative compact contact matrices of L residues. This estimate 3 only holds above the freezing temperature, while the free energy is kept constant below the 4 freezing temperature (Derrida 1981). We assume for simplicity that the conformational 5 entropy, S(Cij), is approximately the same for all compact structures including the native 6 one, and it can be neglected for computing free energy diﬀerences. The mean values of 7 the energy can be computed from the mean values of the contacts, which are computed 8 at the beginning and tabulated to accelerate the computation: ⟨E⟩= P i<j ⟨Cij⟩Uij, 9 ⟨(E −⟨E⟩)⟩= P i<j,k<l (⟨CijCkl⟩−⟨Cij⟩⟨Ckl⟩) UijUkl with Uij = U(Ai, Aj). We also 10 adopt the approximation that ⟨Cij⟩only depends on \|i −j\| (Minning, Porto & Bastolla 11 2013). 12 ) Putting together these free energy estimates, we obtain the free energy diﬀerence 13 between the native and the non-native states as 14 ∆G(Cnat, A) = Gnat −kT log e−Gmisf/kT + e−GU/kT , (14) (14) where the free energy of the non-native state is computed as a Boltzmann average, which 15 is essentially equal to Gmisf when the sequence is hydrophobic (Gmisf −GU/kT ≪−kT) 16 and is essentially equal to GU when the sequence is hydrophylic (Gmisf −GU/kT ≫kT). 17 For neglecting stability against misfolding, we compute ∆G = Gnat(Cnat, A) + LSU. 18 Modelling stability against unfolded and misfolded states 12 Finally, for completeness we descibe here how we estimate the folding free energy ∆G of 13 the experimentally known native state of a protein. 14 For this purpose, we adopt the contact matrix representation of the protein structure, 15 consisting in the following: For each pair of residues at positions i and j along the polypep- 16 tidic chain, Cij equals one if the residues are in contact and zero otherwise. We deﬁne 17 two residues to be in contact if any pair of their heavy atoms are closer than 4.5˚A. Since 18 contacts with \|i−j\| ≤2 are formed in almost all structures, they do not contribute to the 19 free energy diﬀerence between the native and the misfolded ensemble, and we set Cij = 0 20 if \|i −j\| ≤2. The free energy of a protein in the mesoscopic structure described by Cij is 21 modelled as a sum of contact interactions, E(C, A) = P i<j CijU(Ai, Aj), which depends 22 on the type of amino acids in contact Ai and Aj and on 210 contact interaction parame- 23 ters U(a, b), for which we adopt the parameters determined in (Bastolla, Vendruscolo & 24 Knapp, 2000). 25 For simplicity, we neglect the conformational entropy of the folded native state and 26 estimate its free energy as Gnat(Cnat, A) ≈P i<j Cnat ij U(Ai, Aj). Regarding the unfolded 27 state, we neglect their contact interactions and estimate its free energy as GU ≈−TLSU, 28 where T is the temperature in units in which kB = 1, L is chain length and SU is the 29 conformational entropy per residue of an unfolded chain. We compute the free energy of 30 the misfolded state from the partition function of the contact energy E(C, A) over a set of 31 compact contact matrices C of L residues that are obtained from the PDB. In agreement 32 with previous studies (Garel & Orland 1988; Shakhnovich & Gutin 1989; Bryngelson et 33 al. 1995), the resulting free energy is approximately described by the Random Energy 34 Model (REM) (Derrida 1981), with the addition of the third moment of the contact energy 35 10 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Dependence of the substitution rate on the global exchangeability 20 model 21 In this work we studied two measures of the evolutionary variability of protein sites, 22 sequence entropy and substitution rate, predicted through the site-speciﬁc stability con- 23 strained substitution models that we introduced and studied recently (Arenas Sanchez- 24 Cobos & Bastolla 2015; Jimenez, Arenas & Bastolla 2018). The results that we present 25 arise from the predictions of our computational models, not from the analysis of natural 26 protein sequences, thus they ignore important aspects of protein biology such as active 27 sites and protein-protein interactions. We have to pay this price in order to address 28 general questions such as the comparison between the two measures of evolutionary vari- 29 ability and how they are aﬀected by the mutational process and by selection on protein 30 stability. 31 11 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Figure 1: Schematic representation of the site-speciﬁc stability constrained substitution models studied in this work. Figure 1: Schematic representation of the site-speciﬁc stability constrained substitution models studied in this work. Figure 1: Schematic representation of the site-speciﬁc stability constrained substitution models studied in this work. The models adopted in this study are described in Methods and schematically repre- 1 sented in Fig.1. They simulate an evolutionary process where mutations follow a global 2 mutational process modelled through a global amino acid distribution P mut a and exchange- 3 ability matrix Emut ab (Fig.1A), and they modify the stability of the protein ∆G, which 4 determines ﬁtness through Eq.(1) (Fig.1B). We adopt the approximation that sites of the 5 protein evolve independently under site-speciﬁc substitution processes globally governed 6 by the ﬁtness function of Eq.(1) (Fig.1C). Since the stability ∆G depends on pairwise 7 interactions, the ﬁtness eﬀect of a mutation at a site depends on the amino acids present 8 at all other sites. In order to compute independent amino acid frequencies at each site, 9 we perform two type of approximations: the mean-ﬁeld approximation (Fig.1D; Arenas 10 Sanchez-Cobos & Bastolla 2015), which evaluates the eﬀect of the mutation on stability 11 considering the mean-ﬁeld of the other sites, and the wild-type approximation (Fig.1E; 12 Jimenez, Arenas & Bastolla 2018), which evaluates the eﬀect of the mutation on stability 13 when the other sites are occupied by the same amino acid as in the PDB sequence. Dependence of the substitution rate on the global exchangeability 20 model 21 We 14 then compute site-speciﬁc substitution processes obtained by combining the global muta- 15 tional process and the site-speciﬁc ﬁxation probabilities computed from the site-speciﬁc 16 amino acid frequencies through the Halpern-Bruno formula, Eq.(10). 17 We brieﬂy report here previous results obtained with the stability-constrained models. 18 Since the main component of contact interaction matrices like the one that we adopt here 19 is hydrophobicity (i.e., U(a, b) ≈ǫh(a)h(b), where h(a) is related with the hydrophobicity 20 12 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed 28 We found that empirical exchangeability matrices (emp) produce the larger substitu- 29 ti t (Fi 2B d C) Th t i t k i t t b th th t ti We considered three models of global exchangeability matrices (see Materials and 18 Methods): (1) Empirical (emp) exchangeability matrices, such as the familiar WAG (Whe- 19 lan & Goldman 2001) or JTT (Jones, Taylor & Thornton 1992) matrices; (2) Flux ex- 20 changeability matrices (ﬂux), which are obtained from empirical exchangeability matrices 21 removing the selective factors represented in the stability-constrained mean-ﬁeld model, 22 so that the average ﬂux predicted by the model between any pair of amino acids coincides 23 acids obtained from a mutational process at the nucleotide level with parameters opti- 25 mized by maximizing the likelihood of the observed amino acid composition (nuc opt); (4) 26 Exchangeability matrices obtained from a mutational process at the nucleotide level with 27 varying parameters, that allows studying the eﬀect of varying hydrophobicity (nuc var). 28 We found that empirical exchangeability matrices (emp) produce the larger substitu- 29 tion rates (Fig 2B and C) These matrices take into account both the mutation process We found that empirical exchangeability matrices (emp) produce the larger substitu- 29 tion rates (Fig.2B and C). These matrices take into account both the mutation process 30 and the selection process, since they have been obtained from substitutions that have 31 been ﬁxed through natural selection. From Eq.(10) we can see that the substitution 32 rate is enhanced when the site-speciﬁc selective factors F sel,i a F sel,i b are correlated with the 33 global mutational ﬂux (P mut a P mut b Emut ab ). Since the empirical substitution models were 34 determined in such a way that their ﬂux equals the ﬂux observed in real data, which 35 accounts both for the mutational process and for selection, we expect and ﬁnd that the 36 empirical ﬂux is strongly correlated with the selective factors F sel,i a F sel,i b averaged across 37 all protein sites. This argument explains why the empirical exchangeability matrices yield 38 the highest substitution rates. 39 The ﬂux exchangeability matrices remove from the empirical exchangeability matrix 40 the eﬀect of natural selection that is represented in the mean-ﬁeld model averaged across 41 sites. Manuscript to be reviewed of amino acid a), the site-speciﬁc amino acid frequencies obtained under the stability- 1 constrained model yield high average hydrophobicity at buried sites at which the native 2 contact matrix has many contacts. More precisely, sites constrained to be highly hy- 3 drophobic have large components of the principal eigenvector of the contact matrix (Bas- 4 tolla et al. 2005) or, almost equivalently, large eﬀective connectivity (Bastolla et al. 2008). 5 These structural descriptors are strongly related with the number of contacts but do not 6 coincide with them. These stability constraints in turn inﬂuence the variability of the 7 amino acid distribution (Porto et al. 2005) in such a way that sites that are constrained 8 to have either very high or very low average hydrophobic (averaged over the site-speciﬁc 9 amino acid distribution) are characterized by low entropy, while sites that can accomodate 10 both polar and hydrophobic amino acids are characterized by high entropy. Because of 11 this reason, the plot of the sequence entropy versus hydrophobicity and versus number of 12 contacts has a bell shape. 13 These stability constraints that inﬂuence the sequence entropy also inﬂuence the site- 14 speciﬁc substitution rates. However, we found that the substitution rates depend not 15 only on the selective forces that act speciﬁcally at each protein site, but also on the global 16 exchangeability matrix that represents the mutation process. 17 g y p p We considered three models of global exchangeability matrices (see Materials and 18 Methods): (1) Empirical (emp) exchangeability matrices, such as the familiar WAG (Whe- 19 lan & Goldman 2001) or JTT (Jones, Taylor & Thornton 1992) matrices; (2) Flux ex- 20 changeability matrices (ﬂux), which are obtained from empirical exchangeability matrices 21 removing the selective factors represented in the stability-constrained mean-ﬁeld model, 22 so that the average ﬂux predicted by the model between any pair of amino acids coincides 23 with the observed empirical ﬂux, see Eq.(12); (3) Exchangeability matrices between amino 24 acids obtained from a mutational process at the nucleotide level with parameters opti- 25 mized by maximizing the likelihood of the observed amino acid composition (nuc opt); (4) 26 Exchangeability matrices obtained from a mutational process at the nucleotide level with 27 varying parameters, that allows studying the eﬀect of varying hydrophobicity (nuc var). Manuscript to be reviewed 28 We found that empirical exchangeability matrices (emp) produce the larger substitu- 29 tion rates (Fig.2B and C). These matrices take into account both the mutation process 30 and the selection process, since they have been obtained from substitutions that have 31 been ﬁxed through natural selection. From Eq.(10) we can see that the substitution 32 rate is enhanced when the site-speciﬁc selective factors F sel,i a F sel,i b are correlated with the 33 global mutational ﬂux (P mut a P mut b Emut ab ). Since the empirical substitution models were 34 determined in such a way that their ﬂux equals the ﬂux observed in real data, which 35 accounts both for the mutational process and for selection, we expect and ﬁnd that the 36 empirical ﬂux is strongly correlated with the selective factors F sel,i a F sel,i b averaged across 37 all protein sites. This argument explains why the empirical exchangeability matrices yield 38 the highest substitution rates. 39 We considered three models of global exchangeability matrices (see Materials and 18 Methods): (1) Empirical (emp) exchangeability matrices, such as the familiar WAG (Whe- 19 lan & Goldman 2001) or JTT (Jones, Taylor & Thornton 1992) matrices; (2) Flux ex- 20 changeability matrices (ﬂux), which are obtained from empirical exchangeability matrices 21 removing the selective factors represented in the stability-constrained mean-ﬁeld model, 22 so that the average ﬂux predicted by the model between any pair of amino acids coincides 23 with the observed empirical ﬂux, see Eq.(12); (3) Exchangeability matrices between amino 24 acids obtained from a mutational process at the nucleotide level with parameters opti- 25 mized by maximizing the likelihood of the observed amino acid composition (nuc opt); (4) 26 Exchangeability matrices obtained from a mutational process at the nucleotide level with 27 varying parameters, that allows studying the eﬀect of varying hydrophobicity (nuc var). PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Substitution rates are diﬀerent for hydrophobic and hydrophylic 9 sites with the same entropy 10 Next, we investigated more in detail the relationship between site-speciﬁc sequence en- 11 tropies and substitution rates. Since the ﬂux between any pair of amino acids a and b, 12 Eq.(10), decreases when their diﬀerence of ﬁtness increases, Halpern and Bruno argued 13 that the substitution rate Ri is higher at position with higher sequence entropy (Halpern 14 & Bruno 1998). However, this argument is not rigorously valid, since it neglects the 15 fact that the substitution rate is enhanced at sites where the site-speciﬁc selective factors 16 F sel,i a F sel,i b are correlated with the global mutational ﬂux (P mut a P mut b Emut ab ). One can see in 17 Fig.2B-C that sites with larger entropy tend to have on the average larger substitution 18 rates, as predicted by Halpern and Bruno (1998), but the substitution rate is not a strictly 19 increasing function of sequence entropy, not even when it is averaged over diﬀerent sites. 20 We then show in Fig.3 the detailed plot of the substitution rate versus the sequence 21 entropy for all sites of a small protein, chosen in such a way that we can spot all of the sites. 22 We can clearly see in Fig.3 two branches that correspond to diﬀerent numbers of native 23 contacts. Sites with few contacts, which tend to be occupied by polar amino acids, evolve 24 faster than sites with many contacts, occupied by hydrophobic amino acids, even if their 25 sequence entropy is equal. With the ﬂux model of the mutation process, which we consider 26 the most reliable model since it reproduces the empirically observed ﬂux between all pairs 27 of amino acids, this happens for both the MF and the WT model of natural selection 28 (plots C and D). All other studied proteins present the same trend (see Supplementary 29 Figures S1 and S2), but for large proteins the representation is less clear. Manuscript to be reviewed Thus, these results suggest that the ﬂux model associated with the WT model is eﬀective 1 in removing selective constraints for sites with many contacts, but less eﬀective for sites 2 with few contacts and high entropy. 3 Note that the curves that represent the substitution rates versus the sequence en- 4 tropy tend to collapse for very high rates. This is due to the fact that all the global 5 exchangeability matrices are normalized in such a way that their average ﬂux equals one, 6 P ab P mut a P mut b Emut ab = 1. This ﬂux is achieved at neutral sites where the selective factors 7 F sel,i a are equal and the entropy is maximal. 8 Thus, these results suggest that the ﬂux model associated with the WT model is eﬀective 1 in removing selective constraints for sites with many contacts, but less eﬀective for sites 2 with few contacts and high entropy. 3 Note that the curves that represent the substitution rates versus the sequence en- 4 tropy tend to collapse for very high rates. This is due to the fact that all the global 5 exchangeability matrices are normalized in such a way that their average ﬂux equals one, 6 P ab P mut a P mut b Emut ab = 1. This ﬂux is achieved at neutral sites where the selective factors 7 F sel,i a are equal and the entropy is maximal. 8 Manuscript to be reviewed Consistently, we ﬁnd that the substitution rates determined through the ﬂux model 42 13 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Figure 2: Eﬀect of the exchangeability model on substitution rates. The plots represent substitution rate versus sequence entropy (A-C) and versus number of native contacts (D-F) for MSA (A,D), WT model (B,E) and MF model (C,F). Simulations are performed with the emp, ﬂux and nuc opt models of the global exchangeability matrix. In all cases the emp model produces the highest substitution rates, consistent with the fact that this model also represents selection. Figure 2: Eﬀect of the exchangeability model on substitution rates. The plots represent substitution rate versus sequence entropy (A-C) and versus number of native contacts (D-F) for MSA (A,D), WT model (B,E) and MF model (C,F). Simulations are performed with the emp, ﬂux and nuc opt models of the global exchangeability matrix. In all cases the emp model produces the highest substitution rates, consistent with the fact that this model also represents selection. are smaller than those determined with the emp model (Fig.2B and C). We also found 1 in previous work that the ﬂux model yields larger likelihood in phylogenetic inference 2 (Arenas, Sanchez-Cobos & Bastolla, 2015). Because of these results, the ﬂux model is our 3 default exchangeability model. 4 Fig. 2C shows that the nuc opt model with mutations at the nucleotide level and opti- 5 mized parameters produces lower substitution rates than the ﬂux model when associated 6 with the MF model of selective constraints, which indicates that the ﬂux model combined 7 with the MF model may still include some selection. However, when the WT model 8 of selection is applied, the nuc opt model again produces lower substitution rates than 9 the ﬂux model for exposed sites with few contacts and high entropy, but the ﬂux model 10 produces lower substitution rates for buried sites with many contacts and low entropy 11 (Fig. 2B and E). Note that the WT model represents stronger selective constraints than 12 the MF model, since it generally predicts lower sequence entropies and substitution rates. 13 14 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Substitution rates are diﬀerent for hydrophobic and hydrophylic 9 sites with the same entropy 10 When we model 30 the mutation process at the codon level through the model nuc opt, whose parameters are 31 separately optimized for each protein, the diﬀerences between the two branches decrease 32 considerably, in particular when we apply the WT model of selection (Fig.3B) and the 33 trend may change from one protein to the other, since diﬀerent proteins evolve under 34 diﬀerent mutation processes, in such a way that the buried branch may evolve faster than 35 the exposed branch for some protein, see Supplementary Figures S3 and S4. 36 Since sequence entropy is a measure of the selective constraints, this diﬀerence be- 37 tween sites with equal sequence entropy must be attributed to the mutation process 38 embodied in the exchangeability matrix, not to natural selection. The explanation of 39 this ﬁnding is based on Eq.(10): polar amino acids tend to have higher mutational ﬂuxes 40 15 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed 0.5 1 Site-specific substitution rate 2 2.5 3 Site-specific sequence entropy 0.5 1 Site-specific substitution rate 2 2.5 3 Site-specific sequence entropy Mean-field model Exchangeability: nuc_opt Wild-type model Exchangeability: nuc_opt Mean-field model Exchangeability: WAG flux Wild-type model Exchangeability: WAG flux white <6 contacts black >=6 contacts A B C D Figure 3: Each point represents the substitution rate versus the sequence entropy for all sites of the ribonuclease protein with PDB code 1pyl, which is representative of our data set and makes the ﬁgure easier to interpret because of its small number of sites. One can spot two branches, corresponding to sites that evolve faster and slower for the same sequence entropy. The two branches correspond to polar sites with few contacts (white circles) and hydrophobic sites with many contacts (black circles), respectively. D Figure 3: Each point represents the substitution rate versus the sequence entropy for all sites of the ribonuclease protein with PDB code 1pyl, which is representative of our data set and makes the ﬁgure easier to interpret because of its small number of sites. One can spot two branches, corresponding to sites that evolve faster and slower for the same sequence entropy. The two branches correspond to polar sites with few contacts (white circles) and hydrophobic sites with many contacts (black circles), respectively. PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Substitution rates are diﬀerent for hydrophobic and hydrophylic 9 sites with the same entropy 10 On the other hand, mutation processes with extreme properties (very high or very low hydrophobicity) tend to increase the substitution rate. Substitution rates are diﬀerent for hydrophobic and hydrophylic 9 sites with the same entropy 10 (P mut a P mut b Emut ab ) than hydrophobic amino acids, therefore exposed sites in which polar 1 amino acids have larger site-speciﬁc selective factors F sel,i a F sel,i b tend to evolve faster than 2 buried sites. 3 This result contradicts the expectation that the site-speciﬁsubstitution rates are 4 monotonic functions of the sequence entropy (Halpern & Bruno 1998): one can see from 5 Fig.3 that sites characterized by lower entropy can substitute faster if they are exposed 6 sites occupied by polar amino acids. 7 One may note that in Fig.3 some white points with small number of contacts overlap 8 with the black points with large number of contacts. This is due to the fact that the 9 number of contacts is only an approximate predictor of the selective factors favoring 10 hydrophobicity, while better correlation with the average hydrophobicity is achieved using 11 the principal eigenvector of the contact matrix (Bastolla et al. 2005) or the eﬀective 12 connectivity proﬁle that generalizes it for multidomain proteins (Bastolla et al. 2008), 13 which are not merely local descriptors as the number of contacts but also represent the 14 global topology of the native contact matrix. 15 16 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Figure 4: In this plot each point represents a group of proteins with similar mean hy- drophobicity in the evolutionary model, and each curve is obtained by varying the global mutational distribution and exchangeability matrix, which represent the mutation pro- cess. One can see that, for the same mutation process, more hydrophobic proteins tend to evolve more slowly, except when the mutation process induces very high hydrophobic- ity, in which case the substitution rate becomes an increasing function of hydrophobicity. On the other hand, mutation processes with extreme properties (very high or very low hydrophobicity) tend to increase the substitution rate. Figure 4: In this plot each point represents a group of proteins with similar mean hy- drophobicity in the evolutionary model, and each curve is obtained by varying the global mutational distribution and exchangeability matrix, which represent the mutation pro- cess. One can see that, for the same mutation process, more hydrophobic proteins tend to evolve more slowly, except when the mutation process induces very high hydrophobic- ity, in which case the substitution rate becomes an increasing function of hydrophobicity. Manuscript to be reviewed We then consider the same mutation process for all proteins, parameterized by the 1 G+C content at the mutational equilibrium (nuc var model). Since Thymine at second 2 codon position almost always codes for hydrophobic amino acids, there is a negative 3 correlation between G+C content of the mutation model and the average hydrophobicity 4 of the protein sequence. Varying the mutation bias we construct diﬀerent sets of model 5 proteins that present varying hydrophobicity and are characterized by diﬀerent mutational 6 ﬂuxes. In this way, we can investigate how the mutation bias inﬂucence the biophysical 7 properties (hydrophobicity) and the evolutionary properties (substitution rate, sequence 8 entropy) of an evolving protein. When the GC content is high and the hydrophobicity is 9 low (GC≥0.35 with the MF model, Fig.4A and GC≥0.5 with the WT model, Fig.4B) 10 more polar proteins tend to evolve faster, as we observed with the ﬂux model. However, 11 when the mutation process induces high hydrophobicity (GC=0.2 in Fig.4A and GC< 0.5 12 in Fig.4B), the substitution rate becomes an increasing function of hydrophobicity. This 13 is easily rationalized by the fact that, when the background distribution P mut a is biased 14 towards hydrophobic amino acids, the mutational ﬂux P mut a P mut b Emut ab is higher between 15 pairs of hydrophobic residues, and the other way round. 16 Diﬀerent G+C content in Fig.4 represent diﬀerent mutational processes, which may 17 be interpreted as bacterial species characterized by diﬀerent GC bias. For the MF model 18 (plot A), one can see that mutational processes with extreme bias (very high or very low 19 G+C content and hydrophobicity) tend to increase the average substitution rate, while for 20 the WT model (plot B) mutational processes biased towards hydrophobic residues such 21 as GC=0.2 have higher substitution rates. Therefore, although hydrophobicity is nega- 22 tively correlated with the substitution rate when we compare diﬀerent proteins evolving 23 with the ﬂux mutational process (black points in Fig.4 and Fig.3C and D) or with muta- 24 tional processes with high G+C, the correlation becomes positive if we compare diﬀerent 25 mutational processes. 26 More hydrophobic proteins substitute more slowly, but mutation 1 bias towards hydrophobicity increases the substitution rates 2 After investigating the relationship between hydrophobicity and substitution rates com- 3 paring individual sites, we perform the same analysis comparing diﬀerent proteins. For 4 this purpose, we group the 213 proteins in our data set according to their predicted av- 5 erage hydrophobicity under the same mutational process and compare the substitution 6 rates of groups characterized by diﬀerent hydophobicity. In Fig.4, each point represents a 7 group of proteins with similar mean hydrophobicity. Each curve is obtained for a diﬀer- 8 ent mutation process with its background distribution P mut and exchangeability matrix 9 Emut. One can see that, for the ﬂux mutation process (black circles in Fig.4), more polar 10 proteins tend to evolve more rapidly, consistent with what we observed for polar sites in 11 Fig.3. Once again, this behavior may be explained by considering that more polar amino 12 acids have higher mutational ﬂuxes. 13 The above also holds for the nuc opt model, in which the mutation process is separately 14 optimized for each protein, but in this case when the WT selection model is used the 15 maximum of the substitution rate is achieved for proteins of intermediate hydrophobicity 16 (see red squares in Fig.4B), consistent with the observation that with the nuc opt model 17 the hydrophobic sites may evolve slightly faster than the polar sites in some proteins 18 characterized by a mutation process that favors hydrophobic residues, see Supplementary 19 Figures S3 and S4). 20 17 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Manuscript to be reviewed Figure 5: Each point represents a set of protein sites with similar average hydrophobicity in the evolutionary model. The sequence entropy has a universal shape as a function of the average hydrophobicity, with a maximum at 0.14, which is the mean hydrophobicity of the uniform distribution of amino acids. Changes in the background distribution mostly shift the sequence entropy curves without changing the position of the maximum, but they aﬀect the values of entropy. The largest entropies are obtained for the mutation model optimized for each protein sequence (thick black line) and for the mutation bias with GC content equal to 0.40, which yield only slightly hydrophobic sequences. Figure 5: Each point represents a set of protein sites with similar average hydrophobicity in the evolutionary model. The sequence entropy has a universal shape as a function of the average hydrophobicity, with a maximum at 0.14, which is the mean hydrophobicity of the uniform distribution of amino acids. Changes in the background distribution mostly shift the sequence entropy curves without changing the position of the maximum, but they aﬀect the values of entropy. The largest entropies are obtained for the mutation model optimized for each protein sequence (thick black line) and for the mutation bias with GC content equal to 0.40, which yield only slightly hydrophobic sequences. which has a small bias towards slightly hydrophobic sequences. Extreme mutation bias 1 both towards hydrophobic (low GC) and polar (high GC) amino acids yield very reduced 2 sequence entropies, which means that tselection must impose stronger constraints in order 3 to preserve the average hydrophobicity needed for stable proteins. This result is consistent 4 with the ﬁnding that, for equal population size, the average ﬁtness achieved in evolution 5 has a maximum as a function of the mutation bias, and it is low for extreme mutation 6 bias either toward hydrophobic or towards hydrophylic sequences (Mendez et al 2010; 7 Bastolla, Dehouck & Echave, 2017). 8 Inﬂuence of the mutation bias on sequence entropies 27 Next, we study how the shape of the entropy-hydrophobicity curve depends on the muta- 28 tion bias. In Fig.5 each point represents a set of protein sites with similar hydrophobicity 29 in the stability-constrained evolutionary model. The sequence entropy has an almost uni- 30 versal bell shape as a function of the average hydrophobicity of the site, with a maximum 31 when the average hydrophobicity is approximately 0.14, which is the average hydropho- 32 bicity of the uniform distribution of amino acids. This result is of course not surprising, 33 since sites with very high or low average hydrophobicity have distributions that favor only 34 the most hydrophobic or polar amino acids, while all amino acids are allowed when the 35 average hydrophobicity equals the one of the uniform distribution. 36 Changes in the mutation bias do not change the position of the maximum, but they 37 strongly shift the sequence entropy curves in the vertical direction. The largest entropies 38 are obtained for the mutation model nuc opt optimized separately from each PDB se- 39 quence (thick black line) and for the mutation bias with G+C content equal to 0.40, 40 18 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Inﬂuence of hydrophobicity on substitution rates 9 We now study the relationship between site-speciﬁc hydrophobicity and site-speciﬁc sub- 10 stitution rate. As in the previous ﬁgure, also in Fig.6 each point represents a set of protein 11 sites with similar average hydrophobicity in the stability-constrained evolutionary model, 12 and we plot the average substitution rate versus the average hydrophobicity. Diﬀerent 13 from the shape of the sequence entropy, the shape of the substitution rate curve clearly 14 depends on the mutation bias. The average hydropobicity at which the maximum substi- 15 tution rate is achieved decreases with the G+C content or, equivalently, it increases with 16 the average hydrophobicity of the mutation process. In other words, when the mutation 17 process favors the exchange of hydrophobic amino acids, the maximum of the substitution 18 rate is achieved at sites that are more hydrophobic, as expected on the basis of Eq.(10) 19 19 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Figure 6: Each point represents a set of protein sites with similar hydrophobicity in the evolutionary model. The substitution rate shows a maximum whose position depends on the mutation process. The hydropobicity at which the maximum rate is achieved increases with the mean hydrophobicity of the mutation process (lower GC content). The substitution rates tend to increase for mutation processes that yield higher hydrophobicity (lower GC content), but for the MF model (plot A) they also increase for extremely polar mutation bias (GC content 0.8). Figure 6: Each point represents a set of protein sites with similar hydrophobicity in the evolutionary model. The substitution rate shows a maximum whose position depends on the mutation process. The hydropobicity at which the maximum rate is achieved increases with the mean hydrophobicity of the mutation process (lower GC content). The substitution rates tend to increase for mutation processes that yield higher hydrophobicity (lower GC content), but for the MF model (plot A) they also increase for extremely polar mutation bias (GC content 0.8). that suggests that sites where the site-speciﬁc selective factors are correlated with the 1 mutational ﬂux have higher substitution rates. This result conﬁrms that the mutation 2 process has a strong inﬂuence on the substitution rates. Inﬂuence of hydrophobicity on substitution rates 9 3 Consistent with Fig.4, the substitution rate at the maximum tends to increase for 4 mutation processes that favor higher hydrophobicity (lower G+C bias), but for the MF 5 model (plot A) they also increase for extremely polar mutation bias (G+C content 0.8). 6 Consistent with the results reported in Fig.2, the ﬂux model of the exchangeability matrix 7 (thick black line) predicts higher substitution rates than the nuc opt model (red) when 8 applied together with the MF model (Fig.4A), but when it is applied together with the 9 WT model it predicts lower substitution rates at hydrophobic sites with many contacts 10 (Fig.4B), suggesting that the WT model is eﬀective at removing the eﬀect of selection 11 from the ﬂux model at these buried sites. 12 Manuscript to be reviewed measures are positively correlated, as seen from Fig.2, which shows that the substitution 1 rate tends to increase for sites with higher sequence entropy. Nevertheless, sites with 2 the same sequence entropy are characterized by diﬀerent substitution rates, which are 3 systematically higher for polar sites than for hydrophobic sites (Fig.3). This diﬀerence is 4 not due to diﬀerent selective constraints, which are quantiﬁed by sequence entropy, but 5 it is due to the diﬀerent exchangeability of polar and hydrophobic amino acids, which 6 does not inﬂuence the sequence entropy but inﬂuences the substitution rate according 7 to Eq.(10). This equation shows that the substitution rate is larger at sites where the 8 site-speciﬁc selective factors are correlated with the global mutational ﬂux. In particular, 9 at exposed sites polar residues have higher selective factors, and under the ﬂux model 10 these residues are characterized by high mutational ﬂuxes, which explains why exposed 11 sites tend to evolve faster than buried sites with the same sequence entropy. This result is 12 independent of the protein structure and robust with respect to changes of the selection 13 model (WT or MF). As a consequence, more polar proteins are predicted to evolve faster 14 than proteins with large mean hydrophobicity (Fig.4). 15 When we apply the nuc opt codon mutation model based on nucleotide frequencies 16 separately optimized for each protein, we still observe that sites with the same entropy 17 evolve with diﬀerent rates depending on whether they are exposed or buried, but the dif- 18 ferences decrease (Fig.3) and, for some proteins, buried sites may evolve faster than polar 19 sites with the same entropy (Supplementary Fig.S3 and S4), although there is still a nega- 20 tive correlation between the substitution rate of a protein and its hydrophobicity (Fig.4). 21 This is consistent with the observation that, for low G+C mutation bias that favor hy- 22 drophobic residues, the correlation between the substitution rate and the hydrophobicity 23 of proteins becomes positive (see Fig.4), as expected based on Eq.(10). 24 We then compare diﬀerent mutation biases applied to all proteins of our data set. 25 The average substitution rates tend to be larger for mutation bias favoring hydrophobic 26 residues (low G+C) (Fig.4 and Fig.6). Discussion and conclusions 13 Here we studied how the evolutionary variability of proteins is inﬂuenced by the under- 14 lying mutation process, adopting a model of stability-constrained protein evolution with 15 selection on the stability of the native state against both unfolding and misfolding. 16 We found that the sequence entropy and the substitution rate are not equivalent 17 measures of the evolutionary variability of the protein sites, as it was expected based on 18 the arguments presented in the seminal paper by Halpern and Bruno (1998). These two 19 20 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Thus, exposed sites are aﬀected by weaker selective constraints than buried sites 24 (they have higher entropy, see Fig.5), but these selective constraints become more severe 25 when the mutation bias becomes extreme. 26 Therefore, the substitution rate is systematically inﬂuenced by the mutation bias 7 (Fig.6), in such a way that when the mutation bias favors more hydophobic proteins (low 8 G+C) the substitution rate increases and its maximum is achieved at sites that are more 9 hydrophobic, as expected from Eq.10). On the contrary, the curve of the sequence entropy 10 versus the average hydrophobicity has a shape that does not depend on the mutation bias. 11 In particular, the position of the maximum always coincides with the value 0.14, which is 12 the average hydrophobicity of the uniform distribution of amino acids. As a consequence, 13 the site-speciﬁc average hydrophobicity at which the sequence entropy is maximal does 14 not coincide in general with the average hydrophobicity at which the substitution rate is 15 maximal, and this discrepancy becomes larger when the mutation bias is more extreme, 16 causing larger diﬀerences between the two measures of evolutionary variability. 17 Finally, changes of mutation bias severely aﬀect the selective constraints imposed 18 on the protein sites, attaining maximum values of the entropy when the mutation bias 19 is G+C=0.40 and decreasing the site-speciﬁc sequence entropies when the mutation bias 20 becomes more extreme both towards hydrophobic and towards polar residues (Fig.5). The 21 sequence entropy of exposed polar sites decreases with the mutation bias more strongly 22 than the entropy of buried hydrophobic sites, where the diﬀerent curves in Fig.5 tend to 23 collapse. Thus, exposed sites are aﬀected by weaker selective constraints than buried sites 24 (they have higher entropy, see Fig.5), but these selective constraints become more severe 25 when the mutation bias becomes extreme. 26 Acknowledgements 27 We thank Juli´an Echave for providing us the empirical multiple sequence alignments that 28 were used for the leftmost plots in Fig.2, and we acknowledge interesting discussions with 29 him, Alberto Pascual-Garc´ıa and Nick Goldman. Research at CBMSO is facilitated by 30 the Fundaci´on Ram´on Areces. 31 Manuscript to be reviewed When the mutation bias towards A+T increases, the mutational ﬂux P mut a P mut b Emut ab of 1 hydrophobic residues increases and the site-speciﬁc substitution rate given by Eq.(10) 2 increases at buried sites characterized by skewed F sel,i a more rapidly than it decreases at 3 exposed sites, as one can see from Fig.6 that shows a large increase of the substitution rate 4 at buried sites with large average hydrophobicity while the substitution rate at exposed 5 sites depends little on the mutational bias. 6 exposed sites, as one can see from Fig.6 that shows a large increase of the substitution rate 4 at buried sites with large average hydrophobicity while the substitution rate at exposed 5 sites depends little on the mutational bias. 6 Therefore, the substitution rate is systematically inﬂuenced by the mutation bias 7 (Fig.6), in such a way that when the mutation bias favors more hydophobic proteins (low 8 G+C) the substitution rate increases and its maximum is achieved at sites that are more 9 hydrophobic, as expected from Eq.10). On the contrary, the curve of the sequence entropy 10 versus the average hydrophobicity has a shape that does not depend on the mutation bias. 11 In particular, the position of the maximum always coincides with the value 0.14, which is 12 the average hydrophobicity of the uniform distribution of amino acids. As a consequence, 13 the site-speciﬁc average hydrophobicity at which the sequence entropy is maximal does 14 not coincide in general with the average hydrophobicity at which the substitution rate is 15 maximal, and this discrepancy becomes larger when the mutation bias is more extreme, 16 causing larger diﬀerences between the two measures of evolutionary variability. 17 Finally, changes of mutation bias severely aﬀect the selective constraints imposed 18 on the protein sites, attaining maximum values of the entropy when the mutation bias 19 is G+C=0.40 and decreasing the site-speciﬁc sequence entropies when the mutation bias 20 becomes more extreme both towards hydrophobic and towards polar residues (Fig.5). The 21 sequence entropy of exposed polar sites decreases with the mutation bias more strongly 22 than the entropy of buried hydrophobic sites, where the diﬀerent curves in Fig.5 tend to 23 collapse. Manuscript to be reviewed Thus, the comparison of proteins with diﬀerent 27 hydrophobicity under the same mutation model and the comparison between diﬀerent 28 mutation processes, such as those happening in diﬀerent bacterial genomes, yield con- 29 trasting results for the substitution rates: substitution rates tend to be higher for more 30 polar proteins evolving under the same mutation process, but they tend to be higher 31 in organisms with mutation bias towards A+T that favor hydrophobic residues, such as 32 intracellular bacteria. Note that the substitution rates also increase for mutation bias 33 favoring very polar amino acids (high G+C), but the latter happens only when the MF 34 model of selection is applied. 35 As Eq.(10) shows, the higher substitution rate for equal sequence entropy observed at 36 exposed sites is attributable to the higher exchangeability of polar residues, which is a 37 property of the mutational process. In contrast, the diﬀerences in substitution rates that 38 we observe for proteins evolving under diﬀerent mutational processes (Fig.6) is likely to be 39 caused at least in part by natural selection. In fact, buried sites are characterized by lower 40 entropy than exposed sites (Fig.5), which indicates that they experience stronger selective 41 constraints and their selective factors F sel,i a are more skewed towards hydrophobic residues. 42 21 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) References 32 [1] Arenas M., Sanchez-Cobos A. and Bastolla, U. (2015) Maximum likelihood phyloge- 33 netic inference with selection on protein folding stability. Mol Biol. Evol. 32:2195-207. 34 22 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed [2] Arenas M., Weber C.C., Liberles D.A. and Bastolla, U. (2017) ProtASR: An Evolu- 1 tionary Framework for Ancestral Protein Reconstruction with Selection on Folding 2 Stability. Syst Biol. 66:1054-1064. 3 [3] Bastolla U, Porto M, Roman HE, Vendruscolo M (2005) Principal eigenvector of 4 contact matrices and hydrophobicity proﬁles in proteins. Proteins 58:22-30. 5 [4] Bastolla U., Porto, M., Roman, H.E.and Vendruscolo, M. 2006. A protein evolution 6 model with independent sites that reproduces site-speciﬁc amino acid distributions 7 from the Protein Data Bank BMC Evol. Biol. 6:43. 8 [5] Bastolla U, Ortiz AR, Porto M, Teichert F. (2008) Eﬀective connectivity proﬁle: a 9 structural representation that evidences the relationship between protein structures 10 and sequences. Proteins 73:872-88. 11 [6] Bastolla U, Dehouck Y, Echave J (2017) What evolution tells us about protein 12 physics, and protein physics tells us about evolution. Curr Opin Struct Biol. 42:59-66. 13 [7] Bastolla, U.; Vendruscolo, M.; Knapp, E.W. (2000) A statistical mechanical method 14 to optimize energy functions for protein folding. Proc. Natl. Acad. Sci. USA 97, 15 3977-3981. 16 [8] Berezovsky IN, Zeldovich KB, Shakhnovich EI (2007) Positive and negative design 17 in stability and thermal adaptation of natural proteins. PLoS Comput Biol 3:e52. 18 [9] Bryngelson, J.-D.; Onuchic, J.-N.; Socci, N.-D.; Wolynes, P.-G (1995) Funnels, path- 19 ways, and the energy landscape of protein folding: A synthesis. Proteins 21, 167-195. 20 [10] Derrida, B. (1981) Random Energy Model: An exactly solvable model of disordered 21 systems. Phys. Rev. B 24, 2613. 22 [11] Echave J. (2008) Evolutionary divergence of protein structure: The linearly forced 23 elastic network model. Chem Phys Lett 457, 413-416. 24 [12] Echave J., Jackson E.L. and Wilke C.O. (2015) Relationship between protein ther- 25 modynamic constraints and variation of evolutionary rates among sites. Phys Biol. 26 12:025002. 27 [13] Echave J, Spielman SJ, Wilke CO (2016) Causes of evolutionary rate variation among 28 protein sites. Nat Rev Genet. 17:109-21. 29 [14] Franzosa, E. A., and Y. Xia, 2009 Structural determinants of protein evolution are 30 context-sensitive at the residue level. Mol. Biol. Evol. 26: 2387-2395. 31 [15] Garel, T.; Orland, H. (1988) Mean-ﬁeld model for Protein Folding, Europhys Lett, 32 307-310. 33 23 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Manuscript to be reviewed [16] Goldstein RA (2011) The evolution and evolutionary consequences of marginal ther- 1 mostability in proteins. Proteins 79:1396-1407. 2 [17] Goldstein RA and Pollock DD (2017) Sequence entropy of folding and the absolute 3 rate of amino acid substitutions Nat Ecol Evol. 1:1923-1930. 4 [18] Grishin NV; Wolf YI and Koonin EV (2000) From Complete Genomes to Measures 5 of Substitution Rate Variability Within and Between Proteins. Genome Res. 10: 6 991-1000. 7 [19] Halpern A and Bruno WJ (1998) Evolutionary distances for protein-coding se- 8 quences: modeling site-speciﬁc residue frequencies. Mol. Biol. Evol. 15(7):910-917. 9 [20] Huang TT, del Valle Marcos ML, Hwang JK, Echave J (2014) A mechanistic stress 10 model of protein evolution accounts for site-speciﬁc evolutionary rates and their 11 relationship with packing density and ﬂexibility. BMC Evol Biol. 9;14:78. 12 [21] Jimenez MJ, Arenas M, Bastolla U (2018) Substitution rates predicted by stability- 13 constrained models of protein evolution are not consistent with empirical data. Mol 14 Biol Evol. 35, 743-755. 15 [22] Jones DT, Taylor WR, Thornton JM (1992) The rapid generation of mutation data 16 matrices from protein sequences. Comput Appl Biosci 8:275-82. 17 [23] Kimura M (1962) On the probability of ﬁxation of mutant genes in a population. 18 Genetics 4:713-719. 19 [24] Mendez R, Fritsche M, Porto M, Bastolla U (2010= Mutation bias favors protein 20 folding stability in the evolution of small populations. PLoS Comput Biol 6:e1000767. 21 [25] Minning J, Porto M, Bastolla U (2013) Detecting selection for negative design in 22 proteins through an improved model of the misfolded state. Proteins 81:1102-1112. 23 [26] Mustonen V, L¨assig M (2005). Evolutionary population genetics of promoters: pre- 24 dicting binding sites and functional phylogenies. Proc Natl Acad Sci USA 102:15936- 25 41. 26 [27] Noivirt-Brik O, Horovitz A, Unger R (2009) Trade-oﬀbetween positive and negative 27 design of protein stability: from lattice models to real proteins. PLoS Comput Biol 28 5:e1000592. 29 [28] Ohta T (1976) Role of very slightly deleterious mutations in molecular evolution and 30 polymorphism, Theor. Pop. Biol. 10, 254-275. 31 [29] Porto M, Roman HE, Vendruscolo M, Bastolla U (2005) Prediction of site-speciﬁc 32 amino acid distributions and limits of divergent evolutionary changes in protein se- 33 quences. Mol Biol Evol 22:630-8. 34 24 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) Manuscript to be reviewed Manuscript to be reviewed [30] Pupko T, Bell RE, Mayrose I, Glaser F, Ben-Tal N (2002) Rate4Site: an algorithmic 1 tool for the identiﬁcation of functional regions in proteins by surface mapping of 2 evolutionary determinants within their homologues. Bioinformatics. 18 Suppl 1:S71- 3 7. 4 [31] Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstruct- 5 ing phylogenetic trees. Mol Biol Evol 4:406-25. 6 [32] Scherrer MP, Meyer AG, Wilke CO (2012) Modeling coding-sequence evolution 7 within the context of residue solvent accessibility. BMC Evol Biol. 12:179. 8 [33] Sella G, Hirsh AE (2005) The application of statistical physics to evolutionary biol- 9 ogy. Proc Natl Acad Sci USA 102:9541-6. 10 [34] Serohijos AW, Shakhnovich EI (2014) Merging molecular mechanism and evolution: 11 theory and computation at the interface of biophysics and evolutionary population 12 genetics. Curr Opin Struct Biol 2014, 26:84-91. 13 [35] Shakhnovich, E.-I.; Gutin, A.-M. (1989) Formation of unique structure in polypeptide 14 chains, Biophys. Chem. 34, 187. 15 [36] Tirion MM (1996) Large Amplitude Elastic Motions in Proteins from a Single- 16 Parameter, Atomic Analysis. Phys Rev Lett. 77:1905-1908. 17 [37] Whelan S, Goldman N (2001) A general empirical model of protein evolution derived 18 from multiple protein families using a maximum-likelihood approach. Mol Biol Evol 19 18:691-9. 20 [38] Yeh SW, Liu JW, Yu SH, Shih CH, Hwang JK, Echave J. Site-speciﬁc structural 21 constraints on protein sequence evolutionary divergence: local packing density versus 22 solvent exposure. Mol Biol Evol. 2014 31:135-9. 23 25 PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018) PeerJ reviewing PDF \| (2018:04:27285:1:1:NEW 1 Aug 2018)
https://openalex.org/W2283526670	https://biblio.ugent.be/publication/7240462/file/7240467.pdf	English	null	Adaptive initial step size selection for Simultaneous Perturbation Stochastic Approximation	SpringerPlus	2,016	cc-by	8,703	METHODOLOGY Open Access © 2016 Ito and Dhaene. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Correspondence: itokeiic@gmail.com; keiichiito@netscape.net 1 Ghent University - iMinds, INTEC, Gaston Crommenlaan 8 bus 201, Ledeberg, 9050 Ghent, Belgium Full list of author information is available at the end of the article Adaptive initial step size selection for Simultaneous Perturbation Stochastic Approximation Keiichi Ito1,2 and Tom Dhaene1 Ito and Dhaene SpringerPlus (2016) 5:200 DOI 10.1186/s40064-016-1823-3 Ito and Dhaene SpringerPlus (2016) 5:200 DOI 10.1186/s40064-016-1823-3 Abstract A difficulty in using Simultaneous Perturbation Stochastics Approximation (SPSA) is its performance sensitivity to the step sizes chosen at the initial stage of the iteration. If the step size is too large, the solution estimate may fail to converge. The proposed adaptive stepping method automatically reduces the initial step size of the SPSA so that reduction of the objective function value occurs more reliably. Ten mathemati- cal functions each with three different noise levels were used to empirically show the effectiveness of the proposed idea. A parameter estimation example of a nonlinear dynamical system is also included. Keywords: Stochastic approximation, Optimization, Direct method, Noisy function, Parameter estimation Background Equation (2) is analogous to the gradient descent algorithm in nonlinear programming, in which gk is the gradient of the objective function ∇f ( ˆθk). The difference is that in Eq. (2), ˆgk represent gradients stochastically and the effect of the noise or deviation from the true gradient is expected to cancel out as the iteration count k increases. The step sizes ak are normally prescribed in SPSA and FDSA as a function of k just like the Simulated- Annealing’s (Kirkpatrick et al. 1983) cooling schedule. This is because these methods do not assume deterministic responses in the measurements of the objective function values. Thus, unlike the nonlinear programming counterparts, adaptation of step sizes based on gradients and amount of descent achieved (such as in the line search) is usually not done in the stochastic approximation optimization methods. The rationale of the Eq. (2) is intuitively depicted in Fig. 1 for one variable case. where ˆgk( ˆθk) is the estimate of the gradient vector g( ˆθ) at iteration k based on the meas- urements of the objective function. The ak is the step size at iteration k. Equation (2) is analogous to the gradient descent algorithm in nonlinear programming, in which gk is the gradient of the objective function ∇f ( ˆθk). The difference is that in Eq. (2), ˆgk represent gradients stochastically and the effect of the noise or deviation from the true gradient is expected to cancel out as the iteration count k increases. The step sizes ak are normally prescribed in SPSA and FDSA as a function of k just like the Simulated- Annealing’s (Kirkpatrick et al. 1983) cooling schedule. This is because these methods do not assume deterministic responses in the measurements of the objective function values. Thus, unlike the nonlinear programming counterparts, adaptation of step sizes based on gradients and amount of descent achieved (such as in the line search) is usually not done in the stochastic approximation optimization methods. The rationale of the Eq. (2) is intuitively depicted in Fig. 1 for one variable case. Under appropriate conditions, the iteration in Eq. (2) will converge to the optimum θ∗ in some stochastic sense. The hat symbol indicates an “estimate”. Thus, ˆθk denotes the estimate of the optimum θ∗ at iteration k. Let y(·) denote a measurement of the objective Fig. Background Simultaneous Perturbation Stochastic Approximation (SPSA) (Spall 1992) is an opti- mization algorithm that uses only objective function measurements in the search of solutions. Applications of SPSA include model-free predictive control (Dong and Chen 2012a, b; Ko et al. 2008), signal timing for vehicle timing control (Spall and Chin 1997), air traffic network (Kleinman et al. 1997), and marine vessel traffic management (Bur- nett 2004). More applications are mentioned in the introductory article by Spall (1998b). SPSA has been used successfully in many optimization problems that have high-dimen- sional input parameter space and the objective value is not deterministic (SPSA 2001). In this optimization method, the initial design parameter vector θ of D-dimensions is perturbed simultaneously in every dimension, i.e. by adding and subtracting a per- turbation vector of D-dimensions, thus obtaining an estimate of the gradient vector g. Unlike the traditional finite differencing approach, it only takes two function evalu- ations to obtain the estimate of the gradient. Yet, the number of iteration needed for convergence to the optimum is said to be more or less the same with Finite-Difference Stochastic Approximation (FDSA) (Kiefer and Wolfowitz 1952), which in essence is an approximate steepest-descent method that uses finite-differencing to approximate the partial derivatives along each of the D parameters. Thus, the number of function Ito and Dhaene SpringerPlus (2016) 5:200 Page 2 of 18 evaluations of SPSA is D-fold smaller compared to FDSA (Spall 1998b). An extension to this method exists to include second-order (Hessian) effects to accelerate convergence (Spall 2000, 2009; Zhu and Spall 2002). However, we will not treat this enhancement here. The problem solved by SPSA in this paper can be formulated as following. (1) min θ∈Θ f (θ), (1) min θ∈Θ f (θ), where f (θ) is the objective function and θ is a D-dimensional vector of parameters. We assume that each element in the vector θ is a real number and has upper and lower bounds that defines the Cartesian product domain Θ. The SPSA and FDSA procedures are in the general recursive form: (2) ˆθk+1 = ˆθk −ak ˆgk( ˆθk), ˆθk+1 = ˆθk −ak ˆgk( ˆθk), (2) where ˆgk( ˆθk) is the estimate of the gradient vector g( ˆθ) at iteration k based on the meas- urements of the objective function. The ak is the step size at iteration k. Background 1 Objective value minimization using gradient descent (one variable): if gradient g is positive at θk then move to θk+1 < θk, if gradient g is negative then move to θk+1 > θk Fig. 1 Objective value minimization using gradient descent (one variable): if gradient g is positive at θk then move to θk+1 < θk, if gradient g is negative then move to θk+1 > θk Fig. 1 Objective value minimization using gradient descent (one variable): if gradient g is positive at θk the move to θk+1 < θk, if gradient g is negative then move to θk+1 > θk Ito and Dhaene SpringerPlus (2016) 5:200 Page 3 of 18 function f (·) at parameter value denoted by “·” and ck be some small positive number. The measurements are assumed to contain some noise, i.e. y(·) = f (·) + noise. In SPSA, the ith component ˆgki( ˆθk) of the gradient vector ˆgk( ˆθk) is formed from a ratio involving the individual components in the perturbation vector and the difference in the two cor- responding measurements. For two-sided simultaneous perturbations, we have (3) ˆgki( ˆθk) = y( ˆθk + ckk) −y( ˆθk −ckk) 2ck∆ki , (3) where the D-dimensional random perturbation vector where the D-dimensional random perturbation vector (4) k = ∆k0, ∆k1, . . . , ∆k(D−1) T, (4) k = ∆k0, ∆k1, . . . , ∆k(D−1) T, follows a specific statistical distribution criterion. Here, i is the parameter index. A sim- ple choice for each component of k is to use Bernoulli ±1 distribution, which is essen- tially a random switching between +1 and −1. The Bernoulli distribution is proven to be an optimal distribution for the simultaneous perturbation (Sadegh and Spall 1997). Note also that in the Eq. (3), we do not evaluate y( ˆθk). The recursive equation (2) proceeds with only the responses from the two perturbed inputs y( ˆθk + ckk) and y( ˆθk −ckk). The choice of ak and ck is critical to the performance of SPSA and suggested values can be found in Spall (1998a). Background At given iteration k: (5) ak = a (A + k + 1)α , (6) ck = c (k + 1)γ , ak = a (A + k + 1)α , ck = c (k + 1)γ , (5) (6) where α = 0.602, γ = 0.101, c ≃standard deviation of measurement noise , A ≤10 % of maximum number of iterations , a = δ ˆθ0min (A+1)α \|ˆg0i( ˆθ0)\|, k = iteration index starting with 0, δ ˆθ0min = smallest initial change desired in a parameter. The setting for α and γ above are not optimal in the asymptotic sense, but are adapted to finite iteration settings. In practice, one of the drawbacks of SPSA is that one has to find good values for a and c, as both affect the performance of the algorithm Spall (2003, pp. 165–166) (Altaf et al. 2011; Shen et al. 2012; Radac et al. 2011; Easterling et al. 2014; Taflanidis and Beck 2008). However, for c, we have a tangible measure, which is the output measurement error (Spall 1998a), to select a proper value up front. If the function response is noiseless, c is usually not a critical parameter. On the other hand, a is more problematic, because no clear measure exists. It is possible to work with δ ˆθ0min instead of a, but a priori assignment of its value is still non-trivial if little is known about the func- tion that we are trying to optimize. A larger value of a generally produces better results compared to a smaller value of a. However, this also increases the chance that the optimization diverges to a worse solu- tion than the starting point. Very often, the user of SPSA has to find as big a as possible that would not cause divergence. To avoid divergence, an adaptation called “blocking” exists (Easterling et al. 2014; Spall 1998a) in which the objective values at ˆθk is evaluated in addition to the two Ito and Dhaene SpringerPlus (2016) 5:200 Page 4 of 18 perturbations. If the new objective function value is “significantly worse” than the cur- rent objective function value, the updating of ˆθk does not happen. The extra function evaluation at each iteration increases the cost of iteration by 33 %. In addition, a problem dependent threshold parameter to block the ˆθk update needs to be set up by the user. Background Another way to mitigate divergence is to modify the gradient approximation ˆgk by “scaling” and “averaging” (Andradóttir 1996; Xu and Wu 2013). However, the methods proposed in the literature require set up of additional threshold parameters critical to their performance. Furthermore, their methods require additional gradient estimations per iteration. Stochastic Gradient Descent (SGD) methods use noisy information of the gradient of the objective functions. On the other hand, Stochastic Approximation methods such as FDSA and SPSA only uses measurement of noisy objective values. Therefore, adaptive determination of step sizes based on (approximate) gradients and inverse Hessians in SGD literature (such as in Zeiler (2012), Bottou (2010)) may not be directly applicable to or feasible in SPSA. Convergence conditions also differ between the two. Although this does not exclude the possibility of successful import of ideas from SGD literature, in this paper, we will not delve into this direction. This paper provides a solution to determine the appropriate values of a by introduc- ing an adaptive scheme as discussed in “Adaptive initial step sizes” section. It does not require any additional objective function evaluations per iteration nor extra problem dependent parameters to set up. Adaptive initial step sizes To remedy the sensitivity to a, we propose an adaptive stepping algorithm. At the end of each iteration k, we perform the adjustment described in Algorithm 1. Algorithm 1 Adaptive Initial Step 1: if min{y(ˆθk + ck∆k), y(ˆθk −ck∆k)} −y(ˆθ0) ≥0 then 2: ˆθk+1 = ˆθb, where ˆθb gives the best y so far 3: a ←0.5a 4: end if Algorithm 1 Adaptive Initial Step Comments on convergence Currently available theories of stochastic algorithms are almost all based on asymp- totic properties with k →∞, and SPSA is no exception. For given conditions Spall (2003, p. 183), SPSA is proven to converge to a local optima almost surely. However, under limited function evaluation budget, we frequently encounter situations in which SPSA returns worse solution than the initial i.e. divergence. The method we propose is a practical remedy conceived in a finite k setting. We will show, in the next section, its effectiveness empirically via numerical experiments with k in the order of 103. f For ˆθk to converge to the optimal solution θ∗ in infinite steps, the following condi- tions are required for ak and ck (Spall 1992): ak, ck > 0 for all k; ak, ck →0 as k →∞; ∞ k=0 ak = ∞, and ∞ k=0 ak ck 2 < ∞. With Algorithm 1, ∞ k=0 ak = ∞ is not guaran- teed. For example, if the reduction of a happens in every iteration k, the sum is conver- gent. In practice, the numbers of function evaluations are finite, and reductions of a are expected to happen only a limited number of times. Therefore, this violation is expected to pose little problem. The intention of the proposed method is not to modify the asymptotic convergence rate of the original SPSA algorithm Spall (2003, pp. 186–188). The adaptive step takes place only if it is suspected that the objective value has become larger than at the starting point ˆθ0. The probability of Algorithm 1 taking place is expected to go to zero under reason- able signal-to-noise ratio as f ( ˆθk) decreases. The worst situation that can happen is that the every perturbation ckk produces worsening moves and no improvement is obtained compared to the starting point θ0. In “Computational results” section, we will confirm empirically what we have described about the convergence in finite k settings (k ∼103). Another reason to take the objective value at the starting point as the threshold value to judge divergence is that if we update this value with y( ˆθk), where k > 0, we may risk picking a point that is too low due to the noise incurred in the measurement y. This in turn inhibits further improvement of ˆθk for lower objective values. Algorithm 2 Pseudocode of the Proposed Algorithm Algorithm 2 Pseudocode of the Proposed Algorithm 1: Initialize a and c (or set δˆθ0min min (upper bound −lower bound), and c std of response noise). Set maximum number of it- erations maxiter. 2: Obtain initial measurement y(ˆθ0), and let θb = ˆθ0. 3: for k = 0 to maxiter do 4: Compute ∆k and ck. 5: Evaluate y(ˆθk + ck∆k) and y(ˆθk −ck∆k). 6: Record the input parameter vector as ˆθb if better min- imum in y is obtained. 7: Compute ˆgki(ˆθk). 8: Compute ak. 9: ˆθk+1 = ˆθk −akˆgk(ˆθk). 10: Perform Algorithm 1. 11: end for 1: Initialize a and c (or set δˆθ0min min (upper bound −lower bound), and c std of response noise). Set maximum number of it- erations maxiter. 2: Obtain initial measurement y(ˆθ0), and let θb = ˆθ0. 3: for k = 0 to maxiter do 4: Compute ∆k and ck. ˆ Algorithm 1 Adaptive Initial Step 1: if min{y(ˆθk + ck∆k), y(ˆθk −ck∆k)} −y(ˆθ0) ≥0 then 2: ˆθk+1 = ˆθb, where ˆθb gives the best y so far 3: a ←0.5a 4: end if The condition requires that at least one of the two parameter perturbations produce a better (smaller) measurement of the objective function than that of initial guess of parameters ˆθ0 to proceed without modifying a. Therefore, at each iteration k, the smaller of the two measurements of the objective function values of perturbed parameters is compared to that of the initial value at iteration k = 0. If the measurements of the objec- tive values of the perturbed parameters are larger, ˆθk is reset to θb, which is the point that gave the minimum in the history of iteration and a is reduced to half of its previous value. A pseudocode of the proposed SPSA with the adaptive initial step is shown in Algorithm 2. The difference between the standard SPSA and our SPSA is in line 10. Page 5 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 Comments on convergence In the following section, the smallest output of mathematical functions will be sought using the standard SPSA and our adaptive initial stepping SPSA. This will show the sen- sitivity of the function value in the final iteration to the initial step size δ ˆθ0min and so the sensitivity to a, and how the adaptive initial stepping substantially mitigates the difficulty to find the proper initial perturbation magnitude. Page 6 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 Computational results In this section, we will compare the original SPSA and our modified SPSA as described in Algorithm 2 using 10 analytical test functions and a parameter estimation example of a nonlinear dynamic system. Test functionsf To see the effect of the new adaptive stepping algorithm in SPSA, the minimum points of ten different mathematical test functions were sought. Except for Griewank function, the following conditions were applied. The functions’ responses were minimized from arbitrary starting points ˆθ0 ∈[−2, 2]D (D-dimensional product space with lower bound -2 and upper bound 2). If ˆθk = [ ˆθk0, ˆθk1, · · · , ˆθki, · · · , ˆθk(D−1)]T exceeded [−10, 10] in any of its D dimensions, that parameter was replaced by -10 if it was less than -10 or was replaced by 10 if it was larger than 10. For Griewank function, it was randomly started from ˆθ0 ∈[−120, 120]D. If ˆθk exceeded [−600, 600] in any of its D dimensions, that parameter was replaced by −600 if it was less than −600 or was replaced by 600 if it was larger than 600. For all ten functions, the iteration was stopped when 2000 evaluations of the objective function were reached. For convenience, we will label our proposed algo- rithm as “A_SPSA” and the standard SPSA as “SPSA”. The optimizations for each of the ten objective functions were started from 20 differ- ent starting points. After the 2000 iterations, the distributions of objective values were plotted with respect to δ ˆθ0min. Eleven different values of δ ˆθ0min between 1.0 × 10−4 and 1.0 × 101 (up to 1.0 × 102 for Griewank) were used to make the plot. The dimensions of the functions were set to be 20, i.e. D = 20. The definitions of the ten functions are given in the following. The Rosenbrock func- tion is described as (7) f (θ) = D−2 i=0 100(θi+1 −θ2 i )2 + (θi −1)2 , i = 0, 1, . . . , D −1, D > 1, f (θ∗) = 0, θ∗ i = 1. f (θ) = D−2 i=0 100(θi+1 −θ2 i )2 + (θi −1)2 , i = 0, 1, . . . , D −1, D > 1, f (θ∗) = 0, θ∗ i = 1. (7) The Sphere function is described as f (θ) = D−1 i=0 θ2 i , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (8) f (θ) = D−1 i=0 θ2 i , i = 0, 1, . . . Test functionsf , D −1, f (θ∗) = 0, θ∗ i = 0. (8) The Schwefel function is described as f (θ) = D−1 j=0   j i=0 θi   2 , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (9) Page 7 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 The Rastrigin function is described as The Rastrigin function is described as (10) f (θ) = D−1 i=0 θ2 i −10 cos(2πθi) + 10 , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (10) The Skewed Quartic function Spall (2003, ex. 6.6) is described as (11) f (θ) = (Bθ)TBθ + 0.1 D−1 i=0 (Bθ)3 i + 0.01 D−1 i=0 (Bθ)4 i , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (11) where the matrix B in the Skewed Quartic function is a square matrix with upper tri- angular elements set to 1 and the lower triangular elements set to zero. The Griewank function is described as where the matrix B in the Skewed Quartic function is a square matrix with upper tri- angular elements set to 1 and the lower triangular elements set to zero. The Griewank function is described as (12) f (θ) = 1 + D−1 i=0 θ2 i 4000 − D−1 i=0 cos( θi √ i ), i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (12) The Ackley function is described as The Ackley function is described as f (θ) = −20 exp  −0.2 1 D D−1 i=0 θ2 i   −exp 1 D D−1 i=0 cos(2πθi) + 20 −exp(1), i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. f (θ) = −20 exp  −0.2 1 D D−1 i=0 θ2 i   −exp 1 D D−1 i=0 cos(2πθi) + 20 −exp(1), i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (13) The Manevich function is described as The Manevich function is described as f (θ) = D−1 i=0 (1 −θi)2/2j , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 1. (14) The Ellipsoid function is described as The Ellipsoid function is described as f (θ) = D−1 i=0 iθ2 i , f (θ) = D−1 i=0 iθ2 i , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. f (θ) i=0 iθi , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (15) i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. Page 8 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 The Rotated Ellipsoid function is described as The Rotated Ellipsoid function is described as The Rotated Ellipsoid function is described as f (θ) = D−1 i=0   i j=0 θ2 j   2 , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. f (θ) = D−1 i=0   i j=0 θ2 j   2 , i = 0, 1, . . . , D −1, f (θ∗) = 0, θ∗ i = 0. (16) Each of Figs. 2, 3, 4, 5, 6, 7, 8, 9, 10 and 11 show three different cases of noisy measure- ments of the outputs. The subfigures (a) have no noise added, subfigures (b) and (c) have Gaussian noise added to the true output with standard deviation σ of 0.1 and 1.0 respec- tively. In all the three noise levels of the ten functions, c = 0.2 was used.i A general trend observed from the figures is that when the initial step size is large, the original SPSA tends to diverge to big objective values. The SPSA with the proposed initial step size reduction, on the other hand, effectively mitigates this divergence prob- lem producing smaller objective values in general as the (a priori) initial step size is increased. This is because if the two function evaluations in the iteration are not smaller than the starting point value f ( ˆθ0), the algorithm will reduce the step size (by halving a) and restart at ˆθb, which is the point that gave the smallest output in the history of itera- tions. However, note that the iteration index k in ak and ck is not reinitialized. For the ten functions tested, A_SPSA achieved its best performance when δ ˆθ0min was close to 10 or 100 for Griewank function. The Ellipsoid function is described as This indicates that one can simply set the minimum pertur- bation δ ˆθ0min close to the magnitude of the difference between upper and lower bound a b c Fig. 2 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rosenbrock”. a No noise, b σ = 0.10, c σ = 1.0 b a b a c c Fig. 2 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rosenbrock”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 2 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rosenbrock”. a No noise, b σ = 0.10, c σ = 1.0 Page 9 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 a b c Fig. 3 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Sphere”. a No noise, b σ = 0.10, c σ = 1.0 a b b a c c Fig. 3 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Sphere”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 3 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Sphere”. a No noise, b σ = 0.10, c σ = 1.0 a b c Fig. 4 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Schwefel”. a No noise, b σ = 0.10, c σ = 1.0 b a b a a c c Fig. 4 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Schwefel”. a No noise, b σ = 0.10, c σ = 1.0 of the parameter in consideration. This may not be a guarantee for the best results but doing so does not cause the optimization to diverge to large responses and the results achieved are not substantially worse than the cases with best settings for a. of the parameter in consideration. This may not be a guarantee for the best results but doing so does not cause the optimization to diverge to large responses and the results achieved are not substantially worse than the cases with best settings for a. Page 10 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 a b c Fig. The Ellipsoid function is described as 5 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rastri- gin”. a No noise, b σ = 0.10, c σ = 1.0 a b b a a b c c Fig. 5 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rastri- gin”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 5 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rastri- gin”. a No noise, b σ = 0.10, c σ = 1.0 a b c Fig. 6 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Skewed Quartic”. a No noise, b σ = 0.10, c σ = 1.0 a b b a a b c c Fig. 6 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Skewed Quartic”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 6 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Skewed Quartic”. a No noise, b σ = 0.10, c σ = 1.0 As mentioned earlier, the value for c is important when the measurements of y con- tain noise. Figure 12 shows how the choice of c affects the outcome of optimizations. The figure shows the case of the 20 dimensional Sphere Function with Gaussian noise Page 11 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 a b c Fig. 7 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Grie- wank”. a No noise, b σ = 0.10, c σ = 1.0 b a b a c c Fig. 7 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Grie- wank”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 7 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Grie wank”. a No noise, b σ = 0.10, c σ = 1.0 having standard deviation σ = 0.1. Among the three values of c, namely 0.01, 0.1 and 1.0, c = σ = 0.1 gave the best results for A_SPSA. At c = 1.0, however, A_SPSA showed little improvement in the objective value regardless of δ ˆθ0min magnitude. This is caused a b c Fig. The Ellipsoid function is described as 8 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Ack- ley”. a No noise, b σ = 0.10, c σ = 1.0 a b b a c c g. 8 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Ack- ”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 8 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Ack- ley”. a No noise, b σ = 0.10, c σ = 1.0 having standard deviation σ = 0.1. Among the three values of c, namely 0.01, 0.1 and 1.0, c = σ = 0.1 gave the best results for A_SPSA. At c = 1.0, however, A_SPSA showed little improvement in the objective value regardless of δ ˆθ0min magnitude. This is caused Page 12 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 a b c Fig. 9 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Manevich”. a No noise, b σ = 0.10, c σ = 1.0 a b b a c c Fig. 9 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Manevich”. a No noise, b σ = 0.10, c σ = 1.0 Fig. 9 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Manevich”. a No noise, b σ = 0.10, c σ = 1.0 a b c Fig. 10 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Ellipsoid”. a No noise, b σ = 0.10, c σ = 1.0 a b b a a c c Fig. 10 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Ellipsoid”. a No noise, b σ = 0.10, c σ = 1.0 by a becoming prematurely too small in the divergent early iterations. On the other hand, the standard SPSA showed a good reduction at log10(δ ˆθ0min) = −2.0, and −1.5. at both c = 0.1 and 1.0. This implies that for A_SPSA, a range of values of good c can be by a becoming prematurely too small in the divergent early iterations. On the other hand, the standard SPSA showed a good reduction at log10(δ ˆθ0min) = −2.0, and −1.5. at both c = 0.1 and 1.0. The Ellipsoid function is described as This implies that for A_SPSA, a range of values of good c can be by a becoming prematurely too small in the divergent early iterations. On the other hand, the standard SPSA showed a good reduction at log10(δ ˆθ0min) = −2.0, and −1.5. at both c = 0.1 and 1.0. This implies that for A_SPSA, a range of values of good c can be Page 13 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 a b c Fig. 11 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rotated Ellipsoid”. a No noise, b σ = 0.10, c σ = 1.0 a b b a c c Fig. 11 Initial parameter change δ ˆθ0min and distribution of responses after 2000 function evaluations for “Rotated Ellipsoid”. a No noise, b σ = 0.10, c σ = 1.0 a b c Fig. 12 Effect of choice of c to the final response of “Sphere” with Gaussian noise of σ = 0.1 after 2000 func- tion evaluations. a c = 0.01, b c = 0.10, c c = 1.00 a b b a c c Fig. 12 Effect of choice of c to the final response of “Sphere” with Gaussian noise of σ = 0.1 after 2000 func- tion evaluations. a c = 0.01, b c = 0.10, c c = 1.00 Page 14 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 narrower than that of the standard SPSA. On the other hand, the choice of δ ˆθ0min (and therefore a) is much easier for A_SPSA. We can, for example, let δ ˆθ0min ≃min(U − L), where min(U − L) is the minimum difference between upper and lower bounds of the domain of parameter vector θ. In practice, it is better to scale all the input dimensions to fall in similar or equal intervals. Figure 13 shows the results of optimizing the Rosenbrock and Rastrigin functions using three different values of multiplication factor of a: 0.1, 0.5, and 0.9. The difference in multiplication factor does not change the general trend that larger δ ˆθ0min produces better results and that divergence does not occur. One could tune the value of the mul- tiplication factor, but the default value of 0.5 that we showed in the Algorithm 1 gen- erally produces satisfactory results compared to other values of multiplication factors between 0 and 1. The Fig. The Ellipsoid function is described as 13 (b) also shows that δ ˆθ0min ≃min(U −L) may not be an optimal setting since smaller value δ ˆθ0min ≃10−1.5 is shown to produce better optimiza- tion results when the reduction rate is slow at 0.9. This implies that in a bumpy (highly multimodal) function like Rastrigin, the slow decrease in a can adversely affect the mini- mization of the objective value by a large number of resets to θb. The opposite is true with Rosenbrock function in (a), in which the slow reduction factor 0.9 gave the best result at δ ˆθ0min ≃101. For all the mathematical functions tested in this paper, optimization using SPSA diverges almost surely if the δ ˆθ0min is large. However, A_SPSA and SPSA give closely matching results when the initial step sizes are relatively small (i.e., the left hand side of the plots in Figs. 2, 3, 4, 5, 6, 7, 8, 9, 10 and 11). This is because, in cases that divergence does not happen, the adaptation of a does not take place in A_SPSA and therefore SPSA and A_SPSA have identical behavior. This is a confirmation that Algorithm 1 does not alter, in any significant way, the finite sample convergence characteristics of the original SPSA when the divergence does not manifest. Nonlinear dynamics example We consider a parameter estimation problem with Lorenz attractor. Its nonlinear dynamics is described as (17) dx1 dt = s(x2 −x1), dx1 dt = s(x2 −x1), (17) b he responses after 2000 function evaluations. a a b Fig. 13 Effect of choice of the reduction factor of a to the responses after 2000 function evaluations. a Rosenbrock (no noise), b Rastrigin (no noise) a a Fig. 13 Effect of choice of the reduction factor of a to the responses after 2000 function evaluations. a Rosenbrock (no noise), b Rastrigin (no noise) Page 15 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 (18) dx2 dt = x1(r −x3) −x2, (18) dx2 dt = x1(r −x3) −x2, dx2 dt = x1(r −x3) −x2, (18) dx3 dt = x1x2 −bx1. (19) dx3 dt = x1x2 −bx1. (19) We seek to identify the system parameters θ = [s, r, b] by minimizing the one-time-step- ahead prediction error Lk of the state xk+1 given the current state xk = [xk1, xk2, xk3]T. We use fourth-order Runge–Kutta method to obtain xk+1. We seek to identify the system parameters θ = [s, r, b] by minimizing the one-time-step- ahead prediction error Lk of the state xk+1 given the current state xk = [xk1, xk2, xk3]T. We use fourth-order Runge–Kutta method to obtain xk+1. Let us denote ˆxk+1 as one-time-step-ahead prediction given by the estimated system with parameters ˆθk. Then, we can define the prediction error as (20) Lk(xk, ˆθk) = [xk+1 −ˆxk+1]T · [xk+1 −ˆxk+1]. (20) Thus, the optimization to be solved is Thus, the optimization to be solved is Thus, the optimization to be solved is Thus, the optimization to be solved is (21) min θ∈Θ Lk(xk, θ). min θ∈Θ Lk(xk, θ). (21) The index k above is the same as the index k in the SPSA algorithms. So the SPSA itera- tion proceeds along with the time steps of the dynamic system to compute Lk. We set the true parameters to be θ = [10, 28, 8/3] and pretend to not to know them. We set the time increment to be ∆t = 0.005 and simulate from t = 0 to 20, obtaining target state xk with k = 0, 1, 2, . . . , 4000. We let δ ˆθ0min ∈{0.001, 0.01, 1, 10, 100, 1000} and at each value of δ ˆθ0min we run both A_SPSA and SPSA 20 times. Nonlinear dynamics example For this problem, we set the parameter space as three-dimensional product space = [0, 500]3. The initial state is x0 = [2, 3, 4]T. The initial guess (starting point) of the parameter set ˆθ0 is a random pick from . Figure 14 show the box plots of final Lk when started from different values of δ ˆθ0min . The smallest median of final Lk is obtained at δ ˆθ0min = 10 for SPSA and δ ˆθ0min = 100 and 1000 for A_SPSA. The best medians of final Lk obtained for A_SPSA (5.62 × 10−15) is Fig. 14 Initial parameter change δ ˆθ0min and distribution of L4000 (after 8000 function evaluations) Fig. 14 Initial parameter change δ ˆθ0min and distribution of L4000 (after 8000 function evaluations) Ito and Dhaene SpringerPlus (2016) 5:200 Page 16 of 18 smaller compared to that of SPSA (3.10 × 10−13). However, both SPSA and A_SPSA had some runs that did not converge to the above mentioned near-zero Lk values even at these δ ˆθ0min. Again, for A_SPSA, the best setting were obtained when δ ˆθ0min was set to large val- ues near the order of magnitude of the distance between upper and lower bound of the domain, while for SPSA, the best δ ˆθ0min was at an interior value between 10−3 and 103. Figure 15 shows the trajectory of the reference Lorenz attractor and the simulation of the Lorenz attractor whose system parameters s, r, and b were successfully identified by A_SPSA. The time t is run from 0 to 20 starting from the same initial condition used in the identification. The figure shows excellent match. ihi Figure 16 shows the box plots of parameters estimated by A_SPSA and SPSA starting at their best δ ˆθ0min settings. The corresponding statistics are shown in Tables 1 and 2. The boxes appear collapsed as single horizontal lines at medians since the spaces between Fig. 15 State evolution of the target and identified Lorenz attractor, t = 0 to 20 boxes appear collapsed as single horizontal lines at medians since the spaces between b a a b Fig. 16 Distribution of the parameters identified by A_SPSA and SPSA. a A_SPSA with δ ˆθ0min = 100, b SPSA with δ ˆθ0min = 10 a a Fig. 16 Distribution of the parameters identified by A_SPSA and SPSA. Nonlinear dynamics example a A_SPSA with δ ˆθ0min = 100, b SPSA with δ ˆθ0min = 10 Fig. 16 Distribution of the parameters identified by A_SPSA and SPSA. a A_SPSA with δ ˆθ0min = 100, b SPSA with δ ˆθ0min = 10 Page 17 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 Table 1 Statistics of identified Lorenz Attractor parameters by 20 SPSA runs at δ ˆθ0min = 10 Method s r b Pred. Err. L4000 1 A_SPSA: 0 Min.: 0.00 Min.: 8.017 Min.: 0.000 Min.: 0.0000 2 SPSA: 20 1st Qu.: 10.00 1st Qu.: 28.000 1st Qu.: 2.642 1st Qu.: 0.0000 3 Median: 10.00 Median: 28.000 Median: 2.667 Median: 0.0000 4 Mean: 55.94 Mean: 45.534 Mean: 2.311 Mean: 1.3645 5 3rd Qu.: 11.11 3rd Qu.: 36.817 3rd Qu.: 2.667 3rd Qu.: 0.1017 6 Max.: 477.04 Max.: 328.504 Max.: 3.261 Max.: 19.6773 Table 1 Statistics of identified Lorenz Attractor parameters by 20 SPSA runs at δ ˆθ0min = 10 Table 2 Statistics of identified Lorenz attractor parameters by 20 A_SPSA runs at δ ˆθ0min = 100 Method s r b Pred. Err. L4000 1 A_SPSA: 20 Min.: 0.000 Min.: 0.000 Min.: 0.0000 Min.: 0.0000 2 SPSA: 0 1st Qu.: 10.000 1st Qu.: 28.000 1st Qu.: 2.6667 1st Qu.: 0.0000 3 Median: 10.000 Median: 28.000 Median: 2.6667 Median: 0.0000 4 Mean: 68.069 Mean: 31.816 Mean: 24.8487 Mean: 1.2328 5 3rd Qu.: 10.000 3rd Qu.: 28.000 3rd Qu.: 2.6667 3rd Qu.: 0.0000 6 Max.: 500.000 Max.: 156.811 Max.: 438.8246 Max.: 15.6654 Table 2 Statistics of identified Lorenz attractor parameters by 20 A_SPSA runs at δ ˆθ0min = 100 first quartiles and third quartiles are very narrow. Some non-converging cases are visible as dots on the figure. The figure and the tables show that the parameter estimates are more consistent from run to run in A_SPSA than that of SPSA as A_SPSA has narrower first and third quartile differences. first quartiles and third quartiles are very narrow. Some non-converging cases are visible as dots on the figure. The figure and the tables show that the parameter estimates are more consistent from run to run in A_SPSA than that of SPSA as A_SPSA has narrower first and third quartile differences. Conclusion With the adaptive initial step algorithm, one can avoid divergence in SPSA iterations. Moreover, with a large initial step size, the SPSA algorithm with the adaptive initial step algorithm was able to find equal or better solutions compared to the original SPSA for all the ten mathematical function minimization problems that we have tested. In the non- linear dynamics example, the new algorithm was able to find system parameters more precisely. The proposed method may not eliminate the need of tuning the parameters of SPSA algorithms, but it facilitates the process by eliminating the risk of solution diver- gence and reducing the trial-and-error effort. Further testing of the algorithm with dif- ferent test functions, noise distributions, and industrial use-cases would be beneficial. The improvement proposed in this paper is expected to be valuable when the objective functions are costly to evaluate or if the algorithm is employed inside another algorithm such as machine learning or target tracking, for manual tuning of the parameters would be cumbersome in such cases. As a future work, it would be beneficial to investigate under what conditions the probability of the proposed adaptation (i.e. going into if- branch in Algorithm 1) happening tends to zero as iteration k tends to infinity. With the adaptive initial step algorithm, one can avoid divergence in SPSA iterations. Moreover, with a large initial step size, the SPSA algorithm with the adaptive initial step algorithm was able to find equal or better solutions compared to the original SPSA for all the ten mathematical function minimization problems that we have tested. In the non- linear dynamics example, the new algorithm was able to find system parameters more precisely. The proposed method may not eliminate the need of tuning the parameters of SPSA algorithms, but it facilitates the process by eliminating the risk of solution diver- gence and reducing the trial-and-error effort. Further testing of the algorithm with dif- ferent test functions, noise distributions, and industrial use-cases would be beneficial. The improvement proposed in this paper is expected to be valuable when the objective functions are costly to evaluate or if the algorithm is employed inside another algorithm such as machine learning or target tracking, for manual tuning of the parameters would be cumbersome in such cases. As a future work, it would be beneficial to investigate under what conditions the probability of the proposed adaptation (i.e. Conclusion going into if- branch in Algorithm 1) happening tends to zero as iteration k tends to infinity. Authors contributions KI has conceived the Algorithm 1, conducted the numerical experiments, and has written this manuscript. TD has pro- vided guidance by supervising the progress of the research, and has reviewed the manuscript for intellectual feedback and editing. Author details 1 Ghent University - iMinds, INTEC, Gaston Crommenlaan 8 bus 201, Ledeberg, 9050 Ghent, Belgium. 2 Noesis Solutions, Gaston Geenslaan 11 B4, 3001 Louvain, Belgium. Authors’ contributions Authors contributions KI has conceived the Algorithm 1, conducted the numerical experiments, and has written this manuscript. TD has pro- vided guidance by supervising the progress of the research, and has reviewed the manuscript for intellectual feedback and editing. Page 18 of 18 Ito and Dhaene SpringerPlus (2016) 5:200 Received: 13 September 2015 Accepted: 15 February 2016 Received: 13 September 2015 Accepted: 15 February 2016 Received: 13 September 2015 Accepted: 15 February 2016 Received: 13 September 2015 Accepted: 15 February 2016 Acknowledgements The authors would like to thank James C. Spall for constructive comments on the proposed method. Keiichi Ito has been funded by the Institute for the Promotion of Innovation through Science and Technology (IWT) through the Baekeland Mandate program. This research has also been funded by the Interuniversity Attraction Poles Programme BESTCOM initi- ated by the Belgian Science Policy Office. References l f Altaf MU, Heemink AW, Verlaan M, Hoteit I (2011) Simultaneous perturbation stochastic approximation for tidal models. Ocean Dyn 61:1093–1105 y Andradóttir S (1996) A scaled stochastic approximation algorithm. Manag Sci 42(4):475–498. doi:10.1287/mnsc.42.4.475 Bottou L (2010) Large-scale machine learning with stochastic gradient descent In: Proceedings of COMPSTAT’2010 Andradóttir S (1996) A scaled stochastic approximation algorithm. Manag Sci 42(4):475–498. doi:10.1287/mnsc.42.4.475 Bottou L (2010) Large-scale machine learning with stochastic gradient descent. In: Proceedings of COMPSTAT’2010. Springer, pp 177–186 Bottou L (2010) Large-scale machine learning with stochastic gradient descent. In: Proceedings of COMPSTAT’2010. Springer, pp 177–186 Burnett R (2004) Application of stochastic optimization to collision avoidance. In: Proceedings of the American control conference. Massachusetts, USA, Boston, pp 2789–2794 Dong N, Chen Z (2012) A novel ADP based model-free predictive control. Nonlinear Dyn 69(1–2):89–97. doi:10.1007/ s11071-011-0248-3 Dong N, Chen Z (2012) A novel data based control method based upon neural network and simultaneous perturb stochastic approximation Nonlinear Dyn 67(2):957–963 doi:10 1007/s11071-011-0039-x Dong N, Chen Z (2012) A novel data based control method based upon neural network and sim stochastic approximation. Nonlinear Dyn 67(2):957–963. doi:10.1007/s11071-011-0039-x Dong N, Chen Z (2012) A novel data based control method based upon neural network and simultaneous perturba stochastic approximation. Nonlinear Dyn 67(2):957–963. doi:10.1007/s11071-011-0039-x Easterling D, Watson L, Madigan M, Castle B, Trosset M (2014) Parallel deterministic and stochastic global minimization of functions with very many minima Comput Optim Appl 57(2):469–492 doi:10 1007/s10589-013-9592-1 Easterling D, Watson L, Madigan M, Castle B, Trosset M (2014) Parallel deterministic and stochastic global f h O l ( ) d / functions with very many minima. Comput Optim Appl 57(2):469–492. doi:10.1007/s10589-013-9592-1 functions with very many minima. Comput Optim Appl 57(2):469–492. doi:10.1007/s10589-013-9592-1 Kiefer J, Wolfowitz J (1952) Stochastic estimation of the maximum of a regression function. Ann Math Stat 23:452–466 Kiefer J, Wolfowitz J (1952) Stochastic estimation of the maximum of a regression function. Ann Math Stat 23:452–4 Kirkpatrick S, Gelatt CD, Vecchi MP (1983) Optimization by simulated annealing. Science 220(4598):671–680 Kleinman NL, Hill SD, Ilenda VA (1997) SPSA/SIMMOD optimization of air traffic delay cost. In: Proceedings of the Am Kleinman NL, Hill SD, Ilenda VA (1997) SPSA/SIMMOD optimization of air traffic delay cost. In: Proceedings of the Ameri- can control conference. Albuquerque, New Mexico, USA, pp 1121–1125 , , ( ) pfi y g can control conference. Albuquerque, New Mexico, USA, pp 1121–1125 an control conference. References l f Albuquerque, New Mexico, USA, pp 1121–1 Ko HS, Lee KY, Kim HC (2008) A simultaneous perturbation stochastic approximation (SPSA)-based model approximation and its application for power system stabilizers. Int J Control Autom Syst 6(4):506–514 Ko HS, Lee KY, Kim HC (2008) A simultaneous perturbation stochastic approximation (SPSA)-b and its application for power system stabilizers Int J Control Autom Syst 6(4):506–514 Ko HS, Lee KY, Kim HC (2008) A simultaneous perturbation stochastic approximation (SPSA)-based and its application for power system stabilizers. Int J Control Autom Syst 6(4):506–514 Radac M, Precup R, Petriu E, Preitl S (2011) Application of IFT and SPSA to servo system control. IEEE Trans Neural Netw 22(12):2363–2375. doi:10.1109/TNN.2011.2173804 Sadegh P, Spall JC (1997) Optimal random perturbations for stochastic approximation using a simultaneous perturbation gradient approximation. In: Proceedings of the American control conference. Albuquerque, NM, USA, pp 3582–3586 Sadegh P, Spall JC (1997) Optimal random perturbations for stochastic approximation using a simultaneous perturbation gradient approximation. In: Proceedings of the American control conference. Albuquerque, NM, USA, pp 3582–3586 Shen X, Yao M, Jia W, Yuan D (2012) Adaptive complementary filter using fuzzy logic and simultaneous perturbation sto- chastic approximation algorithm Measurement 45(5):1257–1265 doi:10 1016/j measurement 2012 01 011 http:// gradient approximation. In: Proceedings of the American control conference. Albuquerque, NM, USA, pp 3582 3586 Shen X, Yao M, Jia W, Yuan D (2012) Adaptive complementary filter using fuzzy logic and simultaneous perturbation sto- chastic approximation algorithm. Measurement 45(5):1257–1265. doi:10.1016/j.measurement.2012.01.011. http:// www.sciencedirect.com/science/article/pii/S0263224112000267 g g Shen X, Yao M, Jia W, Yuan D (2012) Adaptive complementary filter using fuzzy logic and simultaneous perturbation sto- chastic approximation algorithm. Measurement 45(5):1257–1265. doi:10.1016/j.measurement.2012.01.011. http:// www.sciencedirect.com/science/article/pii/S0263224112000267 p Spall JC (1992) Multivariate stochastic approximation using a simultaneous perturbation gradient approximation. IEEE Trans Autom Control 37(3):332–341 p Spall JC (1992) Multivariate stochastic approximation using a simultaneous perturbation gradient approximation. IEEE Trans Autom Control 37(3):332–341 Spall JC (1998) Implementation of the simultaneous perturbation algorithm for stochastic optimization. IEEE Trans Aerosp Electron Syst 34(3):817–823 Spall JC (1998) Implementation of the simultaneous perturbation algorithm for stochastic optimization. IEEE Trans Aerosp Electron Syst 34(3):817–823 Spall JC (1998) An overview of the simultaneous perturbation method for efficient optimization. Johns Hopkins APL Tech Dig 19(4):482–492 Spall JC (2000) Adaptive stochastic approximation by the simultaneous perturbation method. Trans Autom Control 45(10):1839–1853 Spall JC (2003) Introduction to stochastic search and optimization, estimation, simulation and control. Competing interests h l d Competing interests Keiichi Ito is employed at Noesis Solutions, a private company, which has interests in optimization methods. Keiichi Ito is employed at Noesis Solutions, a private company, which has interests in optimization methods References l f Wiley-Interscience, New York Spall JC (2009) Feedback and weighting mechanisms for improving jacobian estimates in the adaptive simultaneous perturbation algorithm. IEEE Trans Autom Control 54(6):1216–1229 Spall JC, Chin DC (1997) Traffic-responsive signal timing for system-wide traffic control. Transp Res 5(Part C):153–163 SPSA (2001) Simultaneous perturbation stochastic approximation: a method for system optimization. http://www.jhuapl. edu/SPSA/index.htmll Taflanidis A, Beck J (2008) Stochastic subset optimization for optimal reliability problems. Probab Eng Mech 23(2–3):324–338. doi:10.1016/j.probengmech.2007.12.011. http://www.sciencedirect.com/science/article/pii/ S0266892007000501 5th international conference on computational stochastic mechanics Taflanidis A, Beck J (2008) Stochastic subset optimization for optimal reliability problems. Probab Eng Mech 23(2–3):324–338. doi:10.1016/j.probengmech.2007.12.011. http://www.sciencedirect.com/science/article/ S0266892007000501. 5th international conference on computational stochastic mechanics 23(2–3):324–338. doi:10.1016/j.probengmech.2007.12.011. http://www.sciencedirect.com/science/article/pii/ S0266892007000501. 5th international conference on computational stochastic mechanics S0266892007000501. 5th international conference on computational stochastic mechanics Xu Z, Wu X (2013) A new hybrid stochastic approximation algorithm. Optim Lett 7(3):593–606. doi:10.1007/ s11590-012-0443-2 Zeiler MD (2012) ADADELTA: an adaptive learning rate method. arXiv:1212.5701v1 [cs.LG] Zeiler MD (2012) ADADELTA: an adaptive learning rate method. arXiv:1212.5701v1 [cs.LG] Zhu X, Spall JC (2002) A modified second-order SPSA optimization algorithm for finite samples. Int J Adapt Control Signal Process 16:397–409 p g Zhu X, Spall JC (2002) A modified second-order SPSA optimization algorithm for finite samples. Int J Adapt Control Signal Process 16:397–409 Zhu X, Spall JC (2002) A modified second-order SPSA optimization algorithm for finite samp Process 16:397–409
https://openalex.org/W2801768842	https://www.beilstein-journals.org/bjoc/content/pdf/1860-5397-14-77.pdf	English	null	Volatiles from three genome sequenced fungi from the genus <i>Aspergillus</i>	Beilstein journal of organic chemistry	2,018	cc-by	5,876	Abstract The volatiles emitted by agar plate cultures of three genome sequenced fungal strains from the genus Aspergillus were analysed by GC–MS. All three strains produced terpenes for which a biosynthetic relationship is discussed. The obtained data were also corre- lated to genetic information about the encoded terpene synthases for each strain. Besides terpenes, a series of aromatic compounds and volatiles derived from fatty acid and branched amino acid metabolism were identified. Some of these compounds have not been described as fungal metabolites before. For the compound ethyl (E)-hept-4-enoate known from cantaloupe a structural revision to the Z stereoisomer is proposed. Ethyl (Z)-hept-4-enoate also occurs in Aspergillus clavatus and was identified by synthesis of an authentic standard. Address: © 2018 Dickschat et al.; licensee Beilstein-Institut. License and terms: see end of document. Full Research Paper Open Access Beilstein J. Org. Chem. 2018, 14, 900–910. doi:10.3762/bjoc.14.77 Received: 15 January 2018 Accepted: 16 April 2018 Published: 24 April 2018 Associate Editor: S. Bräse © 2018 Dickschat et al.; licensee Beilstein-Institut. License and terms: see end of document. Introduction mushrooms such as the penny bun (Boletus edulis) [7], but can also inhibit the growth of other fungi [8] which likely contrib- utes to the induction of systemic resistance in plants by Tricho- derma [9]. Fungal volatiles can also act as self-inhibitors of fungal germination [10] or as attractants for insects involved in spore distribution [11]. Furthermore, volatiles can be used as taxonomic markers [12] and can serve as indicators for fungal toxin production, e.g., the fungal emission of the sesquiterpene hydrocarbon trichodiene points to the production of trichothecene mycotoxins [13]. Ascomycete fungi are a highly productive and biosynthetically exceptionally creative source of secondary metabolites from all classes of natural products. Many prominent compounds such as lovastatin from Aspergillus terreus [1] or the penicillin anti- biotics from Penicillium [2] are used for human wellfare, whilst others including aflatoxin from Aspergillus flavus [3] or the amatoxins from the death cap (Amanita phalloides) [4] are extremely toxic for humans. Recently, also volatile secondary metabolites from fungi attracted considerable interest [5,6]. Volatiles not only contribute to the pleasant aroma of edible 900 900 Beilstein J. Org. Chem. 2018, 14, 900–910. different enzyme, because no homolog of the geosmin synthase is encoded in the genome of A. fischeri or of any other fungus. Furthermore, the diterpene pimara-8(14),15-diene (7) was one of the main compounds in the bouquet of A. fischeri. The bio- synthesis of this compound is a two-step process that requires cyclisation of geranylgeranyl diphosphate (GGPP) to copalyl diphosphate (CPP) by a class II TS, followed by a second cycli- sation event by a class I TS [32]. These reactions are likely cat- alysed by the only corresponding two-domain enzyme encoded in the A. fischeri genome (accession number XP_001264196, locus tag NFIA_009790). A phylogenetic analysis of 878 fungal terpene synthase homologs (Figure S1 in Supporting Informa- tion File 1) demonstrates that this enzyme is closely related to the bifunctional ent-copalyl diphosphate synthase/ent-kaurene synthase from Fusarium fujikuroi [33]. The N-terminal domain shows the DXDD motif that is typical for class II TSs (312DADD) and the C-terminal domain exhibits an aspartate- rich motif DDXXD, in this case with two of the usually found asparate residues exchanged by glutamate, and an NSE triad, a motif with highly conserved Asn, Ser and Glu residues, for Mg2+ binding as in class I TSs (349DEFME and 847NDYGSLARD). Aspergillus is a well-described genus comprising several hundreds of known species. Introduction Some of these species are human pathogens, e.g., Aspergillus fumigatus can cause infections especially in immunocompromised patients, while other species are safe, e.g., Aspergillus oryzae is traditionally used in Japanese sake brewing. The genus has a rich secondary metabo- lism with 807 compounds from 675 species that were recently summarised in the Aspergillus secondary metabolome database [14]. However, only a few studies about volatile natural prod- ucts from Aspergillus are available [15-22]. Here we report on the volatiles released by three genome sequenced strains of Aspergillus fischeri, A. kawachii and A. clavatus and a correlation of the obtained analytical data to genome informa- tion. Results and Discussion The volatiles released by agar plate cultures of A. fischeri NRRL 181, A. kawachii NBRC 4308 and A. clavatus NRRL 1 grown on medium 129 were collected on charcoal filters with a closed loop stripping apparatus (CLSA) [23]. After solvent ex- traction (CH2Cl2) of the filters the extracts were analysed by GC–MS and compounds were identified by comparison of the recorded EI mass spectra to mass spectral libraries and of calcu- lated retention indices to literature data. For each investigated fungus a representative chromatogram of a headspace extract is shown in Figure 1. Furthermore, two groups of structurally and biosynthetically related sesquiterpenes were found that could each arise from one sesquiterpene synthase (STS). The first of these groups comprised the main compound α-acoradiene (15), accompanied by minor amounts of β-sesquiphellandrene (8), ar-curcumene (9), β-bisabolene (10), (E)-γ-bisabolene (11), trans-α-bergamotene (12), δ-cuprenene (13), and cuparene (14). All these sesquiterpenes arise through a 1,6-cyclisation of farnesyl diphosphate (FPP, via nerolidyl diphosphate, NPP) to the bisabolyl cation (A, Scheme 2). A mixture of sesquiter- penes arising via cation A with the main product trichodiene was previously reported from Fusarium [34]. Aspergillus fisheri A. fischeri produced mainly terpenes, besides traces of the typical fungal volatile oct-1-en-3-ol (1) and a related com- pound that was tentatively identified as (Z)-octa-1,5-dien-3-ol (2) from its mass spectrum (Table 1 and Scheme 1A). Unfortu- nately, a retention index for 2 is not available from the litera- ture, but the mass spectral database hit was very good and the assigned structure for 2 is biosynthetically reasonable: For com- pound 1 a biosynthetic pathway from linoleic acid via its hydro- peroxide has been suggested [24-26], and if the same biosyn- thetic steps would proceed from linolenic acid, this would result in the assigned structure of 2 (Scheme 1B). The compounds 10 and 11 are directly formed from this cation by deprotonation. A 1,3-hydride shift to B and deprotonation yields 8 and γ-curcumene (C). Instead of the latter compound its autoxidation product 9 is observed. From A, a second cycli- sation event results in D that yields 12 upon deprotonation. Al- ternatively, A can react by a 1,5-proton shift to E, followed by cyclisation to F and deprotonation to H. This mechanism is favoured by quantum chemical calculations [35] and provides a reasonable alternative to a previously suggested cyclisation of A to a less stable secondary cation, followed by 1,4-hydride migration to yield the same intermediate F [36]. Again, the dihydrobenzene derivative H is not observed, but instead its autoxidation product 14 is detected. Finally, the main product 15 arises from A by a 1,2-hydride shift to the homobisabolyl cation I, cyclisation to J, and deprotonation. The other compounds identified in the headspace extracts of A. fischeri were all terpenes, including traces of the widespread monoterpenes limonene (3) and linalool (4). The C12 com- pounds (8S,9R,10S)-8,10-dimethyl-1-octalin (5) and (8S,10R)-8,10-dimethyl-1(9)-octalin (6) are intermediates of the biosynthesis of the earthy odorant geosmin that is itself a degraded sesquiterpene [29,30], but geosmin could not be ob- served as a volatile of A. fischeri. The bacterial geosmin synthase is a class I terpene synthase (TS) with two domains [31] that occurs in many actinomycetes, cyanobacteria and myxobacteria, but fungal geosmin biosynthesis must require a 901 Beilstein J. Org. Chem. 2018, 14, 900–910. Figure 1: Total ion chromatograms of headspace extracts from A) Aspergillus fischeri NRRL 181, B) Aspergillus kawachii NBRC 4308, and C) Aspergillus clavatus NRRL 1. Numbers at peaks refer to compound numbers as defined in Table 1 and Schemes 1–5. Aspergillus fisheri Figure 1: Total ion chromatograms of headspace extracts from A) Aspergillus fischeri NRRL 181, B) Aspergillus kawachii NBRC 4308, a C) Aspergillus clavatus NRRL 1. Numbers at peaks refer to compound numbers as defined in Table 1 and Schemes 1–5. The second group of biosynthetically related sesquiterpenes is composed of daucene (19), the main component in the head- space extracts from A. fischeri, and its congeners dauca-4(11),8- diene (17), isodaucene (18), and trans-dauca-8,11-diene (20). The biosynthesis of these compounds requires isomerisation of FPP to NPP, followed by cyclisation to K that results in 17 and 18 by deprotonation (Scheme 3). A 1,2-hydride shift to L and loss of a proton explains the main product 19. For compound 20 a cyclisation of NPP with a different stereochemical course to M is required, followed by deprotonation. Two class I TSs are encoded in the genome of A. fischeri (accession numbers XP_001265719 and XP_001262485, locus tags NFIA_033880 and NFIA_030200), and it seems likely that each of these en- zymes catalyses the formation of one of the two groups of sesquiterpenes with the main compounds α-acoradiene (15) and daucene (19). The enzyme XP_001262485 is closely related to the α-acorenol synthase from Fusarium fujikuroi [37] (Figure S1 in Supporting Information File 1) that produces α-acorenol (16) by quenching of cation J with water (box in Scheme 2), suggesting that this enzyme is responsible for the biosynthesis of 15 in A. fischeri. Therefore, the enzyme XP_001265719 is likely responsible for the biosynthesis of 19 and its byproducts. 902 Beilstein J. Org. Chem. 2018, 14, 900–910. Table 1: Volatiles emitted by Aspergillus fischeri NRRL 181. Aspergillus fisheri compounda Ib I (lit.)c ident.d integrale (Z)-octa-1,5-dien-3-ol (2) 974 ms (834) 0.1% oct-1-en-3-ol (1) 979 974 [27] ms (948), ri 0.5% limonene (3) 1023 1024 [27] ms (920), ri, std 0.2% linalool (4) 1097 1095 [27] ms (923), ri, std 0.1% (8S,9R,10S)-8,10-dimethyl-1-octalin (5) 1221 1224 [28] ms (832), ri <0.1% (8S,10R)-8,10-dimethyl-1(9)-octalin (6) 1231 1233 [28] ms (815), ri <0.1% daucene (19) 1378 1380 [27] ms (917), ri 24.2% trans-dauca-8,11-diene (20) 1424 ms (901) 3.4% trans-α-bergamotene (12) 1434 1432 [27] ms (898), ri 0.4% α-acoradiene (15) 1465 1464 [27] ms (937), ri 12.6% ar-curcumene (9) 1482 1479 [27] ms (882), ri 0.5% isodaucene (18) 1496 1500 [27] ms (918), ri 7.1% cuparene (14) 1503 1504 [27] ms (903), ri 2.5% β-bisabolene (10) 1507 1505 [27] ms (865), ri 1.7% β-sesquiphellandrene (8) 1522 1521 [27] ms (937), ri 1.6% dauca-4(11),8-diene (17) 1526 1530 [27] ms (964), ri 1.4% (E)-γ-bisabolene (11) 1530 1529 [27] ms (844), ri 0.5% δ-cuprenene (13) 1539 1542 [27] ms (818), ri 0.4% pimara-8(14),15-diene (7) 1952 1948 [27] ms (920), ri 25.6% aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. Aspergillus fisheri dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. Scheme 1: Volatiles from Aspergillus fischeri. For all chiral compounds in Schemes 1–5 the relative configurations are shown. Scheme 1: Volatiles from Aspergillus fischeri. For all chiral compounds in Schemes 1–5 the relative configurations are shown. 903 Beilstein J. Org. Chem. 2018, 14, 900–910. Scheme 2: Biosynthesis of bisabolanes and related terpenes in A. fischeri. The biosynthetic origin of the observed traces of monoterpenes is unclear, but these compounds may be formed by a side activi- Aspergillus kawachii The bouquet of A. kawachii was dominated by the alcohols Scheme 2: Biosynthesis of bisabolanes and related terpenes in A. fischeri. Aspergillus kawachii Aspergillus kawachii The biosynthetic origin of the observed traces of monoterpenes is unclear, but these compounds may be formed by a side activi- ty of one of the TSs on GPP. The bouquet of A. kawachii was dominated by the alcohols 3-methylbutan-1-ol (25) and 2-phenylethanol (26) that likely 904 Beilstein J. Org. Chem. 2018, 14, 900–910. Scheme 3: Biosynthesis of daucanes in A. fischeri. Scheme 3: Biosynthesis of daucanes in A. fischeri. in Aspergillus niger. Production of these compounds was shown to be upregulated in a knockout mutant of the MAP kinase Fus3 [39]. arise from leucine and phenylalanine metabolism, respectively (Figure 1, Table 2 and Scheme 4). In addition, small amounts of the sesquiterpenes β-elemene (21a), germacrene D (22), β-ylan- gene (23) and its stereoisomer β-copaene (24) were found. All these sesquiterpenes require a 1,10-cyclisation of FPP to the (E,E)-germacradienyl cation (N). Its deprotonation leads to germacrene A (21) that is known to undergo a Cope rearrange- ment to 21a caused by the thermal impact during GC–MS anal- ysis [38]. A 1,3-hydride shift transforms N into O that yields 22 by loss of a proton. Its reprotonation can induce a second cycli- sation event via R and S to 24, or with a different stereochemi- cal course via P and Q to 23. The mixture of 22, 23 and 24, accompanied by geosmin and the octalin 5, has also been found The genome of A. kawachii contains three genes for TS homologs (accession numbers GAA83682, GAA88217 and GAA91251, locus tags AKAW_01797, AKAW_06331 and AKAW_09365). The first enzyme GAA83682 shows close homology to the bifunctional ent-kaurene synthases from Fusarium and is likely involved in diterpene biosynthesis. The fact that no corresponding diterpene was observed may point to a low gene expression under laboratory culture conditions. It is currently not possible to conclude which of the other two TSs Table 2: Volatiles emitted by Aspergillus kawachii NBRC 4308. compounda Ib I (lit.)c ident.d integrale 3-methylbutan-1-ol (25) <800 731 [27] ms (916), std 61.2% 2-phenylethanol (26) 1110 1106 [27] ms (927), ri, std 36.3% β-elemene (21a) 1391 1389 [27] ms (855), ri 0.1% β-ylangene (23) 1419 1419 [27] ms (901), ri 0.2% β-copaene (24) 1430 1430 [27] ms (903), ri 0.1% germacrene D (22) 1483 1484 [27] ms (934), ri 0.9% aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. Aspergillus kawachii cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. Table 2: Volatiles emitted by Aspergillus kawachii NBRC 4308. aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. 905 Beilstein J. Org. Chem. 2018, 14, 900–910. Scheme 4: Volatiles from A. kawachii. A) Proposed biosynthesis of sesquiterpenes, B) other identified volatiles. Scheme 4: Volatiles from A. kawachii. A) Proposed biosynthesis of sesquiterpenes, B) other identified volatiles. the enzyme XP_001265719 from A. fischeri for the biosynthe- sis of this compound, because the phylogenetic analysis of fungal TSs (Figure S1 in Supporting Information File 1) reveals a closely homologous enzyme in A. clavatus (accession number XP_001273061, locus tag ACLA_093340). Only the gene expression level in laboratory cultures of A. clavatus seems to be much lower than for A. fischeri, since 19 was the main head- space constituent of A. fischeri, but only emitted in traces by A. clavatus. In addition, β-bisabolene (10) and trans-α-berg- amotene (12) were released by A. clavatus. These compounds have also been observed in A. fischeri where they seem to be side products of α-acoradiene biosynthesis. At the current stage it remains elusive which of the three other TS homologs (accession numbers XP_001272213, XP_001273847 and are involved in the biosynthesis of the observed sesquiterpenes from A. kawachii. Notably, both enzymes GAA88217 and GAA91251 are closely related to fungal germacrene D synthases [40] (Figure S1 in Supporting Information File 1) and are good candidates for the formation of 22 and the compounds derived from it in A. kawachi. Aspergillus clavatus aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. XP_001273868, locus tags ACLA_052600, ACLA_063920 and ACLA_064130) in A. clavatus may catalyse the formation of 10 and 12. The enzyme XP_001276070 (ACLA_076850), likely a bifunctional diterpene synthase (DTS), seems to be not expressed in laboratory culture, but another function cannot be excluded for this enzyme. XP_001273868, locus tags ACLA_052600, ACLA_063920 and ACLA_064130) in A. clavatus may catalyse the formation of 10 and 12. The enzyme XP_001276070 (ACLA_076850), likely a bifunctional diterpene synthase (DTS), seems to be not expressed in laboratory culture, but another function cannot be excluded for this enzyme. reduction of acetyl-CoA to ethanol and its esterification by an acyl transferase. In Neurospora crassa acids of short chain alco- hols are formed from alcohols and aldehydes via hemiacetals that are oxidised to the corresponding esters by an alcohol dehy- drogenase [43]. The main compounds were ethyl pentanoate (35) and ethyl heptanoate (37), accompanied by small amounts of ethyl hexa- noate (36), ethyl octanoate (38) and the unsaturated ester ethyl (Z)-hept-4-enoate (41) that was unambiguously identified by synthesis of a reference compound by esterification of (Z)-hept- 4-enoic acid with ethanol. A compound with the retention index I = 1090 was reported from cantaloupe (Curcumis melo) and tentatively identified as ethyl (E)-hept-4-enoate [44], but the E stereoisomer should elute significantly later than the Z isomer. Likely, the reported compound is the same as found here and the structure requires correction to ethyl (Z)-hept-4-enoate. Further ethyl esters were ethyl 2-methylbutyrate (27) and ethyl Esters were the predominant class of compounds emitted by A. clavatus. The observed pattern of volatiles was very unusual, because many ethyl esters and esters derived from carboxylic acids with an odd number of carbons were found. Aspergillus clavatus The headspace extracts from A. clavatus contained small amounts of oct-1-en-3-ol (1) and terpenes, including limonene (3) and pinene (44), and the octalins 5 and 6, in this species accompanied by geosmin (45) (Figure 1, Table 3 and Scheme 5). Furthermore, daucene (19) was observed in small amounts, which further supports the functional assignment of 906 Beilstein J. Org. Chem. 2018, 14, 900–910. Table 3: Volatiles emitted by Aspergillus clavatus NRRL 1. compounda Ib I (lit.)c ident.d integrale ethyl 2-methylbutyrate (27) 849 850 [41] ms (911) 0.3% ethyl 3-methylbutyrate (28) 853 849 [27] ms (915), ri 0.3% 3-methylbutyl acetate (29) 879 869 [27] ms (958), ri 1.0% 2-methylbutyl acetate (32) 883 875 [27] ms (955), ri 1.4% ethyl pentanoate (35) 908 901 [27] ms (935), ri 33.0% α-pinene (44) 932 932 [27] ms (929), ri 1.0% oct-1-en-3-ol (1) 979 974 [27] ms (899), ri 0.7% ethyl hexanoate (36) 1000 997 [27] ms (955), ri 1.9% limonene (3) 1024 1024 [27] ms (857), ri 0.2% isobutyl pentanoate (39) 1052 std 0.9% ethyl (Z)-hept-4-enoate (41) 1092 std 0.4% ethyl heptanoate (37) 1096 1097 [27] ms (961), ri 11.4% 3-methylbutyl pentanoate (30) 1150 1152 [42] std 0.3% 2-methylbutyl pentanoate (33) 1153 std 0.1% ethyl benzoate (42) 1167 1169 [27] ms (938), ri 9.9% ethyl octanoate (38) 1194 1196 [27] ms (896), ri 0.5% (8S,9R,10S)-8,10-dimethyl-1-octalin (5) 1221 1224 [28] ms (868), ri 0.3% (8S,10R*)-8,10-dimethyl-1(9)-octalin (6) 1231 1233 [28] ms (850), ri 0.1% ethyl phenylacetate (43) 1242 1243 [27] ms (867), ri 0.2% isobutyl heptanoate (40) 1246 std 0.4% 3-methylbutyl heptanoate (31) 1343 std <0.1% 2-methylbutyl heptanoate (34) 1347 std <0.1% daucene (19) 1378 1380 [27] ms (851), ri 0.2% geosmin (45) 1395 1399 [27] ms (895), ri 0.4% trans-α-bergamotene (12) 1434 1432 [27] ms (964), ri 13.5% β-bisabolene (10) 1507 1505 [27] ms (926), ri 0.5% aUnidentified compounds and contaminants such as plasticisers are not listed. bRetention index on a HP5-MS GC column. cRetention index data from the literature. dCompound identification is based on matching mass spectrum to a library spectrum (ms, match factor given in brackets, identical mass spectra would produce a match factor of 1000), identical or closely matching retention index (ri), comparison to an authentic standard (std). ePercent of total peak area of the total ion chromatogram. The sum of integrals is lower than 100%, because unidentified compounds and contaminants are not included. Aspergillus clavatus Since the carboxylic acid portion usually derives from fatty acid biosyn- thesis, a process in which the C2 starter acetyl-CoA is elongat- ed with C2 units, esters from carboxylic acids with an even number of carbons are much more widespread. Furthermore, esterification with S-adenosyl-l-methionine (SAM) by a methyl- transferase is a very common process in nature, while ethyl esters are rarer and likely require a two-step pathway through 907 Beilstein J. Org. Chem. 2018, 14, 900–910. Scheme 5: Volatiles from A. clavatus. Scheme 5: Volatiles from A. clavatus. of 3-methylbutan-1-ol (25) and 2-phenylethanol (26), besides traces of germacrene D (22) and a few other terpenes derived from it. This organism encodes two TSs that are closely related to fungal germacrene D synthases that could both be involved in the biosynthesis of the observed sesquiterpenes. The volatiles profile of A. fischeri was dominated by the terpenes α-acora- diene (15), daucene (19) and pimara-8(14),15-diene (7) which matches the genome sequence information from this organism: there is one encoded bifunctional DTS likely for the biosynthe- sis of 7, and two STSs likely for 15 and 19. One of these en- zymes is similar to the α-acorenol synthase from F. fujikuroi, suggesting that this enzyme is responsible for the biosynthesis of 15. The remaining STS can be assigned to 19 which is sup- ported by the occurrence of a closely related enzyme in A. clavatus that also produces small amounts of 19. The biosyn- thetic byproduct dauca-4(11),8-diene (17) has recently also been described from the sponge isolate Dichotomomyces cejpii [48], but the genome of this fungus is not sequenced and the presence of a closely related TS in this organism is currently unknown. While these analyses demonstrate that a reasonable correlation of data from chemical analyses to genomic data allows for a tentative assignment of functions to biosynthetic enzymes, this work cannot replace their necessary biochemical characterisation, but the approach presented here can help to identify interesting candidate enzymes for further investigation. Furthermore, this work demonstrates that fungal volatiles are an interesting subject of study, as many of the compounds such as several of the identified esters from A. clavatus have not been reported from fungi before. 3-methylbutyrate (28), and the aromatic esters ethyl benzoate (42) and ethyl phenylacetate (43). Aspergillus clavatus The esters 27 and 28 were re- ported previously from an Aspergillus parasiticus knockout mutant of the global regulator VeA, but not from the wildtype [20], while 27 is known from Schizophyllum commune [45] and 28 was recently found in truffle [46]. The ester 42 has been de- scribed from A. clavatus before [16]. Besides ethyl esters, a series of esters derived from branched short chain alcohols was identified, including the widespread compounds [5] 3-methylbutyl acetate (29) and 2-methylbutyl acetate (32). The alcohol portion of these esters likely origi- nates from leucine and isoleucine through transamination to the corresponding α-ketocarboxylic acid, oxidative decarboxyl- ation and reduction. For the pentanoate and heptanoate esters not only the corresponding compounds 3- and 2-methylbutyl pentanoate (30 and 33) and -heptanoate (31 and 34), but also the valine-derived analogues isobutyl pentanoate (39) and -heptanoate (40) were detected. Since only for 30 a published retention index was available [42], all six esters were synthesised for their unambiguous identification. Compound 30 was previously tentatively identified from Nodulisporium [47], while the other esters of this series 31, 33, 34, 39, and 40 have not been reported from fungi before. Supporting Information 16.Seifert, R. M.; King, A. D., Jr. J. Agric. Food Chem. 1982, 30, 786–790. doi:10.1021/jf00112a044 Supporting Information File 1 Phylogenetic tree of fungal type I terpene synthases, experimental procedures and NMR spectra of synthetic compounds. Supporting Information File 1 Phylogenetic tree of fungal type I terpene synthases, experimental procedures and NMR spectra of synthetic compounds. 17.Ito, K.; Yoshida, K.; Ishikawa, T.; Kobayashi, S. J. Ferment. Bioeng. 1990, 70, 169–172. doi:10.1016/0922-338X(90)90178-Y 17.Ito, K.; Yoshida, K.; Ishikawa, T.; Kobayashi, S. J. Ferment. Bio 1990, 70, 169–172. doi:10.1016/0922-338X(90)90178-Y 18.Sunesson, A.-L.; Vaes, W. H. J.; Nilsson, C.-A.; Blomquist, G.; 18.Sunesson, A.-L.; Vaes, W. H. J.; Nilsson, C.-A.; Blomquist, G.; Andersson, B.; Carlson, R. Appl. Environ. Microbiol. 1995, 61, 2911–2918. Andersson, B.; Carlson, R. Appl. Environ. Microbiol. 1995, 61, [https://www.beilstein-journals.org/bjoc/content/ supplementary/1860-5397-14-77-S1.pdf] [https://www.beilstein-journals.org/bjoc/content/ supplementary/1860-5397-14-77-S1.pdf] 19.Jeleń, H. H.; Grabarkiewicz-Szczesna, J. J. Agric. Food Chem. 2005, 53, 1678–1683. doi:10.1021/jf0487396 53, 1678–1683. doi:10.1021/jf0487396 20.Roze, L. V.; Chanda, A.; Laivenieks, M.; Beaudry, R. M.; 20.Roze, L. V.; Chanda, A.; Laivenieks, M.; Beaudry, R. M.; Artymovich, K. A.; Koptina, A. V.; Awad, D. W.; Valeeva, D.; Jones, A. D.; Linz, J. E. BMC Biochem. 2010, 11, 33. doi:10.1186/1471-2091-11-33 ORCID® iDs Jeroen S. Dickschat - https://orcid.org/0000-0002-0102-0631 22.Heddergott, C.; Calvo, A. M.; Latgé, J. P. Eukaryotic Cell 2014, 13, 1014–1025. doi:10.1128/EC.00074-14 Acknowledgements Jones, A. D.; Linz, J. E. BMC Biochem. 2010, 11, 33. This work was funded by the DFG (DI1536/9-1). We thank the US Department of Agriculture for A. clavatus and A. fischeri. 21.Takeuchi, T.; Kimura, T.; Tanaka, H.; Kaneko, S.; Ichii, S.; Kiuchi, M.; Suzuki, T. Surf. Interface Anal. 2012, 44, 694–698. doi:10.1002/sia.4870 Conclusion In summary, the volatiles from three species of the genus Aspergillus were identified. A. kawachii released large amounts 908 Beilstein J. Org. Chem. 2018, 14, 900–910. 15.Kaminski, E.; Stawicki, S.; Wasowizc, E. Appl. Microbiol. 1974, 27, 1001–1004. Supporting Information Supporting Information File 1 Phylogenetic tree of fungal type I terpene synthases, experimental procedures and NMR spectra of synthetic compounds. [https://www.beilstein-journals.org/bjoc/content/ supplementary/1860-5397-14-77-S1.pdf] References 23.Grob, K.; Zürcher, F. J. Chromatogr. 1976, 117, 285–294. doi:10.1016/0021-9673(76)80005-2 23.Grob, K.; Zürcher, F. J. Chromatogr. 1976, 117, 285–294. 1. Moore, R. N.; Bigam, G.; Chan, J. K.; Hogg, A. M.; Nakashima, T. T.; Vederas, J. C. J. Am. Chem. Soc. 1985, 107, 3694–3701. doi:10.1021/ja00298a046 24.Matsui, K.; Sasahara, S.; Akakabe, Y.; Kajiwara, T. Biosci., Biotechnol., Biochem. 2003, 67, 2280–2282. 24.Matsui, K.; Sasahara, S.; Akakabe, Y.; Kajiwara, T. Biosci., Biotechnol., Biochem. 2003, 67, 2280–2282. 25.Kuribayashi, T.; Kaise, H.; Uno, C.; Hara, T.; Hayakawa, T.; Joh, T. J. Agric. Food Chem. 2002, 50, 1247–1253. doi:10.1021/jf0112217 2. Hamed, R. B.; Gomez-Castellanos, J. R.; Henry, L.; Ducho, C.; McDonough, M. A.; Schofield, C. J. Nat. Prod. Rep. 2013, 30, 21–107. doi:10.1039/C2NP20065A 2. Hamed, R. B.; Gomez-Castellanos, J. R.; Henry, L.; Ducho, C.; McDonough, M. A.; Schofield, C. J. Nat. Prod. Rep. 2013, 30, 21–107. doi:10.1039/C2NP20065A 26.Assaf, S.; Hadar, Y.; Dosoretz, C. G. J. Agric. Food Chem. 1995, 43, 2173–2178. doi:10.1021/jf00056a040 3. Minto, R. E.; Townsend, C. Chem. Rev. 1997, 97, 2537–2556. doi:10.1021/cr960032y 27.Adams, R. P. Identification of Essential Oil Components by Gas Chromatography/Mass Spectrometry; Allured: Carol Stream, 2009. Chromatography/Mass Spectrometry; Allured: Carol Stream, 2009 4. Hallen, H. E.; Luo, H.; Scott-Craig, J. S.; Walton, J. D. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 19097–19101. doi:10.1073/pnas.0707340104 28.Citron, C. A.; Gleitzmann, J.; Laurenzano, G.; Pukall, R.; Dickschat, J. S. ChemBioChem 2012, 13, 202–214. 28.Citron, C. A.; Gleitzmann, J.; Laurenzano, G.; Pukall, R. Dickschat, J. S. ChemBioChem 2012, 13, 202–214. doi:10.1002/cbic.201100641 doi:10.1002/cbic.201100641 5. Dickschat, J. S. Nat. Prod. Rep. 2017, 34, 310–328. 5. Dickschat, J. S. Nat. Prod. Rep. 2017, 34, 310–328. doi:10.1039/C7NP00003K 29.Dickschat, J. S.; Bode, H. B.; Mahmud, T.; Müller, R.; Schulz, S. J. Org. Chem. 2005, 70, 5174–5182. doi:10.1021/jo050449g 6. Lemfack, M. C.; Gohlke, B.-O.; Toguem, S. M. T.; Preissner, S.; 6. Lemfack, M. C.; Gohlke, B.-O.; Toguem, S. M. T.; Preissner, S.; Piechulla, B.; Preissner, R. Nucleic Acids Res. 2018, 46, D1261–D1265. doi:10.1093/nar/gkx1016 30.Nawrath, T.; Dickschat, J. S.; Müller, R.; Jiang, J.; Cane, D. E.; Schulz, S. J. Am. Chem. Soc. 2008, 130, 430–431. doi:10.1021/ja077790y Piechulla, B.; Preissner, R. Nucleic Acids Res. 2018, 46, 7. Zawirksa-Wojtasiak, R. Food Chem. 2004, 86, 113–118. 7. Zawirksa-Wojtasiak, R. Food Chem. 2004, 86, 113–118. doi:10.1016/j.foodchem.2003.08.016 31.Jiang, J.; He, X.; Cane, D. E. Nat. Chem. Biol. 2007, 3, 711–715 31.Jiang, J.; He, X.; Cane, D. E. Nat. Chem. Biol. 2007, 3, 711–715. doi:10.1038/nchembio.2007.29 doi:10.1016/j.foodchem.2003.08.016 8. Simon, A.; Dunlop, R. W.; Ghisalberti, E. L.; Sivasithamparam, K. Soil Biol. Biochem. 1988, 20, 263–264. doi:10.1016/0038-0717(88)90050-8 32.Peters, R. J. Nat. Prod. Rep. 2010, 27, 1521–1530. doi:10.1039/c0np00019a 32.Peters, R. J. Nat. Prod. Rep. 2010, 27, 1521–1530. doi:10.1039/c0np00019a 33.Tudzynski, B.; Kawaide, H.; Kamiya, Y. Curr. Genet. 1998, 34 33.Tudzynski, B.; Kawaide, H.; Kamiya, Y. Curr. Genet. 1998, 34, 234–240. doi:10.1007/s002940050392 33.Tudzynski, B.; Kawaide, H.; Kamiya, Y. C 234–240. doi:10.1007/s002940050392 9. Worasatit, N.; Sivasithamparam, K.; Ghisalberti, E. L.; Rowland, C. Mycol. Res. 1994, 98, 1357–1363. doi:10.1016/S0953-7562(09)81063-0 234–240. doi:10.1007/s002940050392 34.Dickschat, J. S.; Brock, N. L.; Citron, C. A.; Tudzynski, B 34.Dickschat, J. S.; Brock, N. L.; Citron, C. A.; Tudzynski, B. ChemBioChem 2011, 12, 2088–2095. doi:10.1002/cbic.20110026 doi:10.1016/S0953-7562(09)81063-0 doi:10.1016/S0953-7562(09)81063-0 34.Dickschat, J. S.; Brock, N. L.; Citron, C. A.; Tudzynski, B. ChemBioChem 2011, 12, 2088–2095. doi:10.1002/cbic.201100268 ChemBioChem 2011, 12, 2088–2095. doi:10.1002/cbic.201100 10.Chitarra, C. S.; Abee, T.; Rombouts, F. M.; Posthumus, M. A.; 10.Chitarra, C. S.; Abee, T.; Rombouts, F. M.; Posthumus, M. A.; Dijksterhuis, J. Appl. Environ. Microbiol. 2004, 70, 2823–2829. doi:10.1128/AEM.70.5.2823-2829.2004 35.Hong, Y. J.; Tantillo, D. J. Org. Lett. 2006, 8, 4601–4604. doi:10.1021/ol061884f 35.Hong, Y. J.; Tantillo, D. J. Org. Lett. 2006, 8, 4601–4604. Dijksterhuis, J. Appl. Environ. Microbiol. 2004, 70, 2823–2829. doi:10.1128/AEM.70.5.2823-2829.2004 36.Cane, D. E.; Ha, H. J. J. Am. Chem. Soc. 1986, 108, 3097–3099. doi:10.1021/ja00271a046 11.Fäldt, J.; Jonsell, M.; Nordlander, G.; Borg-Karlson, A.-K. J. Chem. Ecol. 1999, 25, 567–590. doi:10.1023/A:10209580050 doi:10.1021/ja00271a046 45.Schalchli, H.; Hormazabal, E.; Becerra, J.; Birkett, M.; Alvear, M.; Vid l J Q i A Ch E l 2011 27 503 513 Vidal, J.; Quiroz, A. Chem. Ecol. 2011, 27, 503–513. 46.Schmidberger, P. C.; Schieberle, P. J. Agric. Food Chem. 2017, 65, 9287–9296. doi:10.1021/acs.jafc.7b04073 47.Suwannarach, N.; Kumla, J.; Bussaban, B.; Nuangmek, W.; Matsui, K.; Lumyong, S. Crop Prot. 2013, 45, 63–70. doi:10.1016/j.cropro.2012.11.015 48.Barra, L.; Barac, P.; König, G. M.; Crüsemann, M.; Dickschat, J. S. Org. Biomol. Chem. 2017, 15, 7411–7421. doi:10.1039/C7OB01837A 48.Barra, L.; Barac, P.; König, G. M.; Crüsemann, M.; Dickschat, J. S. Org. Biomol. Chem. 2017, 15, 7411–7421. doi:10.1039/C7OB01837A License and Terms This is an Open Access article 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. The license is subject to the Beilstein Journal of Organic Chemistry terms and conditions: (https://www.beilstein-journals.org/bjoc) The definitive version of this article is the electronic one which can be found at: doi:10.3762/bjoc.14.77 License and Terms This is an Open Access article 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. The license is subject to the Beilstein Journal of Organic Chemistry terms and conditions: (https://www.beilstein-journals.org/bjoc) The definitive version of this article is the electronic one which can be found at: doi:10.3762/bjoc.14.77 License and Terms This is an Open Access article 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. The license is subject to the Beilstein Journal of Organic Chemistry terms and conditions: (https://www.beilstein-journals.org/bjoc) The definitive version of this article is the electronic one which can be found at: doi:10.3762/bjoc.14.77 Beilstein J. Org. Chem. 2018, 14, 900–910. doi:10.1021/ja00271a046 J. Chem. Ecol. 1999, 25, 567–590. doi:10.1023/A:1020958005023 37.Brock, N. L.; Huss, K.; Tudzynski, B.; Dickschat, J. S. ChemBioChem 37.Brock, N. L.; Huss, K.; Tudzynski, B.; Dickschat, J. 37.Brock, N. L.; Huss, K.; Tudzynski, B.; Dickschat, J. S. ChemBioChe 2013, 14, 311–315. doi:10.1002/cbic.201200695 37.Brock, N. L.; Huss, K.; Tudzynski, B.; Dickschat, J 2013, 14, 311–315. doi:10.1002/cbic.201200695 12.Stadler, M.; Læssøe, T.; Fournier, J.; Decock, C.; Schmieschek, B.; Tichy, H.-V.; Peršoh, D. Stud. Mycol. 2014, 77, 1–143. doi:10.3114/sim0016 2013, 14, 311–315. doi:10.1002/cbic.201200695 38.Faraldos, J. A.; Wu, S.; Chappell, J.; Coates, R. M. Tetrahedron 2007, 63, 7733–7742. doi:10.1016/j.tet.2007.04.037 13.Zamir, L. O.; Gauthier, M. J.; Devor, K. A.; Nadeau, Y.; Sauriol, F. J. Chem. Soc., Chem. Commun. 1989, 598–600. doi:10.1039/c39890000598 9.Priegnitz, B.-E.; Brandt, U.; Pahirulzaman, K. A. K.; Dickschat, J. S 39.Priegnitz, B.-E.; Brandt, U.; Pahirulzaman, K. A. K Fleißner, A. Eukaryotic Cell 2015, 14, 602–615. doi:10.1128/EC.00018-15 Fleißner, A. Eukaryotic Cell 2015, 14, 602–615. 14.Vadlapudi, V.; Borah, N.; Yellusani, K. R.; Gade, S.; Reddy, P.; Rajamanikyam, M.; Vempati, L. N. S.; Gubbala, S. P.; Chopra, P.; Upadhyayula, S. M.; Amanchy, R. Sci. Rep. 2017, 7, No. 7325. doi:10.1038/s41598-017-07436-w 14.Vadlapudi, V.; Borah, N.; Yellusani, K. R.; Gade, S.; Reddy, 909 Beilstein J. Org. Chem. 2018, 14, 900–910. 40.Niehaus, E.-M.; Schumacher, J.; Burkhardt, I.; Rabe, P.; Münsterkötter, M.; Güldener, U.; Sieber, C. M. K.; Dickschat, J. S.; Tudzynski, B. Front. Microbiol. 2017, 8, No. 1175. doi:10.3389/fmicb.2017.01175 Tudzynski, B. Front. Microbiol. 2017, 8, No. 1175. 41.Siegmund, B.; Derler, K.; Pfannhauser, W. J. Agric. Food Chem. 2001, 49, 3244–3252. doi:10.1021/jf010116u 49, 3244–3252. doi:10.1021/jf010116u 42.Fan, W.; Qian, M. C. J. Agric. Food Chem. 2006, 54, 2695–2704. doi:10.1021/jf052635t 42.Fan, W.; Qian, M. C. J. Agric. Food Chem. 2006, 54, 2695–2 43.Park, Y.-C.; San, K.-Y.; Bennett, G. N. Appl. Microbiol. Biotechnol 43.Park, Y.-C.; San, K.-Y.; Bennett, G. N. Appl. Micr 43.Park, Y.-C.; San, K.-Y.; Bennett, G. N. Appl. Micr 2007, 76, 349. doi:10.1007/s00253-007-0998-5 2007, 76, 349. doi:10.1007/s00253-007-0998-5 44.Beaulieu, J. C.; Lancaster, V. A. J. Agric. Food Chem. 2007, 55 44.Beaulieu, J. C.; Lancaster, V. A. J. Agric. Food Chem. 2007, 55, 9503–9513. doi:10.1021/jf070282n 9503–9513. doi:10.1021/jf070282n 45.Schalchli, H.; Hormazabal, E.; Becerra, J.; Birkett, M.; Alvear, M.; Vidal, J.; Quiroz, A. Chem. Ecol. 2011, 27, 503–513. doi:10.1080/02757540.2011.596832 46 Schmidberger P C ; Schieberle P J Agric Food Chem 2017, 65 45.Schalchli, H.; Hormazabal, E.; Becerra, J.; Birkett, M.; Alvear, M.; Vidal, J.; Quiroz, A. Chem. Ecol. 2011, 27, 503–513. License and Terms 910
https://openalex.org/W4309365116	https://hal.science/hal-03892903/file/a-1981-1763.pdf	English	null	Assessment of DOAC in GEriatrics (Adage Study): Rivaroxaban/Apixaban Concentrations and Thrombin Generation Profiles in NVAF Very Elderly Patients	Thrombosis and haemostasis	2,022	cc-by	9,956	Assessment of DOAC in GEriatrics (ADAGE study): rivaroxaban/apixaban concentrations and thrombin generation profiles in NVAF very elderly patients Geoffrey Foulon-Pinto, Carmelo Lafuente, Georges Jourdi, Julien Le Guen, Fatoumata Tall, Etienne Puymirat, Maxime Delrue, Léa Riviere, Flora Ketz, Isabelle Gouin-Thibault, et al. Assessment of DOAC in GEriatrics (ADAGE study): rivaroxaban/apixaban concentrations and thrombin generation profiles in NVAF very elderly patients Geoffrey Foulon-Pinto, Carmelo Lafuente, Georges Jourdi, Julien Le Guen, Fatoumata Tall, Etienne Puymirat, Maxime Delrue, Léa Riviere, Flora Ketz, Isabelle Gouin-Thibault, et al. To cite this version: Geoffrey Foulon-Pinto, Carmelo Lafuente, Georges Jourdi, Julien Le Guen, Fatoumata Tall, et al.. As- sessment of DOAC in GEriatrics (ADAGE study): rivaroxaban/apixaban concentrations and throm- bin generation profiles in NVAF very elderly patients. Thrombosis and Haemostasis, 2023, 123 (04), pp.402-414. 10.1055/a-1981-1763. hal-03892903 © 2023. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/ licenses/by-nc-nd/4.0/) Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany Thromb Haemost Keywords ►apixaban ►atrial ﬁbrillation ►elderly ►rivaroxaban ►thrombin generation Abstract Background Although a growing number of very elderly patients with atrial ﬁbrilla- tion (AF), multiple conditions, and polypharmacy receive direct oral anticoagulants (DOACs), few studies speciﬁcally investigated both apixaban/rivaroxaban pharmacoki- netics and pharmacodynamics in such patients. Aims To investigate: (1) DOAC concentration–time proﬁles; (2) thrombin generation (TG); and (3) clinical outcomes 6 months after inclusion in very elderly AF in-patients receiving rivaroxaban or apixaban. Methods ADAGE-NCT02464488 was an academic prospective exploratory multicenter study, enrolling AF in-patients aged 80 years, receiving DOAC for at least 4 days. Each patient had one to ﬁve blood samples at different time points over 20 days. DOAC concentrations were determined using chromogenic assays. TG was investigated using ST-Genesia (STG-ThromboScreen, STG-DrugScreen). These authors share equal authorship. received August 9, 2022 accepted after revision October 18, 2022 accepted manuscript online November 17, 2022 DOI https://doi.org/ 10.1055/a-1981-1763. ISSN 0340-6245. © 2023. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/ licenses/by-nc-nd/4.0/) Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany HAL Id: hal-03892903 https://hal.science/hal-03892903v1 Submitted on 23 Jan 2023 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. Accepted Manuscript online: 2022-11-17 Article published online: 2023-01-09 Accepted Manuscript online: 2022-11-17 Article published online: 2023-01-09 Coagulation and Fibrinolysis Assessment of DOAC in GEriatrics (ADAGE Study): Rivaroxaban/Apixaban Concentrations and Thrombin Generation Proﬁles in NVAF Very Elderly Patients Address for correspondence Virginie Siguret, PhD, INSERM UMR_S1140, Université Paris Cité, 4 avenue de l’Observatoire, 75 270 Paris cedex 06, France (e-mail: virginie.siguret@aphp.fr). 1Université Paris Cité, INSERM UMR-S-1140, Innovations 1Université Paris Cité, INSERM UMR-S-1140, Innovations Thérapeutiques en Hémostase, Paris, France 2Service d’Hématologie Biologique, AP-HP. Université Paris Cité, Hôpital Lariboisière, Paris, France 9Service de gériatrie aiguë polyvalente, Hôpital Charles-Foix, AP-HP S b U i i é I S i F UFR Méd i 9Service de gériatrie aiguë polyvalente, Hôpital Charles-Foix, AP-HP b é éd p 3Service de gériatrie à orientation cardiologique et neurologique, AP-HP, Sorbonne Université, Hôpitaux universitaires Pitié- Salpêtrière-Charles Foix, Ivry-sur-Seine, France 3Service de gériatrie à orientation cardiologique et neurologique, AP-HP, Sorbonne Université, Hôpitaux universitaires Pitié- 9Service de gériatrie aiguë polyvalente, Hôpital Charles-Foix, AP-HP Sorbonne Université, Ivry-sur-Seine, France, UFR Médecine Salpêtrière-Charles Foix, Ivry-sur-Seine, France Sorbonne Université, Paris, France Sorbonne Université, Paris, France 10INSERM, CIC 1414 (Centre d’Investigation Clinique de Rennes), Université de Rennes, CHU de Rennes, Rennes, France Sorbonne Université, Paris, France 10INSERM, CIC 1414 (Centre d’Investig Sorbonne Université, Paris, France 10INSERM, CIC 1414 (Centre d’Investigation Clinique de Rennes), 4CEpiA Team (Clinical Epidemiology and Ageing), Université Paris Est Créteil, INSERM, IMRB, Créteil, France , ( g q ), Université de Rennes, CHU de Rennes, Rennes, France 11 d é l l d Université de Rennes, CHU de Rennes, Rennes, France 5Research Center, Institut de Cardiologie de Montréal - Université de Montréal, Montréal, QC, Canada 11Service d’Hématologie Biologique, CHU de Rennes, Rennes, Fran 12Department of Laboratory Medicine, Namur Thrombosis and Hemostasis Center (NTHC), Université Catholique de Louvain, Yvoir, Belgium 6Service de Gériatrie, AP-HP. Université Paris Cité, Hôpital Européen Georges Pompidou, Paris, France Georges Pompidou, Paris, France 7Service de Gériatrie, AP-HP. Université Paris Cité, Hôpital Rothschild, Paris, France ematology-Hemostasis Laboratory, CHU UCL Namur, Yvoir, Belgi 14Service d’Hématologie Biologique, AP-HP. Université Par Hôpital Européen Georges Pompidou, Paris, France 8Service de Cardiologie, AP-HP. Université de Paris Cité, Hôpital Européen Georges Pompidou, Paris, France Hôpital Européen Georges Pompidou, Paris, France 15Université de Lorraine, Faculté de médecine de Nancy, Nancy, France 16 16Université de Paris Cité, UR 7537 BioSTM (Biostatistics), Faculté de Pharmacie, Paris, France Thromb Haemost 1Université Paris Cité, INSERM UMR-S-1140, Innovations Abstract Background Although a growing number of very elderly patients with atrial ﬁbrilla- tion (AF), multiple conditions, and polypharmacy receive direct oral anticoagulants (DOACs), few studies speciﬁcally investigated both apixaban/rivaroxaban pharmacoki- netics and pharmacodynamics in such patients. Aims To investigate: (1) DOAC concentration–time proﬁles; (2) thrombin generation (TG); and (3) clinical outcomes 6 months after inclusion in very elderly AF in-patients receiving rivaroxaban or apixaban. Keywords ►apixaban ►atrial ﬁbrillation ►elderly ►rivaroxaban ►thrombin generation Methods ADAGE-NCT02464488 was an academic prospective exploratory multicenter study, enrolling AF in-patients aged 80 years, receiving DOAC for at least 4 days. Each patient had one to ﬁve blood samples at different time points over 20 days. DOAC concentrations were determined using chromogenic assays. TG was investigated using ST-Genesia (STG-ThromboScreen, STG-DrugScreen). These authors share equal authorship. DOI https://doi.org/ 10.1055/a-1981-1763. ISSN 0340-6245. Adage Study Foulon-Pinto et al. Results We included 215 patients (women 71.1%, mean age: 87 4 years), 104 rivaroxaban and 111 apixaban, and 79.5% receiving reduced-dose regimen. We observed important inter-individual variabilities (coefﬁcient of variation) whatever the regimen, at Cmax [49–46%] and Cmin [75–61%] in 15 mg rivaroxaban and 2.5 mg apixaban patients, respectively. The dose regimen was associated with Cmax and Cmin plasma concentrations in apixaban (p ¼ 0.0058 and p ¼ 0.0222, respectively), but not in rivaroxaban samples (multivariate analysis). Moreover, substantial variability of thrombin peak height (STG-ThromboScreen) was noticed at a given plasma concentra- tion for both xabans, suggesting an impact of the underlying coagulation status on TG in elderly in-patients. After 6-month follow-up, major bleeding/thromboembolic event/death rates were 6.7%/1.0%/17.3% in rivaroxaban and 5.4%/3.6%/18.9% in apixaban patients, respectively. Conclusion Our study provides original data in very elderly patients receiving DOAC in a real-life setting, showing great inter-individual variability in plasma concentrations and TG parameters. Further research is needed to understand the potential clinical impact of these ﬁndings. cokinetics (PK) and pharmacodynamics (PD) with a wide therapeutic index: therefore, they are administered at ﬁxed doses without laboratory monitoring.4–11 However, the management of very elderly patients receiving antico- agulant treatment may be challenging because they are at both high risk of thrombotic and bleeding complications even though meta-analysis and real-life studies provided reassuring data regarding the clinical beneﬁt–risk balance of DOACs.12–15 Phase II/III clinical trials and phase IV studies evaluating the PK/PD proﬁles of DOACs in healthy Thrombosis and Haemostasis © 2023. The Author(s). Methods ADAGE Study Design and Inclusion Criteria The Assessment of Direct oral Anticoagulants in GEriatrics (ADAGE) study is an academic multicenter prospective observa- tional study (www.clinicaltrials.gov; NCT02464488). We recruited hospitalized patients in university hospital geriatric departments (Assistance Publique-Hôpitaux de Paris AP-HP, France; CHU-UCL Namur-site-Godinne, Belgium) (Janu- ary 2015 to June 2019). Patients aged 80 years and receiving rivaroxaban or apixaban for NVAF since at least 4 days were eligible. The DOAC dose regimen was at the discretion of the physician. Exclusion criteria included any acute unstable comorbid condition or an estimated life expectancy of a few weeks. The study was approved by the local ethics committee (Comité consultatif de protection des personnes pour la recherche biomédicale – Île-de-France-6). All participants gavetheir writteninformedconsent toparticipateinthestudy. Physicians prospectively collected demographic data, lab- Sample Collection in ADAGE Patients p Blood samples were collected by venipuncture into a citrate tube (buffered trisodium citrate 0.109 M, 9 vol/1vol) in addition to other samples taken as part of usual care between 7.00 a.m. and 7.00 p.m. No additional venipuncture was speciﬁcally performed for our study purpose. Sampling and last DOAC intake times were systematically recorded on a standardized request form. Sampling time points were at the discretion of the physicians. At most ﬁve samples per patient were collected over a maximal 20-day period follow- ing inclusion. Time points were classiﬁed according to the time elapsed from the last oral intake (T, hours). Especially, for rivaroxaban patients, T1 to T4 samples corresponded to time to peak (TTP) level (namely Tmax) and T20 to T24 corresponded to time to trough level (Tmin); for patients receiving apixaban (twice daily [b.i.d.]), T1 and T4 samples corresponded to Tmax, and T10 to T12 samples to Tmin. Hence, we conducted an academic prospective explorato- ry study in a cohort of NVAF-hospitalized patients aged 80 years receiving rivaroxaban or apixaban. The primary objec- tives were to characterize rivaroxaban and apixaban concen- tration–time proﬁles and determine the potential impact of demographic, clinical, therapeutic, and pharmacogenetic factors on their peak and trough concentrations. The secondary objectives were to perform TG assays in a subset of patients under several experimental conditions and evaluate the potential inﬂuence of individual characteristics on TG parameter variability at peak and trough levels. The tertiary objectives were to report clinical outcomes (bleeding and thromboembolic complications, death) at 6 months. Tubes were sent to each center’s laboratory within 2 hours after sampling. They were immediately double-centrifuged for 15 minutes at 2,000 g at 15 to 22°C to obtain platelet-poor plasma (PPP) distributed in 500 μL aliquots; pellets were kept for genotyping. All samples were stored at 80°C, prior to onward shipment on dry ice for central analysis (Haematol- ogy laboratory, Hôpital Lariboisière AP-HP, Paris, France). PPP samples were thawed for 3 to 5 minutes in a 37°C water- bath just before use. Introduction Nonvalvular atrial ﬁbrillation (NVAF) is the most common heart rhythm disorder with a prevalence exceeding 10% in octogenarians.1 Current guidelines recommend the use of direct oral anticoagulants (DOACs) targeting either factor Xa (FXa; such as rivaroxaban or apixaban) or thrombin (dabigatran) as ﬁrst-line treatment for the prevention of stroke and systemic thromboembolism in patients with NVAF.2,3 DOACs are characterized by predictable pharma- Adage Study Foulon-Pinto et al. allowing the calculation of the CHA2DS2VASc, HEMOR- R2HAGES, and the 13-item Cumulative Illness Rating Scale- Geriatrics (CIRS-G) scores.37 We recorded DOAC regimen and concomitant medications (excluding vitamins). None of the patients had received heparin derivatives in the last 5 days prior the ﬁrst sampling. Drugs were considered as modu- lators (substrate or moderate/strong inhibitors) of P-gp/CYP3A4/5 according to Geneva University Hospitals classiﬁcation.38 Finally, geriatricians were responsible for the clinical follow-up of the patients at 6 months through a phone call and/or the use of medical records, and recorded bleedings (site, severity), thromboembolic events (site), and deaths. Two geriatricians (F.K., L.R.) blinded to the PK/PD study results adjudicated bleeding events as major or non- major clinically relevant bleedings according to the Interna- tional Society on Thrombosis and Haemostasis criteria39; minor bleedings were not systematically recorded. volunteers and NVAF patients have provided consistent PK and PD data with substantial inter-individual variability.16–26 However, little is known regarding specif- ic concentration–time proﬁles of rivaroxaban and apixa- ban in polymedicated NVAF patients aged 80 years.27–29 This age group is characterized by the presence of multiple comorbidities (including renal impairment) and polyphar- macy, potentially altering both drug exposure and effect. Both xabans are metabolized via cytochrome P450 (CYP) 3A4/5 and CYP2J2, and substrates of the drug-efﬂux trans- porter P-glycoprotein (P-gp). Several studies have evaluat- ed the effect of moderate CYP3A4/P-gp inhibitor concomitant use on rivaroxaban or apixaban exposure, especially in healthy volunteers.4,20 One question is whether drug–drug interactions could inﬂuence PK re- sponse variability in the very elderly. Regarding the assessment of DOAC PD, the study of throm- bin generation (TG) has been shown as a suitable and reliable assay to assess in vitro or ex vivo anticoagulant effect of direct FXa inhibitors in contrast to prothrombin time.30–36 Never- theless, to the best of our knowledge, no speciﬁc TG data regarding elderly patients aged 80 years and over receiving rivaroxaban or apixaban have been published yet. Thrombosis and Haemostasis © 2023. The Author(s). Rivaroxaban and Apixaban Concentrations and Thrombin Generation Assessment Anti-Xa chromogenic assays (STA-Liquid-Anti-Xa, Stago, Asnières-sur-Seine, France) were used to determine rivarox- aban and apixaban plasma concentrations using dedicated calibrators and controls on STA-R-Evolution (Stago) (lower limit of quantiﬁcation: 20 ng/mL). The maximal observed plasma DOAC concentration (Cmax) was deﬁned as the level measured at Tmax (T1–T4), whereas the minimal one (Cmin) was deﬁned as the level measured at Tmin (i.e., T20–T24 for rivaroxaban, T10–T12 for apixaban). Physicians prospectively collected demographic data, lab- oratory data (including creatinine clearance [CrCl] calculated with the Cockcroft–Gault formula), comorbid conditions TGwasstudiedusingtheST-Genesiasystem(Stago)whenever sufﬁcient plasma was available. ST-Genesia is a fully automated Thrombosis and Haemostasis © 2023. The Author(s). Adage Study Foulon-Pinto et al. system enabling quantitative standardized assessment of TG derived from Hemker’s ﬂuorescence method, using dedicated reagents, calibrator, and quality controls.35 Three experimental conditions (as a function of the reagent used) were tested: STG- DrugScreen and STG-ThromboScreen, the latter in presence/ absence of thrombomodulin (þTM/ TM) according to the man- ufacturer’s instructions. STG-DrugScreen contains a mixture of phospholipidsandrecombinanthumantissuefactor(TF)atahigh pM concentration; STG-ThromboScreen contains TF at a lower, intermediate, pM concentration.35 TM is contained in the reagent at a concentration chosen to inhibit 50% of the endogenous thrombin potential (ETP) assessed in normal pooled plasma. Four parameters were analyzed: lag time (LT; in minutes, time from test triggering to signal detection), TTP (in minutes, time necessaryforthrombinconcentrationtoreachitsmaximalvalue), peak height (PH; in nM: maximal thrombin concentration), and ETP (in nM·min: area under the thrombin time–concentration curve).35 Results are presented as absolute values and inhibition percentages in presence of TM. trationsorPH).Predictorsincludedinthisanalysiswereselected based on either their associationwith the studied parameter (p- value less than 0.10 in the univariate analysis) or their clinical pertinence. Results are given as 95% conﬁdence interval of the parameters in the model. Model quality and assumptions were graphically checked. A p-value less than 0.05 was considered statistically signiﬁcant. All p-values are given uncorrected. ABCB1, CYP3A5, and CYP2J2 Genotyping J yp g DNA was extracted from peripheral blood leukocytes using the kits E.Z.N.A. SQ (Omega Bio-tek, Norcross, Georgia, United States). We analyzed four loss-of-function variants in three genes potentially involved in the DOAC metabolism (CYP2J2, CYP3A5) or transport (ABCB1 encoding P-gp) vari- ability: CYP2J2 -76G > T (rs890293;7), CYP3A5 6981A >G (rs776746;3), ABCB1 rs2032582 (2677G > T/A), and ABCB1 rs1045642 (3435C > T) (PharmGKB; www.pharmvar.org). They were analyzed using real-time polymerase chain reac- tion with TaqMan Assay-Reagents for allelic discrimination (Applied Biosystems, Courtabœuf, France). Most patients had multiple comorbidities (mean CIRS-G score: 10.4 4.5); 68.8% had moderate renal impairment (CrCl: 30–59 mL/min). Most patients were polymedicated (mean number of comedications: 6.2 2.7 per patient). Overall, patient characteristics did not signiﬁcantly differ between the two DOAC groups, except for CIRS-G and HEMORR2HAGE scores which were higher in the apixaban group (p ¼ 0.0023 and p ¼ 0.0218, respectively). ABCB1 2677G > T/A, ABCB1 3435C > T, CYP2J27, and CYP3A53 allelic frequencies in ADAGE patients are shown in ►Supplementary Table S1 (available in the online version). Thrombosis and Haemostasis © 2023. The Author(s). Patient Characteristics From November 2015 to July 2019, 215 in-patients (153 females, 62 males) with NVAF were included in the ADAGE study, with a mean age of 86.6 4.3 years (min–max: 80– 100). Patient characteristics are summarized in ►Table 1. The rivaroxaban group comprised 104 patients: 18 (17.3%) received the 20-mg full dose o.d., 83 (79.8%) the 15-mg dose, and 3 (2.9%) the 10-mg dose. The apixaban group comprised 111 patients: 26 (23.4%) received the 5-mg full dose b.i.d. and 85 (76.6%) the 2.5-mg dose b.i.d.; 72 (33.5%) patients did not receive the dose recommended according to the Summary of Product Characteristics10,11 (i.e., 15-mg rivaroxaban in patients with CrCl 50 mL/min, and 2.5-mg apixaban in patients 80 years with creatinine level 133 µM or body weight 60 kg): 62 patients received a lower dose regimen than recommended (28 on rivaroxaban and 34 on apixaban), whereas 10 patients received a higher one (8 on rivaroxaban and 2 on apixaban). TG in Plasma Samples Spiked with Xabans Frozen normal pooled plasma samples (Cryocheck, Mont- pellier, France) were spiked with increasing concentrations of rivaroxaban or apixaban ranging from 0 to 450 ng/mL as previously described.35 We measured effective plasma con- centrations in each sample with speciﬁc anti-Xa assays. Concentration–Time Proﬁles of Rivaroxaban and Apixaban in ADAGE Patients All statistical analyses were run on R-software (version 3.5.1) using the nlme package.40,41 Quantitative data were described asmedian(interquartilerange[IQR])ormeanvalues(standard deviation), and minimal and maximal values. Genetic variants were coded as 0 (wild type), 1 (heterozygous), or 2 (homozy- gous) to model additive allelic effects. Frequencies of genotypes were compared to the Hardy–Weinberg equilibrium expected frequencies using the χ2 test. The association between concen- trations (at Tmax and Tmin) or PD parameters and each covariate was tested using either Spearman’s rank correlation (quantita- tive covariate) or the Kruskal–Wallis test (qualitative covariate). We prespeciﬁed that PH measured using STG-ThromboScreen without TM would be chosen as the PD parameter for statistical analyses (given its optimal sensitivity to a wide range of DOAC concentrations according to previous data of our group).35 Multivariate analysis was done using multiplelinear regression, after log-transformation of the predicted parameter (concen- p Atotalof456bloodsampleswerecollected(average2.1samples per patient): 232 from 104 rivaroxaban patients and 224 from 111 apixaban patients. ►Fig. 1 shows DOAC plasma concentra- tion–time proﬁles. Among ADAGE patients receiving 15-mg rivar- oxaban o.d., median Cmax was 251ng/mL (IQR: 176–377; range: 64–676); median Cmin was 42ng/mL (IQR: 28–70, range: <20– 241). Among patients receiving 2.5-mg apixaban b.i.d., the median Cmax was 180ng/mL (IQR: 130–227; range: 69–469) and the median Cmin was 74ng/mL (IQR: 49–116; range: <20– 251). At Tmax, 30.6% of patients on rivaroxaban 15mg o.d. and 31.8% of patients on 2.5-mg apixaban b.i.d. had plasma concen- trations above the 95th percentile of values reported in the Summary of Product Characteristics obtained in younger patients included in pivotal clinical trials.10,11,21 At Tmin, only 3.1 and 9.7% were above the 95th percentile, whereas 21.5 and 9.7% were below the 5th percentile, respectively. Adage Study Foulon-Pinto et al. Concentration–Time Proﬁles of Rivaroxaban and Apixaban in ADAGE Patients Table 1 Characteristics of ADAGE patients at inclusion Total Rivaroxaban Apixaban p-Value n ¼ 215 n ¼ 104 n ¼ 111 Demographic characteristics Age (y) 86.6 4.3 86.4 4.4 86.8 4.3 0.491 Females (%) 71.2 68.3 73.9 0.371 Body weight (kg) 65.6 14.7 64.9 12.6 66.4 16.5 0.543 DOAC regimen Full dose (%)a 20.5 17.3 23.4 0.312 Reduced dose (%)b 79.5 82.7 76.6 0.312 Clinical characteristics Type of AF Persistent (%) 9.1 8.7 9.4 0.978 Paroxysmal (%) 34.0 35.0 33.0 0.978 Permanent (%) 56.9 56.3 57.6 0.978 Previous stroke or TIA (%) 24.9 21.4 28.2 0.250 Diabetes mellitus (%)†† 19.0 10.9 26.6 0.005 Hypertension (%) 76.1 72.8 79.1 0.336 Heart failure (%) 49.8 46.6 52.7 0.412 Dyslipidemia (%) 26.4 24.3 28.4 0.535 Clinical scores CHA2DS2VASc 5.0 1.4 4.9 1.4 5.2 1.4 0.122 HEMORR2HAGES† 2.3 1.0 2.1 1.0 2.5 1.0 0.022 CIRS-G† 10.4 4.5 9.3 3.7 11.4 4.8 0.007 Laboratory data Creatinine clearance (mL/min)c 49.1 16.5 50.5 17.1 47.9 15.9 0.314 Albumin (g/L) 33.2 5.2 33.4 4.2 33.0 5.9 0.756 C-reactive protein (mg/L) 28.8 39.8 28.6 41.9 28.9 38.0 0.374 Hemoglobin (g/dL) 12.1 1.7 12.2 1.7 12.0 1.8 0.400 Fibrinogen (g/L) 5.0 1.5 4.9 1.6 5.0 1.4 0.309 Therapeutic data Number of co-medications 6.2 2.7 5.9 2.5 6.5 2.9 0.256 1 P-gp substrate/inhibitor including amio. (%) 48.3 44.6 51.8 0.292 1 CYP3A4/5 inhibitor including amio. (%) 22.3 17.8 26.4 0.136 Amiodarone (%) 21.8 17.8 25.5 0.187 Antiplatelet drugs (%) 15.0 10.0 20.0 0.055 Abbreviations: AF atrial ﬁbrillation; Amio amiodarone; TIA transient ischemic attack Abbreviations: AF, atrial ﬁbrillation; Amio, amiodarone; TIA, transient ischemic attack. Note: Data shown as percentage or mean standard deviation. Statistically signiﬁcantly different values between the rivaroxaban and apixaban groups are indicated: †p < 0.05 or ††p < 0.01. a20 mg o.d. (rivaroxaban) or 5 mg b.i.d. (apixaban). b15 or 10 mg o.d. (rivaroxaban) or 2.5 mg b.i.d. (apixaban). cEstimated using the Cockcroft–Gault formula. b15 or 10 mg o.d. (rivaroxaban) or 2.5 mg b.i.d. (apixaban). cEstimated using the Cockcroft–Gault formula. At Tmin, patients on amiodarone displayed signiﬁcantly higher rivaroxaban levels compared to the other patients (60 vs. 41 ng/mL; þ47.8%; p ¼ 0.0315). Concentration–Time Proﬁles of Rivaroxaban and Apixaban in ADAGE Patients The intake of a CYP3A4/5 modula- tor other than amiodarone also contributed to increase Cmin (p¼ 0.0361). Cmin increased when renal function worsened (þ8.1% by 10mL/min CrCl decrease, p¼ 0.093). Finally, there was a trend to higher concentrations in ABCB13435C> T car- riers, without reaching signiﬁcance (þ16.9% per mutated allele, p¼ 0.0788). The multivariate analysis included gender, Thrombosis and Haemostasis © 2023. The Author(s). Abbreviations: AF, atrial ﬁbrillation; Amio, amiodarone; TIA, transient ischemic attack. Note: Data shown as percentage or mean standard deviation. Statistically signiﬁcantly different values between the rivaroxaban and apixaban groups are indicated: †p < 0.05 or ††p < 0.01. a20 mg o.d. (rivaroxaban) or 5 mg b.i.d. (apixaban). b15 or 10 mg o.d. (rivaroxaban) or 2.5 mg b.i.d. (apixaban). cEstimated using the Cockcroft–Gault formula. Association of Covariates with Rivaroxaban Plasma Concentration at Cmax and Cmin Concentration at Cmax and Cmin In the rivaroxaban group, we observed an important inter- individual variability of Cmax (coefﬁcients of variation [CVs] 49 and 47%) and Cmin (75 and 37%) for 15- and 20-mg regimens, respectively. No covariates were found signiﬁcantly (p < 0.05) associated with Cmax in univariate analysis (►Table 2). Rivaroxaban concentrations slightly increased with the daily dose (þ20.3% in 20-mg patients compared to 15mg, p¼ 0.346). Adage Study Foulon-Pinto et al. Fig. 1 Plasma concentrations of rivaroxaban (A) and apixaban (B) in ADAGE patients as a function of time since the last oral intake. (A) Patients receiving rivaroxaban 10, 15, or 20 mg o.d. are represented with yellow, red, and grey circles, respectively (noteworthy, 10 mg off- labeled in AF). In order to facilitate interpretation, we represented the geometric mean Cmax and Cmin and the 5th–95th percentiles, i.e., 229 ng/mL (5th–95th: 178–313) and 57 ng/mL (5th–95th: 18–136), as reported in ROCKET-AF patients (median age: 73 years) receiving 15 mg according to Girgis et al21 with solid and dashed lines, respectively. (B) Patients receiving apixaban 2.5 or 5 mg b.i.d. are represented with blue and grey circles, respectively. We represented the median Cmax and Cmin and 5th–95th percentiles, i.e., 123 ng/mL (5th–95th: 69–221) and 79 ng/mL (5th–95th: 34–162), as reported in patients receiving apixaban 2.5 mg b.i.d. in ARISTOTLE trial11 (median age: 70 years), with solid and dashed lines, respectively. AF, atrial ﬁbrillation; b.i.d., twice daily; o.d., once daily. amiodarone therapy (mean Cmax 251 vs. 175 ng/mL, þ43.5%, p ¼ 0.0098) and female gender (mean Cmax 208 ng/mL vs. 145, þ44.0% in women, p ¼ 0.0165). At Tmin, higher concentrations were observed in ABCB1 3435C> T carriers, without reaching statistical signiﬁcance(þ14.8% per mutated allele, p ¼ 0.0867). No other variables apparently explained the apixaban concen- tration variability at Tmax or Tmin. The multivariate analysis included the same predictors than for rivaroxaban. The dose regimen was signiﬁcantly associated with apixaban concen- trations at both Tmax (þ64.8%, p ¼ 0.0058) and Tmin (þ54.8%, p ¼ 0.0222). Amiodarone might also contribute to higher con- centrations at Tmax (p ¼ 0.0319). Association of Covariates with DOAC TG Peak Height/ETP at Tmax and Tmin ►Supplementary Table S2 (available in the online version) summarizes TG results focused on patient variability at Tmax and at Tmin. PH CV conﬁrmed the substantial variability, both at Tmax and Tmin. We sought to identify covariates, namely DOAC plasma concentrations together with individual char- acteristics, potentially associated with PH and ETP measured using STG-ThromboScreen in the absence of TM at Tmax and Tmin (►Table 3). In the univariate analysis performed in the rivaroxaban group at Tmax, two variables were signiﬁcantly associated with PH decrease: renal failure assessed using CrCl (16.5% per 10 mL/min decrease, p ¼ 0.0016) and heart failure (36.6%, p ¼ 0.0414). In multivariate analysis, CrCl was the only variable associated with PH at Tmax (p ¼ 0.0085). At Tmin, amiodarone intake was associated with PH decrease (52.2%, p ¼ 0.0116) as well as rivaroxaban concentrations Thrombin Generation Proﬁles of Rivaroxaban and Apixaban in ADAGE Patients (B) Patients receiving apixaban 2.5 or 5 mg b.i.d. are represented with blue and grey circles, respectively. We represented the median Cmax and Cmin and 5th–95th percentiles, i.e., 123 ng/mL (5th–95th: 69–221) and 79 ng/mL (5th–95th: 34–162), as reported in patients receiving apixaban 2.5 mg b.i.d. in ARISTOTLE trial11 (median age: 70 years), with solid and dashed lines, respectively. AF, atrial ﬁbrillation; b.i.d., twice daily; o.d., once daily. Fig. 1 Plasma concentrations of rivaroxaban (A) and apixaban (B) in ADAGE patients as a function of time since the last oral intake. (A) Patients receiving rivaroxaban 10, 15, or 20 mg o.d. are represented with yellow, red, and grey circles, respectively (noteworthy, 10 mg off- labeled in AF). In order to facilitate interpretation, we represented the geometric mean Cmax and Cmin and the 5th–95th percentiles, i.e., 229 ng/mL (5th–95th: 178–313) and 57 ng/mL (5th–95th: 18–136), as reported in ROCKET-AF patients (median age: 73 years) receiving 15 mg according to Girgis et al21 with solid and dashed lines, respectively. (B) Patients receiving apixaban 2.5 or 5 mg b.i.d. are represented with blue and grey circles, respectively. We represented the median Cmax and Cmin and 5th–95th percentiles, i.e., 123 ng/mL (5th–95th: 69–221) and 79 ng/mL (5th–95th: 34–162), as reported in patients receiving apixaban 2.5 mg b.i.d. in ARISTOTLE trial11 (median age: 70 years), with solid and dashed lines, respectively. AF, atrial ﬁbrillation; b.i.d., twice daily; o.d., once daily. Interestingly, ADAGE patients’ PH and ETP values presented substantial variability with both xabans for a given plasma concentration, independently of the dose regimen and the experimental conditions (►Supplementary Fig. S1 [available in the online version] and ►Fig. 2). Adding TM toTG triggered with intermediate TF concentration markedly reduced ETP and PH, with strong associations with rivaroxaban and apixaban concentrations (p < 104) (►Fig. 2). A marked variability in ETP and PH percentages of inhibition was also observed, reﬂecting the important variability in the protein C-based negative feedback in very elderly patients. amiodarone intake, CYP3A4/5- or P-gp-modulator intake in addition to amiodarone, DOAC daily dose (20 vs. 15mg), CrCl, and ABCB1 3435C> T genotype. The model also included inter- actiontermsbetweenamiodaroneandothermodulatorsintake. This analysis showed signiﬁcantly higher Cmin in patients re- ceiving either amiodarone (p¼ 0.0232) or a CYP3A4/5 modula- tor other than amiodarone (p ¼ 0.0228). Thrombin Generation Proﬁles of Rivaroxaban and Apixaban in ADAGE Patients TG parameters were measured in 250 plasma samples from 128 ADAGE patients: 128 samples from 62 rivaroxaban patients and 122 from 66 apixaban patients. Patient characteristics did not differ signiﬁcantly between both groups (not shown). Temporal parameters (LT and TTP) were prolonged and PHs were decreased in a DOAC concentration-dependent manner (p < 104), using either STG-DrugScreen (►Supplementary Fig. S1, available in the online version) or intermediate pM TF concentration (STG-ThromboScreen without TM [►Fig. 2], LT and TTP not shown); ETP was associated with apixaban concentrations under these both experimental conditions (p ¼ 0.0102 and p <104, respec- tively), but not with rivaroxaban concentrations (p ¼ 0.1028 and 0.1846, respectively). Fig. 1 Plasma concentrations of rivaroxaban (A) and apixaban (B) in ADAGE patients as a function of time since the last oral intake. (A) Patients receiving rivaroxaban 10, 15, or 20 mg o.d. are represented with yellow, red, and grey circles, respectively (noteworthy, 10 mg off- labeled in AF). In order to facilitate interpretation, we represented the geometric mean Cmax and Cmin and the 5th–95th percentiles, i.e., 229 ng/mL (5th–95th: 178–313) and 57 ng/mL (5th–95th: 18–136), as reported in ROCKET-AF patients (median age: 73 years) receiving 15 mg according to Girgis et al21 with solid and dashed lines, respectively. (B) Patients receiving apixaban 2.5 or 5 mg b.i.d. are represented with blue and grey circles, respectively. We represented the median Cmax and Cmin and 5th–95th percentiles, i.e., 123 ng/mL (5th–95th: 69–221) and 79 ng/mL (5th–95th: 34–162), as reported in patients receiving apixaban 2.5 mg b.i.d. in ARISTOTLE trial11 (median age: 70 years), with solid and dashed lines, respectively. AF, atrial ﬁbrillation; b.i.d., twice daily; o.d., once daily. Fig. 1 Plasma concentrations of rivaroxaban (A) and apixaban (B) in ADAGE patients as a function of time since the last oral intake. (A) Fig. 1 Plasma concentrations of rivaroxaban (A) and apixaban (B) in ADAGE patients as a function of time since the last oral intake. (A) Patients receiving rivaroxaban 10, 15, or 20 mg o.d. are represented with yellow, red, and grey circles, respectively (noteworthy, 10 mg off- labeled in AF). In order to facilitate interpretation, we represented the geometric mean Cmax and Cmin and the 5th–95th percentiles, i.e., 229 ng/mL (5th–95th: 178–313) and 57 ng/mL (5th–95th: 18–136), as reported in ROCKET-AF patients (median age: 73 years) receiving 15 mg according to Girgis et al21 with solid and dashed lines, respectively. Thrombosis and Haemostasis © 2023. The Author(s). Association of Covariates with Apixaban Plasma Concentrations at Cmax and Cmin We observed a substantial inter-individual variability in ADAGE patientsonapixabanwithCVsof46and 47% atTmax and 61and 68% at Tmin for 2.5 and 5.0 mg b.i.d. regimens, respectively. In univariate analysis, the apixaban dose regimen was signiﬁ- cantly associated with apixaban concentrations, which were 44.9% higher (p ¼ 0.0247) at Tmax and 45.6% higher at Tmin (p ¼ 0.0317) in patients receiving the full dose (►Table 2). Two other variables were associated with higher apixaban Cmax: Adage Study Foulon-Pinto et al. Thrombosis and Haemostasis © 2023. The Author(s). Association of Covariates with Apixaban Plasma Concentrations at Cmax and Cmin 2 Peak height and endogenous thrombin potential as a function of DOAC concentrations in ADAGE patients and in normal po Fig. 2 Peak height and endogenous thrombin potential as a function of DOAC concentrations in ADAGE patients and in normal pooled plasma samples spiked with DOAC, using STG-Thrombo-Screen reagent in the absence of thrombomodulin, and inhibition percentage in the presence of thrombomodulin. Absolute values of TG parameters in the absence of thrombomodulin (TM) (left) and percentage of inhibition in the presence of TM (right). Samples from patients receiving 10, 15, or 20 mg of rivaroxaban o.d. are represented with yellow, red, and grey circles, respectively. Samples from patients receiving 2.5 or 5 mg of apixaban b.i.d. are represented with blue and grey circles, respectively. Samples from pooled normal plasma spiked with DOAC at varying concentrations are represented with green diamonds.35 b.i.d., twice daily; DOAC, direct oral anticoagulant; o.d., once daily; TG, thrombin generation. Fig. 2 Peak height and endogenous thrombin potential as a function of DOAC concentrations in ADAGE patients and in normal pooled plasma samples spiked with DOAC, using STG-Thrombo-Screen reagent in the absence of thrombomodulin, and inhibition percentage in the presence of thrombomodulin. Absolute values of TG parameters in the absence of thrombomodulin (TM) (left) and percentage of inhibition in the presence of TM (right). Samples from patients receiving 10, 15, or 20 mg of rivaroxaban o.d. are represented with yellow, red, and grey circles, respectively. Samples from patients receiving 2.5 or 5 mg of apixaban b.i.d. are represented with blue and grey circles, respectively. Samples from pooled normal plasma spiked with DOAC at varying concentrations are represented with green diamonds.35 b.i.d., twice daily; DOAC, direct oral anticoagulant; o.d., once daily; TG, thrombin generation. (p ¼ 0.0005), the latter explaining 18.1% of PH variance in univariate analysis. aban: n ¼ 6 [5.4%]); the rate of clinically relevant nonmajor bleeding was 3.3% (rivaroxaban: n ¼ 2 [1.9%]; apixaban: n ¼ 5 [4.5%]). The main bleeding sites were gastrointestinal (n ¼ 7), intracranial (n ¼ 3), and genitourinary tract (n ¼ 3). Five patients (2.3%) (rivaroxaban: n ¼ 1 [1.0%], apixaban: n ¼ 4 [3.6%]) had a thromboembolic event. Finally, 39 patients died (18.1%), 18 on rivaroxaban (17.3%) and 21 on apixaban (18.9%), with a median 3.4 month delay (1.8–4.6) after inclusion. No relationships were found between clinical events and PK/PD parameters. Association of Covariates with Apixaban Plasma Concentrations at Cmax and Cmin Table 2 Inﬂuence of ADAGE patients’ characteristics on xaban plasma concentrations at Tmax and Tmin Covariates Univariate analysis Multivariate analysis Cmax Cmin Cmax Cmin Effect % (explained variability) p Effect % (explained variability) p Effect p Effect p Rivaroxaban Dose regimen þ20.3%a [16.7; þ73.5] 2.23% 0.346 3.8%a [35.1; þ42.6] 0.72% 0.974 29.4%a [14.2; 95.2] 0.210 3.2%a [35.1; 44.6] 0.873 Female gender þ12.2% [17.7; 53.0] 4.09% 0.28 14.2% [36.0; 15.1] 3.34% 0.424 þ11.8% [20.8; 57.9] 0.515 10.1% [35.2; 22.3] 0.466 Creatinine clearance þ1.89%b [8.2; 13.1] 0.72% 0.83 þ8.1%b [3.9; 15.8] 0.57% 0.093 þ3.4%b [7.6; 15.6] 0.553 þ5.0%b [4.9; 16.0] 0.325 Amiodarone þ4.73% [31.5; 60.2] 0.07% 0.99 þ47.8% [2.8; 109.4] 2.23% 0.0315 26.1%b [75.8; 125.5] 0.345 99.8%c [10.3; 262.0] 0.023 P-gp inhibitorc 9.95% [35.3; 25.4] 1.42% 0.60 15.9% [37.6; 13.5] 1.47% 0.264 6.31% [35.8; 36.6] 0.200 8.7% [34.2; 26.8] 0.162 CYP3A4/5 modulatorc þ9.5% [21.0; 51.6] 1.20% 0.52 þ31.1% [24.7; 72.2] 3.45% 0.0361 þ10.6% [23.1; 58.9] 0.417 þ43.1% [5.3; 94.3] 0.023 ABCB1 3435C > T þ18.3%d [5.3; 36.6] 4.25% 0.20 þ16.9%d [6.6; 31.0] 1.39% 0.0788 þ18.3%d [10.9; 39.8] 0.188 þ10.4%d [14.9; 30.1] 0.381 Apixaban Dose regimen þ44.9%e [7.14; 96.1] 12.20% 0.025 þ45.6%e [4.7; 102.6] 7.45% 0.0317 þ64.8%e [16.6; 132.9] 0.006 þ54.8%e [6.65; 124.6] 0.022 Female gender þ44.0% [7.4; 93.1] 9.80% 0.017 þ15.7% [14.9; 57.1] 1.12% 0.348 þ28.7% [4.5; 73.4] 0.095 14.5% [17.3; 58.6] 0.410 Creatinine clearance þ3.3%b [4.28; 10.31] 0.98% 0.289 þ4.6%b [4.2; 12.7] 3.23% 0.166 þ4.5%b [3.0; 11.4] 0.223 þ4.5%b [4.9; 13.1] 0.332 Amiodarone þ43.5% [7.9; 90.9] 10.25% 0.0098 þ14.4% [18.5; 60.5] 0.58% 0.433 þ79.3% [5.5; 204.7] 0.032 þ44.1% [31.3; 202.2] 0.492 P-gp inhibitorc 12.7% [37.1; 21.0] 2.12% 0.408 þ5.7% [26.6; 52.1] 0.23% 0.764 4.0% [36.3; 44.8] 0.187 10.7% [43.7; 41.6] 0.808 CYP3A4/5 modulatorc þ4.0% [20.4; 35.9] 0.23% 0.769 þ20.8% [9.8; 61.7] 2.42% 0.203 þ12.1% [17.3; 51.9] 0.284 þ23.4% [11.1; 71.3] 0.330 ABCB1 3435C > T þ2.1%d [18.7; 19.3] 0.13% 0.938 þ14.8%d [4.1; 30.3] 1.70% 0.087 þ10.1%d [8.3; 25.3] 0.256 þ15.2%d [4.1; 30.9] 0.113 Note: signiﬁcance threshold p < 0.05 (bold values). a15 mg o.d. vs. 20 mg o.d. bPer decrease of 10 mL/min. cPatients who received a P-gp or CYP3A4/5 modulator in addition to amiodarone. dPer mutated allele. The 3 patients who received 10 mg o.d. were excluded from the statistical analysis. e2.5 mg b.i.d. vs. 5 mg b.i.d. Adage Study Foulon-Pinto et al. Fig. Association of Covariates with Apixaban Plasma Concentrations at Cmax and Cmin The univariate analysis performed in the apixaban group showed a PH decrease signiﬁcantly associated with increas- ing concentrations of apixaban, both at Tmax and Tmin (p ¼ 0.0224 and p ¼ 0.0005) (►Table 3), accounting for 12.4 and 16.6% of the observed variance, respectively. In addition, patients receiving amiodarone had a lower PH at Tmin than those without amiodarone intake (37.2%, p ¼ 0.0426). The multivariate analysis conﬁrmed that apixaban concentra- tions were the only determinant of the PH variability at Tmin (p ¼ 0.0074) and Tmax (p ¼ 0.0516) (►Table 3). Thrombosis and Haemostasis © 2023. The Author(s). Thrombosis and Haemostasis © 2023. The Author(s). Six-Month Follow-Up One of the original aims of our work was to explore the potential inﬂuence of pharmacogenetic factors on DOAC concentrations. We showed that the presence of CYP3A53 and CYP2J27 variants was not signiﬁcantly associated with the variability of rivaroxaban or apixaban, in agree- ment with previous studies.48–51 We observed a trend to higher concentrations in ABCB1-3435T carriers: interest- ingly, this variant has been shown to be associated with a reduced risk of thromboembolic outcomes in a large Finnish cohort study.52 Interestingly, an impact of an ABCG2 variant coding breast resistance coding protein was reported in a younger cohort and could be investigat- ed in ADAGE patients.50,53 Regarding apixaban, ADAGE patients receiving 2.5 mg b.i.d. had a median concentration at Tmax close to that of patients receiving 5 mg b.i.d., while the residual concentration in ADAGE patients was close to that reported in the Summary of Product Characteristics for 2.5 mg b.i.d.; patients receiving 5 mg b.i.d. had higher median Cmax and Cmin.11 In addition, inter-individual variabilities were higher than those reported in SmPC (i.e., around 30% from the ARISTOTLE data set, mean age 70 years) and in previous studies.11,18,19,23,24,27–29,42,43 Interestingly, the apixaban dose regimen was the only covariate that was signiﬁcantly associated with plasma Cmax and Cmin in multivariate analy- sis, in contrast to rivaroxaban; moreover, the apixaban regimen subsequently signiﬁcantly impacted TG peak height. If the apixaban dose regimen had been given accord- ing to the patient characteristics, one would have expected similar concentrations as for rivaroxaban. One explanation is that the apixaban reduced dose (2.5 mg b.i.d.) corresponds to half of the full dose (5 mg bid), while the rivaroxaban reduced dose (15 mg) corresponds to three-quarters of the full dose (20 mg). It is important to note that a substantial proportion of patients were theoretically underdosed for both xabans (31% of patients on apixaban, 27% on rivaroxaban). The fear of hemorrhagic complications probably explains the higher Another strength ofour study was toevaluate the TG proﬁle of rivaroxaban and apixaban in a subset of ADAGE patients using a standardized TG analyzer (ST-Genesia). Six-Month Follow-Up However, our study was underpowered to conclude whether dose reduction impacts clinical outcomes: this deserves speciﬁc future investigations in geriatrics. One strength of the ADAGE study was to provide DOAC PK data which are scarce in this speciﬁc setting of hospitalized very elderly patients.28,29 Indeed, knowing the extent of DOAC concentrations in octo- and nonagenarians should facilitate the interpretation of DOAC concentrations when monitoring such patients in speciﬁc situations. Mean Cmax and Cmin of ADAGE patients receiving rivaroxaban 15 mg were remarkably close to those reported in ROCKET-AF patients (mean age 73 years) receiving 20 mg,7 whereas mean con- centrations of ADAGE patients receiving 20 mg o.d. were higher, especially at Tmax. Moreover, we observed a larger inter-individual variability of Cmax and Cmin for both regi- mens compared to Mueck et al’s data, i.e., around 30%.7 Our data show similar or even greater CVs compared to those of previous observational studies conducted in patients rather younger than ours.21–23,25–27,42,43 Among covariates, female sex was signiﬁcantly associated with high apixaban Cmax in univariate analysis, as already reported by Frost et al in a PK study19 and real-life studies.48 Moreover, we also showed that the intake of amiodarone was a signiﬁcant determinant of the DOAC variability of Cmax in univariate analysis. However, although ADAGE patients were highly polymedicated with several co-prescribed CYP3A4/5 and/or P-gp inhibitors, no strong association was shown with DOAC concentrations in multivariate analysis within the limits of the study. Our results reinforce the literature data suggesting that rivaroxaban and apixaban do not interact with those drugs to the extent of a clinically relevant impact justifying a subsequent dose reduction. It should be noted that no patients were taking strong CYP3A4/5/P-gp inhib- itors or inducers, highlighting the compliance with DOAC contraindications in this cohort. Interestingly, we found that the intake of amiodarone, which is both a CYP3A4/5 and a P-gp inhibitor, as well as the additional intake of another CYP3A4/5 modulator, signiﬁ- cantly contributed to the increase of rivaroxaban trough levels, explaining part of the concentration variability con- sistently with data from pivotal trials.20 Besides, renal func- tion is known to inﬂuence the clearance of rivaroxaban, justifying dose adjustment in case of moderate renal im- pairment: we only observed a trend towards higher concen- trations at Tmin with low CrCl, suggesting that the 25% rivaroxaban dose reduction was appropriate in our patients. Six-Month Follow-Up ADAGE is, to the best of our knowledge, the ﬁrst study speciﬁ- cally focused on the assessment of both rivaroxaban/apix- aban concentrations and effects evaluated by TG in a real-life cohort of very elderly NVAF patients. Patients were very old We could not obtain data on bleedings and thrombosis in 11 patients (5 on rivaroxaban, 6 on apixaban). The overall rate of major bleedings was 6.0% (rivaroxaban: n ¼ 7 [6.7%]; apix- Thrombosis and Haemostasis © 2023. The Author(s). Adage Study Foulon-Pinto et al. e 3 Association of ADAGE patients’ characteristics including xaban plasma concentration and peak height (using STG-ThromboScreen reagent without thrombomodulin) variates Univariate Multivariate Tmax Tmin Tmax Tmin Effect % p Effect % p Effect p Effect p aroxaban Plasma concentrations þ23.9%a [11.4; 73.2] 8.4% 0.537 þ38.7%a [12.4;71.1] 22.1% 0.0018 –5.0%a [27.9; 25.2] 0.700 þ21.0%a [4.0; 52.4] 0.103 Creatinine clearance 12.8%b [26.3; 3.0] 22.9% 0.027 6.1%b [16.2; 5.2 ] 3.2% 0.184 16.4%b [26.3; 5.0] 0.0085 6.76%b [15.5; 2.8] 0.156 Heart failure 37.5% [59.8; 2.7] 19.34% 0.041 31.4% [52.2; 1.7] 10.8% 0.055 22.3% [45.0; 9.8] 0.143 12.7% [37.7; 22.4] 0.422 Amiodaronec N.Ac N.Ac. N.A. 51.5% [70.1; 21.2] 12.7% 0.014 N.A.c N.A.c 44.6% [75.1; 23.2] 0.228 ixaban Plasma concentrations þ80.2%a 10.0; 195.0] 22.6% 0.022 37.1%a 13.1; 66.2] 20.8% 0.0005 þ86.8%a [0.48; 250.5] 0.052 þ32.7%a [8.28; 62.59] 0.0074 Creatinine clearance þ0.42%b [4.70; 36.9] 0.08% 0.903 1.7%b [10.7; 15.8] 0.49% 0.566 þ2.4%b [17.9; 27.8] 0.825 1.9%b [13.4; 11.1] 0.753 Heart failure 26.7% [66.7; 61.4] 1.8% 0.426 3.2% [30.5; 53.3] 0.01% 0.935 9.2% [62.9; 122.6] 0.826 þ0.84% [30.8; 46.9] 0.965 Amiodaronec þ3.7% [58.7; 160.1] 0.5% 0.936 37.2% [60.0; 1.5] 6.30% 0.043 26.0% [91.8; 564.1] 0.759 57.47% [83.2; 7.9] 0.129 : Signiﬁcance threshold: p < 0 05 bold values Adage Study Foulon-Pinto et al. (mean age, 87 years), had substantial comorbidities and polypharmacy, frequently including the intake of amiodar- one and/or another CYP3A4/5 or P-gp inhibitor. Thus, we believe that the characteristics of the participants in ADAGE study are representative of the complex medical issues associated with anticoagulant treatment in NVAF patients of 80 years and over. proportion of off-label reduced dose among ADAGE patients compared with those described in real-life registries.44–47 Herein, we showed that the impact of the DOAC regimen choice, reduced versus full dose, on plasma concentrations and PD parameters was higher with apixaban than with rivaroxaban. Thrombosis and Haemostasis © 2023. The Author(s). References 1 Chugh SS, Havmoeller R, Narayanan K, et al. Worldwide epidemi- ology of atrial ﬁbrillation: a Global Burden of Disease 2010 Study. Circulation 2014;129(08):837–847 2 Hindricks G, Potpara T, Dagres N, et al; ESC Scientiﬁc Document Group. 2020 ESC Guidelines for the diagnosis and management of atrial ﬁbrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): the Task Force for the diagnosis and management of atrial ﬁbrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J 2021;42(05):373–498 Six-Month Follow-Up We conﬁrmed that both xabans exerted a concentration-dependent effect on TG parameters using both TF conditions; PHs were consistently more affected than ETP in agreement with previous ﬁndings in various series of muchyounger patients.31,35,54–57 The optimal choice between intermediate (STG-ThromboScreen) and high (STG-DrugScreen) pM-TF concentrations may depend on DOAC concentration, preferentially intermediate up to 300 ng/mL and high above 300ng/mL.57 Remarkably, we were able to evidence a highinter-individualvariabilityofTG parameters among patients receiving both full and reduced-dose xabans. Spiking pooled normal plasma with increasing concentrations of xabans allowed a better understanding of the inter-individ- ual variability and highlighted an underlying hypercoagulable state in some ADAGE patients, leading to shorter LT and TTP combined with higher PH and ETP. We had primarily checked that the pooled normal plasma contained normal procoagu- lant factors and natural coagulation inhibitor levels.35 More- over, for a given concentration, we observed under the same Adage Study Foulon-Pinto et al. conditions higher PH and ETP results in ADAGE rivaroxaban patients when compared to those obtained in DRIVING rivarox- aban healthy volunteers, aged 18 to 45 years, thus supporting this hypothesis.35,57 It is well known that advancing age and chronic comorbid conditions lead to an imbalance between procoagulant factors and coagulation natural inhibitors, po- tentially causes of hypercoagulability.58 This issue regarding TG variability deserves further studies in geriatric setting. Acknowledgements We are grateful to Guillaume Paris for technical assistance. Conﬂict of Interest V.S., I.G.-T., E.Pa, E.Pu, F.M., M.D., and P.G. have received honoraria for participating in expert meetings on dabiga- tran (from Boehringer Ingelheim), rivaroxaban (from Bayer Healthcare AG), or apixaban (Bristol Myers Squibb-Pﬁzer). G.F. and E.Pu received a grant from Bayer. The other authors declare no conﬂicts of interest. In conclusion, our study provides new real-life data regarding DOAC concentrations and TG inter-individual var- iability in elderly in-patients with NVAF, multiple comorbid- ities, and medications. The DOAC regimen choice (reduced vs. full dose) was signiﬁcantly associated with plasma con- centrations and TG PHs for apixaban, but not for rivaroxaban. The underlying coagulation status of elderly patients could affect TG proﬁles. Further research is needed to better understand the substantial PK and PD variability and its potential clinical impact in the rapidly growing population of very old patients with multiple conditions who need long- term anticoagulant treatment. Funding Information Reagents were purchased thanks to a CONNY-MAEVA Charitable Foundation grant and the Société Française de Cardiologie. The funding sources had no role in the design and conduct of the study, collection, management, analysis, and interpretation of the data. Thrombosis and Haemostasis © 2023. The Author(s). Author Contributions Our study has several limitations. First, although one to ﬁve samples per patient were planned over a 20-day period with an average of three samples per patient, a smaller number of blood samples were drawn (i.e., 2.1) because of short hospital stays. Moreover, the sample size clearly limits analyses of associations of outcomes with measures. Second, the results regarding the inﬂuence of individual factors on concentrations at Tmax and Tmin and PD obtained in this exploratory study must be conﬁrmed in an external cohort of hospitalized patients. Third, most patients were recruited during hospitalization: these results cannot be extrapolated to stable outpatients of similar age even though CIRS-scores took into account both chronic and acute comorbid condi- tions. Finally, our study was not designed to detect possible associations between PK/PD parameters measured only over a short period at inclusion and clinical events at 6 months. Author Contributions G.F.-P., V.S., G.J., and E.C. wrote the manuscript; C.L-L, V.S., E.P., and I.G.-T designed the research; C.L-L, J.L, E.P., and M. D. were clinical investigators; G.F.-P., V.S., G.J., F.T., F.K., and L.R. performed the research; G.F.-P., V.S., G.J., T.L., and E.C. analyzed the data; all authors reviewed the manuscript. What does this paper add? • Our academic prospective study provides original data in very elderly in-patients receiving xaban in real-life setting, showing great inter-individual variability in plasma concentrations and PD parameters. • Our academic prospective study provides original data in very elderly in-patients receiving xaban in real-life setting, showing great inter-individual variability in plasma concentrations and PD parameters. • Substantial variability of thrombin peak-heights was noticed at a given plasma concentration for both xabans, suggesting an impact of the underlying coagu- lation status of elderly in-patients. • Substantial variability of thrombin peak-heights was noticed at a given plasma concentration for both xabans, suggesting an impact of the underlying coagu- lation status of elderly in-patients. The overall major bleeding rate of 6.0% at 6 months was not surprisingly higher than that reported in phase III trials in patients >75 years given the characteristics of ADAGE patients.8,9 The predominance of gastrointestinal bleedings could be expected in these patients, as shown in previous studies and meta-analysis.13–15,59 The mortality rate (18.1%) was similar in patients treated with rivaroxaban and apixaban. The recruitment of ADAGE patients in geriatrics units probably reﬂects a high degree of frailty in ADAGE patients. • The dose regimen (reduced dose vs. full dose) had a signiﬁcant impact on plasma concentrations at Tmax and Tmin in apixaban, but not in rivaroxaban patients. • The dose regimen (reduced dose vs. full dose) had a signiﬁcant impact on plasma concentrations at Tmax and Tmin in apixaban, but not in rivaroxaban patients. What is known about this topic? • Very elderly patients (80 years) are characterized by numerous comorbid conditions (e.g., renal im- pairment) and polypharmacy, potentially altering drug PK/PD. • Few speciﬁc DOAC pharmacokinetics (PK) and phar- macodynamics (PD) data are available in this age group with numerous comorbid conditions and polypharmacy. 3 January CT, Wann LS, Calkins H, et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society in Adage Study Foulon-Pinto et al. Collaboration With the Society of Thoracic Surgeons. Circulation 2019;140(02):e125–e151 rivaroxaban, and apixaban. Can J Cardiol 2013;29(7, Suppl): S24–S33 21 Girgis IG, Patel MR, Peters GR, et al. Population pharmacokinetics and pharmacodynamics of rivaroxaban in patients with non- valvular atrial ﬁbrillation: results from ROCKET AF. J Clin Phar- macol 2014;54(08):917–927 4 Hanna MS, Mohan P, Knabb R, Gupta E, Frost C, Lawrence JH. Development of apixaban: a novel anticoagulant for prevention of stroke in patients with atrial ﬁbrillation. Ann N Y Acad Sci 2014; 1329:93–106 5 Kubitza D, Becka M, Roth A, Mueck W. Dose-escalation study of the pharmacokinetics and pharmacodynamics of rivaroxaban in healthy elderly subjects. Curr Med Res Opin 2008;24(10): 2757–2765 22 Al-Aieshy F, Malmström RE, Antovic J, et al. Clinical evaluation of laboratory methods to monitor exposure of rivaroxaban at trough and peak in patients with atrial ﬁbrillation. Eur J Clin Pharmacol 2016;72(06):671–679 23 Testa S, Tripodi A, Legnani C, et al; START-Laboratory Register. Plasma levels of direct oral anticoagulants in real life patients with atrial ﬁbrillation: results observed in four anticoagulation clinics. Thromb Res 2016;137:178–183 6 Kubitza D, Becka M, Voith B, Zuehlsdorf M, Wensing G. Safety, pharmacodynamics, and pharmacokinetics of single doses of BAY 59-7939, an oral, direct factor Xa inhibitor. Clin Pharmacol Ther 2005;78(04):412–421 7 Mueck W, Stampfuss J, Kubitza D, Becka M. Clinical pharmacoki- netic and pharmacodynamic proﬁle of rivaroxaban. Clin Pharma- cokinet 2014;53(01):1–16 24 Cirincione B, Kowalski K, Nielsen J, et al. Population pharmacoki- netics of apixaban in subjects with nonvalvular atrial ﬁbrillation. CPT Pharmacometrics Syst Pharmacol 2018;7(11):728–738 8 Patel MR, Mahaffey KW, Garg J, et al; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial ﬁbrillation. N Engl J Med 2011;365(10):883–891 25 Testa S, Paoletti O, Legnani C, et al. What is known about this topic? Low drug levels and thrombotic complications in high-risk atrial ﬁbrillation patients treated with direct oral anticoagulants. J Thromb Haemost 2018;16(05): 842–848 9 Granger CB, Alexander JH, McMurray JJV, et al; ARISTOTLE Com- mittees and Investigators. Apixaban versus warfarin in patients with atrial ﬁbrillation. N Engl J Med 2011;365(11):981–992 26 Miklič M, Mavri A, Vene N, et al. Intra- and inter- individual rivaroxaban concentrations and potential bleeding risk in patients with atrial ﬁbrillation. Eur J Clin Pharmacol 2019;75 (08):1069–1075 10 BayerAG.Xarelto 20 mgﬁlm-coatedtabletsSmPC. Publishedonline January 17, 2020. Accessed October 8, 2020 at: https://www.ema. europa.eu/en/documents/product-information/xarelto-epar-prod- uct-information_en.pdf 27 Bendayan M, Mardigyan V, Williamson D, et al. Muscle mass and direct oral anticoagulant activity in older adults with atrial ﬁbrillation. J Am Geriatr Soc 2021;69(04):1012–1018 11 Bristol-Myers Squibb/Pﬁzer EEIG. Eliquis 5mg ﬁlm-coated tablets SmPC. Published online September 16, 2020. Accessed October 8, 2020 at: https://www.ema.europa.eu/en/documents/product-infor- mation/eliquis-epar-product-information_en.pdf 28 Nissan R, Spectre G, Hershkovitz A, et al. Apixaban levels in octogenarian patients with non-valvular atrial ﬁbrillation. Drugs Aging 2019;36(02):165–177 29 Sukumar S, Gulilat M, Linton B, et al. Apixaban concentrations with lower than recommended dosing in older adults with atrial ﬁbrillation. J Am Geriatr Soc 2019;67(09):1902–1906 12 Sharma M, Cornelius VR, Patel JP, Davies JG, Molokhia M. Efﬁcacy and harms of direct oral anticoagulants in the elderly for stroke prevention in atrial ﬁbrillation and secondary prevention of venous thromboembolism: systematic review and meta-analysis. Circulation 2015;132(03):194–204 30 Tripodi A. Thrombin generation assay and its application in the clinical laboratory. Clin Chem 2016;62(05):699–707 13 Mitchell A, Watson MC, Welsh T, McGrogan A. Effectiveness and safety of direct oral anticoagulants versus vitamin k antagonists for people aged 75 years and over with atrial ﬁbrillation: a systematic review and meta-analyses of observational studies. J Clin Med 2019;8(04):E554 31 Pfrepper C, Metze M, Siegemund A, Klöter T, Siegemund T, Petros S. Direct oral anticoagulant plasma levels and thrombin genera- tion on ST Genesia system. Res Pract Thromb Haemost 2020;4 (04):619–627 32 Binder NB, Depasse F, Mueller J, et al. Clinical use of thrombin generation assays. J Thromb Haemost 2021;19(12):2918–2929 14 Alexander JH, Andersson U, Lopes RD, et al; Apixaban for Reduc- tion of Stroke and Other Thromboembolic Complications in Atrial Fibrillation (ARISTOTLE) Investigators. Apixaban 5 mg twice daily and clinical outcomes in patients with atrial ﬁbrillation and advanced age, low body weight, or high creatinine: a secondary analysis of a randomized clinical trial. Thrombosis and Haemostasis © 2023. The Author(s). What is known about this topic? Deﬁnition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3(04):692–694 International Society on Thrombosis and Haemostasis. Deﬁnition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3(04):692–694 50 Ueshima S, Hira D, Fujii R, et al. Impact of ABCB1, ABCG2, and CYP3A5 polymorphisms on plasma trough concentrations of apixaban in Japanese patients with atrial ﬁbrillation. Pharmaco- genet Genomics 2017;27(09):329–336 40 Pinheiro J, Bates D, DebRoy S, Sarkar D Nlme: linear and nonlinear mixed effects models. 2021. Accessed December 13, 2022 at: https://CRAN.R-project.org/package¼nlme 51 Gouin-Thibault I, Delavenne X, Blanchard A, et al. Interindividual variability in dabigatran and rivaroxaban exposure: contribution of ABCB1 genetic polymorphisms and interaction with clarithro- mycin. J Thromb Haemost 2017;15(02):273–283 41 R Core Team R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. 2020. Accessed December 13, 2022 at: https://www.R- project.org/ 52 Lähteenmäki J, Vuorinen AL, Pajula J, et al. Pharmacogenetics of bleeding and thromboembolic events in direct oral anticoagulant users. Clin Pharmacol Ther 2021;110(03):768–776 42 Hirsh Raccah B, Rottenstreich A, Zacks N, et al. Appropriateness of direct oral anticoagulant dosing and its relation to drug levels in atrial ﬁbrillation patients. J Thromb Thrombolysis 2019;47(04): 550–557 53 Gulilat M, Keller D, Linton B, et al. Drug interactions and phar- macogenetic factors contribute to variation in apixaban concen- tration in atrial ﬁbrillation patients in routine care. J Thromb Thrombolysis 2020;49(02):294–303 43 Comans AL, Sennesael AL, Bihin B, Regnier M, Mullier F, de Saint- Hubert M. Inappropriate low dosing of direct oral anticoagulants in older patients with non-valvular atrial ﬁbrillation: impact on plasma drug levels. Thromb Res 2021;201:139–142 54 Kreutz R, Persson PB, Kubitza D, et al. Dissociation between the pharmacokinetics and pharmacodynamics of once-daily rivarox- aban and twice-daily apixaban: a randomized crossover study. J Thromb Haemost 2017;15(10):2017–2028 44 Steinberg BA, Shrader P, Thomas L, et al; Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) Inves- tigators and Patients. Factors associated with non-vitamin K antagonist oral anticoagulants for stroke prevention in patients with new-onset atrial ﬁbrillation: results from the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation II (ORBIT-AF II). Am Heart J 2017;189:40–47 55 Artang R, Anderson M, Riley P, Nielsen JD. Assessment of the effect of direct oral anticoagulants dabigatran, rivaroxaban, and apix- aban in healthy male volunteers using a thrombin generation assay. What is known about this topic? JAMA Cardiol 2016;1(06): 673–681 33 Meihandoest T, Studt JD, Mendez A, et al. Automated thrombin generation assay for rivaroxaban, apixaban, and edoxaban meas- urements. Front Cardiovasc Med 2021;8:717939 34 Hemker HC, Al Dieri R, De Smedt E, Béguin S. Thrombin genera- tion, a function test of the haemostatic-thrombotic system. Thromb Haemost 2006;96(05):553–561 35 Foulon-Pinto G, Jourdi G, Perrin J, et al. Study of thrombin generation with St Genesia to evaluate xaban pharmacodynam- ics: analytical performances over 18 months. Int J Lab Hematol 2021;43(04):821–830 15 Deitelzweig S, Keshishian A, Li X, et al. Comparisons between oral anticoagulants among older nonvalvular atrial ﬁbrillation patients. J Am Geriatr Soc 2019;67(08):1662–1671 16 Kubitza D, Becka M, Roth A, Mueck W. The inﬂuence of age and gender on the pharmacokinetics and pharmacodynamics of rivaroxaban–an oral, direct Factor Xa inhibitor. J Clin Pharmacol 2013;53(03):249–255 36 Douxﬁls J, Adcock DM, Bates SM, et al. 2021 Update of the International Council for Standardization in Haematology recom- mendations for laboratory measurement of direct oral antico- agulants. Thromb Haemost 2021;121(08):1008–1020 17 Mueck W, Schwers S, Stampfuss J. Rivaroxaban and other novel oral anticoagulants: pharmacokinetics in healthy subjects, spe- ciﬁc patient populations and relevance of coagulation monitor- ing. Thromb J 2013;11(01):10 37 Miller MD, Paradis CF, Houck PR, et al. Rating chronic medical illness burden in geropsychiatric practice and research: applica- tion of the Cumulative Illness Rating Scale. Psychiatry Res 1992; 41(03):237–248 18 Frost C, Wang J, Nepal S, et al. Apixaban, an oral, direct factor Xa inhibitor: single dose safety, pharmacokinetics, pharmacody- namics and food effect in healthy subjects. Br J Clin Pharmacol 2013;75(02):476–487 38 Service de pharmacologie et toxicologie cliniques - Hôpitaux universitaires de Genève. Interactions médicamenteuses, cyto- chromes P450 et P-glycoprotéine (Pgp). Published 2020. Accessed June 20, 2022 at: https://www.hug.ch/sites/interhug/ﬁles/ structures/pharmacologie_et_toxicologie_cliniques/images/car- te_des_cytochromes_2020.pdf 19 Frost CE, Song Y, Shenker A, et al. Effects of age and sex on the single-dose pharmacokinetics and pharmacodynamics of apixa- ban. Clin Pharmacokinet 2015;54(06):651–662 20 Gong IY, Kim RB. Importance of pharmacokinetic proﬁle and variability as determinants of dose and response to dabigatran, 39 Schulman S, Kearon CSubcommittee on Control of Anticoagula- tion of the Scientiﬁc and Standardization Committee of the Adage Study Foulon-Pinto et al. 49 Raymond J, Imbert L, CousinT, et al. Pharmacogenetics of direct oral anticoagulants: a systematic review. J Pers Med 2021;11(01):37 International Society on Thrombosis and Haemostasis. Thrombosis and Haemostasis © 2023. The Author(s). What is known about this topic? Res Pract Thromb Haemost 2017;1(02):194–201 56 Bertaggia-Calderara D, Kröll D, Gerschheimer C, et al. Effect of rivaroxaban on thrombin generation in vivo. A study in obese patients. Int J Lab Hematol 2018;40(01):e11–e14 45 Yao X, Tangri N, Gersh BJ, et al. Renal outcomes in anticoagulated patients with atrial ﬁbrillation. J Am Coll Cardiol 2017;70(21): 2621–2632 57 Siguret V, Abdoul J, Delavenne X, et al. Rivaroxaban pharmaco- dynamics in healthy volunteers evaluated with thrombin gener- ation and the active protein C system: modeling and assessing interindividual variability. J Thromb Haemost 2019;17(10): 1670–1682 46 Camm AJ, Cools F, Virdone S, et al; GARFIELD-AF Investigators. Mortality in patients with atrial ﬁbrillation receiving nonrecom- mended doses of direct oral anticoagulants. J Am Coll Cardiol 2020;76(12):1425–1436 47 Sugrue A, Sanborn D, Amin M, et al. Inappropriate dosing of direct oral anticoagulants in patients with atrial ﬁbrillation. Am J Cardiol 2021;144:52–59 58 Sepúlveda C, Palomo I, Fuentes E. Primary and secondary hae- mostasis changes related to aging. Mech Ageing Dev 2015; 150:46–54 48 Roşian AN, Roşian ŞH, Kiss B, et al. Interindividual variability of apixaban plasma concentrations: inﬂuence of clinical and genetic factors in a real-life cohort of atrial ﬁbrillation patients. Genes (Basel) 2020;11(04):E438 59 Hanon O, Vidal JS, Pisica-Donose G, et al; SAFIR study group. Bleeding risk with rivaroxaban compared with vitamin K antag- onists in patients aged 80 years or older with atrial ﬁbrillation. Heart 2021;107(17):1376–1382 Thrombosis and Haemostasis © 2023. The Author(s).
https://openalex.org/W2297232337	https://europepmc.org/articles/pmc4799593?pdf=render	English	null	Methicillin-susceptible Staphylococcus aureus skin infections among military conscripts undergoing basic training in Bangkok, Thailand, in 2014	BMC research notes	2,016	cc-by	4,628	© 2016 Nivesvivat et al. 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. Background On June 4, 2014, 25 conscripts were found to possess abscesses on their arms and elbows. This outbreak of infection resulted in one death. The Clinical Epidemiol- ogy Unit investigated other soldiers at the basic military camp on that day to determine the cause(s) of the infec- tion and recommend appropriate prevention and con- trol strategies. In April each year, approximately 9–10 % of men aged 21 years with a good health profile from all provinces are randomly selected to serve in the army, beginning with 10 weeks of basic military training. The Correspondence: rrangsin@pmk.ac.th 1 Department of Military and Community Medicine, Phramongkutklao College of Medicine, 315, Ratchawithi Road, Ratchathewi, Bangkok 10400, Thailand Full list of author information is available at the end of the article Abstract Background: Skin and soft tissue infections are common among military conscripts undergoing close-contact training activities. On June 4, 2014, an outbreak of Staphylococcus aureus skin infection was reported among military conscripts undergoing basic training in Bangkok, Thailand. An investigation was performed to verify the outbreak and recommend future prevention and control strategies. Case presentation: The outbreak resulted in a rate of infection of 19.2 % and a fatality rate of 2.5 % (one death). All were Thai men aged 21.2 ± 1.0 years. Risk factors associated with infection were multiple erythematous papules and training in certain subunits. Randomly selected isolates were evaluated by pulsed field gel electrophoresis to confirm the clonal identity. Conclusions: This report confirms that S. aureus skin infection can be fatal. Our study highlights the role of military personnel in the early detection, prompt treatment, and containment of outbreaks of skin infection, as well as other health issues among conscripts. Keywords: Methicillin-susceptible Staphylococcus aureus, Skin infection, Military, Outbreak reported outbreak occurred 2–6 weeks after training started, a time when training becomes more intensive. Methicillin‑susceptible Staphylococcus aureus skin infections among military conscripts undergoing basic training in Bangkok, Thailand, in 2014 Thirapa Nivesvivat1, Dusit Janthayanont2, Mathirut Mungthin3, Julphat Intarasuphit4, Siriwan Paojinda2, Kanya Phanitorn5, Paijit Permpool5, Saowapap Kasinant5, Onuma Pengpinij5, Parichart Yingprasert6, Wanida Thaochelee6 and Ram Rangsin1 BMC Research Notes BMC Research Notes Nivesvivat et al. BMC Res Notes (2016) 9:179 DOI 10.1186/s13104-016-1989-3 Case presentationh Pus were collected from the sites of infection. The clini- cal isolates were sent for bacterial culture at the Division of Microbiology, Department of Clinical Pathology, Phra- mongkutklao Hospital. Standard bacteriological cultur- ing by inoculation of blood agar was performed, as well as antibiotic susceptibility testing using the disk diffusion method. During our investigation, antibiotic suscepti- bilities to oxacillin, gentamicin, ciprofloxacin, levofloxa- cin, trimethoprim-sulfamethoxazole, chloramphenicol, tetracycline, clindamycin, fusidic acid, fosfomycin, lin- ezolid, vancomycin, and erythromycin were determined. Isolates that were positive for S. aureus were forwarded to the Reference Laboratory of the National Institute of Health, Department of Medical Sciences, Ministry of Public Health, Thailand, where they were confirmed to be S. aureus by standard culture techniques. The PCR tech- nique adopted by the National Institute of Health accom- panied sensitivity analysis by means of disk diffusion (Kirby Bauer) and E-test (vancomycin) assays to differen- tiate MSSA and methicillin-resistant S. aureus (MRSA) [2]. We randomly selected isolates to be further evaluated by pulsed field gel electrophoresis (PFGE) [3] to confirm the clonal identity. A 21-year-old Thai male conscript was admitted to the hospital after fainting with urinary and stool inconti- nence during combat training on June 4, 2014. He was primarily assumed to be suffering from heat syncope. He was a previously healthy man from subunit number 2, who was 20 years old with a BMI of 21.4 kg/m2. Six days before being admitted, abscesses were detected on both arms and elbows and the military camp prescribed oral amoxicillin. Three days later, he developed a high fever, nausea, and vomiting. The Emergency Department examined his vital signs, revealing a body temperature at 38 °C, dyspnea and tachypnea (30/min), tachycardia (140/min), and a blood pressure of 130/80 mmHg. Later, his blood pressure dropped to 85/50 mmHg. The doc- tor administered normal saline infusion, accessed the central venous line for inotropic drugs to increase his blood pressure, and gave broad-spectrum antibiotics. Ceftriaxone (2 g per day) and clindamycin (600 mg every 8 h) were administered intravenously. The conscript was then sent to the intensive care unit, where he remained conscious. Physical examination revealed multiple rup- tured abscesses on an erythematous edematous base of the left elbow. Laboratory tests revealed a raised level of creatinine phosphokinase (10 times higher than normal), as well as an electrolyte imbalance. Liver and renal pan- els were also found to be abnormal. Urinalysis showed hematuria and pyuria. Environmental investigation Environmental samples, comprising combat uniforms, bathing water, and bath reservoirs shared among recruits were collected for bacterial culture using blood agar. The morning after the investigation was completed, the inves- tigation team prescribed antibiotics for all cases. A case was defined as a person who was clinically diag- nosed with one or more abscesses from mid-May to the first 2 weeks of June. Of 213 recruits, 41 (19.2 %), includ- ing the deceased patient, were documented as contract- ing the infection, which was used as the case rate for the whole unit. The index case is thought to have occurred on May 15, 2014 (Fig. 1). Approximately 2 weeks later, four cases were reported from May 30 to 31. On June 1, Epidemiological investigationh The descriptive data were collected on June 5, 2014. Of 213 recruits, 212 completed the self-reported question- naire. The information in the questionnaire included demographic data, clinical manifestation, date of onset, number of lesions, and location(s) of lesions on their body. Demographic questions included age, sex, edu- cation, underlying diseases, and unit types. Clinical manifestations comprised (1) abscesses (2) multiple erythematous papules and (3) fever. Those reporting abscesses were considered as probable cases. The height and weight of the recruits were obtained from heat stroke Nivesvivat et al. BMC Res Notes (2016) 9:179 Page 2 of 6 screening data to calculate the body mass index (BMI) [1] and identify overweight or obese cases. screening data to calculate the body mass index (BMI) [1] and identify overweight or obese cases. The methodologies employed were univariate and mul- tivariate analyses. Using STATA software, the method- ologies resulted in crude odds ratio and adjusted odds ratio, respectively, and 95 % confidence intervals. The variables with significant p values less than 0.05 were fur- ther considered. A case was defined as a person who was clinically diag- nosed with an infectious skin lesion, including abscesses and furuncles, from mid-May to the first 2 weeks of June. The infecting bacteria were confirmed to be methicillin- susceptible Staphylococcus aureus (MSSA) using stand- ard culturing techniques. Healthcare providers and recruits were interviewed about their daily activities, such as hand-to-hand combat training, the characteris- tics of physical training, and hygiene. Ethical approval was obtained from the Institutional Review Board of the Royal Thai Army Medical Depart- ment. Confidentiality was maintained throughout and after the study. Names were not included on the questionnaire for confidentiality reasons. The written informed consent for publication of the clinical details from the next of kin of the deceased patient was also obtained. In the case of the deceased patient, his medical records, signs and symptoms, laboratory information, treatment, and cause of death were reconsidered. Case presentationh The pus culture collected from his elbow and hemoculture indicated MSSA infection. Despite prompt treatment, the patient passed away.i Statistical analysish The case rates were calculated both within each unit and for the whole battalion. Descriptive data were gathered by means of a questionnaire. These data were used to find an association between cases and risk factors. Case fatal- ity was calculated. Nivesvivat et al. BMC Res Notes (2016) 9:179 Page 3 of 6 Fig. 1 Epidemic curve of skin abscesses in conscripts by date of onset (n = 41) Fig. 1 Epidemic curve of skin abscesses in conscripts by date of onset (n = 41) Table 1 Characteristics of all conscripts during basic mili- tary training in Bangkok, Thailand (n = 212) Demographic data Total (%) Age (mean ± SD) 21.2 (±1.0) 20 19 (8.9) 21 156 (73.2) 22 21 (9.9) 23 10 (4.7) ≥24 7 (3.3) Education Bachelor’s 8 (4) High vocational certificate 14 (7) Vocational certificate 22 (10.9) High school 106 (52.7) Secondary school 51 (25.4) Subunit 1 54 (25.4) 2 51 (23.9) 3 55 (25.8) 4 53 (24.9) Weight (kg) (mean ± SD) 64.3 ± 11.4 Height (cm) (mean ± SD) 170.2 ± 6.0 BMI (kg/m2) (mean ± SD) 22.2 ± 3.6 Underweight <18.5 27 (12.7) Normal range (18.5–22.9) 119 (55.9) At risk of obesity (23–24.9) 29 (13.6) Obese I (25–29.9) 29 (13.6) Obese II (≥30) 9 (4.2) Symptoms Abscess 41 (19.2) Number of abscess lesions (mean ± SD) 2.2 ± 2 Multiple erythematous papules 86 (39.0) Table 1 Characteristics of all conscripts during basic mili- tary training in Bangkok, Thailand (n = 212) the number of patients abruptly rose to 17 cases (41.5 %) in 1 day, reaching the highest peak for skin infections. The case fatality rate was 2.4 % (1/41). Table 1 Characteristics of all conscripts during basic mili- tary training in Bangkok, Thailand (n = 212) h All conscripts at this facility were male, with a mean age of 21.2 ± 1.0 years. They were divided into four subunits in the same facility. The clinical manifestations included abscesses (19.2 %), multiple erythematous pap- ules (39.0 %), and fever (16.9 %). The lesions were located on elbows (46.3 %), shoulders (24.4 %), and legs (14.6 %).h The percentage of new conscripts who reported mul- tiple erythematous papules were 51.9, 37.3, 32.7, and 39.6 % for subunits 1, 2, 3, and 4, respectively (p = 0.211). Statistical analysish of isolates tested (%) Oxa Gen Cip Co-tri Chlo Tet Cli Van Ery 22 22 (100.0 %) 22 (100.0 %) 22 (100.0 %) 22 (100 %) 14 (63.6 %) 4 (18.1 %) 19 (86.6 %) 22 (100.0 %) 22 (100.0 %) Table 2 Univariate and multivariate analysis for risk factors of acquiring skin abscesses Risk factors Total (%) Case (%) P value Crude OR (95 % CI) P value Adjusted OR (95 % C Age (mean ± SD) 21.2 (±1.0) 21.0 (±1.0) 0.033 0.5 (0.3–0.9) 0.036 0.5 (0.3–1.0) Education Bachelor 8 (4) 3 (37.5) 1 High vocational certificate 14 (7) 4 (28.6) 0.666 0.7 (0.1–4.2) Vocational certificate 22 (10.9) 5 (22.7) 0.423 0.5 (0.1–2.8) High school 106 (52.7) 19 (17.9) 0.191 0.4 (0.1–1.7) Secondary school 51 (25.4) 7 (13.7) 0.112 0.3 (0.1–1.4) Subunit 1 54 (25.4) 20 (37) 0.001 30.6 (3.9–238.6) 0.001 32.6 (4.1–260.6) 2 51 (23.9) 16 (31.4) 0.003 23.8 (3.0–187.5) 0.002 26.9 (3.3–218.2) 3 55 (25.8) 4 (7.3) 0.216 4.1 (0.4–37.7) 0.178 4.6 (0.5–44.1) 4 53 (24.9) 1 (1.9) 1 BMI (kg/m2) Underweight <18.5 27 (12.7) 7 (25.9) 0.275 1.7 (0.6–4.6) Normal range (18.5–22.9) 119 (55.9) 20 (16.8) 1 At risk of obesity (23–24.9) 29 (13.6) 5 (17.2) 0.955 1.0 (0.4–3.0) Obese I (25–29.9) 29 (13.6) 7 (24.1) 0.362 1.6 (0.6–4.2) Obese II (≥30) 9 (4.2) 2 (22.2) 0.679 1.4 (0.3–7.3) Multiple erythematous papules Yes 86 (39.0) 27 (31.4) <0.01 3.7 (1.8–7.6) <0.01 3.4 (1.5–7.5) No 127 (59.6) 14 (11) 1 Table 2 Univariate and multivariate analysis for risk factors of acquiring skin abscesses Table 3 Staphylococcus aureus susceptibility test results for all available isolates oxa oxacillin, gen gentamicin, cip ciprofloxacin, co-tri trimethoprim sulfamethoxazole, chlo chloramphenicol, tet tetracycline, cli clindamycin, van vancomycin, ery erythromycin No. of isolates Antimicrobial, no. of isolates tested (%) Oxa Gen Cip Co-tri Chlo Tet Cli Van Ery 22 22 (100.0 %) 22 (100.0 %) 22 (100.0 %) 22 (100 %) 14 (63.6 %) 4 (18.1 %) 19 (86.6 %) 22 (100.0 %) 22 (100.0 %) Table 3 Staphylococcus aureus susceptibility test results for all available isolates dicloxacillin (250 mg per oral qid) was prescribed for all cases. reported fever; 86 % were from subunits 1 or 2, and all had a history of skin abscess (es). In addition, three cul- tures (10.3 %) were positive for Bacillus spp. Statistical analysish BMC Res Notes (2016) 9:179 Page 4 of 6 Table 2 Univariate and multivariate analysis for risk factors of acquiring skin abscesses Risk factors Total (%) Case (%) P value Crude OR (95 % CI) P value Adjusted OR (95 % CI) Age (mean ± SD) 21.2 (±1.0) 21.0 (±1.0) 0.033 0.5 (0.3–0.9) 0.036 0.5 (0.3–1.0) Education Bachelor 8 (4) 3 (37.5) 1 High vocational certificate 14 (7) 4 (28.6) 0.666 0.7 (0.1–4.2) Vocational certificate 22 (10.9) 5 (22.7) 0.423 0.5 (0.1–2.8) High school 106 (52.7) 19 (17.9) 0.191 0.4 (0.1–1.7) Secondary school 51 (25.4) 7 (13.7) 0.112 0.3 (0.1–1.4) Subunit 1 54 (25.4) 20 (37) 0.001 30.6 (3.9–238.6) 0.001 32.6 (4.1–260.6) 2 51 (23.9) 16 (31.4) 0.003 23.8 (3.0–187.5) 0.002 26.9 (3.3–218.2) 3 55 (25.8) 4 (7.3) 0.216 4.1 (0.4–37.7) 0.178 4.6 (0.5–44.1) 4 53 (24.9) 1 (1.9) 1 BMI (kg/m2) Underweight <18.5 27 (12.7) 7 (25.9) 0.275 1.7 (0.6–4.6) Normal range (18.5–22.9) 119 (55.9) 20 (16.8) 1 At risk of obesity (23–24.9) 29 (13.6) 5 (17.2) 0.955 1.0 (0.4–3.0) Obese I (25–29.9) 29 (13.6) 7 (24.1) 0.362 1.6 (0.6–4.2) Obese II (≥30) 9 (4.2) 2 (22.2) 0.679 1.4 (0.3–7.3) Multiple erythematous papules Yes 86 (39.0) 27 (31.4) <0.01 3.7 (1.8–7.6) <0.01 3.4 (1.5–7.5) No 127 (59.6) 14 (11) 1 Table 3 Staphylococcus aureus susceptibility test results for all available isolates oxa oxacillin, gen gentamicin, cip ciprofloxacin, co-tri trimethoprim sulfamethoxazole, chlo chloramphenicol, tet tetracycline, cli clindamycin, van vancomycin, ery erythromycin No. of isolates Antimicrobial, no. Statistical analysish and one culture (3.4 %) was positive for the following organisms: Corynebacterium spp., Streptococcus pyogenes, Klebsiella pneumoniae (ESBL-producing strain), Enterococcus fae- calis, Citrobacter spp., Acinetobacter baumannii, and Pseudomonas stutzeri. Statistical analysish While the percentage of conscripts who displayed abscesses were 37.0, 31.4, 7.3, and 1.9 % for subunits 1, 2, 3, and 4, respectively (p < 0.01) (Table 1). Adjusted odds ratios for belonging to subunits 1 and 2 were 32.6 (4.1– 260.6, 95 % CI) and 26.9 (3.3–218.2, 95 % CI) respectively. Multiple erythematous papules were significantly asso- ciated (p < 0.01) with abscesses. The adjusted odds ratio was 3.4 (1.5–7.5, 95 % CI) (Table 2). h The data gathered from interviews with healthcare providers and conscripts revealed that 1–2 weeks before the outbreak, part of the training of subunit 1 and 2 con- scripts involved low crawling on a concrete floor. l A dermatologist from the investigation team was able to identify various kinds of skin lesions including multi- ple erythematous papules, multiple discrete small dermal pustules with surrounding erythema, and skin abscesses. Pus from those suffering from abscesses was collected on June 5, 2014. From 29 clinical cultures, 22 (75.9 %), including one blood culture of the deceased patient, grew MSSA isolates. Among all 22 subjects who were posi- tive for MSSA culture (Table 3): one person died; 50.0 % Nivesvivat et al. Conclusions Prevalence of MSSA nasal colonization among healthy young adults in Thailand has been reported to be 15 %, while prevalence of MRSA nasal carriage was 1 % [4]. Patients infected with MSSA were also reportedly more prevalent than patients infected with MRSA [5]. Among soldiers in the US army, a study reported that 38 % were colonized with MSSA and 3.9 % were colonized with MRSA [6]. A total of 11 randomly selected isolates, including one isolate collected from blood culture of the deceased patient, were evaluated by PFGE to confirm clonal iden- tity revealing four genetic patterns (Fig. 2). Sa 35/37 and Sa 43/57 had 100 % genetic similarity. Sa 34/57 (from the deceased patient), Sa 36/57, Sa 37/57, Sa 38/57, Sa 41/57, Sa 42/57 and Sa 44/57 also showed 100 % genetic similar- ity. Sa 39/57 and Sa 40/57 had their own genetic patterns. All environmental swabs were negative for bacterial cul- tures. The morning after the investigation was completed, Investigations revealed that the pathogen responsible for abscesses in military populations in this present study was MSSA. This organism has generally been reported to be less harmful than MRSA, and more susceptible to antimicrobial therapy [7]. However, here we report a con- script who displayed abscesses and consequently died of Page 5 of 6 Nivesvivat et al. BMC Res Notes (2016) 9:179 Fig. 2 Dendrogram of methicillin-susceptible Staphylococcus aureus, as analyzed by the BioNumerics software Fig. 2 Dendrogram of methicillin-susceptible Staphylococcus aureus, as analyzed by the BioNumerics software MSSA skin infection was associated with the PVL gene. In one study, some MSSA isolates from blood culture or skin lesions were found to be positive for the PVL gene [5]. MSSA septic shock. The cultures from skin tissue and hemoculture were positive for MSSA. Considering the laboratory results and the fact that prompt treatment was initiated but was ineffective, the patient most likely suf- fered septic shock and multiorgan failure. Staphylococcus aureus usually colonizes the anterior nares [16, 17] and may be transmitted to clothes and other parts of the skin by touching. PFGE revealed that MSSA isolated from the conscripts exhibited four DNA pat- terns, one of which was shared by 7 of 11 clinical isolates. Among the other four isolates, two shared an identical DNA pattern. References 1. Razak F, Anand SS, Shannon H, Vuksan V, Davis B, Jacobs R, Teo KK, McQueen M, Yusuf S. Defining obesity cut points in a multiethnic popula- tion. Circulation. 2007;115(16):2111–8. 2. Clinical and Laboratory Standards Institute. Clinical and Laboratory Stand- ards Institute [document]. In: Wayne PA, editor. Clinical and Laboratory Standards Institute; 2005. p. v. 3. Smith CL, Klco SR, Cantor CR. Pulsed-field gel electrophoresis and the technology of large DNA molecules. In: Davies KE, editor. Genome analy- sis: a practical approach. Washington DC: IRL Press; 1988. p. 41–72. 4. Kitti T, Boonyonying K, Sitthisak S. Prevalence of methicillin-resistant Staphylococcus aureus among university students in Thailand. Southeast Asian J Trop Med Public Health. 2011;42(6):1498–504. 1. Razak F, Anand SS, Shannon H, Vuksan V, Davis B, Jacobs R, Teo KK, McQueen M, Yusuf S. Defining obesity cut points in a multiethnic popula- tion. Circulation. 2007;115(16):2111–8. 2. Clinical and Laboratory Standards Institute. Clinical and Laboratory Stand- ards Institute [document]. In: Wayne PA, editor. Clinical and Laboratory Standards Institute; 2005. p. v. 3. Smith CL, Klco SR, Cantor CR. Pulsed-field gel electrophoresis and the technology of large DNA molecules. In: Davies KE, editor. Genome analy- sis: a practical approach. Washington DC: IRL Press; 1988. p. 41–72. 4. Kitti T, Boonyonying K, Sitthisak S. Prevalence of methicillin-resistant Staphylococcus aureus among university students in Thailand. Southeast Asian J Trop Med Public Health. 2011;42(6):1498–504. 1. Razak F, Anand SS, Shannon H, Vuksan V, Davis B, Jacobs R, Teo KK, McQueen M, Yusuf S. Defining obesity cut points in a multiethnic popula- tion. Circulation. 2007;115(16):2111–8. 16. Williams RE. Healthy carriage of Staphylococcus aureus: its prevalence and importance. Bacteriol Rev. 1963;27:56–71. 16. Williams RE. Healthy carriage of Staphylococcus aureus: its prevalence and importance. Bacteriol Rev. 1963;27:56–71. 17. Wertheim HF, Melles DC, Vos MC, van Leeuwen W, van Belkum A, Verbrugh HA, et al. The role of nasal carriage in Staphylococcus aureu infections. Lancet Infect Dis. 2005;5(12):751–62. 2. Clinical and Laboratory Standards Institute. Clinical and Laboratory Stand- ards Institute [document]. In: Wayne PA, editor. Clinical and Laboratory Standards Institute; 2005. p. v. 18. Elias AF, Chaussee MS, McDowell EJ, Huntington MK. Community-based intervention to manage an outbreak of MRSA skin infections in a county jail. J Correct Health Care. 2010;16(3):205–15. 4. Kitti T, Boonyonying K, Sitthisak S. Prevalence of methicillin-resistant Staphylococcus aureus among university students in Thailand. Southeast Asian J Trop Med Public Health. 2011;42(6):1498–504. Consent 5. Nickerson EK, Wuthiekanun V, Wongsuvan G, Limmathurosakul D, Srisa- mang P, Mahavanakul W, Thaipadungpanit J, Shah KR, Arayawichanont A, Amornchai P, et al. Factors predicting and reducing mortality in patients with invasive Staphylococcus aureus disease in a developing country. PLoS One. 2009;4(8):e6512. Patient consent forms were waived because this was an outbreak investigation and the study was approved by the royal Thai army medical department institutional review board. 6. Ellis MW, Griffith ME, Jorgensen JH, Hospenthal DR, Mende K, Patter- son JE. Presence and molecular epidemiology of virulence factors in methicillin-resistant Staphylococcus aureus strains colonizing and infect- ing soldiers. J Clin Microbiol. 2009;47(4):940–5. p g The authors declare that they have no competing interests. The authors declare that they have no competing interests. Received: 31 May 2015 Accepted: 11 March 2016 Received: 31 May 2015 Accepted: 11 March 2016 14. Early GJ, Seifried SE. Risk factors for community-associated Staphylococ- cus aureus skin infection in children of Maui. Hawaii J Med Public Health. 2012;71(8):218–23. 15. Hsiang MS, Shiau R, Nadle J, Chan L, Lee B, Chambers HF, et al. Epidemio- logic similarities in pediatric community-associated methicillin-resistant and methicillin-sensitive in the San Francisco Bay area. J Pediatric Infect Dis Soc. 2012;1(3):200–11. 15. Hsiang MS, Shiau R, Nadle J, Chan L, Lee B, Chambers HF, et al. Epidemio- logic similarities in pediatric community-associated methicillin-resistant and methicillin-sensitive in the San Francisco Bay area. J Pediatric Infect Dis Soc. 2012;1(3):200–11. Author details 1 1 Department of Military and Community Medicine, Phramongkutklao College of Medicine, 315, Ratchawithi Road, Ratchathewi, Bangkok 10400, Thailand. 2 Department of Outpatients, Phramongkutklao Hospital, Bangkok, Thailand. 3 Department of Parasitology, Phramongkutklao College of Medicine, 315, Ratchawithi Road, Ratchathewi, Bangkok 10400, Thailand. 4 Division of Derma- tology, Department of Internal Medicine, Phramongkutklao Hospital, Bangkok, Thailand. 5 Infection Control Unit of Phramongkutklao Hospital, Bangkok, 9. May L, Porter C, Tribble D, Armstrong A, Mostafa M, Riddle M. Self- reported incidence of skin and soft tissue infections among deployed US military. Travel Med Infect Dis. 2011;9(4):213–20. 10. Landrum ML, Neumann C, Cook C, Chukwuma U, Ellis MW, Hospenthal DR, et al. Epidemiology of Staphylococcus aureus blood and skin and soft tissue infections in the US military health system, 2005–2010. JAMA. 2012;308(1):50–9. g p g Thailand. 6 Clinical Epidemiology Unit of Phramongkutklao Hospital, Bangkok, Thailand. 11. Miller LG, Perdreau-Remington F, Bayer AS, Diep B, Tan N, Bharadwa K, et al. Clinical and epidemiologic characteristics cannot distinguish community-associated methicillin-resistant Staphylococcus aureus infection from methicillin-susceptible S. aureus infection: a prospective investigation. Clin Infect Dis. 2007;44(4):471–82. Conclusions However, different DNA patterns may be observed for the same strain, and similar DNA patterns may not necessarily indicate the same family of isolates [18]. The varying health status and behavioral activities of individuals in a community mean that some pathogens may be transmitted into the environment and passed on to others. Researchers have advocated that clones identi- fied by PFGE are indistinguishable from outbreak isolates and suggest person-to-person transmission [13]. Previous studies have identified the prevalence and risk factors of community-associated methicillin-resistant S. aureus (CA-MRSA) [8–10]. Although the risk fac- tors associated with acquiring CA-MRSA and MSSA infection are similar, the information regarding MSSA infection is limited [8, 11–13]. Inadequate hygiene and physical trauma are likely to lead to the emergence of MRSA infection among military trainees [12, 13]. The higher case rates observed for subunits 1 and 2 compared with subunits 3 and 4 in our study might be explained by the training activity involving low crawling on a concrete floor prior to the onset of abscess skin lesions among conscripts in subunits 1 and 2, while conscripts in subu- nits 3 and 4 did not undergo any such activity during the same period. Low crawl training often causes skin abra- sions on elbows and 43 % of cases had abscesses on their elbows. In summary, we identified an outbreak of MSSA infection among new military conscripts in a crowded environment during basic training. The risk factors asso- ciated with infection were identified as training exercises that involved crawling on a hard floor often resulting in skin abrasions, and the sharing of towels. Delayed diag- nosis and ineffective antibiotic treatment led to severe infection and fatality, for this normally treatable condi- tion. We recommend that a survey of Staphylococcus carriage among military camps be undertaken by the military medical agency to prevent similar outbreaks in the future. Furthermore, multiple erythematous papules are another important risk factor associated with abscess development. A recent study revealed that control of eczema can reduce the chance of S. aureus infection [14] and compromised skin integrity has been found to be associated with CA-MSSA infection [15]. A report of skin infection caused by MSSA or MRSA revealed the Panton-Valentine leukocidin (PVL) gene as a virulence factor [6]. It is possible that this outbreak of Nivesvivat et al. BMC Res Notes (2016) 9:179 Page 6 of 6 Authors’ contributions TN, DJ, MM, JI, SP, KP, PP, SK, OP, PY, WT, and RR conducted the outbreak inves- tigation. TN analyzed the data and wrote the manuscript. All authors read and approved the final manuscript. 7. Miller LG, Kaplan SL. Staphylococcus aureus: a community pathogen. Infect Dis Clin N Am. 2009;23(1):35–52. 7. Miller LG, Kaplan SL. Staphylococcus aureus: a community pathogen. Infect Dis Clin N Am. 2009;23(1):35–52. 8. Zinderman CE, Conner B, Malakooti MA, LaMar JE, Armstrong A, Bohnker BK. Community-acquired methicillin-resistant Staphylococcus aureus among military recruits. Emerg Infect Dis. 2004;10(5):941–4. 8. Zinderman CE, Conner B, Malakooti MA, LaMar JE, Armstrong A, Bohnker BK. Community-acquired methicillin-resistant Staphylococcus aureus among military recruits. Emerg Infect Dis. 2004;10(5):941–4. Acknowledgements We thank the laboratory at the Division of Microbiology, Department of Clini- cal Pathology, Phramongkutklao Hospital, and the Reference Laboratory of the National Institute of Health, Department of Medical Sciences, Ministry of Public Health, Thailand, for participating in this study. 12. Campbell KM, Vaughn AF, Russell KL, Smith B, Jimenez DL, Barrozo CP, et al. Risk factors for community-associated methicillin-resistant Staphy- lococcus aureus infections in an outbreak of disease among military train- ees in San Diego, California, in 2002. J Clin Microbiol. 2004;42(9):4050–3. 1. Razak F, Anand SS, Shannon H, Vuksan V, Davis B, Jacobs R, Teo KK, McQueen M, Yusuf S. Defining obesity cut points in a multiethnic popula- tion. Circulation. 2007;115(16):2111–8. 2. Clinical and Laboratory Standards Institute. Clinical and Laboratory Stand- ards Institute [document]. In: Wayne PA, editor. Clinical and Laboratory Standards Institute; 2005. p. v. 3. Smith CL, Klco SR, Cantor CR. Pulsed-field gel electrophoresis and the technology of large DNA molecules. In: Davies KE, editor. Genome analy- sis: a practical approach. Washington DC: IRL Press; 1988. p. 41–72. 4. Kitti T, Boonyonying K, Sitthisak S. Prevalence of methicillin-resistant Staphylococcus aureus among university students in Thailand. Southeast Asian J Trop Med Public Health. 2011;42(6):1498–504. Competing interests 13. Aiello AE, Lowy FD, Wright LN, Larson EL. Meticillin-resistant Staphylo- coccus aureus among US prisoners and military personnel: review and recommendations for future studies. Lancet Infect Dis. 2006;6(6):335–41. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit p y y p • We accept pre-submission inquiries
https://openalex.org/W2789998274	https://hal.sorbonne-universite.fr/hal-01832923/document	English	null	Superconducting magnetic bearings simulation using an <i>H</i>-formulation finite element model	Superconductor science and technology/Superconductor science & technology	2,018	cc-by	12,379	To cite this version: Loic Queval, Kun Liu, Wenjiao Yang, Victor M.R. Zermeno, Guangtong Ma. Superconducting mag- netic bearings simulation using an H -formulation finite element model. Superconductor Science and Technology, 2018, 31 (8), pp.084001. 10.1088/1361-6668/aac55d. hal-01832923 Superconducting magnetic bearings simulation using an H -formulation finite element model Loic Queval, Kun Liu, Wenjiao Yang, Victor M.R. Zermeno, Guangtong Ma Superconducting magnetic bearings simulation using an H -formulation finite element model Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-01832923 https://hal.sorbonne-universite.fr/hal-01832923v1 Submitted on 9 Jul 2018 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Superconductor Science and Technology Superconducting magnetic bearings simulation using an H-formulation finite element model View the article online for updates and enhancements. H-formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors - H-formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors F Sass, G G Sotelo, R de Andrade Junior et al. - This content was downloaded from IP address 134.157.80.157 on 09/07/2018 at 09:54 Related content Related content Experiment and simulation of superconducting magnetic levitation with REBCO coated conductor stacks Kun Liu, Wenjiao Yang, Guangtong Ma et al. - Superconducting magnetic bearings with bulks and 2G HTS stacks: comparison between simulations using H and A-V formulations with measurements F Sass, D H N Dias, G G Sotelo et al. - H-formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors F Sass, G G Sotelo, R de Andrade Junior et al. - Related content Experiment and simulation of superconducting magnetic levitation with REBCO coated conductor stacks Kun Liu, Wenjiao Yang, Guangtong Ma et al. - Superconducting magnetic bearings with bulks and 2G HTS stacks: comparison between simulations using H and A-V formulations with measurements F Sass, D H N Dias, G G Sotelo et al. - H-formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors F Sass, G G Sotelo, R de Andrade Junior et al. - This content was downloaded from IP address 134.157.80.157 on 09/07/2018 at 09:54 Superconductor Science and Technology https://doi.org/10.1088/1361-6668/aac55d Superconductor Science and Technology Supercond. Sci. Technol. 31 (2018) 084001 (14pp) Abstract The modeling of superconducting magnetic bearings (SMBs) is of great signiﬁcance for predicting and optimizing their levitation performance before construction. Although much effort has been made in this area, there still remains some space for improvements. Thus the goal of this work is to report a ﬂexible, fast and trustworthy H-formulation ﬁnite element model. First the methodology for modeling and calibrating both bulk-type and stack-type SMBs is summarized. Then its effectiveness for simulating SMBs in 2D, 2D axisymmetric and 3D is evaluated by comparison with measurements. In particular, original solutions to overcome several obstacles are given: clariﬁcation of the calibration procedure for stack-type and bulk-type SMBs, details on the experimental protocol to obtain reproducible measurements, validation of the 2D model for a stack-type SMB modeling the tapes’ real thickness, implementation of a 2D axisymmetric SMB model, implementation of a 3D SMB model, and extensive validation of the models by comparison with experimental results for ﬁeld cooling and zero ﬁeld cooling, for both vertical and lateral movements. The accuracy of the model being having proven, it now has a strong potential for speeding up the development of numerous applications including maglev vehicles, magnetic launchers, ﬂywheel energy storage systems, motor bearings and cosmic microwave background polarimeters. Keywords: superconducting magnetic bearing (SMB), maglev, H-formulation ﬁnite element model, modeling and simulation Keywords: superconducting magnetic bearing (SMB), maglev, H-formulation ﬁnite element model, modeling and simulation (Some ﬁgures may appear in colour only in the online journal) 4 Author to whom any correspondence should be addressed. Loïc Quéval1 , Guangtong Ma2 1 Group of electrical engineering-Paris (GeePs), CNRS UMR 8507, CentraleSupélec, UPSud, UPMC, Gif- sur-Yvette, France 2 Applied Superconductivity Laboratory (ASCLab), State Key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu, Sichuan 610031, People’s Republic of China 3 Karlsruhe Institute of Technology, Hermann-von-Helmholtz Platz 1, D-76344 Eggenstein-Leopoldshafen, Germany 2 Applied Superconductivity Laboratory (ASCLab), State Key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu, Sichuan 610031, People’s Republic of China 3 Karlsruhe Institute of Technology, Hermann-von-Helmholtz Platz 1, D-76344 Eggenstein-Leopoldshafen, Germany E-mail: loic.queval@geeps.centralesupelec.fr, gtma@swjtu.edu.cn and victor.zermeno@kit.edu Received 16 March 2018, revised 2 May 2018 Accepted for publication 16 May 2018 Published 21 June 2018 1. Introduction 2D 2D axi 3D A-V Homemade Hofmann et al [25] Sugiura et al [26] Ueda et al [27] Dias et al [28–30] Takeda et al [31] Ma et al [32–34] Chun et al [35] Ruiz-Alonso et al [36] Wang et al [37] Sotelo et al [38] Software — Li et al [39] Hauser [40] T-Ω Homemade Zhang et al [41] Zheng et al [42] Uesaka et al [43, 44] Gou et al [45] Tsuchimoto et al [46] Tsuda et al [47–49] Ma et al [50, 51] Pratap et al [52] Software — — — E Homemade — — — Software — — — H Homemade Lu et al [53] — Lu et al [54] Yu et al [55] Software Sass et al [56] Patel et al [57] Patel et al [57] Quéval et al [58] This work Quéval et al [58] This work This work There are various analytical and numerical models that are able to predict, more or less accurately, the maglev perfor- mances of SMBs. A detailed review is provided by Navau et al in [22]. Among them, ﬁnite element (FE) models using various formulations are being intensively developed. The formulations are named after the state variables to be solved: A-V-formulation for the magnetic potential vector and the electric potential, T-Ω- formulation for the current potential vector and the magnetic potential, E-formulation for the electric ﬁeld and H-formulation for the magnetic ﬁeld. The critical state model [23] or the E-J power law model [24] is then commonly used together with one of these formulations to model the nonlinear resistivity of the superconductor. A summary of the formulation used by inde- pendent groups to model SMBs with homemade FE codes and free/paid for FE software is proposed in table 1. analyze a SMB made of a cylindrical PM and a cylindrical HTS bulk [55]. A substantial effort was made there to experimentally validate the model for both zero ﬁeld cooling and ﬁeld cooling, but only for vertical displacements. Surprisingly, the simulated levitation force did not go back to zero when the gap increased. And the levitation force loop proved difﬁcult to reproduce for the ﬁeld cooling case. g FE software has also been employed to simulate SMBs using the H-formulation. 1. Introduction or repulsive depending on the arrangement and on the oper- ating conditions. It can even provide passive stable levitation. This unique feature motivated the development of super- conducting magnetic bearings (SMBs) [1–4]. They have been customized for numerous applications, including maglev vehicles [5–7], magnetic launchers [8, 9], ﬂywheel energy storage systems [10–17], motors [18], and cosmic microwave background polarimeters [19–21]. The relative movement between a permanent magnet (PM) and a high temperature superconductor (HTS) can induce super- currents in the HTS. By interacting with the PM static magnetic ﬁeld, these supercurrents produce a force that can be attractive © 2018 IOP Publishing Ltd Printed in the UK 0953-2048/18/084001+14$33.00 1 L Quéval et al Supercond. Sci. Technol. 31 (2018) 084001 Table 1. SMB ﬁnite element models. 2D 2D axi 3D A-V Homemade Hofmann et al [25] Sugiura et al [26] Ueda et al [27] Dias et al [28–30] Takeda et al [31] Ma et al [32–34] Chun et al [35] Ruiz-Alonso et al [36] Wang et al [37] Sotelo et al [38] Software — Li et al [39] Hauser [40] T-Ω Homemade Zhang et al [41] Zheng et al [42] Uesaka et al [43, 44] Gou et al [45] Tsuchimoto et al [46] Tsuda et al [47–49] Ma et al [50, 51] Pratap et al [52] Software — — — E Homemade — — — Software — — — H Homemade Lu et al [53] — Lu et al [54] Yu et al [55] Software Sass et al [56] Patel et al [57] Patel et al [57] Quéval et al [58] This work Quéval et al [58] This work This work Table 1. SMB ﬁnite element models. 1. Introduction Actually the groups listed in table 1 all used COMSOL Multiphysics [62], either with the magnetic ﬁeld formulation (mfh) physic available in the AC/DC module, or by manually implementing the partial differential equations (PDEs) with the PDE module. Sass et al developed a 2D model [56] to obtain the levitation force between a PM and an YBCO bulk or stacks of YBCO tapes. The ﬁeld of the PM was obtained using analytical equations. To model the movement, the ﬁeld gener- ated by the PM was applied as a time-dependent Dirichlet boundary condition on a boundary close to the HTS domain. To reduce the computing time, a symmetry axis was used, restricting the movement to vertical displacements. To model the stacks, an anisotropic homogenized model was adopted [63]. The agreement with measurements for ﬁeld cooling and zero ﬁeld cooling was good. A similar model was developed by Quéval et al [58] to include the PM assembly real geometry and the iron nonlinearity. To do so, the ﬁeld of the PM assembly was obtained using a magnetostatic FEM. Besides, the model was able to deal with any relative movement, making it possible to optimize the SMB on a realistic displacement sequence. A similar 3D model was mentioned in [58] but without details about its implementation. Patel et al introduced a 2D All these models have their own features and limitations. Focusing on the H-formulation, important efforts have been made to simulate SMBs using homemade codes. Lu et al wrote a FE code in FORTRAN to estimate the levitation force between a PM and an HTS bulk in 2D [53]. This is probably an evolution of the code reported in [54] for the 3D simulation of a cylindrical HTS bulk over a PM guideway. In those articles, the ﬁeld of the moving PM, obtained analytically, was applied as a time- dependent Dirichlet boundary condition on the outer boundary of a model including only the HTS domain and a thin air domain. But it is not clear if the self-ﬁeld of the HTS bulk was included. The model and its extensions to other PM guideways geometries, ﬁeld cooling and lateral movements [59–61] provided interesting guidelines but the authors provided no convincing experimental validation of it. Yu et al implemented a similar 3D model to 2 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 1. Modeling approach. (a) PM assembly model. 1. Introduction O is the origin of the coordinate system. The outer boundary is not shown. The time- dependent coordinates of M in the PM assembly reference frame describe the relative movement of the HTS and PM assemblies. (b) HTS assembly model. M is the origin of the coordinate system. axisymmetric H-formulation FEM in [57] to estimate the levi- tation force between a PM and stacks of YBCO tapes. The PM was modeled by a thin current domain approximating the ideal equivalent 2D axisymmetric current sheet. To model the move- ment, this thin domain was moved along the z-direction by deﬁning it with a time and space dependent current density. With this modeling strategy, the boundary conditions are ﬁxed but many elements are required to mesh the ‘moving’ PM assembly thus limiting the applicability of the model to simple geometries. To model the stacks, an isotropic homogenized model was used. The simulated levitation force, limited to the ﬁrst magnetization, was compared with measurements from 20 to 77 K in a ﬁeld cooling condition only. The agreement for a SMB with a rolled stack was fair at 20 K and reasonable at 77 K [57]. For a SMB with a stack of annuli [64], the agreement was good. Similarly, a 3D model was built to study the current pattern for the SMB with the rolled stack with limited discussion and validation [57]. axisymmetric H-formulation FEM in [57] to estimate the levi- tation force between a PM and stacks of YBCO tapes. The PM was modeled by a thin current domain approximating the ideal equivalent 2D axisymmetric current sheet. To model the move- ment, this thin domain was moved along the z-direction by deﬁning it with a time and space dependent current density. With this modeling strategy, the boundary conditions are ﬁxed but many elements are required to mesh the ‘moving’ PM assembly thus limiting the applicability of the model to simple geometries. To model the stacks an isotropic homogenized model was used Figure 1. Modeling approach. (a) PM assembly model. O is the origin of the coordinate system. The outer boundary is not shown. The time- dependent coordinates of M in the PM assembly reference frame describe the relative movement of the HTS and PM assemblies. (b) HTS assembly model. M is the origin of the coordinate system. 1. Introduction The motivation behind this work is to develop ﬂexible, fast and trustworthy H-formulation FE models able to predict the maglev performances of SMBs in 2D, 2D axisymmetric and 3D conﬁgurations. Key advancements with respect to previous models include: clariﬁcation of the calibration pro- cedure for stack-type and bulk-type SMBs, details on the experimental protocol to obtain reproducible measurements, validation of the 2D model for a stack-type SMB considering the tapes real thickness, implementation of a 2D axisymmetric SMB model, implementation of a 3D SMB model, and extensive validation of the models by comparison with experimental results for ﬁeld cooling and zero ﬁeld cooling, for both vertical and lateral movements. The test cases reported here have been selected to serve as benchmarks, with the aim to help focus the effort of the numerical modeling community towards the most relevant approaches [65]. materials are non-magnetic. To mathematically model the HTS assembly, the H-formulation is used [67, 68], r m ´ ´ = - ¶ ¶ W ( ) t H H in , 1 0 r m ´ ´ = - ¶ ¶ W ( ) t H H in , 1 0 = + G ( ) H H H on , 2 self ext m = ⋅ = ( ) ∣ ( ) ( ) t H H H 0, 3 0 0 0 0 ( ) 1 ( ) 2 ( ) 3 ( ) 3 ( ) 3 where H is the magnetic ﬁeld strength, r is the material resistivity, m0 is the vacuum magnetic permeability, W is the computational domain and G is the outer domain boundary. Neumann boundary conditions are used for inner boundaries ann boundary conditions are used for inner boundarie On G, Dirichlet boundary conditions are used to impose the self-ﬁeld Hself (the one created by the supercurrent) and the external ﬁeld Hext (the one created by the PM assembly). The current density J, the electric ﬁeld E and the magnetic ﬂux density B can be obtained from H using, 2.1. PM assembly model The PM assembly is an arrangement of any number of PMs and ferromagnetic pieces surrounded by air (or any coolant). For simple geometries, analytical formulas could be used [56, 66]. But it is modeled here using a magnetostatic A-formulation FE model. This allows us to include the iron nonlinear B–H curve and to consider complex PM assembly geometries [58]. r = - (∣∣ ) ( ) ( ) ( ) E J J J B B J B , , 7 n sc c c c 1 ( ) 7 where Jc is the ﬁeld dependent local critical current density, Ec is the critical current criterion and n is a material para- meter. To impose a transport current in a conductor, an int- egral constraint on the current density can be used ò = ⋅ W ( ) ( ) I J s t d , 8 tr c ( ) 8 2. Superconducting magnetic bearing model The SMB model is built by unidirectional coupling between the PM assembly model and the HTS assembly model. The coupling is done by applying the sum of the external ﬁeld Hext and the self-ﬁeld Hself on the outer boundaries G of the HTS assembly model (ﬁgure 1). = ´ ( ) J H, 4 r = ( ) E J, 5 m = ( ) B H. 6 0 ( ) 4 ( ) 4 ( ) 5 ( ) 6 ( ) 5 ( ) 6 ( ) 6 The resistivity rsc of the HTS is represented by a power law, The resistivity rsc of the HTS is represented by a power law, 2.2. HTS assembly model The arrows indicate the PM magnetization direction. The mesh of the air/coolant is not shown. Inset: zoom on the stack; the blue lines show the superconductor layers. The HTS is said to be ‘ﬁeld cooled’ (FC) when the cooling is achieved close to the PM assembly, and ‘zero ﬁeld cooled’ (ZFC) when the PM is far enough so that the applied ﬁeld is negligible. We assume that during the cooling all the ﬂux is pinned [69] and that no macroscopic currents are induced in the HTS [70]. This is experimentally validated by the fact that the forces after cooling but before any movements are null [29]. To simulate the FC case, we can therefore disregard Hself and set = ( ) T x y z H H , , . t 0 PM 0 By doing so, equation (3) is respected because the divergence of the ﬁeld generated by the PM is zero. Note that we implicitly make here the hypothesis that the ﬁeld generated by the super- current does not inﬂuence the PM’s remanent ﬁeld. obtained by three step motors and screw rods. The 3D posi- tion is recorded by three linear displacement sensors. The 3D force is measured by a 3D load cell. The time, the 3D position and the 3D force are recorded at 1 kS s−1. The measured data presented here corresponds to a 500 point moving average. The PM assembly is at room temperature while the HTS assembly is at liquid nitrogen temperature. A 1 mm sheet of aerogel paper CT200-Z is used to thermally insulate the PM and avoid a shift of its remanent ﬂux density with the temperature during the measurement [71]. The z-direction force recorded by the load cell includes the weight of the HTS assembly: therefore the initial force (i.e. the weight) was subtracted from the measurements to remove any force not produced by the supercurrent in the measured data presented here. The liquid nitrogen container is mounted so that its weight is not measured by the load cell. The self-ﬁeld Hself is obtained from the HTS assembly model at each time step by numerical integration of the Biot– Savart law. The consideration of Hself is required to make the problem self-consistent since the air/coolant layer around the HTS domain is slim. Indeed Hext is applied on a boundary that is close to the HTS domain. 2.2. HTS assembly model where Wc is the conductor cross section and s d is the differ- ential cross-sectional area vector. For the FE discretization, we use linear edge elements [67]. The HTS assembly is an arrangement of any number of normal and superconducting pieces (bulks or conductors) surrounded by air (or any coolant). It is assumed that the where Wc is the conductor cross section and s d is the differ- ential cross-sectional area vector. For the FE discretization, we use linear edge elements [67]. 3 Supercond. Sci. Technol. 31 (2018) 084001 Figure 3. SMB geometry and mesh for the 2D case. The point O is located at the center of the PM assembly top surface. The point M is located at the center of the stack bottom surface. The dimension of the PM and HTS assemblies in the x-direction are 240 mm and 100 mm, respectively. The arrows indicate the PM magnetization direction. The mesh of the air/coolant is not shown. Inset: zoom on the stack; the blue lines show the superconductor layers. L Quéval et al L Quéval et al Figure 2. Test rig used to measure the 3D forces of the SMBs. The PM assembly was ﬁxed to the base moving in the xy-direction. The HTS assembly was ﬁxed to the 3D load cell moving in the z-direction. Figure 2. Test rig used to measure the 3D forces of the SMBs. The PM assembly was ﬁxed to the base moving in the xy-direction. The HTS assembly was ﬁxed to the 3D load cell moving in the z-direction. The external ﬁeld Hext is obtained from the PM assembly model. The static magnetic ﬁeld generated by the PM assembly HPM needs to be modiﬁed to take the relative movement into account. This is done by = ( ) ( ) ( ) x y z t T x y z H H , , , , , , 9 t ext PM ( ) 9 where Tt is the translation operator that describes the time- dependent position of the HTS assembly in the PM assembly reference frame. Figure 3. SMB geometry and mesh for the 2D case. The point O is located at the center of the PM assembly top surface. The point M is located at the center of the stack bottom surface. The dimension of the PM and HTS assemblies in the x-direction are 240 mm and 100 mm, respectively. 2.2. HTS assembly model The force F (in N) between the PM assembly and the HTS assembly is obtained with ò = ´ W ( ) s F J B d , 10 sc ( ) 10 where Wsc is the HTS assembly cross section and s d the differential cross-sectional area. where Wsc is the HTS assembly cross section and s d the differential cross-sectional area. 4.1. Geometry The force measurements were carried out using a test rig developed at ASCLab (ﬁgure 2). The 3D relative motion is The linear SMB and the coordinate system adopted in this section are shown in ﬁgure 3. The PM assembly is made of 4 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 4. 2D model calibration: magnetic ﬂux density at 2 and 5 mm above the PM. cuboidal Nd–Fe–B PMs and iron slabs arranged in ﬂux concentration. The HTS assembly is a stack of 120 YBCO tapes (SuperPower SCS12050-AP). The HTS assembly can only move along the yz-plane ( = ( ) x t 0 M ). 4.2. Sequences In this section, we consider three displacement sequences. They are described by the successive positions of M(y z , M M) relative to O (in millimeters). The ﬁrst position of each sequence is the cooling position. The moving speed is 1 mm s−1 representing a quasistatic process. • ZFC100: (y z , M M)={(0, 100), (0, 6), (0, 100)} • FC25: (y z , M M)={(0, 25), (0, 6), (0, 25)} • FC25_LD: (y z , M M)={(0, 25), (0, 6), (6, 6), (−6, 6), (6, 6)}. • ZFC100: (y z , M M)={(0, 100), (0, 6), (0, 100)} • ZFC100: (y z , M M)={(0, 100), (0, 6), (0, 10 • FC25: (y z , M M)={(0, 25), (0, 6), (0, 25)} Figure 4. 2D model calibration: magnetic ﬂux density at 2 and 5 mm above the PM. yM • FC25_LD: (y z , M M)={(0, 25), (0, 6), (6, 6), (−6, 6), (6, 6)}. M • FC25_LD: (y z , M M)={(0, 25), (0, 6), (6, 6), (−6, 6), (6, 6)}. M • FC25_LD: (y z , M M)={(0, 25), (0, 6), (6, 6), (−6, 6), (6, 6)}. the Biot–Savart law, the Biot–Savart law, p = - ¢ ¢ ⋅ - ¢ - ¢ + - ¢ ¢ ¢ W ∬ ( ) ( ) ( ) ( ) ( ) ( ) H y z t J y z t z z y y z z y z , , 1 2 , , d d , 13 y x self, 2 2 sc p = ¢ ¢ ⋅ - ¢ - ¢ + - ¢ ¢ ¢ W ∬ ( ) ( ) ( ) ( ) ( ) ( ) H y z t J y z t y y y y z z y z , , 1 2 , , d d , 14 z x self, 2 2 sc 4.3. Modeling To mesh the super- conducting layer, we use a mapped mesh [75] with ten elements distributed symmetrically following an arithmetic sequence in the width and one element in the thickness. Such mesh proved to be a good compromise between speed and accuracy. The outer boundary of the HTS assembly model G is located at a distance of 1.5 mm from the HTS stack. This is less than the minimum levitation gap so that the coupling boundary is always inside the air gap. 4.3. Modeling Equations (1)–(10) are implemented in COMSOL Multi- physics 4.3a PDE mode application in a 2D space. More details about such implementation can be found in [72] for example. The HTS assembly is a stack of YBCO tapes: we model only the superconducting layers taking their real thickness into account. Each tape has a net current enforced to zero by means of an integral constraint. An anisotropic Kim- like model [73] is used to describe the dependence of the critical current density on the magnetic ﬁeld, ( ) 13 p + W ( ) ( ) ( ) y y z z 2 14 sc ( ) 14 where Wsc is the HTS assembly domain. This completes and corrects [56]. From equation (10) in 2D, the lateral force Fy and the levitation force Fz (in N) between the PM assembly and the HTS assembly are given by = + + a ^ ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟ ( ) ( ) // J J k B B B B 1 , 11 c c0 2 2 2 0 = + + a ^ ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟ ( ) ( ) // J J k B B B B 1 , 11 c c0 2 2 2 0 = - ¢ ¢ ⋅ ¢ ¢ ¢ ¢ ⋅ W ∬ ( ) ( ) ( ) ( ) F t J y z t B y z t y z d , , , , d d , 15 y x z sc sc = ¢ ¢ ⋅ ¢ ¢ ¢ ¢ ⋅ W ∬ ( ) ( ) ( ) ( ) F t J y z t B y z t y z d , , , , d d , 16 z x y sc sc ( ) 15 ( ) 11 ( ) 16 where Wsc is the HTS assembly domain and dsc is the dimension of the HTS assembly in the x-direction. where Wsc is the HTS assembly domain and dsc is the dimension of the HTS assembly in the x-direction. where B// and B⊥are the ﬁeld components parallel and perpendicular to the tape, respectively. J , c0 B , 0 k and α are material parameters. Equation (11) provides a reasonable description of the anisotropic behavior of HTS coated con- ductors (without artiﬁcial pinning) [74]. 4.5. Model validation Figure 6. 2D model validation: levitation force for the FC25 sequence. To validate the 2D model, we consider the FC25 and FC25_LD sequences. The force calculated with the 2D model is in good agreement with the measured force (ﬁgures 6 and 7). This validates the modeling approach adopted. Similar simulations (not reported here) were performed for the stack-type SMB of [56] giving similar agreements with the measurements, and comparing well with the anisotropic homogenized bulk SMB model [63] implemented by Sass et al. Note that, by obtaining Hext using a magnetostatic FEM instead of analytical formulas, and by considering both vertical and lateral displacements, the current model overcomes the limitations of the previous one. Besides, by modeling here the stack with the tapes real thick- ness, we open the possibility of considering complex HTS assemblies. Table 2. Parameters for the 2D case. Symbol Quantity Value Br PM remanent ﬂux density 1.12 T (side) 0.975 T (middle) Ec Critical current criterion 1 × 10−4 V m−1 n HTS parameter 31 Jc0 HTS parameter 3.225 × 1010 A m−2 B0 HTS parameter 0.0525 T k HTS parameter 0.256 α HTS parameter 0.58 rair Air resistivity 1 Ω m [76] m0 Air/HTS permeability 4π × 10−7 H m−1 Table 2. Parameters for the 2D case. Symbol Quantity Value Br PM remanent ﬂux density 1.12 T (side) 0.975 T (middle) Ec Critical current criterion 1 × 10−4 V m−1 n HTS parameter 31 Jc0 HTS parameter 3.225 × 1010 A m−2 B0 HTS parameter 0.0525 T k HTS parameter 0.256 α HTS parameter 0.58 rair Air resistivity 1 Ω m [76] m0 Air/HTS permeability 4π × 10−7 H m−1 4.4. Model calibration Note that, by obtaining Hext using a magnetostatic FEM instead of analytical formulas, and by considering both vertical and lateral displacements, the current model overcomes the limitations of the previous one. Besides, by modeling here the stack with the tapes real thick- ness, we open the possibility of considering complex HTS assemblies. 5. 2D axisymmetric case: axisymmetric SMB 5.1. Geometry The axisymmetric SMB and the coordinate system adopted in this section are shown in ﬁgure 8. The PM assembly is a Figure 5. 2D model calibration: levitation force for the ZFC100 sequence. Figure 6. 2D model validation: levitation force for the FC25 sequence. Figure 7. 2D model validation: (a) levitation force and (b) lateral force for the FC25_LD sequence. Table 2. Parameters for the 2D case. Symbol Quantity Value Br PM remanent ﬂux density 1.12 T (side) 0.975 T (middle) Ec Critical current criterion 1 × 10−4 V m−1 n HTS parameter 31 Jc0 HTS parameter 3.225 × 1010 A m−2 B0 HTS parameter 0.0525 T k HTS parameter 0.256 α HTS parameter 0.58 rair Air resistivity 1 Ω m [76] m0 Air/HTS permeability 4π × 10−7 H m−1 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 5. 2D model calibration: levitation force for the ZFC100 sequence. Figure 7. 2D model validation: (a) levitation force and (b) lateral force for the FC25_LD sequence. 5. 2D axisymmetric case: axisymmetric SMB levitation force during the ZFC100 sequence is equal to the measured value (ﬁgure 5). The procedure used here is applicable for any stack-type SMB with the advantage that only three parameters are obtained by trial-and-error. The parameters of the 2D case are summarized in table 2. 4.4. Model calibration To calibrate the PM assembly model, we need to know the B–H curve of the iron and the remanent ﬂux density Br of the PM. The assumed B–H curve is given in the appendix. To obtain the remanent ﬂux density Br of the PM, we measured the magnetic ﬂux density at several distances above the PM. By a trial-and-error process, we obtained Br that minimizes the difference between the measured data and the PM assembly model (ﬁgure 4). To calibrate the HTS assembly model, we need to get the values of ﬁve parameters: J , c0 n, B , 0 k and α. To obtain J , c0 it is a common practice to use the maximum levitation force obtained for a zero ﬁeld cooling sequence [29, 43, 51, 54]. The procedure used here is dif- ferent. Jc0 and n are obtained by ﬁtting the power law to the measured current–voltage curve of a short sample of the same conductor. The measurement was made at 77 K using the four probe method. The other HTS tape parameters B , 0 k and α are obtained by trial-and-error so that the simulated maximum From equation (9) in 2D, with the conventions of ﬁgure 2, the expression for Hext becomes ( ) 12 = + + ( ) ( ( ) ( )) ( ) y z y y z z H H , , t t , t , 12 M M ext PM where ( ) y z , M M is the time-dependent position of the HTS assembly relative to O. Hself is obtained by 2D integration of 5 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al L Quéval et al levitation force during the ZFC100 sequence is equal to the measured value (ﬁgure 5). The procedure used here is applicable for any stack-type SMB with the advantage that only three parameters are obtained by trial-and-error. The 4.5. Model validation To validate the 2D model, we consider the FC25 and FC25_LD sequences. The force calculated with the 2D model is in good agreement with the measured force (ﬁgures 6 and 7). This validates the modeling approach adopted. Similar simulations (not reported here) were performed for the stack-type SMB of [56] giving similar agreements with the measurements, and comparing well with the anisotropic homogenized bulk SMB model [63] implemented by Sass et al. 5.1. Geometry The axisymmetric SMB and the coordinate system adopted in this section are shown in ﬁgure 8. The PM assembly is a cylindrical Nd–Fe–B magnet. The HTS assembly is a 6 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al L Quéval et al Figure 8. SMB geometry and mesh for the 2D axisymmetric case. The point O is located at the center of the PM top surface. The point M is located at the center of the bulk bottom surface. The arrow indicates the PM magnetization direction. The mesh of the air/ coolant is not shown. Figure 8. SMB geometry and mesh for the 2D axisymmetric case. Th i O i l d h f h PM f Th i Figure 9. 2D axi/3D model calibration: magnetic ﬂux density at 2 and 5 mm above the PM. Figure 9. 2D axi/3D model calibration: magnetic ﬂux density at 2 and 5 mm above the PM. Figure 9. 2D axi/3D model calibration: magnetic ﬂux density at 2 and 5 mm above the PM. Figure 8. SMB geometry and mesh for the 2D axisymmetric case. The point O is located at the center of the PM top surface. The point M is located at the center of the bulk bottom surface. The arrow indicates the PM magnetization direction. The mesh of the air/ coolant is not shown. From (9) in 2D axisymmetric, with the conventions of ﬁgure 8, the expression for Hext becomes, = + ( ) ( ( )) ( ) r z r z z H H , , t , t , 18 M ext PM ( ) 18 cylindrical single domain melt-textured YBCO bulk. The HTS assembly can only move along the z-direc- tion ( = ( ) r t 0 M ). where zM is the time-dependent position of the HTS assembly relative to O. 5.3. Modeling Equations (1)–(10) are implemented in COMSOL Multi- physics 4.3a PDE mode application in a 2D axisymmetric space. More details about such implementation can be found in [77] for example. The HTS assembly is a bulk, thus an isotropic Kim-like model [73] is used to describe the dependence of the critical current density on the magnetic ﬁeld, ò a a = - p ( ) ( ) K m m d 1 sin , 22 0 2 2 ò a a = - p ( ) ( ) E m m 1 sin d . 23 0 2 2 ( ) 22 ( ) 23 From equation (10) in 2D axisymmetric, the levitation force Fz (in N) between the PM assembly and the HTS assembly is given by = + ( ) ∣∣ ( ) J J B B B 1 , 17 c c0 0 ( ) 17 = - ¢ ¢ ⋅ ¢ ¢ p ¢ ¢ ¢ q W ∬ ( ) ( ) ( ) ( ) F t J r z t B r z t r r z , , , , 2 d d , 24 z r sc ( ) 24 where Wsc is the HTS assembly domain. where Wsc is the HTS assembly domain. where Jc0 and B0 are material parameters. To mesh the HTS bulk, we use the mapped mesh shown in ﬁgure 8 with 8×8 elements distributed following arithmetic sequences in the rz-plane. The outer boundary of the HTS assembly model G is located at 2.5 mm from the HTS bulk, corresponding to half of the minimum levitation gap. 5.2. Sequences In this section, we consider three displacement sequences. They are described by the successive positions of M(r z , M M) relative to O (in millimeters). The ﬁrst position of each sequence is the cooling position. The moving speed is 1 mm s−1 representing a quasistatic process. • ZFC100: ( ) r z , M M ={(0, 100), (0, 5), (0, 100)} • FC25: (r z , M M)={(0, 25), (0, 5), (0, 100), (0, 5)} • FC5: (r z , M M)={(0, 5), (0, 100), (0, 5), (0, 100)}. • ZFC100: ( ) r z , M M ={(0, 100), (0, 5), (0, 100)} • FC25: (r z , M M)={(0, 25), (0, 5), (0, 100), (0, 5)} • FC5: (r z , M M)={(0, 5), (0, 100), (0, 5), (0, 100)}. ( ) 20 = ¢ + ¢ + - ¢ ( ) ( ) ( ) m rr r r z z 4 , 21 2 2 ( ) 21 = + ¢ + - ¢ ( ) ( ) ( ) m r r z z , 21 2 2 where Wsc is the HTS assembly domain, and K and E are the complete elliptic integrals of the ﬁrst and second kind, 5.1. Geometry Hself is obtained by 2D axisymmetric integra- tion of the Biot–Savart law, p = - ¢ ¢ ¢ - ¢ ´ - - - ¢ ¢ q W ⎡ ⎣⎢ ⎤ ⎦⎥ ∬ ( ) ( ) ( ) ( ) ( ) ( ) ( ) H r z t J r z t m r r z z K m m m E m r z , , 1 4 , , 2 2 1 d d , 19 r self, 3 sc p = ¢ ¢ ¢ ´ + + ¢ - - ¢ ¢ q W ⎡ ⎣⎢ ⎤ ⎦⎥ ∬ ( ) ( ) ( ) ( ) ( ) ( ) ( ) H r z t J r z t m r r r K m m r r r r m E m r z , , 1 4 , , 2 2 1 d d , 20 z self, 3 sc = ¢ + ¢ + - ¢ ( ) ( ) ( ) m rr r r z z 4 , 21 2 2 5.4. Model calibration To obtain the remanent ﬂux density Br of the PM cylinder, we measured the magnetic ﬂux density at several distances above the PM. By a trial-and-error process, we obtained Br that 7 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 10. 2D axi/3D model calibration: levitation force for the ZFC100 sequence. Figure 11. 2D axi/3D model validation: levitation force for the FC25 sequence. Figure 12. 2D axi/3D model validation: levitation force for the FC5sequence. Figure 13. SMB geometry and mesh for the 3D case. The point O is located at the center of the PM top surface. The point M is located at the center of the bulk bottom surface. The arrow indicates the PM p ( ) Figure 10. 2D axi/3D model calibration: levitation force for the ZFC100 sequence. Figure 12. 2D axi/3D model validation: levitation force for the FC5sequence. Figure 13. SMB geometry and mesh for the 3D case. The point O is located at the center of the PM top surface. The point M is located at the center of the bulk bottom surface. The arrow indicates the PM magnetization direction. The mesh of the air/coolant is not shown. Figure 11. 2D axi/3D model validation: levitation force for the FC25 sequence. Table 3. Parameters for the 2D axisymmetric and 3D cases. Symbol Quantity Value Br PM remanent ﬂux density 1.27 T Ec Critical current criterion 1 × 10−4 V m−1 Jc0 HTS parameter 2.4 × 108 A m−2 n HTS parameter 21 [79] B0 HTS parameter 0.37 T rair Air resistivity 1 Ω m [76] m0 Air/HTS permeability 4π × 10−7 H m−1 Table 3. Parameters for the 2D axisymmetric and 3D cases. set to commonly used values. Note that the value of n weakly affects the calculated results if higher than 15. The parameters of the 2D axisymmetric case are summarized in table 3. 5.5. Model validation To validate the 2D axisymmetric model, we consider the FC25 and FC5 sequences. The force calculated with the 2D axisymmetric model is in good agreement with the measured force (ﬁgures 11 and 12). This serves as a validation. minimizes the difference between the measured data and the PM assembly model (ﬁgure 9). To obtain J , c0 it is common practice to use the maximum levitation force obtained for a zero ﬁeld cooling sequence [29, 36, 51, 54]. Accordingly, the critical current density Jc0 is set here at 2.4 × 108 A m−2, so that the simulated maximum levitation force during the ZFC100 sequence is equal to the measured value (ﬁgure 10). Alternatively, Jc0 could have been determined beforehand as done for the 2D case, for example by measuring it by VSM (vibrating sample magnetometer) for a small piece from the bulk as reported in [78]. The other HTS bulk parameters are 6.3. Modeling Equations (1)–(10) are implemented in COMSOL Multi- physics 4.3a PDE mode application in a 3D space. More details about such implementation can be found in [80] for example. To mesh the HTS bulk, we swept the mesh shown in ﬁgure 8 following a 360° circular path to obtain the hex- ahedral mesh shown in ﬁgure 13. The outer boundary of the HTS assembly model G is here again located at 2.5 mm from the HTS bulk, corresponding to half the minimum levita- tion gap. Figure 14. 3D model validation: (a) levitation force and (b) lateral force for the ZFC100, ZFC100_Y7.5 and ZFC100_Y15 sequences. p = ´ - ¢ - - ¢ - ¢ + - ¢ + - ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) ( ) H x y z t J y y J x x x x y y z z x y z , , , 1 4 d d d , 28 z x y self, 2 2 2 3 sc ( ) 28 From equation (9) in 3D, with the conventions of ﬁgure 13, the expression for Hext becomes where Wsc is the HTS assembly domain. From equation (10) in 3D, the forces Fx, Fy and Fz (in N) between the PM assembly and the HTS assembly are given by = + + + ( ) ( ( ) ( ) ( )) y z t x x t y y t z z t H H , , , , , M M M ext PM = + + + ( ) ( ( ) ( ) ( )) ( ) y z t x x t y y t z z t H H , , , , , 25 M M M ext PM ( ) 25 where ( ) x y z , , M M M is the time-dependent position of the HTS assembly relative to O. 6.2. Sequences In this section, we consider six displacement sequences. They are described by the successive positions of M(x y z , , M M M) relative to O (in millimeters). The ﬁrst position of each sequence is the cooling position. The moving speed is 1 mm s−1 representing a quasistatic process. • ZFC100: (x y z , , M M M)={(0, 0, 100), (0, 0, 5), (0, 0, 100)} • FC25: (x y z , , M M M)={(0, 0, 25), (0, 0, 5), (0, 0, 100), (0, 0, 5)} • FC5: (x y z , , M M M)={(0, 0, 5), (0, 0, 100), (0, 0, 5), (0, 0, 100)} • ZFC100_Y7.5: (x y z , , M M M)={(0, 7.5, 100), (0, 7.5, 5), (0, 7.5, 100)} • ZFC100_Y15: (x y z , , M M M)={(0, 15, 100), (0, 15, 5), (0, 15, 100)} • FC25_LD: ( ) x y z , , M M M ={(0, 0, 25), (0, 0, 5), (0, 7.5, 5), (0, −7.5, 5), (0, 7.5, 5), (0, −7.5, 5), (0, 7.5, 5), (0, 0, 5)}. The ZFC100, FC25 and FC5 sequences are similar to the 2D axisymmetric case. The ZFC100_Y7.5 and ZFC100_Y15 sequences are similar to the ZFC100 sequences but the HTS bulk is off-axis. 6.1. Geometry The 3D SMB and the coordinate system adopted in this section are shown in ﬁgure 13. The SMB is the same as that for the 2D axisymmetric case but the HTS assembly can now move along any direction. 8 Supercond. Sci. Technol. 31 (2018) 084001 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 14. 3D model validation: (a) levitation force and (b) lateral force for the ZFC100, ZFC100_Y7.5 and ZFC100_Y15 sequences. Figure 14. 3D model validation: (a) levitation force and (b) lateral force for the ZFC100, ZFC100_Y7.5 and ZFC100_Y15 sequences. 6.2. Sequences 7. Discussion The test cases considered above have been selected carefully to serve as benchmarks. For the 2D case, we selected a stack-type SMB for its true 2D nature. Indeed bulk-type SMB suffer from several factors that make them difﬁcult to be simulated accurately in 2D. In particular, large bulks with homogeneous properties are difﬁcult to obtain. The end effects and the impact of intragrain currents should then be taken into account [58, 81]. For the 2D axisymmetric and 3D cases, we selected a simplistic bulk-type SMB that allows comparison of the results for axial displacement sequences. Finally, we considered on purpose repetitive dis- placements. This is because simpliﬁed models, such as Meissner- limit and frozen-ﬁeld models, can often estimate the ﬁrst section of the force loop but generally fail to predict the rest [22, 57]. For the FE discretization, we use linear edge elements [67]. The degrees of freedom of the edge elements being associated with the tangential components along the edges of the elements, it is only possible to impose the tangential component of the ﬁeld. Nevertheless, in practice a thin layer of air/coolant is sufﬁcient to obtain accurate estimation of the maglev performances as demonstrated in this work. Figure 15. 3D model validation: (a) levitation force and (b) lateral force for the FC25_LD sequence. 6.3. Modeling Sci. Technol. 31 (2018) 084001 L Quéval et al Figure 15. 3D model validation: (a) levitation force and (b) lateral force for the FC25_LD sequence. the small amplitude of the lateral fo model. 7. Discussion The test cases considered above have serve as benchmarks. For the 2D cas SMB for its true 2D nature. Indeed b several factors that make them difﬁcul in 2D. In particular, large bulks with h difﬁcult to obtain. The end effects an currents should then be taken into ac axisymmetric and 3D cases, we sele SMB that allows comparison of the re sequences. Finally, we considered o placements. This is because simpliﬁed limit and frozen-ﬁeld models, can ofte of the force loop but generally fail to For the FE discretization, we [67]. The degrees of freedom of associated with the tangential comp the elements, it is only possible Figure 15. 3D model validation: (a) levitation force and (b) lateral force for the FC25_LD sequence. Table 4. Degree of freedom, time st DOF Time ste 2D ZFC100 20270 27213 2D axi ZFC100 539 25 3D ZFC100 21068 61 Table 4. Degree of freedom, time steps and computing time. DOF Time steps Computing time (s) 2D ZFC100 20270 272139 392529 2D axi ZFC100 539 252 13038 3D ZFC100 21068 610 27777 Table 4. Degree of freedom, time steps and computing time. the small amplitude of the lateral force. This validates the 3D model. the small amplitude of the lateral force. This validates the 3D model. 6.3. Modeling Hself is obtained by 3D inte- gration of the Biot–Savart law, = ¢ ¢ ¢ ⋅ ¢ ¢ ¢ - ¢ ¢ ¢ ⋅ ¢ ¢ ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) F t J x y z t B x y z t J x y z t B x y z t x y z , , , , , , , , , , , , d d d , 29 x y z z y sc ( ) 29 p = ´ - ¢ - - ¢ - ¢ + - ¢ + - ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) ( ) H x y z t J z z J y y x x y y z z x y z , , , 1 4 d d d , 26 x y z self, 2 2 2 3 sc p = ´ - ¢ - - ¢ - ¢ + - ¢ + - ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) ( ) H x y z t J x x J z z x x y y z z x y z , , , 1 4 d d d , 27 y z x self, 2 2 2 3 sc = ¢ ¢ ¢ ⋅ ¢ ¢ ¢ - ¢ ¢ ¢ ⋅ ¢ ¢ ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) F t J x y z t B x y z t J x y z t B x y z t x y z , , , , , , , , , , , , d d d , 30 y z x x z sc ( ) 30 ( ) 26 = ¢ ¢ ¢ ⋅ ¢ ¢ ¢ - ¢ ¢ ¢ ⋅ ¢ ¢ ¢ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) ( ) ( ) ( ) F t J x y z t B x y z t J x y z t B x y z t x y z , , , , , , , , , , , , d d d , 31 z x y y x sc = ¢ ¢ ¢ ⋅ ¢ ¢ ¢ W ∭ ( ) ( ) ( ) F t J x y z t B x y z t , , , , , , z x y sc 9 Supercond. 6.4. Model calibration We use the same parameters as that for the 2D axisymmetric case (table 3). Melt-textured YBCO bulks have an anisotropic critical current density: it is larger in the ab-plane than along the c-axis [82]. This is the reason why most of previous 3D SMB models used an anisotropic bulk model. This was either achieved by stacking multiple 2D layers [27, 42–44, 46–48], by superimposing two virtual HTS bulks [54] or by con- sidering a tensor of resistivity [50]. As it is still not clear how to model HTS in 3D to include experimental phenomena such as ﬂux cutting, ﬂux ﬂow and magnetically anisotropic critical current densities [65, 80, 83], we adopted here a simplistic isotropic bulk model. This probably explains the difference between simulation and measurements for the 3D sequences ZFC100_Y7.5, ZFC100_Y15 and FC25_LD. Indeed, for these sequences the bulk is off-axis and a current is induced along the c-axis. Nevertheless, the present results show that maglev performance of a bulk-type SMB can be reasonably well predicted using a 3D isotropic bulk model. 2D case: iron B–H curve (B, H)={(0.0, 0.0), (0.5, 90.0), (1.0, 270.0), (1.1, 318.25), (1.2, 384.50), (1.3, 479.50), (1.3875, 608.562), (1.45, 755.437), (1.5, 939.185), (1.545, 1188.93), (1.575, 1407.93), (1.6275, 2077.31), (1.673 75, 3117.93), (1.702 25, 3969.37), (1.7275, 4843.66), (1.758 25, 6081.34), (1.808 75, 8581.09), (1.85, 11 066.4), (1.9025, 14 985.7), (2.05, 33 003.3), (2.15, 59 203.3), (2.226 25, 93 214.9), (2.27, 118 884.0), (2.333 75, 163 558.0), (2.4075, 220 788.0), (2.6, 373 973.0), (3.0, 692 281.0)}. B in T, H in A m−1. 6.5. Model validation The 3D model should be able to reproduce the results obtained with the 2D axisymmetric model for the ZFC100, FC25 and FC5 sequences. The levitation force calculated with the 3D model has been added to ﬁgures 10–12, showing similar results. To further validate the 3D model, we consider the ZFC100_Y7.5 and ZFC100_Y15 sequences. The levita- tion and lateral forces calculated with the 3D model are in fair agreement with the measured force (ﬁgure 14). The calculated forces are somewhat smaller than the measured ones, but globally the force reduction as a function of the off-axis position is predicted correctly. Similar results have been obtained for a ﬁeld cooling height of 5 mm (not reported here). Finally, we consider the FC25_LD sequence. The levitation and lateral forces calculated with the 3D model are plotted together with the measured data in ﬁgure 15. Note the instable behavior of the bearing: when the lateral position increases, the lateral force increases too. Here again, the agreement is fair considering the length of the sequence and To simulate (zero) ﬁeld cooling, we applied an initial ﬁeld H0 according to equation (3). But because of inherent num- erical approximations (mesh, linear elements, etc), the curl of this ﬁeld is not perfectly null. In 2D, this would be equivalent to apply a set of Dirichlet’s boundary conditions that does not satisfy the Ampère’s circuital law on the outer boundary G [76]. 10 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al Table 5. Relative error at the maximum force. Measured (N) Model (N) Relative error (%) 2D FC25 110.7 114.6 3.5 2D axi FC25 41.1 42.8 4.1 3D FC25 41.1 42.4 3.2 Table 5. Relative error at the maximum force. the relative movement is the input of the simulation but in reality it is a consequence of the efforts exerted on the bearing [86]. Finally, further vetting and reﬁning of the models could help developing and improving lumped parameter SMB models [87, 88], as a mean of drastically speeding up simulations. 8. Conclusion We reported here our experience on simulating SMBs with a commercial ﬁnite element software using the H-formulation in 2D, 2D axisymmetric and 3D. The main difﬁculty is linked to the task of modeling a moving magnet. To address this problem, we chose the approach consisting in modeling the movement via time-dependent Dirichlet boundary conditions. It requires (a) only one static solution of the PM assembly ﬁnite element model, and (b) a reduced air/coolant domain around the superconducting material in the HTS assembly model. With a proper calibration procedure, we showed that the proposed model can predict accurately the observed behavior of both stack-tape and bulk-type bearings, for var- ious cooling conditions and various displacement sequences. This comprehensive validation is a necessary step before using such models for designing and optimizing realistic bearings. Besides, the test cases have been selected so that they could be used as a benchmark for other models. ORCID iDs Loïc Quéval https://orcid.org/0000-0003-3934-4372 Guangtong Ma https://orcid.org/0000-0002-0249-4310 Acknowledgments As a result, some unphysical induced current might ﬂow in the superconducting domain following equation (4) during (zero) ﬁeld cooling. A ﬁne mesh was selected here to limit this effect. This work was supported in part by the National Natural Science Foundation of China under Grants 51475389, and 51722706, in part by the Fundamental Research Funds for the Central Universities under Grant 2682016ZY05, in part by the Sichuan Youth Science & Technology Foundation under Grant 2016JQ0003, and in part by the State Key Laboratory of Traction Power under Grant TPL1712. The computing time for each model depends on many factors such as: mesh quality, number of time steps, HTS parameters and displacement sequence. All the calculations were performed using COMSOL Multiphysics 4.3a [62] and a standard desktop computer (Intel i7-4770 s, 3.10 GHz, 8 GB RAM). The state variables were scaled to 107, and the relative and absolute tolerances were set to 10−2 and 10−3, respectively. Table 4 gives a summary of the computational effort for the ZFC sequences. It can seem prohibitive for some applications, in particular when considering complex 3D SMB geometries. But we used here a rather ﬁne mesh with the goal to obtain good agreements with measurements. As a result, the relative error at the maximum levitation force stayed below 5% for the FC25 sequences (table 5). Actually coarser meshes can often help to decrease the computing time to few seconds for 2D cases, without losing too much information [56, 58]. Appendix 2D case: iron B–H curve References [1] Moon F C and Chang P Z 1990 High-speed rotation of magnets on high Tc superconducting bearings Appl. Phys. Lett. 56 397–9 [2] Weinberger B R, Lynds L, Hull J R and Balachandran U 1991 Low friction in high temperature superconductor bearings Appl. Phys. Lett. 59 1132–4 [3] Hull J R and Cansiz A 1999 Vertical and lateral forces between a permanent magnet and a high-temperature superconductor J. Appl. Phys. 86 6396 [4] Werfel F N, Floegel-Delor U, Rothfeld R, Riedel T, Goebel B, Wippich D and Schirrmeister P 2012 Superconductor bearings, ﬂywheels and transportation Supercond. Sci. Technol. 25 014007 Future efforts could be dedicated to reducing the computing time of such models. For stack-type bearings, the anisotropic homogenization proposed in [63] and extended in [84] is a good alternative. But it should be used with caution, and the ﬁrst validations proposed in [56, 85] should be extended to other geometries and other test conditions. Another necessary step is the coupling of such models with motion equations, in order to predict the dynamic behavior of the loaded bearing. Indeed, here [5] Wang J S et al 2002 The ﬁrst man-loading high temperature superconducting maglev test vehicle in the world Physica C 378–381 809–14 [6] Schultz L, de Haas O, Verges P, Beyer C, Rohlig S, Olsen H, Kuhn L, Berger D, Noteboom U and Funk U 2005 Superconductively levitated transport system—the 11 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al material with a ﬁnite-element method Supercond. Sci. Technol. 14 34–40 SupraTrans project IEEE Trans. Appl. Supercond. 15 2301–5 [7] Mattos L S, Rodriguez E, Costa F, Sotelo G G, de Andrade R and Stephan R M 2016 MagLev-Cobra operational tests IEEE Trans. Appl. Supercond. 26 3600704 [26] Sugihara T, Hashizume H and Miya K 1991 Numerical electromagnetic ﬁeld analysis of type-II superconductors Int. J. Appl. Electromagn. Mater. 2 183–96 [8] Yang W J, Wen Z, Duan Y, Chen X D, Qiu M, Liu Y and Lin L Z 2006 Construction and performance of HTS maglev launch assist test vehicle IEEE Trans. Appl. Supercond. 16 1108–11 [27] Ueda H, Azumaya S, Tsuchiya S and Ishiyama A 2006 3D electro-magnetic analysis of levitating transporter using bulk superconductor IEEE Trans. Appl. Supercond. References 16 1092–5 [28] Dias D H N, Motta E S, Sotelo G G, de Andrade R Jr, Stephan R M, Kuehn L, de Haas O and Schultz L 2009 Simulations and tests of superconducting linear bearings for a maglev prototype IEEE Trans. Appl. Supercond. 19 2120–3 [9] Wang J S, Wang S, Deng C, Zheng J, Song H, He Q and Zeng Y 2007 Laboratory-scale high temperature superconducting maglev launch system IEEE Trans. Appl. Supercond. 17 2091–4 [10] Bornemann H, Ritter T, Urban C, Paitsev O, Peber K and Rietschel H 1994 Low friction in a ﬂywheel system with passive superconducting magnetic bearings IEEE Trans. Appl. Supercond. 2 439–47 [29] Dias D H N, Motta E S, Sotelo G G and de Andrade R Jr 2010 Experimental validation of ﬁeld cooling simulations for linear superconducting magnetic bearings Supercond. Sci. Technol. 23 075013 [11] Chen Q Y, Xia Z, Ma K B, McMichael C K, Lamb M, Coolep R S, Fopler P C and Chu W K 1994 Hybrid high Tc superconducting magnetic bearings for ﬂywheel energy storage system IEEE Trans. Appl. Supercond. 2 457–64 [30] Dias D H N, Sotelo G G and de Andrade R Jr 2011 Study of the lateral force behavior in a ﬁeld cooled superconducting linear bearing IEEE Trans. Appl. Supercond. 21 1533–37 g pp p [31] Takeda N, Uesaka M and Miya K 1994 Computation and experiments on the static and dynamic characteristics of high Tc superconducting levitation Cryogenics 34 745–52 g y [12] Miyagawa Y, Kameno H, Takahata R and Ueyama H 1999 A 0.5 kWh ﬂywheel energy storage system using a high-Tc superconducting magnetic bearing IEEE Trans. Appl. Supercond. 9 996–9 p g y g [32] Ma G T 2013 Considerations on the ﬁnite-element simulation of high-temperature superconductors for magnetic levitation purposes IEEE Trans. Appl. Supercond. 23 3601609 [13] Coombs T, Campbell A M, Storey R and Weller R 1999 Superconducting magnetic bearings for energy storage ﬂywheels IEEE Trans. Appl. Supercond. 9 968–71 [33] Ma G T, Liu H, Li X T, Zhang H and Xu Y Y 2014 Numerical simulations of the mutual effect among the superconducting constituents in a levitation system with translational symmetry J. Appl. Phys. 115 083908 [14] Ichihara T et al 2005 Application of superconducting magnetic bearings to a 10 kWh-class ﬂywheel energy storage system IEEE Trans. Appl. Supercond. References 15 2245–8 y y [34] Ye C Q, Ma G T and Wang J S 2016 Calculation and optimization of high-temperature superconducting levitation by a vector potential method IEEE Trans. Appl. Supercond. 26 3603309 [15] Werfel F N, Floegel-Delor U, Riedel T, Rothfeld R, Wippich D, Goebel B, Reiner G and Wehlau N 2007 A compact HTS 5 kWh/250 kW ﬂywheel energy storage system IEEE Trans. Appl. Supercond. 17 2138–41 [35] Chun Y D, Kim Y H, Lee J, Hong J P and Lee J W 2001 Finite element analysis of magnetic ﬁeld in high temperature bulk superconductor IEEE Trans. Appl. Supercond. 11 2000–3 [16] Strasik M et al 2007 Design, fabrication, and test of a 5 kWh/ 100 kW ﬂywheel energy storage utilizing a high-temperature superconducting bearing IEEE Trans. Appl. Supercond. 17 2133–7 [17] Mukoyama S et al 2017 Development of superconducting magnetic bearing for 300 kW ﬂywheel energy storage system IEEE Trans. Appl. Supercond. 27 3600804 [36] Ruiz-Alonso D, Coombs T A and Campbell A M 2004 Numerical analysis of high-temperature superconductors with the critical-state model IEEE Trans. Appl. Supercond. 14 2053–63 [18] Nagaya K, Kosugi Y, Suzuki T and Murakami I 1999 Pulse motor with high-temperature superconducting levitation IEEE Trans. Appl. Supercond. 9 4688–94 [37] Wang L, Wang H H and Wang Q L 2006 Finite element analysis of magnetic levitation force in superconducting magnetic levitation system Cryog. Supercond. 34 190–3 (in Chinese) pp p [19] Hull J R, Hanany S, Matsumura T, Johnson B and Jones T 2005 Characterization of a high-temperature superconducting bearing for use in a cosmic microwave background polarimeter Supercond. Sci. Technol. 18 S1–S5 [38] Sotelo G G, de Andrade R Jr and Ferreira A C 2009 Test and simulation of superconducting magnetic bearings IEEE Trans. Appl. Supercond. 19 1681–6 [20] Matsumura T, Kataza H, Utsunomiya S, Yamamoto R, Hazumi M and Katayama N 2016 Design and performance of a prototype polarization modulator rotational system for use in space using a superconducting magnetic bearing IEEE Trans. Appl. Supercond. 26 3602304 [39] Li Y L, Fang J, Guo M Z, Xiao L, Zheng M H and Jiao Y L 2008 ANSYS-based analysis of levitation force in the HTS hybrid magnetic bearings Cryog. Supercond. 36 40–4 (in Chinese) [40] Hauser A O 1997 Calculation of superconducting magnetic bearings using a commercial FE-program (ANSYS) IEEE Trans. Magn. References 33 1572–5 [21] EBEX Collaboration 2017 The EBEX balloon borne experiment—optics, receiver, and polarimetry Astrophys. J. Suppl. submitted (arXiv:1703.03847) [41] Zhang J, Zeng Y, Cheng J and Tang X 2008 Optimization of permanent magnet guideway for HTS maglev vehicle with numerical methods IEEE Trans. Appl. Supercond. 18 1681–6 [22] Navau C, Del-Valle N and Sanchez A 2013 Macroscopic modeling of magnetization and levitation of hard type-II superconductors: the critical-state model IEEE Trans. Appl. Supercond. 23 8201023 [42] Zheng X J and Yang Y 2007 Transition cooling height of high- temperature superconductor levitation system IEEE Trans. Appl. Supercond. 17 3862–6 [23] Bean C P 1962 Magnetization of hard superconductors Phys. Rev. Lett. 8 250–3 [24] Rhyner J 1993 Magnetic properties and AC-losses of superconductors with power law current–voltage characteristics Physica C 202 292–300 [43] Uesaka M, Yoshida Y, Takeda N and Miya K 1993 Experimental and numerical analysis of three-dimensional high-Tc superconducting levitation systems Int. J. Appl. Electromagn. Mater. 4 13–25 [25] Hofmann C and Ries G 2001 Modelling the interactions between magnets and granular high-Tc superconductor [25] Hofmann C and Ries G 2001 Modelling the interactions between magnets and granular high-Tc superconductor 12 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al [60] Lu Y Y and Zhuang S J 2012 Magnetic forces simulation of bulk HTS over permanent magnetic railway with numerical method J. Low Temp. Phys. 169 111–21 [44] Yoshida Y, Uesaka M and Miya K 1994 Magnetic ﬁeld and force analysis of high Tc superconductor with ﬂux ﬂow and creep IEEE Trans. Magn. 30 3503–6 [45] Gou X F, Zheng X J and Zhou Y H 2007 Drift of levitated/ suspended body in high-Tc superconducting levitation systems under vibration: I. A criterion based on magnetic force-gap relation for gap varying with time IEEE Trans. Appl. Supercond. 17 3795–802 [61] Lu Y Y and Dang Q H 2012 Magnetic forces investigation of bulk HTS over permanent magnetic guideway under different lateral offset with 3D-model numerical method Adv. Mater. Sci. Eng. 2012 640497 [62] COMSOL Multiphysics version 4.3a www.comsol.com [46] Tsuchimoto M and Honma T 1994 Numerical evaluation of levitation force of HTSC ﬂywheel IEEE Trans. Magn. 4 211–5 [63] Rodriguez-Zermeño V M, Abrahamsen A B, Mijatovic N, Jensen B B and Sørensen M P 2013 Calculation of alternating current losses in stacks and coils made of second generation high temperature superconducting tapes for large scale applications J. References Appl. Phys. 114 173901 [47] Tsuda M, Lee H and Iwasa Y 1998 Electromaglev (active- maglev)-magnetic levitation of a superconducting disk with a DC ﬁeld generated by electromagnets: III. Theoretical results on levitation height and stability Cryogenics 38 743–56 [64] Patel A, Hahn S, Voccio J, Baskys A, Hopkins S C and Glowacki B A 2017 Magnetic levitation using a stack of high temperature superconducting tape annuli Supercond. Sci. Technol. 30 024007 [48] Tsuda M, Lee H, Noguchi S and Iwasa Y 1998 Electromaglev (active-maglev)-magnetic levitation of a superconducting disk with a DC ﬁeld generated by electromagnets: IV. Theoretical and experimental results on supercurrent distributions in ﬁeld-cooled YBCO disks Cryogenics 39 893–903 [65] Sirois F and Grilli F 2015 Potential and limits of numerical modelling for supporting the development of HTS devices Supercond. Sci. Technol. 28 043002 [66] Ma G T, Liu H F, Wang J S, Wang S Y and Li X C 2009 3D modeling permanent magnet guideway for high temperature superconducting maglev vehicle application J. Supercond. Novel Magn. 22 841–7 [49] Ueda H and Ishiyama A 2004 Dynamic characteristics and ﬁnite element analysis of a magnetic levitation system using a YBCO bulk superconductor Supercond. Sci. Technol. 17 S170–5 [67] Brambilla R, Grilli F and Martini L 2007 Development of an edge- element model for AC loss computation of high-temperature superconductors Supercond. Sci. Technol. 20 16–24 [50] Ma G T, Wang J S and Wang S Y 2010 3D modeling of high-Tc superconductor for magnetic levitation/suspension application: I. Introduction to the method IEEE Trans. Appl. Supercond. 20 2219–27 [68] Rodriguez-Zermeño V M, Grilli F and Sirois F 2013 A full 3D time-dependent electromagnetic model for Roebel cables Supercond. Sci. Technol. 26 052001 [69] Huang H, Zheng J, Zheng B T, Qian N, Li H T, Li J and Deng Z G 2017 Correlations between magnetic ﬂux and levitation force of HTS bulk above a permanent magnet guideway J. Low Temp. Phys. 189 42–52 [51] Ma G T, Wang J S and Wang S Y 2010 3D modeling of high-Tc superconductor for magnetic levitation/suspension application: II. Validation with experiment IEEE Trans. Appl. Supercond. 20 2228–34 [70] Navau C, Sanchez A, Pardo E and Chen D-X 2004 Equilibrium positions due to different cooling processes in superconducting levitation systems Supercond. Sci. Technol. References 17 828–32 [52] Pratap S and Hearn C S 2015 3D transient modeling of bulk high-temperature superconducting material in passive magnetic bearing applications IEEE Trans. Appl. Supercond. 25 5203910 [71] Lu Y Y, Qin Y J, Dang Q H and Wang J S 2010 Inﬂuence of experimental methods on crossing in magnetic force–gap hysteresis curve of HTS maglev system Physica C 470 1994–7 [53] Lu Y and Qin Y 2015 Inﬂuence of critical current density on magnetic force of HTSC bulk above PMR with 3D- modeling numerical solutions Int. J. Mod. Phys. B 29 1542038 [72] Hong Z, Campbell A M and Coombs T A 2006 Numerical solution of critical state in superconductivity by ﬁnite element software Supercond. Sci. Technol. 19 1246–52 [54] Lu Y Y, Wang J S, Wang S Y and Zheng J 2008 3D-modeling numerical solutions of electromagnetic behavior of HTSC bulk above permanent magnetic guideway J. Supercond. Nov. Magn. 21 467–72 [73] Kim Y B, Hempstead C F and Strnad A R 1962 Critical persistent currents in hard superconductors Phys. Rev. Lett. 9 306 [55] Yu Z Q, Zhang G M, Qiu Q Q and Hu L 2015 Numerical simulation of levitation characteristics of a cylindrical permanent magnet and a high-temperature superconductor based on the 3D ﬁnite-element method Trans. China Electrotech. Soc. 30 32–8 (in Chinese) [74] Grilli F, Sirois F, Rodriguez-Zermeño V M and Vojenčiak M 2014 Self-consistent modeling of the Ic of HTS devices: how accurate do models really need to be? IEEE Trans. Appl. Supercond. 24 8000508 p [75] Rodriguez-Zermeño V M, Mijatovic N, Traeholt C, Zirngibl T, Seiler E, Abrahamsen A B, Pedersen N F and Sorensen M P 2011 Towards faster FEM simulation of thin ﬁlm superconductors: a multiscale approach IEEE Trans. Appl. Supercond. 21 3273–6 [56] Sass F, Sotelo G G, de Andrade R Jr and Sirois F 2015 H- formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors Supercond. Sci. Technol. 28 125012 [57] Patel A, Hopkins S C, Baskys A, Kalitka V, Molodyk A and Glowacki B A 2015 Magnetic levitation using high temperature superconducting pancake coils as composite bulk cylinders Supercond. Sci. Technol. 28 115007 [76] Lahtinen V, Lyly M, Stenvall A and Tarhasaari T 2012 Comparison of three eddy current formulations for superconductor hysteresis loss modelling Supercond. Sci. Technol. References 25 115001 [77] Zhang M, Kvitkovic J, Pamidi S V and Coombs T A 2012 Experimental and numerical study of a YBCO pancake coil with a magnetic substrate Supercond. Sci. Technol. 25 125020 [58] Quéval L, Sotelo G G, Kharmiz Y, Dias D H N, Sass F, Zermeño V M R and Gottkehaskamp R 2016 Optimization of the superconducting linear magnetic bearing of a maglev vehicle IEEE Trans. Appl. Supercond. 26 3601905 [78] Sawamura M and Tsuchimoto M 2000 Numerical analysis for superconductor in sheet and bulk form Japan J. Ind. Appl. Math. 17 199–208 [59] Lu Y Y, Lu B J and Wang S Y 2011 The relationship of magnetic stiffness between single and multiple YBCO superconductors over permanent magnet guideway J. Low Temp. Phys. 164 279–86 [79] Ainslie M D and Fujishiro H 2015 Modelling of bulk super- conductor magnetization Supercond. Sci. Technol. 28 053002 13 Supercond. Sci. Technol. 31 (2018) 084001 L Quéval et al applications of HTS coated conductors Supercond. Sci. Technol. 29 024007 [80] Zhang M and Coombs T A 2012 3D modeling of high-Tc superconductors by ﬁnite element software Supercond. Sci. Technol. 25 015009 applications of HTS coated conductors Supercond. Sci. Technol. 29 024007 [85] Liu K, Yang W, Ma G T, Quéval L, Gong T, Ye C, Li X and Luo Z 2017 Experiment and simulation of superconducting magnetic levitation with REBCO coated conductor stacks Supercond. Sci. Technol. 31 015013 [81] Deng Z et al 2012 Trapped ﬂux and levitation properties of multiseeded YBCO bulks for HTS magnetic device applications: I. Grain and current features IEEE Trans. Appl. Supercond. 22 6800110 [86] Dias D H N, Sotelo G G, Rodriguez E F, de Andrade R Jr and Stephan R M 2013 Emulation of a full scale maglev vehicle behavior under operational conditions IEEE Trans. Appl. Supercond. 23 3601105 [82] Murakami M, Oyama T, Fujimoto H, Gotoh S, Yamaguchi K, Shiohara Y, Koshizuaka N and Tanaka S 1991 Melt processing of bulk high Tc superconductors and their application IEEE Trans. Magn. 27 1479–86 [87] Hearn C S, Pratap S B, Chen D and Longoria R G 2014 Lumped-parameter model to describe dynamic translational interaction for high-temperature superconducting bearings IEEE Trans. Appl. Supercond. 24 46–53 [83] Badía-Majós A and López C 2012 Electromagnetics close beyond the critical state: thermodynamic prospect Supercond. Sci. Technol. References 25 104004 [88] Hearn C S, Pratap S B, Chen D and Longoria R G 2016 Dynamic performance of lumped parameter model for superconducting levitation IEEE Trans. Appl. Supercond. 26 3602608 [84] Quéval L, Rodriguez-Zermeño V M and Grilli F 2016 Numerical models for AC loss calculation in large-scale 14 14
https://openalex.org/W3009293129	http://oar.icrisat.org/11771/1/The-effects-of-gypsum-on-pod-yield-and-pre-harvest-aflatoxin-contamination-in-selected-peanut-cultivars-of-Zambia.pdf	English	null	The effects of gypsum on pod-yield and pre-harvest aflatoxin contamination in selected peanut cultivars of Zambia	African journal of plant science	2,020	cc-by	3,208	Full Length Research Paper The effects of gypsum on pod-yield and pre-harvest aflatoxin contamination in selected peanut cultivars of Zambia Full Length Research Paper 1Department of Soil Science, University of Zambia, P. O. Box 32379, Lusaka, Zambia. 2Chitedze Agricultural Research Station, International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), P. O. Box 1096, Lilongwe, Malawi. 3Department of Entomology and Plant Pathology, North Carolina State University, Box 7613, Raleigh, North Carolina, USA. Good agricultural practices are an effective means of minimizing pre-harvest aflatoxin contamination in peanuts. A field experiment was conducted to evaluate the effect of gypsum on pod yield and aflatoxin contamination in three peanut cultivars (Kadononga, MGV 4 and MGV 5) in Zambia. The experiment was conducted in Chongwe and Lusaka districts. Gypsum (15.6 % calcium) was applied at rates of 0 and 400 kg/ha at flowering stage. Although gypsum had no significant effect on aflatoxin contamination, there were significant differences (p = 0.009) in cultivar susceptibility to aflatoxin contamination. The cultivar with the smallest kernels had 18.8% lower aflatoxin content than the large-kernelled cultivar. Additionally, gypsum did not have a clear effect on pod yield. For instance, gypsum was associated with 44.8% more grain-filled pods in Kadononga (p = 0.005) at the site in Lusaka, but this result did not apply to the other two cultivars. At the site in Chongwe, gypsum was associated with 34.6% higher pod yield of MGV 5 only (p = 0.006). These results further suggest that plant factors such as kernel size may have an influence on natural resistance to aflatoxin contamination in peanuts. Key words: Aflatoxin, gypsum, peanut cultivar, pod-yield, Zambia. Corresponding author. E-mail: hendrix.chalwe@unza.zm. Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License Vol. 14(3), pp. 134-138, March 2020 DOI: 10.5897/AJPS2019.1807 Article Number: A7B6D2263193 ISSN 1996-0824 Copyright © 2020 Author(s) retain the copyright of this article http://www.academicjournals.org/AJPS Vol. 14(3), pp. 134-138, March 2020 DOI: 10.5897/AJPS2019.1807 Article Number: A7B6D2263193 ISSN 1996-0824 Copyright © 2020 Author(s) retain the copyright of this article http://www.academicjournals.org/AJPS Vol. 14(3), pp. 134-138, March 2020 DOI: 10.5897/AJPS2019.1807 Article Number: A7B6D2263193 ISSN 1996-0824 Copyright © 2020 Author(s) retain the copyright of this article http://www.academicjournals.org/AJPS INTRODUCTION The prevalence of high aflatoxin contamination in peanuts is a recurrent problem in most tropical climates including Zambia (Njoroge et al., 2017). This has prompted concerted efforts to combat aflatoxin contamination at various stages of the peanut value The prevalence of high aflatoxin contamination in peanuts is a recurrent problem in most tropical climates including Zambia (Njoroge et al., 2017). This has prompted concerted efforts to combat aflatoxin contamination at various stages of the peanut value chain. In the pre-harvest stages, there is need to implement good agricultural practices to minimize contamination during crop growth. Measures that minimize plant stress especially during the pod development stage are recommended so as to minimize Corresponding author. E-mail: hendrix.chalwe@unza.zm. Chalwe et al. Chalwe et al. 135 colonization of pods by toxigenic Aspergillus fungi and subsequent aflatoxin contamination in kernels (Waliyar et al., 2013; Torres et al., 2014). bunch type growth habit. MGV 5 and MGV 4 are Virginia market types taking between 120 and 130 days to physiological maturity with potential yields of 2.5 to 3.0 metric ton/ha and 1.5 to 2.5 metric ton/ha, respectively. Kadononga is a Spanish type that takes between 90 and 100 days to maturity with a yield potential of 1.0 to 1.5 metric ton/ha. In terms of seed size, MGV 5 is large-seeded; MGV 4 is medium-seeded while Kadononga is a small-seeded cultivar. The treatments were laid out in split-plot design with gypsum as the main plot factor and cultivar as the sub-plot factor, respectively. The experimental plots measured 4 m by 4 m with a 1 m isle between plots. Each factor was replicated thrice resulting in 18 experimental plots. Gypsum was applied on the soil surface covering the entire plant row span at flowering stage rates of 0 and 400 kg/ha. The time of application and the rate of 400 kg/ha of gypsum were adopted from the literature (Waliyar et al., 2013). One of the critical elements in the development of sound groundnut pods and kernels is calcium (Cox et al., 1976; Jain et al., 2011). Well-developed mature pods are not easily perforated by insects and this minimizes the entry of fungi into the seed tissue. It is on this principle that calcium-containing soil amendments such as gypsum are used to minimize pre-harvest aflatoxin contamination in peanuts (Reding et al., 1993; Gebreselassie et al., 2014). Harvesting and determination of pod-yield Harvesting of peanut pods was done at physiological maturity. Plants were dug out using a hand hoe. Pod yield was determined by counting grain-filled pods from each of the 6 randomly selected plants from the middle plant rows of each experimental plot. The pods were then dried in an electric vacuum oven (Hereanus, Germany) set at 45°C to a gravimetric moisture content of 10%. The dried pods were then shelled by hand and prepared for aflatoxin testing. Seeding and field management Seedbed preparation was done by ploughing with a tractor- mounted disc plough followed by levelling using a disc harrow. Seeding was done on a flat seedbed in 5 cm-deep planting holes made using a hand-hoe. The recommended seeding rate for each cultivar was followed. The fields were kept weed, pest and disease- free throughout the growing season. Weeds were uprooted by hand or dug-out using a hoe just as soon as they appeared. Appropriate pesticides and fungicides were sprayed regularly to control pests and fungal infections, respectively. However, cultivar response to calcium inputs partly depends on kernel size. Large seeded cultivars require higher inputs of calcium than small-seeded kernels (Jordan et al., 2014). The objective of this study was to assess the effect of gypsum amendment on kernel yield and pre-harvest aflatoxin contamination on aflatoxin- susceptible cultivars. Pod-yield and aflatoxin contamination in three peanut cultivars of Zambia were evaluated following a gypsum amendment on two soils with contrasting exchangeable calcium content. Location and soil properties A field experiment was conducted at the University of Zambia (UNZA), Field Research Station (15° 28.646’ S, 28° 20. 278’ E) in Lusaka district and at Kasisi Agricultural Training Centre (KATC15°14.989’ S, 28° 29.013’ E) in Chongwe district. The experiment was done under rain-fed conditions from mid-December 2016 to April 2017. Both research sites are situated in the agro- ecological region II of Zambia with mean annual rainfall ranging from 800 to 1000 mm (Soil Survey Branch, 2002). The two research sites were characterised by soil with contrasting chemical properties in terms of soil pH and exchangeable calcium content. The soil at KATC was characterised by strong acidity (pH = 4.22) and very low exchangeable calcium (0.06 cmol/kg) while the soil at UNZA had near neutral pH of 6.98 and high exchangeable calcium content of 5.14 cmol/kg. It should be noted that most acidic soils are associated with calcium deficiency (Brady and Weil, 2010). Sampling and aflatoxin analysis One third of the total kernel yield per treatment constituted the laboratory sample. The sample was constituted by aggregating several 100 g scoops from a single bulk sample. The bulk sample was shaken after each 100 g sub-sample was taken. Duly constituted laboratory samples were then ground into fine flour using an ordinary kitchen grinder (LM2211BM, Moulinex, China). Ground samples were homogenized by thorough shaking. Total aflatoxin content in the flour was extracted using 65% ethanol reconstituted from an original product (UN1170, Xilong Scientific Co., Shantou City, China) with a concentration of 95%. For each treatment, three sub-samples each weighing 10.0 g were mixed with 30 ml of ethanol and shaken on a rotary shaker (ISO-9001- 2000, Navyug, India) at 120 rpm for 3 min. After shaking, the mixture was filtered through Whitman 42 filter paper. Total aflatoxin concentration in each sample was determined using Neogen Afla Reveal® Q+ aflatoxin kit (Neogen Corporation, USA). The lower and upper limits of detection of aflatoxin concentration were 1 and 50 µg/kg, respectively. All the tested samples had total aflatoxin concentrations within detection limits. Treatments and experimental design The treatments in the experiment were two rates of gypsum and three cultivars. The cultivars were: MGV 5, MGV 4 and Kadononga. The chosen cultivars were among the most popular ones among small-holder farmers and seed companies. MGV 5 and MGV 4 were among the most common commercial cultivars while Kadononga was among the most common local landrace mostly preferred for its early maturity and ‘tasty’ kernels. All the three cultivars have a INTRODUCTION Additionally, amending soils with gypsum at the rate of 250 kg/ha was associated with higher grain yields compared with the control (Bairagi et al., 2017). According to Kabir et al. (2013) amending the soil with gypsum as a source of calcium resulted in higher number of pods per plant and 100 pod weights. Therefore, sufficient calcium fertilization in peanuts can both minimize aflatoxin contamination and increase kernel yield. RESULTS AND DISCUSSION Effects of gypsum on pod-yield in selected peanut cultivars Statistical analysis Data on pod yield and aflatoxin content were subjected to the 136 Afr. J. Plant Sci. Table 1. Effects of gypsum on the number of grain-filled pods per plant for each variety. gyp g p p p y Site Cultivar Gypsum level (kg/ha) Number of pods per plant ± standard error mean P-value UNZA Kadononga 0 29 ± 3.1 0.005a 400 42 ± 3.2 MGV 4 0 51 ± 5.9 0.410 400 44 ± 4.3 MGV 5 0 57 ± 3.8 0.284 400 65 ± 6.1 KATC Kadononga 0 23 ± 2.2 0.199 400 20 ± 1.4 MGV 4 0 30 ± 2.2 0.509 400 28 ± 2.6 MGV 5 0 26 ± 1.5 0.006a 400 35 ± 2.7 aGypsum was associated with higher number of grain-filled pods per plant in selected cases. analysis of variance test at 95% confidence interval. The separation of statistically significant treatment means was done using Fisher’s protected Least Significant Difference. analysis of variance test at 95% confidence interval. The separation of statistically significant treatment means was done using Fisher’s protected Least Significant Difference. characteristics such as the exchangeable calcium content. Recommendations must therefore consider soil type and in particular the native calcium content at which additional amounts would trigger a response. The choice of 400 kg/ha in the current study was based on the recommendation by Waliyar et al. (2013). The poor response to gypsum observed in the study suggests that the chosen rate of gypsum application may not be adequate for the soil types at the two sites. Effects of gypsum on pod-yield in selected peanut cultivars q yp The observed result in this study could be attributed to the native exchangeable calcium content of the soil at the two experimental sites. Although the soil at Kasisi had low exchangeable calcium, it was still able to meet the calcium requirements for Kadononga and MGV 4. As for MGV 5 at the same site, the positive response to gypsum input could signify the need for extra calcium for optimal growth. In the case of the soil at UNZA that had high native exchangeable calcium content, the logical expectation would be that the small-kernelled Kadononga would not respond to additional calcium, unlike the larger- kernelled MGV 4 and MGV 5. On the contrary, only the small-seeded Kadononga responded, making it difficult to attribute the response to gypsum amendment. Gypsum applied at flowering stage of peanuts resulted in higher number of grain-filled pods only in selected cultivars (Table 1). Although gypsum was associated with higher number of pods for Kadononga at UNZA, no similar result was observed at KATC. Similarly, gypsum was associated with a significantly higher number of grain-filled pods for MGV 5 at KATC and not at UNZA. MGV 4 did not show any response to gypsum at both sites. In general, these results are contrary to literature (Cox et al., 1976; Jain et al., 2011; Jordan et al., 2014; Bairagi et al., 2017) that report higher yields due to calcium-containing inputs such as gypsum. According to Kabir et al. (2013) peanut plants fertilized with calcium inputs recorded higher 100 pod weights compared with plants that did not receive a calcium input. As earlier noted by Smith et al. (1993) calcium is an essential element for pod filling in peanuts and a lack of it is reported to cause fruit abortions, resulting in fewer grain- filled pods. Additionally, differences in plant nutrient and moisture requirements between cultivars have an effect on crop performance. For instance, small-kernelled peanut cultivars have a lower nutrient and soil moisture requirement than larger ones (Jordan et al., 2014). Thus, if there is a limited supply of nutrients, especially exchangeable calcium and plant-available-water in the soil, the small-kernelled cultivars would grow normally According to Cox et al. (1976), peanut response to calcium-containing inputs is dependent on soil 137 Chalwe et al. Figure 1. Total aflatoxin content of peanuts kernels from respective cultivar. Error bars represent standard error of the mean. Effects of gypsum on pod-yield in selected peanut cultivars Letters within each data bar indicate statistical significance between treatments for each site. Figure 1. Total aflatoxin content of peanuts kernels from respective cultivar. Error bars represent standard error of the mean. Letters within each data bar indicate statistical significance between treatments for each site. under given soil conditions while the larger-kernelled cultivars would need external inputs. explained by the moisture status. Reding et al. (1993) attributed the reduction in aflatoxin contamination following the application of gypsum amendment to inherent plant factors. Similarly, results from the current study suggest that there could be plant factors such as pod strength that may influence aflatoxin contamination. According to Waliyar et al. (2013), well-developed mature pods tend to be less susceptible to fungal infection and subsequent aflatoxin contamination than immature and weak pods. In a study involving ten peanut genotypes, it was reported that although all the studied cultivars supported the growth of Aspergillus flavus, one cultivar recorded significantly less aflatoxin contamination (< 20 ppb) regardless of the amount of fungal accumulation (Korani et al., 2017). This result may suggest that some peanut genotypes are naturally more resistant to aflatoxin contamination than others. Effect of gypsum and cultivar on total aflatoxin content in kernels at harvest Results from the study showed that the gypsum amendment had no effect on total aflatoxin content in kernels (p > 0.05). In contrast, other authors such as Reding et al. (1993) and Gebreselassie et al. (2014) reported decreased aflatoxin content in peanut kernels due to gypsum amendment. Nonetheless, there were significant differences (p < 0.01) in mean total aflatoxin content in the three cultivars (Figure 1). The aflatoxin content varied according to kernel size. The cultivar with smallest kernel size was the least contaminated. This pattern was observed at both experimental sites. The mean total aflatoxin concentrations across the two sites were 10.1, 10.8 and 12 ppb for Kadononga, MGV 4 and MGV 5, respectively, in the same order as their kernel size starting with the smallest to the largest. Although the aflatoxin contamination in all the cultivars was within the permissible limits of less than 15 ppb according to the Zambia Bureau of Standards ZS 723 safety standard, the result in this study suggests that more effort is needed to manage aflatoxin contamination in larger-kernelled cultivars than in smaller ones. The authors have not declared any conflict of interests. The authors have not declared any conflict of interests. Njoroge SMC, Matumba L, Kanenga K, Siambi M, Waliyar F, Maruwo J, Machinjiri N, Monyo ES (2017). Aflatoxin B1 levels in groundnut products from local markets in Zambia. Mycotoxin Research 33(2):113-119. ACKNOWLEDGEMENTS Reding CLC, Harrison MA, Kvien CK (1993). Aspergillus parasiticus growth and aflatoxin synthesis on florunner peanuts grown in gypsum-supplemented soil. Journal of Food Protection 56(7):593- 611. This study was funded by the Peanut and Mycotoxin Innovation Laboratory under the Southern African Value Chain Project through the U.S. Agency for International Development, under the terms of Award No. AID-ECG-A- 00-07-0001 to The University of Georgia as management entity for the U.S. Feed the Future Innovation Lab on Peanut Productivity and Mycotoxin Control. Smith DH, Wells MA, Porter DM, Cox FR (1993). Nutrient deficiencies and toxicities in crop plants’, In: Peanuts, APS Press pp. 193-247. Soil Survey Branch (2002). Agro-ecological map of Zambia. Republic of Zambia, Ministry of Agriculture, Mt. Makulu Research Station, Department of Agriculture, Private Bag 7, Chilanga, Zambia. Torres AM, Barros GG, Palacios SA, Chulze SN, Battilani P (2014). Review on pre- and post-harvest management of peanuts to minimize aflatoxin contamination. Food Research International 62:11- 19. Conclusions Gypsum at the rate of 400 kg/ha did not have a clear influence on pod yield and no significant effect on total aflatoxin contamination in harvested kernels. However, this study showed significant differences in cultivar susceptibility to aflatoxin contamination. For the three cultivars in the study, results showed a negative relationship between pre-harvest aflatoxin content and kernel size indicating that inherent factors such as kernel size may have a role in determining aflatoxin resistance. Further, it can be concluded that management of aflatoxin contamination in larger-kernelled cultivars requires more effort than in smaller ones. Soil moisture status assessed in terms of mean daily rainfall received during the pod-development phase of the crop did not show significant differences (p > 0.05) across cultivars at each of the two sites. Therefore, the observed differences in mean total aflatoxin levels cannot be Afr. J. Plant Sci. 138 CONFLICT OF INTERESTS Korani WA, Chu Y, Holbrook C, Clevenger J, Ozias-Akins P (2017). Genotypic regulation of aﬂatoxin accumulation but not Aspergillus fungal growth upon post-harvest infection of peanut (Arachis hypogaea L.) seeds. Toxins 9(7):218. REFERENCES Waliyar F, Osiru M, Sudini HK, Njoroge SMC (2013). Aflatoxins: Finding solutions for improved food safety-Reducing Aflatoxins in Groundnuts through Integrated Management and Biocontrol. International Food Policy Research Institute, 2033 K Street, NW, Washington, DC 20006-1002 USA. Bairagi MD, David AA, Thomas T, Gurjar PC (2017). Effect of different level of N P K and gypsum on soil properties and yield of groundnut (Arachis hypogaea L.) var. Jyoti. International Journal of Current Microbiology and Applied Sciences 6(6):984-991. gy pp ( ) Brady NC, Weil RR (2010). Elements of the nature and properties of soils. 3rd ed. Prentice Hall, Upper Saddle River, New Jersey, USA. Cox FR, Sullivan GA, Martin CK (1976). Effect of calcium and irrigation treatments on peanut yield, grade and seed quality, Peanut Science 3:81-85. Gebreselassie R, Dereje A, Solomon H (2014). On-farm pre harvest agronomic management practices of Aspergillus infection on groundnut in Abergelle, Tigray. Journal of Plant Pathology and Microbiology 5(2):1-6. gy ( ) Jain M, Pathak BP, Harmon AC, Tillman BL, Gallo M (2011). Calcium dependent protein kinase (CDPK) expression during fruit development in cultivated peanut (Arachis hypogaea) under Ca2+- sufficient and -deficient growth regimens. Journal of Plant Physiology 168(18):2272-2277. Jordan DL, Brandenburg RL, Brown BA, Bullen GS, Roberson GT, Shew B (2014). Peanut Information. North Carolina Cooperative Extension Service, College of Agriculture and Life Sciences, North Carolina State University, NC, USA. Kabir R, Yeasmin S, Islam AKMM, Sarkar MAR (2013). Effect of phosphorus, calcium and boron on the growth and yield of groundnut (Arachis hypogea L.), International Journal of Bio-Science and Bio- Technology 5(3):51-60.
https://openalex.org/W4239341178	https://bmcnephrol.biomedcentral.com/track/pdf/10.1186/s12882-020-01913-7	English	null	The Correlation Analysis between the Oxford Classification of Chinese IgA Nephropathy Children and Renal Outcome -A retrospective cohort study	Research Square (Research Square)	2,020	cc-by	7,501	Open Access Heyan Wu1,2, Zhengkun Xia1* , Chunlin Gao1, Pei Zhang1, Xiao Yang1, Ren Wang3, Meiqiu Wang1 and Yingchao Peng1 Heyan Wu1,2, Zhengkun Xia1 , Chunlin Gao1, Pei Zhang1, Xiao Yang1, Ren Wang3, Meiqiu Wang1 and Yingchao Peng1 Abstract Background: The 2016 Oxford Classification’s MEST-C scoring system predicts outcomes in adults with IgA nephropathy (IgAN), but it lacks tremendous cohort validation in children with IgAN in China. We sought to verify whether the Oxford classification could be used to predict the renal outcome of children with IgAN. Methods: In this retrospective cohort study, 1243 Chinese IgAN children who underwent renal biopsy in Jinling Hospital were enregistered from 2000 to 2017. The combined endpoint was defined as either a ≥50% reduction in estimated glomerular filtration rate (eGFR) or end-stage renal disease (ESRD). We probed into the relevance betwixt the Oxford classification and renal prognosis. Results: There were 29% of children with mesangial proliferation(M1), 35% with endocapillary proliferation (E1), 37% with segmental sclerosis/adhesion lesion (S1), 23% with moderate tubular atrophy/interstitial fibrosis (T1 25–50% of cortical area involved), 4.3% with severe tubular atrophy/interstitial fibrosis (T2 > 50% of cortical area involved), 44% with crescent in< 25% of glomeruli(C1), and 4.6% with crescent in> 25% of glomeruli (C2). All children were followed for a medial of 7.2 (4.6–11.7) years, 171 children (14%) arrived at the combined endpoint. The multivariate COX regression model revealed that the presence of lesions S (HR2.7,95%CI 1.8 ~ 4.2, P<0.001) and T (HR6.6,95%CI 3.9 ~ 11.3, P<0.001) may be the reason for poorer prognosis in the whole cohort. In contrast, C lesion showed a significant association with the outcome only in children received no immunosuppressive treatment. Conclusions: This study revealed that S and T lesions were useful as the long-term renal prognostic factors among Chinese IgAN children. Keywords: IgA nephropathy, Oxford classification, Children, Renal outcome Correspondence: njxzk@126.com; shuangmu34@163.com 1Department of Pediatrics, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, China Full list of author information is available at the end of the article * Correspondence: njxzk@126.com; shuangmu34@163.com 1Department of Pediatrics, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, China Full list of author information is available at the end of the article Wu et al. BMC Nephrology (2020) 21:247 https://doi.org/10.1186/s12882-020-01913-7 Wu et al. BMC Nephrology (2020) 21:247 https://doi.org/10.1186/s12882-020-01913-7 * Correspondence: njxzk@126.com; shuangmu34@163.com 1Department of Pediatrics, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, China Full list of author information is available at the end of the article Inclusion criteria and clinical data set Children with IgAN who underwent renal biopsy in Jinling Hospital were enrolled from January 1, 2000, to December 31, 2017, in this retrospective cohort study. The inclusion criteria were follow-up ≥12 months, age at the time of biopsy ≤18 years, a minimum of eight glomeruli and an ini- tial eGFR≥15 ml/min /1.73m2. The exclusion criteria were secondary IgAN caused by Henoch-Schonlein purpura, liver cirrhosis, systemic lupus erythematosus, hepatitis B virus infection, tumour, ankylosing spondylitis and psoria- sis. The following clinical data including age, gender, dur- ation from onset to renal biopsy, estimate glomerular filtration rate (eGFR), mean arterial pressure (MAP), proteinuria and therapeutic regimen [Renin angioten- sin aldosterone system blockade (RASB), glucocortic- oid (GC) and other immunosuppressant agents] were collected during biopsy and follow-up. Immunosup- pressive therapy after kidney biopsy was defined as treatment with any immunosuppressive agent, regard- less of duration or dose. Pathology review Pathology review Renal pathology was scored according to the Oxford classification of IgAN [4], assessed by the following parameters: mesangial hypercellularity (M), scored as ab- sent (M0) or (M1) if≥50% of glomeruli had more than three cells per mesangial area; endocapillary hypercellu- larity (E), scored as E0 if absent or E1 if present; seg- mental glomerulosclerosis or adhesion (S), scored (S0) if absent or (S1) if present; tubular atrophy/interstitial fibrosis(T), scored as T0 (0–25% of cortical area), T1 (26–50% of cortical area), or T2(>50% of cortical area); cellular/ fibrocellular crescents scored as C0 (no cres- cents), C1 (crescents in at least one but < 25% of glom- eruli), or C2 (crescents in more than 25% of glomeruli). Immunofluorescence studies were performed (IgG, IgA, IgM, C3) and showed at least 1+ (on scale from 0 to 3+) mesangial deposition of IgA, with IgA being the dominant immunoglobulin deposited in the glomeruli. The intensity of deposits as determined via immunofluorescence mi- croscopy was graded semi-quantitatively on a scale from 0 to 3+, where 0 = no, 1+ = slight, 2+ = moderate, and 3+ = intense. At least two pathologists blinded to patient out- comes at the time of review confirmed the pathological results. Statistical analyses Continuous variables were presented as mean ± standard deviation (normally distributed variables) or median with interquartile range (IQR, non-normally distributed vari- ables) for data distribution. Categorical variables are presented as percentages. The Kruskal-Wallis test and chi-squared test were used to compare continuous and categorical variables, respectively. The Kaplan–Meier curve was used to illustrate univariate differences be- tween groups of pathological variables, and the log-rank test were used to test the two curves differences. The Cox proportional hazards regression model was con- ducted on univariable and multivariable analyses of Oxford classification in the IgAN children. Univariable Cox regression analysis was performed for each patho- logical variable. Multivariable Cox regression analysis (Backward: LR approach) adjusting for other clinically essential factors including initial eGFR, initial mean arterial pressure, and initial proteinuria was performed to appraise the function of MEST-C scoring system on the renal outcome. Hazard ratio (HR) with 95% confi- dence interval (CI) for each variable was estimated. All probabilities were two-tailed, and P-value< 0.05 was con- sidered statistically significant. © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Page 2 of 10 Page 2 of 10 Background of biopsy until the date of the last follow-up, the inci- dence of the event of interest, or March 2019 (end of the study). The combined endpoint was defined as either ≥50% reduction eGFR or ESRD or death. g IgA nephropathy (IgAN), being particularly frequent in Asia, is the primary reason for end-stage renal disease (ESRD) in all ages [1]. Since IgAN does not exhibit a specific serologic profile, a percutaneous kidney biopsy remains a definitive tool to establish the diagnosis of IgAN [2]. Additionally, the prognostic value of histo- logical data has become increasingly recognized in the past decade [3]. As a new pathological classification standard to judge the renal prognosis of IgAN, the Oxford classification [4] has been put forward in recent years. The purpose of the Oxford classification was to consider the pathological features associated with clin- ical outcomes independently of clinical data and to im- prove the current ability to predict outcomes in IgAN patients. Although the classification standard has been formulated by rigorous methodology, its clinical applica- tion in children needs to be further verified. What’s more, IgAN has regional and ethnic differences, which determines that the Oxford classification needs to be re- fined in different populations and races. In this study, the clinical and pathological data of IgAN followed up for a long time in our department were retrospectively analyzed to assess the predictability of Oxford classifica- tion among Chinese children. Pathological findings 9.4% had a fraction of glomeruli with crescents one- tenth or more, 6.6% had a fraction of glomeruli with crescents one-sixth or more, and only 4.6% had a frac- tion of glomeruli with crescents one-fourth or more. The percentage of immunoglobulins deposited only in the mesangial region was 68%, while 32% of immuno- globulins were deposited in both the mesangial and ca- pillary loop regions. 25% of children showed positive glomerular staining for IgG, 44% showed positive glom- erular staining for IgM, 84% showed positive glomerular staining for C3, and 1.1% showed positive glomerular staining for C4. The immunofluorescence intensity of IgA was between 1+ and 3+, including 5.6% of 1+, 13% of 2+ and 81% of 3 + . Definitions eGFR was calculated using the Schwartz formula [5], and the CKD-EPI equation [6] was used in adolescents aged > 16 years at the time of biopsy. ESRD was defined as eGFR < 15 mL/min/1.73m2, initiation of dialysis or transplantation. Survival time was defined from the time Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Page 3 of 10 Data analysis was executed using SPSS for windows version 26 (IBM Corporation, Armonk, NY). Table 2 Pathological findings at the time of biopsy in children with IgA nephropathy (n = 1243) Pathology findings Values at renal biopsy The number of glomeruli per biopsy 20.4 ± 4.7 MEST-C score % of total biopsies M1 29 E1 35 S1 37 T1 23 T2 4.3 C1 44 C2 4.6 Deposition site of immunoglobulins % of total biopsies Pure-mesangium 68 Mesangium+ capillary loop 32 Immunoglobulins deposits % of total biopsies Glomerular IgG deposition 25 Glomerular IgM deposition 44 Glomerular C3 deposition 84 Glomerular C4 deposition 1.1 Intensity of IgA deposits a % of total biopsies 1+ 5.6 2+ 13 3+ 81 Clinical features We enrolled 1243 Chinese children diagnosed with pri- mary IgAN from 2000 to 2017 in Jinling Hospital. Table 1 outlines the clinical, treatment and follow-up characteristics of the study children. The average age of children at diagnosis was 14 ± 4 years, with male- dominated (68%). The average value of MBP was 89 ± 16 mmHg, initial proteinuria was 0.6 (interquartile ranges, 0.3–1.4) g/day per 1.73m2, and the initial eGFR was 102 ± 20 ml/min per 1.73m2. All children were followed for a medial of 7.2 (4.6–11.7) years, 171 chil- dren (14%) arrived at combined endpoint (ESRD, n = 82; ≥50% eGFR decline, n = 89). During the follow-up period, 70% of children were treated with RASB, 45% were treated with GC, and 19% received GC combined other immunosuppressive drugs. Pathological findings The average of glomeruli was 20.4 ± 4.7 per biopsy. Based on Oxford classification,29% of the children showed M1, 35% showed E1, 37% showed S1, 23% showed T1, 4.3% showed T2, 44% showed C1 and 4.6% showed C2 (Table 2). The distribution of the percentage of crescents observed in every child was shown in Fig. 1. Of 48.6% children with any cellular/ fibrocellular cres- cents, 28% had crescents in<10% of glomeruli, whereas Table 1 Baseline and follow-up characteristics (n = 1243) At renal biopsy Values at renal biopsy Table 1 Baseline and follow-up characteristics (n = 1243) Table 1 Baseline and follow-up characteristics (n = 1243) At renal biopsy Values at renal biopsy Male, % 68 Age, years 14 ± 4 Duration from onset to renal biopsy (months) 12.0 (0.8,96.5) eGFR, ml/min per 1.73 m2 102 ± 20 MAP, mmHg 89 ± 16 Proteinuria, g/day per 1.73m2 0.6 (0.3–1.4) Follow-up parameters Values during follow-up Length of follow-up, years 7.2 (4.6–11.7) RASB, % 70 Any immunosuppression, % 64 GC, % 45 GC + IS, % 19 Combined event, % 14 ESRD, % 6.6 50% reduction in initial eGFR,% 7.2 Values are expressed as mean ± SD; medians (interquartile ranges), or percentages Abbreviations: eGFR Estimate glomerular filtration rate, MAP Mean arterial pressure, RASB Renin angiotensin aldosterone system blockade, GC Glucocorticoid, IS Other immunosuppressant, ESRD End-stage of renal disease Abbreviations: M1 Mesangial hypercellularity>0.5, E1 Presence of endocapillary hypercellularity, S1 Presence of segmental glomerulosclerosis, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/ interstitial fibrosis≥50% of cortical area, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli 9.4% had a fraction of glomeruli with crescents one- tenth or more, 6.6% had a fraction of glomeruli with crescents one-sixth or more, and only 4.6% had a frac- tion of glomeruli with crescents one-fourth or more. The percentage of immunoglobulins deposited only in the mesangial region was 68%, while 32% of immuno- globulins were deposited in both the mesangial and ca- pillary loop regions. 25% of children showed positive glomerular staining for IgG, 44% showed positive glom- erular staining for IgM, 84% showed positive glomerular staining for C3, and 1.1% showed positive glomerular staining for C4. The immunofluorescence intensity of IgA was between 1+ and 3+, including 5.6% of 1+, 13% of 2+ and 81% of 3 + . Values are expressed as mean ± SD or number (percentage); The intensity of IgA deposits as determined via immunofluorescence microscopy was graded semi-quantitatively on a scale from 0 to 3+, where 0 no, 1 + = slight, 2 + = moderate, and 3 + = intense Abbreviations: M1 Mesangial hypercellularity>0.5, E1 Presence of endocapillary hypercellularity, S1 Presence of segmental glomerulosclerosis, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/ interstitial fibrosis≥50% of cortical area, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli Associations between clinical and histologic variables Associations between clinical and histologic variables Linear regression analysis of Oxford classification with the most robust indicators for estimating renal decline (eGFR, MAP and proteinuria) was performed to explore the correlation between Oxford classification and clinical signs. As shown in Table 4, Children with S1, T1–2 and C1–2 lesions were associated with a reduced initial eGFR at the time of biopsy. All histological lesions (M1, E1, S1, T1–2, and C1–2) were associated individually with higher initial proteinuria at the time of biopsy. All histological lesions were associated with more upper ini- tial MAP at the time of biopsy. Table 3 Comparison of all kinds of lesions with different time of onset to renal biopsy (n = 1243) Table 3 Comparison of all kinds of lesions with different time of onset to renal biopsy (n = 1243) Variables Time from onset to renal biopsy χ2 P-value ≤12 months >12 months M0/M1 449/172 433/189 1.1 0.296 E0/ E1 411/210 401/221 0.4 0.525 S0/ S1 554/67 235/387 354.5 <0.001 T0/ T1/ T2 595/21/5 315/259/48 323.3 <0.001 C0/ C1/ C2 506/104/11 138/438/46 437.6 <0.001 Abbreviations: M0 Mesangial hypercellularity≤0.5, M1 Mesangial hypercellularity>0.5, E0 Absence of endocapillary hypercellularity, E1 Presence of endocapillary hypercellularity; S0 absence of segmental glomerulosclerosis, S1 Presence of segmental glomerulosclerosis, T0 Tubular atrophy/ interstitial fibrosis 0–25% of cortical area, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/interstitial fibrosis≥50% of cortical area, C0 Absence of crescents, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli Effects of different kidney biopsy time on the variables in Oxford classification The median time (12 months) of onset to renal biopsy was selected as the cut-off point to analyze the effect of Wu et al. BMC Nephrology (2020) 21:247 Page 4 of 10 Wu et al. BMC Nephrology Fig. 1 Distribution of the percentage of glomeruli with crescents in biopsies with any crescents. Crescents were present in 599(48%) of 1243 total biopsies the percentage of glomeruli with crescents in biopsies with any crescents. Crescents were present in 599(48%) of 1243 Fig. 1 Distribution of the percentage of glomeruli with crescents in biopsies with any crescents. Crescents were present in 599(48%) of total biopsies Fig. 1 Distribution of the percentage of glomeruli with crescents in biopsies with any crescents. Crescents were present in 599(48%) of 1243 total biopsies lesions, there was no significant difference in time from onset to renal biopsy within available data. biopsy time on variables in the Oxford classification From Table 3. It showed that when the time of onset to renal biopsy was less than 12 months, the patient’s le- sions were milder, dominated by S0(χ2 = 354.5, P<0.001), T0 (χ2 = 323.3, P<0.001), and C0(χ2 = 437.6, P<0.001). On the contrary, when the time of onset to renal biopsy was longer than 12 months, the lesions were corrected mainly by S1, T1–2 and C1–2. Concerning E and M biopsy time on variables in the Oxford classification From Table 3. It showed that when the time of onset to renal biopsy was less than 12 months, the patient’s le- sions were milder, dominated by S0(χ2 = 354.5, P<0.001), T0 (χ2 = 323.3, P<0.001), and C0(χ2 = 437.6, P<0.001). On the contrary, when the time of onset to renal biopsy was longer than 12 months, the lesions were corrected mainly by S1, T1–2 and C1–2. Concerning E and M Associations between pathologic features and renal outcome E were not significant variables, which may weaken their predictive values because of the low percentage in the cohort. The independent predictive value of pathology MEST-C score was reduced by immunosuppressive therapy. The associations between pathological features and renal outcomes were analyzed in a COX regression model (Table 5). With univariate COX regression model, renal outcome from a combined event were both associated significantly with lesions S (HR3.5, 95%CI 2.3 ~ 5.3, P< 0.001), T (HR 2.6, 95%CI 2.1 ~ 3.3, P<0.001) and C (HR 2.1, 95%CI 1.5 ~ 2.8, P<0.001). However, the lesion of M (HR 1.8, 95%CI 1.3 ~ 2.3, P = 0.115), and E (HR 1.4, 95%CI 0.9 ~ 2.1, P = 0.326) were not associated with renal outcome. When adjusted for clinically essential data in the multivariate COX regression model, only S (HR2.7, 95%CI1.8 ~ 4.2, P<0.001) and T (HR6.6, 95%CI 3.9 ~ 11.3, P<0.001) lesions remained as independent predictors of renal outcome. Our results suggested that patients with severe patho- logical lesions (e.g. S、T、C) were associated with lower eGFR, higher blood pressure and higher proteinuria, which were consistent with other findings [7–9]. Glomerular hypertension may mediate progressive renal damage by leading to glomerular hyperfiltration and glomerular enlargement [7]. For the control target of blood pres- sure in IgAN, the KDIGO guidelines [8] pointed out that when proteinuria > 0.3 g/day, the recommended target blood pressure (BP) was < 130/80 mmHg, and when proteinuria > 1 g/d, the recommended target BP was < 125/75 mmHg. Bellur et al. [9] showed that S was strongly associated with proteinuria and lower eGFR levels, which was consistent with our conclusion. Previ- ous studies [10] have shown that T was an independent risk factor for poor renal prognosis and associated with BP. Some scholars [11, 12] had found that the level of eGFR was lower in patients with IgAN with extensive crescent formation, and there was a negative correl- ation between eGFR and the proportion of crescents. Thus, it can be concluded that the most critical risk factors for the progression of IgAN (proteinuria, eGFR, MAP) are significantly correlated with the pathological damage found by renal biopsy, which reflects the value of the combination of clinical and pathological risk fac- tors in judging the prognosis of IgAN children. Predictive value of lesion M, E and C between immunosuppressive and without immunosuppressive groups We further assessed the predictive value of lesions M, E and C in children who received no immunosuppressive treatment to assess their natural predictive value. Chil- dren with crescent who didn’t receive immunosuppres- sive therapy experienced worse survival from the combined event (Fig. 3e), but this difference disappeared after received immunosuppression (Fig. 3f). The predict- ive value of lesion M and E were not changed by adding immunosuppressive treatment (Fig. 3a-d). Renal survival IgAN children according to Oxford classification As presented in Fig. 2, the Kaplan-Meier analyses revealed that lesion S (Log-Rank, χ2 = 14.796, P<0.001; Fig. 2c) and T (χ2 = 48.976, P<0.001; Fig. 2d) were each correlation with renal survival. However, lesions M (χ2 = 1.459, P = 0.177, Fig. 2a), E (χ2 = 2.399, P = 0.331, Fig. 2b) and C (χ2 = 6.218, P = 0.054, Fig. 2e) were not associated with renal outcome. Page 5 of 10 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology Table 4 Linear Regression Analysis of Oxford Classification and Clinical Indicators at Renal Biopsy Clinical indicators MAP (mmHg) eGFR (ml/min/1.73m2) Proteinuria (g/day/1.73m2) R P-value R P-value R P-value M1 0.342 <0.001 0.044 0.133 0.569 <0.001 E1 0.338 <0.001 −0.331 0.389 0.527 <0.001 S1 0.541 0.042 −0.744 0.007 0.604 <0.001 T1–2 0.532 <0.001 −0.578 <0.001 0.689 <0.001 C1–2 0.549 0.008 −0.447 <0.001 0.447 <0.001 Linear Regression results are results from separate models for each independent variable Abbreviations: MAP Mean arterial blood pressure, eGFR Estimate glomerular filtration rate, M1 Mesangial hypercellularity>0.5, E1 Presence of endocapillary hypercellularity, S1 Presence of segmental glomerulosclerosis, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/interstitial fibrosis≥50% of cortical area, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli Table 4 Linear Regression Analysis of Oxford Classification and Clinical Indicators at Renal Biopsy Linear Regression results are results from separate models for each independent variable Abbreviations: MAP Mean arterial blood pressure, eGFR Estimate glomerular filtration rate, M1 Mesangial hypercellularity>0.5, E1 Presence of endocapillary hypercellularity, S1 Presence of segmental glomerulosclerosis, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/interstitial fibrosis≥50% of cortical area, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli Discussion The lesion M was not a significant risk for renal out- come in our cohort. We speculated that it might be diffi- cult to address its value because of its low prevalence in our study. But the value of lesion M as an independent risk for progression is debated. On the one hand, the VALIGA cohort [13] and a Chinese adult cohort [14] confirmed M lesion as a significant factor for progres- sion, but on the other Shima et al. [15] reported that M had lost predictive value in patients receiving immuno- suppressive therapy. By and large, the presence of M This study investigated the clinical and histopathologic predictors of poor prognosis in pediatric patients with IgAN. The median duration from onset to renal biopsy was 12 months in our cohort. An early diagnosis seemed to be the primary reason for a low frequency of chronic and severe lesions such as lesion S, T and C. In our co- hort, we confirmed that S and T were independent risk factors associated with renal outcomes. The lesion C en- hanced the ability to predict progression only in those who did not receive immunosuppression. Lesion M and Wu et al. BMC Nephrology Wu et al. BMC Nephrology (2020) 21:247 Page 6 of 10 lesion appears to have a negligible correlation with renal outcomes therapy [17, 18] reported that E1 was independently as- sociated with more rapid loss of renal function and Fig. 2 Renal survival according to pathological variables. a Effect of the presence of mesangial hypercellularity on survival from a combined event in all patients. b Effect of the presence of endocapillary hypercellularity on survival from a combined event in all patients. c Effect of the presence of segmental sclerosis on survival from a combined event in all patients. d Effect of the presence of interstitial fibrosis/tubular atrophy on survival from a combined event in all patients. e Effect of the presence of cellular/fibrocellular crescents on survival from a combined event in all patients Fig. 2 Renal survival according to pathological variables. a Effect of the presence of mesangial hypercellularity on survival from a combined event in all patients. b Effect of the presence of endocapillary hypercellularity on survival from a combined event in all patients. c Effect of the presence of segmental sclerosis on survival from a combined event in all patients. Discussion Multivariate Cox Regression model: adjusted for initial eGFR, initial mean arterial pressure, and initial proteinuria p Abbreviations: CI Confidence interval, HR Hazard ratio, M0 Mesangial hypercellularity≤0.5, M1 Mesangial hypercellularity>0.5, E0 Absence of endocapillary hypercellularity, E1 Presence of endocapillary hypercellularity, S0 Absence of segmental glomerulosclerosis, S1 Presence of segmental glomerulosclerosis, T0 Tubular atrophy/ interstitial fibrosis 0–25% of cortical area, T1 Tubular atrophy/ interstitial fibrosis 26–50% of cortical area, T2 Tubular atrophy/interstitial fibrosis≥50% of cortical area, C0 Absence of crescents, C1 Crescents in at least one but < 25% of glomeruli, C2 Crescents in more than 25% of glomeruli The validation differences among the above different child cohorts are mainly related to the regional and eth- nic differences in IgAN, the selection criteria, follow-up time and treatment measures of each study, which em- phasizes the need to generate a large database for IgAN children to address the problem of insufficient statistical power due to the small number of progressive cases, es- pecially the relatively short follow-up period. Our data showed relevance between lesion S and renal prognosis, which further confirmed that S was a particu- lar index to judge the prognosis. Many data from the children’s cohort have proved the independent predictive value of S lesion. Children’s group from France con- firmed that lesion S was the only histological variable predicting a decline in renal function and was not asso- ciated with clinical data at the time of renal biopsy and whether they received immunosuppressive therapy [19]. Studies [20] have revealed S lesion develops from the organization of previous segmental necrotizing or endo- capillary inflammatory lesions or in response to podo- cyte injury and detachment. Therefore, it has also been suggested that in children with active glomerular lesions, special attention needs to be paid to the relationship be- tween lesion S and M and E. Our cohort validated the significance of the Oxford classification in a large number of Chinese IgAN chil- dren. A comprehensive analysis of the renal pathological features and clinical conditions represented in the co- hort suggests that Oxford classification must be consid- ered in conjunction with clinical features (including proteinuria levels and eGFR values) and treatment given after renal biopsy. This also suggests that treatment operations after biopsy may regulate some pathological risk factors. Discussion Multivariate Cox Regression model: adjusted for initial eGFR, initial mean arterial pressure, and Table 5 Factors at biopsy influencing renal outcome from ESRD or 50% drop in eGFR by univariate and multivariate Cox The lesion T was confirmed to be associated with renal failure, which was accord with almost all previous adult validation studies [4, 8, 10]. This may not be surprising because T lesion represents a more chronic and late stage of IgAN renal damage. However, most children validation studies, such as Japan cohort [15], Sweden cohort [21] and VALIGA cohort [13], failed to confirm that T lesion could maintain independent pre- dictive value in children. Only the cohort from China by Le et al. [22] and our cohort confirmed that T lesion has independent predictive value in children population. This difference may be due to only a small proportion of children arrived at a composite terminal during the follow-up in these child studies [13, 15, 21]. In our research, particular interest was given to chil- dren with lesion C as they had a predictive value (HR 2.1, 95% CI 1.5–2.8) in the univariate analysis, although it did not retain its significance in the multivariate ana- lysis. C lesion was seen in 49% of the children, however, with 28% having crescents in<10% of glomeruli. At the same time, a higher percentage of children with C were treated with immunosuppression than children without this lesion. Overall, crescents predicted a higher risk of a combined event, although this remained significant only in children not receiving immunosuppression. Thus, crescents in a minority of glomeruli may represent a lesion reversible by immunosuppressive therapy. Our findings suggest that children whose biopsies show these active lesions should be considered for immunosuppres- sive treatment, which was consistent with a multicenter study [23]. Segmental glomerulosclerosis S0 1 S1 3.5 (2.3 ~ 5.3) P-value <0.001 Tubular atrophy/interstitial fibrosis Univariate Cox Regression model: unadjusted. Multivariate Cox Regression model: adjusted for initial eGFR, initial mean arterial pressure, and initial proteinuria Univariate Cox Regression model: unadjusted. Multivariate Cox Regression model: adjusted for initial eGFR, initial mean arterial pressure, and initial proteinuria Univariate Cox Regression model: unadjusted. Discussion d Effect of the presence of interstitial fibrosis/tubular atrophy on survival from a combined event in all patients. e Effect of the presence of cellular/fibrocellular crescents on survival from a combined event in all patients Fig. 2 Renal survival according to pathological variables. a Effect of the presence of mesangial hypercellularity on survival from a combined event in all patients. b Effect of the presence of endocapillary hypercellularity on survival from a combined event in all patients. c Effect of the presence of segmental sclerosis on survival from a combined event in all patients. d Effect of the presence of interstitial fibrosis/tubular atrophy on survival from a combined event in all patients. e Effect of the presence of cellular/fibrocellular crescents on survival from a combined event in all patients lesion appears to have a negligible correlation with renal outcomes. therapy [17, 18] reported that E1 was independently as- sociated with more rapid loss of renal function and worse renal survival, which indirectly suggested that pro- liferative lesions were treatment-responsive. We conjec- tured that the widespread use of immunosuppressants might have influenced the absence of correlation be- tween E lesion and outcome. The lesion E, observed in 35% of the children, did not independently predict clinical outcome in the whole co- hort. This was highly consistent with the findings in the Oxford classification cohort [16]. However, two studies in which patients did not receive immunosuppressive Page 7 of 10 Page 7 of 10 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Table 5 Factors at biopsy influencing renal outcome from ESRD or 50% drop in eGFR by univariate and multivariate Cox regression Risk factors Univariate Cox Regression Multivariate Cox Regression HR(95%CI) HR(95%CI) Mesangial hypercellularity M0 1 M1 1.8 (1.3 ~ 2.3) P-value 0.115 Endocapillary hypercellularity E0 1 E1 1.4 (0.9 ~ 2.1) P-value 0.326 Segmental glomerulosclerosis S0 1 1 S1 3.5 (2.3 ~ 5.3) 2.7 (1.8 ~ 4.2) P-value <0.001 <0.001 Tubular atrophy/interstitial fibrosis T0 1 1 T1 or T2 2.6 (2.1 ~ 3.3) 6.6 (3.9 ~ 11.3) P-value <0.001 <0.001 Crescent C0 1 1 C1 or C2 2.1 (1.5 ~ 2.8) 1.8 (1.2 ~ 2.5) P-value <0.001 0.212 Univariate Cox Regression model: unadjusted. Discussion To explore the risk factors and their impact on disease progression by studying the clinical and pathological features of IgAN, the level of diagnosis and treatment of IgAN will ultimately be improved. The limitations of this study must be recognized. First, retrospective design makes the control of measured Wu et al. BMC Nephrology (2020) 21:247 Page 8 of 10 Wu et al. BMC Nephrology Fig. 3 Predictive value of mesangial hypercellularity, endocapillary hypercellularity and cellular/fibrocellular crescents between immunosuppressive and without immunosuppressive groups. a Effect of the presence of mesangial hypercellularity on survival from a combined event in patients without immunosuppression. b Effect of the presence of mesangial hypercellularity on survival from a combined event in patients with immunosuppression. c Effect of the presence of endocapillary hypercellularity on survival from a combined event in patients without immunosuppression. d Effect of the presence of endocapillary hypercellularity on survival from a combined event in patients with immunosuppression. e Effect of the presence of cellular/ fibrocellular crescents on survival from a combined event in patients without immunosuppression. f Effect of the presence of cellular/fibrocellular crescents on survival from a combined event in patients with immunosuppression Fig. 3 Predictive value of mesangial hypercellularity, endocapillary hypercellularity and cellular/fibrocellular crescents between immunosuppressive and without immunosuppressive groups. a Effect of the presence of mesangial hypercellularity on survival from a combined event in patients without immunosuppression. b Effect of the presence of mesangial hypercellularity on survival from a combined event in patients with immunosuppression. c Effect of the presence of endocapillary hypercellularity on survival from a combined event in patients without immunosuppression. d Effect of the presence of endocapillary hypercellularity on survival from a combined event in patients with immunosuppression. e Effect of the presence of cellular/ fibrocellular crescents on survival from a combined event in patients without immunosuppression. f Effect of the presence of cellular/fibrocellular crescents on survival from a combined event in patients with immunosuppression However, some features of this study may increase the strength of these findings, including the broad set of data collected over many years and long-term follow-up by the same team with a well-established clinical proto- col, as well as the careful re-evaluation of all renal variables difficult. Second, our results may not be extrap- olated to other ethnic groups due to geographical vari- ability in IgAN outcomes. Discussion Final, due to the limitations of retrospective studies, not all children were treated with RASB, which may weaken the rigour of the study. Page 9 of 10 Page 9 of 10 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 biopsies by two expert pathologist blinded to clinical data and outcome. Received: 19 August 2019 Accepted: 26 June 2020 Received: 19 August 2019 Accepted: 26 June 2020 Funding h h The authors acknowledge support from the Clinical Advanced Techniques, Primary Research & Development Plan of Jiangsu Province (BE2017719) and the Pediatric Medical Innovation Team of Jiangsu Province (CXTDA2017022). The cost of publishing paper was funded in this study. The funders had no role in the design of the study, data collection, analysis and data interpretation, or writing of the manuscript. 11. Nasri H, Mubarak M. Extracapillary proliferation in IgA nephropathy; recent findings and new ideas. J Nephropathol. 2015;4(1):1–5. 11. Nasri H, Mubarak M. Extracapillary proliferation in IgA nephropathy; recent findings and new ideas. J Nephropathol. 2015;4(1):1–5. 12. Rodrigues JC, Haas M, Reich HN. IgA nephropathy. Clin J Am Soc Nephrol. 2017;12(4):677–86. 12. Rodrigues JC, Haas M, Reich HN. IgA nephropathy. Clin J Am Soc Nephrol. 2017;12(4):677–86. 13. Coppo R, Troyanov S, Bellur S, Cattran D, Cook HT, Feehally J, Roberts IS, Morando L, Camilla R, Tesar V, Lunberg S, Gesualdo L, Emma F, Rollino C, Amore A, Praga M, Feriozzi S, Segoloni G, Pani A, Cancarini G, Durlik M, Moggia E, Mazzucco G, Giannakakis C, Honsova E, Sundelin BB, Di Palma AM, Ferrario F, Gutierrez E, Asunis AM, Barratt J, Tardanico R, Perkowska- Ptasinska A, VALIGA study of the ERA-EDTA Immunonephrology Working Group. Validation of the Oxford classification of IgA nephropathy in cohorts with different presentations and treatments. Kidney Int. 2014;86(4):828–36. Availability of data and materials The datasets used and analyzed in this study are available from the first author and corresponding author on reasonable request. Acknowledgements Not applicable. 8. Beck L, Bomback AS, Choi MJ, Holzman LB, Langford C, Mariani LH, Somers MJ, Trachtman H, Waldman M. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for glomerulonephritis. Am J Kidney Dis. 2013; 62(3):403–41. 8. Beck L, Bomback AS, Choi MJ, Holzman LB, Langford C, Mariani LH, Somers MJ, Trachtman H, Waldman M. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for glomerulonephritis. Am J Kidney Dis. 2013; 62(3):403–41. Authors’ contributions HW, ZX and CG performed the data collection and analysis and participated in manuscript writing. PZ, XY, RW, MW and YP performed the database setup and statistical analysis. HW, ZX and CG participated in the study design and coordination and helped to draft the manuscript. All of the authors have read and approved the final manuscript. 9. Bellur SS, Lepeytre F, Vorobyeva O, Troyanov S, Cook HT, Roberts IS. International IgA nephropathy working group. Evidence from the Oxford classification cohort supports the clinical value of subclassification of focal segmental glomerulosclerosis in IgA nephropathy. Kidney Int. 2017;91(1): 235–43. 9. Bellur SS, Lepeytre F, Vorobyeva O, Troyanov S, Cook HT, Roberts IS. International IgA nephropathy working group. Evidence from the Oxford classification cohort supports the clinical value of subclassification of focal segmental glomerulosclerosis in IgA nephropathy. Kidney Int. 2017;91(1): 235–43. 10. Zhang L, Li J, Yang S, Huang N, Zhou Q, Yang Q, Yu X. Clinicopathological features and risk factors analysis of IgA nephropathy associated with acute kidney injury. Ren Fail. 2016;38(5):799–805. 10. Zhang L, Li J, Yang S, Huang N, Zhou Q, Yang Q, Yu X. Clinicopathological features and risk factors analysis of IgA nephropathy associated with acute kidney injury. Ren Fail. 2016;38(5):799–805. Conclusions 1. Suzuki H. IgA nephropathy. Nihon Jinzo Gakkai Shi. 2015;57(8):1349–53. 2. Canetta PA, Kiryluk K, Appel GB. Glomerular diseases: emerging tests and therapies for IgA nephropathy. Clin J Am Soc Nephrol. 2014;9(3):617–25. 3. Roberts IS. Oxford classification of immunoglobulin a nephropathy: an update. Curr Opin Nephrol Hypertens. 2013;22(3):281–6. In summary, this study showed that T and S lesions were independently linked to poor renal outcome in Chinese IgAN children. In contrast, C lesion showed significant association with prognosis only in children received no immunosuppressive treatment. M and E lesions appeared to be unrelated to renal prognosis. 4. Trimarchi H, Barratt J, Cattran DC, Cook HT, Coppo R, Haas M, Liu ZH, Roberts IS, Yuzawa Y, Zhang H, Feehally J. IgAN classification working Group of the International IgA nephropathy network and the Renal Pathology Society; conference participants. Oxford classification of IgA nephropathy 2016: an update from the IgA nephropathy classification working group. Kidney Int. 2017;91(5):1014–21. Consent for publication 15. Shima Y, Nakanishi K, Kamei K, Togawa H, Nozu K, Tanaka R, Sasaki S, Iijima K, Yoshikawa N. Disappearance of glomerular IgA deposits in childhood IgA nephropathy showing diffuse mesangial proliferation after 2 years of combination/prednisolone therapy. Nephrol Dial Transplant. 2011;26(1):163–9. Not applicable. Ethics approval and consent to participate The study complied with the Declaration of Helsinki and was approved by the Medical Ethics Committee of Jinling Hospital (Approval number: 2019NZGKJ-266). The committee waived the requirement for informed consent in consideration of the retrospective nature of the study and anonymous data analyses. 14. Zeng CH, Le W, Ni Z, Zhang M, Miao L, Luo P, Wang R, Lv Z, Chen J, Tian J, Chen N, Pan X, Fu P, Hu Z, Wang L, Fan Q, Zheng H, Zhang D, Wang Y, Huo Y, Lin H, Chen S, Sun S, Wang Y, Liu Z, Liu D, Ma L, Pan T, Zhang A, Jiang X, Xing C, Sun B, Zhou Q, Tang W, Liu F, Liu Y, Liang S, Xu F, Huang Q, Shen H, Wang J, Shyr Y, Phillips S, Troyanov S, Fogo A, Liu ZH. A multicenter application and evaluation of the oxford classification of IgA nephropathy in adult Chinese patients. Am J Kidney Dis. 2012;60(5):812–20. 16. Working Group of the International IgA Nephropathy Network and the Renal Pathology Society, Cattran DC, Coppo R, Cook HT, Feehally J, Roberts IS, Troyanov S, Alpers CE, Amore A, Barratt J, Berthoux F, Bonsib S, Bruijn JA, D'Agati V, D'Amico G, Emancipator S, Emma F, Ferrario F, Fervenza FC, Florquin S, Fogo A, Geddes CC, Groene HJ, Haas M, Herzenberg AM, Hill PA, Hogg RJ, Hsu SI, Jennette JC, Joh K, Julian BA, Kawamura T, Lai FM, Leung CB, Li LS, Li PK, Liu ZH, Mackinnon B, Mezzano S, Schena FP, Tomino Y, Walker PD, Wang H, Weening JJ, Yoshikawa N, Zhang H. The Oxford Competing interests Competing interests The authors declare that they have no competing interests. Abbreviations 5. Schwartz GJ, Schneider MF, Maier PS, Moxey-Mims M, Dharnidharka VR, Warady BA, Furth SL, Muñoz A. Improved equations estimating GFR in children with chronic kidney disease using an immunonephelometric determination of cystatin C. Kidney Int. 2012;82(4):445–53. 5. Schwartz GJ, Schneider MF, Maier PS, Moxey-Mims M, Dharnidharka VR, Warady BA, Furth SL, Muñoz A. Improved equations estimating GFR in children with chronic kidney disease using an immunonephelometric determination of cystatin C. Kidney Int. 2012;82(4):445–53. C: Crescent; CI: Confidence intervals; eGFR: Estimated glomerular filtration rate; E: Endocapillary proliferation; ESRD: End-stage renal disease; GC: Glucocorticoid; IgAN: IgA Nephropathy; HR: Hazard ratio; IQR: Interquartile range; M: Mesangial proliferation; MAP: Mean arterial pressure; KDIGO: The Kidney Disease Improving Global Outcomes; RASB: Renin angiotensin aldosterone system blockade; S: Segmental sclerosis/adhesion lesion; T: Tubular atrophy/interstitial fibrosis; VALIGA: European Validation Study Of The Oxford Classification Of IgA Nephropathy 6. Selistre L, Rabilloud M, Cochat P, de Souza V, Iwaz J, Lemoine S, Beyerle F, Poli-de-Figueiredo CE, Dubourg L. Comparison of the Schwartz and CKD-EPI equations for estimating glomerular filtration rate in children, adolescents, and adults: a retrospective cross-sectional study. PLoS Med. 2016;13(3): e1001979. 6. Selistre L, Rabilloud M, Cochat P, de Souza V, Iwaz J, Lemoine S, Beyerle F, Poli-de-Figueiredo CE, Dubourg L. Comparison of the Schwartz and CKD-EPI equations for estimating glomerular filtration rate in children, adolescents, and adults: a retrospective cross-sectional study. PLoS Med. 2016;13(3): e1001979. 7. Ku E, Sarnak MJ, Toto R, McCulloch CE, Lin F, Smogorzewski M, Hsu CY. Effect of blood pressure control on long-term risk of end-stage renal disease and death among subgroups of patients with chronic kidney disease. J Am Heart Assoc. 2019;8(16):e012749. 7. Ku E, Sarnak MJ, Toto R, McCulloch CE, Lin F, Smogorzewski M, Hsu CY. Effect of blood pressure control on long-term risk of end-stage renal disease and death among subgroups of patients with chronic kidney disease. J Am Heart Assoc. 2019;8(16):e012749. Author details 1 Le W, Zeng CH, Liu Z, Liu D, Yang Q, Lin RX, Xia ZK, Fan ZM, Zhu G, Wu Y, Xu H, Zhai Y, Ding Y, Yang X, Liang S, Chen H, Xu F, Huang Q, Shen H, Wang J, Fogo AB, Liu ZH. Validation of the Oxford classification of IgA nephropathy for pediatric patients from China. BMC Nephrol. 2012;13:158. 23. Haas M, Verhave JC, Liu ZH, Alpers CE, Barratt J, Becker JU, Cattran D, Cook HT, Coppo R, Feehally J, Pani A, Perkowska-Ptasinska A, Roberts IS, Soares MF, Trimarchi H, Wang S, Yuzawa Y, Zhang H, Troyanov S, Katafuchi R. A multicenter study of the predictive value of crescents in IgA nephropathy. J Am Soc Nephrol. 2017;28(2):691–701. Author details 1 Author details 1Department of Pediatrics, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, China. 2Department of Pediatrics, Nanfang Hospital, Southern Medical University, Guangzhou, China. 3Department of Pediatrics, Jinling Hospital, Nanjing Medical University, Nanjing, China. Page 10 of 10 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 Wu et al. BMC Nephrology (2020) 21:247 classification of IgA nephropathy: rationale, clinicopathological correlations, and classification. Kidney Int. 2009;76(5):534–45. classification of IgA nephropathy: rationale, clinicopathological correlations, and classification. Kidney Int. 2009;76(5):534–45. 17. El Karoui K, Hill GS, Karras A, Jacquot C, Moulonguet L, Kourilsky O, Frémeaux-Bacchi V, Delahousse M, Duong Van Huyen JP, Loupy A, Bruneval P, Nochy D. A clinicopathologic study of thrombotic microangiopathy in IgA nephropathy. J Am Soc Nephrol. 2012;23(1):137–48. 17. El Karoui K, Hill GS, Karras A, Jacquot C, Moulonguet L, Kourilsky O, Frémeaux-Bacchi V, Delahousse M, Duong Van Huyen JP, Loupy A, Bruneval P, Nochy D. A clinicopathologic study of thrombotic microangiopathy in IgA nephropathy. J Am Soc Nephrol. 2012;23(1):137–48. 18. Chakera A, MacEwen C, Bellur SS, Chompuk LO, Lunn D, Roberts ISD. Prognostic value of endocapillary hypercellularity in IgA nephropathy patients with no immunosuppression. J Nephrol. 2016;29(3):367–75. 18. Chakera A, MacEwen C, Bellur SS, Chompuk LO, Lunn D, Roberts ISD. Prognostic value of endocapillary hypercellularity in IgA nephropathy patients with no immunosuppression. J Nephrol. 2016;29(3):367–75. 19. Cambier A, Rabant M, Peuchmaur M, Hertig A, Deschenes G, Couchoud C, Kolko A, Salomon R, Hogan J, Robert T. Immunosuppressive treatment in children with IgA nephropathy and the clinical value of Podocytopathic features. Kidney Int Rep. 2018;3(4):916–25. 20. Trimarchi H, Coppo R. Podocytopathy in the mesangial proliferative immunoglobulin a nephropathy: new insights into the mechanisms of damage and progression. Nephrol Dial Transplant. 2019;34(8):1280–5. 21. Edström Halling S, Söderberg MP, Berg UB. Predictors of outcome in paediatric IgA nephropathy with regard to clinical and histopathological variables (Oxford classification). Nephrol Dial Transplant. 2012;27(2):715–22. 22. Le W, Zeng CH, Liu Z, Liu D, Yang Q, Lin RX, Xia ZK, Fan ZM, Zhu G, Wu Y, Xu H, Zhai Y, Ding Y, Yang X, Liang S, Chen H, Xu F, Huang Q, Shen H, Wang J, Fogo AB, Liu ZH. Validation of the Oxford classification of IgA nephropathy for pediatric patients from China. BMC Nephrol. 2012;13:158. 22. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2595812771	https://www.epj-conferences.org/articles/epjconf/pdf/2016/21/epjconf_icnfp2016_05011.pdf	English	null	Search for long-lived neutral particles decaying into lepton-jets in 20.3 fb<sup>−1</sup>proton-proton collisions at √s = 8 TeV with the ATLAS detector	EPJ web of conferences	2,016	cc-by	2,766	Search for long-lived neutral particles decaying into lepton-jets in 20.3 fb−1 proton-proton collisions at √s = 8 TeV with the ATLAS detector Marco Schioppa1,a on behalf of the ATLAS Collaboration 1Universita’ della Calabria and INFN Cosenza, Rende, Italy Abstract. Several models of elementary particle physics beyond the Standard Model, predict the existence of neutral particles that can decay in jets of leptons and light hadrons (lepton-jets). The present contribution collects the results about the lepton-jet search with the ATLAS experiment at the proton-proton LHC collider at √s=8 TeV during the entire 2012 data taking (20 fb−1). No excess of events have been observed over the expected background and the exclusion limits for two diﬀerent models, that predict the Higgs boson to decay in lepton-jets, have been computed. A new lepton-jet search is underway for the new LHC era at ( √s=13 TeV). © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/). ae-mail: marco.schioppa@cern.ch , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 DOI: 10.1051/ 12605011 epjconf/2016 2 Lepton-jet deﬁnition and selection The search adopts a generic deﬁnition of lepton-jet in order to make the analysis as model-independent as possible. The lepton-jet contains at least one γd highly isolated in the inner detector, contained in a narrow angular cone (ΔR) and decaying far from the primary vertex to pairs of electrons/muons/pions. A lepton-jet gun Monte Carlo generator has been developed to optimize search criteria and to produce detection eﬃciency curves. This MC allows to study the response of the detector to lepton-jets and guide their characterization and the identiﬁcation of the variables useful for the selection of signal. Only three topologies of lepton-jets are considered: TYPE0, TYPE1 and TYPE2. The ﬁrst one is a cluster of only muons identiﬁed into the MS and no jets in a cone opening ΔR=0.5. It is the signature of a lepton-jet with all γd decaying to muon pairs. The second one is a cluster of two muons and jets in a cone opening ΔR=0.5. It the signature of one γd decaying to muon pair and the other one in electron or pion pair. The third one are jets with low electromagnetic fraction, narrow width and no muons in a cone opening ΔR=0.5. It is the signature of a lepton-jet with one or two γd both decaying to electron/pion pairs. Figures 1 shown a sketch of the three lepton-jet types. p p g p j yp The triggers used to select displaced lepton-jets of TYPE0 and TYPE1 is the unprescaled multi-muon trigger with at least three reconstructed muons in the MS with pT ≥6 GeV. For the TYPE2 lepton-jets the single jet trigger with lo electromagnetic fraction is used. The events that pass the trigger ﬁlters have to satisfy additional requirements. To separate the signal to the background the following major requirements are made: exactly two reconstructed lepton-jets in the event, both lepton-jet are isolated in the ID (the highest ID pT in the event must be ≤3 GeV), the absolute value of the azimuthal angle Δφ between the two lepton-jets must be ≥1, the muons must have no ID track match and the jets must be in timing with the bunch crossing (−1ns ≤tjet ≤5ns). Figure 1. 1 Introduction The Standard Model (SM) of particle physics and its mutual interactions has proven up to now ex- tremely successful, but despite that it still remains incomplete, e.g. it does not explain the nature of dark matter. New long lived light particles are predicted in several extensions of the SM, such as hidden sector scenarios. If the mass of these particles is in the MeV to GeV range, they would decay mainly to leptons and light mesons [1]. One of the most promising model is that in which the hidden sector and the visible one are coupled via the vector portal: a light hidden photon (γd) mixes kinetically with the SM photon. If the γd is the lightest state of the hidden sector, it decays back to SM particles. Due to the small mass of the γd and the size of the kinetic mixing parameter (that control both the BR and lifetime of the γd) these particles are produced with a large boost. As a consequence the decay products of the γd are collimated jets of pairs of electrons and/or muons and or pions that can be produced far from the IP of the event (displaced lepton-jet). The high resolution and high granularity measurement capability of the ATLAS air-core muon-spectrometer (MS) [2] is ideal for this kind of search. This note refers to the full dataset collected by ATLAS during the 2012 run at √s = 8 TeV, corresponding to an integrate luminosity of 20.3 fb−1 as reported in [3]. Search for non-prompt muonic lepton-jet has been performed at ATLAS during the 2011 data taking at √s = 7 TeV as reported in [4]. DOI: 10.1051/ 12605011 epjconf/2016 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 2 Lepton-jet deﬁnition and selection Sketch of the three lepton-jets classiﬁcation: left TYPE0 lepton-jet (only muons), centre TYPE1 lepton-jet (muons and jets), right TYPE2 lepton-jet (only jets). If the lepton-jet contains only one γd it contributes only to TYPE0 and TYPE2. Figure 1. Sketch of the three lepton-jets classiﬁcation: left TYPE0 lepton-jet (only muons), centre TYPE1 lepton-jet (muons and jets), right TYPE2 lepton-jet (only jets). If the lepton-jet contains only one γd it contributes only to TYPE0 and TYPE2. 3 Background The potential backgrounds in the signal sample include all processes with prompt muons with or without associated jets (e.g. W+jets, Z+jets, ttbar, single-top, etc.), cosmic-rays (e.g. muon bundles in cosmic-ray air showers), beam induced background. The cosmic-ray events are reduced using jet timing. For muons there is an additional requirement: the perigee to the beam line of the extrapolated MS track has to be close to the PV. For the evaluation of the remain CR contribution the triggers have been used in empty bunches of 2012 data. QCD multi-jet 2 2 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 DOI: 10.1051/ 12605011 epjconf/2016 is reduced using isolation in the ID (less dependent to pile-up) and jet EM fraction/width cuts. The isolation variable is the pT of the ID tracks associated to the PV in a cone ΔR=0.5 around the lepton- jet line of ﬂight (ptrack T ≥0.5 GeV). Figure 2-left shows the ID isolation distribution in the control sample of the 2012 data selected by single-jet triggers. The ID isolation is validated using the muons from Z→μμ decays. The remain QCD multi-jet background is evaluated using a data-driven matrix method. As uncorrelated variables the \|Δφ\| and max{ pT} are used. The signal region is deﬁned as \|Δφ\| ≥1rad and max{ pT} ≤3 GeV. Figure 2-right shows the event distribution in this plane before the requirements on \|Δφ\| and max{ pT}. Indicating with "A" the signal region, the residual background in A can be predicted from the population of the other three regions: NA = NB x ND/NC. Figure 2. Left plot is the distribution of pT: (ﬁlled dot) control sample of the ﬁrst 2 fb−1 of 2012 data, (ﬁlled square) Z →μμ in 2012 data and (solid line) Z →μμ MC sample. All distributions are normalized to unit area. Right plot is the distribution of lepton-jet events in the ABCD plane before the requirements max{ pT} ≤3 GeV and \|Δφ\| ≥1. Taken from [3]. Figure 2. Left plot is the distribution of pT: (ﬁlled dot) control sample of the ﬁrst 2 fb−1 of 2012 data, (ﬁlled square) Z →μμ in 2012 data and (solid line) Z →μμ MC sample. All distributions are normalized to unit area. Right plot is the distribution of lepton-jet events in the ABCD plane before the requirements max{ pT} ≤3 GeV and \|Δφ\| ≥1. Taken from [3]. 4 Lepton-jet benchmark models The performance of the lepton-jets search criteria are evaluated setting limits on two Falkowski- Ruderman-Volansky-Zupan (FRVZ) models [5] which predict non-SM Higgs boson decays to lepton- jets. Figure 3 shows the diagrams for the decay of the Higgs to lepton-jets in the two models. The two diagrams refer to the production of 2γd and 4γd respectively. In both models the decay chain contain two lighter hidden fermions HLSP (Hidden Lightest Stable Particle) that escape to the detection. 5 Results Summary of the lepton-jet (LJ) selection applied to data and background in the full 201 The ﬁrst uncertainty is statistical, while the second is systematic. All LJ pair types TYPE2-TYPE2 LJs excluded Data 119 29 Cosmic-rays 40 ± 11 ± 9 29 ± 9 ± 29 Multi-jets 70 ± 58 ± 11 12 ± 9 ± 2 Total background 110 ± 59 ± 14 41 ± 12 ± 29 (top plots) and for those in which the TYPE2-TYPE2 events are excluded. The corresponding limits with TYPE2-TYPE2 events excluded are shown in table 2. For the case of a hidden photon which kinetically mixes with the SM photon, these limits can be converted into exclusion limits on the kinetic mixing parameter ϵ. These results are also interpreted in the context of the Vector portal model as exclusion contours in the kinetic mixing parameter ϵ vs γd mass plane as shown in ﬁgure 5. Table 2. Ranges of γd lifetime cτ excluded at 95% CL for H →2γd + X and H →4γd + X, assuming 10% BR and the Higgs boson SM gluon fusion production cross section. The events TYPE2-TYPE2 are not used. FRVZ model Expected cτ [mm] BR(100%) Higgs →2γd + X 14<cτ<140 Higgs →4γd + X 15<cτ<260 5 Results Table 1 summarizes the data and background results of the search for lepton-jets in the 2012 data sample. Both for all lepton-jet pair events and for the case where the TYPE2-TYPE2 lepton-jets are excluded, the data agree with the background expectation. The results of the lepton-jets search are used to set upper limits on the Higgs boson decay branching fraction to lepton-jets as a function of the γd mean lifetime, according to the FRVZ models. The resulting exclusion limits on the σ x BR, assuming the Higgs boson SM gluon fusion production cross section σS M = 19.2 pb, are shown in ﬁgure 4 as a function of the γd mean lifetime (expressed as cτ) for the two benchmark models and for the cases in which all the lepton-jet type events are included 3 3 DOI: 10.1051/ 12605011 epjconf/2016 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 γd h fd 2 fd 2 γd HLSP HLSP ℓ + ℓ - ℓ + ℓ - γd h fd 2 fd 2 γd HLSP HLSP γd γd sd 1 sd 1 ℓ + ℓ - ℓ + ℓ - ℓ + ℓ - ℓ + ℓ - Figure 3. Diagrams of the two FRVZ models used as benchmarks in this search. l+ l−corresponds to elec- tron/muon/pion pair decay in the ﬁnal state. γd h fd 2 fd 2 γd HLSP HLSP γd γd sd 1 sd 1 ℓ + ℓ - ℓ + ℓ - ℓ + ℓ - ℓ + ℓ - Figure 3. Diagrams of the two FRVZ models used as benchmarks in this search. l+ l−corresponds to elec- tron/muon/pion pair decay in the ﬁnal state. Table 1. Summary of the lepton-jet (LJ) selection applied to data and background in the full 2012 data sample. The ﬁrst uncertainty is statistical, while the second is systematic. All LJ pair types TYPE2-TYPE2 LJs excluded Data 119 29 Cosmic-rays 40 ± 11 ± 9 29 ± 9 ± 29 Multi-jets 70 ± 58 ± 11 12 ± 9 ± 2 Total background 110 ± 59 ± 14 41 ± 12 ± 29 Table 1. Summary of the lepton-jet (LJ) selection applied to data and background in the full 2012 data sample. The ﬁrst uncertainty is statistical, while the second is systematic. Table 1. 6 To be continue with Run II data The opportunity oﬀered by LHC in the era at highest energy ( √s = 13 TeV) and high luminosity was caught and a new lepton-jet search is started. An ambitious physics program for this new work is underway. It will include new triggers (e.g. single muon trigger), new TYPEs with EM clusters, the RPC timing for cosmic-rays rejection, the lepton-jet vertex search and the harmonization with prompt- lepton-jet search and extend the analysis to other lepton-jet production mechanisms (e.g. associated production with vector bosons W and Z). 4 4 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 DOI: 10.1051/ 12605011 epjconf/2016 igure 4. Top plots: 95% upper limits on the σ x BR for the processes H →2γd + X (left) and H →4γd ight), as a function of the γd lifetime (cτ) for the FRVZ benchmark samples. Bottom plots: 95% upper lim xcluding the TYPE2-TYPE2 events. The expected limit is shown as the dashed curve and the almost ident olid curve shows the observed limit. The horizontal lines correspond to σ x BR for two values of the BR of iggs boson decay to dark photons. Taken from [3]. Figure 4. Top plots: 95% upper limits on the σ x BR for the processes H →2γd + X (left) and H →4γd + X (right), as a function of the γd lifetime (cτ) for the FRVZ benchmark samples. Bottom plots: 95% upper limits excluding the TYPE2-TYPE2 events. The expected limit is shown as the dashed curve and the almost identical solid curve shows the observed limit. The horizontal lines correspond to σ x BR for two values of the BR of the Higgs boson decay to dark photons. Taken from [3]. 7 Conclusions The search for new phenomena beyond the Standard Model is a very active ﬁeld at the ATLAS ex- periment. Recent results for lepton-jet search at ATLAS are presented in this note. No excess over the Standard Model expectation has been observed and limits are placed on the two FRVZ benchmark models. 5 5 5 DOI: 10.1051/ 12605011 epjconf/2016 , 126 EPJ Web of Conferences 05011 (2016) ICNFP 2015 [GeV] dγ m -3 10 -2 10 -1 10 1 ∈ kinetic mixing parameter -11 10 -10 10 -9 10 -8 10 -7 10 -6 10 -5 10 -4 10 -3 10 -2 10 ATLAS = 8 TeV s -1 20.3 fb SN LSND BaBar A1 APEX KLOE HADES CHARM E137 U70 Orsay E141 E774 e a σ ,5 μ a favoured σ 2 ±,μ a 90% CL ATLAS BR 40% BR 20% BR 10% BR 5% Figure 5. Parameter space exclusion plot for γd as a function of its mass and of the kinetic mixing parameter ϵ. The 90% CL exclusion limits are presented assuming the FRVZ model H →2γd + X with decay branching fraction to γd of 5/10/20/40% and the NNLO gluon fusion Higgs production cross section. There are shown the existing 90% CL exclusion regions from other experiments like beam dump, electron and muon anomalous magnetic moment, constraints from astrophysical observations. Taken from [3]. [GeV] dγ m -3 10 -2 10 -1 10 1 ∈ kinetic mixing parameter -11 10 -10 10 -9 10 -8 10 -7 10 -6 10 -5 10 -4 10 -3 10 -2 10 ATLAS = 8 TeV s -1 20.3 fb SN LSND BaBar A1 APEX KLOE HADES CHARM E137 U70 Orsay E141 E774 e a σ ,5 μ a favoured σ 2 ±,μ a 90% CL ATLAS BR 40% BR 20% BR 10% BR 5% Figure 5. Parameter space exclusion plot for γd as a function of its mass and of the kinetic mixing parameter ϵ. The 90% CL exclusion limits are presented assuming the FRVZ model H →2γd + X with decay branching fraction to γd of 5/10/20/40% and the NNLO gluon fusion Higgs production cross section. There are shown the existing 90% CL exclusion regions from other experiments like beam dump, electron and muon anomalous magnetic moment, constraints from astrophysical observations. Taken from [3]. References [1] M. J. Strassler and K. M. Zurek, Phys. Lett. B 651 374 (2007) [2] The ATLAS Collaboration, JINST 3, S08003 (2008) [3] The ATLAS Collaboration, JHEP 11, 088 (2014) [4] The ATLAS Collaboration, Phys. Lett. B 721, 32 (2013) [5] A. Falkowski, J. T. Ruderman, T. Volansky and J. Zupan, JHEP 05 077 (2010) 6
https://openalex.org/W3163503573	https://zenodo.org/record/4745520/files/1109-pre.pdf	English	null	The experience of women students in engineering and mathematics careers: a focus group study	null	2,021	cc-by	6,676	The experience of women students in engineering and mathematics careers: a focus group study Alicia García-Holgado GRIAL Research Group, Computer Science Department University of Salamanca Salamanca, Spain aliciagh@usal.es Sonia Verdugo-Castro GRIAL Research Group, Department of Didactic, Organization and Research Methods University of Salamanca Salamanca, Spain soniavercas@usal.es Angeles Dominguez Tecnológico de Monterrey Monterrey, Mexico Universidad Andrés Bello Chile angeles.dominguez@tec.mx Francisco J García-Peñalvo GRIAL Research Group, Computer Science Department University of Salamanca Salamanca, Spain fgarcia@usal.es Mª Cruz Sánchez-Gómez GRIAL Research Group, Department of Didactic, Organization and Research Methods University of Salamanca Salamanca, Spain mcsago@usal.es Andrea Vázquez-Ingelmo GRIAL Research Group, Computer Science Department University of Salamanca Salamanca, Spain andreavazquez@usal.es Itzel Hernández-Armenta Tecnológico de Monterrey Monterrey, Mexico armenta.itz@gmail.com Itzel Hernández-Armenta Tecnológico de Monterrey Monterrey, Mexico armenta.itz@gmail.com Abstract—The gender gap is a problem that occurs in different forms in regions and countries around the world. It is a goal of large international organisations, governments, companies and other entities. Although it is not a new issue, it is important to continue studying it and seek mechanisms and strategies to attract and maintain more women in these areas. In particular, in the field of education and employment, the STEM areas present large gender gaps whose reduction would not only impact the equality of men and women but would also have an impact on the economy of the countries and on improving the economic situation of women. In this context, there are initiatives in Latin America working on this issue, but it is necessary to look more deeply into the elements that influence the decision to study careers in these areas. In this context, two focus groups have been held as roundtables with STEM women from different Latin American and European countries, to answer a series of questions centred on their motivations and decisions before and during their university studies. The results obtained have provided some inputs for defining gender equality action plans in ten Higher Education Institutions from Chile, Colombia, Costa Rica, Ecuador, and Mexico. Furthermore, the results show similarities with previous studies involving STEM women with different Latin American profiles. PRE in tertiary education enrolled in engineering, manufacturing, and construction programs is between 6% and 7% between 2015-2018; in contrast, the percentage of male students choosing these careers is around 20-21%. NT Implementing mechanisms and strategies in the higher education system are measures to close the gender gap. A. García-Holgado et al., "The experience of women students in engineering and mathematics careers: a focus group study," in 2021 IEEE Global Engineering Education Conference (EDUCON) (21-23 April 2021, Vienna, Austria). USA: IEEE, 2021, pp. 50-56. A. García-Holgado et al., "The experience of women students in engineering and mathematics careers: a focus group study," in 2021 IEEE Global Engineering Education Conference (EDUCON) (21-23 April 2021, Vienna, Austria). USA: IEEE, 2021, pp. 50-56. The experience of women students in engineering and mathematics careers: a focus group study However, it requires a holistic approach that involves all educational and professional stages, integrating universities, industry, and government efforts to develop national policies towards reducing the gender gap. In this context, the W- STEM project, "Building the future of Latin America: engaging women into STEM," aims to involve higher education institutions across Latin America and Europe to establish mechanisms of attraction, access, and guidance of women into STEM careers. E-PRINT Several factors influence the gender gap in STEM disciplines. Different studies focused on the contextual influences during active phases of educational or career decision making [1-5]. According to the Social Cognitive Career Theory (SCCT) [6] and the subsequent adaptations [7, 8], some of the main influential factors are self-perception, self-efficacy, interest in science, expectations of results, previous educational experiences, family and social context, personal characteristics of the person and his/her objectives when deciding which studies to take. Before enrolling in STEM careers, the support or barriers can impact the number of women enrolled in STEM careers [9]. However, the effort to reduce the gender gap does not finish at this stage; it starts in the first educational stages and continues during professional life, according to the Leaky Pipeline phenomenon. There are also a high number of women who drop out in STEM studies. During their studies, the university context can have an impact on the drop out numbers, comfort in the university environment [10], teachers support [1] or the support received from the faculty [11] are some factors identified in previous studies. Keywords—gender gap, women students, STEM, qualitative analysis, engineer, mathematician. y , g , I. INTRODUCTION I. INTRODUCTION Nowadays, the gender gap in STEM (Science, Technology, Engineering, and Mathematics) is a challenge. According to the UNESCO Sustainability Development Goals (SDG), organisations and governments worldwide have plans to reduce the gender gap, not only in STEM but also in all the societies' areas, according to the UNESCO Sustainability Development Goals (SDG). This is not only a problem that affects women; it is crucial for the future of society. According to the World Economic Forum, the forthcoming future is characterised by the transformation of the industries favouring technological skills. The new technologies will merge the physical, digital, and biological worlds, impacting all disciplines and economies. STEM careers will have a crucial role in this transformation. However, women are underrepresented in those areas. According to the UNESCO Institute for Statistics, the mean percentage of female students This study aims to analyse the perception of undergraduate and graduate students of STEM studies to identify the support and barriers that influenced their career decision-making, focusing on the socio-cultural context, and the support and barriers perception of women in the university context during P2 Universidad del Norte Colombia Systems and Computer Engineering Student P3 Oulu University Spain Telecommunicatio ns Engineering PhD student P4 Politecnico di Torino Colombia Computer Science PhD student P5 Technologic al University Dublin Venezuela Systems Engineering PhD student P6 Tecnológico de Monterrey Mexico Business and Information Technology Engineering Student P7 Universidad de Guadalajara Mexico Computer Science Student P8 Universidad Técnica Federico Santa María Chile Civil Computer Engineering Student P9 Pontificia Universidad Católica de Valparaíso Chile Civil Computer Engineering Graduate P10 Universidad Tecnológica de Bolívar Colombia Systems Engineering Student P11 Instituto Tecnológico de Costa Rica Costa Rica Computer Engineering Student P12 Universidad de Costa Rica Costa Rica Computer Science Graduate P13 Universidad Técnica Particular de Loja Ecuador Electronics and Telecommunicatio ns Engineer Graduate P14 Universidad Técnica del Norte Ecuador Mechatronics engineering Student E-PRINT their STEM studies. A set of research questions were defined to achieve this objective: their STEM studies. A set of research questions were defined to achieve this objective: • RQ1: Which motivations and ideas motivated or made it more difficult to choose a STEM career? • RQ2: What influence has the socio-cultural context (family, friends, couple, etc.) had on the educational path? • RQ3: What is the perception of support and barriers women have concerning the university context during their studies in STEM areas? y , g , I. INTRODUCTION The rest of the document is organised as follows. Section 2 describes the roundtables organised with engineering and mathematics students. Section 3 presents the results concerning gender equality. Finally, the last section summarises the main conclusions derived from this work. II. APPROACH PARTICIPANTS IN THE ROUNDTABLE ABOUT GIRLS IN ICT TABLE I. PARTICIPANTS IN THE ROUNDTABLE ABOUT GIRLS IN ICT Institution Country Degree Situation P1 University of Salamanca Spain Computer Science Student TABLE I. PARTICIPANTS IN THE ROUNDTABLE ABOUT GIRLS IN ICT Institution Country Degree Situation P1 University of Salamanca Spain Computer Science Student II. APPROACH The study was carried out in April and May 2020, during the COVID-19 crisis. Two international focus groups were organised to answer the research questions. The focus groups were developed as online roundtables discussions with female students and graduates in STEM studies. The roundtables were organised as online events streamed through YouTube and Facebook Live (Figure 1). Furthermore, the events are framed as part of the W-STEM project's activities (Building the future of Latin America: engaging women into STEM), a cooperation project between Europe and Latin America funded by the European Union. The project aims to reduce the gender gap in STEM careers in Latin America, establishing mechanism and strategies in the processes of attraction, access, retention and orientation in Higher Education Institutions [12, 13]. RE Fig. 1. Flyers associated with the two roundtables. PRE The second roundtable was related to the International Women in Mathematics Day on 12 May. We involved 15 female students and graduates from studies with a high mathematical component such as Industrial Physical Engineering, Mathematics or Civil Mathematical Engineering. The same countries were involved (Table II). It is also important to note that the participants invited to both roundtables are STEM women who are not actively involved in bridging the gender gap in these areas and have no connection to the W-STEM project. This approach was aimed at reducing biases as much as possible. Besides, the participants had no prior instructions on how the discussion would be conducted. Fig. 1. Flyers associated with the two roundtables. Each roundtable was focused on a STEM area and related to an International Day to engage more people. The first roundtable was on 23 April during the Girls in ICT Day and it was focused on ICT-related studies such as Computer Engineering, Mechatronics Engineering or Systems Engineering. This roundtable involved 14 female students and graduates from Chile, Colombia, Costa Rica, Ecuador, Finland, Ireland, Italy, Mexico, United Kingdom, and Spain. It is important to notice that the roundtable was in Spanish, so the participants from non-Spanish speakers' countries were women who speak Spanish (Table I). TABLE II. PARTICIPANTS IN THE ROUNDTABLE ABOUT WOMEN IN MATHEMATICS Institution Country Degree Situation P1 Oulu University Venezuela Chemical Engineering Graduate P2 Politecnico di Torino Italy Mathematics Master student P3 Pontificia Universidad Católica de Valparaíso Chile Mathematics PhD Student TABLE II. PARTICIPANTS IN THE ROUNDTABLE ABOUT WOMEN IN MATHEMATICS TABLE I. TABLE I. P4 Technological University Dublin Mexico Mathematics Graduate P5 Tecnológico de Monterrey Mexico Industrial Physical Engineering Graduate P6 Tecnológico de Monterrey Mexico Engineering Physics Master student P7 Universidad de Costa Rica Costa Rica Mathematics Student P8 Universidad de Guadalajara Mexico Mathematics Student P9 Universidad de Guadalajara Mexico Physical & Mathematical Sciences Graduate P10 Universidad del Norte Colombia Mathematics PhD Student P11 Universidad Técnica del Norte Ecuador Electronics and communications engineering Graduate P12 Universidad Técnica Federico Santa María Chile Civil Mathematical Engineering Student P13 Universidad Técnica Particular de Loja Ecuador Civil Engineering Student P14 Universidad Tecnológica de Bolívar Colombia Mathematics Graduate P15 University of Salamanca Spain Mathematics Student groups. The first one related to the reason for the choice of studies, which could be mathematics or ICT. The second question related to the experience as a student of mathematics or ICT and issues to be highlighted. The third question related to how to motivate other women and girls to study mathematics or ICT. groups. The first one related to the reason for the choice of studies, which could be mathematics or ICT. The second question related to the experience as a student of mathematics or ICT and issues to be highlighted. The third question related to how to motivate other women and girls to study mathematics or ICT. From the answers given by the participants of the two focus groups held, it has been observed that, about the reason for choosing these studies, the following stand out: tradition, care for others, participation in initiatives such as competitions, championships and Olympics related to science, interest in these studies, passion for the chosen studies and the usefulness found in these studies (mathematics or ICT). Some of the women indicate that they chose such studies because they were good at the subject at school and therefore did not question other study options but were clear that it should be that option. This does not mean that family tradition also plays a role; if the parents or siblings, mainly the boys, have taken these studies, the daughters follow the professional path. This can be seen from the statements of two participants in the ICT focus group: P4. Because at school I was doing well in physics and mathematics and in my time, it was like natural selection. If we were doing well in those subjects, we had to make a career out of them. P3. • What can be done to encourage more women and girls to study ICT/mathematics? I chose to study engineering because I did not feel as capable as a doctor although I wanted to help people. And since in sixth grade, we all want to be doctors. Well, they want to be doctors. So I did not feel with the strength and energy to be a doctor, because I could not be with a person and help them and cure them, no. So, I said "how can I help people from somewhere else", and engineering seemed like something that I had an affinity for and was doing well, so I decided. The qualitative analysis of the content has been carried out by having the literal speeches transcribed. Once the transcriptions have been read, the meta-categories and categories of the concept map have been created from the collected discourses. For the construction of the categories, the elements highlighted by the participants in the focus groups, in the answers to the questions, have been considered. Subsequently, it has been possible to coding the contents in the categorical system in order to link the narratives with the concepts of the map. On the other hand, there is a transparent motivating element for the choice of mathematics and ICT studies, and that is the participation in championships, Olympics, etc. Two women from the mathematics focus group say: P4. I like mathematics, just as I wanted physics, just as I liked even philosophy. So, since I was in secondary school, I have been in the mathematics Olympics, and that is where I liked the form, the approach that they give to it, that is to say, it is not This analysis process can be carried out through qualitative analysis programmes that support the process of creating the meta-categories and subcategories, to facilitate codification [14]. TABLE I. I liked science, maths and physics at school, but I did have a reference point. My father is a telecommunications engineer, so it was effortless for me to say, "I do the same thing and that's it". PRINT In other cases, the first option was studies linked to the care of other people, such as medicine. Even the family was more supportive of their daughters if they decided to go for this type of study. However, after thinking about it, they decided to opt for other kinds of studies such as mathematics and ICT, where they felt they could also contribute to society. This was stated by one of the participants in the mathematics focus group: P7. I didn't know what to study, many years of my life I thought I would study social work because I had a very deep-seated part of me that I wanted to help people and that was the way I saw that I could help people. After a while, I realised that this career would not be very compatible with my personality. I know that you have to be an influential person. And I thought, "what am I going to do now, what am I going to study", and finally, several factors led me to mathematics. E-PR The roundtables were in Spanish to facilitate the discussion. Each one took 90 minutes and was moderated by two women researchers related to mathematics and software engineering. The roundtables were divided into two phases. A first phase in which each woman answered the focus group questions and a second phase in which women answered the online attendants' questions. The questions were: PRE • Why did you choose to study ICT/mathematics, and what career do you hope to pursue? P • What is your experience as a student of ICT/mathematics? Did anything surprise you? Do you find it fun? Would you recommend it? • What can be done to encourage more women and girls to study ICT/mathematics? III. RESULTS AND DISCUSSION Three main questions have been asked in the focus everyday life. Since I was very young, it was a skill that has been honed, which is also a fundamental tool. so much the process, it is not so much the result, but the procedure that gives it more importance. And I liked that very much. P7. After a data center event, I met a lady who works at NASA. She studied computers and talking to other colleagues; she told us that for NASA, it is essential that people study science careers, and that there is work beyond being a maths teacher. On the other hand, another necessary category of analysis is social influences. These influences can be family, peer group, other references such as teachers, etc. However, these influences can favour or hinder the decision-making process. In the mathematics group, two participants indicate the family influence: P4. It was always something I liked since I was very young, and I always wanted to study by myself and take my big brother's books and see what he was doing. P7. I was helped a lot with the support of my father, who studied computers. Whereas in the case of ICT, you can see two different positions. On the one hand, you can see how family tradition can favour the choice of such studies. However, family tradition can also hinder the decision if it is loaded with stereotypes: P13. My mother told me "no, this is a career for men, you should study medicine or another career", so I asked "but why should this career be for men, why not for a woman? So, I think that is one of the factors that gave me strength, instead of discouragement. P12. It was a bit of divine enlightenment. Maybe computer science was my thing. And at the same time, it was also a bit about blood, because my parents studied the same thing, my older sister studied engineering, and my younger sister is also studying engineering. So basically, the genes and the family nucleus were an excellent motivation for me to study computer science. RINT A woman from the ICT focus group highlights: P6. Yes, it was a process that changed in many areas, and I think something that helped me a lot was that I went to an event, to a camp, for fun. III. RESULTS AND DISCUSSION The event is called Campus Party, and it takes place in Mexico and various parts of Europe. It is a week developed so that the people who participate, of all ages, can learn about Information Technology. I went for pure fun, and when I went, I left with the idea of "how could I not have been presented with this option before", in other words, it was self-evident. My whole life profile pointed directly to an IT career; however, I never saw it, and it wasn't because someone had deprived me of the option to choose an IT career, it just hadn't come up. The passion for mathematics and ICT studies is also of great value in deciding which tasks to choose. This is revealed in the mathematics group: P3. My passion for mathematics started at a very young age, and I found mathematics easy, but this is not what called me into this world. What encouraged me to enter this world was a contest that I had the opportunity to attend, which was of mathematical logic. It was this contest that changed my perspective on mathematics. And it is also indicated in the ICT group: P5. I always say and tell my friends that if I were born again, I would also study the same thing. Why? Because I love to dream. I love to dream. And maybe what I discovered with this career is that everything begins with an idea. And you bring ideas to life, bring projects to life, and the limit is in the imagination. It's not about being super smart; it's not about being the best at maths or programming. No. It's merely about having the idea and working on it. So, one of the best things that have happened to me in my life is this study of computers. P13. I think it's one of the parts of my career that I'm most passionate about. Engineering has a wide, vast field of application. It's amazing. You can do so many things. For example, in my career, I can be in telecommunications, I can be in electronics. PRE However, it is not only the family that is an essential socialising element in influencing the decision, but also the teachers. In the case of mathematics: P9. III. RESULTS AND DISCUSSION I had an approach, but more than anything else I think what motivated me to study the physics-mathematics degree was a lecture I heard from a physicist while I was in high school, where, with a lot of passion and overflowing with an interest for his area and his specialisation, he spoke about the need we human beings have to understand the environment around us, to understand many of the phenomena that occur. And in the case of ICT: P2. I entered Systems Engineering, and I didn't know what I was getting into and I chose it. It was because of a physics teacher at the school who considered me to be very good at the subject and who had potential and told me "study this career, it has potential". This career has made me fall in love. E-PRIN By looking at the elements that the participating women highlight in their studies, benefits and difficulties can be identified. Especially in ICT, the variety of areas in which a person can specialise, and work has been highlighted: P9. The good thing I find about ICT is that you can work in any area and learn any area of the same computer science. And accompanied by the passion for studies is also the interest, as three participants indicate: P4. In a moment of introspection, because I already had to do my paperwork to get into university, I started to think "it's effortless to say that you love something when it's going well, but when you're going through a difficult time, and you're still there persevering, you realise that you like this". That's when I decided to study for a degree in mathematics, and it's a decision I made almost two years ago, and I don't regret it (Math). P12. I was always very entertained by maths; it was like the only subject where I came with a desire to learn. After everything the teacher taught us, I would go home to find exercises and solve them like crazy (Math). P11. It was like love at first sight. It was as if my mind exploded, and I realised that I was super excited and wanted to know more and more (ICT). Interestingly, of the five difficulties mostly encountered in training by these women, two are related to women's invisibility, and another problem is related to the invisibility of studies. hunger, people don't get jobs". them practice in the area of engineering. The number decreases, it decreases as a woman advances, so it's part of the challenge, and it's part of our responsibility to solve it. However, part of the origin of the gender gap is the gender stereotypes that are widespread in the contexts in which we develop, because of culture and the reproduction of biases. Therefore, it is necessary to be clear about some issues of gender equality and equal opportunities such as that all people have the right to access and representation, as indicated in the group of ICT: P5. Girls are as good or better than boys, that we have a disadvantage because we are girls is false; we have a history of being very good at programming. Once this point has been reached, women and girls need to be motivated to pursue the studies that they like and are excited about. We must move away from gender biases, from stereotypes. We must also move away from preconceived ideas that women are better able to care for others or will find more employment if they go to other studies. Positive factors must be promoted, such as those social references that female students have that encourage them to follow the path that they like. Therefore, to close the focus group sessions, participants are asked what they would say to other girls and women to encourage them to study mathematics and ICT. In the mathematics group, the following stand out: P3. I want the girls who did not have the option like me to find that love for mathematics in that competition, to know that mathematics is much more. Maths is much more than calculus; it is super beautiful, and I would love to invite them to learn a little more about mathematics. Mathematics at heart is everywhere. P1. I see the faces of all these young girls and what I would like to say to them is that it is not easy, but there is something in their heart that tells them "that is the right way to go", it is a passion. Although it is the most challenging path, although it will not give you the immediate satisfaction that love, that passion is what will allow them to move forward. And in the ICT group, the following participant stands out: P11. III. RESULTS AND DISCUSSION Therefore, we should ask ourselves what is happening socially so that women feel that they are being left out of the scientific and technical education system or that they are not reaching it. In ICT, the gender gap is indicated: P6. It is impressive that there are far fewer women and there is far less presence of women, but there is also so much need for critical thinking that we women generate, and in my case, we are starting more than we are advancing. So, I think there are several challenges. The main challenge is that we need more women to study engineering careers. But now, how do you stay on course? You often check a career in engineering, and when you finish, because the field is so vast, not all of Also, in the case of mathematics, the usefulness found in studies makes up a key element: P11. What has encouraged me since I was very young about mathematics is that it is unpredictable and straightforward for every aspect of hunger, people don't get jobs". I believe that both women and men can take risks with the things or decisions they make. It is not a question of gender. It's each person, each person has different thoughts, and each person is going to make other decisions, so I think we must move away from gender. PRINT On the lack of knowledge about the studies, in the group of mathematics, they point out: P14. In the search, I found the mathematics, without really knowing everything that I was going to find within the career. Because it is not a well-known career within the entire portfolio of university careers. Let's say that I entered without really knowing much, and I know that many people may not know what can be achieved in a science career. P5. It wasn't easy to find information about careers offering this type of study, because there aren't that many. And the ones that do exist are not very publicised. However, two other difficulties identified are the difficulty of the studies themselves, as indicated in ICT: P8. The truth is that race has been a rollercoaster ride of emotions. I'll be super honest; I wasn't very close to maths and physics, so when I took the challenge to enter an engineering school, it was scary and difficult. It cost me a lot. But with studies, with work, with dedication and the fact of not giving up, one manages to get everything out. And you also identify the difficulty of finding a job, as indicated in the maths group: P1. There is something important, in the subject of studying mathematics it has always been saying "don't study mathematics, why would you study that, people are dying of RE Figure 2 shows the concept map with the system of meta- categories and categories. E-P Fig. 2. Concepts map. PRE Fig. 2. Concepts map. Fig. 2. Concepts map. Following the results, it should be indicated that [15] produced similar results to those described in this study, within another experience of the W-STEM project carried out with women different from the current study. In this previous study, women also expressed their passion for their studies and pointed out that family and teachers have a unique role in choosing their studies. In some cases, the influence has been positive, and in other cases, the impact has contributed to women questioning their abilities, feeling less valid. RQ1: Which motivations and ideas motivated or made it more difficult to choose a STEM career? RQ1: Which motivations and ideas motivated or made it more difficult to choose a STEM career? In general, all the women in the study give motivational messages to those women and girls who are attracted to mathematics and ICT, but who do not dare to pursue these studies. In some cases, fear holds back women who want to study science and technology. This is why these women are more self-confident and move away from the biases that exist towards women in STEM. PRINT The women participating in the study highlight some of the motivations that encouraged them to study their studies. On the one hand, some of them have always been attracted to higher education; some even have a strong passion for it. Another motivation that these women find is the variety of areas they can specialise and work in, within mathematics and ICT. In this sense, the awareness of the usefulness of their studies has been a driving force in their decision to pursue them. Some of the women have previously participated in scientific-technical initiatives such as the Academic Olympiads, competitions and championships. Initially, when they participated in these initiatives, they were not clear about their professional vocation. However, participation in these initiatives helped to dispel their doubts. This is why it is important to encourage participation in initiatives of these characteristics, intending to explore personal interests. In another hemisphere, one finds motivations such as tradition. In this sense, some women have chosen such studies following the family tradition of their parents or brothers. However, these studies are also selected for their knowledge of the comfort zone. Those women who felt that they were good at the subjects associated with mathematics and ICT, in some cases, decide to follow that educational option. And finally, there are some cases where women wanted to devote themselves to caring for others, mainly through medicine. However, after trying this option or seriously considering it, they chose mathematics and ICT as a means of contributing to society. PRE Some of the problems and difficulties they identify in the university context during their STEM studies are the gender gap, gender stereotypes, the lack of visibility of women in STEM, the lack of information about the studies, and also, the level of difficulty of the studies and the subsequent difficulty in finding work about it. RQ1: Which motivations and ideas motivated or made it more difficult to choose a STEM career? E-P They point out that they felt strange in the classrooms as they were the only ones among so many men. Although they stress that they were well treated by their colleagues, they also highlight that the rates of representation of women are lower. Furthermore, the stereotypes in which they have grown up cannot be ignored. These stereotypes underestimate their ability to perform STEM work. However, sometimes there is not enough information about the studies they want to take, which also makes it difficult to access them. IV. CONCLUSIONS Among the outputs of W-STEM is the definition of gender equality action plans for the Latin American higher education institutions involved in the project. The action plans are focused on reflecting the strategies and mechanisms to improve the processes of attraction, access and guidance of women into STEM studies at higher education institutions. hunger, people don't get jobs". Also, in [16], women were also strongly motivated to get other women and girls to fight for their dreams, away from the fear of daring. On the other hand, the profession is also underestimated, conveying that "if you study that, you will starve and not find work". However, not everything is pessimistic. Some families support their daughters' decisions through modelling. There are women who follow this educational option because their family has been an example to follow. On the other hand, teachers also provide motivating references for women, from which they can follow their example. This is the case with some teachers who verbally motivate the girls and even transmit their subject with passion. Regarding teachers and families as a whole, some women stress that caution is required. In some cases, women are told "if you want, I can help you", assuming that they cannot do it alone. Doing this can make the person feel undervalued, so we must give the woman the possibility to explore her capacities and abilities, and if necessary, let her be the one to ask for help. As a society, we must work for equal opportunities; we must fight to mitigate the gender biases and stereotypes that still perpetuate inequalities. Prejudices that professions are gendered must be uprooted. And from the educational levels, interventions must also be directed towards the family and the environment of girls and women. Finally, answering the three research questions posed in the study: RQ3: What is the perception of support and barriers women have concerning the university context during their studies in STEM areas? T RQ2: What influence has the socio-cultural context (family, friends, couple, etc.) had on the educational path? RQ2: What influence has the socio-cultural context (family, friends, couple, etc.) had on the educational path? The focus group study carried out with students from different universities across Latin America (Chile, Colombia, Costa Rica, Ecuador, México) and Europe (Finland, Ireland, Italy, Spain, United Kingdom) has provided some inputs for the definition of gender equality action plans in the involved institutions. Although it is not possible to generalise the results to all young women in Latin America, the results complement other quantitative studies. The application of focus groups as a qualitative analysis technique allows discussions on a topic The primary contexts that have conditioned the choice have been family and teachers. The family has conditioned the preference of their studies for and against. In contrast, gender biases and stereotypes continue to be reproduced. The preconceived idea that women should care for others and also teach continues to spread. Some women are sent the message that they are not good enough to be scientific and technical professionals. [7] M. Cupani, A. E. Azpilicueta, and V. Sialle, "Evaluación de un modelo social-cognitivo de la elección de la carrera desde la tipología de Holland en estudiantes de la escuela secundaria," Revista Española de Orientación y Psicopedagogía, vol. 28, no. 3, pp. 18–24, 2017, doi: 10.5944/reop.vol.28.num.3.2017.21615. among different people, thus exposing their different experiences and perceptions. This in turn allows conclusions to be drawn about the overall experience of the people involved in the focus group. On the other hand, other methodological techniques such as the individual interview do not allow the presentation of the diversity of experiences, and the final reading of the data collection is a single case. [8] I. M. a. B. B. Vázquez Romero, A., "Factores sociocognitivos asociados a la elección de estudios científico-matemáticos. Un análisis diferencial por sexo y curso en la Educación Secundaria," Revista de investigación educativa, RIE, vol. 37, no. 1, pp. 269–286, 2019, doi: 10.6018/rie.37.1.303531. ACKNOWLEDGEMENT [9] A. García-Holgado, C. S. González-González, and A. Peixoto, "A comparative study on the support in engineering courses: a case study in Brazil and Spain," IEEE Access, vol. 8, pp. 125179-125190, 2020, doi: 10.1109/ACCESS.2020.3007711. This work has been possible with the support of the Erasmus+ Programme of the European Union in its Key Action 2 "Capacity-building in Higher Education". Project W- STEM "Building the future of Latin America: engaging women into STEM" (Reference number 598923-EPP-1-2018- 1-ES-EPPKA2-CBHE-JP). The content of this publication does not reflect the official opinion of the European Union. Responsibility for the information and views expressed in the publication lies entirely with the authors. [10] A. M. Gloria and S. E. Robinson Kurpius, "Influences of self-beliefs, social support, and comfort in the university environment on the academic nonpersistence decisions of American Indian undergraduates," Cultural Diversity and Ethnic Minority Psychology, vol. 7, no. 1, pp. 88-102, 2001, doi: 10.1037/1099-9809.7.1.88. [11] S. L. Eddy and S. E. Brownell, "Beneath the numbers: A review of gender disparities in undergraduate education across science, technology, engineering, and math disciplines," Physical Review Physics Education Research, vol. 12, no. 2, p. 020106, 2016, doi: 10.1103/PhysRevPhysEducRes.12.020106. This research was partially supported by the Spanish Ministerio de Ciencia, Inovación y Universidades under a FPU fellowship (FPU017/01252 and FPU17/03276). [12] A. García-Holgado, A. Camacho Díaz, and F. J. García-Peñalvo, "Engaging women into STEM in Latin America: W-STEM project," in Proceedings of the 7th International Conference on Technological Ecosystems for Enhancing Multiculturality (TEEM 2019) (León, Spain, October 16-18, 2019), M. Á. Conde-González, F. J. Rodríguez Sedano, C. Fernández Llamas, and F. J. García-Peñalvo Eds., (ACM International Conference Proceeding Series (ICPS). New York, NY, USA: ACM, 2019, pp. 232-239. INT REFERENCES [1] C. A. Moss-Racusin, J. F. Dovidio, V. L. Brescoll, M. J. Graham, and J. Handelsman, "Science faculty's subtle gender biases favor male students," Proceedings of the National Academy of Sciences, vol. 109, no. 41, pp. 16474-16479, 2012, doi: 10.1073/pnas.1211286109. [13] F. J. García-Peñalvo, A. Bello, A. Domínguez, and R. M. Romero Chacón, "Gender Balance Actions, Policies and Strategies for STEM: Results from a World Café Conversation," Education in the Knowledge Society, vol. 20, 31, 2019, doi: 10.14201/eks2019_20_a3. PRIN [2] R. W. Lent, S. D. Brown, and G. Hackett, "Contextual supports and barriers to career choice: A social cognitive analysis," Journal of Counseling Psychology, vol. 47, no. 1, pp. 36-49, 2000, doi: 10.1037/0022-0167.47.1.36. [14] M. C. Sánchez-Gómez and M. V. Martín-Cilleros, "Implementation of focus group in health research," Studies in Systems, Decision and Control, vol. 71, pp. 49-61, 2017, doi: 10.1007/978-3-319-43271-7_5. E-PR [3] S. O. Salami, "Influence of culture, family and individual differences on choice of gender‐dominated occupations among female students in tertiary institutions," Women in Management Review, vol. 22, no. 8, pp. 650-665, 2007, doi: 10.1108/09649420710836326. RE [15] S. Verdugo-Castro, A. García-Holgado, and M. C. Sánchez-Gómez, "Interviews of Spanish women in STEM: a multimedia analysis about their experiences," in Proceedings of the 8th International Conference on Technological Ecosystems for Enhancing Multiculturality (TEEM 2020) (Salamanca, Spain, October 21-23, 2020), F. J. García-Peñalvo Ed. New York, NY, USA: ACM, 2020. E- [4] M. C. Cadaret, P. J. Hartung, L. M. Subich, and I. K. Weigold, "Stereotype threat as a barrier to women entering engineering careers," Journal of Vocational Behavior, vol. 99, pp. 40-51, 2017, doi: 10.1016/j.jvb.2016.12.002. PRE [5] C. Leaper and C. R. Starr, "Helping and Hindering Undergraduate Women's STEM Motivation: Experiences With STEM Encouragement, STEM-Related Gender Bias, and Sexual Harassment," Psychology of Women Quarterly, vol. 43, no. 2, pp. 165-183, 2019, doi: 10.1177/0361684318806302. PR [16] S. Verdugo-Castro, M. C. Sánchez-Gómez, A. García-Holgado, and M. Bakieva, "Pilot study on university students' opinion about STEM studies at higher education," in Proceedings of the 8th International Conference on Technological Ecosystems for Enhancing Multiculturality (TEEM 2020) (Salamanca, Spain, October 21-23, 2020), F. J. García-Peñalvo Ed. New York, NY, USA: ACM, 2020. [6] R. W. Lent, S. D. Brown, and G. Hackett, "Toward a Unifying Social Cognitive Theory of Career and Academic Interest, Choice, and Performance," Journal of Vocational Behavior, vol. 45, no. 1, pp. 79– 122, 1994, doi: 10.1006/jvbe.1994.1027.
https://openalex.org/W2605691527	https://www.scielo.br/j/ibju/a/cKWyqCSyvdWxsxph8T7Yz8S/?lang=en&format=pdf	English	null	Selection of best videos of the year for 2016	International Braz J Urol	2,017	cc-by	858	EDITORIAL Volume 43 \| number 2 \| March . April, 2017 Dear esteemed colleagues and friends, As we begin a new year, I would like to take this opportunity to thank each of you for your commitment in supporting our journal and expanding its reader- ship. This past year has been truly exceptional in the quality of submissions we have received within the video section of the International Brazilian Journal of Urology. We are committed in publishing the highest quality videos detailing novel surgical techniques and approaches by leading surgical teams from across the world. Simi- larly, we encourage groups pushing the envelope in terms of how we can continually refine the art of surgery and ultimately improve the outcomes of our patients. In this regard, I am pleased to share with you this year’s selection for best videos of the year for 2016. Many individual criteria were taken into account in making this selection including novelty, superior quality in terms of video depiction and narration, and lastly vides that best depict re-defining surgical approaches to urological diseases. On that note, here are the selections: First Prize: Robotic ureteroureterostomy for treatment of a proximal ureteral stricture by Andrade HS et al. from the Glickman Urological and Kidney Institute, Cleveland Cli- nic (Cleveland, Ohio) published in volume 42(5):1041-1042, September-October 2016 (available at: http://www.intbrazjurol.com.br/video-section/andrade_1041_1042) (1). This video is truly a masterful depiction on how minimally invasive robotic sur- gery can be used to tackle benign ureteral stricture disease. Excellent perioperative surgical outcomes are reported and the authors provide sufficient follow-up of 27 months to insure these favorable outcomes are maintained with adequate follow-up. There is no question this surgical approach and specifically this video can be used by surgeons skilled and interested in conducting such surgical procedures. Selection of best videos of the year for 2016 Dear esteemed colleagues and friends, REFERENCES 1. Andrade HS, Kaouk JH, Zargar H, Caputo PA, Akca O, Ramirez D, Autorino R, Noble M, Stein RJ. Robotic Ureteroureterostomy for Treatment of a Proximal Ureteric Stricture. Int Braz J Urol. 2016;42:1041-1042. 3. Tobias-Machado M, Nunes-Silva I, Hidaka AK, Sato LL, Almeida R, Colombo JR Jr, Zampolli HC, Pompeo AC. Retzus-sparing robotic-assisted laparoscopic radical prostatectomy: a step-by-step technique description of this first brazilian experience. Int Braz J Urol. 2016;42:1250. 2. Jairath A, B SS, Mishra S, Ganpule A, Sabnis R, Desai M. Robotic repair of vesicovaginal fistula - initial experience. Int Braz J Urol. 2016;42:168-9. 2. Jairath A, B SS, Mishra S, Ganpule A, Sabnis R, Desai M. Robotic repair of vesicovaginal fistula - initial experience. Int Braz J Urol. 2016;42:168-9. EDITORIAL Volume 43 \| number 2 \| March . April, 2017 \| INT BRAZ J UROL doi: 10.1590/S1677-5538.IBJU.2017.02.02 Volume 43 \| number 2 \| March . April, 2017 robotic assisted laparoscopic transabdominal extravesical approach. The authors report excellent outcomes with no recurrences at short-term follow-up of 3 months. robotic assisted laparoscopic transabdominal extravesical approach. The authors report excellent outcomes with no recurrences at short-term follow-up of 3 months. Second Prize: Robotic repair of vesicovaginal fistula- initial experience by Jairath A et al. from the Department of Urology, Muljbhai Patel Urological Hospital (Nadiad, India) published in volume 42(1):168-169, January-February 2016 (available at: http://www.intbrazjurol.com.br/video-section/video-library/jairath_168_169) (2). The authors presents a surgical series of 8 vesicovaginal fistulas approached using a 182 Third Prize: Retzus-sparing robotic-assisted laparoscopic radical prostatectomy: A step-by-step technique description of this first Brazilian experience by Tobias-Machado M et al. from the Departmento de Urologia, Faculdade de Medicina do ABC, Santo Andre, SP, Brazil and other centers of excellence in Sao Paulo, SP, Brazil published in volume 42(6):1250, November-December 2016 (available at: http://www.intbrazjurol.com.br/video-section/tobias-machado_1250_1250) (3). This surgical vi- deo is quite elegant in its demonstration that Retzus sparing RRP is not only feasible and reproducible but can enhance continence recovery following initial catheter removal. This is furthermore an excel- lent depiction of the wonderful surgical collaborations taking place in Brazil. I would like to conclude this editorial by once again thanking each of you for the support of the International Brazilian Journal of Urology. We are committed in publishing the very best original articles and videos from across the world and similarly will continue to do so in a very timely manner, with a rigorous peer review process by the very best subject leaders. Lastly, I send my very best wishes to you and your families for 2017. Video Section Editor, International Brazilian of Urology Associate member, Department of GU Oncology Moffitt Cancer Center Associate Professor, Department of Urology University of South Florida Tampa, Florida, USA E-mail: philippe.spiess@moffitt.org 1. Andrade HS, Kaouk JH, Zargar H, Caputo PA, Akca O, Ramirez D, Autorino R, Noble M, Stein RJ. Robotic Ureteroureterostomy for Treatment of a Proximal Ureteric Stricture. Int Braz J Urol. 2016;42:1041-1042. Philippe E Spiess, M.D, MS, FACS, FRCS(C) Video Section Editor, International Brazilian of Urology Associate member, Department of GU Oncology Moffitt Cancer Center Associate Professor, Department of Urology University of South Florida Tampa, Florida, USA E-mail: philippe.spiess@moffitt.org Video Section Editor, International Brazilian of Urology Associate member, Department of GU Oncology Moffitt Cancer Center Associate Professor, Department of Urology University of South Florida Tampa, Florida, USA E-mail: philippe.spiess@moffitt.org 183
https://openalex.org/W4385809375	https://www.cambridge.org/core/services/aop-cambridge-core/content/view/FAE9B655E9F24282D52B5B8B45D696BE/S0022112023005566a.pdf/div-class-title-modelling-the-wall-friction-coefficient-for-a-simple-shear-granular-flow-in-view-of-the-degradation-mechanism-div.pdf	English	null	Modelling the wall friction coefficient for a simple shear granular flow in view of the degradation mechanism	Journal of fluid mechanics	2,023	cc-by	7,567	J. Fluid Mech. (2023), vol. 969, A7, doi:10.1017/jfm.2023.556 J. Fluid Mech. (2023), vol. 969, A7, doi:10.1017/jfm.2023.556 J. Fluid Mech. (2023), vol. 969, A7, doi:10.1017/jfm.2023.556 Modelling the wall friction coeﬃcient for a simple shear granular ﬂow in view of the degradation mechanism Cheng-Chuan Lin1,2, Riccardo Artoni3, Fu-Ling Yang4,† and Patrick Richard3 1Department of Mechanical Engineering, National Taipei University of Technology, 10608 Taipei, Taiwan 2Graduate Institute of Manufacturing Technology, National Taipei University of Technology, 10608 Taipei, Taiwan 3MAST-GPEM, Univ Gustave Eiffel, IFSTTAR, 44344 Bouguenais, France MAST-GPEM, Univ Gustave Eiffel, IFSTTAR, 44344 Bouguenais, France 4Department of Mechanical Engineering, National Taiwan University, 106319 Taipei, Taiwan 4Department of Mechanical Engineering, National Taiwan University, 106319 Taipei, Taiw (Received 21 February 2023; revised 5 June 2023; accepted 3 July 2023) A steady granular ﬂow experiment was performed in a conﬁned annular shear cell to examine how the wall friction coefﬁcient μw degrades from the intrinsic sliding friction coefﬁcient f between the grains and the container wall. Two existing models are invoked to examine the decay trend of μw/f in view of the ratio of shear velocity to the square root of granular temperature χ (Artoni & Richard, Phys. Rev. Lett., vol. 115, 2015, 158001) and the ratio of grain angular and slip velocities Ω (Yang & Huang, Granul. Matt., vol. 18, issue 4, 2016, p. 77), respectively. As both models correlate μw/f to different ﬂow properties, a hidden relation is speculated between χ and Ω, or equivalently, between the granular temperature and the grain rotation speed. We used experiment data to conﬁrm and reveal this hidden relation. From there, a uniﬁed μw/f −χ model is proposed with physical meanings for the model coefﬁcients and to show general agreement with the measured trend. Hence we may conclude that both the ﬂuctuations in grain translations and their mean rotation are the crucial yet equivalent mechanisms to degrade μw/f. Key words: dry granular material © The Author(s), 2023. Published by Cambridge University Press. 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, provided the original article is properly cited. 969 A7-1 1. Introduction A continuum description for dense granular ﬂows helps to predict ﬂow behaviours in geophysical hazards and industrial applications. The description requires both a constitutive equation and a boundary condition that is often assigned as a velocity https://doi.org/10.1017/jfm.2023.556 Pub † Email address for correspondence: fulingyang@ntu.edu.tw † Email address for correspondence: fulingyang@ntu.edu.tw © The Author(s), 2023. Published by Cambridge University Press. 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, provided the original article is properly cited. 969 A7-1 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard condition – no-slip (Mills, Loggia & Tixier 1999; Mills, Tixier & Loggia 2000; Ancey & Evesque 2000; Ancey 2001; Lin & Yang 2020) or slip (Hui et al. 1984; Jenkins & Richman 1986; Johnson & Jackson 1987; Richman 1988; Jenkins & Berzi 2010; Artoni & Santomaso 2014) – or a Coulomb-type tangential stress condition for a solid wall, τw = μwp. Here, a hydrostatic pressure p is assumed along with a constant effective wall friction coefﬁcient μw (Courrech du Pont et al. 2003; Taberlet et al. 2003; Jop, Forterre & Pouliquen 2005; Orpe & Khakhar 2007). While the hydrostatic p has been conﬁrmed in both experiments and discrete element (DE) simulations (Artoni et al. 2018; Lin, Jiang & Yang 2020), DE simulation reveals a non-constant μw that decays from the sphere-wall sliding friction coefﬁcient f towards the creeping regime in steady, simple shear and surface ﬂows (Richard et al. 2008; Brodu, Richard & Delannay 2013; Artoni & Richard 2015b; Artoni et al. 2018). To explain the weakening, Richard et al. (2008) consider how intermittent and random sphere ﬂuctuations render individual sphere-wall friction events in different directions that cancel each other out to give lower bulk friction. Later, Artoni & Richard (2015b) deﬁned χ = u∥/T∥, using the shear velocity u∥and the streamwise bulk granular temperature T∥, to indicate that a ﬂow at small χ experiences more incoherent grain dynamics. How μw degrades from f has been ﬁtted, with two constants (Artoni & Richard 2015b), as μw f = χB A + χB . (1.1) (1.1) From a different aspect, Shojaaee et al. (2012) simulated two-dimensional shear ﬂows to study how individual disk rotation may destroy pure sliding motion at the contact points to render μw lower than f. 1. Introduction Such a rotation-induced degradation of μw is also reported from a simulated three-dimensional granular avalanche by Yang & Huang (2016), together with a degradation model developed in view of the single-grain dynamics as follows. When a grain moves against the lateral wall at translation velocity upi and angular velocity ωpi, it would develop a total velocity at the contact point as upi + R × ωpi, with R denoting the radius of gyration from the grain centre. A non-zero ωpi would divert the surface sliding velocity, together with the generated Coulomb friction, away from upi. Hence the freedom of individual grain rotation provides a different friction-cancelling mechanism at the grain-size level that can accumulate to degrade μw mechanically according to the degree of rotation-induced diversion. The authors proposed a dimensionless rotation index Ω = \|ω\| D/2u∥with mean angular speed \|ω\| = ⟨\|ωpi\|⟩, and sphere diameter D. They proposed a different model as μw f = 1 + aΩ (1 + aΩ)2 + (bΩ)2 , (1.2) (1.2) with its coefﬁcients a = ⟨ω⊥⟩/\|ω\| and b = ⟨ω∥⟩/\|ω\| depending on the relative strength of the mean angular velocity components perpendicular and parallel to u∥, respectively. In fact, Louge (1994) also studies the sphere-wall friction force for a rapid granular ﬂow in simple shear. After normalizing the tangential stress by the normal stress at the wall, a non-constant μw/f is found to correlate well with a dimensionless relative velocity at the contact plane, r = (\|u∥\| + \|ω\| D/2)/2T∥. Though this work is conducted in a ﬂow regime different from that in the aforementioned investigations, it pioneers the concept of how different grain dynamics may affect the ﬂow boundary condition by linking all the three velocity scales involved in χ and Ω. However, we recall that the μw/f(r) relation is reported for the rapid ﬂow regime, unlike how (1.1) and (1.2) describe a monotonic decay 969 A7-2 Modelling wall friction coefﬁcient with degradation mechanism Modelling wall friction coefﬁcient with degradation mechanism (a) (b) (c) Mw H Z O Figure 1. Experiment facilities. (a) Annular shear cell ﬁlled with POM spheres to a height H and conﬁned by the top loading Mw. The blue rectangle marks the observation window. (b) The base bumpy wall was rotated at rate O. (c) A force sensor was mounted on the lateral wall of the cell. (a) Mw H Z O (b) (b) (c) ) Figure 1. 1. Introduction Experiment facilities. (a) Annular shear cell ﬁlled with POM spheres to a height H and conﬁned by the top loading Mw. The blue rectangle marks the observation window. (b) The base bumpy wall was rotated at rate O. (c) A force sensor was mounted on the lateral wall of the cell. of μw/f towards the creeping regime. Hence this work will focus on the two more recent models as an attempt to understand the degradation mechanisms for μw. Both (1.1) and (1.2) describe a monotonic decay of μw/f with the strength of the associated cancelling mechanism, which suggests a correlation between the model variable χ and Ω. Equivalently, we may infer a hidden relation between the microscopic grain ﬂuctuation and rotation dynamics. Hence this work aims to investigate experimentally the boundary stress of a conﬁned simply sheared granular ﬂow in view of the grain dynamics to broaden our understanding of the granular ﬂow boundary friction coefﬁcient. Section 2 describes the experiment set-up, while the image and data processing methods are presented in § 3, together with the measured bulk dynamics. Section 4 discusses the scaling relations between \|ω\| and T∥, and the measured data are used to consolidate a wall friction model. We conclude the work in § 5. 2. Experiment set-up and force measurement Experiments were conducted in the annular shear cell in ﬁgure 1(a), identical to that used in Artoni et al. (2018). We used 2.1 kg of polyoxymethylene (POM) spheres of diameter D = 5.9 ± 0.1 mm and density ρs = 1400 kg m−3 packed to height H = 15D. A top weight of mass Mw = 0.22, 1.1 and 5.4 kg was added but allowed to move in the vertical direction to apply constant loading stress over its projection area SA. The conﬁned spheres were driven by a bumpy bottom wall (see ﬁgure 1b) at different rates O = 5.9, 23.4 and 117.2 r.p.m., which combines with the three Mw to achieve nine ﬂow conditions, as summarized in table 1. A six-axis force sensor (ATI Nano17) was mounted to a cut wall piece of the same curvature and material to the shear cell container (shown in ﬁgure 1c). The sensing module was installed to a 20 mm × 20 mm square window centred at different heights z = 15, 30, 50 and 70 mm from the base to measure the local boundary force. The force signals were averaged over time for three repeated measurements under each loading and driving condition to estimate the mean stress components. The normal stress σrr in general follows the hydrostatic pressure proﬁle σrr(z) = Mwg/SA + ρg(H −z), as reported in 969 A7-3 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard System M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 Mw (kg) 0.22 0.22 0.22 1.1 1.1 1.1 5.4 5.4 5.4 O (r.p.m.) 5.9 23.4 117.2 5.9 23.4 117.2 5.9 23.4 117.2 Table 1. The details of the nine ﬂow conditions considered in the experiments. Table 1. The details of the nine ﬂow conditions considered in the experiments. ﬁgure 4(a) of Artoni et al. (2018). The wall tangential stress components in the shearing and the vertical directions, σrθ and σrz, were used to calculate an effective wall friction coefﬁcient by μw = (σrθ)2 + (σrz)2/\|σrr\|. We follow Savage & Hutter (1991) to conduct a set of sliding table tests to measure the intrinsic grain–wall sliding friction coefﬁcient f = 0.24 ± 0.01 (see the Appendix for more details). This value was used to scale μw before we present the depth proﬁle of μW/f in ﬁgure 2(a). 2. Experiment set-up and force measurement Degradation of μw/f away from the moving base is observed when the loading is sufﬁcient (Mw ≥1.1) to engage sphere dynamics to the base motion. With the lowest M1, the spheres may remain loosely in contact with each other, and a nearly constant μw/f was measured for z/D < 9 but starts to grow toward the top plate with μw/f > 1, suggesting the top load effect. As we aim to study the wall friction weakening, the data of these three cases with the lowest M1 will not be considered in the following analysis. 3. Image analysis and bulk dynamics To understand the depth variation of μw/f in view of (1.1) and (1.2), we obtain the needed velocity information from the individual grain dynamics. The sphere motion over the region boxed by the thick blue rectangle in ﬁgure 1(a) was recorded by a lateral high-speed camera (Phantom Miro 320S) at a rate of 24–3400 frames per second (FPS). A front LED lamp (EFFILux EFFISharp) was used to generate a bright reﬂection spot on each sphere surface to permit particle tracking, and two LED panels (FOTGA LED430) were placed aside to provide nearly uniform illumination. A snapshot of one recorded image is shown in ﬁgure 3(a). g ( ) The reﬂection spot on each sphere was nearly circular and can be located by the method of circular Hough transformation (Hough 1962) and found off-centred as the LED lamp was not aligned with the camera. We compared the bright spot positions to the manually located true centres in one image to evaluate a mean deviation distance and orientation, as r/D = 0.09 ± 6.1 × 10−3 and θ = 1.15 ± 0.09 rad from the horizontal, with both standard deviations (STDs) lower than 7%. We exploited this nearly uniform deviation to offset the bright spot locations to present the sphere centres, as shown by the red plus signs in ﬁgure 3(a). Refer to Lin & Yang (2018) for more details. As the illumination and camera set-up remained unchanged, we applied the same offset strategy throughout the image processing routine. The nearest neighbour method was applied to track each sphere centre in consecutive images to evaluate instantaneous particle velocity vectors upi (Yang & Huang 2016). To measure grain rotation, a nearly circular small black marker of diameter (d = 1 ± 0.21 mm) was spray-painted on each sphere. We followed Lin & Yang (2018) to extend a search circle of radius D/2 from each sphere centre to locate the associated marker pixels using a greyscale threshold. We used the mean position of these identiﬁed marker pixels to present the marker centroid, as shown by the green points in ﬁgure 3(a). The same nearest neighbour method was applied to locate the same marker in the subsequent image. 3. Image analysis and bulk dynamics 969 A7-4 969 A7-4 Modelling wall friction coefﬁcient with degradation mechanism g f fﬁ g 0.5 μw/f 1.0 1.5 0 5 10 15 (a) (b) (c) (d) 10–3 10–2 u∥/OR 10–1 100 0 5 10 15 10–1 10–4 0 5 z/D z/D 10 15 M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 100 101 0 5 10 15 T∥/(OR)2 \|ω\|/O Figure 2. Depth proﬁles of bulk (a) scaled effective wall friction coefﬁcient, (b) scaled streamwise velocity, (c) scaled streamwise granular temperature, and (d) scaled angular speed. A dashed horizontal line separates the shear zone (open symbol) and the creeping zone (ﬁlled symbol). 0.5 μw/f 1.0 1.5 0 5 10 15 (a) (b) 10–3 10–2 u∥/OR 10–1 100 0 5 10 15 z/D 0.5 μw/f 1.0 1.5 0 5 10 15 (a) z/D b) 10–3 10–2 u∥/OR 10–1 100 0 5 10 15 μw/f (d) 100 101 0 5 10 15 \|ω\|/O (c) 10–1 10–4 0 5 z/D 10 15 M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 T∥/(OR)2 \|ω\|/O Figure 2. Depth proﬁles of bulk (a) scaled effective wall friction coefﬁcient, (b) scaled streamwise velocity, (c) scaled streamwise granular temperature, and (d) scaled angular speed. A dashed horizontal line separates the shear zone (open symbol) and the creeping zone (ﬁlled symbol). Under the assumption that the rotation axis remains unchanged over a short observation duration of 10−2 s, we can ﬁt a plane to the arc swept by ﬁve consecutive marker positions in the least squares sense. The normal vector of this ﬁtted plane when forced to pass the instantaneous sphere centre found in the ﬁrst instant speciﬁes the desired rotation axis vector that can be normalized to unity magnitude. From this, we can extract the radius of gyration rpi for the marker-swept arc together with its tangential velocity Vθ,pi, and the division of the two scalars determines the angular speed \|ωpi\| = Vθ,pi/rpi. A further combination with the normalized rotation axis vector gives the desired angular velocity vector ωpi. More details regarding the measurement algorithms can be found in Lin & Yang (2018). https://doi.org/10.1017/jfm.2023.556 Pub To estimate a bulk property from these instantaneous sphere dynamics, an averaging routine was implemented as described below. An averaging box that spans width Lb = 6D 969 A7-5 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard C. C. Lin, R. 3. Image analysis and bulk dynamics Artoni, F. L. Yang and P. Richard z (a) (b) 1D Lb = 6D Hb = 2D z Figure 3. Portion of a raw image in greyscale. (a) The sphere (red plus) and marker (green asterisk) locations are marked. (b) The red and blue (moving 1D from the centre of the red box towards the bottom) boxes denote the averaging box used to extract the bulk properties. (b) 1D Lb = 6D Hb = 2D z z (a) (b) Figure 3. Portion of a raw image in greyscale. (a) The sphere (red plus) and marker (green asterisk) locations are marked. (b) The red and blue (moving 1D from the centre of the red box towards the bottom) boxes denote the averaging box used to extract the bulk properties. along the streamwise direction and a vertical span Hb = 2D was employed, as illustrated by the thin red rectangle in ﬁgure 3(b). The instantaneous information of the spheres that fell in the averaging box over an observation period was collected to compute a temporal mean to present a quasi-steady bulk property at the mid-height of the average box. Together, we calculated the standard deviation of the bulk value over the observation period. We took a 5 s observation period if FPS ≤1000, and changed to a 1 s period if FPS > 1000. Next, we shifted the averaging box by 1D across the observation window in ﬁgure 1(a) to evaluate the depth proﬁles of bulk dynamic properties to be discussed in the following. p p y p p g First, the bulk streamwise velocity was evaluated as u∥(z) = i up∥,i(t)Ai(t)/ i Ai(t), using the instantaneous velocity component parallel to the base up∥,i(t) of the ith particle and its projection area in the averaging box Ai(t). Here, Ai(t)/ i Ai(t) represents a weighting factor considering all the other particle information that fell within the averaging box at the same moment. The velocity perpendicular to the base, uz(z), nearly vanished across the depth for all ﬂow conditions and hence is not considered. Figure 2(b) shows the velocity depth proﬁle u∥(z) after being scaled by the base rim tangential velocity (OR). Distinctive decay rates in the velocity depth proﬁles are observed. We follow Artoni et al. 3. Image analysis and bulk dynamics (2018) to deﬁne a fast-moving shear zone by the segment showing exponential decay to zero next to the moving base, and identify a much slower segment with a reduced decay rate in the creeping zone near the top plate. We ﬁtted a line to each velocity segment and use their intersection to determine a transition height at z/D ≈9–11, as marked midway at z/D = 10. The velocity proﬁle can be well described by u∥(z)/OR = ub + (u0 −ub) exp(−z/δ), as reported in Artoni et al. (2018). Here, ub and u0 are the scaled velocities at the top and bottom, respectively, and δ is the decay length of the proﬁle. The depth proﬁle of the dominating streamwise granular temperature was evaluated next 969 A7-6 969 A7-6 Modelling wall friction coefﬁcient with degradation mechanism by T∥(z) = i up∥,i − u∥(zc) + (zi −zc) ∂u∥ ∂z zc 2 , (3.1) (3.1) where the velocity gradient at the averaging box centre zc was taken into account to offset the possible effect from the chosen averaging box height (Artoni & Richard 2015a). It is worth noting that the measurement errors of granular temperature could be discussed further in Xu, Reeves & Louge (2004). The scaled results, T∥/(OR)2, in ﬁgure 2(c) exhibit another sharp trend variation around the velocity transition height z/D = 10. Next, we examine the overall grain angular speed \|ω\| in ﬁgure 2(d) after scaling by O. As grain rotation is induced by unbalanced torque from its interaction with the neighbouring grains and the shear cell wall, it ﬂuctuates more severely in time than the properties deduced from grain translation motion. Hence we present the expected value of ωpi collected over the same time interval as that used for the velocity. All the \|ω\|/O values exhibited an exponential decay across the shear zone, and a trend conversion into the creeping zone around the same z/D = 10, similar to that observed on T∥(z)/(OR)2. For both \|ω\|/O and T∥(z)/(OR)2, an increase of the top load caused a steeper decay in the shear zone and a stronger variation in the creeping zone, compared to that found for u∥(z). This implies a hidden link between T∥and \|ω\| that has never been reported, and suggests an association between the micro-events considered in the two μw models. 3. Image analysis and bulk dynamics Finally, we also examined the solid volume fraction proﬁle φ(z, t) = i Ai(t)/LbHb (not shown) to conﬁrm a nearly consistent value at φ ≈0.7 across the creeping zone and most of the shear zone, while φ(z) drops near the moving base when z/D < 5 due to shear-induced dilatancy. The higher value of φ may be attributed to the ordering induced by the wall. However, the fact that the transition of u∥, T∥and \|ω\| occurs in the plateau of φ will support that the grain packing condition is not the primary cause for the phenomenon, but some other microscopic grain dynamics. 4.1. Scaling relations (a) Examination of the scaled angular speed D \|ω\|/2 with T∥in the creeping (inset) and fast shear zones. (b) Rotation index Ω versus χ. The contact dynamic simulation (CDM) data were adopted from Artoni & Richard (2015b), denoted by grey solid diamonds. velocity at the contact point (i.e. \|ω\| D/2u∥≪1) so that the grain–wall interaction is more like a pure sliding motion. Meanwhile, a small Ω suggests a large χ, which can also occur when the translation ﬂuctuation is negligible as compared to the bulk motion (i.e. u∥/T∥≫1). From both viewpoints, we may expect a μw closer to the sliding friction coefﬁcient between the grains and the wall. In contrast, when the strength of streamwise ﬂuctuation increases (χ < 1), the intermittent grain motions are promoted to cause ﬂuctuating interaction forces and unbalanced torque on each grain. The rotation index Ω rises accordingly to suggest how grain rotation diverts the sliding friction force at each contact to degrade bulk sliding. In an attempt to link these micro-mechanisms to other ﬂow variables, we recall that the streamwise ﬂow ﬂuctuation is often associated with the bulk shear strain rate \| ˙γ \|, which is a pertinent variable in understanding fast ﬂow behaviours. We evaluated \| ˙γ \| from u∥(z) and examine how the two variables vary with \| ˙γ \| in ﬁgure 5. We detect a clean power-law relation T∥∼\| ˙γ \|0.7 in the fast shear zone but the goodness of ﬁt is lost in the creeping regime. The current power exponent in the shear zone is greater than the values, 0.5 to 0.625, reported from a simulated conﬁned shear ﬂow but losing the constant exponent in the creeping regime is a shared feature (Richard et al. 2020, 2022). Further, it is worth noting that our power exponent falls in the range of 0.5–1.0 measured in the ﬂowing layer of a quasi-two-dimensional narrow rotating drum ﬂow (Orpe & Khakhar 2007) and below the exact 1.0 predicted for pure collisional ﬂows in simple shear (Jenkins & Savage 1983; Campbell 1990). The clean correlation between \|ω\| and T∥in ﬁgure 4(a) explains the similar variation trend for \|ω\| ∼\| ˙γ \|0.84 in the fast shear zone and the scatterings in the creeping zone. To the best of our knowledge, how \|ω\| varies with ˙γ at the lateral wall has never been reported. 4.1. Scaling relations First, we examine the D\|ω\|/2 as a function of T∥in ﬁgure 4(a), where the bulk variables are manipulated to possess identical dimension. A nearly linear correlation between the two variables is discovered in the creeping and fast shear zones. To provide further evidence, we also extracted the data from contact dynamics simulations of conﬁned shear ﬂows as presented in Artoni & Richard (2015a,b). The simulation results were processed in the same way as we analysed the experimental data and are shown by the grey solid diamonds in the plot of the shear zone in ﬁgure 4(a) to conﬁrm a similar trend. Fitting to each data group on the log-log plot gives the best-ﬁt slope, which diminishes slightly with the degree of streamwise ﬂuctuation (≈T∥), from slopes of approximately 1.2 and 1.05 to unity. The nearly linear \|ω\|–T∥relation across the two ﬂow zones suggests a direct correspondence of the two friction model parameters as shown in ﬁgure 4(b) where the trend is ﬁtted to all data as https://doi.org/10.1017/jfm.2023.556 Pub Ω = αχ−β, (4.1) (4.1) with α = 1.3 and a near unity β = 1.03. A much reduced Ω occurs when the translation velocity is large, which condition would suppress the effect of the rotation-induced with α = 1.3 and a near unity β = 1.03. A much reduced Ω occurs when the translation velocity is large, which condition would suppress the effect of the rotation-induced C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard 10–2 100 102 χ 10–1 100 D\|ω\|/2 101 102 (a) (b) 10–2 100 102 10–1 100 101 1 2 3 4 0 2 4 6 Ω 8 10 12 CDM M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 Creeping Slope = 1.0 Slope(CDM) = 1.05 Slope(Exp) = 1.2 T\|\| Figure 4. (a) Examination of the scaled angular speed D \|ω\|/2 with T∥in the creeping (inset) and fast shear zones. (b) Rotation index Ω versus χ. The contact dynamic simulation (CDM) data were adopted from Artoni & Richard (2015b), denoted by grey solid diamonds. 10–2 100 102 10–1 100 D\|ω\|/2 101 102 (a) 10–2 100 102 10–1 100 101 Creeping Slope = 1.0 Slope(CDM) = 1.05 Slope(Exp) = 1.2 T\|\| χ (b) 1 2 3 4 0 2 4 6 Ω 8 10 12 CDM M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 Figure 4. Modelling wall friction coefﬁcient with degradation mechanism 10–2 100 10–1 100 101 102 10–1 100 101 102 10–1 100 101 102 (a) 100 \|γ˙\|D \|γ˙\| \|ω\| 102 Slope = 0.7 10–2 100 10-1 100 101 102 103 100 102 Shear Shear Creeping Creeping Slope = 0.84 T\|\| (b) Figure 5. Comparison of (a) T∥and (b) \|ω\| with respect to bulk shear strain rate \| ˙γ \|, manipulated to give dimension consistency. 10–1 100 101 102 \|γ˙\| \|ω\| 10–2 100 10-1 100 101 102 103 100 102 Shear Creeping Slope = 0.84 (b) 10–2 100 10–1 100 101 102 10–1 100 101 102 (a) 100 \|γ˙\|D 102 Slope = 0.7 Shear Creeping T\|\| \|γ˙\|D \|γ˙\| Figure 5. Comparison of (a) T∥and (b) \|ω\| with respect to bulk shear strain rate \| ˙γ \|, manipulated to give dimension consistency. 10–5 10–2 I I 100 0 1 2 χ Ω 3 4 5 (a) (b) M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 10–5 10–2 100 0 2 4 6 8 10 12 Figure 6. Examination of (a) χ and (b) Ω with respect to the local inertial number. 10–5 10–2 I 100 0 1 2 χ 3 4 5 (a) M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 I Ω (b) 10–5 10–2 100 0 2 4 6 8 10 12 Figure 6. Examination of (a) χ and (b) Ω with respect to the local inertial number. This inertial number measures the tendency of grain streamwise motion due to \| ˙γ \| relative to its transverse settling under σrr. Hence a ﬂow at large I is in a well-ﬂuidized fast state so that u∥can be large and the ﬂuctuation-induced dynamic variations are comparably irrelevant in modifying bulk dynamics. This explains why ﬁgure 6 shows a general monotonic rise for χ but decay in Ω with increasing I. In particular, Ω saturates to a constant at large I to suggest a constant μw/f from (1.2). Likewise, the discovery that χ rises rapidly with I also leads to a limiting constant μw/f ≈1. These ﬁndings should explain why assigning a constant μw works for fast ﬂow prediction. 2 1 On the other hand, we notice a sudden rise of Ω for I ≲10−2–10−1, suggesting a variation in μw/f. The particular inertial number I = 10−2 has been marked to signal the transition from the quasi-static to the inertial ﬂow regime in Midi (2004). 4.1. Scaling relations We may conclude that \| ˙γ \| is no longer effective to characterize the particle level ﬂuctuations in either their translation (T∥) or rotation (\|ω\|) in the creeping zone. Some may worry about the further inﬂuence from the top load or, equivalently, the conﬁning normal stress, σrr, as their values can modify how inter-grain friction engages in a ﬂow. Considering the combined effect of \| ˙γ \| and σrr, a dimensionless inertial number I = D \| ˙γ \|/√σrr/ρs has been deﬁned and found useful to categorize dense granular ﬂow regimes (Midi 2004). 969 A7-8 969 A7-8 Modelling wall friction coefﬁcient with degradation mechanism In fact, Lu, Brodsky & Kavehpour (2007) also report a transitional regime in between the quasi-static and the inertial ﬂow regimes over 10−3 ≲I ≲10−1 (where I is converted here from the Savage number therein, which is exactly I2). Hence we may expect the transition from the shear (fast inertial) ﬂow regime to the creeping (transitional, towards the quasi-static) regime for the current ﬂow condition. 969 A7-9 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard 1.0 0.8 0.6 0.4 0.2 0 10–1 100 101 a¯ + a = –0.1322 b¯ + b = 0.1765 a¯ – a = –0.3653 a¯ = –0.2478, b¯ = 0.3145 b¯ + b = 0.4526 M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 Best fit to (1.1) Best fit to (4.2) a¯ < 0 b¯ > 0 ω∥ ω⊥ χ μw/f Figure 7. Comparison of the experimental data of μw/f–χ and the best ﬁtted models in (1.1) and (4.2). Upper and lower bounds of μw/f–χ are also portrayed, using the measured angular speeds as the model coefﬁcients. 1.0 0.8 0.6 0.4 0.2 0 10–1 100 101 a¯ + a = –0.1322 b¯ + b = 0.1765 a¯ – a = –0.3653 a¯ = –0.2478, b¯ = 0.3145 b¯ + b = 0.4526 M1O1 M1O2 M1O3 M2O1 M2O2 M2O3 M3O1 M3O2 M3O3 Best fit to (1.1) Best fit to (4.2) a¯ < 0 b¯ > 0 ω∥ ω⊥ χ μw/f Figure 7. Comparison of the experimental data of μw/f–χ and the best ﬁtted models in (1.1) and (4.2). Upper and lower bounds of μw/f–χ are also portrayed, using the measured angular speeds as the model coefﬁcients. Figure 7. Comparison of the experimental data of μw/f–χ and the best ﬁtted models in (1.1) and (4.2). Upper and lower bounds of μw/f–χ are also portrayed, using the measured angular speeds as the model coefﬁcients. Finally, we would like to comment on the correlation between χ, Ω and I. Judging from ﬁgure 6(a), it is clear that the χ–I data do not collapse onto a universal trend, and that the Ω–I data show a pronounced scatter in the creeping regime in ﬁgure 6(b), both contrasting the universality of the Ω–χ relation in ﬁgure 4(b). Hence we would like to suggest that at least one of T∥and \|ω\| should be considered in a boundary condition model, in addition to the commonly used I. 4.2. Wall friction model Finally, ﬁgure 7 examines the measured μw/f with χ, and a monotonic decay with decreasing χ is observed, conﬁrming the effect of how promoted ﬂuctuations, relative to bulk u∥, can degrade the bulk friction coefﬁcient from that of pure sliding. For the ﬁrst time in the literature, we conﬁrm in experiments that the μw/f–χ model in (1.1) can capture the general trend of wall friction weakening with the ﬁtted coefﬁcients A = 0.1 and B = 2.05. The decay trend of μw/f with decreasing χ is shown by the black solid line. However promising it is, one may be concerned that granular temperature is generally more difﬁcult to measure than the shear strain rate, and may hope to express T∥by the ﬁtted relation T∥∼\| ˙γ \|0.7 found in ﬁgure 5(a). However, such a nice correlation is not preserved in the creeping regime, where we detect the pronounced weakening of μw/f. Hence the efforts seem critical and indispensable unless we may extract other robust correlations for T∥–\| ˙γ \| in the creeping regime. The other μw/f–Ω model in (1.2) is proposed, with its two coefﬁcients a and b representing the mean angular velocity components in the directions perpendicular and parallel to u∥. It would be meaningful to see how well this model performs while keeping the dependence on the just-conﬁrmed parameter χ. Thus we further simplify (4.1) into Ω −χ−1 and rewrite (1.2) as Ω −χ−1 and rewrite (1.2) as μw f = 1 + ¯a/χ (1 + ¯a/χ)2 + (¯b/χ)2 . (4.2) 969 A7-10 μw f = 1 + ¯a/χ (1 + ¯a/χ)2 + (¯b/χ)2 . (4.2) (4.2) 969 A7-10 Modelling wall friction coefﬁcient with degradation mechanism A best ﬁt to the experimental data gives ¯a = −0.1 and ¯b = 0.39, and the new model shows equally good agreement, as portrayed by the grey dashed line. The negative ¯a indicates that ⟨ω⊥⟩(∼¯a) points in the direction opposing u∥as grains rotate under the action of streamwise friction at the wall contact points. However, it is surprising to observe a positive b ∼⟨ω∥⟩, which suggests that the individual grain–wall contact friction in the vertical direction did not cancel but showed a tendency towards the moving base even though the bulk transverse velocity is averaged to nearly zero. 4.2. Wall friction model This phenomenon may be attributed to the grain–wall friction coefﬁcient being much smaller than the grain internal friction coefﬁcient fint = 0.394. This coefﬁcient fint was estimated by the tangent of the static repose angle 21.5◦measured on a grain pile on the bed of the same experimental spheres. The grain–wall friction cannot compete with the inter-grain friction so that the grains adjacent to the container wall tended to rotate as sketched in the inset. To support this speculation, we extracted the angular speed components of individual grains and used their temporal and spatial averages to estimate the bulk angular velocity components ⟨ω⊥⟩and ⟨ω∥⟩. The mean values ¯a = −0.2478 and ¯b = 0.3145 agree in sign with the above-ﬁtted values, supporting the speculated mechanism further, and provide another pair of model coefﬁcients. In addition, the standard deviations from the mean values a = ⟨ω⊥⟩/\|ω\| and b = ⟨ω∥⟩/\|ω\| measure the strength variation in the microscopic events and can be used to estimate a set of model coefﬁcients. In ﬁgure 7, the green dotted line ﬁrst portrays the predicted μw/f trend using the mean (¯a, ¯b), which seems a bit off-trend with the measured data. However, if we take into account the deviation in the mean (¯a, ¯b), an upper bound of μw/f is found taking the highest ¯a + a and the lowest ¯b − b, as shown by the blue dotted line. Likewise, the orange dotted line gives the lower bound of μw/f with the smallest ¯a − a and the largest ¯b + b. It is interesting to note that the measured μw/f–χ data are enveloped by these two bounding curves, supporting the proposed model and the current understanding that grain-level ﬂuctuations – in either translation or rotation – are the key to degrading μw from f. 5. Conclusions This experimental work reports a non-constant lateral wall friction coefﬁcient that weakens with the distance to the moving boundary when a loaded conﬁned granular material was sheared at the base. To examine the data from the perspective of two existing models in (1.1) and (1.2), we measured the intrinsic sliding friction coefﬁcient between the grains and the container wall, f, bulk shear velocity u∥, streamwise granular temperature T∥, and averaged grain rotation speed \|ω\|. The independence of the two models suggests a correlation between the model parameters χ = u∥/T∥and Ω = \|ω\| D/2u∥. In turn, we discover a hidden correlation between \|ω\| and T∥across the ﬂow ﬁeld, even though the two variables actually exhibit distinctive depth variations in the shear and creeping regimes (separated at z/D ≈10), just like u∥. This ﬁnding further helps to consolidate the two μw/f models to propose (4.2) in terms of χ and physics-embedded model coefﬁcients. Finally, we also examine χ and Ω with respect to the local inertial number I and observe a sharp rise of Ω when I ≲10−2 in the creeping regime, suggesting that grain rotation or the equivalent translational ﬂuctuations may bring non-negligible microscopic mechanisms to be considered in the creeping ﬂow rheology. To further test the robustness of the proposed model, the same conﬁned shear ﬂows but with different packing heights H are highly desired, as the thicknesses of the shear and creeping zones may change accordingly. Moreover, investigation in other 969 A7-11 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard Top view Side view Downstream Upstream Sliding direction g θ A B C 120° × 3 (a) (b) Figure 8. Experimental set-up of the sliding table test. (a) The top and side views of the circular disk. Three spheres were inserted into the disk at positions A, B and C. (b) The sliding table: upstream and downstream positions are the initial placements of the disk. Downstream Upstream Sliding direction g θ (b) (b) Top view (a) Top view (a) Upstream Upstream Upstream Downstream Figure 8. Experimental set-up of the sliding table test. (a) The top and side views of the circular disk. Three spheres were inserted into the disk at positions A, B and C. (b) The sliding table: upstream and downstream positions are the initial placements of the disk. ﬂow conﬁgurations and an extension to the rapid granular ﬂow regime would also be intriguing. Author ORCIDs. Author ORCIDs. Cheng-Chuan Lin https://orcid.org/0000-0002-9085-5410; Riccardo Artoni https://orcid.org/0000-0001-9757-4489; Fu-Ling Yang https://orcid.org/0000-0002-6633-6311; Patrick Richard https://orcid.org/0000-0003-2380-6552. 5. Conclusions Further careful investigations into the correlation between χ, Ω and r in wider ﬂow conditions and their possible correlations with the inertial number are undoubtedly important. The ﬁndings will deepen our knowledge on the degradation mechanism for μw/f and help to advance the boundary condition model for granular ﬂows. Supplementary material. The experiment movies under the M1O1, M2O2 and M3O3 driving conditions are available in the supplementary material at https://doi.org/10.1017/jfm.2023.556. Funding. The authors would like to acknowledge ﬁnancial support from the National Science and Technology Council, Taiwan (109-2628-E-002-002-MY3, 111-2923-E-002-003) Campus France (40943RF). Declaration of interests. The authors report no conﬂict of interest. Declaration of interests. The authors report no conﬂict of interest. Modelling wall friction coefﬁcient with degradation mechanism Modelling wall friction coefﬁcient with degradation mechanism To reduce the possible effects from the set-ups, we alternated the orientation of the disk placement by positioning one of the glued spheres (labelled A, B, C) on the downstream side before the test. Once set in motion, the disk will slide as if the chosen sphere led the sliding down the plate, as illustrated in ﬁgure 8(b). We also alternated the plate placement by swapping the upstream and downstream sides. For each set-up, ﬁve repeated experiments were conducted and the obtained 30 fk values were averaged to give the mean value of 0.24 with a standard deviation of 0.01. REFERENCES ANCEY, C. 2001 Dry granular ﬂows down an inclined channel: experimental investigations on the frictional–collisional regime. Phys. Rev. E 65, 011304. g y ANCEY, C. & EVESQUE, P. 2000 Frictional–collisional regime for granular suspension ﬂows down an inclined channel. Phys. Rev. E 62, 8349–8360. ARTONI, R. & RICHARD, P. 2015a Average balance equations, scale dependence, and energy cascade for granular materials. Phys. Rev. E 91, 032202. ARTONI, R. & RICHARD, P. 2015b Effective wall friction in wall-bounded 3D dense granular ﬂows. Phys. Rev. Lett. 115, 158001. ARTONI, R. & SANTOMASO, A. 2014 Effective wall slip in chutes and channels: experiments and discrete element simulations. Granul. Matt. 16, 377–382. ARTONI, R., SOLIGO, A., PAUL, J.-M. & RICHARD, P. 2018 Shear localization and wall friction in conﬁned dense granular ﬂows. J. Fluid Mech. 849, 395–418. BRODU, N., RICHARD, P. & DELANNAY, R. 2013 Shallow granular ﬂows down ﬂat frictional chann ﬂows and longitudinal vortices. Phys. Rev. E 87, 022202. g y CAMPBELL, C.S. 1990 Rapid granular ﬂows. Annu. Rev. Fluid Mech. 22, 57–90. CAMPBELL, C.S. 1990 Rapid granular ﬂows. Annu. Rev. Fluid Mech. 22, 57–90. COURRECH DU PONT, S., GONDRET, P., PERRIN, B. & RABAUD, M. 2003 Wall effects on granular heap stability. Europhys. Lett. 61, 492–498. y p y , HOUGH, P.V.C. 1962 Method and means for recognizing complex patterns. US Patent 30696541962. & 1984 d di i f hi h h i HUI, K., HAFF, P.K., UNGAR, J.E. & JACKSON, R. 1984 Boundary conditions for high-shear grain ﬂows. J. Fluid Mech. 145, 223–233. JENKINS, J.T. & BERZI, D. 2010 Dense inclined ﬂows of inelastic spheres: tests of an extension of kinetic theory. Granul. Matt. 12, 151–158. JENKINS, J.T. & RICHMAN, M.W. 1986 Boundary conditions for plane ﬂows of smooth, nearly elastic, circular disks. J. Fluid Mech. 171, 53–69. JENKINS, J.T. & SAVAGE, S.B. 1983 A theory for the rapid ﬂow of identical, smooth, nearly elastic, spherical particles. J. Fluid Mech. 130, 187–202. p JOHNSON, P.C. & JACKSON, R. 1987 Frictional–collisional constitutive relations for granular mate application to plane shearing. J. Fluid Mech. 176, 67–93. JOP, P., FORTERRE, Y. & POULIQUEN, O. 2005 Crucial role of sidewalls in granular surface ﬂows: consequences for the rheology. J. Fluid Mech. 541, 167–192. q gy , LIN, C.-C., JIANG, M.-Z. & YANG, F.-L. 2020 Developing a rheological relation for transient dense granular ﬂows via discrete element simulation in a rotating drum. J. Mech. Appendix. Sliding table tests To measure the intrinsic grain–wall sliding friction coefﬁcient f, we conducted a sliding table test inspired by Savage & Hutter (1991). We designed an acrylic circular disk on which three experimental POM spheres were glued as the vertices of an equilateral triangle, as shown in ﬁgure 8(a). Special care was taken to ensure that the spheres were glued to the same height so that the disk can stand horizontally on these spheres. The disk was placed gently on an inclined plate made of poly(methyl methacrylate) (PMMA) in ﬁgure 8(b), identical to the material used for the cylinder wall of the shear cell. After we checked that the three spheres were in balanced contact with the plate, we slowly tilted the plate until the disk started to slide. At that moment, we recorded the inclined angle θk by the electronic level meter so that the sliding friction coefﬁcient can be estimated by fk = tan(θk). 969 A7-12 REFERENCES 36, 707–719. LIN, C.-C. & YANG, F.-L. 2018 A new image processing algorithm for three-dimensional angular velocity measurement and its application in a granular avalanche. Adv. Powder Technol. 29, 506–517. LIN, C.-C. & YANG, F.-L. 2020 Continuum simulation for regularized non-local μ(I) of dense granular ﬂows. J. Comput. Phys. 420, 109708. LOUGE, M.Y. 1994 Computer simulations of rapid granular ﬂows of spheres interacting with a ﬂat, frictional boundary. Phys. Fluids 6, 2253–2269. LU, K., BRODSKY, E.E. & KAVEHPOUR, H.P. 2007 Shear-weakening of the transitional regime for granular ﬂow. J. Fluid Mech. 587, 347–372. MIDI, G.D.R. 2004 On dense granular ﬂows. Eur. Phys. J. E 14, 341–365. g MILLS, P., LOGGIA, D. & TIXIER, M. 1999 Model for a stationary dense granular ﬂow along an inclined wall. Europhys. Lett. 45, 733–738. MILLS, P., TIXIER, M. & LOGGIA, D. 2000 Inﬂuence of roughness and dilatancy for dense granular ﬂow along an inclined wall. Eur. Phys. J. E 1, 5–8. g y ORPE, A.V. & KHAKHAR, D.V. 2007 Rheology of surface granular ﬂows. J. Fluid Mech. 571, 1–32. 969 A7-13 C.-C. Lin, R. Artoni, F.-L. Yang and P. Richard RICHARD, P., ARTONI, R., VALANCE, A. & DELANNAY, R. 2020 Inﬂuence of lateral conﬁ granular ﬂows: comparison between shear-driven and gravity-driven ﬂows. Granul. Matt. 22, 8 granular ﬂows: comparison between shear driven and gravity driven ﬂows. Granul. Matt. 22, 81. RICHARD, P., VALANCE, A., DELANNAY, R. & BOLTENHAGEN, P. 2022 Granular surface ﬂows conﬁned between ﬂat frictional walls Part 1 Kinematics J Fluid Mech 940 A30 RICHARD, P., VALANCE, A., DELANNAY, R. & BOLTENHAGEN, P. 2022 Granular surface ﬂows conﬁned between ﬂat, frictional walls. Part 1. Kinematics. J. Fluid Mech. 940, A30. RICHARD, P., VALANCE, A., MÉTAYER, J.-F., SANCHEZ, P., CRASSOUS, J., LOUGE, M. & DELANNAY, R. 2008 Rheology of conﬁned granular ﬂows: scale invariance, glass transition, and friction weakening. Phys. Rev. Lett. 101, 248002. RICHMAN, M.W. 1988 Boundary conditions based upon a modiﬁed Maxwellian velocity distributi of identical, smooth, nearly elastic spheres. Acta Mech. 75, 227–240. SAVAGE, S.B. & HUTTER, K. 1991 The dynamics of avalanches of granular materials from initiation to runout. Part I. Analysis. Acta Mech. 86, 201–223. y SHOJAAEE, Z., BRENDEL, L., TÖRÖK, J. & WOLF, D.E. 2012 Shear ﬂow of dense granular materials near smooth walls. II. Block formation and suppression of slip by rolling friction. Phys. Rev. E 86, 011302. TABERLET, N., RICHARD, P., VALANCE, A., LOSERT, W., PASINI, J.M., JENKINS, J.T. & DELANNAY, R. 2003 Superstable granular heap in a thin channel. Phys. Rev. Lett. 91, 264301. XU, H., REEVES, A.P. & LOUGE, M.Y. 2004 Measurement errors in the mean and ﬂuctuation velocities of spherical grains from a computer analysis of digital images. Rev. Sci. Instrum. 75, 811–819. YANG, F.-L. & HUANG, Y.-T. 2016 New aspects for friction coefﬁcients of ﬁnite granular avalanche down a ﬂat narrow reservoir. Granul. Matt. 18 (4), 77. 969 A7-14 969 A7-14
https://openalex.org/W2123350162	https://www.scielo.br/j/jmoea/a/gd7wf9qPvMcMPtLgydmvxWr/?lang=en&format=pdf	English	null	A compact wide slot antenna with dual band-notch characteristic for ultra wideband applications	Journal of Microwaves, Optoelectronics and Electromagnetic Applications	2,011	cc-by	3,470	55 55 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 A Compact Wide slot antenna with dual band- notch characteristic for Ultra Wideband Applications Cheng-yuan Liu1 and Tao Jiang1,2,3 1 College of Information and Communications Engineering, Harbin Engineering University, Harbin, Heilongjiang, 150001, P. R. China 2. Research Centre of War-Ship EMC, Harbin Engineering University， Harbin, Heilongjiang,150001, P. R. China 3．Research Centre of Telecommunication, Harbin Institute of Technology, Harbin, Heilongjiang, 150080, P. R. China {Liuchengyuan, jiangtao}@hrbeu.edu.cn Ying-song Li College of Information and Communications Engineering, Harbin Engineering University, Harbin, Heilongjiang, 150001, P. R. China liyingsong@hrbeu.edu.cn Abstract— A compact CPW-fed ultra-wideband antenna with dual band-notch characteristic is presented. Two notched frequency bands are obtained by embedding two U-shaped slots in the radiation patch and a rectangle slot in the ground plane. The two notched bands can be controlled by adjusting the length of the responding slots. The proposed antenna is successfully simulated, fabricated and tested. Experimental and numerical antenna shows that the proposed antenna with compact size of 21×28mm2, has an impedance bandwidth range from 3.1GHz to more than 11.0GHz for voltage standing-wave ratio less than 2, expect two notch band frequency 5GHz-6GHz for WLAN and 7.7GHz-8.5GHz for X-band for satellite communications in China. Index Terms—CPW-fed; ultra-wideband (UWB) antenna; dual band-notch; omni-directional antenna. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 p y Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accept Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag IS Index Terms—CPW-fed; ultra-wideband (UWB) antenna; dual band-notch; omni-directional antenna. Brazilian Microwave and Optoelectronics Society-SBMO p y Brazilian Society of Electromagnetism-SBMag I. INTRODUCTION With the development of the modern wireless communications, the ultra-wideband (UWB) systems have attached much attention recently because of its advantages including high speed data, small size, low cost, low complexity [1-4]. As the important part of the UWB systems, the antenna has received increased attention due to its impedance bandwidth, simple structure and omni-directional radiation pattern. Recently, a lot of UWB antennas have been realized for 3.1GHz-10.6GHz applications [1-8], such as spline-shaped antenna[1], diamond antennas [2-3], annual ring antenna [4], bow-tie antennas [5-6],triangular patch antennas [7], square monopole antenna with inverted T-Shaped notch in the ground plane [8]. However, several narrow bandwidth systems have been used for a long time, such as WLAN (5-6GHz) and X-band (7.7-8.5GHz). Therefore, plenty of UWB antennas with band notch antennas have been proposed for reducing the potential interference between UWB and narrow band received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 56 systems. But most of the proposed antenna only has one band-notch characteristics [9-13], such as pie slot antenna [9], U-slot antennas [10], parasitic elements[11], slots [12-14]. Three UWB antennas using SSRs [15], C-shaped parasitic strip and slots [16] and E-shaped slots have been realized [17]. But the allocation shapes have intensive influence to band notch characteristics. It is difficult to fabricate and adjust the central frequency of the notch band. In this paper, a compact CPW-fed UWB antenna with dual band notch characteristic is investigated numerically and experimentally. By using two U-shaped slots in the radiation patch and an rectangle slot in the CPW ground, two band-notched frequency will be appeared, which reduce the potential interference. The antenna was successfully optimized by Ansoft high frequency structure simulator (HFSS) 10, fabricated and tested. It is found that the designed antenna satisfies all the requirements in the UWB frequency band except 5-6GHz for WLAN and 7.7-8.5GHz for X-band. Details of the antenna design are presented herein and the measured voltage standing-wave ratio, radiation pattern and the gain are given. The article is divided as follows: Section II discusses the antenna model and the configuration; Section III gives the studies on the key parameters; Section IV shows the measured results of the VSWR, radiation pattern and the gain. Section V concludes the article. II. ANTENA DESIGN Fig.1 Geometry of the antenna. II. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag I. INTRODUCTION ANTENA DESIGN Fig.1 Geometry of the antenna. Fig.1 illustrates the geometry of the proposed CPW-fed UWB antenna with dual band notch characteristic. The antenna is printed on a substrate with relative permittivity 2.65, a loss tangent of 0.002 and a thickness of 1.6mm. The size of the antenna is 21×28mm2, and a 50Ω CPW feeding structure is employed. The notch bands of the proposed antenna are caused by a rectangle slot with width 0.8mm in the CPW ground and a simple square patch with two U-shaped slots with width 0.2mm. Two U-shaped slots which determine the notch band 7.7GHz-8.5GHz are etched in the radiation patch. The rectangle slot embedded in the CPW ground plays an important role in 5-6GHz. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 57 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 Fig.2 The photograph of the proposed antenna. Fig.2 The photograph of the proposed antenna. The photograph of the fabricated antenna is shown in Fig.2. In order to obtain the characteristic of the antenna, the current distributions of the proposed antenna have been investigated by using the Ansoft HFSS 10. Fig.3 Simulated current of the proposed antenna. Fig.3 Simulated current of the proposed antenna. Fig.3 (a) and (d) show the current distributions at 3.5GHz and 9.0GHz, respectively. The current distributions mainly flow along the CPW ground and the patch, while around the slots the current are small. On the contrary, in Fig.3 (b) and (c) the current distributions around slots are obtained at 5.5GHz and 8GHz. The current distributions are mainly flow though along the rectangle slot and the two U-shaped slots. Therefore, the surface current produced by the slots can excite the notch band frequencies. Brazilian Microwave and Optoelectronics Society-SBMO p y Brazilian Society of Electromagnetism-SBMag received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN III. PARAMETERS STUDIES The length of the embedded rectangle slot L2, the length of the U-shaped slots L3, the distance g between the radiation and the CPW ground and the dimension m of the square radiation patch have large effects on the proposed antenna. So they are selected to obtain the optimized results. In the investigated process, only one parameter is changed with other parameters fixed at one time. The effects of the parameters L2, L3, g and m on the VSWR vs. frequency are plotted in Fig.4. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 58 d Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 201 tromagnetism-SBMag © 2011 SBMO/SBMag (a) (b) (c) Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN (a) (b) (c) (a) (a) (b) (c) (c) Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism SBMag received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 59 (d) Fig.4 simulated VSWRs Vs. frequency with various parameters. (d) Fig.4 simulated VSWRs Vs. frequency with various parameters. (d) Fig.4 simulated VSWRs Vs. frequency with various parameters. Fig.4 simulated VSWRs Vs. frequency with various parameters. It can be seen from the Fig.4 (a), the rejection frequency can be changed from 5GHz to 6GHz by increasing length L2 of the rectangle slot. At the same time, the other rejection frequency at X-band is also changed. Form the Fig.3; we can see that changing current distribution at the X-band could alter rejection frequency. Therefore the length of the rectangle is the key parameter, then, it should be selected carefully in the design. Fig.4 (b) shows that with the increasing in the length of L3, the higher rejection frequency produced by U-shaped would moves to the lower level, while lower the rejection frequency only changed slightly at 5-6GHz. Fig.4 (c) indicates that the width g is a crucial factor for the rejection frequency at X-band. The higher rejection frequency can changed rapidly with the increasing of g. p Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Microwave and Optoelectronics Society SBMO ece ed 3 o 0 0; o e e 5 o Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepte Brazilian Society of Electromagnetism SBMag © 2011 SBMO/SBMag ISS received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 N III. PARAMETERS STUDIES The changed current distributions between the radiation and the CPW ground cause this situation. Fig.4 (d) describes the influence of the dimension of the square radiation patch m. With the increasing of the dimension of the radiation patch, the higher rejection frequency at X-band removes to the lower band. While the lower rejection frequency at 5-6GHz for WLAN changed slightly. The entire above can implies that the proposed antenna has two rejection frequencies .The lower at 5-6GHz for WLAN is mainly determined by the rectangle slot and the higher at 7.7-8.5GHz for X-band is caused by the U-shaped slots and the distance between radiation patch and CPW ground (g). As the electrical size has more effect on impedance bandwidth, they need optimized by tradeoff in this design. For the length of the slots have great influence on the notch band. The length of the embedded rectangle slot and the two U-shaped slots can be postulated as [16] 2 notch re c f L ε = (1) (1) where L is the total length of the U-shaped slots and rectangle slot, re ε is the effective dielectric constant, and c is the speed of light. We take (1) and the parameters studies above into consideration Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 60 in achieving the dimensions of the rectangle slot and the U-shaped slots at the beginning of the design and then adjust the geometry for the final design. Based on the parameters study above, the proposed antenna with dual band notch characteristics is optimized and manufactured after several adjustments of different parameters. The antenna is also optimized by using Ansoft HFSS 10. The optimized parameters of the antenna are as follow: L=28mm, W=21mm, L1=15 mm, W1=16.8 mm, L2=19mm, W2=0.8mm, L3=5.6mm, W3=2.8mm, m=10.8mm, S=0.3 mm, W4=1.4mm, g=1.2mm. p y Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN 2179-1074 received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN Brazilian Microwave and Optoelectronics Society SBMO rec Brazilian Society of Electromagnetism-SBMag © 2011 IV. RESULTS AND DISCUSSIONS To evaluate the performance of the optimized antenna, the proposed antenna was implemented and tested. The VSWR of the antenna is obtained by using the HP8757D vector network analyzer. In order to compare the simulation results of the antenna, the proposed antennas with rectangle slot and U-shaped slots and without all the slots are manufactured and measured. The VSWRs of the antennas are shown in Fig.5. Fig.5 the VSWR of the antenna with and without the slots. Fig.5 the VSWR of the antenna with and without the slots. From the Fig.5, the measured result is seen well agreed with the simulated result which helps to verify the design accuracy. The differences between the simulated and measured values may be due to the errors of the manufactured antenna and the SMA connector to CPW-fed transition, which is included in the measurements but not taken into account in the calculated results. Two notch bands characteristic are found from Fig.5. The radiation patters were measured in an anechoic chamber. The measured radiation patterns mainly at 3.5GHz, 6.5GHz and 9.5GHz are shown in Fig.6. It shows that the antenna can give a nearly omni-directional characteristic in the x-y plane and quasi omni- directional pattern in the x-z plane. As can be seen in Fig.6, the radiation patterns in the x-y plane deteriorate more or less with the frequency increasing, but the radiation patterns are still nearly omni- directional. The peak gains of the proposed antenna at these frequencies are achieved by compared to a wire dipole antenna. A stable gain can be obtained throughout the operation band expect the two notched frequencies. In order to compare, the proposed antenna without slots is also measured. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 ournal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 61 3.5GHz (x-z plane) 3.5GHz(x-y plane) d Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted ctromagnetism-SBMag © 2011 SBMO/SBMag ISSN 3.5GHz (x-z plane) 3.5GHz(x-y plane) 6.5GHz (x-z plane) Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 IV. RESULTS AND DISCUSSIONS 2010; accepted 3 June 2011 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN 2179-1074 3.5GHz (x-z plane) 3.5GHz(x-y plane) 6.5GHz (x-z plane) 3.5GHz (x-z plane) 3 5GH ( l ) 3.5GHz (x-z plane) 3.5GHz(x-y plane) 3.5GHz(x y plane) 6.5GHz (x-z plane) 6.5GHz (x-z plane) Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO B ili S i t f El t ti SBM Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 ournal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 62 6.5GHz(x-y plane) 9.5GHz (x-z plane) Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; romagnetism-SBMag © 2011 SBMO/SBMag 6.5GHz(x-y plane) 9.5GHz (x-z plane) 9.5GHz(x-y plane) nd Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 ectromagnetism-SBMag © 2011 SBMO/SBMag ISSN 2179-1074 6.5GHz(x-y plane) 9.5GHz (x-z plane) 9.5GHz(x-y plane) 6.5GHz(x-y plane) 6.5GHz(x-y plane) 9.5GHz (x-z plane) 9.5GHz (x-z plane) 9.5GHz(x-y plane) 9.5GHz(x-y plane) Brazilian Microwave and Optoelectronics Society-SBMO received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 Brazilian Society of Electromagnetism-SBMag © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag Brazilian Microwave and Optoelectronics Society-SBMO B ili S i f El i SBM received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 63 Fig.6 Measured radiation pattern of the proposed antenna. Fig.7 The gain of the antenna with and without slots. Fig.6 Measured radiation pattern of the proposed antenna. Fig.6 Measured radiation pattern of the proposed antenna. Fig.7 The gain of the antenna with and without slots. The gain of the proposed antenna with and without slot is shown in Fig.7. In the operation band, two sharp gains decreased in the vicinity of 5.5GHz and 8.0GHz. The gain drops to -3.6dBi and - 2.4dBi at the notch band, respectively. ACKNOWLEDGMENT This is partially supported by the National Nature Science Fund of China (No.60902014) , Nature Science Fund of Heilongjiang (No.2006F11), Core Young Teacher Fund of Harbin Engineering University (No.0812) . V. CONCLUSIONS A CPW-fed ultra-wideband antenna with dual band-notch characteristic is proposed for UWB applications. Dual stop band is achieved by cutting two U-shaped slots in the radiation patch and a rectangle in the CPW ground. The antenna is successfully optimized, fabricated, tested. The results show that the antenna not only has dual band notch characteristics but also has good radiation pattern. Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 Brazilian Microwave and Optoelectronics Society-SBMO Brazilian Society of Electromagnetism-SBMag [9] Chin-Ju Pan, Chungnan Lee,,Chih-Yu Huang, Hsiao-Cheng Lin, “band-notched ultra-wideband slot antenna”, Microwave And Optical Technology Letters , 48(12):2444-2446, 2006. Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 10, No.1, June 2011 p gy , ( ) , [10] Yuan, T., Qiu, C.W., Li L. W., Leong M. S.,and Zhang,Q., “Elliptically shaped ultra wideband patch antenna with band-notch features”, Microw. Opt. Technol. Lett. 50(1):736-738, 2008. REFERENCES [1] Lizzi,L.,Viani,F.,Azaro,R.,Massa,A., “A PSO-Driven Spline-Based Shaping Approach for Ultrawideband (UWB) Antenna Synthesis”, IEEE Trans. Antennas Propag,56(8):2613-2621, 2008. [1] Lizzi,L.,Viani,F.,Azaro,R.,Massa,A., “A PSO-Driven Spline-Based Shaping Approach for Ultrawideband (UWB) Antenna Synthesis”, IEEE Trans. Antennas Propag,56(8):2613-2621, 2008. [2] Lu, G., vonder Mark, S., Korisch, I., Greenstein, L. J. and Spasojevic, P., “Diamond and rounded diamond antennas for ultrawide-band communications”, IEEE Antennas Wireless Propagation Lett., 3(1): 249-252, 2004. [3] Koohestani M., and Golpour M., “Compact rectangular slot antenna with a novel coplanar waveguide fed dim for ultra wideband applications”, Microw. Opt. Technol. Lett., 52(1):331-334, 2010. [4] Ren, Y.J.,Chang,K., “An annual ring antenna for UWB communications”, IEEE Antennas Wireless Propag Lett, 5(1):274-276, 2006. ( ) [5] Karacolak, T.,Topsakal,E., “A double-sided rounded bow-tie antenna (DSRBA) for UWB communication”, IEEE Antennas Wireless Propag Lett, 5(1):446-449, 2006. [6] Kiminami, K., Hirata, A., Shiozawa, T., “Double-sided printed bow-tie antenna for UWB communications”, IEEE Antennas Wireless Propag. Lett., 3(1):152-153, 2004. [7] Choi, S. T., Hamaguchi, K., and Kohno, R., “Small printed CPW-fed triangular monopole antenna for ultra- applications”, Microw. Opt. Technol. Lett., 51(1):1180-1182, 2009. pp , p , ( ) , [8] Mohammad Ojaroudi, Changiz Ghobadi, Javad Nourinia , “Small Square Monopole Antenna With Inverted T-Shaped Notch in the Ground Plane for UWB Application” , IEEE Antennas And Wireless Propagation Letters, 8(1):728-731, 2009. received 3 Nov. 2010; for review 5 Nov. 2010; accepted 3 June 2011 © 2011 SBMO/SBMag ISSN 2179-1074 64 [9] Chin-Ju Pan, Chungnan Lee,,Chih-Yu Huang, Hsiao-Cheng Lin, “band-notched ultra-wideband slot antenna”, Microwave And Optical Technology Letters , 48(12):2444-2446, 2006. p gy , ( ) , [10] Yuan, T., Qiu, C.W., Li L. W., Leong M. S.,and Zhang,Q., “Elliptically shaped ultra wideband patch antenna with band-notch features”, Microw. Opt. Technol. Lett. 50(1):736-738, 2008. [11] A. M. Abbosh and M. E. Bialkowski, “Design of UWB Planar Band-Notched Antenna Using Parasitic Elements”, IEEE Transactions On Antennas And Propagation, 57(3):796-799, 2009. [12] Carla R. Medeiros, Jorge R. Costa, Carlos A. Fernandes, “Compact Tapered Slot UWB Antenna with WLAN Band Rejection”, IEEE Antennas and Wireless Propagation Letters, 8(1):661-664, 2009. [12] Carla R. Medeiros, Jorge R. Costa, Carlos A. Fernandes, “Compact Tapered Slo Rejection”, IEEE Antennas and Wireless Propagation Letters, 8(1):661-664, 2009. L.-H. Ye and Q.-X. Chu, “Improved band-notched UWB slot antenna”, Electronics Letters, 45(25): 1283-1285, 2009. [14] Joon-Won Jang and Hee-Yong Hwang, “An Improved Band-Rejection UWB Antenna with Resonant Patches and a Slot”, IEEE Antennas and Wireless Propagation Letters, 8(1): 299-302, 2009. REFERENCES p g [15] Zha, F. T., Gong S. X., Liu G., Yang H. Y., and Lin S. G., “compact slot antenna for 2.4GHz/UWB with d notched characteristic”, Microw. Opt. Technol. Lett., 48(1):1859-1862, 2009. [16] Chu, Q. X., and Yang, Y. Y., “A compact ultra wideband antenna with 3.4/5.5GHz dual band-notched characteristics”, IEEE Trans. Antennas Propag., 56(12):3637-3644, 2008. IEEE Trans. Antennas Propag., 56(12):3637-3644, 200 p g , ( ) , [17] Luo, L., Cui, Z.,Xiong J. P., Zhang, X. M., and Jiao, Y. C., “Compact printed ultra-wideband monopole antenna with dual band-notch characteristic”, Electron. Lett., 44(1):1106-1107, 2008.
https://openalex.org/W3036249939	https://www.mdpi.com/1424-8220/20/12/3512/pdf?version=1592732057	English	null	Piezoelectric Energy Harvesting Solutions: A Review	Sensors	2,020	cc-by	25,138	Received: 25 May 2020; Accepted: 18 June 2020; Published: 21 June 2020 Abstract: The goal of this paper is to review current methods of energy harvesting, while focusing on piezoelectric energy harvesting. The piezoelectric energy harvesting technique is based on the materials’ property of generating an electric ﬁeld when a mechanical force is applied. This phenomenon is known as the direct piezoelectric eﬀect. Piezoelectric transducers can be of diﬀerent shapes and materials, making them suitable for a multitude of applications. To optimize the use of piezoelectric devices in applications, a model is needed to observe the behavior in the time and frequency domain. In addition to diﬀerent aspects of piezoelectric modeling, this paper also presents several circuits used to maximize the energy harvested. Keywords: energy harvesting; piezoelectric materials; piezoelectric transducer types; modeling; frequency response; energy harvesting electronic circuits; SPICE simulation Sensors 2020, 20, 3512; doi:10.3390/s20123512 sensors Review Piezoelectric Energy Harvesting Solutions: A Review Corina Covaci * and Aurel Gontean Applied Electronics Department, Politehnica University Timisoara, 300006 Timisoara, Romania; aurel.gontean@upt.ro * Correspondence: corina.covaci@student.upt.ro Review Piezoelectric Energy Harvesting Solutions: A Review Corina Covaci * and Aurel Gontean Corina Covaci * and Aurel Gontean 1. Introduction In recent decades, wireless technologies and microelectronics have led to the development of wearable devices, such as items of clothing and accessories, in which power is supplied either by batteries or energy harvesting devices [1]. In conjunction with these approaches is the concept of the Internet of Things (IoT), where wireless sensors networks are commonly used [2]. IoT has led to smart equipment being placed in remote areas or in places where it can be diﬃcult or impossible to charge batteries (e.g., health care devices placed inside the human body, and smart buildings). Despite the progress made by low-power integrated circuit technology, the energy density of chemical batteries needs to be improved, since the power requirement for the mentioned applications is diﬃcult to fulﬁl [3,4]. Therefore, it is necessary to develop new energy harvesting techniques to sustain such self-powered systems. Not only is energy harvesting in sustaining self-powered systems necessary as a feasible and economically practical alternative to batteries, it also reduces the danger of greenhouse gas emission and sustains the environment [5]. Typically, an energy harvesting system has three parts [6]: • The energy source: represents the energy from which the electrical power will be scavenged—this energy can be ambient (available in the ambient environment, e.g., sunlight, ambient heat or wind) or external (energy sources that are explicitly deployed, e.g., lightning, human heat or vibrations) [7]; • The harvesting mechanism: consists of the structure which converts the ambient energy into electrical energy; • The load: the sink which consumes or stores the electrical output energy. • The load: the sink which consumes or stores the electrical output energy. The most common small-scale energy sources are sunlight, electromagnetic radiation, environmental mechanical energy, human body heat, and human body mechanical energy. Unlike solar energy, electromagnetic radiation, and environmental mechanical energy, which are highly dependent Sensors 2020, 20, 3512; doi:10.3390/s20123512 www.mdpi.com/journal/sensors www.mdpi.com/journal/sensors Sensors 2020, 20, 3512 2 of 37 on the environment, human body energy harvesters can be integrated into daily human activities to power a variety of devices [3,8]. Sensors 2020, 20, x FOR PEER REVIEW 2 of 36 Human body heat can be harvested using the principle of thermoelectric power generators, based on the Seebeck eﬀect of materials. Using this material’s principle, one can generate electrical energy using the diﬀerence between the human body and the ambient temperature. 1. Introduction In Figure 1 [9], the approximate working-frequency level for different mechanical energy sources is shown. Figure 1. Frequency level for different mechanical energy sources [9]. Figure 1. Frequency level for diﬀerent mechanical energy sources [9]. Figure 1. Frequency level for different mechanical energy sources [9]. Figure 1. Frequency level for diﬀerent mechanical energy sources [9]. Energy harvesting efficiency can be defined as the ratio between the power consumed on the external load resistance and the total input mechanical power. Mechanical efficiency is the converted electrical power divided by the total input mechanical power, and the electrical efficiency can be defined as the ratio between the power consumed on the external load resistance and the converted electrical power [12]. Mechanical energy 𝐸௠ (Equation (2)), electrical energy 𝐸௘ (Equation (3)) and energy conversion efficiency 𝐸% (Equation (4)) are defined using the following equations [13]: Energy harvesting eﬃciency can be deﬁned as the ratio between the power consumed on the external load resistance and the total input mechanical power. Mechanical eﬃciency is the converted electrical power divided by the total input mechanical power, and the electrical eﬃciency can be deﬁned as the ratio between the power consumed on the external load resistance and the converted electrical power [12]. Mechanical energy Em (Equation (2)), electrical energy Ee (Equation (3)) and energy conversion eﬃciency E% (Equation (4)) are deﬁned using the following equations [13]: 𝐸௠= න𝐹 𝑑(𝑡)𝑑𝑡 ∆௧ ଴ (2) 𝐸௘= 𝑃 ∆𝑡= 𝑉ଶ 𝑅 ∆𝑡 (3) 𝐸% = 𝐸௘ 𝐸௠ 100 (4) Em = Z ∆t 0 F d(t)dt (2) Ee = P ∆t = V2 R ∆t (3) E% = Ee Em 100 (4) (2) (2) (3) (3) (4) (4) 𝐸௠ where 𝐹 is the applied force, 𝑑 is the movement distance while the force is applied, ∆𝑡 is the generation time 𝑃is the output power 𝑉is the output voltage and 𝑅is the resistive load applied where F is the applied force, d is the movement distance while the force is applied, ∆t is the generation time, P is the output power, V is the output voltage, and R is the resistive load applied to the harvester. generation time, 𝑃 is the output power, 𝑉 is the output voltage, and 𝑅 is the resistive load applied to the harvester. A mechanical energy harvester can be used to harvest the energy generated by human walking. 1. Introduction The inconvenience is that a considerable diﬀerence in temperature is needed to have a stable system [3]. Human body heat can be harvested using the principle of thermoelectric power generators, based on the Seebeck effect of materials. Using this material’s principle, one can generate electrical energy using the difference between the human body and the ambient temperature. The Human body mechanical energy and environmental mechanical energy are widely exploited due to their abundance in daily life. Every motion in nature can be a potential source of kinetic energy. Therefore, mechanical energy is the most prevalent form of energy [9]. Using mechanical energy scavenging, suﬃcient power can be provided to ensure long-term autonomy for self-powered systems [1]. For example, around 10 mW can be harvested from the motion of the upper limbs, 1 mW can be obtained from a typing motion, breathing generates around 100 mW and, by walking, we generate up to 1 W [10]. The harvested power density Pres for mechanical energy depends on the motion frequency and magnitude, as shown in the resonance power Formula (1) [11]: inconvenience is that a considerable difference in temperature is needed to have a stable system [3]. Human body mechanical energy and environmental mechanical energy are widely exploited due to their abundance in daily life. Every motion in nature can be a potential source of kinetic energy. Therefore, mechanical energy is the most prevalent form of energy [9]. Using mechanical energy scavenging, sufficient power can be provided to ensure long-term autonomy for self-powered systems [1]. For example, around 10 mW can be harvested from the motion of the upper limbs, 1 mW can be obtained from a typing motion, breathing generates around 100 mW and, by walking, we generate up to 1 W [10]. The harvested power density 𝑃௥௘௦ for mechanical energy depends on the f d d h h l (1) (1) Pres = 4π3mf 3 resyZmax (1) p ( ) 𝑃௥௘௦= 4 𝜋ଷ 𝑚 𝑓௥௘௦ ଷ 𝑦 𝑍௠௔௫ (1) where m is the inertial mass, Zmax is the maximum displacement, fres is the resonance frequency, and y is the amplitude of vibration of the housing. where 𝑚 is the inertial mass, 𝑍௠௔௫ is the maximum displacement, 𝑓௥௘௦ is the resonance frequency, and 𝑦 is the amplitude of vibration of the housing. In Figure 1 [9], the approximate working-frequency level for diﬀerent mechanical energy sources is shown. 1. Introduction For this, there are two types of mechanical energy storage devices: flywheels (which have high energy densities, but also require a considerable amount of space and have complex structures) and springs (which have a low energy density but are simple and reliable) [3] Combining these harvesters with A mechanical energy harvester can be used to harvest the energy generated by human walking. For this, there are two types of mechanical energy storage devices: ﬂywheels (which have high energy densities, but also require a considerable amount of space and have complex structures) and springs (which have a low energy density, but are simple and reliable) [3]. Combining these harvesters with a ratchet, the mechanical harvested energy can be stored for later use. (which have a low energy density, but are simple and reliable) [3]. Combining these harvesters with a ratchet, the mechanical harvested energy can be stored for later use. There are three typical ways to convert mechanical energy into electrical energy: There are three typical ways to convert mechanical energy into electrical energy: electromagnetic, electrostatic/triboelectric, and piezoelectric [14]. If high eﬃciency in the mechanical-to-electrical energy Sensors 2020, 20, 3512 3 of 37 transfer is required, electromagnetic systems are the most suitable because they usually involve coils and magnets, but this also means bulky and complicated mechanisms [15]. The choice between the three methods is highly dependent on the application but, of these, piezoelectricity is the most widely studied [15]. If the application needs high voltage, high energy density, high capacitance, and little mechanical damping [16–18], piezoelectric energy harvesting is the solution, with the observation that piezoelectric materials can be brittle or rigid and can be toxic [19,20]. For applications that do not need outside sources, instead of piezoelectric transducers, electromagnetic devices can also be used [17]. They have high output current, low output impedance, and do not need contacts [18,21], but they usually have coil losses, low eﬃciency at low frequencies, and low output voltages [18,22]. Due to the power requirement, speed growth gears are added to meet desirable rotating velocity. Therefore, electromagnetic devices are not compatible with fabrication at the microscale level suitable for human body applications [23]. Triboelectric energy harvesting presents a multitude of advantages compared with piezoelectric and electromagnetic energy harvesting, such as high power density, high conversion, and device ﬂexibility [23,24]. 1. Introduction Still, it also has reliability and durability issues, and, in addition, its working mechanism is not yet fully understood [23,25]. Each energy harvesting method has advantages and disadvantages, and diﬀerent approaches for harvesting energy eﬀectively from human body motion are proposed in the literature. Techniques to increase eﬃciency for piezoelectric energy harvesting include: • Non linearity [26,27]; • Double pendulum system [28]; • Frequency up conversion [29]; • Circuit management [10,30]. Techniques to increase eﬃciency for electromagnetic energy harvesting include: • Sprung eccentric rotor [31]; • Frequency up conversion [32]; • Spring clockwork mechanism [33]; • Spring-less system [34]; • Non linearity [35]. Techniques to increase eﬃciency for triboelectric energy harvesting include: • Ultrathin ﬂexible single-electrode [36]; • Core-shell structure mechanism [37]; • Air-cushion mechanism [38]; Li id l l d [39] • Non linearity [26,27]; • Double pendulum system [28]; • Frequency up conversion [29]; • Circuit management [10,30]. Techniques to increase eﬃciency for electromagnetic energy harvesting include: • Double pendulum system [28]; • Frequency up conversion [29]; • Circuit management [10,30]. Techniques to increase eﬃciency for electromagnetic energy harvesting include: • Sprung eccentric rotor [31]; Techniques to increase eﬃciency for triboelectric energy harvesting include: • Ultrathin ﬂexible single-electrode [36]; • Ultrathin ﬂexible single-electrode [36]; • Core-shell structure mechanism [37]; • Air-cushion mechanism [38]; • Liquid metal electrode [39]. In Figure 2, the classiﬁcation of energy harvesting sources is represented. In Figure 2, the classiﬁcation of energy harvesting sources is represented. For self-powered systems, small-scale energy harvesters are ideal due to their advantages of small volumes, long lives, and low or non-existent need of maintenance [40]. For self-powered systems, small-scale energy harvesters are ideal due to their advantages of smal umes, long lives, and low or non-existent need of maintenance [40]. In this review, we will focus on piezoelectricity and on the methods to harvest piezoelectric energy. The paper is organized as follows: in Section 2, a brief introduction to piezoelectricity is presented, followed by a classiﬁcation of piezoelectric materials (Section 3) and a small piezoelectric transducers description (Section 4). Section 5 presents the parameters that need to be highlighted in piezoelectric modeling, while Section 6 shows the frequency response of a piezoelectric energy harvester. After the behavior of piezoelectric transducers is known, diﬀerent electronic circuits for energy harvesting are presented in Section 7. The paper ends with a brief mention of some applications from the literature, some personal contributions, and conclusions. 4 of 37 Sensors 2020, 20, 3512 Figure 2. Classification of energy harvesting sources. Figure 2. Classiﬁcation of energy harvesting sources. Figure 2. Classification of energy harvesting sources. Figure 2. Classiﬁcation of energy harvesting sources. For self-powe 2. Piezoelectricity small volumes, long lives, and low or non-existent need of maintenance [40]. In this review, we will focus on piezoelectricity and on the methods to harvest piezoelectric energy. The paper is organized as follows: in Section 2, a brief introduction to piezoelectricity is presented, followed by a classification of piezoelectric materials (Section 3) and a small piezoelectric transducers description (Section 4). Section 5 presents the parameters that need to be highlighted in piezoelectric modeling, while Section 6 shows the frequency response of a piezoelectric energy harvester. After the behavior of piezoelectric transducers is known, different electronic circuits for energy harvesting are presented in Section 7 The paper ends with a brief mention of some Briscoe and Dunn [41] deﬁned piezoelectricity as “electric charge that accumulates in response to applied mechanical stress in materials that have non-centrosymmetric crystal structures”, while Erturk and Inman [42] deﬁned piezoelectricity as “a form of coupling between the mechanical and electrical behaviors of ceramics and crystals belonging to certain classes”. The Greek origin of the word “piezoelectricity” is “squeeze or press” [5], which refers to the propriety of the piezoelectric materials to generate an electric ﬁeld when a mechanical force is applied, a phenomenon called the direct piezoelectric eﬀect [43]. energy harvesting are presented in Section 7. The paper ends with a brief mention of some applications from the literature, some personal contributions, and conclusions. 2. Piezoelectricity Briscoe and Dunn [41] defined piezoelectricity as “electric charge that accumulates in response to applied mechanical stress in materials that have non-centrosymmetric crystal structures”, while Erturk and Inman [42] defined piezoelectricity as “a form of coupling between the mechanical and The piezoelectric eﬀect is divided into two phenomena: the direct piezoelectric eﬀect and the converse piezoelectric eﬀect [42]. The property of some materials to generate an electric ﬁeld when a strain is applied (direct piezoelectric eﬀect) was discovered by Pierre and Jacques Curie in 1880 [5]. The converse or inverse piezoelectric eﬀect was mathematically deduced from the principles of thermodynamics a year later by Lippmann [44], and it states that a piezoelectric material will deform if an electric ﬁeld is applied to it [43]. These two eﬀects coexist in a piezoelectric material, therefore ignoring the presence of one eﬀect in an application would be thermodynamically inconsistent [45]. electrical behaviors of ceramics and crystals belonging to certain classes”. For self-powe 2. Piezoelectricity The Greek origin of the word “piezoelectricity” is “squeeze or press” [5], which refers to the propriety of the piezoelectric The electrical behavior of piezoelectric materials can be described using Hooke’s law [46]. The electrical behavior of a material is represented by Equation (5): (5) pp , p two phenomena: the direct piezoelectric effect and the D = ε E, (5) The piezoelectric effect is divided into two phenomena: the direct piezoelectric effect and the converse piezoelectric effect [42]. The property of some materials to generate an electric field when a strain is applied (direct piezoelectric effect) was discovered by Pierre and Jacques Curie in 1880 [5]. where D is the displacement of charge density, ε is the permittivity, and E is the applied electric ﬁeld strength. The piezoelectric effect is divided into two phenomena: the direct piezoelectric effect and the converse piezoelectric effect [42]. The property of some materials to generate an electric field when a strain is applied (direct piezoelectric effect) was discovered by Pierre and Jacques Curie in 1880 [5]. where D is the displacement of charge density, ε is the permittivity, and E is the applied electric ﬁeld strength. strain is applied (direct piezoelectric effect) was discovered by Pierre and Jacques Curie in 1880 The converse or inverse piezoelectric effect was mathematically deduced from the principle To deﬁne a system, Hooke’s Law states that: pp ( p ) The converse or inverse piezoelectric effect wa To deﬁne a system, Hooke’s Law states that: 44], and it states that a piezoelectric material will deform two effects coexist in a piezoelectric material, therefore S = s T, (6) 44], and it s two effect S = s T, orm efore (6) ignoring the presence of one effect in an application would be thermodynamically inconsistent [45]. The electrical behavior of piezoelectric materials can be described using Hooke’s law [46]. The electrical behavior of a material is represented by Equation (5): where S is the strain, s is the compliance, and T is the stress. Equations (5) and (6) are combined to form a new relationship: ignoring the presence of one effect in an application would be thermodynamically inconsistent [45]. The electrical behavior of piezoelectric materials can be described using Hooke’s law [46]. The electrical behavior of a material is represented by Equation (5): where S is the strain, s is the compliance, and T is the stress. For self-powe 2. Piezoelectricity Equations (5) and (6) are combined to form a new relationship: The electrical behavior of piezoelectric materials can be describ electrical behavior of a material is represented by Equation (5): where S is the strain, s is the compliance, and T is the stress. Equations (5) and (6) are combined to form a new relationship: The electrical behavior of piezoelectric materials can be described using Hooke s law [46]. trical behavior of a material is represented by Equation (5): Equations (5) and (6) are combined to form a new relationship: The electrical behavior of piezoelectric materials can be describ electrical behavior of a material is represented by Equation (5): Equations (5) and (6) are combined to form a new relationship: 𝐷= 𝜀 𝐸, (5) harge density, 𝜀 is the permittivity, and 𝐸 is the applied electric  {S} = h sEi {T} + [d] {E} {D} = h dti {T} + h εTi {E} (7) 𝐷= 𝜀 𝐸, harge density, 𝜀 is the permit  {S} = h sEi {T} + [d] {E} {D} = h dti {T} + h εTi {E} (5) (7) 5 of 37 Sensors 2020, 20, 3512 where [d] is the direct piezoelectric eﬀect matrix, h dti is the matrix which describes the converse piezoelectric eﬀect, E indicates that a zero electric ﬁeld, or a constant electric ﬁeld, is found in the system, T indicates a zero stress ﬁeld, or a constant stress ﬁeld across the system and t determines the transposition matrix. The four piezoelectric coeﬃcients, dij, eij, gij, hij are deﬁned, as shown in Equation (8):  dij = ∂Di ∂Tj E = ∂Sj ∂Ei T eij = ∂Di ∂Sj E = ∂Tj ∂Ei S gij = ∂Ei ∂Tj D = ∂Sj ∂Di T hij = ∂Ei ∂Sj D = ∂Tj ∂Di S (8)  dij = ∂Di ∂Tj E = ∂Sj ∂Ei T eij = ∂Di ∂Sj E = ∂Tj ∂Ei S gij = ∂Ei ∂Tj D = ∂Sj ∂Di T hij = ∂Ei ∂Sj D = ∂Tj ∂Di S (8) (8) where the ﬁrst terms are related to the direct piezoelectric eﬀect, and the second terms are related to the converse piezoelectric eﬀect. For self-powe 2. Piezoelectricity A simpliﬁed method to describe both the direct and converse piezoelectric eﬀects is represented Equation (9) [47]: ( D = d T + ε E S = s T + d E (9) (9) where S is the strain, T is the stress, E is the electric ﬁeld intensity, D is the electric displacement, d is the piezoelectric coeﬃcient, ε is the permittivity, and s is the elastic compliance. where S is the strain, T is the stress, E is the electric ﬁeld intensity, D is the electric displacement, d is the piezoelectric coeﬃcient, ε is the permittivity, and s is the elastic compliance. The performance of the piezoelectric energy harvesters is dependent on the transducer’s mechanical-electrical conversion eﬃciency E% [1]. The energy conversion eﬃciency for a piezoelectric transducer can be calculated using Equation (10) [48]: E% = Pout Pin 100 (10) (10) where Pout is the electrical output power, deﬁned in Equation (11), and Pin is the mechanical input power (Equation (12)). where Pout is the electrical output power, deﬁned in Equation (11), and Pin is the mechanical input power (Equation (12)). (11) Pout = vp ip (11) Pin = F v (12) (11) (12) (12) where vp is the overall voltage between the transducer’s electrodes, ip is the current ﬂowing through the piezoelectric transducer when the circuit is closed, F is the external mechanical force, and v is the speed of the moving object. where vp is the overall voltage between the transducer’s electrodes, ip is the current ﬂowing through the piezoelectric transducer when the circuit is closed, F is the external mechanical force, and v is the speed of the moving object. Another important phenomenon that must be taken into consideration when working with piezoelectric materials is the change of polarization under mechanical stress. Three factors can inﬂuence the direction and strength of the polarization: the orientation of polarization within the crystal, the crystal symmetry, and the stress applied by the mechanical deformation of the system. Any change in the polarization can be measured as the change in surface charge density at the crystal faces, and is measured in Cm−2 or, more commonly seen, µC/cm2. For self-powe 2. Piezoelectricity • It can be fabricated at both macro- and micro-scales [52]. It can be fabricated at both macro- and micro-scales [52]. It can be fabricated at both macro- and micro-scales [52]. Despite all of these advantages, piezoelectric energy harvesting systems also have a few disadvantages: Despite all of these advantages, piezoelectric energy harvesting systems also have a few disadvantages: Despite all of these advantages, piezoelectric energy harvesting systems also have a few disadvantages: Despite all of these advantages, piezoelectric energy harvesting systems also have a few disadvantages: • The power harvested is low [5] compared with other harvesting techniques (e.g., Thermoelec • The power harvested is low [5] compared with other harvesting techniques (e.g., Thermoelec • The power harvested is low [5] compared with other harvesting techniques (e.g., Thermoelectric Generator - TEG devices generate up to 125 W); • The power harvested is low [5] compared with other harvesting techniques (e.g., Thermoelectric Generator - TEG devices generate up to 125 W); e e a o E e i e ge e a e up o ); • The harvesters require rectification, maximum power extraction, and output voltage regulation [51]; • Due to their high generated voltage and low output current, piezoelectric energy harvesters are not always suitable to be used with low voltage CMOS process [54–56]. • The harvesters require rectiﬁcation, maximum power extraction, and output voltage regulation [51]; • Due to their high generated voltage and low output current, piezoelectric energy harvesters are not always suitable to be used with low voltage CMOS process [54–56]. e e a o E e i e ge e a e up o ); • The harvesters require rectification, maximum power extraction, and output voltage regulation [51]; D t th i hi h t d lt d l t t t i l t i h t • The harvesters require rectiﬁcation, maximum power extraction, and output voltage regulation [51]; • Due to their high generated voltage and low output current piezoelectric energy harvesters are In 1770, a French scientist, Abraham-Louis Perrelet, used the energy harvested from arm movements to design a completely autonomous, self-winding pedometer watch. This was the ﬁrst record in history when somebody harvested the energy generated by body movement. Centuries later, human-based kinetic energy generators remain unexploited [5]. A healthy person takes around 10,000 steps a day [15], therefore, human walking can generate signiﬁcant power if harvested [57]. For self-powe 2. Piezoelectricity Piezoelectric energy harvesting presents a multitude of advantages: Piezoelectric energy harvesting presents a multitude of advantages: Piezoelectric energy harvesting presents a multitude of advantages: Piezoelectric energy harvesting presents a multitude of advantages: • High energy and power density [5,49,51,52]; • Simple structure [5,49]; • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; • Ease of application [52]; • High energy and power density [5,49,51,52]; • Simple structure [5,49]; • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; • High energy and power density [5,49,51,52]; • Simple structure [5,49]; • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; • Ease of application [52]; • High energy and power density [5,49,51,52]; • Simple structure [5,49]; • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; • High energy and power density [5,49,51,52]; • High energy and power density [5,49,51,52]; p [ , ]; • It does not need an external voltage source [5,52]; • It does not need an external voltage source [5,52]; p • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • It does not need an external voltage source [5,52]; • Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5]; • Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; g Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5] Good scalability [51]; Piezoelectric materials can be meshed into hybrid materials to produce a broad range of voltages [5] Good scalability [51]; • Piezoelectric transducers have versatile shapes [51]; • Piezoelectric transducers have versatile shapes [51]; Piezoelectric transducers have versatile shapes [51]; Piezoelectric transducers have versatile shapes [51]; Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; M i l i i l h hi h C i ( h b hi h h Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; E f li ti [52] • Piezoelectric transducers can be easily incorporated in energy harvesting structures [53]; • Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; E f li i [52] Many piezoelectric materials have a high Curie temperature (the temperature above which the materials lose piezoelectricity) [53]; Ease of application [52]; It can be fabricated at both macro and micro scales [52] Ease of application [52]; • Ease of application [52]; • It can be fabricated at bot • Ease of application [52]; • It can be fabricated at both macro- and micro-scales [52]. For self-powe 2. Piezoelectricity A piezoelectric energy harvester has two basic parts [49]: A piezoelectric energy harvester has two basic parts [49]: • The mechanical module—generates electrical energy; • The mechanical module—generates electrical energy; • The mechanical module—generates electrical energy; • The electrical module—comprises an electrical circuit which converts and rectiﬁes the generated voltage. • The electrical module—comprises an electrical circuit which converts and rectiﬁes the generated voltage. Therefore, the eﬀectiveness of the energy harvester is not only dependent on the piezoelectric transducer, but also on its integration with the electrical circuit [49]. Piezoelectric energy harvesting systems are in general associated with three phases (Figure 3) [1,50] Sensors 2020, 20, 3512 transducer, but als Piezoelectric 6 of 37 gure 3) 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; [1,50]: 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; [ , ] 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the piezoelectric energy harvester structure and piezoelectric coeﬃcients; [ , ] 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the piezoelectric energy harvester structure and piezoelectric coeﬃcients; 1. Mechanical—mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the piezoelectric energy harvester structure and piezoelectric coeﬃcients; 1. Mechanical mechanical energy conversion: associated with the mechanical strength of the piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. For self-powe 2. Piezoelectricity Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the piezoelectric energy harvester structure and piezoelectric coeﬃcients; piezoelectric energy harvester under enormous stresses and the matching of mechanical impedance; 2. Mechanical—electrical energy conversion: involves the electromechanical coupling factor of the p gy p 3. Electrical—electrical energy conversion: comprising electrical impedance matching piezoelectric energy harvester structure and piezoelectric coefficients; 3 Electrical—electrical energy conversion: comprising electrical impedance matching 3. Electrical—electrical energy conversion: comprising electrical impedance matching piezoelectric energy harvester structure and piezoelectric coefficients; 3 Electrical—electrical energy conversion: comprising electrical impedance matching Figure 3. The three phases associated with piezoelectric energy harvesting. Figure 3. The three phases associated with piezoelectric energy harvesting. Figure 3. The three phases associated with piezoelectric energy harvesting. Figure 3. The three phases associated with piezoelectric energy harvesting. For self-powe 2. Piezoelectricity Human steps have a frequency between 1.2 Hz to 2.2 Hz when considering 1.4 m/s as an average walking speed [58]. If we consider that the human walking frequency is 1 Hz, power density for piezoelectric Sensors 2020, 20, 3512 Human steps hav alki eed [58 7 of 37 verage ity for transduction can be as high as 343 mW/cm3 (theoretical value) and 19 mW/cm3 (practical value) [5]. Therefore, in addition to providing a space for walking, the pavement can serve as a solution to recover and harvest the energy generated by walking [5]. piezoelectric transduction can be as high as 343 mW/cm (theoretical value) and 19 mW/cm (practical value) [5]. Therefore, in addition to providing a space for walking, the pavement can serve as a solution to recover and harvest the energy generated by walking [5]. The main challenge in developing a piezoelectric energy harvesting pavement is the low The main challenge in developing a piezoelectric energy harvesting pavement is the low frequency of walking compared with the high operating bandwidth of piezoelectric energy harvesting systems. To obtain the best results, the operating excitation frequency must be within the range of the resonant frequency of the harvester [16]. For this, a frequency up-conversion technique must be used [29]. g p g p gy g p frequency of walking compared with the high operating bandwidth of piezoelectric energy harvesting systems. To obtain the best results, the operating excitation frequency must be within the range of the resonant frequency of the harvester [16]. For this, a frequency up-conversion technique must be used [29]. In addition to the frequency up-conversion, this kind of energy harvesting also needs to extract the maximum power from the piezoelectric transducer to maximize the net energy ﬂowing into the storage device [51]. Since a piezoelectric transducer has a relatively large capacitive term with a low resonant frequency, extracting the maximum power requires an impractically large inductor (tens of hundreds Henry) [51]. A solution to this problem can be resistive matching [59,60] or using a nonlinear technique called Synchronized Harvesting on Inductor (SSHI) [61]. [ ] In addition to the frequency up-conversion, this kind of energy harvesting also needs to extract the maximum power from the piezoelectric transducer to maximize the net energy flowing into the storage device [51]. 3. Piezoelectric Materials 3. Piezoelectric Materials Figure 4 [41] presents the piezoelectric materials in the crystal classes. Figure 4 [41] presents the piezoelectric materials in the crystal classes. Figure 4 [41] presents the piezoelectric materials in the crystal classes. Figure 4 [41] presents the piezoelectric materials in the crystal classes. Figure 4. The relationship between piezo-, pyro-, and ferroelectric materials. Figure 4. The relationship between piezo-, pyro-, and ferroelectric materials. Figure 4. The relationship between piezo-, pyro-, and ferroelectric materials. Figure 4. The relationship between piezo-, pyro-, and ferroelectric materials. Non-centrosymmetric materials are materials lacking a center of inversion. There are 32 crystal classes of which 20 possess direct piezoelectricity, and 10 of these are polar crystals (in the absence of mechanical stress, they exhibit spontaneous polarization). These polar crystals will show pyroelectricity—in the presence of an oscillating thermal gradient, they will generate a charge. Moreover, the materials are ferroelectric if the dipole moment is reversible when a sufficiently large electric field is applied. Therefore, ferroelectric materials are also piezoelectric, but they exhibit semiconductor properties that are similar to the properties found in mechanically stressed piezoelectric materials [41]. Non-centrosymmetric materials are materials lacking a center of inversion. There are 32 crystal classes of which 20 possess direct piezoelectricity, and 10 of these are polar crystals (in the absence of mechanical stress, they exhibit spontaneous polarization). These polar crystals will show pyroelectricity—in the presence of an oscillating thermal gradient, they will generate a charge. Moreover, the materials are ferroelectric if the dipole moment is reversible when a suﬃciently large electric ﬁeld is applied. Therefore, ferroelectric materials are also piezoelectric, but they exhibit semiconductor properties that are similar to the properties found in mechanically stressed piezoelectric materials [41]. For self-powe 2. Piezoelectricity Since a piezoelectric transducer has a relatively large capacitive term with a low resonant frequency, extracting the maximum power requires an impractically large inductor (tens of hundreds Henry) [51]. A solution to this problem can be resistive matching [59,60] or using a nonlinear technique called Synchronized Harvesting on Inductor (SSHI) [61]. 3. Piezoelectric Materials 3. Piezoelectric Materials There are around 200 piezoelectric materials used in energy harvesting applications [62], found in four main categories: There are around 200 piezoelectric materials used in energy harvesting applications [62], found in four main categories: • Single crystals (Rochelle salt, lithium niobite, quartz crystals); • Single crystals (Rochelle salt, lithium niobite, quartz crystals); g y q y • Ceramics (barium titanate (BaTiO3), lead-zirconate-titanate (PZT), potassium niobate (KNbO3)); • Ceramics (barium titanate (BaTiO3), lead-zirconate-titanate (PZT), potassium niobate (KNbO3)); g y q y • Ceramics (barium titanate (BaTiO3), lead-zirconate-titanate (PZT), potassium niobate (KNbO3)); • Ceramics (barium titanate (BaTiO3), lead-zirconate-titanate (PZT), potassium niobate (KNbO3)); • Polymers (polylactic acid (PLA), polyvinylidene ﬂuoride (PVDF), co-polymers, cellulose and derivatives); ( ( ) ( ) p ( )) • Polymers (polylactic acid (PLA), polyvinylidene ﬂuoride (PVDF), co-polymers, cellulose and derivatives); • Polymer composites or nanocomposites (polyvinylidene ﬂuoride-zinc oxide (PVDF-ZnO), cellulose BaTiO3, polyimides-PZT) [63]. • Polymer composites or nanocomposites (polyvinylidene ﬂuoride-zinc oxide (PVDF-ZnO), cellulose BaTiO3, polyimides-PZT) [63]. Another classiﬁcation of piezoelectric materials is [64]: Sensors 2020, 20, 3512 8 of 37 • Naturally occurring: Quartz, Rochelle salt, Topaz, Tourmaline group; • Naturally occurring: Quartz, Rochelle salt, Topaz, Tourmaline group; • Synthetic: Barium titanate, lead titanate, lithium niobite, lead zirconat • Synthetic: Barium titanate, lead titanate, lithium niobite, lead zirconate titanate. The choice of a piezoelectric material depends not only on their piezoelectric properties and the functionality of the application sector but also on parameters such as design ﬂexibility, application frequency, and available volume [63]. Although quartz is more precise and has a high acoustic quality, it costs more than piezoceramic zirconate titanate ceramics and has a lower sensitivity, which limits the resolution of quartz charge mode sensors [65]. The most commonly used materials for piezoelectric energy harvesting devices used to be lead-based materials such as PZT. However, due to legislative measures regarding the toxicity of lead [66], PZT was replaced by other lead-free materials such as BaTiO3, which has a lower transduction eﬃciency [9]. Despite their better piezoelectric properties, some of the disadvantages associated with piezoelectric ceramics are: rigidity, brittleness, high density, lower voltage coeﬃcient, physical limitation in producing small-sized piezoelectric ceramics, and lack of design ﬂexibility [53,67,68]. Piezoelectric polymers are a better candidate for piezoelectric energy harvesting applications since they are mechanically ﬂexible, so they can withstand high strain. 3. Piezoelectric Materials 3. Piezoelectric Materials They also generate suitable voltage with suﬃcient power output, despite their low power density, and they can resist higher driving ﬁelds because they have a higher dielectric breakdown and possess maximum functional ﬁeld strength; they have a low fabrication cost, and the processing is faster compared with ceramic-based composites [9,53,63]. p [ ] Maamet et al. [9] presented the main characteristics of piezoelectric material types, as shown in Table 1 [63,66–82]. Table 1. Characteristics of piezoelectric material types [9]. Type Description and Characteristics Existing Solutions and Examples Single crystal materials • Monocrystals vertically grown on a substrate with Bridgeman method or Flux method, etc.; • Outstanding piezoelectric properties and are mostly used in sensors and actuators applications; • Depending on the growing technique, can have diﬀerent nanostructure forms. • Zinc-Oxide (ZnO); • Lead Magnesium Niobate (or PMN) based nanostructures: PMN-PT. Lead-based Piezoceramics • Polycrystalline materials with perovskite crystal structure; • High piezoelectric eﬀect and low dielectric loss; • Simple fabrication process, compatible with MEMS fabrication; • Highly toxic due to the presence of lead. Most common are modiﬁed or doped PZT, such as: Lead Magnesium Niobate-PZT (PMN-PZT), PZT-5A, Zinc Oxide enhanced PZT (PZT-ZnO), etc. Lead-free Piezoceramics • Non-toxic piezoceramics; • Have lower transduction eﬃciency; • Competitive lead-free materials are perovskite crystal structured type. • BaTiO3; • Bismuth Sodium Titanate (BNT-BKT); • Potassium Sodium Niobate (KNN) – based material: LS45, KNLNTS. Table 1. Characteristics of piezoelectric material types [9]. Table 1. Characteristics of piezoelectric material types [9]. Table 1. Characteristics of piezoelectric material types [9]. • Zinc-Oxide (ZnO); • Lead Magnesium Niobate (or PMN) based nanostructures: PMN-PT. 9 of 37 Sensors 2020, 20, 3512 Table 1. Cont. Type Description and Characteristics Existing Solutions and Examples Piezopolymers • Electroactive Polymer (EAP); • Flexible, non-toxic and light-weighted; • Small electromechanical coupling than piezoceramics; • Low manufacturing cost and rapid processing; • Biocompatible, biodegradable, and low power consumption compared to other piezoelectric materials. • Can be used for piezo-MEMS fabrication; • Polyvinylidene Fluoride (PVDS) derived polymers. Description and Characteristics ption and Characteristics Existing Solutions and Examples Existing Solutions and Examples The work of Mishra et al. [63] highlighted the most important parameters of piezoelectric materials, based on [83–85], and a brief synthesis is reproduced in Table 2. Table 2. Comparison between piezoelectric ceramics and piezoelectric polymers [63]. Table 2. Comparison between piezoelectric ceramics and piezoelectric polymers [63]. In Table 3, d31 is the piezoelectric constant, sE 11 is the elastic compliance at constant electric ﬁeld, εT 33/ε0 is the dielectric constant, where ε0 is the permittivity of free space, and ρ is the mass density. In Table 4, diﬀerent power levels are presented for several piezoelectric materials, considering the transducer’s volume and application frequency. 3. Piezoelectric Materials 3. Piezoelectric Materials Properties/Parameters Piezoelectric Ceramics (PZT) Piezoelectric Polymers (PVDF) Piezoelectricity High Low Acoustic impedance (106 kg m−2 s−1) 1 High (30) Low (2.7) Density (103 kg m−3) 7.5 1.78 Relative permittivity (ε/ε0) 1200 12 Piezoelectric strain constant (10−12 C N−1) d31 = 110, d33 = 225–590 d31 = 23, d33 = −33 Piezoelectric stress constant (10−3 V m N−1) 1 g31 = 10, g33 = 26 g31=216, g33 = −330 Electromechanical coupling factor 2 k31 = 30 k31 = 12 Dielectric constant 1180 10–15 Mechanical ﬂexibility 1 Poor Outstanding Curie temperature (◦C) 386 80 1 Exceptional properties of PVDF for energy harvesting application vis-à-vis PZT ceramics. 2 % at 1 kHz. Erturk and Inman [86] emphasized the diﬀerence between several ceramics (PZT-5A and PZT-5H) and single crystals (PMN-PT and PMN-PZT) in Table 3. Table 3. Properties of interest for various ceramics and single crystals [86]. Table 3. Properties of interest for various ceramics and single crystals [86]. Piezoelectric Material d31(pm/V) sE 11(pm2/N) εT 33/ε0 ρ (kg/m3) PZT-5A −171 16.4 1700 7750 PZT-5H −274 16.5 3400 7500 PMN-PT (30% PT) −921 52 7800 8040 PMN-PT (33% PT) −1330 69 8200 8060 PMN-PZT −2252 127 5000 7900 Average −989.6 56.2 5220 7850 In Table 3, d31 is the piezoelectric constant, sE 11 is the elastic compliance at constant electric ﬁeld, εT 33/ε0 is the dielectric constant, where ε0 is the permittivity of free space, and ρ is the mass density. In Table 4, diﬀerent power levels are presented for several piezoelectric materials, considering the transducer’s volume and application frequency. In Table 3, d31 is the piezoelectric constant, sE 11 is the elastic compliance at constant electric ﬁeld 11 ε0 is the dielectric constant, where ε0 is the permittivity of free space, and ρ is the mass density. p y p ρ y In Table 4, diﬀerent power levels are presented for several piezoelectric materials, considering the sducer’s volume and application frequency. 10 of 37 Sensors 2020, 20, 3512 Table 4. Generated peak power for piezoelectric materials. 3. Piezoelectric Materials 3. Piezoelectric Materials Piezoelectric Material Peak Power (mW) Volume Frequency (Hz) PVDF [87] 0.61 72 × 16 × 0.41 mm 2 PZT ceramic [88] 52 1.5 cm3 100 PZT ﬁber [89] 120 2.2 cm3 - PMN-PZT single crystal [90] 0.015 20 × 5 × 0.5 mm 1744 PMN-PT single crystal [91] 3.7 25 × 5 × 1 mm 102 Erturk and Inman [92] also classiﬁed piezoelectric materials as hard single crystals/ceramics and soft single crystals/ceramics. Hard single crystals/ceramics have the mechanical damping with one order of magnitude less than their counterparts, but they produce power with one order of magnitude higher than soft single crystals/ceramics for excitations at their respective resonance frequencies. Despite its toxicity and its low conversion eﬃciency, lead zirconate titanate (PZT) is the most popular piezoelectric ceramic material [42]. PZT was developed in the 1950s at the Tokyo Institute of Technology, and its versions PZT-5A and PZT-5H are the most widely implemented piezoelectric ceramic materials [93]. The popularity of PZT is due to the fact that it is one of the most eﬃcient and cost-acceptable materials [15]. To improve conversion eﬃciency, two mechanisms of force ampliﬁcation frames are oﬀered: concave shape [94] and convex shape [95]. For further improvement of the conversion eﬃciency, Wang et al. [96] used a multilayer piezoelectric stack with a ﬂexure-free convex force ampliﬁcation frame. Sensors 2020, 20, x FOR PEER REVIEW 10 of 36 cost-acceptable materials [15]. To improve conversion efficiency, two mechanisms of force amplification frames are offered: concave shape [94] and convex shape [95]. For further improvement of the conversion efficiency, Wang et al. [96] used a multilayer piezoelectric stack with a flexure-free convex force amplification frame Sensors 2020, 20, x FOR PEER REVIEW 10 of 36 cost-acceptable materials [15]. To improve conversion efficiency, two mechanisms of force amplification frames are offered: concave shape [94] and convex shape [95]. For further improvement of the conversion efficiency, Wang et al. [96] used a multilayer piezoelectric stack with a flexure-free convex force amplification frame. 4. Types of Transducers 4 Types of Transducers 4. Types of Transducers Piezoelectric transducers can be found in diﬀerent shapes: 4. Types of Transducers Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in diﬀerent shapes: 4. Types of Transducers Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in diﬀerent shapes: 4. Types of Transducers Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in different shapes: Piezoelectric transducers can be found in diﬀerent shapes: • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Piezoelectric transducers can be found in different shapes: • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Figure 5. Cantilever beam transducer. (a) (b) Figure 6. Circular diaphragm transducer: (a) Front view; (b) Side view. Figure 5. Cantilever beam transducer. Piezoelectric transducers can be found in different shapes: • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Figure 5. Cantilever beam transducer. (a) (b) Figure 6. Circular diaphragm transducer: (a) Front view; (b) Side view. Figure 6. Circular diaphragm transducer: (a) Front view; (b) Side view. • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Piezoelectric transducers can be found in different shapes: • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Figure 5. Cantilever beam transducer. Figure 5. Cantilever beam transducer. • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Figure 5. Cantilever beam transducer. • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8) • Cantilever beam (Figure 5); • Circular diaphragm (Figure 6); • Cymbal type (Figure 7); • Stack type (Figure 8). Figure 5. Cantilever beam transducer. Figure 5. Cantilever beam transducer. g Figure 5. Cantilever beam transducer. Figure 5. Cantilever beam transducer. 4. Types of Transducers 4 Types of Transducers 4. Types of Transducers g (a) (a) (b) (b) (b) (b) (a) (a) (a) (b) Figure 6 Circular diaphragm transducer: (a) Front view; (b) Side view Figure 6. Circular diaphragm transducer: (a) Front view; (b) Side view. Figure 6. Circular diaphragm transducer: (a) Front view; (b) Side view. 11 of 37 Sensors 2020, 20, 3512 Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 8. Stack piezoelectric transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 7. Cymbal transducer. Figure 8. Stack piezoelectric transducer. The cantilever beam structure consists of a thin piezoelectric layer (or two layers) and a non-piezoelectric layer (usually a conductive metallic layer) ﬁxed at one end to achieve a structure operating in its ﬂexural mode (as shown in Figure 5), and is the most widely used due to its simple geometry and generation of the maximum amount of strain. If only one piezoelectric layer is bonded to the metallic layer, the conﬁguration is called “unimorph”; if there are two piezoelectric layers, the conﬁguration is called “bimorph”. The bimorph structure is more popular in piezoelectric energy harvesting devices because it doubles the electrical energy output, without making any remarkable changes in the device volume [63]. The circular diaphragm structure consists of a thin disk-shaped piezoelectric layer attached to a metal shim ﬁxed on the edges of the clamping ring, as shown in Figure 6. At the core of the diaphragm is attached a proof mass to intensify the performance under low-frequency operation and to improve the power output [63]. Cymbal transducers consist of a piezoelectric layer placed between two metal end caps on both sides (Figure 7), and they are useful in applications with higher impact forces. Applying axial stress on the metal end caps, the stress is ampliﬁed and converted into radial stress, which leads to a higher piezoelectric coeﬃcient and, therefore, higher charge generation from the piezoelectric energy harvester [97]. Stack piezoelectric transducers consist of multiple piezoelectric layers stacked over each other in a way that the poling direction of the layers aligns with the applied force (as shown in Figure 8), and are used in applications which demand high pressure [63]. The advantages and disadvantages of diﬀerent types of piezoelectric transducers are summarized in Table 5 [63]. 5. Piezoelectric Modeling For piezoelectric energy harvesting modeling, it is essential that the model represents both direct and converse piezoelectric eﬀects. Therefore, it must show both forward and feedback interaction between the electrical and mechanical domains [48]. In addition, the model should not only represent the piezoelectric transducer, but it also should be integrated with the electrical circuit [49]. Hence, for design optimization and comprehension of the behavior, it is important to use an electromechanical model [101]. 4. Types of Transducers 4 Types of Transducers 4. Types of Transducers For vibration energy harvesting, the most studied piezoelectric structures are unimorph and bimorph cantilever conﬁgurations [53]. For example, Rundy et al. [14] used a 1.75 cm PZT bimorph cantilever to harvest energy from low-level vibrations. The harvester’s natural frequency was 100 Hz, and they attached a proof mass to the tip of the cantilever to lower the resonance frequency. In this way, they obtained 60 µW of power. In another example, Sodano et al. [98] used a 63.5 × 60.3 × 0.27 mm PZT-5H cantilever on an electromagnetic shaker at 50 Hz charging a 1000 mAh NiMH rechargeable battery to 90% of its capacity within 22 h (or 160 mW average power). 12 of 37 12 of 37 Sensors 2020, 20, 3512 Table 5. Advantages and disadvantages of diﬀerent conﬁgurations for piezoelectric transducers. Type of Conﬁguration Features/Advantages Disadvantages Unimorph/Bimorph cantilever beam • Simple structure • Low fabrication cost • Lower resonance frequency • Power output is proportional to proof mass • High mechanical quality factor • Inability to resist a high impact force Circular diaphragm • Compatible with pressure mode operation • Stiﬀer than a cantilever of the same size • Higher resonance frequencies Cymbal transducer • High energy output • Withstands high impact force • Limited to applications demanding high-magnitude vibration sources Stacked structures • Withstands high mechanical load • Suitable for pressure mode operation • Higher output from d33 mode • High stiﬀness Table 5. Advantages and disadvantages of diﬀerent conﬁgurations for piezoelectric transducers. Jasim et al. [99] used another type of transducer to harvest the energy produced by trucks driving on the urban expressway Route 55 in New Jersey. In a previous paper [100], they optimized the geometry considering the balance of energy harvesting and mechanical stress, resulting in a bridge transducer. The bridge transducer consists of a square PZT5X disk placed between two metal end caps and was designed to have a layered poling pattern and electrode conﬁguration that increased the piezoelectric coeﬃcient and capacitance, producing 2.1 mW under repeated loading of 270 kg. 5.1. Piezoelectric Working Modes When a piezoelectric energy harvester is modeled, one of the characteristics that must be represented is the working mode. There are three piezoelectric working modes [102,103]: • Transverse mode (d33 mode); • Longitudinal mode (d31 mode); • Piezotronic mode. • Piezotronic mode. In general, piezoelectric transducers consist of several layers of piezoelectric, elastic, conducting, or insulating materials. When an electric potential is applied between the conducting layers, an actuation force is produced in the piezoelectric layers, and the direction of stress-strain components and the electric ﬁeld indicate the modes of operation (transverse or longitudinal), as shown in Figure 9 [16]. Sensors 2020, 20, 3512 force is produced in field indicate the modes of operation (transverse or longitudinal), as shown in Figure 9 [16]. (a) (b) Figure 9. Working mode: (a) d33 mode (transverse mode) (b) d31 mode (longitudinal mode). Figure 9. Working mode: (a) d33 mode (transverse mode) (b) d31 mode (longitudinal mode). ngitudinal), as shown in Figure 9 [16]. (b) p (a) (a) (b) Figure 9. Working mode: (a) d33 mode (transverse mode) (b) d31 mode (longitudinal mode). Figure 9. Working mode: (a) d33 mode (transverse mode) (b) d31 mode (longitudinal mode). In Figure 9a, the standard form of a piezoelectric transducer working in transverse mode is shown. An example of a piezoelectric transducer working in d33 mode is the stack type transducer. The working mode is based on the fact that the transducer has several piezoelectric thin films connected in parallel, and the mechanical connection is serial, layered upon each other. Because of the thin films mechanically connected in series, the total displacement of the piezoelectric stack transducer is the product between the total number of films and the movement of each film [16]. In Figure 9a, the standard form of a piezoelectric transducer working in transverse mode is shown. An example of a piezoelectric transducer working in d33 mode is the stack type transducer. The working mode is based on the fact that the transducer has several piezoelectric thin ﬁlms connected in parallel, and the mechanical connection is serial, layered upon each other. Because of the thin ﬁlms mechanically connected in series, the total displacement of the piezoelectric stack transducer is the product between the total number of ﬁlms and the movement of each ﬁlm [16]. Longitudinal mode (presented in Figure 9b) was widely studied using cymbal, unimorph, and bimorph transducers. In this mode, piezoelectric thin films are fixed upon a supporting structure and placed between the electrode layers. When an electric field is applied in a vertical direction, a stress/strain is produced along the horizontal direction. • Piezotronic mode. Longitudinal mode (presented in Figure 9b) was widely studied using cymbal, unimorph, and bimorph transducers. In this mode, piezoelectric thin ﬁlms are ﬁxed upon a supporting structure and placed between the electrode layers. When an electric ﬁeld is applied in a vertical direction, a stress/strain is produced along the horizontal direction. The last working mode, piezotronic, appeared with the discovery of ZnO nano wires with n-type conductivity [104]. In piezotronic working mode, due to the piezoelectric potential, a Schottky barrier is created between the piezoelectric nanowire and the electrode that regulates the flow of electrons [103]. The last working mode, piezotronic, appeared with the discovery of ZnO nano wires with n-type conductivity [104]. In piezotronic working mode, due to the piezoelectric potential, a Schottky barrier is created between the piezoelectric nanowire and the electrode that regulates the ﬂow of electrons [103]. 5.2. Input–Output Dependency A piezoelectric energy harvester can be modeled like a black box, where the dependency between the input and output characteristics is known. For example, the piezoelectric transducer’s thickness is closely related to the generated output voltage [105], therefore a piezoelectric energy harvester model should represent this dependency and highlight the critical thickness where the output voltage becomes extremely small [68]. Such a dependency is presented in [106], where the authors deﬁned the open circuit output voltage V of a piezoceramic based energy harvester. Because the environment’s frequency range is less than 100 Hz, which is outside the frequency range of a piezoceramic transducer [98,107], they deﬁned the output voltage in the oﬀ-resonance condition as [106]: V = E t = −g F t A (13) (13) where E is the electric ﬁeld, t is the piezoceramic’s thickness, A is the piezoceramic’s area, and g is the piezoelectric voltage coeﬃcient, expressed as in Equation (14): where E is the electric ﬁeld, t is the piezoceramic’s thickness, A is the piezoceramic’s area, and g is the piezoelectric voltage coeﬃcient, expressed as in Equation (14): g = d ε0εr (14) (14) where d is the piezoelectric charge constant, εr is the relative dielectric constant and ε0 is the vacuum dielectric constant. where d is the piezoelectric charge constant, εr is the relative dielectric constant and ε0 is the vacuum dielectric constant. Sensors 2020, 20, 3512 14 of 37 According to Equations (13) and (14), g reﬂects the sensitive capabilities of piezoelectric materials [108], therefore a material with a higher piezoelectric voltage coeﬃcient is expected to generate a higher voltage. If the input force is a bending stress, the open circuit output voltage V3j is deﬁned as presented in ation (15) [109]: Equation (15) [109]: (15) V3j = σxx g3j Lj (15) where σxx is the bending constant, g3j the piezoelectric constant and Lj is the distance between the electrodes. Therefore, depending on the type of traductor, it represents either the thickness of the piezoelectric ﬁlm or the lateral gap between the electrodes. Similar to Equation (14), g3j is deﬁned as presented in Equation (16): where σxx is the bending constant, g3j the piezoelectric constant and Lj is the distance between the electrodes. Therefore, depending on the type of traductor, it represents either the thickness of the piezoelectric ﬁlm or the lateral gap between the electrodes. 5.2. Input–Output Dependency Similar to Equation (14), g3j is deﬁned as presented in Equation (16): d g33 = d33 ε0εr (16) (16) Another critical parameter that has to be deﬁned in a piezoelectric transducer’s model is the short-circuit current I, which can be deﬁned using the time derivative of the displacement ﬁeld D, as presented in Equation (17) [110]: I = dD dt = ddT dt = ddS dt (17) (17) where T is stress, and S is strain. where T is stress, and S is strain. where T is stress, and S is strain. Based on Equation (17), the short-circuit current depends on the rate of stress application and the resulting strain rate of the piezoelectric material. Therefore, a higher current can be obtained by applying stress more rapidly. So far, we have formulas for the open circuit voltage and for the short-circuit current. Both are useful pieces of information for a piezoelectric energy harvester model, but since these two values are measured in completely diﬀerent scenarios, the output power cannot be deﬁned as their product. The most basic way to determine the maximum power output P of an energy harvester is to measure the voltage V across a range of resistive loads connected to the device and the current I through them [111]: 2 P = V2 R = I2 R (18) (18) In Equation (18) the maximum output power is deﬁned. If the average output power is the one needed for the model representation, it has to be calculated depending on the type of the input force. For example, if a regularly oscillating AC source is applied at the input of the harvester, the average power is calculated using Equation (19) [111]: Pavg = VRMS IRMS = Vpeak √ 2 Ipeak √ 2 = Vpeak Ipeak 2 (19) (19) Therefore, in this case, the RMS power is half of the peak power. Nevertheless, if the input force is an impulse-type excitation, the average power should be calculated using Equation (20) [111]: Therefore, in this case, the RMS power is half of the peak power. Nevertheless, if the input force is an impulse-type excitation, the average power should be calculated using Equation (20) [111]: Pavg = 1 Tpulse Z t2 t1 V(t)2 R dt = 1 Tpulse Z t2 t1 I(t)2 R dt (20) (20) where R is the resistive load connected on the device, t1 and t2 are the start and the end moments of the pulse, respectively, and Tpulse is the pulse duration. p When the input force is an impulse-type excitation, the duration time of the generated output voltage is too short to collect energy, therefore, a capacitor placed at the harvester’s output is 15 of 37 output Sensors 2020, 20, 3512 of the pulse, respe When the inp recommended. 5.3. Impedance Analysis 5.3. Impedance Analysis Another important electrical parameter that has to be represented in a piezoelectric transducer model is the impedance. An impedance analysis consists in representing the resistive impedance (frequency-independent component, ZRe) and reactive impedance (frequency-dependent component, ZIm) using a Nyquist plot. In this way, the impedance properties of the harvester can be visualized. Usually, a piezoelectric energy harvester is modelled as an RC circuit, which gives a semi-circular relationship between ZRe and ZIm, represented in Figure 10 [112], where the real (resistive) impedance is equal to the diameter of the curve on the real axis and the value of capacitance can be found from the frequency fC, where −ZIm is at a maximum. Another important electrical parameter that has to be represented in a piezoelectric transducer model is the impedance. An impedance analysis consists in representing the resistive impedance (frequency-independent component, 𝑍ோ௘) and reactive impedance (frequency-dependent component, 𝑍ூ௠) using a Nyquist plot. In this way, the impedance properties of the harvester can be visualized. Usually, a piezoelectric energy harvester is modelled as an RC circuit, which gives a semi- circular relationship between 𝑍ோ௘ and 𝑍ூ௠, represented in Figure 10 [112], where the real (resistive) impedance is equal to the diameter of the curve on the real axis and the value of capacitance can be found from the frequency 𝑓஼, where −𝑍ூ௠ is at a maximum. Figure 10. Nyquist plot of the ideal semi-circular response from a RC circuit. Figure 10. Nyquist plot of the ideal semi-circular response from a RC circuit. Figure 10. Nyquist plot of the ideal semi-circular response from a RC circuit. Figure 10. Nyquist plot of the ideal semi-circular response from a RC circuit. The impedance analysis is used to identify the optimum load impedance for maximum power transfer, as this will be close to the internal impedance of the harvester. By modeling the piezoelectric energy harvester as an RC circuit, the resistance and capacitance are obtained from the Nyquist plot, using Equation (22): The impedance analysis is used to identify the optimum load impedance for maximum power transfer, as this will be close to the internal impedance of the harvester. By modeling the piezoelectric energy harvester as an RC circuit, the resistance and capacitance are obtained from the Nyquist plot, using Equation (22): 1 𝑅 𝐶= 1 2 𝑓 (22) R C = 1 2π fc (22) (22) 2 𝜋 𝑓௖ ( Lumped Parameter Electromechanical Model where T is stress, and S is strain. In this case, the generated energy E can be calculated using the output voltage V and the capacitance C, as shown in Equation (21) [68]: g gy, , p p p recommended. In this case, the generated energy 𝐸 can be calculated using the output voltage 𝑉 and the capacitance 𝐶, as shown in Equation (21) [68]: recommended. In this case, the generated energy E can be calculated using the output voltage V and the capacitance C, as shown in Equation (21) [68]: g gy, , p p p recommended. In this case, the generated energy 𝐸 can be calculated using the output voltage 𝑉 and the capacitance 𝐶, as shown in Equation (21) [68]: E = 1 2 C V2 (21) 𝐸= 1 2 𝐶 𝑉ଶ (21) E = 1 2 C V2 𝐸= 1 2 𝐶 𝑉ଶ (21) (21) 5.4. Lumped Parameter Electromechanical Model (25) qc = C vt (25) qc = C vt where C is the electrical capacitance of the transducer and vt is the voltage across the capacitor. where C is the electrical capacitance of the transducer and vt is the voltage across the capacitor. Because of the direct piezoelectric eﬀect, there is a coupled charge induced by the mechanical domain due to the relative displacement x. Furthermore, because of the converse piezoelectric eﬀect, there is a transduced force from the electrical domain proportional to the capacitor voltage, deﬁned in Equation (27). Considering the constant ratio T, the electromechanical coupling between mechanical and electrical domain is given by [48]: qt = T x (26) Ft = T vt (27) (26) qt = T x (26) (26) (27) Ft = T vt (27) (27) Ft = T vt Rate-independent hysteresis exists only in the electrical domain, between the applied actuator voltage and resulting charge, thereby introducing non-linearities into the electrical domain, as shown in Equation (28): vh = H (q) (28) (28) The overall voltage between the transducer’s electrodes is given by Kirchhoﬀ’s second law, highlighted in Equation (29): The overall voltage between the transducer’s electrodes is given by Kirchhoﬀ’s second law, highlighted in Equation (29): (29) vp = vt + vh (29) From the lumped parameter electromechanical model, represented in Equations (23)–(29), it is possible to draw the piezoelectric transducer that couples the electrical and mechanical domains, as shown in Figure 11, where ip = .qp is the current ﬂowing through the transducer when the circuit is closed. Sensors 2020, 20, x FOR PEER REVIEW 16 of 36 Figure 11. Two-port network model for the piezoelectric transducer. Figure 11. Two-port network model for the piezoelectric transducer. Figure 11. Two-port network model for the piezoelectric transducer. Figure 11. Two-port network model for the piezoelectric transducer. 5.5. The Finite Element Method 5.5. The Finite Element Method 5.4. Lumped Parameter Electromechanical Model 5.4. Lumped Parameter Electromechanical Model Most models follow a cascade model structure, where decoupled sub-models are connected to describe the behavior of a piezoelectric transducer. These cascade models take into consideration only one-way coupling since they are control-oriented approximations [113], but a model should represent the energy harvesting and also the damping control of the piezoelectric device. Therefore, a physical Most models follow a cascade model structure, where decoupled sub-models are connected to describe the behavior of a piezoelectric transducer. These cascade models take into consideration only one-way coupling since they are control-oriented approximations [113], but a model should represent the energy harvesting and also the damping control of the piezoelectric device. Therefore, a physical electromechanical interpretation should be compatible with a two-port network model [114] and capable of incorporating two-way piezoelectric coupling eﬀects [48]. Goldfarb and Celanovic [115,116] proposed in their work the lumped parameter electromechanical model, managing to postulate the nonintuitive behavior of piezoelectric transducers. An extended lumped parameter was proposed by Ruderman et al. [117], where they incorporated state-varying capacitance and Voigt–Kelvin-type linear creep eﬀects into the model. If we consider the piezoelectric transducer as a one-mass m with stiﬀness k and damping b, the mechanical domain is governed by the diﬀerential equation (Equation (23)) [48]: m ..x + b .x + k x = Ft + F (23) (23) re Ft is the transduced force from electrical domain and F is the external mechanical force. Piezoelectric materials are dielectric; therefore, the electric domain is mainly deﬁned by a capacitive behavior. As was experimentally observed, between voltage and displacement, as well as between 16 of 37 Sensors 2020, 20, 3512 force and displacement, is exhibited a rate-independent hysteresis. This hysteresis is observed in closed leads between displacement and charge, allowing the postulation that the net electrical charge qp across the piezoelectric transducer can be deﬁned as the sum between two components, as shown in Equation (24) [48]: force and displacement, is exhibited a rate-independent hysteresis. This hysteresis is observed in closed leads between displacement and charge, allowing the postulation that the net electrical charge qp across the piezoelectric transducer can be deﬁned as the sum between two components, as shown in Equation (24) [48]: (24) qp = qc + qt (24) where qc is deﬁned in (25) and qt is deﬁned in Equation (26). 5.5. The Finite Element Method 5.5. The Finite Element Method where U(t) is the (N × 1) vector of the mechanical coordinate displacements, V(t) is the (P × 1) voltage vector, Q(t) is the (P × 1) electric charge vector, Mm is the (N × N) mass matrix, cm is the (N × N) damping matrix, Km is the (N × N) stiﬀness matrix, Kc is (N × P) electromechanical coupling matrix and Ke is the diagonal (P × P) capacitance matrix. Using Equation (30), the natural frequencies of the piezoelectric energy harvester can be calculated. Piezoelectric energy harvesters have two natural frequencies, which are deﬁned as characteristic limit bounds for each vibration mode: • The short-circuit natural frequencies—where it is assumed that no potential diﬀerence exists across the piezoelectric layers in free vibration, therefore the electromechanical coupling in Equation (30) is omitted [118]; • The open circuit natural frequencies—where it is assumed that no charge ﬂows in the electrical circuit [122]. In practice, the short-circuit condition corresponds to a low resistive load connected to the electrodes of the piezoelectric layers (R →0), and the open-circuit condition corresponds to a high resistive load (R →∞). In practice, the short-circuit condition corresponds to a low resistive load connected to the electrodes of the piezoelectric layers (R →0), and the open-circuit condition corresponds to a high resistive load (R →∞). For the ﬁnite element method, the sensitivity analysis is a useful tool for modeling and designing analysis of mechatronic systems and is applied to solve engineering problems such as structural analysis, model updating, design optimization of structures, system control, and uncertainty propagation [123]. The method enables the evaluation of the degree of inﬂuence of the input parameters on the output responses of piezoelectric models, and is considered the most critical step in design optimization [124]. 5.5. The Finite Element Method 5.5. The Finite Element Method Sensitivity analysis techniques are classiﬁed into: • The local method, also referred to as one factor at time analysis—based on the approximation of partial derivatives to evaluate how uncertainty in one factor aﬀects the response of the model keeping the other factors ﬁxed at their nominal values; • The local method, also referred to as one factor at time analysis—based on the approximation of partial derivatives to evaluate how uncertainty in one factor aﬀects the response of the model keeping the other factors ﬁxed at their nominal values; • The global method—evaluates the eﬀect of parameters that are varying within the considered multidimensional space [125], as seen in the Aloui et al. [118] work mentioned above. 5.5. The Finite Element Method 5.5. The Finite Element Method In literature, several approaches to develop piezoelectric energy harvester models are presented, but two of them are the most widely used, showing a close correlation between theoretical and experimental data [118]: In literature, several approaches to develop piezoelectric energy harvester models are presented, but two of them are the most widely used, showing a close correlation between theoretical and experimental data [118]: • The analytical distributed parameter model; • The analytical distributed parameter model; • The analytical distributed p • The finite element model • The ﬁnite element model. The first was introduced by Erturk and Inman [119] and is applied for unimorph and bimorph piezoelectric energy harvesters to obtain a modal solution of second-order ordinary differential equations The ﬁrst was introduced by Erturk and Inman [119] and is applied for unimorph and bimorph piezoelectric energy harvesters to obtain a modal solution of second-order ordinary diﬀerential Sensors 2020, 20, 3512 17 of 37 equations. The second was adapted by De Marqui Junior et al. [120,121], and uses standard discretization to provide discrete models with less constraining hypotheses on the global electrical variables. Aloui et al. [118] studied the ﬁnite element modeling method of a piezoelectric composite beam with P piezoelectric layers and used a standard discretization of N mechanical degrees of freedom. They expressed the general damped electromechanical equations, as shown in Equations (30) and (31), where Equation (30) corresponds to the mechanical equation of motion and Equation (31) corresponds to the electrical equation. Mm .. U(t) + cm . U(t) + Km U(t) + Kc V(t) = F(t) (30) Ke V(t) −KT c U(t) + Q(t) = 0 (31) (30) (31) where U(t) is the (N × 1) vector of the mechanical coordinate displacements, V(t) is the (P × 1) voltage vector, Q(t) is the (P × 1) electric charge vector, Mm is the (N × N) mass matrix, cm is the (N × N) damping matrix, Km is the (N × N) stiﬀness matrix, Kc is (N × P) electromechanical coupling matrix and Ke is the diagonal (P × P) capacitance matrix. 5.6. Equivalent Electric Circuit Model Rjafallah et al. [1] modeled a PU-50%-vol%-PZT composite using the equivalent electric scheme presented in Figure 12. Rjafallah et al. [1] modeled a PU-50%-vol%-PZT composite using the equivalent electric scheme presented in Figure 12. 18 of 37 scheme Sensors 2020, 20, 3512 Rjafallah et al. presented in Figure Fi 12 Th i l t i it d l [1] Figure 12. The equivalent circuit model [1]. Fi u e 12 The e ui ale t i uit odel [1] Figure 12. The equivalent circuit model [1]. Figure 12. The equivalent circuit model [1]. Figure 12. The equivalent circuit model [1]. In the equivalent circuit, there are two blocks interconnected by 𝑅௜, which represents the electrical connection between PU and PZT-ceramic materials. One block models the behavior generated by vibrations, where 𝐼௔௖ is the generated current, 𝐶௘(𝜔) is the capacitance of PU and 𝑅௘(𝜔) is the resistance representing the dielectric losses of the PU matrix, and the other block models the short-circuit current 𝐼௖௖, the capacitance 𝐶௣(𝜔), and the dielectric losses of PZT-ceramic particles 𝑅௣(𝜔). 𝑉ௗ௖ is the bias voltage source, 𝑅௦ is the resistance of the electrodes and wires, 𝐿 is the In the equivalent circuit, there are two blocks interconnected by Ri, which represents the electrical connection between PU and PZT-ceramic materials. One block models the behavior generated by vibrations, where Iac is the generated current, Ce(ω) is the capacitance of PU and Re(ω) is the resistance representing the dielectric losses of the PU matrix, and the other block models the short-circuit current Icc, the capacitance Cp(ω), and the dielectric losses of PZT-ceramic particles Rp(ω). Vdc is the bias voltage source, Rs is the resistance of the electrodes and wires, L is the inductance of the liaisons, and R represents the electrical resistance of the electrical load. ௣( ) ௗ௖ g , ௦ , inductance of the liaisons, and 𝑅 represents the electrical resistance of the electrical load. 5.6. Equivalent Electric Circuit Model The total current 𝐼் traversing the resistance 𝑅 is obtained using the superposition theorem, The total current IT traversing the resistance R is obtained using the superposition theorem, presented in Equation (32): ( ) ( ) ( ) ( ) he resistance 𝑅 is obtained using the superposition theorem, ( ) ( ) ( ) ( ) I(T) R = I(1) R + I(2) R + I(3) R (32) (32) 𝐼ோ (்) = 𝐼ோ (ଵ) + 𝐼ோ (ଶ) + 𝐼ோ (ଷ) (32) where 𝐼ோ (ଵ) is the static electric current provided with the voltage source 𝑉ௗ௖ when all current sources from Figure 12 are replaced by open circuits, 𝐼ோ (ଶ) is the current provided during the electrical- mechanical conversion, performed by PU, where 𝐼௔௖ is active while 𝐼௖௖ is replaced by open circuit and 𝑉ௗ௖ is replaced by short-circuit, and 𝐼ோ (ଷ) corresponds to the electric current due to the mechanical-electrical conversion, carried out by the PZT-ceramic particles, when 𝐼௖௖ is active while 𝐼௔௖ is replaced by open circuit and 𝑉ௗ௖ is replaced by short-circuit. 𝐼ோ (ଶ) and 𝐼ோ (ଷ) have the same where I(1) R is the static electric current provided with the voltage source Vdc when all current sources from Figure 12 are replaced by open circuits, I(2) R is the current provided during the electrical-mechanical conversion, performed by PU, where Iac is active while Icc is replaced by open circuit and Vdc is replaced by short-circuit, and I(3) R corresponds to the electric current due to the mechanical-electrical conversion, carried out by the PZT-ceramic particles, when Icc is active while Iac is replaced by open circuit and Vdc is replaced by short-circuit. I(2) R and I(3) R have the same frequency as that of the mechanical vibration. Taking into consideration all of these, Equation (32) is equivalent to Equation (33) [1]: frequency as that of the mechanical vibration. 6. Frequency Response One of the main characteristics in piezoelectric energy harvesting is the frequency response, since the energy harvesters perform best when their resonance frequency matches their input frequency [63,126]. Currently, most piezoelectric energy harvesters are resonance-based devices; this means that the harvester’s resonant frequency matches the source’s frequency in order to achieve high eﬃciency [127]. Otherwise, a small mismatch can generate a signiﬁcant reduction in voltage and power output [63]. Therefore, the size and shape of the piezoelectric layers are designed according to the natural frequency of the system and the piezoelectric material is chosen to match the application frequency. For example, piezoelectric ceramics are used in applications that demand high vibration frequencies (above 100 Hz) due to their high stiﬀness and superior piezoelectric properties. In comparison, piezoelectric polymers and composites are used in applications that require lower vibration frequencies (up to 30 Hz) [63]. There are two types of frequency response functions [12]: There are two types of frequency response functions [12]: • The frequency response function of linear ﬁeld variables (FRF) or mobility function: is a function of the excitation force and output response velocity; • The frequency response function of linear ﬁeld variables (FRF) or mobility function: is a function of the excitation force and output response velocity; • The frequency response function for power variables (FRFP): is a function of output power or squared voltage and input power. • The frequency response function for power variables (FRFP): is a function of output power or squared voltage and input power. In order to maximize the harvested power, the device’s resonant frequency should correspond to the ambient frequency, but the energy of ambient vibrations is often distributed over a wide frequency spectrum. Therefore, several optimization techniques are used to broaden the harvester bandwidth [9]. Usually, PZT sensors are characterized by a large bandwidth, e.g., Hemmasian Ettefagh et al. [128] studied a sensor that showed high performance within the 32–6400 Hz sensing bandwidth. To adjust the piezoelectric energy harvester’s resonance frequency after fabrication, natural frequency tuning can be implemented using diﬀerent techniques [9]: • Geometrical tuning—the technique consists of adjusting some of the system’s geometrical parameters (inertial mass [129,130] and shape [131,132]) to tune its resonance frequency. The technique implies a simple design without aﬀecting the damping, and before installation, a ﬁne-tuning is possible. • Preload application—the stiﬀness is reduced by applying an axial preload to the beam. 5.6. Equivalent Electric Circuit Model Taking into consideration all of these, Equation (32) is equivalent to Equation (33) [1]: I(T) R = Vdc Rs + Re + Ri + Rp + R + Ze Iac Rs + jLω + Ze + Ri + Zp + R + Zp Icc Rs + jLω + Ze + Ri + Zp + R (33) (33) The root mean square harvested power can be calculated using Equation (34) [1]: PRMS = s 1 T Z T 0 (R i2(t))2dt = s 1 T Z T 0 (R i(2) R (t) + i(3) r (t) 2 ) 2 dt (34) (34) where i(2) R (t) and i(3) R (t) are the temporal electric currents corresponding to the complex electric currents I(2) R and I(3) R . R R The model can be simpliﬁed by removing Rs and L from the equivalent electric scheme due to their weak values compared to the electrical impedances of PU and PZT [1]. Sensors 2020, 20, 3512 19 of 37 19 of 37 6. Frequency Response The technique has a larger eﬀective operating region, but it aﬀects the damping, and it is not suitable for ﬁne-tuning [9,133]. • Extensional mode resonator—the concept uses extension deformation and adjusts the distance between the vibrating beams. The technique has a large eﬀective operating region but also implies complex design, and it is delicate for ﬁne-tuning [9,134]. • Stiﬀness variation—a magnetic stiﬀness is added to the system. The technique is easy to implement and has a simple design, but by adding a magnet, the resonance power is reduced, and it also has a complicated nonlinear behavior [9,135–138]. To increase the operating frequency range, a multi-frequency system, which consists of an array of sub-systems with their own resonance frequency, is needed. There are two techniques for a multi-frequency sub-system [9]: To increase the operating frequency range, a multi-frequency system, which consists of an array of sub-systems with their own resonance frequency, is needed. There are two techniques for a multi-frequency sub-system [9]: • Cantilever array—the technique uses an array of cantilevers with diﬀerent dimensions, each with its own resonance frequency. The concept allows having better control over the ﬁnal frequency range by adjusting the number and dimensions of the cantilevers, it has a simple design and a uniform frequency spectrum, but the total volume is increased [9,139]. • Multimodal system—several seismic masses are used in order to have diﬀerent vibration modes with diﬀerent resonance frequencies. The technique implies a complex design and a non-uniform frequency spectrum, while the total volume is increased [9,140]. Sensors 2020, 20, 3512 20 of 37 Another optimization method is to use a nonlinear system that can be implemented using [9]: Another optimization method is to use a nonlinear system that can be implemented using [9]: Another optimization method is to use a nonlinear system that can be implemented using [9]: • Levitation based systems—the technique uses magnetic repulsion to keep a permanent magnet in levitated oscillations. It is used for low-frequency energy harvesting, and it has no mechanical friction, but the concept is rarely used in piezoelectric energy harvesting [9,141–143]. • Levitation based systems—the technique uses magnetic repulsion to keep a permanent magnet in levitated oscillations. It is used for low-frequency energy harvesting, and it has no mechanical friction, but the concept is rarely used in piezoelectric energy harvesting [9,141–143]. 6. Frequency Response • Vibrating beam with magnetic interaction—the technique combines vibrating beams with magnetic interactions by exploiting nonlinear magnetic forces to change the dynamics of a vibrating structure. It can adjust the frequency response, but the design is complex, and the total volume is increased [9,144–146]. • Mechanical nonlinearity—the nonlinearity is created by a preloaded clamped-clamped structure. The technique is used for frequency tuning with preload conﬁguration, and it does not need an external system, but the design is complex [9,147–150]. • Amplitude limiter conﬁguration—the technique uses a mechanical stopper to broaden the operating frequency range by changing the damping and the stiﬀness of the system at impact. The design is simple, but it has energy losses due to mechanical impact and risks of mechanical impact fatigue [9,151–153]. All of the widen operating frequency systems presented above are resonant, and most operate at frequencies above 50 Hz. Below 50 Hz, energy harvesting with resonant systems becomes more challenging. Moreover, below 10 Hz, using a resonant system becomes unrealistic. Therefore, for this frequency range, non-resonant systems are used, considering two approaches: frequency up-conversion and free moving mass [154]. Frequency up-conversion systems combine a low-frequency system (which absorbs the energy when a low-frequency excitation occurs) and a high-frequency system (which absorbs the energy transferred by the low-frequency system and converts it into electrical energy). However, frequency up-conversion systems need a relatively high amplitude of excitation in order to have an interaction between the low-frequency and high-frequency systems. For this, the following techniques are used [9]: • Resonators-based frequency up-conversion—both low-frequency and high-frequency systems are resonant, and most of them consist of vibrating beams with high and low inertial mass, respectively. The design is simple, and it has a wide bandwidth, but the total volume is increased, and it has risks of mechanical impact fatigue. • Free moving ball-based frequency up-convertors—in this technique, the low-frequency system is replaced by a free moving ball which transfers energy to a vibrating structure. The design is simple, and it harvests energy from arbitrary motion, but it has a non-uniform frequency spectrum, and it is rarely used in piezoelectric energy harvesting. Unlike the frequency up-conversion systems, which require at least one resonating structure for transduction, free moving mass harvesters are entirely non-resonant. Their principle is based on the arbitrary motion of a mass and is used for harvesting from very low frequencies (below 10 Hz), such as human motion. 7. Piezoelectric SPICE Models rotational movements) but, unfo 7. Piezoelectric SPICE Models As presented in the previous sections, piezoelectric transducers are classiﬁed into numerous types, use diﬀerent materials, and their behavior is dependable on the input parameters. Therefore, choosing a suitable transducer for diﬀerent applications implies the use of several SPICE models. 7. Piezoelectric SPICE Models As presented in the previous sections, piezoelectric transducers are classified into numerous types, use different materials, and their behavior is dependable on the input parameters. Therefore, As presented in the previous sections, piezoelectric transducers are classified into numerous types, use different materials, and their behavior is dependable on the input parameters. Therefore, choosing a suitable transducer for different applications implies the use of several SPICE models. Mouapi and Hakem [158] developed a piezoelectric SPICE model for a QP20W transducer, which consists of a sine current source with a parasitic capacitance and resistance, as presented in Figure 13. types, use different materials, and their behavior is dependable on the input parameters. Therefore, choosing a suitable transducer for different applications implies the use of several SPICE models. Mouapi and Hakem [158] developed a piezoelectric SPICE model for a QP20W transducer, which consists of a sine current source with a parasitic capacitance and resistance, as presented in Figure 13. choosing a suitable transducer for different applications implies the use of several SPICE models. Mouapi and Hakem [158] developed a piezoelectric SPICE model for a QP20W transducer, which consists of a sine current source with a parasitic capacitance and resistance, as presented in Figure 13. Figure 13. Piezoelectric SPICE model for QP20W transducer. Figure 13. Piezoelectric SPICE model for QP20W transducer. Figure 13. Piezoelectric SPICE model for QP20W transducer. Figure 13. Piezoelectric SPICE model for QP20W transducer. Figure 13. Piezoelectric SPICE model for QP20W transducer. Figure 13. Piezoelectric SPICE model for QP20W transducer. Another approach is to use a sine voltage source to simulate a piezoelectric transducer, as Linear Technology did for the MIDE V22BL transducer [159] (Figure 14). Another approach is to use a sine voltage source to simulate a piezoelectric transducer, as Linear Technology did for the MIDE V22BL transducer [159] (Figure 14). Another approach is to use a sine voltage source to simulate a piezoelectric transducer, as Linear Technology did for the MIDE V22BL transducer [159] (Figure 14). Fi 14 Pi l t i SPICE d l f MIDE V22BL t d Figure 14. Piezoelectric SPICE model for MIDE V22BL transducer. Figure 14. 6. Frequency Response There are two free moving mass techniques: • Free moving object—the harvester has a free moving object that can be suspended by a rope [155], a rod [156], or rolling inside a cage [157]. The design is simple and generates relatively high power, but the ball’s movements are unpredictable. • Free moving object—the harvester has a free moving object that can be suspended by a rope [155], a rod [156], or rolling inside a cage [157]. The design is simple and generates relatively high power, but the ball’s movements are unpredictable. • Free moving liquid—the technique uses ferroﬂuid motions to vary a magnetic ﬁeld across a coil. The design is simple and it detects inﬁnitely low displacement, but it generates low power and the technique is rarely used in piezoelectric energy harvesting. • Free moving liquid—the technique uses ferroﬂuid motions to vary a magnetic ﬁeld across a coil. The design is simple and it detects inﬁnitely low displacement, but it generates low power and the technique is rarely used in piezoelectric energy harvesting. Since the mass movement is arbitrary and multidirectional, these techniques also imply 3D harvesting. Three-dimensional harvesting systems harvest energy from all directions (translation and rotational movements) but, unfortunately, this technological area is the least found in the literature [9]. 21 of 37 mply 3D i d p y ion and [9] Sensors 2020, 20, 3512 Since the m h h harvesting. Three-d l 7. Piezoelectric SPICE Models rotational movements) but, unfo 7. Piezoelectric SPICE Models Piezoelectric SPICE model for MIDE V22BL transducer. Figure 14. Piezoelectric SPICE model for MIDE V22BL transducer. Figure 14. Piezoelectric SPICE model for MIDE V22BL transducer. Figure 14. Piezoelectric SPICE model for M 8. Electronic Circuits for Piezoelectric Energy Harvesting 8.1. AC-DC Piezoelectric Energy Harvesting Circuit 8.1. AC DC Piezoelectric Energy Harvesting Circuit To rectify the AC power generated by the To rectify the AC power generated by the piezoelectric transducers, an AC-DC energy conditioning circuit is needed between the transducer and the energy storage [15]. There are several electrical schemes used to obtain the highest energy conversion eﬃciency, but the most popular circuit uses a bridge rectiﬁer to convert the output voltage of both polarities to a single polarity. This circuit is used to charge a capacitor CL to a required voltage, after which a switch S is closed, connecting the capacitor to the load RL, as presented in Figure 15 [15,160], where the piezoelectric element is modeled as an equivalent current source Ieq in parallel with a capacitor Cp and the internal leakage resistance Rp. o ecti y t e AC powe ge e ated by t e pie oe ect ic t a sduce s, a AC C e e gy conditioning circuit is needed between the transducer and the energy storage [15]. There are several electrical schemes used to obtain the highest energy conversion efficiency, but the most popular circuit uses a bridge rectifier to convert the output voltage of both polarities to a single polarity. This circuit is used to charge a capacitor 𝐶௅ to a required voltage, after which a switch 𝑆 is closed, connecting the capacitor to the load 𝑅௅, as presented in Figure 15 [15,160], where the piezoelectric element is modeled as an equivalent current source 𝐼௘௤ in parallel with a capacitor 𝐶௣ and the internal leakage resistance 𝑅௣. Fi 1 S d d AC DC h i i i i h l i i l i d l Figure 15. Standard AC-DC energy harvesting circuit with electric piezoelectric model. Figure 15. Standard AC-DC energy harvesting circuit with electric piezoelectric model. Figure 15. Standard AC-DC energy harvesting circuit with electric piezoelectric model. Figure 14. Piezoelectric SPICE model for M 8. Electronic Circuits for Piezoelectric Energy Harvesting As previously explained, when an external mechanical force is applied to a piezoelectric transducer, it will generate an electric ﬁeld due to the direct piezoelectric eﬀect. However, the generated energy varies due to the speed and magnitude of the input force. This means that the only application in which the piezoelectric transducers can be used alone is measuring the input force unless the input force is mechanically regulated, which implies external power. Therefore, piezoelectric energy harvesting applications do not use stand-alone transducers, they need to rectify and store the output to use the generated energy in external sensors and transmitters [160]. For this purpose, diﬀerent electronic circuits can be used (the most popular techniques are summarized in Table 6), but there are also several 22 of 37 ement 22 of 37 ement Sensors 2020, 20, 3512 Table 6), but there a optimization e g M commercially available integrated circuits for power management optimization, e.g., MB39C811, LTC3588-1, LTC3599-2, and MAX17710 [161]. Table 6. Summary of the most popular circuits used in piezoelectric energy harvesting. Circuit 𝐃𝐞𝐬𝐜𝐫𝐢𝐩𝐭𝐢𝐨𝐧 Table 6. Summary of the most popular circuits used in piezoelectric energy harvesting. Circuit Description AC-DC piezoelectric energy harvesting circuit Rectiﬁes the AC power generated by the piezoelectric transducers Two stage piezoelectric energy harvesting circuit Maximize the rectiﬁed power SSHI technique Uses a nonlinear technique to nullify the eﬀect of the capacitive term and increases the output power SECE technique Compared with other techniques, it does not depend on the load Th i it t d i T bl 6 d t il d i th t b ti AC-DC piezoelectric energy harvesting circuit Rectifies the AC power generated by the piezoelectric transducers Two stage piezoelectric energy harvesting circuit Maximize the rectified power SSHI technique Uses a nonlinear technique to nullify the effect of the capacitive term and increases the output power SECE technique Compared with other techniques, it does not depend on the load The circuits presented in Table 6 are detailed in the next subsections. The circuits presented in Table 6 are detailed in the next subsections. p AC DC Pi l t i E H ti Ci it 8.1. AC-DC Piezoelectric Energy Harvesting Circuit 8.1. AC DC Piezoelectric Energy Harvesting Circuit To rectify the AC power generated by th 8.1. AC-DC Piezoelectric Energy Harvesting Circuit 8.1. AC DC Piezoelectric Energy Harvesting Circuit To rectify the AC power generated by the The AC-DC energy harvesting circuit can also be represented using a lumped-parameter electromechanical model with a single degree of freedom, as presented in Figure 16 [162,163], where u(t) is the transverse displacement of the transducer, F(t) is the excitation force, 𝑀 is the effective The AC-DC energy harvesting circuit can also be represented using a lumped-parameter electromechanical model with a single degree of freedom, as presented in Figure 16 [162,163], where u(t) is the transverse displacement of the transducer, F(t) is the excitation force, M is the eﬀective mass, K is the eﬀective stiﬀness, η is the eﬀective damping coeﬃcient, Θ is the eﬀective piezoelectric coupling coeﬃcient and Cp is the eﬀective capacitance. Sensors 2020, 20, x FOR PEER REVIEW 22 of 36 mass, 𝐾 is the effective stiffness, 𝜂 is the effective damping coefficient, Θ is the effective piezoelectric coupling coefficient and 𝐶௣ is the effective capacitance. Figure 16. Standard AC-DC energy harvesting circuit with electromechanical piezoelectric model. Figure 16. Standard AC-DC energy harvesting circuit with electromechanical piezoelectric model. Sensors 2020, 20, 3512 23 of 37 8.2. Two-Stage Piezoelectric Energy Harvesting Circuit Figure 16. Standard AC-DC energy harvesting circu g gy g 8 2 Two Stage Piezoelectric Energy Harvesting Circuit Figure 17 Generalized two-stage piezoelectric energy harvesting circuit Figure 17. Generalized two-stage piezoelectric energy harvesting circuit. Figure 17. Generalized two-stage piezoelectric energy harvesting circuit. g g p gy g The DC DC converter block is presented in detail in Figure 18 [166]: The DC-DC converter block is presented in detail in Figure 18 [166]: The DC-DC converter block is presented in detail in Figure 18 [166]: g g p gy g The DC-DC converter block is presented in detail in Figure 18 [166]: The DC-DC converter block is presented in detail in Figure 18 [166]: The DC-DC converter block is presented in detail in Figure 18 [166]: Fi 1 DC DC i h h i h Figure 18. DC-DC converter in the two-stage energy harvesting scheme. Figure 18. DC-DC converter in the two-stage energy harvesting scheme. Figure 18. DC-DC converter in the two-stage energy harvesting scheme. Figure 18. DC-DC converter in the two-stage energy harvesting scheme. For the circuit presented in Figure 18 the converter efficiency 𝜂 is calculated in Equation (35) [166]: For the circuit presented in Figure 18, the converter eﬃciency ηC is calculated in Equation (35) [166]: n Figure 18, the converter efficiency 𝜂஼ is calculated in Equation (35) [166]: in Figure 18, the converter efficiency 𝜂஼ is calculated in Equation (35) [166]: ηC = Vrect + VD −Vces Vrect Vesd Vesd + VD (35) (35) where Vrect is the rectiﬁed voltage, VD is the forward bias of the diode D1, Vces is the voltage drop of the internal switch Tr and Vesd is the voltage of the energy storage device. 8.3. Synchronized Switch Harvesting on Inductor (SSHI) 8.3. Synchronized Switch Harvesting on Inductor (SSHI) 8.2. Two-Stage Piezoelectric Energy Harvesting Circuit Figure 16. Standard AC-DC energy harvesting circu g gy g 8 2 Two Stage Piezoelectric Energy Harvesting Circuit 8.2. Two-Stage Piezoelectric Energy Harvesting Circuit Figure 16. Standard AC-DC energy harvesting circu g gy g 8 2 Two-Stage Piezoelectric Energy Harvesting Circuit Since the excitation force level is not constant, the rectiﬁed voltage level is not constant either. Therefore, a DC-DC converter can be used after the rectiﬁer to maximize the power transferred to the storage device [59,164]. Another solution is to use a step-down DC-DC converter operating in discontinuous conduction mode (DCM) if the generated voltage is higher than the required voltage level of the storage device [164,165]. In this case, if the duty cycle is optimized for the step-down converter, the energy ﬂow to the battery can be tripled [165]. In Figure 17, Guan and Liao [166] presented an optimized two-stage energy harvester, where C0 is a temporary storage capacitor. 8.2. Two-Stage Piezoelectric Energy Harvesting Circuit Since the excitation force level is not constant, the rectified voltage level is not constant either. Therefore, a DC-DC converter can be used after the rectifier to maximize the power transferred to the storage device [59,164]. Another solution is to use a step-down DC-DC converter operating in discontinuous conduction mode (DCM) if the generated voltage is higher than the required voltage level of the storage device [164,165]. In this case, if the duty cycle is optimized for the step-down converter, the energy flow to the battery can be tripled [165]. In Figure 17, Guan and Liao [166] presented an optimized two-stage energy harvester, where 𝐶଴ is a temporary storage capacitor. o age ie oe e i E e gy a e i g i ui Since the excitation force level is not constant, the rectified voltage level is not constant either. Therefore, a DC-DC converter can be used after the rectifier to maximize the power transferred to the storage device [59,164]. Another solution is to use a step-down DC-DC converter operating in discontinuous conduction mode (DCM) if the generated voltage is higher than the required voltage level of the storage device [164,165]. In this case, if the duty cycle is optimized for the step-down converter, the energy flow to the battery can be tripled [165]. In Figure 17, Guan and Liao [166] presented an optimized two-stage energy harvester, where 𝐶଴ is a temporary storage capacitor. Figure 17. Generalized two-stage piezoelectric energy harvesting circuit. Figure 17. Generalized two-stage piezoelectric energy harvesting circuit. Figure 17. Generalized two-stage piezoelectric energy harvesting circuit. 8.3. Synchronized Switch Harvesting on Inductor (SSHI) If the system is weakly coupled (and not only in this situation), the harvested power can be increased using a nonlinear technique called Synchronized Switch Harvesting on Inductor (SSHI) [167]. This technique is based on the resonant circuit created by the internal capacitor of the piezoelectric transducer and an external inductor which ﬂips the capacitor voltage instantly to nullify the eﬀect of the capacitive term. In this way, the energy used to charge the internal capacitor in order to conduct the diode bridge, which would have been wasted, is now harvested [51]. In the classical AC-DC energy harvesting circuit (Figure 15), a negative power is produced because the output current and generated voltage cannot keep the same phase; this means that the harvested energy may return to the mechanical part, losing some of the harvested power [168]. The SSHI circuit (Figures 19 and 20) overcomes this problem by adding a switch path (which contains the switch S1 and the inductor L1). The switch opens instantly when the capacitor voltage reaches the maximum in the opposite polarity to result in the ﬂipping of the capacitor voltage. This enables the circuit to harvest the capacitor charge instead of being discharged to waste [51]. Sensors 2020, 20, 3512 24 of 37 output voltage [169]; the only difference between them is whether the switch path is connected in parallel or in series with the rectifier [15]. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. p [ ] Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 20. Series-SSHI energy harvesting circuit. Sensors 2020, 20, 3512 output voltage [169 parallel or in series 24 of 37 cted in Sensors 2020, 20, 3512 24 of 37 parallel or in series with the rectifier [15]. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. parallel or in series with the rectifier [15]. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit Figure 19. Parallel-Synchronized Switch Harvesting on Inductor (SSHI) energy harvesting circuit. Figure 20. Series-SSHI energy harvesting circuit. Figure 20. 8.3. Synchronized Switch Harvesting on Inductor (SSHI) Series-SSHI energy harvesting circuit. Figure 20. Series-SSHI energy harvesting circuit. Figure 20. Series-SSHI energy harvesting circuit. There are two SSHI circuits: parallel (Figure 19) and series (Figure 20). Both can increase the output voltage [169]; the only diﬀerence between them is whether the switch path is connected in parallel or in series with the rectiﬁer [15]. Sensors 2020, 20, x FOR PEER REVIEW 24 of 36 Lalla t a d Guyo a [169] de i ed a elf o e ed SSHI i te fa e able to auto ati ally e fo Lallart and Guyomar [169] designed a self-powered SSHI interface able to automatically perform switching actions once the output voltage reaches its maximum. The schematics of the self-powered parallel-SSHI and series-SSHI are presented in Figures 21 and 22. Lallart and Guyomar [169] designed a self-powered SSHI interface able to automatically perform switching actions once the output voltage reaches its maximum. The schematics of the self-powered parallel-SSHI and series-SSHI are presented in Figures 21 and 22. Figure 21. Self-powered parallel-SSHI energy harvesting circuit. Figure 21. Self-powered parallel-SSHI energy harvesting circuit. Figure 21. Self-powered parallel-SSHI energy harvesting circuit. Figure 21. Self-powered parallel-SSHI energy harvesting circuit. 25 of 37 25 of 37 Sensors 2020, 20, 3512 Figure 22. Self-powered series-SSHI energy harvesting circuit. Figure 22. Self-powered series-SSHI energy harvesting circuit. Figure 22 Self powered series SSHI energy harvesting circuit Figure 22. Self-powered series-SSHI energy harvesting circuit. In Figures 21 and 22, the diodes are 1N4004 model, 𝑇ଵ and 𝑇ଷ are TIP32C, 𝑇ଶ and 𝑇ସ are TIP31C 𝐿 has 120 mH and 97 Ω𝐶 and 𝐶 are 23 nF capacitors and 𝐶 has 47 µF [169] In Figures 21 and 22, the diodes are 1N4004 model, T1 and T3 are TIP32C, T2 and T4 are TIP31C, L1 has 120 mH and 97 Ω, C1 and C2 are 23 nF capacitors and CL has 47 µF [169]. TIP31C, 𝐿ଵ has 120 mH and 97 Ω, 𝐶ଵ and 𝐶ଶ are 23 nF capacitors and 𝐶௅ has 47 µF [169]. For the self-powered series-SSHI, a current flowing through the circuit is generated by the piezoelectric transducer. When the output voltage across the piezoelectric transducer is increasing, 𝐶ଵ will be charged, while 𝑇ଵ and 𝑇ଶ are blocked, and when this voltage reaches the peak value, it starts For the self-powered series-SSHI, a current ﬂowing through the circuit is generated by the piezoelectric transducer. 8.3. Synchronized Switch Harvesting on Inductor (SSHI) When the output voltage across the piezoelectric transducer is increasing, C1 will be charged, while T1 and T2 are blocked, and when this voltage reaches the peak value, it starts decreasing and, between the emitter and the base of T1, appears a voltage diﬀerence. When this voltage diﬀerence reaches the threshold voltage of T1, it will start conducting and at the same time, C1 provides the voltage for T2’s base since D1 is blocked. The conduction of T1 and T2 initiates the inductance-capacitance resonant circuit and Cp is quickly discharged through D3, T2, RL, D8, and L1. With the presence of inductor L1, the voltage across Cp is inverted. Similarly, when the voltage reaches the negative peak, it will also be inversed due to the circuit’s symmetrical topology [15]. For the self-powered parallel-SSHI, the working principle is similar. The diﬀerence consists in the fact that when Cp is discharging, the current ﬂows only through D3, T2 and L1, or D4, T3 and L1. This means that the voltage inverting process is not inﬂuenced by RL. piezoelectric transducer returns to an open circuit, storage capacitor When the rectified voltage reac 9. Piezoelectric Energy Harvesting Applications g p g p g , y y mechanical displacement, the next electric charge extraction sequence occurs [170]. In theory, with determined input power, the flyback converter delivers a constant output power ndependent of the load. However, in practice, due to the components’ imperfections, the effective output voltage domain is restricted [170]. 9. Piezoelectric Energy Harvesting Applications In the literature, numerous examples of applications for piezoelectric energy harvesting exist. For example, Liu et al. [15] used a monolithic multilayer piezoelectric stack manufactured by Smart Material, which consists of 300 layers of piezoelectric patches with the internal capacitance of 2.1 µF, In the literature, numerous examples of applications for piezoelectric energy harvesting exist. For example, Liu et al. [15] used a monolithic multilayer piezoelectric stack manufactured by Smart Material, which consists of 300 layers of piezoelectric patches with the internal capacitance of 2.1 µF, to test the eﬃciency of a footstep piezoelectric-stack energy harvester, in a laboratory experiment. The piezoelectric device was placed in an ampliﬁcation frame stimulated by a shaker. The input force was measured by a force transducer placed between the ampliﬁcation frame and the shaker, while the ampliﬁed force was measured by another force transducer placed between the interface of the piezoelectric stack harvester and the ampliﬁcation frame. They considered the force excitation of human locomotion 114 N, and the ampliﬁed force resulted being 846 N. With this setup, they used three diﬀerent circuits: • When they used a standard energy harvesting circuit (a simple AC-DC converter) with a 50 load, they obtained an output power of 1.35 mW; • When they used a series-SSHI circuit with a 60 kΩload, they obtained an output powe 1.33 mW; • When they used a parallel-SSHI with a load of 80 kΩ, they obtained an output power of 2.35 mW, and the harvesting eﬃciency was increased by 74% in comparison with the ﬁrst circuit. • When they used a parallel-SSHI with a load of 80 kΩ, they obtained an output power of 2.35 mW, and the harvesting eﬃciency was increased by 74% in comparison with the ﬁrst circuit. A diﬀerent way to achieve this result without using a shaker is to use a common loudspeaker excited by an audio power ampliﬁer. This is a convenient way to modify the output frequency and the excitation level. He et al. 8.4. Synchronous Electrical Charge Extraction (SECE) Another method for piezoelectric energy harvesting electrical circuits is Synchronous Electrical Charge Extraction (SECE). Unlike other methods, the generated power using the SECE method does not depend on the load, therefore, the load can vary without aﬀecting its eﬃciency. The disadvantage of this method consists of the complex controller of the circuit [51]. The electrical circuit of the SECE method is presented in Figure 23 [170]. In Figure 23, after the AC-DC converter, which consists of four UF4004 rectiﬁer diodes, a DC-DC power converter was implemented using a ﬂyback switching mode converter. The latter uses an IRFD220 MOSFET transistor T, a Myrra 74,010 coupled inductor L, an SB540 Schottky diode D5 and a 1000 µF storage capacitor CL [170]. Sensors 2020, 20, 3512 of this method cons The ele t i al 26 of 37 Figure 23. Synchronous Electrical Charge Extraction (SECE) energy harvesting circuit. Figure 23. Synchronous Electrical Charge Extraction (SECE) energy harvesting circuit. Figure 23. Synchronous Electrical Charge Extraction (SECE) energy harvesting circuit. In Figure 23, after the AC-DC converter, which consists of four UF4004 rectifier diodes, a DC- DC power converter was implemented using a flyback switching mode converter. The latter uses an RFD220 MOSFET transistor 𝑇, a Myrra 74,010 coupled inductor 𝐿, an SB540 Schottky diode 𝐷ହ and a 1000 µF storage capacitor 𝐶௅ [170]. The gate voltage of the MOSFET transistor, 𝑢௚, is determined by a control circuit measuring the ectified voltage 𝑉௥௘௖௧, and controls the power converter operation. When the rectified voltage eaches the peak value, a 15 V voltage is applied on the 𝑇gate, the transistor starts conducting and The gate voltage of the MOSFET transistor, ug, is determined by a control circuit measuring the rectiﬁed voltage Vrect, and controls the power converter operation. When the rectiﬁed voltage reaches the peak value, a 15 V voltage is applied on the T gate, the transistor starts conducting and transfers the electric charge from the piezoelectric transducer to the coupled inductor. When the piezoelectric transducer’s electrical charge is completely extracted, the control circuit detects the cancellation of the rectiﬁed voltage, and T is blocked. Now that the transistor is blocked, the piezoelectric transducer returns to an open circuit, and the energy stored by L is transferred to the storage capacitor. When the rectiﬁed voltage reaches the peak value again, synchronously with mechanical displacement, the next electric charge extraction sequence occurs [170]. 8.4. Synchronous Electrical Charge Extraction (SECE) reaches the peak value, a 15 V voltage is applied on the 𝑇 gate, the transistor starts conducting and transfers the electric charge from the piezoelectric transducer to the coupled inductor. When the piezoelectric transducer’s electrical charge is completely extracted, the control circuit detects the cancellation of the rectified voltage, and 𝑇 is blocked. Now that the transistor is blocked, the In theory, with determined input power, the ﬂyback converter delivers a constant output power independent of the load. However, in practice, due to the components’ imperfections, the eﬀective output voltage domain is restricted [170]. piezoelectric transducer returns to an open circuit, storage capacitor When the rectified voltage reac 9. Piezoelectric Energy Harvesting Applications [125] designed a piezoelectric energy harvester structure using a double-layer squeezing structure and a piezoelectric beam array. During the experiment, a 60-kg person stepped on and oﬀ the structure with diﬀerent frequencies. The maximum output power obtained by one piezoelectric beam was 134.2 µW under a step frequency of 1.81 Hz, but the authors considered that a total of 40 piezoelectric beams inside the ﬂoor structure could reach 5.368 mW. Sensors 2020, 20, 3512 27 of 37 Zhang et al. [3] integrated a stiﬀness spring with an energy generator to develop a device for harvesting the mechanical energy of the foot. The harvester was placed in a shoe near the heel, therefore, when the person’s foot touches the ground, the heel compressed a pedal, and a piezoelectric beam was bent, generating electrical energy. When the foot lifted oﬀthe ground, the stiﬀness spring already stored elastic potential energy, which was converted to kinetic energy, bending the piezoelectric beam once again. A 60-kg person could obtain 235.2 mJ per step. A comparison study of vibrational energy harvesting using piezoelectric tiles in walkways vs. stairways was presented in [171]. The authors concluded that piezoelectric tiles placed in a stairway perform better than the ones placed in walkways due to the natural increased pedestrian work demanded in traversing the stairs. They also indicated that the tile design should consider the naturally random characteristics of pedestrian traﬃc in order to increase the level of harvested power. p p Cho et al. [13] developed a road-compatible piezoelectric energy harvester using piezoelectric transducers ﬁxed onto both ends, the stress being converged toward the center of the device via a rigid bar. The harvester was used in actual road conditions for ﬁve months, being stressed by the vehicles traveling at speeds of 10–50 km/h when they entered a highway rest area. At 50 km/h, the harvester’s output power was 2.381 W, while at 10 km/h, the output power was 576 mW. The generated energy was used to power the LED indicators and to transmit real-time information about the sensor’s leak, temperature, and strain. Cha et al. [172] analyzed the possibility of harvesting the energy generated by the mouse click motion, using a unimorph PVDF beam structure. piezoelectric transducer returns to an open circuit, storage capacitor When the rectified voltage reac 9. Piezoelectric Energy Harvesting Applications The maximum harvested energy, obtained when the load resistance matched the impedance of the piezoelectric transducer for the fundamental harmonic, was in the range of 1–10 nJ, but this level could be increased by using multiple piezoelectric layers or alternative materials with higher eﬃciency. Funding: This research received no external funding. Author Contributions: Both authors have contributed to this research. C.C. designed the structure of the material and wrote the paper. Both authors carried out simulations and analyzed the data. A.G. reviewed the manuscript. All authors have read and agreed to the published version of the manuscript 10. Conclusions The development of the Internet of Things concept, wearables devices, and wireless technologies has led to the need for self-powered systems due to the inaccessibility of batteries for changing. A solution for these self-powered systems is to harvest mechanical energy using piezoelectricity. Piezoelectric materials have the property to generate an electric ﬁeld when a mechanical force is applied. This phenomenon is known as the direct piezoelectric eﬀect. Piezoelectric energy harvesting has several advantages, such as high energy and power density, low cost, good scalability, and ease of application. However, due to its main disadvantages (low level of harvested power and the need for rectiﬁcation, maximum power extraction, and output voltage regulation), piezoelectric transducers cannot be used alone to harvest mechanical energy. Therefore, a piezoelectric energy harvester usually contains an AC-DC converter, has a two-stage conversion circuit, or uses non-linear techniques such as SSHI or SECE. The piezoelectric energy harvesting technique can be used in a multitude of applications; therefore, each implementation needs to optimize the technique for its own needs. First, a suitable transducer must be found. Piezoelectric transducers can be found in diﬀerent shapes (cantilever beam, circular diaphragm, cymbal type, stack type, and more) and can be made from diﬀerent materials, each with its own characteristics. When the piezoelectric transducer is chosen, the next step is to develop a model in order to simulate and optimize the behavior in the integrated system. Author Contributions: Both authors have contributed to this research. C.C. designed the structure of the material and wrote the paper. Both authors carried out simulations and analyzed the data. A.G. reviewed the manuscript. All authors have read and agreed to the published version of the manuscript Funding: This research received no external funding. 28 of 37 Sensors 2020, 20, 3512 Acknowledgments: This work was supported by a grant of the Romanian Ministery of Research and Innovation, project number10PFE/16.10.2018, PERFORM-TECH-UPT - The increasing of the institutional performance of the Polytechnic University of Timis,oara by strengthening the research, development and technological transfer capacity in the ﬁeld of “Energy, Environment and Climate Change”, within Program 1 - Development of the national system of Research and Development, Subprogram 1.2 - Institutional Performance - Institutional Development Projects - Excellence Funding Projects in RDI, PNCDI III”. Conﬂicts of Interest: The authors declare no conﬂict of interest. Nomenclature Notations h dti Converse piezoelectric eﬀect matrix [d] Direct piezoelectric eﬀect matrix ∆t Generation time Θ The eﬀective piezoelectric coupling coeﬃcient A Transducer’s area C Electrical capacitance D Displacement of charge density, electric displacement, C m−2 E Electric ﬁeld strength or intensity, N/C E% Mechanical-electrical energy conversion eﬃciency Ee Electrical energy Em Mechanical energy F Applied force Ft The transduced force from electrical domain F(t) The excitation force K The eﬀective stiﬀness Lj The distance between the electrodes M The eﬀective mass P Output power Pavg The average output power Pin Mechanical input power Pout Electrical output power Pres Power density, W/m3 R Resistive load S Strain T Stress, N/m2 (Pa) Tpulse The duration of the pulse V Output voltage V3j The output circuit voltage when the input force is a bending stress Zmax Maximum displacement b damping d Movement distance or piezoelectric coeﬃcient or piezoelectric charge constant d31 Piezoelectric constant dij, eij, gij, hij Piezoelectric coeﬃcients fres Resonance frequency g Piezoelectric voltage coeﬃcient g3j The piezoelectric voltage constant when the input force is a bending stress ip The current through ﬂowing through the piezoelectric transducer when the circuit i k stiﬀness m Inertial mass qc Results from the voltage across the capacitor qp Electrical charge qt Coupled charge induced by the mechanical domain s Elastic compliance, 1/Pa 29 of 37 Sensors 2020, 20, 3512 sE 11 Elastic compliance t Transducer’s thickness t1, t2 The start and the end of the pulse u(t) The transverse displacement of the traductor v The applied force’s speed vh The electrical domain’s non-linearities vp The overall voltage between piezoelectric transducer’s electrodes x Relative displacement y Amplitude of vibration of the housing ε Permittivity, F/m εT 33/ε0 Dielectric constant ε0 Permittivity in free space or vacuum dielectric constant (8.854 × 10−12 F/m) εr The relative dielectric constant η The eﬀective damping coeﬃcient ηC The convertor eﬃciency ρ Mass density σxx The bending stress Abbreviations BaTiO3 Barium titanate BNT Bismuth Sodium Titanate DCM Discontinues Conduction Mode EAP Electrode polymer IoT Internet of Things KNbO3 Potassium niobate MEMS Micro ElectroMechanical Systems PLA Polylactic acid PMN Lead magnesium niobite PU Polyurethane PVDF Polyvinylidene ﬂuoride PZT Lead zirconate titanate SECE Synchronous Electrical Charge Extraction SSHI Synchronized Switch Harvesting on Inductor TEG Thermoelectric Generator ZnO Zinc oxide References In Proceedings of the 2019 5th International Conference on Advanced Computing Communication Systems (ICACCS), Coimbatore, India, 5 March 2019. [CrossRef] 9. Maamer, B.; Boughamoura, A.; El-Bab, A.M.F.; Francis, L.A.; Tounsi, F. A review on design improvements and techniques for mechanical energy harvesting using piezoelectric and electromagnetic schemes. Energy Convers. Manag. 2019, 199, 111973. [CrossRef] 10. Caliò, R.; Rongala, U.; Camboni, D.; Milazzo, M.; Stefanini, C.; Petris, G.D.; Oddo, C. Piezoelectric Energy Harvesting Solutions. Sensors 2014, 14, 4755–4790. [CrossRef] 11. Mitcheson, P.; Green, T.; Yeatman, E.; Holmes, A. Architectures for Vibration-Driven Micropower Generators. J. Microelectromech. Syst. 2004, 13, 429–440. [CrossRef] 12. Wang, X. A study of harvested power and energy harvesting eﬃciency using frequency response analyses of power variables. Mech. Syst. Signal Process. 2019, 133, 106277. [CrossRef] 13. Cho, J.Y.; Kim, K.-B.; Hwang, W.S.; Yang, C.H.; Ahn, J.H.; Hong, S.D.; Sung, T.H. A multifunctional road-compatible piezoelectric energy harvester for autonomous driver-assist LED indicators with a self-monitoring system. Appl. Energy 2019, 242, 294–301. [CrossRef] 14. Roundy, S.; Wright, P.K.; Rabaey, J. A study of low level vibrations as a power source for wireless sensor nodes. Comput. Commun. 2003, 26, 1131–1144. [CrossRef] 15. Liu, H.; Hua, R.; Lu, Y.; Wang, Y.; Salman, E.; Liang, J. Boosting the eﬃciency of a footstep piezoelectric-stack energy harvester using the synchronized switch technology. J. Intell. Mater. Syst. Struct. 2019, 30, 813–822. [CrossRef] 16. Khalid, S.; Raouf, I.; Khan, A.; Kim, N.; Kim, H.S. A Review of Human-Powered Energy Harvesting for Smart Electronics: Recent Progress and Challenges. Int. J. Precis. Eng. Manuf. Green Technol. 2019, 6, 821–851. [CrossRef] 17. Roundy, S.; Wright, P.K.; Rabaey, J.M. Energy Scavenging for Wireless Sensor Networks; Springer: Boston, MA, USA, 2004. [CrossRef] 18. Boisseau, S.; Despesse, G.; Ahmed, B. Electrostatic Conversion for Vibration Energy Harvesting. Small Scale Energy Harvest. 2012. [CrossRef] Luo, Z.; Zhu, D.; Beeby, S.P. Multilayer ferroelectret-based energy harvesting insole. J. Phys. Conf. Ser. 2015 660, 012118. [CrossRef] 20. Kim, K.N.; Chun, J.; Chae, S.A.; Ahn, C.W.; Kim, I.W.; Kim, S.-W.; Baik, J.M. Silk ﬁbroin-based biodegradable piezoelectric composite nanogenerators using lead-free ferroelectric nanoparticles. Nano Energy 2015, 14, 87–94. [CrossRef] 21. Khan, F.; Sassani, F.; Stoeber, B. Nonlinear behaviour of membrane type electromagnetic energy harvester under harmonic and random vibrations. Microsyst. Technol. 2013, 20, 1323–1335. [CrossRef] 22. Miao, P.; Mitcheson, P.D.; Holmes, A.S.; Yeatman, E.M.; Green, T.C.; Stark, B.H. MEMS inertial power generators for biomedical applications. Microsyst. Technol. 2006, 12, 1079–1083. References 1. Rjafallah, A.; Hajjaji, A.; Guyomar, D.; Kandoussi, K.; Belhora, F.; Boughaleb, Y. Modeling of polyurethane/lead zirconate titanate composites for vibration energy harvesting. J. Compos. Mater. 2018, 53, 613–623. [CrossRef] 2. Landaluce, H.; Arjona, L.; Perallos, A.; Falcone, F.; Angulo, I.; Muralter, F. A Review of IoT Sensing 1. Rjafallah, A.; Hajjaji, A.; Guyomar, D.; Kandoussi, K.; Belhora, F.; Boughaleb, Y. Modeling of polyurethane/lead zirconate titanate composites for vibration energy harvesting. J. Compos. Mater. 2018, 53, 613–623. [CrossRef] 1. Rjafallah, A.; Hajjaji, A.; Guyomar, D.; Kandoussi, K.; Belhora, F.; Boughaleb, Y. Modeling of polyurethane/lead zirconate titanate composites for vibration energy harvesting. J. Compos. Mater. 2018, 53, 613–623. [CrossRef] 2. Landaluce, H.; Arjona, L.; Perallos, A.; Falcone, F.; Angulo, I.; Muralter, F. A Review of IoT Sensing Applications and Challenges Using RFID and Wireless Sensor Networks. Sensors 2020, 20, 2495. [CrossRef] [PubMed] 3. Zhang, Y.-H.; Lee, C.-H.; Zhang, X.-R. A novel piezoelectric power generator integrated with a compliant energy storage mechanism. J. Phys. D Appl. Phys. 2019, 52, 455501. [CrossRef] . Zhou, M.; Al-Furjan, M.; Wang, B. Modeling and Eﬃciency Analysis of a Piezoelectric Energy Harve Based on the Flow Induced Vibration of a Piezoelectric Composite Pipe. Sensors 2018, 18, 4277. [CrossR 4. Zhou, M.; Al-Furjan, M.; Wang, B. Modeling and Eﬃciency Analysis of a Piezoelectric Energy Harvester Based on the Flow Induced Vibration of a Piezoelectric Composite Pipe. Sensors 2018, 18, 4277. [CrossRef] 5. Maghsoudi Nia, E.; Wan Abdullah Zawawi, N.A.; Mahinder Singh, B.S. Design of a pavement using piezoelectric materials. Mater. Werkst. 2019, 50, 320–328. [CrossRef] . Maghsoudi Nia, E.; Wan Abdullah Zawawi, N.A.; Mahinder Singh, B.S. Design of a pavement u piezoelectric materials. Mater. Werkst. 2019, 50, 320–328. [CrossRef] 6. Sudevalayam, S.; Kulkarni, P. Energy Harvesting Sensor Nodes: Survey and Implications. IEEE Commun. Surv. Tutor. 2011, 13, 443–461. [CrossRef] 6. Sudevalayam, S.; Kulkarni, P. Energy Harvesting Sensor Nodes: Survey and Implications. IEEE Commun. Surv. Tutor. 2011, 13, 443–461. [CrossRef] 7. Shaikh, F.K.; Zeadally, S. Energy harvesting in wireless sensor networks: A comprehensive review. Renew. Sustain. Energy Rev. 2016, 55, 1041–1054. [CrossRef] 7. Shaikh, F.K.; Zeadally, S. Energy harvesting in wireless sensor networks: A comprehensive review. Renew. Sustain. Energy Rev. 2016, 55, 1041–1054. [CrossRef] Sensors 2020, 20, 3512 30 of 37 8. Kumar, S.S.; Kaviyaraj, R.; Narayanan, L.A.J. Energy Harvesting by Piezoelectric Sensor Array in Road Using Internet of Things. References Bendame, M.; Abdel-Rahman, E.; Soliman, M. Electromagnetic Impact Vibration Energy Harvesters. In Physics Structural Nonlinear Dynamics and Diagnosis; Springer: Cham, Switzerland, 2015; pp. 29–58. [CrossRef] 35. Geisler, M.; Boisseau, S.; Perez, M.; Ait-Ali, I.; Perraud, S. Scaling eﬀects in a non-linear electromagnetic energy harvester for wearable sensors. J. Phys. Conf. Ser. 2016, 773, 012044. [CrossRef] 36. Chen, S.W.; Cao, X.; Wang, N.; Ma, L.; Zhu, H.R.; Willander, M.; Wang, Z.L. An Ultrathin Flexible Single-Electrode Triboelectric-Nanogenerator for Mechanical Energy Harvesting and Instantaneous Force Sensing. Adv. Energy Mater. 2016, 7, 1601255. [CrossRef] 37. Yu, A.; Pu, X.; Wen, R.; Liu, M.; Zhou, T.; Zhang, K.; Wang, Z.L. Core–Shell-Yarn-Based Triboelectric Nanogenerator Textiles as Power Cloths. ACS Nano 2017, 11, 12764–12771. [CrossRef] [PubMed] 38. Zhang, Z.; Du, K.; Chen, X.; Xue, C.; Wang, K. An air-cushion triboelectric nanogenerator integrated with stretchable electrode for human-motion energy harvesting and monitoring. Nano Energy 2018, 53, 108–115. [CrossRef] 39. Tang, W.; Jiang, T.; Fan, F.R.; Yu, A.F.; Zhang, C.; Cao, X.; Wang, Z.L. Liquid-Metal Electrode for High-Performance Triboelectric Nanogenerator at an Instantaneous Energy Conversion Eﬃciency of 70.6%. Adv. Funct. Mater. 2015, 25, 3718–3725. [CrossRef] 40. Zhang, C.; He, X.-F.; Li, S.-Y.; Cheng, Y.-Q.; Rao, Y. A Wind Energy Powered Wireless Temperature Sensor Node. Sensors 2015, 15, 5020–5031. [CrossRef] 41. Briscoe, J.; Dunn, S. 2.1 Background. In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 3–4. 2. Erturk, A.; Inman, D.J. 1.4 Summary of the Theory of Linear Piezoelectricity. In Piezoelectric Energy Harves 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 9–12. 3. de Almeida, B.V.; Pavanello, R. Topology Optimization of the Thickness Proﬁle of Bimorph Piezoele Energy Harvesting Devices. J. Appl. Comput. Mech. 2019, 5, 113–127. [CrossRef] 44. Lippmann, G. Principe de la conservation de lélectricité, ou second principe de la théorie des phénomènes électriques. J. Phys. Theor. Appl. 1881, 10, 381–394. [CrossRef] 45. Koh, S.J.A.; Zhao, X.; Suo, Z. Maximal energy that can be converted by a dielectric elastomer generator. Appl. Phys. Lett. 2009, 94, 262902. [CrossRef] 46. Briscoe, J.; Dunn, S. 2.2 Polarisation. In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 4–6. 47. Chopra, I. Review of state of art of smart structures and integrated systems. AIAA J. 2002, 40, 2145–2187. [CrossRef] 48. Tavares, R.; Ruderman, M. On Energy Harvesting Using Piezoelectric Transducer with Two-Port Model Under Force Excitation. References [CrossRef] 23. Zhang, K.; Wang, X.; Yang, Y.; Wang, Z.L. Hybridized Electromagnetic–Triboelectric Nanogenerator for Scavenging Biomechanical Energy for Sustainably Powering Wearable Electronics. ACS Nano 2015, 9, 3521–3529. [CrossRef] [PubMed] 24. Invernizzi, F.; Dulio, S.; Patrini, M.; Guizzetti, G.; Mustarelli, P. Energy harvesting from human motion: Materials and techniques. Chem. Soc. Rev. 2016, 45, 5455–5473. [CrossRef] [PubMed] 25. Xu, W.; Wong, M.-C.; Hao, J. Strategies and progress on improving robustness and reliability of triboelectric nanogenerators. Nano Energy 2019, 55, 203–215. [CrossRef] 26. Cao, J.; Wang, W.; Zhou, S.; Inman, D.J.; Lin, J. Nonlinear time-varying potential bistable energy harvesting from human motion. Appl. Phys. Lett. 2015, 107, 143904. [CrossRef] 27. Wang, W.; Cao, J.; Bowen, C.R.; Zhou, S.; Lin, J. Optimum resistance analysis and experimental veriﬁcation of nonlinear piezoelectric energy harvesting from human motions. Energy 2017, 118, 221–230. [CrossRef] 28. Lin, J.; Liu, H.; Chen, T.; Yang, Z.; Sun, L. A rotational wearable energy harvester for human motion. In Proceedings of the 2017 IEEE 17th International Conference on Nanotechnology (IEEE-NANO), Pittsburgh, PA, USA, 25–28 July 2017. [CrossRef] 29. Pillatsch, P.; Yeatman, E.M.; Holmes, A.S. A piezoelectric frequency up-converting energy harvester with rotating proof mass for human body applications. Sens. Actuator A Phys. 2014, 206, 178–185. [CrossRef] 31 of 37 Sensors 2020, 20, 3512 30. Pillatsch, P.; Yeatman, E.M.; Holmes, A.S.; Wright, P.K. Wireless power transfer system for a human motion energy harvester. Sens. Actuator A Phys. 2016, 244, 77–85. [CrossRef] 31. Halim, M.; Rantz, R.; Zhang, Q.; Gu, L.; Yang, K.; Roundy, S. An electromagnetic rotational energy harvester using sprung eccentric rotor, driven by pseudo-walking motion. Appl. Energy 2018, 217, 66–74. [CrossRef] 31. Halim, M.; Rantz, R.; Zhang, Q.; Gu, L.; Yang, K.; Roundy, S. An electromagnetic rotational energy harvester using sprung eccentric rotor, driven by pseudo-walking motion. Appl. Energy 2018, 217, 66–74. [CrossRef] 32. Fan, K.; Cai, M.; Liu, H.; Zhang, Y. Capturing energy from ultra-low frequency vibrations and human motion 32. Fan, K.; Cai, M.; Liu, H.; Zhang, Y. Capturing energy from ultra-low frequency vibrations and human motion through a monostable electromagnetic energy harvester. Energy 2019, 169, 356–368. [CrossRef] , ; , ; , ; g, p g gy q y through a monostable electromagnetic energy harvester. Energy 2019, 169, 356–368. [CrossRef] 33. Brunner, S.; Gerst, M.; Pylatiuk, C. Design of a body energy harvesting system for the upper extremity. Curr. Dir. Biomed. Eng. 2017, 3. [CrossRef] 34. References Resistive Impedance Matching Circuit for Piezoelectric Energy H ti J I t ll M t S t St t 2010 21 1293 1302 [C R f] 0. Kong, N.; Ha, D.S.; Erturk, A.; Inman, D.J. Resistive Impedance Matching Circuit for Piezoelectric Ene Harvesting. J. Intell. Mater. Syst. Struct. 2010, 21, 1293–1302. [CrossRef] 61. Guyomar, D.; Badel, A.; Lefeuvre, E.; Richard, C. Toward energy harvesting using active materials and conversion improvement by nonlinear processing. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2005, 52, 584–595. [CrossRef] [PubMed] 62. Priya, S.; Song, H.-C.; Zhou, Y.; Varghese, R.; Chopra, A.; Kim, S.-G.; Polcawich, R.G. A Review on Piezoelectric Energy Harvesting: Materials, Methods, and Circuits. Energyo 2019. [CrossRef] 63. Mishra, S.; Unnikrishnan, L.; Nayak, S.K.; Mohanty, S. Advances in Piezoelectric Polymer Composites for Energy Harvesting Applications: A Systematic Review. Macromol. Mater. Eng. 2018, 304, 1800463. [CrossRef] 64 Sappati K ; Bhadra S Piezoelectric Polymer and Paper Substrates: A Review Sensors 2018 18 3605 63. Mishra, S.; Unnikrishnan, L.; Nayak, S.K.; Mohanty, S. Advances in Piezoelectric Polymer Composites for Energy Harvesting Applications: A Systematic Review. Macromol. Mater. Eng. 2018, 304, 1800463. [CrossRef] 64. Sappati, K.; Bhadra, S. Piezoelectric Polymer and Paper Substrates: A Review. Sensors 2018, 18, 3605. [CrossRef] [PubMed] 64. Sappati, K.; Bhadra, S. Piezoelectric Polymer and Paper Substrates: A Review. Sensors 2018, 18, 3605. [CrossRef] [PubMed] 5. Shenck, N.; Paradiso, J. Energy scavenging with shoe-mounted piezoelectrics. J. IEEE Micro 2001, 21, 30 [CrossRef] 66. Rödel, J.; Webber, K.G.; Dittmer, R.; Jo, W.; Kimura, M.; Damjanovic, D. Transferring lead-free piezo ceramics into application. J. Eur. Ceram. Soc. 2015, 35, 1659–1681. [CrossRef] 67. Ramadan, K.S.; Sameoto, D.; Evoy, S. A review of piezoelectric polymers as functional materials for electromechanical transducers. Smart Mater. Struct. 2014, 23, 033001. [CrossRef] 68. Yoon, S.; Shin, D.-J.; Ko, Y.-H.; Cho, K.-H.; Koh, J.-H. Flexible Energy Harvester Based on Poly (vinylidene ﬂuoride) Composite Films. J. Nanosci. Nanotechnol. 2019, 19, 1289–1294. [CrossRef] 69. Vatansever, D.; Hadimani, R.L.; Shah, T.; Siores, E. An investigation of energy harvesting from renewable sources with PVDF and PZT. Smart Mater. Struct. 2011, 20, 055019. [CrossRef] 70. Wang, Z.L. Piezoelectric Nanogenerators Based on Zinc Oxide Nanowire Arrays. Science 2006, 312, 242–246. [CrossRef] 71. Rajagopalan, P.; Singh, V.; Palani, I.A. Enhancement of ZnO-based ﬂexible nano generators via a sol–gel technique for sensing and energy harvesting applications. Nanotechnology 2018, 29, 105406. [CrossRef] [PubMed] 72. Bhatia, D.; Sharma, H.; Meena, R.; Palkar, V. References In Proceedings of the 2019 IEEE International Conference on Mechatronics (ICM), Ilmenau, Germany, 18–20 March 2019. [CrossRef] 49. Gareh, S.; Kok, B.C.; Yee, M.H.; Borhana, A.A.; Alswed, S.K. Optimization of the Compression-Based Piezoelectric Traﬃc Model (CPTM) for Road Energy Harvesting Application. Int. J. Renew. Energy Res. 2019, 9, 1272–1282. 50. Uchino, K.; Ishii, T. Energy Flow Analysis in Piezoelectric Energy Harvesting Systems. Ferroelectrics 2010, 400, 305–320. [CrossRef] 51. Wu, L.; Do, X.-D.; Lee, S.-G.; Ha, D.S. A Self-Powered and Optimal SSHI Circuit Integrated with an Active Rectiﬁer for Piezoelectric Energy Harvesting. IEEE Trans. Circuits Syst. I Regul. Pap. 2017, 64, 537–549. [CrossRef] 52. Erturk, A.; Inman, D.J. 1.1 Vibration-Based Energy Harvesting Using Piezoelectric Transduction. In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 1–4. 32 of 37 32 of 37 Sensors 2020, 20, 3512 53. Piliposian, G.; Hasanyan, A.; Piliposyan, D. The eﬀect of the location of piezoelectric patches on the sensing, actuating and energy harvesting properties of a composite plate. J. Phys. D Appl. Phys. 2019, 52, 445501. [CrossRef] 54. Zhang, G.; Gao, S.; Liu, H. A utility piezoelectric energy harvester with low frequency and high-output voltage: Theoretical model, experimental veriﬁcation and energy storage. AIP Adv. 2016, 6, 095208. [CrossRef] 55. Woo, M.S.; Ahn, J.H.; Eom, J.H.; Hwang, W.S.; Kim, J.H.; Yang, C.H.; Sung, T.H. Study on increasing output current of piezoelectric energy harvester by fabrication of multilayer thick ﬁlm. Sens. Actuator A Phys. 2018, 269, 524–534. [CrossRef] 56. Akkaya Oy, S. A design of mass-spring type piezoelectric energy harvesting. Sci. Iran. 2020. [CrossRef] 57. Starner, T. Human-powered wearable computing. IBM Syst. J. 1996, 35, 618–629. [CrossRef] 56. Akkaya Oy, S. A design of mass-spring type piezoelectric energy harvesting. Sci. Iran. 2020. [CrossRef] r, T. Human-powered wearable computing. IBM Syst. 58. Plessis, A.J.D.; Huigsloot, M.J.; Discenzo, F.D. Resonant packaged piezoelectric power harvester for machinery health monitoring. In Proceedings of the Smart Structure and Materials 2005: Industrial and Commercial Applications of Smart Structures Technologies, San Diego, CA, USA, 5 May 2005. [CrossRef] 9. Ottman, G.; Hofmann, H.; Bhatt, A.; Lesieutre, G. Adaptive piezoelectric energy harvesting circuit wireless remote power supply. IEEE Trans. Power Electron. 2002, 17, 669–676. [CrossRef] 59. Ottman, G.; Hofmann, H.; Bhatt, A.; Lesieutre, G. Adaptive piezoelectric energy harvesting circuit for wireless remote power supply. IEEE Trans. Power Electron. 2002, 17, 669–676. [CrossRef] 60. Kong, N.; Ha, D.S.; Erturk, A.; Inman, D.J. References A novel ZnO piezoelectric microcantilever energy scavenger: Fabrication and characterization. Sens. Bio-Sens. Res. 2016, 9, 45–52. [CrossRef] 73. Chen, Y.; Zhang, Y.; Zhang, L.; Ding, F.; Schmidt, O.G. Scalable single crystalline PMN-PT nanobelts sculpted from bulk for energy harvesting. Nano Energy 2017, 31, 239–246. [CrossRef] 74. Yang, Z.; Zu, J. Comparison of PZN-PT, PMN-PT single crystals and PZT ceramic for vibration energy harvesting. Energy Convers. Manag. 2016, 122, 321–329. [CrossRef] 75. Karami, M.A.; Bilgen, O.; Inman, D.J.; Friswell, M.I. Experimental and analytical parametric study of single-crystal unimorph beams for vibration energy harvesting. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2011, 58, 1508–1520. [CrossRef] 33 of 37 33 of 37 Sensors 2020, 20, 3512 76. Duan, Z.-X.; Yu, G.-Q.; Liu, J.-B.; Liu, J.; Dong, X.-W.; Han, L.; Jin, P.-Y. Preparation and characterization of PZT thick ﬁlm enhanced by ZnO nanowhiskers for MEMS piezoelectric generators. Prog. Nat. Sci. 2011, 21, 159–163. [CrossRef] Baek, C.; Park, H.; Yun, J.H.; Kim, D.K.; Park, K.-I. Lead-free BaTiO3 Nanowire Arrays-based Piezoelectric Energy Harvester. MRS Adv. 2017, 2, 3415–3420. [CrossRef] 78. Lee, G.; Shin, D.-J.; Kwon, Y.-H.; Jeong, S.-J.; Koh, J.-H. Optimized piezoelectric and structural properties of (Bi,Na)TiO3–(Bi,K)TiO3 ceramics for energy harvester applications. Ceram. Int. 2016, 42, 14355–14363. [CrossRef] 79. Zheng, M.; Hou, Y.; Chao, L.; Zhu, M. Piezoelectric KNN ceramic for energy harvesting from mechanochemically activated precursors. J. Mater. Sci. Mater. Electron. 2018, 29, 9582–9587. [CrossRef] 80. Kim, J.; Kim, J.; Han, S.; Kang, H.-W.; Lee, H.-G.; Cheon, C. (K,Na)NbO3-based ceramics with excess alkali oxide for piezoelectric energy harvester. Ceram. Int. 2016, 42, 5226–5230. [CrossRef] 81. Sodano, H.A.; Granstrom, J.; Feenstra, J.; Farinholt, K. Harvesting of Electrical Energy from a Backpack Using Piezoelectric Shoulder Straps; Active and Passive Smart Structures and Integrated Systems: San Diego, CA, USA, 2007. [CrossRef] 82. Keat, C.K.; Leong, K.S.; Tee, L.K. An experimental investigation of piezoelectric P(VDF-TrFE) thick ﬁlm on ﬂexible substrate as energy harvester. IOP Conf. Ser. Mater. Sci. Eng. 2017, 210, 012078. [CrossRef] 83. Piezo Film Sensors Technical Manual. Available online: https://www.sparkfun.com/datasheets/Sensors/Flex/ MSI-techman.pdf (accessed on 19 May 2020). 84. Jain, A.; KJ, P.; Sharma, A.K.; Jain, A.; PN, R. Dielectric and piezoelectric properties of PVDF/PZT composites: A review. Polym. Eng. Sci. 2015, 55, 1589–1616. [CrossRef] 85. Crossley, S.; Whiter, R.A.; Kar-Narayan, S. Polymer-based nanopiezoelectric generators for energy harvesting applications. Mater. Sci. Technol. 2014, 30, 1613–1624. [CrossRef] 86. Erturk, A.; Inman, D.J. References Performance Analysis of Piezoelectric Energy Har in Pavement: Laboratory Testing and Field Simulation. Transp. Res. Rec. 2019, 2673, 115–124. [Cros 100. Yesner, G.; Jasim, A.; Wang, H.; Basily, B.; Maher, A.; Safari, A. Energy harvesting and evaluation of a novel piezoelectric bridge transducer. Sens. Actuator A Phys. 2019, 285, 348–354. [CrossRef] 101. Gareh, S.; Kok, B.; Uttraphan, C.; Thong, K.; Borhana, A. Evaluation of piezoelectric energy harvester outcomes in road traﬃc applications. In Proceedings of the 4th IET Clean Energy and Technology Conference (CEAT 2016), Kuala Lumpur, Malaysia, 14–15 November 2016. [CrossRef] 102. Nguyen, C.H.; Hanke, U.; Halvorsen, E. Actuation of Piezoelectric Layered Beams With d31 and d33 Coupling. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2018, 65, 815–827. [CrossRef] 103. Jenkins, K.; Nguyen, V.; Zhu, R.; Yang, R. Piezotronic Eﬀect: An Emerging Mechanism for Sensing Applications. Sensors 2015, 15, 22914–22940. [CrossRef] 04. Wang, X.; Zhou, J.; Song, J.; Liu, J.; Xu, N.; Wang, Z.L. Piezoelectric Field Eﬀect Transistor and Nanof Sensor Based on a Single ZnO Nanowire. Nano Lett. 2006, 6, 2768–2772. [CrossRef] 104. Wang, X.; Zhou, J.; Song, J.; Liu, J.; Xu, N.; Wang, Z.L. Piezoelectric Field Eﬀect Transistor and Nanoforce Sensor Based on a Single ZnO Nanowire. Nano Lett. 2006, 6, 2768–2772. [CrossRef] 105. Ahn, J.-H.; Shin, D.-J.; Koh, J.-H. Comparative study on the thickness dependent output energy for 05. Ahn, J.-H.; Shin, D.-J.; Koh, J.-H. Comparative study on the thickness dependent output energy (Bi,Sc)O3-(Pb,Ti)O3 multilayered structure. Ceram. Int. 2017, 43. [CrossRef] 106. Zhao, H.; Hou, Y.; Yu, X.; Fu, J.; Zheng, M.; Zhu, M. A wide temperature insensitive piezoceramics for high-temperature energy harvesting. J. Am. Ceram. Soc. 2019, 102, 5316–5327. [CrossRef] 107. Priya, S. Criterion for material selection in design of bulk piezoelectric energy harvesters. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2010, 57, 2610–2612. [CrossRef] 108. Gao, X.; Wu, J.; Yu, Y.; Chu, Z.; Shi, H.; Dong, S. Giant Piezoelectric Coeﬃcients in Relaxor Piezoelectric Ceramic PNN-PZT for Vibration Energy Harvesting. Adv. Funct. Mater. 2018, 28, 1706895. [CrossRef] 109. Gupta, S.; Bhunia, R.; Fatma, B.; Maurya, D.; Singh, D.; Prateek; Gupta, R.; Priya, S.; Gupta, R.K.; Garg, A. Multifunctional and Flexible Polymeric Nanocomposite Films with Improved Ferroelectric and Piezoelectric Properties for Energy Generation Devices. ACS Appl. Energy Mater. 2019, 2, 6364–6374. [CrossRef] 110. Briscoe, J.; Dunn, S. 3.3.1 Mechanical Input: Methods and Characterization of Device Straining. References 10.1.1 Eﬀective Parameters of Various Soft Ceramics and Single Crystals. In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 301–3022. 87. Li, S.; Yuan, J.; Lipson, H. Ambient wind energy harvesting using cross-ﬂow ﬂuttering. J. Appl. Phys. 2011, 109, 026104. [CrossRef] 8. Kim, H.W.; Priya, S.; Uchino, K.; Newnham, R.E. Piezoelectric Energy Harvesting under High Pre-Stre Cyclic Vibrations. J. Electroceram. 2005, 15, 27–34. [CrossRef] 89. Mohammadi, F.; Khan, A.; Cass, R.B. Electronics on Unconventional Substrates-Electrotextiles and Giant-Area Flexible Circuits; Shur, M.S., Wilson, P.M., Urban, D., Eds.; Materials Research Society: Boston, MA, USA, 2003; Volume 736, pp. 263–268. 90. Erturk, A.; Bilgen, O.; Inman, D.J. Power generation and shunt damping performance of a single crystal lead magnesium niobate-lead zirconate titanate unimorph: Analysis and experiment. Appl. Phys. Lett. 2008, 93, 224102. [CrossRef] 91. Xu, C.; Ren, B.; Di, W.; Liang, Z.; Jiao, J.; Li, L.; Wang, D. Cantilever driving low frequency piezoelectric energy harvester using single crystal material 0.71Pb(Mg1/3Nb2/3)O3-0.29PbTiO3. Appl. Phys. Lett. 2012, 101, 033502. [CrossRef] 92. Erturk, A.; Inman, D.J. 10.4.4 Overall Comparision of Ceramics (PZT-5H, PZT-8) and Single Crystals (PMN-PZT, PMN-PZT-Mn). In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 315–317. 93. Cook-Chennault, K.A.; Thambi, N.; Sastry, A.M. Powering MEMS portable devices—A review of non-regenerative and regenerative power supply systems with special emphasis on piezoelectric energy harvesting systems. Smart Mater. Struct. 2008, 17, 043001. [CrossRef] 94. Feenstra, J.; Granstrom, J.; Sodano, H. Energy harvesting through a backpack employing a mechanically ampliﬁed piezoelectric stack. Mech. Syst. Signal Process. 2008, 22, 721–734. [CrossRef] 95. Liu, J.; O’Connor, W.J.; Ahearne, E.; Byrne, G. Electromechanical modelling for piezoelectric ﬂextensional actuators. Smart Mater. Struct. 2013, 23, 025005. [CrossRef] 96. Wang, Y.; Chen, W.; Guzman, P. Piezoelectric stack energy harvesting with a force ampliﬁcation frame: Modeling and experiment. J. Intell. Mater. Syst. Struct. 2016, 27, 2324–2332. [CrossRef] 97. Li, H.; Tian, C.; Deng, Z.D. Energy harvesting from low frequency applications using piezoelectric materials. Appl. Phys. Rev. 2014, 1, 041301. [CrossRef] 34 of 37 34 of 37 Sensors 2020, 20, 3512 98. Sodano, H.A.; Park, G.; Leo, D.J.; Inman, D.J. Use of piezoelectric energy harvesting devices for charging batteries. In Proceedings of the Smart Structures and Materials 2003: Smart Sensor Technology and Measurement Systems, San Diego, CA, USA, 3–5 March 2003. [CrossRef] 99. Jasim, A.F.; Wang, H.; Yesner, G.; Safari, A.; Szary, P. References In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 43–45. 111. Briscoe, J.; Dunn, S. 3.3.2 Measurement of Nanogenerator Output. In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 45–48. 112. Briscoe, J.; Dunn, S. 3.3.3 Electrical Characterization. In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 48–49. 113. Fleming, A.J.; Leang, K.K. Introduction. In Design, Modeling and Control of Nanopositioning Systems Advances in Industrial Control 2014; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 1–15. [CrossRef] 114. Roundy, S. On the Eﬀectiveness of Vibration-based Energy Harvesting. J. Intell. Mater. Syst. Struct. 2005, 16, 809–823. [CrossRef] 115. Goldfarb, M.; Celanovic, N. A Lumped Parameter Electromechanical Model for Describing the Nonlinear Behavior of Piezoelectric Actuators. J. Dyn. Syst. Meas. Control 1997, 119, 478–485. [CrossRef] 116. Goldfarb, M.; Celanovic, N. Modeling piezoelectric stack actuators for control of micromanipulation. IEEE Control Syst. 1997, 17, 69–79. [CrossRef] 117. Ruderman, M.; Kamiya, Y.; Iwasaki, M. Extended lumped parameter electromechanical model of piezoelectric actuators. In Proceedings of the 2015 IEEE International Conference on Mechatronics (ICM), Nagoya, Japan, 6–8 March 2015. [CrossRef] 118. Aloui, R.; Larbi, W.; Chouchane, M. Global sensitivity analysis of piezoelectric energy harvesters. Compos. Struct. 2019, 228, 111317. [CrossRef] 119. Erturk, A.; Inman, D.J. A Distributed Parameter Electromechanical Model for Cantilevered Piezoelectric Energy Harvesters. J. Vib. Acoust. 2008, 130. [CrossRef] 120. De Marqui, C., Jr.; Erturk, A.; Inman, D.J. An electromechanical ﬁnite element model for piezoelectric energy harvester plates. J. Sound Vib. 2009, 327, 9–25. [CrossRef] Sensors 2020, 20, 3512 35 of 37 121. Rosa, M.; De Marqui, C., Jr. Modeling and Analysis of a Piezoelectric Energy Harvester with Varying Cross-Sectional Area. Shock Vib. 2014, 2014, 1–9. [CrossRef] 122. Larbi, W.; Deü, J.-F. Reduced order ﬁnite element formulations for vibration reduction using piezoelectric shunt damping. Appl. Acoust. 2019, 147, 111–120. [CrossRef] 123. Łasecka-Plura, M.; Lewandowski, R. Design sensitivity analysis of frequency response functions and steady-state response for structures with viscoelastic dampers. Vib. Phys. Syst. 2014, 26, 129–136. 124. Lima, A.D.; Faria, A.; Rade, D. Sensitivity analysis of frequency response functions of composite sandwich plates containing viscoelastic layers. Compos. Struct. 2010, 92, 364–376. [CrossRef] 125. Campolongo, F.; Saltelli, A.; Cariboni, J. From screening to quantitative sensitivity analysis. A uniﬁed approach. Comput. Phys. Commun. 2011, 182, 978–988. [CrossRef] 126. Mösch, M.; Fischerauer, G.; Hoﬀmann, D. A Self-adaptive and Self-Suﬃcient Energy Harvesting System. References Sensors 2020, 20, 2519. [CrossRef] 127. He, M.; Wang, S.; Zhong, X.; Guan, M. Study of a Piezoelectric Energy Harvesting Floor Structure with Force Ampliﬁcation Mechanism. Energies 2019, 12, 3516. [CrossRef] 128. Hemmasian Ettefagh, M.; Boudaud, M.; Bazaei, A.; Chen, Z.; Régnier, S. Piezoelectric Ceramic Transducers as Time-Varying Displacement Sensors in Nanopositioners. Sens. Transducers 2019, 237, 30–36. 129. Wu, X.; Kato, S.; Zhang, K.; Ren, T.; Lin, J.; Liu, L. A frequency adjustable vibration energy ha In Proceedings of the PowerMEMS 2008, Sendai, Japan, 9–12 November 2008; pp. 245–248. 130. Karadag, C.V.; Topaloglu, N. A self-suﬃcient and frequency tunable piezoelectric vibration energy harvester. J. Vib. Acoust. 2016, 139. [CrossRef] 131. Jang, S.J.; Kim, I.H.; Jung, H.J.; Lee, Y.P. A tunable rotational energy harvester for low frequency vibration. Appl. Phys. Lett. 2011, 99, 134102. [CrossRef] 132. Gieras, J.F.; Oh, J.H.; Huzmezan, M.; Sane, H.S. Electromechanical Energy Harvesting System. U.S. Patent 8,222,775, 17 July 2012. 133. Leland, E.S.; Wright, P.K. Resonance tuning of piezoelectric vibration energy scavenging generators using compressive axial preload. Smart Mater. Struct. 2006, 15, 1413–1420. [CrossRef] 134. Morris, D.J.; Youngsman, J.M.; Anderson, M.J.; Bahr, D.F. A resonant frequency tunable, extensional mode piezoelectric vibration harvesting mechanism. Smart Mater. Struct. 2008, 17, 065021. [CrossRef] 135. Zhu, D.; Tudor, M.J.; Beeby, S.P. Strategies for increasing the operating frequency range of vibration energy harvesters: A review. Meas. Sci. Technol. 2009, 21, 022001. [CrossRef] 136. Hsu, J.-C.; Tseng, C.-T.; Chen, Y.-S. Analysis and experiment of self-frequency-tuning piezoelectric energy harvesters for rotational motion. Smart Mater. Struct. 2014, 23, 075013. [CrossRef] 137. Madinei, H.; Khodaparast, H.H.; Adhikari, S.; Friswell, M. Design of MEMS piezoelectric harvesters with electrostatically adjustable resonance frequency. Mech. Syst. Signal Process. 2016, 81, 360–374. [CrossRef] 138. Challa, V.R.; Prasad, M.G.; Shi, Y.; Fisher, F.T. A vibration energy harvesting device with bidirectional resonance frequency tunability. Smart Mater. Struct. 2008, 17, 015035. [CrossRef] 139. Sari, I.; Balkan, T.; Kulah, H. An electromagnetic micro power generator for wideband environmental vibrations. Sens. Actuator A Phys. 2008, 145–146, 405–413. [CrossRef] 140. Toyabur, R.; Salauddin, M.; Cho, H.; Park, J.Y. A multimodal hybrid energy harvester based on piezoelectric-electromagnetic mechanisms for low-frequency ambient vibrations. Energy Convers. Manag. 2018, 168, 454–466. [CrossRef] 141. Saha, C.; O’Donnell, T.; Wang, N.; Mccloskey, P. Electromagnetic generator for harvesting energy from human motion. Sens. Actuator A Phys. 2008, 147, 248–253. [CrossRef] 142. Andò, B.; Baglio, S.; Trigona, C.; Dumas, N.; Latorre, L.; Nouet, P. References Nonlinear mechanism in MEMS devices for energy harvesting applications. J. Micromech. Microeng. 2010, 20, 125020. [CrossRef] 143. Mann, B.; Owens, B. Investigations of a nonlinear energy harvester with a bistable potential well. J. Sound Vib. 2010, 329, 1215–1226. [CrossRef] 144. Sebald, G.; Kuwano, H.; Guyomar, D.; Ducharne, B. Experimental Duﬃng oscillator for broadband piezoelectric energy harvesting. Smart Mater. Struct. 2011, 20, 102001. [CrossRef] 145. Sato, T.; Watanabe, K.; Igarashi, H. Coupled Analysis of Electromagnetic Vibration Energy Harvester with Nonlinear Oscillation. IEEE Trans. Magn. 2014, 50, 313–316. [CrossRef] 146. Barton, D.A.W.; Burrow, S.G.; Clare, L.R. Energy Harvesting from Vibrations with a Nonlinear Oscillator. J. Vib. Acoust. 2010, 132. [CrossRef] Sensors 2020, 20, 3512 36 of 37 147. Nguyen, D.S.; Halvorsen, E.; Jensen, G.U.; Vogl, A. Fabrication and characterization of a wideband MEMS energy harvester utilizing nonlinear springs. J. Micromech. Microeng. 2010, 20, 125009. [CrossRef] 148. Leadenham, S.; Erturk, A. M-shaped asymmetric nonlinear oscillator for broadband vibration energy harvesting: Harmonic balance analysis and experimental validation. J. Sound Vib. 2014, 333, 6209–6223. [CrossRef] 149. Zhang, J.; Zhang, J.; Shu, C.; Fang, Z. Enhanced piezoelectric wind energy harvesting based on a buckled beam. Appl. Phys. Lett. 2017, 110, 183903. [CrossRef] 150. Panyam, M.; Daqaq, M.F.; Emam, S.A. Exploiting the subharmonic parametric resonances of a buckled beam for vibratory energy harvesting. Meccanica 2018, 53, 3545–3564. [CrossRef] 151. Yildirim, T.; Ghayesh, M.H.; Li, W.; Alici, G. A review on performance enhancement techniques for ambient vibration energy harvesters. Renew. Sustain. Energy Rev. 2017, 71, 435–449. [CrossRef] 152. Liu, H.; Lee, C.; Kobayashi, T.; Tay, C.J.; Quan, C. Investigation of a MEMS piezoelectric energy harvester system with a frequency-widened-bandwidth mechanism introduced by mechanical stoppers. Smart Mater. Struct. 2012, 21, 035005. [CrossRef] 153. Soliman, M.S.M.; Abdel-Rahman, E.M.; El-Saadany, E.F.; Mansour, R.R. A wideband vibration-based energy harvester. J. Micomech. Microeng. 2008, 18, 115021. [CrossRef] 153. Soliman, M.S.M.; Abdel-Rahman, E.M.; El-Saadany, E.F.; Mansour, R.R. A wideband vibration-based energy harvester. J. Micomech. Microeng. 2008, 18, 115021. [CrossRef] 154. Halim, M.A.; Cho, H.; Park, J.Y. Design and experiment of a human-limb driven, frequency up-converted 54. Halim, M.A.; Cho, H.; Park, J.Y. Design and experiment of a human-limb driven, frequency up-conve electromagnetic energy harvester. Energy Convers. Manag. 2015, 106, 393–404. [CrossRef] 155. Xu, J.; Tang, J. Piezoelectric cantilever-pendulum for multi-directional energy harvesting with internal resonance. In Proceedings of the Active and Passive Smart Structures and Integrated Systems, San Diego, CA, USA, 9–12 March 2015. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References [CrossRef] 156. Zhao, L.-C.; Zou, H.-X.; Yan, G.; Zhang, W.-M.; Peng, Z.-K.; Meng, G. Arbitrary-directional broadband vibration energy harvesting using magnetically coupled ﬂextensional transducers. Smart Mater. Struct. 2018, 27, 095010. [CrossRef] 157. Bowers, B.J.; Arnold, D.P. Spherical, rolling magnet generators for passive energy harvesting from human motion. J. Micromech. Microeng. 2009, 19, 094008. [CrossRef] 158. Mouapi, A.; Hakem, N. Electrical Optimization of Piezoelectric Energy Harvesting System in Vehicles. In Proceedings of the International Conference on Energy, Environment and Economics, Edinburgh, UK, 16–18 August 2016. g 159. DC1459B-A. Available online: http://www.analog.com/en/design-center/evaluation-hardware-and-software/ evaluation-boards-kits/dc1459b-a.html#eb-overview (accessed on 30 April 2020). 160. Briscoe, J.; Dunn, S. 3.2.6 Applications. In Nanostructured Piezoelectric Energy Harvesters, 1st ed.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 39–42. 161. Gljuš´ci´c, P.; Zelenika, S.; Blaževi´c, D.; Kamenar, E. Kinetic Energy Harvesting for Wearable Medical Sensors. Sensors 2019, 19, 4922. [CrossRef] [PubMed] 162. Erturk, A.; Inman, D.J. 11.1 AC-DC Rectiﬁcation and Analysis of the Rectiﬁed Output. In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 325–331. 163. Shu, Y.C.; Lien, I.C. Analysis of power output for piezoelectric energy harvesting systems. Smart Mater. Struct. 2006, 15, 1499–1512. [CrossRef] 164. Erturk, A.; Inman, D.J. 11.2 Two-Stage Energy Harvesting Circuits: DC-DC Conversion for Impedance Matching. In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 331–336. 165. Ottman, G.; Hofmann, H.; Lesieutre, G. Optimized piezoelectric energy harvesting circuit using step-down converter in discontinuous conduction mode. IEEE Trans. Power Electron. 2003, 18, 696–703. [CrossRef] 166. Guan, M.J.; Liao, W.H. On the eﬃciencies of piezoelectric energy harvesting circuits towards storage device voltages. Smart Mater. Struct. 2007, 16, 498–505. [CrossRef] 167. Erturk, A.; Inman, D.J. 11.3 Synchronized Switcing on Inductor for Piezoelectric Energy Harvesting. In Piezoelectric Energy Harvesting, 1st ed.; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 336–340. 168. Liang, J.; Liao, W. Piezoelectric Energy Harvesting and Dissipation on Structural Damping. J. Intell. Mater. Syst. Struct. 2008, 20, 515–527. [CrossRef] 169. Lallart, M.; Guyomar, D. An optimized self-powered switching circuit for non-linear energy harvesting with low voltage output. Smart Mater. Struct. 2008, 17, 035030. [CrossRef] 170. Lefeuvre, E.; Badel, A.; Richard, C.; Guyomar, D. Piezoelectric Energy Harvesting Device Optimization by Synchronous Electric Charge Extraction. J. Intell. Mater. Syst. Struct. 2005, 16, 865–876. [CrossRef] 37 of 37 Sensors 2020, 20, 3512 171. Edlund, C.; Ramakrishnan, S. 172. Cha, Y.; Hong, J.; Lee, J.; Park, J.-M.; Kim, K. Flexible Piezoelectric Energy Harvesting from Mouse Click Motions. Sensors 2016, 16, 1045. [CrossRef] [PubMed] References An analytic study of vibrational energy harvesting using piezoelectric tiles in stairways subjected to human traﬃc. Eur. J. Appl. Math. 2018, 30, 968–985. [CrossRef] 172. Cha, Y.; Hong, J.; Lee, J.; Park, J.-M.; Kim, K. Flexible Piezoelectric Energy Harvesting from Mouse Click Motions. Sensors 2016, 16, 1045. [CrossRef] [PubMed]
https://openalex.org/W2970656668	https://nottingham-repository.worktribe.com/preview/2520626/animals-09-00622.pdf	English	null	Animal Research, Accountability, Openness and Public Engagement: Report from an International Expert Forum	Animals	2,019	cc-by	7,343	animals animals Communication Animal Research, Accountability, Openness and Public Engagement: Report from an International Expert Forum von Keyserlingk 1,* 1 Animal Welfare Program, Faculty of Land and Food Systems, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 1 Animal Welfare Program, Faculty of Land and Food Systems, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 2 Department of Clinical Sciences, Swedish University of Agricultural Sciences, 75007 Uppsala, Sweden 3 Principal Adviser, Animal Welfare, Ministry for Primary Industries, Wellington 6140, New Zealand 4 The Center for Alternatives to Animal Testing (CAAT), Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA 2 Department of Clinical Sciences, Swedish University of Agricultural Sciences, 75007 Uppsala, Sweden 3 Principal Adviser, Animal Welfare, Ministry for Primary Industries, Wellington 6140, New Zealand 4 The Center for Alternatives to Animal Testing (CAAT), Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA 5 School of Sociology and Social Policy, University of Nottingham, Nottingham NG7 2RD, UK gy y, y g , g , 6 Maurice Young Centre for Applied Ethics, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 7 Department of Zoology, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 8 Director Research Governance, South Eastern Sydney and Illawarra Shoalhaven Local Health Districts, Conjoint Professor, University of New South Wales, Sydney 2036, Australia 9 The Humane Society of the United States, Washington, DC 20037, USA * C d i k li k@ b T l 1 604 822 4898 6 Maurice Young Centre for Applied Ethics, University of British Columbia, Vancouver, BC V6T 1Z4, Canada Maurice Young Centre for Applied Ethics, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 7 Department of Zoology, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 8 Director Research Governance, South Eastern Sydney and Illawarra Shoalhaven Local Health Districts, Conjoint Professor, University of New South Wales, Sydney 2036, Australia 9 Th H S i f h U i d S W hi DC 20037 USA 8 Director Research Governance, South Eastern Sydney and Illawarra Shoalhaven Local Health Districts, Conjoint Professor, University of New South Wales, Sydney 2036, Australia j y y y The Humane Society of the United States, Washington, DC 20037, USA j y y y 9 The Humane Society of the United States, Washington, DC 20037, USA * Correspondence: marina.vonkeyserlingk@ubc.ca; Tel.: +1-604-822-4898 Communication Animal Research, Accountability, Openness and Public Engagement: Report from an International Expert Forum Communication Elisabeth H. Ormandy 1, Daniel M. Weary 1, Katarina Cvek 2, Mark Fisher 3, Kathrin Herrmann 4, Pru Hobson-West 5 , Michael McDonald 6, William Milsom 7, Margaret Rose 8, Andrew Rowan 9, Joanne Zurlo 4 and Marina A.G. Animals 2019, 9, 622; doi:10.3390/ani9090622 animals animals 1. Introduction The use of animals in research presents an interesting balance between the need for openness and conﬁdentiality. Previous work has shown that during a crisis transparent communications lead to higher levels of public trust [1]. Sharing details about animal-based research practices can add legitimacy to such research by increasing public trust. Historically, sharing details about animal research was thought to put institutions at increased risk of exposure to animal rights activism [2,3]. However, such targeting by animal rights activists is currently at a worldwide low [4]. There is also evidence that public acceptance of animal research has continually declined over the past two decades [5,6], perhaps associated with a decrease in public trust of animal research. In 2012, the University of British Columbia (UBC) became the ﬁrst University in North America to publish the species and numbers of animals used in research on campus in a given year, along with information on the level of invasiveness of the procedures used on the animals listed. While UBC made the move to increased openness, it has received criticism because of the controlled nature of the information sharing, and some argued that this eﬀort does not go far enough in terms of promoting public engagement about animal research [7,8]. Practices of “selective openness” seem to be the norm within the culture of the animal research community [3]. Calls for public governance of science have become increasingly prevalent for at least the last 15 years [9–12]. Community engagement in the governance of controversial research is now widely acknowledged to be an important part of the public governance of science [13–15]. Eﬀorts toward public engagement generally seek to democratize science policy-making by expanding expert-driven conversations to include lay citizens [16]. Despite the public attention that this issue has received, and the recognition of a need for public accountability in science generally [17], little consensus has developed on how to better engage members of the public on issues related to animal-based research. Meaningful public engagement would seem to be unrealistic without information sharing. Calls for greater openness have long been central to the discourses of critics: what has now changed is the extent to which this agenda is being embraced by the scientiﬁc community, at least in the UK [18]. Received: 18 July 2019; Accepted: 20 August 2019; Published: 29 August 2019 Received: 18 July 2019; Accepted: 20 August 2019; Published: 29 August 2019 Simple Summary: The issues of openness, transparency and public engagement about animal research have taken focus in several diﬀerent countries in recent years. This paper gives an account of a two-day-long expert forum that brought together policy experts and academics from Australia, Canada, Germany, New Zealand, Sweden, the United Kingdom and the United States. The aim was to share current governance practices regarding openness and transparency of animal research and to brainstorm ideas for better public engagement. The facilitated conversations were transcribed and analysed to create this report and recommendations that encourage international policy-makers and other stakeholders to engage in genuine dialogue about the use of animals in research. of a two day long expert forum that brought together policy experts and academics from Australia, Canada, Germany, New Zealand, Sweden, the United Kingdom and the United States. The aim was to share current governance practices regarding openness and transparency of animal research and to brainstorm ideas for better public engagement. The facilitated conversations were transcribed and analysed to create this report and recommendations that encourage international policy-makers and other stakeholders to engage in genuine dialogue about the use of animals in research. Abstract: In November 2013, a group of international experts in animal research policy (n = 11) gathered in Vancouver, Canada, to discuss openness and accountability in animal research. The primary objective was to bring together participants from various jurisdictions (United States, Sweden, Australia, New Zealand, Germany, Canada and the United Kingdom) to share practices regarding the governance of animals used in research, testing and education, with emphasis on the governance process followed, the methods of community engagement, and the balance of openness versus conﬁdentiality. During the forum, participants came to a broad consensus on the need for: (a) evidence-based metrics to allow a “virtuous feedback” system for evaluation and quality assurance of animal research, (b) the need for increased public access to information, together with opportunities for stakeholder dialogue about animal research, (c) a greater diversity of views to be represented on decision-making committees to allow for greater balance and (d) a standardized and robust ethical decision-making process that incorporates some sort of societal input. These recommendations encourage aspirations beyond merely imparting information and towards a genuine dialogue that represents a shared agenda surrounding laboratory animal use. Received: 18 July 2019; Accepted: 20 August 2019; Published: 29 August 2019 Animals 2019, 9, 622; doi:10.3390/ani9090622 www.mdpi.com/journal/animals www.mdpi.com/journal/animals 2 of 12 Animals 2019, 9, 622 Keywords: animal ethics; animal experimentation; animal welfare; governance; policy; public engagement Keywords: animal ethics; animal experimentation; animal welfare; governance; policy; public engagement 2.1. Recruitment of Expert Participants Participants were recruited via email on the basis of their academic expertise or their current role in policy-making. All participants were required to have working knowledge of the governance of animal research in their country of residence, and before the forum each participant provided a written summary of their governance system. Of the 11 participants, 7 were primary involved in academia, and 4 in policy, but all had previously contributed to either regional or national policy-making processes. The aim was to recruit participants who represent those countries with a range of governance systems for the use of animals in science, so participants from the USA, UK, Canada, Sweden, Australia, New Zealand and Germany were recruited. Together, these seven countries represent a variety of governance styles ranging from fully legislated (e.g., UK), to partially legislated (e.g., USA) to non-legislated (e.g., Canada), a range of national systems, from state governance (e.g., Australia, Germany) to national governance (e.g., UK), as well as varying degrees of openness, from more open (e.g., Sweden) to less open (e.g., USA). More participants could have been recruited, from within these countries and from other nations. However, the organizers wanted to balance the diversity of viewpoints and ensure the ability for constructive discussion over a two-day forum. 1. Introduction For example, the Basel Declaration [19] (signed on 29 November 2010 by a group of 80 biomedical scientists from Sweden, Germany, Switzerland, France and Great Britain following a meeting to discuss issues related to animal research) committed to “promote the dialogue concerning animal welfare in research by transparent and fact-based communications to the public.” Similarly, at the 8th World Congress on Alternatives and Animal Use in the Life Sciences, held in Montreal in 2011, attendees passed the Montreal Declaration calling for an “increase in the transparency of the translation of animal-based research” [20]. In 2012, over 40 organizations involved in bioscience signed The UK Concordat on Openness on Animal Research and committed to be more open about the ways in which animals are used in scientiﬁc, medical and veterinary research. Signatories now number 120. As part of the process, members of the UK public were consulted on the four commitments in the Concordat: (1) “we will be clear about how, when and why we use animals in research”, (2) “we will enhance our communication with the media and the public about our research using animals”, (3) “we will be proactive in providing opportunities for the public to ﬁnd out about research using animals” and (4) “we will report our progress annually and share our experiences” [21]. Similar initiatives have been signed in Belgium, Spain and Portugal [22–24]. There was also public dialogue on openness and animal research, which explored the public’s views on openness and transparency in animal research and found that “The public want the sector to demonstrate its commitment to openness by creating Animals 2019, 9, 622 3 of 12 greater scrutiny of itself” [25]. In response to these initiatives, some authors have called for more discussion of precisely what openness can be expected to achieve [18]. Given the diﬃculties that arise when trying to balance openness and conﬁdentiality, the local nature of this move toward increased openness at UBC provided an incentive for UBC researchers to host a meeting with international experts. The aim of this meeting was to share best practices and discuss international diﬀerences and challenges, and to develop a vision for increasing openness. The meeting also provided an opportunity to articulate why a move towards increased openness is important in terms of building an ethical culture of research and to explore options for more meaningful dialogue about animal research with the broader public. 2.2. Discussion Groups and Facilitation Forum discussions occurred over a two-day period and were divided into small group sessions and plenary sessions. The topics to be discussed were selected by the organizers based on the pre-meeting summaries submitted by the participants. Each topic was presented to the whole group at the beginning of each session. The group then broke out into two smaller groups for discussion. Facilitators moderated the discussion in the small groups, and scribes were present to record the conversation. The scribes wrote notes on a ﬂip chart so that all group members see if the conversation was faithfully captured. Small-group discussions lasted between 45 and 60 min. Once ﬁnished, the small groups reconvened in plenary, and the scribes and facilitators gave feedback to the whole group about what was discussed. This feedback formed the basis of a plenary discussion lasting approximately 60–90 min. All plenary sessions were audio-recorded and transcribed for analysis. 2.3. Analysis Inductive thematic content analysis of the plenary sessions was carried out by the ﬁrst author. The small-group sessions had scribes present to take notes of the topics discussed—these notes were typed up and referred to as part of the analysis. The small groups also gave feedback about their discussions during the plenary sessions to reduce the risk the plenary sessions missed any key points. Thematic content analysis followed the process described by Coﬀey and Atkinson [26] and Burnard et al. [27]. In brief, the transcripts were read and, line by line, labels were assigned to sections (i.e., sentences or phrases) of the text. This method of qualitative coding has been described as the use of “tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study” [28]. The process of tagging was iterative, and some tags were changed as more 4 of 12 Animals 2019, 9, 622 tags were added. The tags were then grouped together into themes to describe the major concepts raised in the discussions. Again, this was an iterative process, and on occasion tags or themes were altered to improve ﬁt and more faithfully represent the content of the conversations. The structure of the forum produced an environment for synergistic discussion among the participants. While direct quotes are used in this paper, they will not be attributed to a speciﬁc individual because the dialogue was co-constructed, meaning that several participants often contributed to the formation of each point. Forum discussion focused on ethics review committees, institutional culture and policy and oversight. 3.1. Ethics Review Committee Participants discussed the structure and functioning of the protocol review committees responsible for approving animal-based research protocols in each of the countries. There was general desire for increased openness and public access to animal research protocols, as well as a desire for balanced selection of committee members and recruitment methods that reduce biases and consider other perspectives. 3.1.1. Protocol Review Committee Structure To relieve any power imbalances between members within protocol review committees (e.g., one community representative versus a majority of university scholars), participants suggested the number of community members be increased, with the intention of increasing the diversity of views. One participant noted that “...at a committee level, it’s more important to have a diversity of views and actors instead of just having more numbers and expect that greater numbers will bring diversity.” Participants also discussed the need for improved recruitment methods to avoid institutional bias in the selection of committee members. As an example, in Sweden local political parties and animal welfare organizations select laypersons to represent community values on ethics review committees. It was also noted that the protocol review process should be independent from the institution where research takes place: “We need more thought on the committee structure to make [for a] more independent review process.” In Sweden, the protocol review committee is not aﬃliated with the institution where the protocol will be implemented, but this was not the case in countries represented by other participants. In all other countries, the protocol review committee is directly aﬃliated with the institution, and community member selection is less democratic (e.g. new community members selected because they were acquaintances of existing committee members). These sentiments echo those highlighted in previous literature (e.g., [29,30]). In particular, Schuppli and Fraser [31] found that, among other factors, the eﬀectiveness of ethics review committees was inﬂuenced by committee composition and dynamics, recruitment of members and member turnover. Schuppli and Fraser [31] highlight a potential bias towards institutional or research interests. This bias results from: (1) membership being dominated by scientists, (2) poor leadership by chairpersons that prevented full participation of community and minority members, which leads to poor committee dynamics, (3) “community members” being aﬃliated with the institution and (4) the motivation of some members to pursue a personal agenda rather than adhere to the committee’s mandate. Other research has looked at the importance of committee structure but considered the role of sub-committees and whether and how these can maximise opportunities for ethical review [32]. 3.1.2. Protocol Review Committee Function There was a discussion on the need for a robust and informed decision-making process; speciﬁcally, a process that people can trust. It was recognized that some decisions to allow animal research to proceed might not be in line with societal values, so there is a need to engage the local and national community in a broader dialogue related to animal research. 5 of 12 Animals 2019, 9, 622 “ . . . there’s a lot riding on the people who actually sit around the table. Whether they’re the researcher, or the community rep, or the veterinarian, or the Chair, there’s a lot tied to that. Individual history, personal experience, personality, lots of things like that. So how do we get to a place where we have a robust decision-making process that is also kind of, you know, almost repeatable?” “I don’t know if the analogy of the jury system might work . . . if you took the same case to 10 diﬀerent juries, you’d get diﬀerent decisions, right? And somehow, for me anyways, that doesn’t bother me. But the case [of] the animal ethics committee maybe bothers me a little bit more. And I think that’s because I’m somehow less trusting that the process being followed in that community is really, fairly reﬂects the, well, that the process works . . . ” Participants speciﬁcally discussed the role of the protocol review committee Chair and highlighted the view that the Chair must provide the leadership to empower all members of the committee to voice their concerns: “We need the right leadership present on a committee to take charge and to speak up on the ethical issues on the protocols and to keep asking the right questions. So, we need a good Chair to allow consensus regarding the protocols we use for approval and rejection.” This was considered especially important when trying to ensure that community representatives are fairly included in the decision-making process and are not overwhelmed with jargon or technical details. It was generally agreed that review of animal research requires a broader value judgment, not just a review of the technical details or the impacts on the animals involved. Broader ethical review may be limited by the expertise of the members or by committees speciﬁcally asking members not to review the social value of the research and only focus on harm mitigation. 3.2.1. Openness within Institutions Participants discussed how to build a virtuous and self-reﬂective system at the institutional level to facilitate decision-making throughout the entire process of doing animal-based research. There was general agreement that institutions in all countries need to build an ethical culture where the burden of decision-making is shared between the protocol review committee and the research community (“we need to share the ethical load . . . there should be a continuum of ethical responsibilities”) and that, globally, governance systems for animal-based science require openness, mentorship and a greater sense of connectedness between individuals within an institution. Participants felt that this would be best achieved through a bottom-up approach. Examples of how to achieve a greater degree of ethical decision-making included providing support (separate from and before protocol review) for young faculty members so that they can initiate their research careers in the ﬁeld of animal-based research in an ethically reﬂective manner: “In a way, what we’re talking about is the . . . deliberate growing of a network of people who can be helpful in various areas . . . There’s (potentially) a list of people who volunteer for this and maybe occasionally—I picture them as occasionally—doing an orientation session or maybe helping out [with] a little bit of education for junior researchers and stuﬀlike that.” In addition to suggestions for orientation sessions, participants suggested that a core advisory team or supporting oﬃce be founded that could guide researchers to other peers who have faced similar problems and to other resources. Participants also identiﬁed the need for institutional support in building a network of people who can be called on to help share expertise and ethical practice: “But there are real, live, people that you’re actually asking to spend their time, and there has to be resourcing for this . . . the institution that has to support it.” It is important to note that the suggested network of people is intended to be an “institutional commitment to changing the culture”, rather than creating a new bureaucracy within institutions. It was also emphasized that greater interdisciplinary communication between departments and researchers should be encouraged to support the building of an ethical culture related to animal research. 3.1.2. Protocol Review Committee Function there should be more of a default to openness, with exceptions for proprietary research that you would write in and say, well, I need to be exempted from this requirement for openness.” 6 of 12 Animals 2019, 9, 622 3.2. Institutional Culture When discussing research institutions, participants tended to talk separately about the need for greater openness within institutions and the need for greater openness between the institutional research community and the public. 3.1.2. Protocol Review Committee Function It has been noted elsewhere that the protocol review process typically focuses more on technical details than on a broader value judgment [33]. The call for a broader value judgment has also been made elsewhere [34,35]. It was also generally agreed by the participants that there should be increased openness from protocol review committee members (and institutions) about: (a) the functioning of the committees and how decisions are arrived at and (b) approved protocols. Participants largely agreed that committee members should not be bound by conﬁdentiality and should be able to share information outside the committee (with the exception of proprietary information). This is particularly important for community representatives who may need broader support for their role in the committee. There was also a suggestion to build on the UK model of publishing lay summaries to create a searchable online resource for all approved protocols, including the numbers and types of animals involved. Providing the public an opportunity to engage and comment on proposed animal research utilizing lay summaries has been tested in Canada [36]. Some researchers may have concern about the public release of certain information in their protocol, so there should be some mechanism to redact the protocol with the provision of footnotes justifying the case. It was agreed that certain information should remain conﬁdential, including the names of any research staﬀ. The discussion about keeping names conﬁdential focused on the perceived risk of animal rights activism; this speaks to the tension between openness and perceived security risk that has been discussed elsewhere [2]. In addition, a certain level of conﬁdentiality for ethics review committees might be important so that members feel safe to engage in open conversation with committee members. Overall, there was agreement that some aspects of animal research should be released into the public domain, whereas others could remain conﬁdential. While there was agreement that lay summaries of approved protocols should be made available to the public, no consensus was reached on which speciﬁc components should be available. However, it was suggested that openness should be the default, but that conﬁdentiality could be requested in special cases to protect researchers’ personal safety and intellectual property, “ . . . 3.2.1. Openness within Institutions The discussion also highlighted that the “Three Rs” principle (Replacement, Reduction and Reﬁnement, see Russell and Burch [37] for description) does not equal ethical review; rather, the implementation of the Three Rs is part of a larger cycle of ethical reﬂection. Figure 1 below summarizes this cycle. 7 of 12 Animals 2019, 9, 622 Figure 1. Proposed virtuous, self-reﬂexive learning loop for animal-based research that would focus on ethical reﬂection. Ethical reﬂection would no longer be limited to the ethics review component (adapted from [38]). Animal manipulation occurs from the procurement of animals to the study termination, so it represents a relatively small part of the process of animal-based research. Figure 1. Proposed virtuous, self-reﬂexive learning loop for animal-based research that would focus on ethical reﬂection. Ethical reﬂection would no longer be limited to the ethics review component (adapted from [38]). Animal manipulation occurs from the procurement of animals to the study termination, so it represents a relatively small part of the process of animal-based research. Figure 1. Proposed virtuous, self-reﬂexive learning loop for animal-based research that would focus on ethical reﬂection. Ethical reﬂection would no longer be limited to the ethics review component (adapted from [38]). Animal manipulation occurs from the procurement of animals to the study termination, so it represents a relatively small part of the process of animal-based research. Participants generally agreed that increased openness would help to raise standards by allowing greater cooperation between researchers and the institution: “I do think the evidence is available, but it means you’ve got to have enough openness so that you can try and bring it out and hold it up to scrutiny.” This encourages the creation of a system that ‘learns’ by having a continuous feedback and evidence-based accountability: “We have to prove that this system works in a way and to produce the evidence for it.” Participants recognized the importance of building an ethical culture but also recognized the importance of being clear that this is a learning process and the system will be reﬁned over time. To achieve this goal, cooperation and commitment are needed from all stakeholders within the institution and research community. Elsewhere, it has been noted that openness alone will do little to improve ethical decision-making in animal research [39]. 3.2.1. Openness within Institutions However, as indicated by the workshop participants, an institutional commitment to building an ethical culture of research requires increased openness, not only at the protocol review committee level but at all levels that make up the system of animal research. Animals 2019, 9, 622 8 of 12 3.2.2. Openness between Institutional Research Community and the Public 3.2.2. Openness between Institutional Research Community and the Public It was generally agreed that institutions should support the dialogue among stakeholders, including the broader public. There is an opportunity for institutions to issue candid reports that show what research has been approved, the number of animals, degrees of invasiveness, etc: “ . . . our general idea was that [the evaluation] report doesn’t just get shared with the units in question—in general terms it has to be shared with the larger world . . . ” Participants recognized a need for providing information on animal research (in a non-self-serving way) and agreed that there should be opportunities to have public discussions about animal research. Simply publishing information about animal research is not enough, there also needs to be a two-way dialogue where members of the public have the opportunity to have their voices heard: “I’m just wondering whether we have a broader understanding of what we mean by openness, what its purpose is . . . Is it part of the communication, so it’s actually part of a dialogue, or is just a public display about opening doors?...to me it should be in eﬀect a dialogue rather than, you know, we’re having an open house, come around and look and see what we’re doing.” One participant highlighted that it is important to keep those involved in animal research out of harms’ way and to protect intellectual property. To this end, it was suggested that public consultation could be carried out to ﬁnd out what the public would like to know about animal research and the areas in which they would like to see greater openness. This suggestion ﬁts well with the aims of the UK Concordat on Openness on Animal Research [25]. Another way of ensuring greater openness regarding animal research is to invest in meta-analysis so that the beneﬁts achieved by animal research are more accurately known. While methods like systematic reviews were not explicitly discussed by workshop participants, the value of meta-analysis has been documented elsewhere [40]. 3.3. Policy and Oversight Participants discussed the ’business of science’ and indicated that funding agencies or granting bodies are an essential part of building an ethical culture of animal research. There was encouragement of further discussions between institutions and granting agencies on developing vehicles for funding agency involvement in the virtuous feedback loop described above. 3.2.1. Openness within Institutions A more accurate assessment of the harms and beneﬁts achieved by animal research will serve to facilitate proper harm: beneﬁt analysis of proposed research and protocol review strategies that employ a full ethical review rather than just mitigation of harms to animals. This is particularly important in light of a recent study that used a systematic review methodology to retrospectively analyse six approved preclinical animal studies. The study authors, Pound and Nicol, concluded that “All the studies were of poor quality. Having weighed the actual harms to animals against the actual clinical beneﬁts accruing from these studies, and considering the quality of the research and its impact, less than 7% of the studies were permissible according to Bateson’s Cube . . . ” [41]. Author Contributions: E.H.O., M.A.G.v.K., D.M.W. and W.M. conceived and designed the idea; E.H.O., M.A.G.v.K and D.M.W. executed the workshop; E.H.O. analysed the data; M.A.G.v.K. and W.M. facilitated the funding; E.H.O. was the lead author, with M.A.G.v.K. and D.M.W. contributing as secondary authors to the manuscript. The remaining authors (K.C., M.F., K.H., P.H.-W., M.M., M.R., A.R., J.Z.) provided critical feedback on previous drafts, new references and suggestions for rewriting, prior to submission for publication. 3.3.2. Findings for Canada Workshop participants discussed systems internationally, but given the location of the event there was some focus on animal research oversight in Canada. In particular, participants discussed introducing a third level of oversight (as opposed to the current two levels—see CCAC http://www.ccac. ca/en/). Participants describe this potential new system as follows: (1) a national board, responsible for setting minimum acceptable standards and enforcement of these standards by a third party, as well as dedicated animal welfare oﬃcers, (2) an advisory animal board to institutions, which involves local or regional stakeholders and the broader community and (3) an institutional board, responsible for internal enforcement, open houses, providing public information on approved protocols, setting internal standards and limits for animal research at that particular institution (which may go above and beyond minimum standards created by the national board). Whether such a system would work will likely depend in large part on the acceptance by those directly involved in animal-based research to incorporating the voices of the broader community; i.e. to what degree transparency and openness are viewed as positive attributes of an ethical oversight system. In addition to the recommendation for restructuring animal research oversight in the ways outlined above, one participant encouraged the establishment of face-to-face annual meetings at the provincial level. This would allow for greater openness and create the opportunity for dialogue about what animal research practices are deemed acceptable: “Face to face was very important . . . There [is] actually time and eﬀort made to invite interested groups to attend, as well as the general public, as well as others, all and sundry.” 3.3.1. Openness at All Levels Participants were in agreement that science in general should be more open, not just animal research: “the problem isn’t just about the use of animals. It’s also about our openness in science about what we’re producing . . . this is a smaller part of a much larger picture where we say, oh the system works, but we aren’t willing to hold the system up to public scrutiny and show that it does work, and be self-critical [of] ourselves in the academy.” However, participants also agreed that when focusing speciﬁcally on animal research, creating an ethical culture that is self-reﬂexive requires openness at all levels (committee, institutions, policy). “It’s not just about what we do on animal care committees and the local committees or the national [governing] bodies. It’s a part of a whole systemic thing around public knowledge . . . put it in the larger context and say we should be trying to justify what we’re doing, and the rest of the system has to do its bit. And if it’s 9 of 12 Animals 2019, 9, 622 not willing to do its bit, you aren’t going to ﬁx it at the local level, you’re not going to ﬁx it at the national, regulatory level either. There’s got to be something larger going on.” Figure 1 illustrates a virtuous, self-reﬂexive learning loop for animal-based research that goes beyond the accounting of harms to animals that happens in animal ethics committees; it is hard to imagine creating such a system without addressing issues of openness at every stage of the system. References 1. Auger, G.A. Trust Me, Trust Me Not: An Experimental Analysis of the Eﬀect of Transparency on Organizations. J. Public Relat. Res. 2014, 26, 325–343. [CrossRef] 1. Auger, G.A. Trust Me, Trust Me Not: An Experimental Analysis of the Eﬀect of Transparency on Organizations. J. Public Relat. Res. 2014, 26, 325–343. [CrossRef] 2. Vastag, B. Openness in biomedical research collides with heightened security concerns. J. Am. Med. Assoc. 2003, 289, 686–690. 2. Vastag, B. Openness in biomedical research collides with heightened security concerns. J. Am. Med. Assoc. 2003, 289, 686–690. 3. Holmberg, T.; Ideland, M. Secret and lies: “Selective openness” in the apparatus of animal experimentation. Public Underst. Sci. 2012, 21, 354–368. [CrossRef] [PubMed] 3. Holmberg, T.; Ideland, M. Secret and lies: “Selective openness” in the apparatus of animal experimentation. Public Underst. Sci. 2012, 21, 354–368. [CrossRef] [PubMed] 4. Understanding Animal Research. Available online: http://www.animalrightsextremism.info (accessed on 13 December 2018). 4. Understanding Animal Research. Available online: http://www.animalrightsextremism.info (accessed on 13 December 2018). 5. PEW Research Centre. Americans are Divided over the Use of Animals in Scientiﬁc Research. Available online: https://www.pewresearch.org/fact-tank/2018/08/16/americans-are-divided-over-the-use-of-animals- in-scientiﬁc-research/ (accessed on 17 July 2019). 5. PEW Research Centre. Americans are Divided over the Use of Animals in Scientiﬁc Research. Available online: https://www.pewresearch.org/fact-tank/2018/08/16/americans-are-divided-over-the-use-of-animals- in-scientiﬁc-research/ (accessed on 17 July 2019). 6. Speaking of Research. 52% of American Public Opposes the Use of Animals in Scientiﬁc Research. Available online: https://speakingofresearch.com/2018/08/30/52-of-american-public-opposes-the-use-of-animals-in- scientiﬁc-research/ (accessed on 17 July 2019). 6. Speaking of Research. 52% of American Public Opposes the Use of Animals in Scientiﬁc Research. Available online: https://speakingofresearch.com/2018/08/30/52-of-american-public-opposes-the-use-of-animals-in- scientiﬁc-research/ (accessed on 17 July 2019). 7. Hadley, J. Telling it like it is: A proposal to improve the transparency in biomedical research. Between Species 2012, 15, 103–126. [CrossRef] 7. Hadley, J. Telling it like it is: A proposal to improve the transparency in biomedical research. Between Species 2012, 15, 103–126. [CrossRef] 8. Ormandy, E.H. Openness and accountability of animal research: A focus group study with local stakeholders at a Canadian University. In Proceedings of the 9th World Congress on Alternatives and Animal Use in the Life Sciences, Prague, Czech Republic, 24–28 August 2014. 9. Dietrich, H.; Schibeci, R. Beyond public perceptions of gene technology: Community participation I public policy in Australia. Public Underst. Sci. 2003, 12, 381–401. [CrossRef] [PubMed] 10. Fukuyama, F.; Furger, F. Beyond Bioethics: A Proposal for Modernizing the Regulation of Human Biotechnologies; Paul H. 4. Conclusions (a) During the forum, several points of consensus were reached among participants: (a) evidence-based metrics to allow a “virtuous feedback” system for evaluation and quality assurance of animal research, (b) the need for increased public access to information together with opportunities for stakeholder dialogue about animal research, (c) a greater diversity of views to be represented on decision-making committees, to allow for greater balance and (d) a standardized and robust ethical decision-making process that incorporates some sort of societal input. Arriving at these policy suggestions was not easy; these emerged out of several days of robust discussion. We hope that this paper encourages further discussion on issues of governance and practice that go beyond openness. Academics and stakeholders based in the UK recently undertook a dialogue process in order to identify priorities for further research [42]. The advantage of shared agenda setting processes is precisely that they are co-created to ensure the broad representation of diﬀerent experiences and perspectives. The points of consensus from our meeting underscore the productivity of collaborative dialogue, as well as how diﬀerent perspectives bring strength to the policy suggestions presented. Regulatory frameworks and the level of openness vary between jurisdictions, but these diﬀerences create opportunities to share best practices and to move the global discussion for increased public openness forward. Author Contributions: E.H.O., M.A.G.v.K., D.M.W. and W.M. conceived and designed the idea; E.H.O., M.A.G.v.K and D.M.W. executed the workshop; E.H.O. analysed the data; M.A.G.v.K. and W.M. facilitated the funding; E.H.O. was the lead author, with M.A.G.v.K. and D.M.W. contributing as secondary authors to the manuscript. The remaining authors (K.C., M.F., K.H., P.H.-W., M.M., M.R., A.R., J.Z.) provided critical feedback on previous drafts, new references and suggestions for rewriting, prior to submission for publication. 10 of 12 10 of 12 Animals 2019, 9, 622 Funding: For the forum was provided in part by a Canadian Institutes of Health Research (CIHR) Dissemination grant (CIHR MET-124776) awarded to M.A.G. von Keyserlingk. The authors also thank UBC Faculty of Medicine, UBC Faculty of Science, UBC Faculty of Land and Food Systems and UBC Vice President and Research International Oﬃce for their generous ﬁnancial contributions that helped facilitate this work. Acknowledgments: This study received ethics approval from the Behavioural Research Ethics Board (H13-00599) at the University of British Columbia. We also thank Michael Brunt of the Animal Welfare Program for his insightful comments on an earlier version of this manuscript. 4. Conclusions 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, and in the decision to publish the results. 16. Rowe, G.; Frewer, L.J. A typology of public engagement mechanisms. Sci. Technol. Hum. Values 2005, 30, 251–290. [CrossRef] References Nitze School of Advanced International Studies: Washington, DC, USA, 2007; pp. 1–355. Available online: http://ieet.org/archive/Fukuyama-BiotechReg2006.pdf (accessed on 26 November 2018). 11. House of Lords. Science and Technology—Third Report. In Proceedings of the Science and Technology Select Committee, London, UK, 23 February 2000; Available online: http://www.publications.parliament.uk/ pa/ld199900/ldselect/ldsctech/38/3801.htm (accessed on 24 August 2019). 12. Walmsley, H. Biobanking, public consultation, and the discursive logics of deliberation: Five lessons from British Columbia. Public Underst. Sci. 2009, 19, 452–468. [CrossRef] [PubMed] 13. Gaskell, G.; Allum, N.; Bauer, M.; Jackson, J.; Howard, S.; Lindsey, N. Ambivalent GM Nation? Public Attitudes to Biotechnology in the UK, 1991–2002. Lond. Sch. Econ. Politi. Sci. Life Sci. Europ. Societ. Rep. 2003, 1–19. Available online: https://collection.sciencemuseumgroup.org.uk/documents/aa110107203/ambivalent- nation-public-attitudes-to-biotechnology-in-the-uk-1991-2002 (accessed on 24 August 2019). 14. Sherwin, S. Toward setting an adequate ethical framework for evaluating biotechnology policy. Presented at the Canadian Biotechnology Advisory Committee, Ottawa, ON, Canada, 4 January 2001. 15. Burgess, M.M.; Tansey, J. Democratic deﬁcit and the politics of “informed and inclusive consultation”. In Hindsight and Foresight on Emerging Technologies; Einseidel, E., Parker, R., Eds.; UBC Press: Vancouver, BC, Canada, 2008; pp. 275–288. 16. Rowe, G.; Frewer, L.J. A typology of public engagement mechanisms. Sci. Technol. Hum. Values 2005, 30, 251–290. [CrossRef] 11 of 12 11 of 12 Animals 2019, 9, 622 17. Mayer, S. Science out of step with the public: The need for public accountability of science in the UK. Sci. Public Policy 2003, 30, 177–181. [CrossRef] 18. McLeod, C.; Hobson-West, P. Opening up animal research and science-society relations? A thematic analysis of transparency discourses in the United Kingdom. Public Underst. Sci. 2016, 25, 791–806. [CrossRef] [PubMed] 19. Basel Declaration. Available online: http://www.basel-declaration.org/doc/basle%20declation_en.pdf (accessed on 20 December 2018). 20. Leenaars, M.; Ritske-Hoitinga, M.; Ormandy, E.H.; Griﬃn, G. Background to the Montréal Declaration on the synthesis of evidence to advance the 3Rs principles in science, as adopted by the 8th World Congress on Alternatives and Animal Use in the Life Sciences, Montréal, Canada on August 2011. ALTEX Proc. WC8 2012, 3–6. [CrossRef] 21. UK Concordat Public Consultation Report. Available online: http://www.understandinganimalresearch.org. k/ﬁl /9314/1207/5101/ d t lt ti t d l d l df ( d 18 D b 2014) 21. UK Concordat Public Consultation Report. Available online: http://www.understandinganimalresearch.org. uk/ﬁles/9314/1207/5101/concordat-consultation-report-download-only.pdf (accessed on 18 December 2014). 22. European Animal Research Association. Belgian Research Organisations Unite in Support of Animal Research. Available online: http://eara.eu/en/21-belgian-research-organisations-unite-in-support-of-animal-research/ (accessed on 24 August 2019). 23. References European Animal Research Association. Transparency Agreement on Animal Research Launched in Spain. Available online: http://eara.eu/en/campaigns/transparency-agreement-spain/ (accessed on 6 November 2018). 24. European Animal Research Association. Transparency Agreement on Animal Research in Portugal. Available online: http://eara.eu/en/transparency-agreement-on-animal-research-in-portugal/ (accessed on 26 November 2018). 25. Ipsos MORI. Openness in Animal Research: The Public’s Views on Openness and Transparency in Animal Research. Available online: http://www.understandinganimalresearch.org.uk/ﬁles/3014/1041/0713/openness- in-animal-r.pdf (accessed on 18 December 2014). 26. Coﬀey, A.; Atkinson, P. Concepts and coding. In Making Sense of Qualitative Data: Complementary Research Strategies; Coﬀey, A., Atkinson, P., Eds.; Sage: Thousand Oaks, CA, USA, 2006; pp. 26–53. 7. Burnard, P.; Gill, P.; Stewart, K.; Treasure, E.; Chadwick, B. Analysing and presenting qualitative data Dent. J. 2008, 204, 429–432. [CrossRef] 8. Miles, M.B.; Huberman, A.M. Qualitative Data Analysis: An. Expanded Sourcebook, 2nd ed.; Sage: Thous Oaks, CA, USA, 1994. 29. Groling, J. University animal ethics committees and the Animals (Scientiﬁc Procedures) Act: Using the Freedom of Information Act as a research tool. In Proceedings of the 2012 Conference Critical Perspectives on Animals in Society; held at the University of Exeter, UK; Calvert, C., Groling, J., Eds.; Critical Perspectives on Animals in Society (CPAS): Exeter, UK, 2013; pp. 49–62. 30. Hansen, L.A.; Goodman, J.R.; Chandna, A. Analysis of animal ethics committee membership at American institutions. Animals 2012, 2, 68–75. [CrossRef] [PubMed] 31. Schuppli, C.A.; Fraser, D. Factors inﬂuencing the eﬀectiveness of research ethics committees. J. Med. Ethic. 2007, 33, 294–301. [CrossRef] [PubMed] 32. Hawkins, P.; Hobson-West, P. Delivering Eﬀective Ethical Review: The AWERB as a Forum for Discussion. Available online: https://view.pagetiger.com/AWERB/AWERB (accessed on 24 August 2019). 33. Ideland, M. Diﬀerent views on ethics: How animal ethics is situated in committee culture. J. Med. Ethic. 2009, 35, 258–261. [CrossRef] [PubMed] 34. Schuppli, C.A. Decisions about the use of animals in research: Ethical reﬂection by animal ethics committee members. Anthrozoos 2011, 24, 409–425. [CrossRef] 35. Rose, M. Ethical decision making: Do we need to reset the GPS? In Proceedings of the 2013 ANZCAART Conference, Sydney, Australia, 23–25 July 2013. 36. Ormandy, E.H.; Kwok, Y.K.E.; Weary, D.M. Public openness in laboratory research: A model for soliciting public input into protocol review. In Proceedings of the 9th World Congress on Alternatives and Animal Use in the Life Sciences, Prague, Czech Republic, 24–28 August 2014. 37. Russell, W.M.S.; Burch, R.L. The Principles of Humane Experimental Technique; Methuen: London, UK, 1959; p. 238. References Animals 2019, 9, 622 12 of 12 38. Anderson, J.A.; Swatzky-Girling, B.; McDonald, M.A.; Pullman, D.; Saginur, R.; Sampson, H.A.; Willison, D.J. Research Ethics, Broadly Writ. Health Law Rev. 2011, 19, 12–24. 39. Johnson, J. Some challenges with Animal Ethics Committees: Can greater transparency help? In Proceedings of the 2013 ANZCAART Conference, Sydney, Australia, 23–25 July 2013. 40. Kimmelmann, J. The XV Collection: Ethical Oversights in Ethical Oversight of Animal Research. PLoS. Blogs 2019. Available online: https://blogs.plos.org/biologue/2019/01/11/the-xv-collection-ethical-oversights-in- ethical-oversight-of-animal-research/ (accessed on 24 August 2019). 41. Pound, P.; Nicol, C.J. Retrospective harm beneﬁt analysis of pre-clinical animal research for six treatment interventions. PLoS ONE 2018. [CrossRef] [PubMed] 42. Davies, G.F.; Greenhough, B.J.; Hobson-West, P.; Kirk, R.G.W.; Applebee, K.; Bellingan, L.C.; Wolfensohn, S. Developing a Collaborative Agenda for Humanities and Social Scientiﬁc Research on Laboratory Animal Science and Welfare. PLoS ONE 2016, 11, e0158791. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W1963744203	https://www.scielo.br/j/cadsc/a/TkLyY3qYywK3bwTPkXZQKyc/?lang=pt&format=pdf	Portuguese	null	O que é a Oncologia Integrativa?	Cadernos Saúde Coletiva	2,013	cc-by	5,400	Trabalho realizado no Departamento de Saúde Coletiva da Faculdade de Ciências Médicas da Universidade Estadual de Campinas (FCM-UNICAMP) – Campinas (SP), Brasil. 1Doutora em Saúde Coletiva; Membro do Laboratório de Práticas Alternativas, Complementares e Integrativas em Saúde (Lapacis), Departamento de Saúde Coletiva da FCM-UNICAMP − Campinas (SP), Brasil. 2Docente do Departamento de Saúde Coletiva − FCM-UNICAMP; Coordenador do Lapacis – Campinas (SP), Brasil. Endereço para correspondência: Pamela Siegel − Laboratório de Práticas Alternativas, Complementares e Integrativas em Saúde da Faculdade de Ciências Médicas da Universidade Estadual de Campinas − Rua Tessália Vieira de Camargo, 126 − Cidade Universitária – CEP: 13083-887 – Campinas (SP), Brasil – E-mail: gfusp@mpc.com.br Fonte de financiamento: Fapesp, projeto 2010/19680-3 e Telsan Engenharia e Serviços Ltda. Conflito de interesses: nada a declarar. Resumo Resumo A Oncologia Integrativa (OI) é um ramo da Medicina Integrativa (MI) que integra à medicina convencional as práticas com plementares, com evidências positivas, classificadas em: práticas baseadas na biologia, técnicas mente-corpo, práticas de manipulação corporal, terapias energéticas e sistemas médicos tradicionais. O objetivo deste artigo foi apresentar os resultados de uma revisão bibliográfica crítica sobre o estado da arte da OI. Foram realizados 2 procedimentos de seleção do material: uma revisão sistemática da literatura no PubMed-MEDLINE, em que foram selecionados 26 artigos; uma seleção de livros publicados a partir de 2006, citados nas referências dos artigos selecionados ou em entrevistas exploratórias com especialistas em 2 eventos internacionais: o IX congresso da Society of Integrative Oncology e o MD Anderson Observer Programme. Conclui-se que o conceito de OI já está enraizado nos principais centros de pesquisa e assistência em câncer na América do Norte, e que há uma produção bibliográfica consistente sobre este novo campo de conhecimento. Conclui-se, também, que quando combinadas com o cuidado convencional as modalidades integrativas podem estimular a efetividade e reduzir os sintomas adversos do câncer. Palavras-chave: oncologia; neoplasias; terapias complementares; medicina integrativa. O que é a Oncologia Integrativa? What is Integrative Oncology? Pamela Siegel1, Nelson Filice de Barros2 INTRODUÇÃO lógicos que contribui para a avaliação e interpretação dos da dos contidos nos artigos selecionados. As informações trazi das nos conjuntos de estudos selecionados podem apresentar resultados conflitantes e coincidentes e, a partir da análise do material, é possível identificar as limitações e evidências, positivas e negativas, produzidas em diferentes investigações. A RSL é composta pelas seguintes etapas, aplicadas neste estu do: a identificação do tema ou questão de pesquisa; estabele cimento de critérios de inclusão/exclusão de artigos; definição das informações a serem extraídas dos estudos selecionados; avaliação dos estudos incluídos; interpretação dos resultados; síntese do conhecimento7-9. A Oncologia Integrativa (OI) é um ramo da Medicina Integrativa (MI) a qual usa práticas baseadas em evidências de forma integrada com a medicina convencional1, a partir da aplicação de 5 categorias de medicinas alternativas e com plementares (MAC) no acompanhamento dos tratamentos convencionais como quimioterapia, cirurgia, radioterapia e terapia molecular, a saber: 1) Práticas baseadas na biologia: vitaminas, remédios à base de ervas e outros suplementos dietéticos; 2) Técnicas mente-corpo: yoga, meditação, visua lização; artes expressivas (arteterapia, musicoterapia, dança); 3) Práticas de manipulação corporal: reflexologia, massagem, exercícios; 4) Terapias energéticas: reiki, toque terapêutico, qigong; 5) Sistemas médicos tradicionais: medicina tradicio nal chinesa e medicina ayurvédica. Com o intuito de identificar quais fundamentos norteiam a construção do conceito de OI, foi realizado o rastreamento do descritor IO por 2 pesquisadores independentes na base de dados PubMed-MEDLINE, por ser o banco de dados mais consultado no mundo, no dia 16 de março de 2011, dispen sando a aplicação de limites e filtros. Setenta e quatro estudos foram encontrados, dos quais 26 foram selecionados. O crité rio de inclusão foi pautado na seleção dos artigos cujos títulos incluíssem termos conceituais sobre OI, como por exemplo, “guidelines”, “state of the art”, “view”, “overview”, “retrospecti ve”, “name”, “concept”, “definition”, “care”, “paradigm”, “ethics”, “principles”, “regulatory issues”, “approach”, “introduction”. Vinte e seis artigos preencheram estes critérios. O critério de exclusão consistiu no descarte de artigos que apresenta ram uma técnica terapêutica específica, ou que continham exclusivamente Ensaios Clínicos Controlados Randomizados (ECCR) ou títulos de resumos não disponíveis on-line. Quarenta e oito artigos preenchiam esses critérios e foram excluídos. INTRODUÇÃO A definição das informações a serem extraídas dos estudos selecionados consistiu das seguintes variáveis de aná lise: o conceito de OI; os protocolos de aplicação envolvendo segurança e eficácia; as questões éticas envolvendo a aplicação da OI; os principais serviços pioneiros que oferecem a OI e as fontes de consulta sobre as terapias complementares baseadas em evidências. Para contextualizar a criação do conceito de OI, apresen tamos um breve histórico de seus antecedentes. Em 1998, foi criado o Office of Cancer Complementary and Alternative Medicine (OCCAM) para coordenar as atividades do National Cancer Institute (NCI) na área das MAC2. O termo Integrative Oncology (IO) foi cunhado pelo Dr. Robert Wittes3, diretor da unidade de tratamento e diagnóstico do câncer, do NCI , em 2000. Pouco tempo depois, em 2003, é fundada a Society for Integrative Oncology (SIO), reunindo um grupo de profissio nais, pesquisadores e docentes desta modalidade, e foi lança do o periódico indexado Journal of the Society for Integrative Oncology 4. A partir de 2004 são publicados estudos no banco de dados PubMed-MEDLINE, usando o termo Integrative Oncology. Fatores como o rápido envelhecimento da popula ção norte-americana e a expansão da indústria do bem-estar impulsionaram uma demanda pelas MAC e, em 2009, sete centros de pesquisa oncológicos ofereciam um programa integrativo, entre eles, o MD Anderson Cancer Center, Dana Farber Cancer Institute, Johns Hopkins University, Memorial Sloan-Kettering Cancer Center, University of California, San Francisco (UCSF), University of California, Los Angeles (UCLA), e a Mayo Clinic5. Atualmente, 36 serviços de saúde e clínicas nos Estados Unidos atuam com base na OI, oferecen do algum tipo ou um conjunto de terapias integrativas6. Com relação à seleção dos livros utilizados foi realizada uma busca nas referências bibliográficas dos 26 artigos sele cionados e colhidas informações bibliográficas, pessoalmente, pela autora principal, durante o congresso internacional da SIO, ocorrido em Cleveland − Ohio, em novembro de 2011; na assistência de palestras; em entrevistas exploratórias com acadêmicos e na feira de livros montada no local do congres so; bem como, durante o Observer Programme cursado no hospital do câncer MD Anderson, Houston − Texas, também em novembro de 2011. A seleção dos livros não pretendeu ser exaustiva, contudo abrangeu os principais autores e suas respectivas publicações entre 2006 e a atualidade. Abstract Integrative Oncology (IO) is a branch of Integrative Medicine (IM), which integrates the following evidence-based complemen tary practices into conventional medicine: biologically based practices; mind-body techniques; body manipulation; energetic practices and whole systems. The purpose of this article is to present the results of a critical literature review on the state of the art of IO. Two procedures for selecting the material were applied: a systematic literature review undertaken in the PubMed- MEDLINE, in which 26 articles were selected; a selection of books published since 2006, mentioned in the references of the selected articles and in exploratory interviews with specialists at two international events: the ninth conference of the Society of Integrative Oncology and the MD Observer Programme. In conclusion, firstly the concept of IO is already steadily being applied in the main North-American cancer health centers and there is a consistent bibliographic production concerning this new field of knowledge. Secondly, when combined with conventional care, the integrative modalities can stimulate the effectiveness and reduce the adverse symptoms of cancer. Keywords: medical oncology; neoplasms; complementary therapies; integrative medicine. ywords: medical oncology; neoplasms; complementary de Coletiva; Membro do Laboratório de Práticas Alternativas, Complementares e Integrativas em Saúde (Lapacis), Departamento de Saúde -UNICAMP − Campinas (SP), Brasil. Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 348 348 Oncologia Integrativa INTRODUÇÃO Foram O objetivo deste artigo foi apresentar os resultados de uma revisão bibliográfica crítica sobre o estado da arte da OI, composta de uma revisão sistemática da literatura efetuada no PubMed-MEDLINE, em que foram selecionados 26 estudos e 10 livros seminais sobre o tema, publicados a partir de 2006. da Literatura e da seleção bibliográfica Na etapa da RSL da avaliação dos estudos e da interpre tação dos resultados, foi possível constatar que os 26 estudos estão distribuídos entre os anos 2004 e 2011, e observa-se que há um consenso5,10-12 quanto ao conceito de OI, enfati zando que enquanto a medicina alternativa abrange terapias promovidas com exclusão da medicina convencional, a me dicina complementar segue uma lógica associativa podendo ser eficaz quando usada em combinação com a medicina convencional, e a medicina integrativa é um desenvolvimento das medicinas alternativas e complementares baseadas em evidências associadas a diferentes tratamentos convencionais. A OI segue, portanto, as bases da medicina integrativa ofer tando práticas complementares para o cuidado de pessoas em tratamento do câncer. Segundo Mumber20, o propósito da MI é eventualmente eliminar os termos “MAC” e “convencional” e chegar a uma forma de medicina que proporcione aos pacientes “aquilo que funciona”. Na seção I do livro, são abordados os seguintes te mas: a pesquisa clínica e as evidências; a formação médica em medicina integrativa; o bem-estar do médico na sua prática clínica; modelos de cuidado; questões legais e levantamento de custos. A seção II traz as diferentes modalidades que fazem parte da OI, a saber: atividade física; nutrição; intervenções mente-corpo; ervas medicinais; terapias de manipulação; medicina energética; espiritualidade e sistemas médicos alter nativos. Seguem capítulos sobre a prevenção e assistência do cuidado, bem como sobre o tabagismo e alcoolismo na OI. O último capítulo do livro é dedicado às diferentes neoplasias, como câncer de mama, próstata, pulmão, colorretal, pele e outras. Em cada subcapítulo, há um tópico sobre os fatores de risco, a detecção precoce da doença, o diagnóstico, o trata mento convencional, a prevenção, tabelas para consulta e um glossário da terminologia específica, e o acompanhamento com cada modalidade da OI. O livro encerra com o tema dos cuidados paliativos e como lidar com os seus diversos sinto mas, tais como dor, estresse, anorexia, fadiga, constipação, dispneia, insônia, náusea e vômito, alimentação parenteral, os efeitos colaterais dos opioides, depressão, ansiedade e solidão. A segurança e a eficácia destas terapias são discutidas em dois estudos13,14, e, em geral, sua aplicação só é desencoraja da no caso de haver evidências que indiquem risco sério ou ineficácia, caso contrário são toleradas, encorajadas e moni toradas, seguindo tabelas de recomendações, pontuações e avaliações dos estudos clínicos na área. METODOLOGIA A Revisão Sistemática da Literatura (RSL) é um processo de identificação e seleção de estudos científicos, com critérios Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 349 349 Pamela Siegel, Nelson Filice de Barros Center17; University of Texas M. D. Anderson Cancer/Center Complementary/Integrative Medicine18 e do United States Pharmacopeia19. Em geral, apresentam listas bastante comple tas das interações entre ervas, alimentos, suplementos e medi camentos, e permitem aos usuários ter acesso a listas de discus são, livros, artigos, monografias e boletins sobre o assunto10,12. identificados 10 principais livros, considerados seminais por sua utilização pela maior parte dos autores que trabalham com o tema. Além de apresentar um resumo do conteúdo de cada livro, as variáveis críticas usadas para analisá-los foram: a diversidade de terapêuticas integrativas abordadas; as respectivas pesquisas apresentadas sobre as terapêuticas integrativas; e as práticas integrativas recomendadas para os sintomas do tratamento oncológico convencional. i Em 2006, foi publicada a obra de Mumber20, contendo 15 capítulos e 517 páginas. No prefácio, o autor revela que a motivação para escrever o livro nasceu das três principais perguntas que os pacientes de câncer lhe faziam durante a sua prática clínica: a) O que mais posso fazer além do tra tamento convencional? b) Onde posso ir para encontrar um profissional que me ajude a aplicar isso? c) Por que o meu oncologista não sabe mais sobre isso? O autor afirma, ainda, que, em meados dos anos 1990, 70% das perguntas feitas ao Office of Alternative Medicine (OAM) eram sobre MAC para câncer. Na Introdução, assinada por David S. Rosenthal, mé dico, docente da Harvard Medical School e ex-presidente do SIO e Henry K. Oliver, diretor da Harvard University Health Services, são apresentados o desenvolvimento do conceito e a prática de OI, cujas perguntas anteriormente mencionadas precisam ser enfrentadas pelos oncologistas integrando pa cientes, seus clínicos e praticantes de MAC. da Literatura e da seleção bibliográfica Com relação à ética envolvendo a aplicação da OI, alguns autores15 afirmam que a sua prática deve ir além da eficácia e segurança, estendendo-se ao impacto do câncer sobre o sofrimento, o pensar, sentir, criar e querer do paciente. Reforçam a ideia de que a comu nicação médico-paciente sobre o uso das práticas integrativas é fundamental para um tratamento holístico e retomam os princípios gerais da ética biomédica. Entre os serviços pioneiros que oferecem a OI nos gran des centros norte-americanos estão: M. D. Anderson Cancer Center, Leonard P. Zakim Center for Integrative Therapies no Dana-Farber Cancer Institute, Memorial Sloan-Kettering Cancer Center em Nova Iorque, e a Cleveland Clinic, con tudo a lista completa destas instituições consta no site do Consortium of Academic Health Centers for Integrative Medicine (CAHCIM)16. As fontes de consulta on-line para a OI, usadas por estas instituições, profissionais de saúde e pacientes, estão en tre outros, nos sítios do Memorial Sloan-Kettering Cancer Em 2008, Cohen e Markman21 publicam uma obra impor tante, com 232 páginas, contendo 13 capítulos distribuídos em Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 350 350 Oncologia Integrativa 3 partes: Introdução; Programas inscritos no NCI e Pesquisa. O livro inicia com uma advertência sobre a necessidade de se realizar uma pesquisa crítica de qualidade para identificar e diminuir os riscos associados ao uso das MAC. A seguir, aborda o tema dos princípios éticos e legais, e menciona a importância da comunicação médico-paciente para guiar as decisões clínicas no uso das MAC. Os seguintes cinco capítu los estão dedicados aos programas de OI dos hospitais M.D. Anderson; Memorial Sloan-Kettering; Dana Farber; Johns Hopkins e Mayo Clinic, respectivamente. Os últimos quatro capítulos tratam de pesquisas sobre a abordagem mente -corpo; as ervas medicinais e suas interações com produtos farmacêuticos; a aplicação da acupuntura e seus efeitos cola terais nos tratamentos oncológicos. de pesquisa no câncer; a importância da verdade e de dizer a verdade na OI. Os últimos dois capítulos tratam da perspecti va de um paciente e do futuro da OI. O segundo trabalho importante publicado em 2009 foi o livro de Weber23, com 240 páginas e 8 capítulos, que apre senta as ervas, os componentes e suplementos que a pesquisa provou serem eficazes na recuperação dos pacientes em tra tamento oncológico. da Literatura e da seleção bibliográfica O autor traz uma lista de métodos das MAC, tanto ocidentais, como chineses, que podem ser apli cados no câncer, na quimioterapia e radioterapia, bem como para o alívio dos efeitos destes tratamentos. O terceiro trabalho relevante é o Annual Report24, que traz as prioridades das pesquisas financiadas pelo NCI em 2008, na área das MAC, para o ano 2009. O relatório procura: 1) identificar novas terapêuticas na farmacopeia dos siste mas médicos tradicionais, como definido pela Organização Mundial da Saúde; 2) usar as abordagens complementares para melhorar a porcentagem das terapias convencionais con tra o câncer; 3) pesquisar modificações do estilo de vida (die ta, exercício, abordagens corpo-mente) e seu impacto sobre o câncer (resposta à terapia oncológica convencional, sobrevi da), algumas das quais estão em andamento e outras conclu ídas. Entre outros, o texto contém estudos sobre vitaminas; algas; 5 plantas africanas; frutas, fitoquímicos e flavonoides. O relatório também aborda a estimulação cranial (cranial mi crocurrent electrical stimulation) com o dispositivo chamado Alpha-Stim, e pesquisa o programa Take Heart, fazendo os pacientes se exercitarem por 30 minutos diariamente. Outro livro publicado em 2008 foi o de Servan-Schreiber22, professor de psiquiatria da Universidade de Pittsburgh, diag nosticado com câncer no cérebro aos 31 anos de idade, em 1992. O autor vive a experiência da quimioterapia e cirurgia cerebral, e pergunta ao seu oncologista se algum estilo de vida poderia prevenir a recorrência. Após receber uma res posta negativa, o autor faz uma imersão profunda em dados científicos sobre as defesas naturais do organismo contra o câncer e acaba concluindo que uma dieta pobre, maus hábi tos como o tabagismo, as influências de alguns hormônios e toxinas ambientais aumentam o risco. O livro é uma síntese de buscas científicas e experiências pessoais que impactou o mundo globalizado. Em 2009, aconteceu a publicação de 3 trabalhos relevantes, o primeiro é o livro de Abrams e Weil1, que contém 30 capítulos escritos por diferentes autores e distribuídos em 601 páginas. Weil, que fundou o Program in Integrative Medicine, em 1998, atualmente o Arizona Center for Integrative Medicine (ACIM), considera que o treinamento em MI deve entrar na formação de todos profissionais de saúde e afirma que, de todas as espe cialidades da medicina, é a oncologia a mais lenta em abraçar a MI. da Literatura e da seleção bibliográfica A razão disso, segundo ele, seria o emocionalismo e o medo que envolve o câncer, porque a doença é debilitante, e os tratamentos convencionais nem sempre atingem o êxito propagado pelos oncologistas, podendo causar mutações e transformações malignas. O livro proporciona informações sobre os principais temas da OI, como questões nutricionais; o uso de ervas medicinais; interações com a quimioterapia; o debate sobre os antioxidantes; o papel da atividade física; massoterapia; terapias mente-corpo; Medicina Tradicional Chinesa, as perspectivas ayurvédica, homeopática e antropo sófica; a medicina energética; os cuidados espirituais; o papel da MI no câncer de mama, próstata e colorretal; a radiotera pia; o manejo de sintomas; o combate à toxicidade da qui mioterapia; as terapias alternativas; as MAC e a metodologia O lançamento do livro de Decker e Lee25 foi a novidade na área da OI em 2010. O livro de 196 páginas, publicado pela Oncology Nursing Society (ONS), analisa como os sintomas podem ser administrados com as MAC e como estas devem ser usadas. O livro contém tabelas detalhadas dos tratamentos baseados em evidências em várias gradações, fortes, mode radas, fracas, negativas e conflituosas, de ervas, extratos, in formações sobre considerações nutricionais para os pacientes oncológicos e apêndices com a posição da ONS sobre o uso das MAC. Dezesseis sintomas oncológicos são abordados. Em 2011, mais 2 trabalhos sobre as aplicações das MAC na oncologia foram publicados, pelos autores Cassileth26 e Forsythe27. O 1º está organizado em 7 partes, e contém 55 ca pítulos em 354 páginas. A autora abrange mais de 50 terapias e segue um padrão apresentando as seguintes informações: a definição; o que os profissionais dizem que a terapia faz; crenças sobre as quais as terapias se baseiam; pesquisa sobre as evidências até o momento; o que essa prática pode fazer por você e onde consegui-la. Ela descarta várias terapias al ternativas e complementares como não tendo base científica, como a medicina ayurvédica, a homeopatia, as formas de cura Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 351 351 Pamela Siegel, Nelson Filice de Barros dos nativos americanos, algumas abordagens naturopáticas, os jejuns, a ingestão de sucos, os florais de Bach, a dieta ma crobiótica, as visualizações, o toque terapêutico e o qigong. da Literatura e da seleção bibliográfica Outras terapias teriam algum efeito sobre os sintomas do câncer e serviriam para relaxar ou aliviar algum incômodo, tais como: a acupuntura, acupressão, massagem, pilates, Alexander, hidroterapia, reflexologia, rolfing, aromaterapia, arteterapia, dança. Ainda outras terapias seriam totalmente desrecomendadas, como por exemplo, a terapia da biologia dentária, a quelação, a desintoxicação do cólon, terapias à base de enzimas, terapia neural, terapia de oxigênio, terapias com cartilagem de tubarão e terapias com cristais. A autora enfatiza, ainda, que nenhuma destas terapias tem poder de cura sobre o câncer e que as remissões espontâneas no câncer ainda não são bem compreendidas ou estudadas. massagem, nutrição, reflexologia, suplementos, visualização e yoga. Em cada capítulo sobre estas práticas, há ilustrações, formas de uso e tabelas. Alguns capítulos são dedicados aos sintomas mais frequentes do câncer e a forma de tratá-los com as MAC. Os sintomas abordados são: disfunção cog nitiva pós-quimioterápica (chemo brain); constipação, de pressão, diarreia, boca seca, fadiga, dor de cabeça, calores, supressão imunológica, insônia, perda de apetite, perda de libido e edema. Consideramos o livro de Mumber um dos mais completos de todos os analisados, contudo, por haver sido publicado no ano de 2006, ele inclui os resultados das pesquisas eficazes realizadas somente até 2005. O livro de Cohen e Markman enfatiza as questões éticas da aplicação da OI e a comunicação médico-paciente, mas nas pesquisas, se limita a mencionar os estudos somente em três áreas: mente-corpo, ervas medicinas e acupuntura. A obra de Servan-Schreiber é única no sentido de transmitir a narrativa de um pesquisador que é ao mesmo tempo um paciente de câncer. O texto de Abrams e Weil se destaca por discutir o estigma e o emocionalismo envolvidos no câncer e a necessidade de dizer a verdade ao paciente. O livro de Weber traz um recorte sobre as ervas medicinais, os extratos, compostos e suplementos, este é o seu ponto forte e, ao mesmo tempo, sua limitação. Já o Annual Report tem a utilidade de apresentar as pesquisas financiadas pelo NCI em 2008, apontando para o uso de fitoquímicos, estimulação cranial e exercícios físicos. A importância do livro de Decker e Lee reside no fato de ele ter sido publicado pela Sociedade de Enfermagem norte-americana e ser um livro de referência na área da enfermagem oncológica. da Literatura e da seleção bibliográfica Com relação ao livro de Cassileth, o ponto que o diferencia dos demais é que ele abrange mais de 50 terapias e serve como uma salvaguarda para os profissionais de saúde e usuários que desejam obter informações claras sobre o uso das MAC no câncer. O livro de Forsyth destaca pontos críticos do uso da quimioterapia e se concentra em quatro tipos de terapias: suplementos, vita minas, nutrição, homeopatia. Finalmente, o livro de Ladas e Kelly serve como um guia prático de estratégias integrativas. Destaca-se que ele enfatiza, entre outras, o uso da aromate rapia, reflexologia, visualização e yoga, modalidades que não recebem tanta atenção nos demais livros. O segundo livro traz uma visão crítica da aplicação dos tratamentos convencionais e analisa os baixos índices de cura e de sobrevida e os altos índices de recorrência do câncer. Durante o seu treinamento como oncologista na Universidade da Califórnia, o autor descobriu que somente 2% dos pacien tes no estágio 4 do câncer sobreviviam após vários ciclos de aplicações da quimioterapia e, mesmo assim, a maioria dos sobreviventes corriam o risco de passar o resto da vida com os sintomas de disfunção cognitiva pós-quimioterápica (chemo brain) e teriam de conviver com problemas cardíacos ou dis funções do fígado, dor constante e perda da sensibilidade nos pés e nos dedões. Segundo o autor, os oncologistas baseiam to das as suas recomendações sobre os últimos estudos clínicos, porém nenhum deles oferece uma resposta de cura completa. Os medicamentos quimioterápicos são aplicados baseados em probabilidades estatísticas de que estas drogas possam ter um efeito positivo sobre o câncer, contudo, segundo o autor, se os oncologistas estiverem equivocados, como acontece com grande frequência, o paciente estará literalmente ingerindo um veneno sem quaisquer efeitos benéficos. O autor defende o fortalecimento do sistema imunológico como a melhor arma no combate ao câncer e aplica suplementos, vitaminas, nutrição, homeopatia e outras terapias. Ele mantém as doses de quimioterapia em níveis mínimos, compatíveis com os dados extraídos de exames sanguíneos de cada paciente. Em 2012, foi lançado o livro de Ladas e Kelly28, que está dividido em 5 capítulos e tem 304 páginas. As autoras en fatizam a importância do papel das MAC no tratamento do câncer e da comunicação do paciente com a sua equipe médi ca, incluindo aí o terapeuta complementar. da Literatura e da seleção bibliográfica Elas apresentam aquelas práticas que, em sua experiência e de acordo com as evidências e o histórico da prática, foram úteis no apoio aos pacientes de câncer. São elas: aromaterapia, Medicina Tradicional Chinesa (MTC) e acupressão, chás, homeopatia, CONCLUSÕES A síntese do conhecimento extraído da RSL dos 26 artigos e dos 10 livros considerados neste texto permite concluir que há abundante material disponível para consolidar a construção do conceito de OI e que ele faz parte de um novo modelo de cuidado que aponta para um pluralismo na saúde. As modali dades das práticas integrativas só são desencorajadas quando causam sérios riscos à saúde dos pacientes ou interferem com a quimioterapia. Além disso, a OI resgata os princípios da bioética, buscando o bem-estar do paciente, procurando evi tar danos e respeitar sua autonomia. No entanto, destaca-se que, nos Estados Unidos, a OI ainda não é oferecida para a população de baixa renda, idosa ou portadora de deficiências, no Medicaid e Medicare. Os desdobramentos desta constatação foram: mais verba federal de pesquisa alocada para as MAC nos Estados Unidos e em outros países; o interesse por parte de empresas privadas de alavancar a produção de fitoquímicos, vitaminas e suple mentos; e a organização de serviços e disciplinas/cursos de formação sobre as MAC nas áreas acadêmicas da saúde. Ao mesmo tempo, houve a necessidade de proteger o cida dão dos efeitos nocivos de algumas práticas alternativas sem embasamento científico e informá-lo sobre o uso das MAC. Daí a disponibilização de informações em sítios confiáveis e as advertências que constam nos livros anteriormente anali sados enfatizando os tratamentos integrativos, baseados em evidências, e que sejam administrados a partir de uma equipe multiprofissional de saúde. Igualmente, o material analisado permite constatar que, quando combinadas com o cuidado convencional, as mo dalidades complementares podem estimular a efetividade e reduzir os sintomas adversos do câncer. No Brasil, acreditamos que a OI possa ser introduzi da como uma extensão da Política Nacional de Práticas Integrativas e Complementares (PNPIC), já que esta Política deve ser entendida como continuidade do processo de im plantação do Sistema Único de Saúde (SUS) que favorece a integralidade da atenção à saúde, contribuindo, também, para a corresponsabilidade dos indivíduos pela saúde e para o exercício da cidadania. Contudo, devido à ausência de dire trizes específicas, as Práticas Integrativas e Complementares (PICs) têm ocorrido na rede pública estadual e municipal de forma desigual e descontinuada, sem o devido registro e fornecimento adequado de insumos ou ações de acompanha mento e avaliação30. CONCLUSÕES Consideramos o material apresentado nesta revisão bi bliográfica crítica de vital importância tanto para os profis sionais de saúde, como para os gestores e usuários do Sistema Único de Saúde brasileiro, considerando que, no Brasil, o conceito de Oncologia Integrativa ainda é desconhecido e nem sequer é mencionado no sítio eletrônico do Instituto Nacional do Câncer; que temos uma Política Nacional de Práticas Integrativas e Complementares que dá suporte legal às práticas complementares; e que a previsão é de, no mínimo, 500 mil novos casos de câncer por ano na população brasileira. DISCUSSÃO A sociedade norte-americana assistiu nas últimas duas décadas a um aumento da demanda por produtos e serviços alternativos e complementares na saúde. Eisenberg29 condu ziu um duplo inquérito sobre o uso de 16 tipos diferentes de medicinas alternativas e complementares em 1990 e 1997. Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 352 352 Oncologia Integrativa Os resultados mostram que houve um aumento significativo de 42% com gastos em medicina alternativa entre 1990 e 1997, pagos pelo bolso do consumidor. Eisenberg sugeriu, ainda, que as agências federais, as corporações privadas, fundações e instituições acadêmicas adotassem uma postura mais pro ativa com relação à implementação de pesquisas clínicas, ao currículo educacional, ao credenciamento profissional e controle de qualidade na área. Outra porta de entrada para a OI é o Programa Nacional de Assistência à Dor e Cuidados Paliativos, cujos objetivos são, entre outros, articular iniciativas governamentais e não governamentais voltadas para a atenção/assistência aos pa cientes com dor e cuidados paliativos31. 6. Integrative Oncology Centers [Internet]. [cited 2011 May 12]. Available from: http://fontherapeutics.com/resources/integrative-oncology-centers/ 7. Sampaio RF, Mancini MC. Estudos de revisão sistemática: um guia para síntese criteriosa da evidência científica. Rev Bras Fisioter. 2007;11(1): 83-9. REFERÊNCIAS 1. Abrams D, Weil A. Integrative Oncology. New York: Oxford University Press; 2009. 2. Office of Cancer Complementary and Alternative Medicine [Internet]. [cited 2011 Apr 6] . Available from: http://www.cancer.gov/cam/ 7. Sampaio RF, Mancini MC. Estudos de revisão sistemática: um guia para síntese criteriosa da evidência científica. Rev Bras Fisioter. 2007;11(1): 83-9. 3. Wittes R. Integrative Oncology: Cancer Care For The Next Millenium [Internet]. [cited 2011 May 8]. Available from: http://legislative.cancer. gov/Files/testimony-2000-06-07.pdf 8. Mendes KDS, Silveira RCCP, Galvao CM. Revisão integrativa: método de pesquisa para a incorporação de evidências na saúde e na enfermagem. Texto Contexto Enferm. 2008;17(4):758-64. 8. Mendes KDS, Silveira RCCP, Galvao CM. Revisão integrativa: método de pesquisa para a incorporação de evidências na saúde e na enfermagem. Texto Contexto Enferm. 2008;17(4):758-64. 4. Journal of the Society for Integrative Oncology [Internet]. [cited 2011 May 10]. Available from: www.integrativeonc.org/ 9. Gonçalo CS, Castro CM, Bonon MM, Mota PMR, Dahdal AB, Batista JC, Hirayama MS, Peres SMP, Barros NF. Planejamento e execução de revisões sistemáticas da literatura. Brasília Med. 2012;49(2): 5. Geffen JR. Integrative oncology for the whole person: a multidimensional approach to cancer care. Integr Cancer Ther. 2010;9(1):105-21. Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 353 353 Pamela Siegel, Nelson Filice de Barros 22. Servan-Schreiber D. Anti-Cancer: Explore a New Way of Life. New York: Viking Adult; 2009. 10. Deng GE, Frenkel M, Cohen L, Cassileth BR, Abrams DI, Capodice JL, Courneya KS, Dryden T, Hanser S, Kumar N, Labriola D, Wardell DW, Sagar S; Society for Integrative Oncology. Evidence-based clinical practice guidelines for integrative oncology: complementary therapies and botanicals. J Soc Integr Oncol. 2009;7(3):85-120. 23. Weber D. Introduction to Integrative Oncology, Herbs, compounds and supplements in the treatment of cancer. London: Panaxea Publishing; 2009. 11. Wesa K, Cassileth BR. Introduction to section on integrative oncology. Curr Treat Options Oncol. 2010;11(3-4):70-2. 24. Office of Cancer Complementary and Alternative Medicine. Annual Report on Complementary and Alternative Medicine, 2009 [Internet]. [cited 2011 Jun 6]. Available from: http://cam.cancer.gov/cam_annual_ report.html 12. Smyth JF. Integrative oncology--what’s in a name? Eur J Cancer. 2006; 42(5):572-3. 25. Decker GM, Lee CO. Handbook of Integrative Oncology Nursing: Evidence-Based Practice. Illinois: Oncology Nursing Society; 2010. 13. Deng GE, Cassileth BR, Cohen L, Gubili J, Johnstone PA, Kumar N, Vickers A; Society for Integrative Oncology Executive Committee, Abrams D, Rosenthal D, Sagar S, Tripathy D. Integrative Oncology Practice Guidelines. J Soc Integr Oncol. 2007;5(2):65-84. 26. Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 354 REFERÊNCIAS Cassileth BR. The Complete Guide to Complementary Therapies in Cancer Care: Essential Information for Patients, Survivors and Health Professionals. Singapore: World Scientific Publishing Co.; 2011. 14. Ben-Arye E, Schiff E, Golan O. Ethical issues in integrative oncology. Hematol Oncol Clin North Am. 2008;22(4):737-53. 27. Forsythe JW. The Compassionate Oncologist. Fix Bay Inc Publishing; 2011. 15. Wesa K, Gubili J, Cassileth B. Integrative oncology: complementary therapies for cancer survivors. Hematol Oncol Clin North Am. 2008;22(2):343-53. 28. Ladas EJ, Kelly KM. Integrative Strategies for Cancer Patients. Singapore: World Scientific Publishing Co.; 2012. 16. Consortium of Academic Health Centers for Integrative Medicine (CAHCIM) [Internet]. [cited 2011 Apr 2]. Available from: http://www. imconsortium.org/ 29. Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, Van Rompay M, Kessler RC. Trends in alternative medicine use in the United States, 1990-1997: results of a follow-up national survey. JAMA. 1998;280(18): 1569-75. [cited 2011 May 12] Available from: http://webpub.allegheny. edu/employee/L/lcoates/CoatesPage/FS101/Articles_PDF/General/ JAMA_Trends_in_AM.pdf 17. Memorial Sloan-Kettering Cancer Center [Internet]. [cited 2011 Apr 5]. Available from: http://www.mskcc.org/mskcc/html/44.cfm; http://www. mskcc.org/aboutherbs 18. University of Texas M. D. Anderson Cancer/Center Complementary/ Integrative Medicine [Internet]. [cited 2011 Apr 5]. Available from: http://www.mdanderson.org/CIMER 30. Brasil. Ministério da Saúde. Política Nacional de Práticas Integrativas e Complementares. Brasília: Ministério da Saúde. 2006. 92 p. [cited 2011 May 20]. Available from: http://portal.saude.gov.br/portal/arquivos/pdf/ pnpic_publicacao.pdf 19. United States Pharmacopeia [Internet]. [cited 2011 Apr 5]. Available from: http://www.usp.org/dietarySupplements 31. Brasil. Ministério da Saúde. Programa Nacional de Assistência à Dor e Cuidados Paliativos. Brasília: Ministério da Saúde. Portaria GM/MS nº 19, de 03 de janeiro de 2002. 20. Mumber PM. Integrative Oncology, Principles and practice. London: Taylor & Francis; 2006. 21. Cohen L, Markman M. Integrative Oncology, Incorporating Complementary Medicine into Conventional Cancer Care. Totowa, NJ: Humana Press; 2008. Recebido em: 03/06/2012 Aprovado em: 24/10/2012 Recebido em: 03/06/2012 Aprovado em: 24/10/2012 Cad. Saúde Colet., 2013, Rio de Janeiro, 21 (3): 348-54 354
https://openalex.org/W2177605184	https://europepmc.org/articles/pmc4702895?pdf=render	English	null	TSGene 2.0: an updated literature-based knowledgebase for tumor suppressor genes	Nucleic acids research	2,015	cc-by	7,431	ABSTRACT normal cell growth with a potential to spread through the body. It often arises from two types of genetic alterations related to the cell proliferation, differentiation, apoptosis and cell-to-cell communication (2,3): the loss-of-function of tumor suppressor genes (TSGs) and the gain-of-function of oncogenes (OCGs). The inactivation or reduced func- tion of protein-coding TSGs can be induced in many ways including promoter methylation changes (4), copy number alterations (5), deregulated mRNA expression due to mi- croRNA (miRNA) activities (6) and competing endogenous long non-coding RNAs (lncRNAs) (7). In general, TSGs play key roles in the cell cycle checkpoints and in maintain- ing genomic stability. Defective TSGs often allow uncon- trolled cell growth without normal DNA repair, apoptosis and normal metabolic regulation (8). Accumulating lines of evidence have shown that non-protein coding RNAs, such as miRNAs (9–11) and lncRNAs (12), can act as TSGs to initiate and promote cancer development. Tumor suppressor genes (TSGs) are a major type of gatekeeper genes in the cell growth. A knowl- edgebase with the systematic collection and cura- tion of TSGs in multiple cancer types is critically important for further studying their biological func- tions as well as for developing therapeutic strategies. Since its development in 2012, the Tumor Suppres- sor Gene database (TSGene), has become a popular resource in the cancer research community. Here, we reported the TSGene version 2.0, which has sub- stantial updates of contents (e.g. up-to-date litera- ture and pan-cancer genomic data collection and curation), data types (noncoding RNAs and protein- coding genes) and content accessibility. Speciﬁcally, the current TSGene 2.0 contains 1217 human TSGs (1018 protein-coding and 199 non-coding genes) cu- rated from over 9000 articles. Additionally, TSGene 2.0 provides thousands of expression and mutation patterns derived from pan-cancer data of The Can- cer Genome Atlas. A new web interface is available at http://bioinfo.mc.vanderbilt.edu/TSGene/. System- atic analyses of 199 non-coding TSGs provide numer- ous cancer-speciﬁc non-coding mutational events for further screening and clinical use. Intriguingly, we identiﬁed 49 protein-coding TSGs that were consistently down-regulated in 11 cancer types. In summary, TSGene 2.0, which is the only available database for TSGs, provides the most updated TSGs and their features in pan-cancer. p p To provide a comprehensive TSG resource for the can- cer research community, we developed the Tumor Sup- pressor Gene database (TSGene 1.0) in 2012 (13), and have been continuously maintaining it since then. C⃝The Author(s) 2015. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. ABSTRACT TSGene 1.0 is the only active data resource specifically designed for TSGs. It has received 82573 web hits based on daily unique internet protocol addresses. Since its release, TS- Gene database has become a popular resource, enjoying wide use for testing drug resistance (14), studying HIV in- tegration of cancer-related genes (15,16), exploring phos- phorylation regulatory networks in cancer cells (17), identi- fying cancer-associated transcript fusions (18), uncovering intronic enhancers through loss of methylation (19) and de- signing genome-scale CRISPR-based gene repression (20). Moreover, TSGene 1.0 has been frequently used as a special gene list in the systems biology-based studies for the cancer genomic data (21–24). I h l h i d h TSGene 2.0: an updated literature-based knowledgebase for tumor suppressor genes Min Zhao1, Pora Kim2, Ramkrishna Mitra2, Junfei Zhao2 and Zhongming Zhao2,3,4,5,* 1School of Engineering, Faculty of Science, Health, Education and Engineering, University of the Sunshine Coast, Maroochydore DC, Queensland 4558, Australia, 2Department of Biomedical Informatics, Vanderbilt University School of Medicine, Nashville, TN 37232, USA, 3Department of Cancer Biology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA, 4Department of Psychiatry, Vanderbilt University School of Medicine, Nashville, TN 37212, USA and 5School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, TX 77030, USA Received September 14, 2015; Revised October 30, 2015; Accepted November 03, 2015 To whom correspondence should be addressed. Tel: +1 615 343 9158; Fax: +1 615 936 8545; Email: zhongming.zhao@vanderbilt.edu Nucleic Acids Research, 2016, Vol. 44, Database issue D1023–D1031 doi: 10.1093/nar/gkv1268 Nucleic Acids Research, 2016, Vol. 44, Database issue D1023–D1031 doi: 10.1093/nar/gkv1268 Published online 20 November 2015 C⃝The Author(s) 2015. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creat permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Received September 14, 2015; Revised October 30, 2015; Accepted November 03, 2015 To whom correspondence should be addressed. Tel: +1 615 343 9158; Fax: +1 615 936 8545; Email: zho INTRODUCTION In the past several years, we have witnessed the unprece- dented growth of cancer genomic data, such as those from Cancer is a large family of diseases that cause millions of death worldwide every year (1). It is characterized by the ab- D1024 Nucleic Acids Research, 2016, Vol. 44, Database issue the Cancer Genome Project (CGP), The Cancer Genome Atlas (TCGA) and the International Cancer Genome Con- sortium (ICGC). Furthermore, many more TSGs have been reported including non-protein coding genes (miRNAs and lncRNAs). Accordingly, there is a strong need to charac- terize the tumor suppressor gene landscape at the genome, epigenome, transcriptome and proteome levels, across all types of cancer. We have addressed this need with substan- tial updates to TSGene 2.0. Our changes include more ex- tensive literature curation, data integration and annotation, and a user-friendly web interface. translated them to the official gene symbols from Entrez Gene database. In this round, we were more cautious for the TSGs with potential oncogenic roles. TSGs may also play differ- ent roles in different cancers or at different stages of dis- ease. For example, histone deacetylase 1 gene (HDAC1) has been reported as a TSG during the cancer initiation, but as an oncogene during the tumor maintenance pro- cess (26). Other TSGs may have oncogenic roles in differ- ent cancer grades. For example, RASSF1 is a TSG mainly reported in neuroendocrine tumors of the lung. One of its isoform acts as an oncogene in some high-grade lung tu- mors (27). The Notch signaling pathway has been identi- fied as oncogene in the hematopoietic cancers (28). How- ever, accumulating lines of evidence suggest that pathway members have growth-suppressive roles in some hematopoi- etic cells, in skin, pancreatic epithelium and hepatocytes (28). SIRT1 can negatively regulate the TGF-beta signal- ing pathway and enhance tumorigenesis, but it also interacts with promyelocytic leukaemia (PML) protein to stabilize TP53 and induce cell senescence (29). WT1, a well-studied tumor suppressor has dual roles in cancer progression de- pending on the presence or absence of regulatory protein partners (30). A few well-studied oncogenes, such as MYC, have weak evidence for tumor suppression (31). We did not include them in TSGene database to avoid potential misuse. INTRODUCTION y As the only literature-based database dedicated to TSGs, TSGene 2.0 provides not only a comprehensive resource for the cancer research community, but also a classified TSG catalog for advanced integrative analyses across multiple cancers. For example, as described in this paper, we ob- served that two lncRNA TSGs, DLEU1 and DLEU2, are not only highly mutated in multiple prostate cancers, but also in cancers of the bladder and ovaries. Our analyses of 24 highly confident miRNA TSGs revealed that they were associated with cancer-specific signaling pathways in multiple cancers. Furthermore, we pinpointed a testable prevalent deletion of has-miR-31 in high-grade glioblas- toma (GBM). We also used a pan-cancer expression data analysis method and found that the TGF-beta signaling pathway is dominated by TSGs that are consistently down- regulated in tumor samples. These cancer genomics-based integrative analyses can provide complementary evidence of novel functions of known TSGs in the new cancer types with potential lethal effects that otherwise might have been overlooked in the analysis of individual cancers. TSGene 2.0’s new web interface has more user-friendly features for browsing the relevant information and querying the func- tionalities of TSGs. The web server is available at http: //bioinfo.mc.vanderbilt.edu/TSGene/. p To create an overview of TSGs that have also been re- ported to act as oncogenes, we compiled a gene list of 320 protein-coding oncogenes. Our information sources in- cluded a classical review of cancer genes (32), a research ar- ticle for oncogenic miRNAs (33), the UniProtKB keyword ‘proto-oncogene’ and 17 TSGs with dual roles from our lit- erature curation (Supplementary Table S1). As the result, we obtained a list of 73 TSGs with potential oncogenic roles (including 54 protein-coding TSGs and 19 miRNA TSGs) (Supplementary Table S2). This list of 73 genes is also avail- able through TSGene 2.0. We will update it frequently. Curation of the known and conflicting TSGs in literature Curation of the known and conflicting TSGs in literature To maintain consistency, we duplicated the literature query- ing strategy used for TSGene 1.0 on the PubMed and GeneRif (Gene Reference Into Function). The most recent systematic PubMed search was conducted on 25 April 2015 using the term: ‘tumor suppressor’ [Title] NOT (P53 [Ti- tle] OR TP53 [Title]). To avoid false results, we searched for matches in titles only. The search returned 6178 PubMed ab- stracts. On the same day, we also extracted 5454 additional short statements associated with 3719 PubMed abstracts from GeneRif database (25), using the term ‘tumor suppres- sor.’ For new gene type, long non-coding RNAs, we per- formed extensive PubMed searches separately by using the expression: ‘long non-coding RNA’ and ‘tumor suppressor.’ This search returned 357 references for further curation. Af- ter removing the 5795 references that have been analyzed in our previous curation, we kept ≥3000 references for the manual check. Following our previous reference curation processes, we first downloaded all the abstracts from the PubMed and grouped them according to semantic similar- ity. Next, we extracted the sentences containing the keyword ‘tumor suppressor’ and manually extracted gene names and DATA COLLECTION AND DISCUSSION In summary, we compiled 1217 human TSGs, including 1018 protein-coding and 199 non-coding genes, from 3354 PubMed abstracts with confirmed literature evidence. We stored all the curated TSGs and relevant annotations in a MySQL relational database. A dynamic web interface was implemented by Perl CGI and JavaScript for data browsing and querying. Representative entries in the TSGene 2.0 Supplementary Figure S1 shows a compilation of informa- tion provided by TSGene 2.0. Annotations for each gene can be obtained by clicking the links at the top left (General information, Expression, etc.). General information (Sup- plementary Figure S1A) displays the gene name, pathway, disease, nucleotide sequence and protein sequence. Supple- mentary Figure S1B shows highlighted summaries from supporting literature and other data sources. The Expres- sion page has differential gene expression plots in 11 cancer types; they are provided in a box view with cancer type and average expression score (Supplementary Figure S1C). Tak- ing BRCA1 as an example, the expression plot shows rela- tively higher expression in ten cancers compared to matched normal tissues (all adjusted P-values <0.05; Student’s t- Nucleic Acids Research, 2016, Vol. 44, Database issue D1025 about the complementary functions in their tumor suppres- sive roles of these genes. test; Supplementary Figure S1C). To help users obtain re- sults of statistical analysis, we present all statistical P-values for each TSG on the Expression page. Experimentally ver- ified miRNA targets and predicted upstream transcription factors for each TSG are provided on the Regulation page (Supplementary Figure S1D). Somatic mutational annota- tions for each TSG from the COSMIC database are shown on the Variant page. The Lollipop plots in Supplementary Figure S1E, the pie charts in Supplementary Figure S1F and the bar plots in Supplementary Figure S1G summa- rize somatic mutational features on protein domains, can- cer types and ratios of loss-of-function over missense mu- tations, respectively. We also used a copy number variation plot to present the copy number profile for each TSG. Fi- nally, the Interaction page shows physical interaction infor- mation. It is supported by integrated high-throughput ex- periments, metabolic and signaling interactions, and helps users explore partners interacting with each TSG (34). g Considering their high alteration frequency, DLEU1 and DLEU2 may be the candidate TSGs for further experi- mental evaluation in prostate cancer. To predict their pos- sible biological functions, we examined their co-expressed protein-coding genes using the TCGA prostate RNAseq quantitative scores with both mRNA and lncRNA expres- sion data (Supplementary Table S3). For DLEU1, only one gene, TRIM13, had a Pearson’s correlation coefficient (PCC) ≥0.5. TRIM13 encodes an E3 ubiquitin-protein lig- ase with three zinc-binding domains and a GTPase activat- ing protein domain in the amino acid terminus (35). Biological features of the 179 miRNA TSGs in humans In recent years, an increasing number of miRNAs have been reported as functioning as tumor suppressors. We have col- lected 179 tumor suppressor miRNAs; they have important roles mainly in suppressing oncogenes, thus, inhibiting tu- mor growth. To assess their relationship to protein-coding TSGs, we picked 13 well-studied TSGs, each having the ev- idence supported by at least 30 publications (Supplemen- tary Table S5). We used a gene ranking tool, ToppGene (36), to prioritize the tumor suppressor miRNAs and found 24 miRNAs that were significantly associated with the 13 TSGs (Supplementary Table S5, ranking P-values < 0.05). To further explore the functional distribution of the top ranked miRNAs, we performed a target-based functional enrichment analysis using DIANA-miRPath (37) (Table 2). Not surprisingly, the majority of the enriched pathways were related to the key cancer pathways, such as the cell cycle and tumorigenesis. By incorporating the somatic sin- gle nucleotide variant (SNV) and copy number variation (CNV) data in these 24 miRNAs, we created a mutational landscape of these critical tumor suppressor miRNAs in pan-cancer manner. They were mutated in 486 cell lines (55.20%) in the Cancer Cell Line Encyclopedia (CCLE) data set. Because miRNAs are relatively short in terms of nucleotide sequence, the high mutational rates observed in the 24 key miRNAs may reflect their critical and complex roles in cancer cells. g ) All TSGs and related information are available for down- load for analysis or large-scale project design. The data browsing functions in TSGene 2.0 are improved, and pro- vide quick access to a specific set of TSGs. Users can browse TSGs using a KEGG pathway, chromosome, gene type, cancer type, summarized expression and mutational feature (Supplementary Figure S2C–F). Representative entries in the TSGene 2.0 It is involved in the formation of intracellular vesicles transport- ing and phospholipase D activation (35). Different from DLEU1, DLEU2 had 17 protein-coding genes with the PCC values ≥0.5, including a tumor suppressor gene (E2F2). Functional enrichment analysis of these 17 genes found that they were mainly associated with the cancer pathways or cancer-related gene interactions (Supplementary Table S4). We also observed a number of amplifications for the col- lected 19 lncRNAs in many cancers (Figure 1A), which might warrant further experimental validation for their po- tential oncogenic roles in those cancers. Text querying, sequence searching and data browsing TSGene 2.0 provides text-based query and BLAST search functions (Supplementary Figure S2). A quick text search box on the top right of each page is used to query by gene symbol and Entrez gene ID. The advanced search page provides access to various TSG annotations, includ- ing gene symbol, Entrez gene ID, genomic location, disease and pathway (Supplementary Figure S2A). Users can also retrieve any specific TSG set by keyword matching. Sophis- ticated logical operators provide a window for the user to retrieve somatic mutation information on customized mu- tations, tumor types and histology types. Moreover, annota- tions on interactors, transcription factors and regulatory in- formation are also searchable (Supplementary Figure S2A). An updated BLAST interface is provided for the newly col- lected TSGs in our TSGene 2.0 database (Supplementary Figure S2B). Functional annotations of 19 lncRNA TSGs TSGene 2.0 contains information on 19 lncRNAs with the tumor suppressive features (Table 1). To explore mu- tational features of these 19 lncRNA TSGs, we mapped their genomic coordinates into TCGA’s publicly available genomic alteration data (Figure 1A). We found genetic al- terations related to them in 54 cancer types. Alteration fre- quencies were ≥10% in 25 cancer types. Prostate cancer was the most prevalent, with >22.6% of TCGA prostate tumor samples harboring copy number losses (Figure 1A). Us- ing the TCGA prostate data, we plotted the alteration pro- files of these lncRNAs. The most frequently altered lncR- NAs were DLEU1 and DLEU2, followed by MEG3 and MT1DP, both of which had the alteration frequency >2% in 313 TCGA prostate cancer patients (Figure 1B). This mu- tational pattern in prostate cancer may provide information As shown in Figure 2A, more patients with ovarian and esophageal cancers carried any type of alteration of the 24 miRNAs, and deletion events were most frequent in glioblastoma multiforme (GBM). In total, there were 10 mutated miRNAs in the TCGA GBM cohort (Figure 2B). This observation is consistent with our previous pathway analysis results (Table 2) showing that the 24 miRNAs were enriched in glioma (corrected P-value = 3.51E-06). Dele- D1026 Nucleic Acids Research, 2016, Vol. 44, Database issue Figure 1. Alteration landscape of the 19 long noncoding RNA (lncRNA) TSGs in pan-cancer. (A) Alteration profiles of 19 lncRNA TSGs cancer types with alteration frequency ≥10% in each cancer type. (B) Sample-based distribution of alterations in 19 lncRNA TSGs in TC cancer data. CNA: copy number alteration. Multiple alterations: more than one type of mutations. All the mutational analyses were conduc the cBIO portal data (45). tion of hsa-miR-31 was detected in 26% of GBM sam- ples (Figure 2B). According to the record in our TSGene also confirmed by the survival analysis in a mo (39). Further screening of patients with has-miR Figure 1. Alteration landscape of the 19 long noncoding RNA (lncRNA) TSGs in pan-cancer. (A) Alteration profiles of 19 lncRNA TSGs in 25 major cancer types with alteration frequency ≥10% in each cancer type. (B) Sample-based distribution of alterations in 19 lncRNA TSGs in TCGA prostate cancer data. CNA: copy number alteration. Multiple alterations: more than one type of mutations. All the mutational analyses were conducted based on the cBIO portal data (45). Functional annotations of 19 lncRNA TSGs tion of hsa-miR-31 was detected in 26% of GBM sam- ples (Figure 2B). According to the record in our TSGene database, Has-miR-31 has been characterized as an miRNA TSG in breast cancer (38). However, systematic study of its role in human cancers is lacking. Given that TSGs mainly operate by the ‘loss-of-function’ model, we further explored the deletion frequency of has-miR-31 in each cancer type. Figure 2C shows that it was highly deleted in both GBM and pancreatic adenocarcinoma (20.0% of the cohort). It is reported that has-miR-31 functions to inhibit GBM migra- tion and invasion (39). Its influence on the risk of GBM was also confirmed by the survival analysis in a mouse model (39). Further screening of patients with has-miR-31 dele- tions may find more clues about its roles in cancer metas- tasis and progression. protein-coding TSGs consistently down-regulated in pan- cancer Top 20 KEGG pathways enriched with the 24 microRNA TSGs KEGG pathway Adjusted P-value* # genes # microRNAs Viral carcinogenesis 5.65E-27 43 14 Cell cycle 6.81E-24 33 13 Pathways in cancer 1.51E-20 48 14 Small cell lung cancer 7.98E-17 21 13 Hepatitis B 7.98E-17 28 14 Chronic myeloid leukemia 1.73E-15 19 13 Colorectal cancer 5.48E-14 16 12 Prostate cancer 5.48E-14 20 12 Endometrial cancer 2.70E-10 12 11 Pancreatic cancer 9.02E-10 16 10 Transcriptional misregulation in cancer 1.21E-09 29 11 HTLV-I infection 1.25E-09 33 13 Bladder cancer 6.19E-09 11 9 p53 signaling pathway 7.70E-08 14 11 PI3K-Akt signaling pathway 4.48E-07 36 14 Non-small cell lung cancer 8.52E-07 11 11 Measles 1.30E-06 20 10 Glioma 3.51E-06 12 13 P-values were calculated by hypergeometric tests followed by the Benjamini-Hochberg multiple testing correction (66). Table 1. Nineteen long non-coding RNAs with tumor suppressor roles Gene symbol Cancer tissue / cell line References ADAMTS9-AS2 Glioma (46) CASC2 Colorectal, endometrial, gliomas (47,48) DLEU1 Leukemia (49) DLEU2 Leukemia (49,50) FER1L4 Gastric (51) GAS5 Breast (52) H19 Breast, hepatocellular, embryonal tumor cell line (53) LINC-PINT N/A (54) LOC401317 Nasopharyngeal (55) MEG3 Meningioma, hepatocellular, leukemia, pituitary tumor (12) MT1DP Hepatocellular (56) PTCSC3 Thyroid (57) PTENP1 Colon, hepatocellular, prostate (12,58) PWAR4 Breast, endometrial (59) TDRG1 Testicular (60) TP53COR1 Lymphoma (12,61) TUSC7 Colon (62,63) XIST Hematologic (64) ZFAS1 Breast (65) Table 2. Top 20 KEGG pathways enriched with the 24 microRNA TSGs KEGG pathway Adjusted P-value # genes # microRNAs Viral carcinogenesis 5.65E-27 43 14 Cell cycle 6.81E-24 33 13 Pathways in cancer 1.51E-20 48 14 Small cell lung cancer 7.98E-17 21 13 Hepatitis B 7.98E-17 28 14 Chronic myeloid leukemia 1.73E-15 19 13 Colorectal cancer 5.48E-14 16 12 Prostate cancer 5.48E-14 20 12 Endometrial cancer 2.70E-10 12 11 Pancreatic cancer 9.02E-10 16 10 Transcriptional misregulation in cancer 1.21E-09 29 11 HTLV-I infection 1.25E-09 33 13 Bladder cancer 6.19E-09 11 9 p53 signaling pathway 7.70E-08 14 11 PI3K-Akt signaling pathway 4.48E-07 36 14 Non-small cell lung cancer 8.52E-07 11 11 Measles 1.30E-06 20 10 Glioma 3.51E-06 12 13 *P-values were calculated by hypergeometric tests followed by the Benjamini-Hochberg multiple testing correction (66). Although these 49 genes were down-regulated in tumor samples, their mutational features have not been systemati- cally explored. To this end, we plotted all the somatic alter- ations, including single nucleotide variants and gene copy number alterations, using all publicly available TCGA can- cer data (Figure 3C). protein-coding TSGs consistently down-regulated in pan- cancer Based on the pan-cancer expression profiles in 11 cancers from TCGA, we surveyed gene expression changes for all TSGs by comparing expression in the tumor and control Nucleic Acids Research, 2016, Vol. 44, Database issue re 2. Mutational landscape of the 24 top-ranked microRNA (miRNA) TSGs in pan-cancer. (A) Somatic mutational patterns of 24 top-ranked s in multiple cancer types. (B) Sample-based copy number alteration covering 9 miRNA TSGs in TCGA high-grade glioblastoma. (C) Global tions of has-miR-31 in multiple cancer types. All the mutational analyses were conducted based on the cBIO portal data (45). Nucleic Acids Research, 2016, Vol. 44, Database issue D1027 Mutational landscape of the 24 top-ranked microRNA (miRNA) TSGs in pan-cancer. (A) Somatic mutational patterns of 24 top-ranked Figure 2. Mutational landscape of the 24 top-ranked microRNA (miRNA) TSGs in pan-cancer. (A) Somatic mutational patterns of 24 top-ranked miRNA TSGs in multiple cancer types. (B) Sample-based copy number alteration covering 9 miRNA TSGs in TCGA high-grade glioblastoma. (C) Global somatic mutations of has-miR-31 in multiple cancer types. All the mutational analyses were conducted based on the cBIO portal data (45). D1028 Nucleic Acids Research, 2016, Vol. 44, Database issue D1028 Nucleic Acids Research, 2016, Vol. 44, Database issue Table 1. Nineteen long non-coding RNAs with tumor suppressor roles Gene symbol Cancer tissue / cell line References ADAMTS9-AS2 Glioma (46) CASC2 Colorectal, endometrial, gliomas (47,48) DLEU1 Leukemia (49) DLEU2 Leukemia (49,50) FER1L4 Gastric (51) GAS5 Breast (52) H19 Breast, hepatocellular, embryonal tumor cell line (53) LINC-PINT N/A (54) LOC401317 Nasopharyngeal (55) MEG3 Meningioma, hepatocellular, leukemia, pituitary tumor (12) MT1DP Hepatocellular (56) PTCSC3 Thyroid (57) PTENP1 Colon, hepatocellular, prostate (12,58) PWAR4 Breast, endometrial (59) TDRG1 Testicular (60) TP53COR1 Lymphoma (12,61) TUSC7 Colon (62,63) XIST Hematologic (64) ZFAS1 Breast (65) Table 2. protein-coding TSGs consistently down-regulated in pan- cancer Remarkably, the 49 genes were highly mutated in 24 cancer data sets, affecting over 50% of pa- tients in each cancer cohort. The cancer in which these genes were most frequently mutated was colorectal, with 92% of tumor samples having at least one mutational event in these 49 genes (Figure 3C). Further examination of the sample- based mutational patterns in this cancer revealed that the APC genes were most frequently mutated among the 49 genes, occurring in 78% of the tumor samples (Supple- mentary Figure S3). Additionally, the mutational frequency was >10% in 8 other TSGs in colorectal cancer: FAT4 (19%), TOPORS (15%), PPP2CB (13%), RASL11A (13%), PCDH9 (12%), PRDM2 (12%), LIFR (11%) and TGFBR2 (11%). Recently, FAT4 was reported as a recurrently mu- tated driver gene in the colorectal cancer (41). However, the samples (Supplementary Table S6). We identified 8351 dif- ferentially down-regulated events among 1027 TSGs in 11 cancers (Student’s t-test, all adjusted P-values <0.01). A to- tal of 1022 TSGs were down-regulated in at least 2 can- cer types (Figure 3A), suggesting that they may be con- sistently down-regulated in multiple cancer types. Impor- tantly, we found 49 TSGs whose expression was consistently decreased in all the 11 cancer types (Supplementary Ta- ble S7). Pathway enrichment analysis indicated that these 49 genes were enriched in the TGF-beta signaling path- way (adjusted P-value = 2.03E-4) (Supplementary Table S8). The Gene Ontology (GO) analysis revealed that the genes are mainly involved in the regulation of cell prolif- eration pathways (Supplementary Table S8). Interestingly, they were enriched in the experimentally verified targets of 16 miRNAs (Figure 3B). Three of these microRNAs are well-characterized oncogenes (has-miR-21, has-miR-372 and has-miR-373) (40). This finding may indicate a similar competitive regulatory pattern between oncogenic miRNAs and TSGs (22). Nucleic Acids Research, 2016, Vol. 44, Database issue D1029 Figure 3. Biological features of 49 down-regulated TSGs in multiple cancer types. (A) Shared down-regulated TSGs in 11 major cancers. Eleven colors represent 11 cancer types, respectively. Lengths of the circularly arranged segments are proportional to the total number of TSGs in the 11 cancers. The three outer rings are stacked bar plots, representing relative overlap of other cancer TSGs to the cancer totals. Ribbons connecting different segments represent the number of shared TSGs between cancer types. (B) A microRNA regulatory network enriched with the 49 down-regulated TSGs. protein-coding TSGs consistently down-regulated in pan- cancer The network contains 16 microRNAs (red) and 45 down-regulated TSGs (blue) targeted by these connected microRNAs. (C) Global view of the somatic mutations in the 49 down-regulated TSGs in pan-cancer. Figure 3. Biological features of 49 down-regulated TSGs in multiple cancer types. (A) Shared down-regulated TSGs in 11 major cancers. Eleven colors represent 11 cancer types, respectively. Lengths of the circularly arranged segments are proportional to the total number of TSGs in the 11 cancers. The three outer rings are stacked bar plots, representing relative overlap of other cancer TSGs to the cancer totals. Ribbons connecting different segments represent the number of shared TSGs between cancer types. (B) A microRNA regulatory network enriched with the 49 down-regulated TSGs. The network contains 16 microRNAs (red) and 45 down-regulated TSGs (blue) targeted by these connected microRNAs. (C) Global view of the somatic mutations in the 49 down-regulated TSGs in pan-cancer. lack of evidence for other TSGs like TOPORS in the col- orectal cancer suggests that future studies are needed. (integrated in our database), which is a predictive score of TSGs and oncogenes based on their mutation profile pat- terns. Following the approach of TUSON, we compared the mutation ratio between the loss-of-function mutations and missense mutations based on the COSMIC annota- tions to provide additional information for the user. We de- fined loss-of-function mutations by extracting the following 7 mutation types from COSMIC: (i) frameshift deletion, (ii) whole gene deletion, (iii) complex in-frame deletion, (iv) in- frame deletion, (v) frameshift insertion, (vi) frameshift and (vii) nonsense substitution. Next, we counted the number of missense mutations for each TSG and calculated the ra- tio of the number of loss-of-function mutations to the num- ber of missense mutations. The results are shown by color bars in our web page: blue denotes the relative abundance of loss-of-function mutations and green denotes the rela- tive abundance of missense mutations. Although the ma- jority of TSGs, such as TP53, had fewer loss-of-function mutations than missense mutations (709 TSGs with ratios between 0 and 0.2), some well-known TSGs like RB1 and APC had more loss-of-function than missense mutations (i.e. ratio >0.5). RB1’s loss-of-function mutation ratio was 0.58. APC, had the highest ratio (0.84). From these results, we may infer the relative loss-of-functional impact. protein-coding TSGs consistently down-regulated in pan- cancer Such a compilation of diverse data sets or information in TSGene 2.0 enables the researchers to assess different lines of evi- dence or features for their specific projects. (integrated in our database), which is a predictive score of TSGs and oncogenes based on their mutation profile pat- terns. Following the approach of TUSON, we compared the mutation ratio between the loss-of-function mutations and missense mutations based on the COSMIC annota- tions to provide additional information for the user. We de- fined loss-of-function mutations by extracting the following 7 mutation types from COSMIC: (i) frameshift deletion, (ii) whole gene deletion, (iii) complex in-frame deletion, (iv) in- frame deletion, (v) frameshift insertion, (vi) frameshift and (vii) nonsense substitution. Next, we counted the number of missense mutations for each TSG and calculated the ra- tio of the number of loss-of-function mutations to the num- ber of missense mutations. The results are shown by color bars in our web page: blue denotes the relative abundance of loss-of-function mutations and green denotes the rela- tive abundance of missense mutations. Although the ma- jority of TSGs, such as TP53, had fewer loss-of-function mutations than missense mutations (709 TSGs with ratios between 0 and 0.2), some well-known TSGs like RB1 and APC had more loss-of-function than missense mutations (i.e. ratio >0.5). RB1’s loss-of-function mutation ratio was 0.58. APC, had the highest ratio (0.84). From these results, we may infer the relative loss-of-functional impact. Such a compilation of diverse data sets or information in TSGene 2.0 enables the researchers to assess different lines of evi- dence or features for their specific projects. SUMMARY AND LIMITATIONS 44, Database issue D1030 Nucleic Acids Research, 2016, Vol. 44, Database issue There are some limitations in our annotations. For exam- ple, we only provide the longest representative DNA and protein sequences for each TSG, and do not cover all po- tential isoform sequences. We added a note to this effect on the page for sequence information, so that users are aware that TSGene 2.0 provides only the longest representative sequences. These sequences are useful for human research, but some homologous sequences may not be found in other species if a user performs a BLAST-based sequence similar- ity search. For protein-coding genes, it may be feasible to collect homologous sequences from widely-used databases like HomoloGene (43). However, there is still a lack of high- quality homologous data for non-coding genes, particularly for long non-coding RNAs. (2011) A cooperative microRNA-tumor suppressor gene network in acute T-cell lymphoblastic leukemia (T-ALL). Nat. Genet., 43, 673–678. 7. Tay,Y., Kats,L., Salmena,L., Weiss,D., Tan,S.M., Ala,U., Karreth,F., Poliseno,L., Provero,P., Di Cunto,F. et al. (2011) Coding-independent regulation of the tumor suppressor PTEN by competing endogenous mRNAs. Cell, 147, 344–357. 8. Sherr,C.J. (2004) Principles of tumor suppression. Cell, 116, 235–246. i 9. Chen,C.Z. (2005) MicroRNAs as oncogenes and tumor suppressors. New Engl. J. Med., 353, 1768–1771. 10. Hammond,S.M. (2007) MicroRNAs as tumor suppressors. Nat. Genet., 39, 582–583. 11. Zhang,B.H., Pan,X.P., Cobb,G.P. and Anderson,T.A. (2007) microRNAs as oncogenes and tumor suppressors. Dev. Biol., 302, 1–12. 12. Prensner,J.R. and Chinnaiyan,A.M. (2011) The emergence of lncRNAs in cancer biology. Cancer Discov., 1, 391–407. We will continue to update the TSGene database as new information appears, especially data related to noncoding RNA, proteomics and metabolomics data. TSGs are in- volved in multiple steps in cancer progression, including initiation, progression and metastasis. For example, a re- cently described cancer metastasis suppressor database (44) has many overlapping between metastasis suppressors and TSGs. Therefore, we also plan to curate more specific in- formation for each TSG, such as tumor type(s) and sub- type(s), as well as data about cancer initiation, progression and metastasis. In addition, we will further integrate high- throughput genomics data like epigenetic changes for each TSG. Finally, we plan to add more useful annotations and add homologous genes in other model species. Our goal is to provide a continuously updated, high quality and content- rich literature-based TSG database to facilitate the TSG studies. 13. Zhao,M., Sun,J. and Zhao,Z. SUMMARY AND LIMITATIONS (2013) TSGene: a web resource for tumor suppressor genes. Nucleic Acids Res., 41, D970–D976. 14. Crystal,A.S., Shaw,A.T., Sequist,L.V., Friboulet,L., Niederst,M.J., Lockerman,E.L., Frias,R.L., Gainor,J.F., Amzallag,A., Greninger,P. et al. (2014) Patient-derived models of acquired resistance can identify effective drug combinations for cancer. Science, 346, 1480–1486. 15. Wagner,T.A., McLaughlin,S., Garg,K., Cheung,C.Y., Larsen,B.B., Styrchak,S., Huang,H.C., Edlefsen,P.T., Mullins,J.I. and Frenkel,L.M. (2014) HIV latency. Proliferation of cells with HIV integrated into cancer genes contributes to persistent infection. Science, 345, 570–573. 16. Cohn,L.B., Silva,I.T., Oliveira,T.Y., Rosales,R.A., Parrish,E.H., Learn,G.H., Hahn,B.H., Czartoski,J.L., McElrath,M.J., Lehmann,C. et al. (2015) HIV-1 integration landscape during latent and active infection. Cell, 160, 420–432. 17. AlQuraishi,M., Koytiger,G., Jenney,A., MacBeath,G. and Sorger,P.K. (2014) A multiscale statistical mechanical framework integrates biophysical and genomic data to assemble cancer networks. Nat. Genet., 46, 1363–1371. 18. Yoshihara,K., Wang,Q., Torres-Garcia,W., Zheng,S., Vegesna,R., Kim,H. and Verhaak,R.G. (2015) The landscape and therapeutic relevance of cancer-associated transcript fusions. Oncogene, 34, 4845–4854. FUNDING 20. Gilbert,L.A., Horlbeck,M.A., Adamson,B., Villalta,J.E., Chen,Y., Whitehead,E.H., Guimaraes,C., Panning,B., Ploegh,H.L., Bassik,M.C. et al. (2014) Genome-Scale CRISPR-Mediated Control of Gene Repression and Activation. Cell, 159, 647–661. National Institutes of Health (NIH) [R01LM011177, P50CA095103, P50CA098131 and P30CA068485 in part] and Ingram Professorship Funds [to Z.Z.]. The funders had no role in study design, data collection and analysis, deci- sion to publish or preparation of the manuscript. Funding for open access charge: Ingram Professorship Funds at Van- derbilt University.l 21. Liu,H., Flores,M.A., Meng,J., Zhang,L., Zhao,X., Rao,M.K., Chen,Y. and Huang,Y. (2015) MeT-DB: a database of transcriptome methylation in mammalian cells. Nucleic Acids Res., 43, D197–D203. 22. Zhao,M., Sun,J. and Zhao,Z. (2012) Distinct and competitive regulatory patterns of tumor suppressor genes and oncogenes in ovarian cancer. PLoS One, 7, e44175. 23. Zhao,M., Sun,J. and Zhao,Z. (2013) Synergetic regulatory networks mediated by oncogene-driven microRNAs and transcription factors in serous ovarian cancer. Mol. Biosyst., 9, 3187–3198. Conflict of interest statement. None declared. Conflict of interest statement. None declared. 24. Zhao,M. and Zhao,Z. (2013) CNVannotator: a comprehensive annotation server for copy number variation in the human genome. PLoS One, 8, e80170. Supplementary Data are available at NAR Online. Supplementary Data are available at NAR Online. SUMMARY AND LIMITATIONS We have updated our TSGene database to version 2.0, which catalogs 1217 human TSGs curated from thousands of publications. This updated database has an additional list of 572 TSGs that were manually curated from the newly published studies. Its comprehensive annotations, includ- ing the pan-cancer gene expression and mutational profiles, provide useful resources for further exploration of the bio- logical functions of TSGs and for cross-cancer comparison of TSGs. In addition, the massive precomputed results and graphic presentations in TSGene 2.0 will be helpful for the cancer-specific TSG identification and subsequent analyses. We have updated our TSGene database to version 2.0, which catalogs 1217 human TSGs curated from thousands of publications. This updated database has an additional list of 572 TSGs that were manually curated from the newly published studies. Its comprehensive annotations, includ- ing the pan-cancer gene expression and mutational profiles, provide useful resources for further exploration of the bio- logical functions of TSGs and for cross-cancer comparison of TSGs. In addition, the massive precomputed results and graphic presentations in TSGene 2.0 will be helpful for the cancer-specific TSG identification and subsequent analyses. Although we performed extensive literature searching and curation, it is important to acknowledge the difficulty of performing an error-free search. For example, our strict search strategy, which matched the keyword to the reference title, caused us to miss some newly reported, but not widely accepted TSGs. We overcame this shortcoming in this up- date by performing an extensive literature search. In our ref- erence curation, we collected a number of TSGs with po- tential oncogenic roles. Based on oncogenes collected from public resources, we pinpointed 73 TSGs with such roles. We would suggest that users use the TUSON score (42) pii q y Although we performed extensive literature searching and curation, it is important to acknowledge the difficulty of performing an error-free search. For example, our strict search strategy, which matched the keyword to the reference title, caused us to miss some newly reported, but not widely accepted TSGs. We overcame this shortcoming in this up- date by performing an extensive literature search. In our ref- erence curation, we collected a number of TSGs with po- tential oncogenic roles. Based on oncogenes collected from public resources, we pinpointed 73 TSGs with such roles. We would suggest that users use the TUSON score (42) D1030 Nucleic Acids Research, 2016, Vol. SUPPLEMENTARY DATA 19. Blattler,A., Yao,L., Witt,H., Guo,Y., Nicolet,C.M., Berman,B.P. and Farnham,P.J. (2014) Global loss of DNA methylation uncovers intronic enhancers in genes showing expression changes. Genome Biol., 15, 469–484. REFERENCES 1. Siegel,R.L., Miller,K.D. and Jemal,A. (2015) Cancer statistics, 2015. CA Cancer J. Clin., 65, 5–29. 25. Lu,Z., Cohen,K.B. and Hunter,L. (2007) GeneRIF quality assurance as summary revision. Pac. Symp. Biocomput., 2007, 269–280. 2. Balmain,A., Gray,J. and Ponder,B. (2003) The genetics and genomics of cancer. Nat. Genet., 33(Suppl), 238–244. 26. Santoro,F., Botrugno,O.A., Dal Zuffo,R., Pallavicini,I., Matthews,G.M., Cluse,L., Barozzi,I., Senese,S., Fornasari,L., Moretti,S. et al. (2013) A dual role for Hdac1: oncosuppressor in tumorigenesis, oncogene in tumor maintenance. Blood, 121, 3459–3468. 3. Haber,D.A. and Settleman,J. (2007) Cancer: drivers and passengers. Nature, 446, 145–146. 4. Esteller,M. (2002) CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future. Oncogene, 21, 5427–5440. 27. Pelosi,G., Fumagalli,C., Trubia,M., Sonzogni,A., Rekhtman,N., Maisonneuve,P., Galetta,D., Spaggiari,L., Veronesi,G., Scarpa,A. et al. (2010) Dual role of RASSF1 as a tumor suppressor and an oncogene in neuroendocrine tumors of the lung. Anticancer Res., 30, 4269–4281. 5. Zack,T.I., Schumacher,S.E., Carter,S.L., Cherniack,A.D., Saksena,G., Tabak,B., Lawrence,M.S., Zhsng,C.Z., Wala,J., Mermel,C.H. et al. (2013) Pan-cancer patterns of somatic copy number alteration. Nat. Genet., 45, 1134–1140. 6. Mavrakis,K.J., Van Der Meulen,J., Wolfe,A.L., Liu,X., Mets,E., Taghon,T., Khan,A.A., Setty,M., Rondou,P., Vandenberghe,P. et al. Nucleic Acids Research, 2016, Vol. 44, Database issue D1031 Nucleic Acids Research, 2016, Vol. 44, Database issue D1031 Nucleic Acids Research, 2016, Vol. 44, Database issue D1031 (1997) Cloning of two candidate tumor suppressor genes within a 10 kb region on chromosome 13q14, frequently deleted in chronic lymphocytic leukemia. Oncogene, 15, 2463–2473. 28. Lobry,C., Oh,P. and Aifantis,I. (2011) Oncogenic and tumor suppressor functions of Notch in cancer: it’s NOTCH what you think. J. Exp. Med., 208, 1931–1935. (1997) Cloning of two candidate tumor suppressor genes within a 10 kb region on chromosome 13q14, frequently deleted in chronic lymphocytic leukemia. Oncogene, 15, 2463–2473. g q q y lymphocytic leukemia. Oncogene, 15, 2463–2473. 29. Lamouille,S. and Derynck,R. (2009) Oncogene and tumour suppressor: the two faces of SnoN. EMBO J., 28, 3459–3460. 50. Bullrich,F., Fujii,H., Calin,G., Mabuchi,H., Negrini,M., Pekarsky,Y., Rassenti,L., Alder,H., Reed,J.C., Keating,M.J. et al. (2001) Characterization of the 13q14 tumor suppressor locus in CLL: identification of ALT1, an alternative splice variant of the LEU2 gene. Cancer Res., 61, 6640–6648. 30. Yang,L., Han,Y., Suarez Saiz,F. and Minden,M.D. (2007) A tumor suppressor and oncogene: the WT1 story. Leukemia, 21, 868–876. 31. Uribesalgo,I., Benitah,S.A. and Di Croce,L. (2012) From oncogene to tumor suppressor: the dual role of Myc in leukemia. Cell Cycle, 11, 1757–1764. 51. REFERENCES Xia,T., Chen,S., Jiang,Z., Shao,Y., Jiang,X., Li,P., Xiao,B. and Guo,J. (2015) Long noncoding RNA FER1L4 suppresses cancer cell growth by acting as a competing endogenous RNA and regulating PTEN expression. Sci. Rep., 5, 13445. 32. Vogelstein,B., Papadopoulos,N., Velculescu,V.E., Zhou,S., Diaz,L.A. Jr and Kinzler,K.W. (2013) Cancer genome landscapes. Science, 339, 1546–1558. 52. Mourtada-Maarabouni,M., Pickard,M.R., Hedge,V.L., Farzaneh,F. and Williams,G.T. (2009) GAS5, a non-protein-coding RNA, controls apoptosis and is downregulated in breast cancer. Oncogene, 28, 195–208. 33. Wang,D., Qiu,C., Zhang,H., Wang,J., Cui,Q. and Yin,Y. (2010) Human microRNA oncogenes and tumor suppressors show significantly different biological patterns: from functions to targets. PLoS One, 5, e13067. 53. Hao,Y., Crenshaw,T., Moulton,T., Newcomb,E. and Tycko,B. (1993) Tumour-suppressor activity of H19 RNA. Nature, 365, 764–767. 34. Cerami,E.G., Gross,B.E., Demir,E., Rodchenkov,I., Babur,O., Anwar,N., Schultz,N., Bader,G.D. and Sander,C. (2011) Pathway Commons, a web resource for biological pathway data. Nucleic Acids Res., 39, D685–D690. 54. Marin-Bejar,O., Marchese,F.P., Athie,A., Sanchez,Y., Gonzalez,J., Segura,V., Huang,L., Moreno,I., Navarro,A., Monzo,M. et al. (2013) Pint lincRNA connects the p53 pathway with epigenetic silencing by the Polycomb repressive complex 2. Genome Biol., 14, R104. 35. Vitale,N., Horiba,K., Ferrans,V.J., Moss,J. and Vaughan,M. (1998) Localization of ADP-ribosylation factor domain protein 1 (ARD1) in lysosomes and Golgi apparatus. Proc. Natl. Acad. Sci. U.S.A., 95, 8613–8618. lycomb repressive complex 2. Genome Biol., 14, R104. 55. Gong,Z., Zhang,S., Zeng,Z., Wu,H., Yang,Q., Xiong,F., Shi,L., Yang,J., Zhang,W., Zhou,Y. et al. (2014) LOC401317, a p53-regulated long non-coding RNA, inhibits cell proliferation and induces apoptosis in the nasopharyngeal carcinoma cell line HNE2. PLoS One, 9, e110674. 36. Chen,J., Bardes,E.E., Aronow,B.J. and Jegga,A.G. (2009) ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res., 37, W305–W311. 56. Yu,W., Qiao,Y., Tang,X., Ma,L., Wang,Y., Zhang,X., Weng,W., Pan,Q., Yu,Y., Sun,F. et al. (2014) Tumor suppressor long non-coding RNA, MT1DP is negatively regulated by YAP and Runx2 to inhibit FoxA1 in liver cancer cells. Cell. Signal., 26, 2961–2968. 37. Vlachos,I.S., Zagganas,K., Paraskevopoulou,M.D., Georgakilas,G., Karagkouni,D., Vergoulis,T., Dalamagas,T. and Hatzigeorgiou,A.G. (2015) DIANA-miRPath v3.0: deciphering microRNA function with experimental support. Nucleic Acids Res., 43, W460–W466. 57. Fan,M., Li,X., Jiang,W., Huang,Y., Li,J. and Wang,Z. (2013) A long non-coding RNA, PTCSC3, as a tumor suppressor and a target of miRNAs in thyroid cancer cells. Exp. Ther. Med., 5, 1143–1146. 38. O’Day,E. and Lal,A. (2010) MicroRNAs and their target gene networks in breast cancer. Breast Cancer Res., 12, 201. 58. Chen,C.L., Tseng,Y.W., Wu,J.C., Chen,G.Y., Lin,K.C., Hwang,S.M. and Hu,Y.C. REFERENCES (2015) Suppression of hepatocellular carcinoma by baculovirus-mediated expression of long non-coding RNA PTENP1 and MicroRNA regulation. Biomaterials, 44, 71–81. 39. Hua,D., Ding,D., Han,X., Zhang,W., Zhao,N., Foltz,G., Lan,Q., Huang,Q. and Lin,B. (2012) Human miR-31 targets radixin and inhibits migration and invasion of glioma cells. Oncol. Rep., 27, 700–706. 40. Kent,O.A. and Mendell,J.T. (2006) A small piece in the cancer puzzle: microRNAs as tumor suppressors and oncogenes. Oncogene, 25, 6188–6196. 59. Zapata-Benavides,P., Mendez-Vazquez,J.L., Gonzalez-Rocha,T.R., Zamora-Avila,D.E., Franco-Molina,M.A., Garza-Garza,R. and Rodriguez-Padilla,C. (2009) Expression of prostate apoptosis response (Par-4) is associated with progesterone receptor in breast cancer. Arch. Med. Res., 40, 595–599. 41. Yu,J., Wu,W.K., Li,X., He,J., Li,X.X., Ng,S.S., Yu,C., Gao,Z., Yang,J., Li,M. et al. (2015) Novel recurrently mutated genes and a prognostic mutation signature in colorectal cancer. Gut, 64, 636–645. 60. Chen,H.Y., Wen,J.M., Xiao,X.W., Li,D.J., Guo,X.L., Long,Z., Dai,Y.B. and Tang,Y.X. (2010) Expression of human testis development related gene 1 in testicular cancer detected by tissue microarray. Zhonghua Nan Ke Xue, 16, 883–886. 42. Davoli,T., Xu,A.W., Mengwasser,K.E., Sack,L.M., Yoon,J.C., Park,P.J. and Elledge,S.J. (2013) Cumulative haploinsufficiency and triplosensitivity drive aneuploidy patterns and shape the cancer genome. Cell, 155, 948–962. 61. Huarte,M., Guttman,M., Feldser,D., Garber,M., Koziol,M.J., Kenzelmann-Broz,D., Khalil,A.M., Zuk,O., Amit,I., Rabani,M. et al. (2010) A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell, 142, 409–419. g 43. NCBI Resource Coordinators. (2015) Database resources of the National Center for Biotechnology Information. Nucleic Acids Res., 43, D6–D17. 44. Zhao,M., Li,Z. and Qu,H. (2015) An evidence-based knowledgebase of metastasis suppressors to identify key pathways relevant to cancer metastasis. Sci. Rep., 5, 15478. 62. Liu,Q., Huang,J., Zhou,N., Zhang,Z., Zhang,A., Lu,Z., Wu,F. and Mo,Y.Y. (2013) LncRNA loc285194 is a p53-regulated tumor suppressor. Nucleic Acids Res., 41, 4976–4987. 45. Gao,J., Aksoy,B.A., Dogrusoz,U., Dresdner,G., Gross,B., Sumer,S.O., Sun,Y., Jacobsen,A., Sinha,R., Larsson,E. et al. (2013) Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci. Signal., 6, pl1. 63. Pasic,I., Shlien,A., Durbin,A.D., Stavropoulos,D.J., Baskin,B., Ray,P.N., Novokmet,A. and Malkin,D. (2010) Recurrent focal copy-number changes and loss of heterozygosity implicate two noncoding RNAs and one tumor suppressor gene at chromosome 3q13.31 in osteosarcoma. Cancer Res., 70, 160–171. 3q13.31 in osteosarcoma. Cancer Res., 70, 160–171. 46. Yao,J., Zhou,B., Zhang,J., Geng,P., Liu,K., Zhu,Y. and Zhu,W. (2014) A new tumor suppressor LncRNA ADAMTS9-AS2 is regulated by DNMT1 and inhibits migration of glioma cells. Tumour Biol., 35, 7935–7944. 64. Yildirim,E., Kirby,J.E., Brown,D.E., Mercier,F.E., Sadreyev,R.I., Scadden,D.T. and Lee,J.T. REFERENCES (2013) Xist RNA is a potent suppressor of hematologic cancer in mice. Cell, 152, 727–742. 47. Wang,P., Liu,Y.H., Yao,Y.L., Li,Z., Li,Z.Q., Ma,J. and Xue,Y.X. (2015) Long non-coding RNA CASC2 suppresses malignancy in human gliomas by miR-21. Cell. Signal., 27, 275–282. 65. Askarian-Amiri,M.E., Crawford,J., French,J.D., Smart,C.E., Smith,M.A., Clark,M.B., Ru,K., Mercer,T.R., Thompson,E.R., Lakhani,S.R. et al. (2011) SNORD-host RNA Zfas1 is a regulator of mammary development and a potential marker for breast cancer. RNA, 17, 878–891. 48. Baldinu,P., Cossu,A., Manca,A., Satta,M.P., Sini,M.C., Palomba,G., Dessole,S., Cherchi,P., Mara,L., Tanda,F. et al. (2007) CASC2a gene is down-regulated in endometrial cancer. Anticancer Res., 27, 235–243. 66. Benjamini,Y. and Hochberg,Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. B Stat. Methodol., 57, 289–300. 49. Liu,Y., Corcoran,M., Rasool,O., Ivanova,G., Ibbotson,R., Grander,D., Iyengar,A., Baranova,A., Kashuba,V., Merup,M. et al.
https://openalex.org/W2794339447	https://europepmc.org/articles/pmc5846787?pdf=render	English	null	Epidemiological and histological findings implicate matrix Gla protein in diastolic left ventricular dysfunction	PloS one	2,018	cc-by	10,794	RESEARCH ARTICLE Objectives Accepted: February 22, 2018 Published: March 12, 2018 A novel paradigm of diastolic left ventricular (LV) dysfunction proposes involvement of the cardiac microvasculature. Vitamin K dependent matrix Gla protein (MGP) plays a role in pre- serving microcirculatory integrity. We hypothesized that LV filling pressure–a measure of diastolic LV dysfunction–increases with higher plasma level of inactive desphospho-uncar- boxylated MGP (dp-ucMGP). We also studied the distribution of active and inactive MGP in human myocardium. Copyright: © 2018 Wei et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. OPEN ACCESS OPEN ACCESS Citation: Wei F-F, Trenson S, Monney P, Yang W- Y, Pruijm M, Zhang Z-Y, et al. (2018) Epidemiological and histological findings implicate matrix Gla protein in diastolic left ventricular dysfunction. PLoS ONE 13(3): e0193967. https:// doi.org/10.1371/journal.pone.0193967 Citation: Wei F-F, Trenson S, Monney P, Yang W- Y, Pruijm M, Zhang Z-Y, et al. (2018) Epidemiological and histological findings implicate matrix Gla protein in diastolic left ventricular dysfunction. PLoS ONE 13(3): e0193967. https:// doi.org/10.1371/journal.pone.0193967 ☯These authors contributed equally to this work. ☯These authors contributed equally to this work. * jan.staessen@med.kuleuven.be ☯These authors contributed equally to this work. * jan.staessen@med.kuleuven.be Abstract Editor: Vincenzo Lionetti, Scuola Superiore Sant’Anna, ITALY Editor: Vincenzo Lionetti, Scuola Superiore Sant’Anna, ITALY Methods Data Availability Statement: Data is available upon request as consent given by study participants did not include data sharing with third parties. Anonymized data can be made available to researchers who meet the criteria for accessing confidential data, and data requests should be submitted via URL http://www.uzleuven.be/ ethische-commissie/onderzoek. The Ethics Committee of the University Hospitals Leuven has imposed the data sharing restrictions. We measured echocardiographic diastolic LV function and plasma dp-ucMGP (ELISA) in 668 Flemish and for replication in 386 Swiss. Epidemiological and histological findings implicate matrix Gla protein in diastolic left ventricular dysfunction Fang-Fei Wei1☯, Sander Trenson2☯, Pierre Monney3, Wen-Yi Yang1, Menno Pruijm4, Zhen- Yu Zhang1, Yassine Bouatou5, Qi-Fang Huang1, Belen Ponte6, Pierre-Yves Martin6, Lutgarde Thijs1, Tatiana Kuznetsova1, Karel Allegaert7,8, Stefan Janssens2, Cees Vermeer9, Peter Verhamme10, Michel Burnier4, Murielle Bochud11, Georg Ehret12,13, Jan A. Staessen1,9* 1 Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium, 2 Division of Cardiology, University Hospitals Leuven, Leuven, Belgium, 3 Department of Cardiology, Centre Hospitalier Universitaire Vaudois, University of Lausanne, Lausanne, Switzerland, 4 Department of Nephrology, University Hospital of Lausanne, Lausanne, Switzerland, 5 Department of Pathology, Academisch Medisch Centrum, Universiteit van Amsterdam, Amsterdam, The Netherlands, 6 Department of Nephrology, University Hospital of Geneva, Geneva, Switzerland, 7 Research Unit Organ Systems, Department of Development and Regeneration, University of Leuven, Leuven, Belgium, 8 Department of Pediatric Surgery, Erasmus Medical Center, Sophia Children’s Hospital, Rotterdam, The Netherlands, 9 R&D Group VitaK, Maastricht University, Maastricht, The Netherlands, 10 Research Unit Molecular and Vascular Biology, Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium, 11 Division of Chronic Disease, Institute of Social and Preventive Medicine, University Hospital of Lausanne, Lausanne, Switzerland, 12 Department of Cardiology, University Hospital of Geneva, Geneva, Switzerland, 13 Center for Complex Disease Genomics, McKusick-Nathans Institute of Genetic Medicine, John Hopkins University, Baltimore, Maryland, United States of America a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ☯These authors contributed equally to this work. * jan.staessen@med.kuleuven.be Conclusions Our study supports a role of activated MGP in maintaining myocardial integrity and diastolic LV performance and can potentially be translated into new strategies for managing diastolic LV dysfunction and preventing its progression to heart failure. Diastolic LV function and matrix Gla protein (P0.026); the odds ratios for having E/e’ 8.5 were 1.99, 3.29 and 2.36, respectively (P0.017). Cardiac biopsies from patients with ischemic or dilated cardiomyopathy and healthy hearts (n = 4 for each) were stained with conformation-specific MGP antibodies. In diseased compared with normal hearts, uncarboxylated inactive MGP was more prevalent (P0.004) in the perivascular matrix and interstitium (204.4 vs. 8.6 μm2 per field) and phos- phorylated active MGP in and around capillaries and interstitial cells (31.3 vs. 6.6 number of positive capillaries and cells per field). Funding: The European Union (HEALTH-F7- 305507 HOMAGE and the European Research Council (Advanced Researcher Grant 2011- 294713-EPLORE and Proof-of-Concept Grant 713601-uPROPHET) and the Fonds voor Wetenschappelijk Onderzoek Vlaanderen, Ministry of the Flemish Community, Brussels, Belgium (G.0881.13 and 11O3416N) currently support the Studies Coordinating Centre in Leuven. The Swiss National Science Foundation funded the Swiss Kidney Project on Genes in Hypertension (FN 33CM30-140331/1). G Ehret is supported by the Fondation pour Recherches Me´dicales, Geneva, Switzerland. The EPLORE grant paid for the plasma dp ucMGP measurement at R&D Group VitaK (Maastricht University, The Netherlands). Jan A. Staessen was a scientific adviser to R&D Group VitaK, but did never receive any monetary or other compensation from VitaK. Cees Vermeer was an employee of and received a salary from R&D Group VitaK, but this did not influence the study design, decision to publish or the preparation of the manuscript. Likewise the funders providing support to the Academic Partners in the form of authors’ salaries (FFW, ST, WYY, ZYZ, QFH, LT) did not have any role in the study design, data collection and analysis, decision to publish, or in preparation of the manuscript. Introduction In view of the demographic transition, heart failure (HF) is a major public health problem [1]. Diastolic HF, also referred to as HF with preserved ejection fraction, accounts for 50% of cases [2]. Mortality of diastolic HF is 30% within one year of the first hospital admission [3]. Sub- clinical diastolic left ventricular (LV) dysfunction has a prevalence of 25% in the general popu- lation [4,5], predisposes to further deterioration of LV function [6], and finally to overt HF [2]. A novel paradigm of diastolic HF focuses on proinflammatory signaling originating from the cardiac microvasculature [2]. Matrix Gla protein (MGP) is a small protein (11 kD) synthesized by vascular smooth muscle cells and the endothelium [7]. Activation of MGP requires two posttranslational modifications: vitamin-K dependent γ-glutamate carboxylation and serine phosphorylation [7]. Activated MGP is a potent locally acting inhibitor of calcification in large arteries [8] and protects against macrovascular complications [9,10]. Competing interests: We confirm that R&D Group VitaK (Maastricht University, The Netherlands) did not alter our adherence to PLOS ONE policies on sharing data and materials. Recent studies support a role of MGP in preserving the integrity of the microcirculation [11] and protecting against calcium deposition [12,13]. Relaxation of the heart requires seques- tration of calcium ions in the endoplasmatic reticulum [14]. In view of the novel paradigm implicating the microcirculation in the pathogenesis of diastolic LV dysfunction [2] and the potential system-wide role of MGP [9–13], we hypothesized that echocardiographically assessed diastolic LV function might be inversely associated with inactive circulating despho- spho-uncarboxylated MGP (dp-ucMGP). We investigated our hypothesis in a Flemish [13,15] and Swiss [8] population sample. At the time of writing of this manuscript, studies dealing with expression of MGP in the heart were scarce and did not provide any detail where in the myocardium MGP was expressed [16]. After having obtained our initial finding of association between diastolic LV function with circulating dp-ucMGP in Flemish and after having it repli- cated in Swiss, we undertook histological studies to confirm the presence of MGP in the heart and to determine its exact localization in healthy and diseased hearts, using conformation-spe- cific MGP antibodies. Competing interests: We confirm that R&D Group VitaK (Maastricht University, The Netherlands) did not alter our adherence to PLOS ONE policies on sharing data and materials. Results Among Flemish and Swiss, E/e’ (6.78 vs. 6.73) and dp-ucMGP (3.94 μg/L vs. 4.20 μg/L) were similarly distributed. In multivariable-adjusted models, for each doubling of dp-ucMGP, E/e’ increased by 0.26, 0.33 and 0.31 in Flemish, Swiss and both cohorts combined 1 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Population studies The Flemish Study on Environment, Genes and Health Outcomes (FLEMENGHO) [13,15] and the Swiss KIdney Project On Genes in Hypertension (SKIPOGH) [8] are family-based PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 2 / 20 Diastolic LV function and matrix Gla protein population studies, which comply with the Helsinki declaration for research in human subjects [17] and received ethical approval. All participants provided informed written consent. FLE- MENGHO participants were recruited from a geographically defined area in Northern Bel- gium. After recruitment from 1985 until 2004 (initial participation rate, 78.0%), participants remained in follow-up. From 2005 until 2014, the re-examination included echocardiography (re-examination rate, 80.3%). Of 964 participants, 696 had dp-ucMGP measured concurrently with echocardiography. The SKIPOGH participants were enrolled from 2009 until 2013 in Berne, Geneva and Lausanne, Switzerland [8]. Of 1129 participants (participation rate, 25.6%), 390 had both echocardiographic images and plasma dp-ucMGP levels available for this analy- sis. For the current analysis, we excluded 17 Flemish, because of atrial fibrillation (n = 6), paced heart rhythm (n = 2) or poor image quality (n = 9) making assessment of diastolic LV function difficult. In addition, we excluded 11 Flemish and 4 Swiss, because indexes of LV function (n = 10) or MGP levels (n = 5) were more than 3 SDs away from the population mean. Thus, the number of participants statistically analyzed totaled 668 Flemish and 386 Swiss. In the two cohorts, echocardiographic images were acquired and analyzed off-line accord- ing to current guidelines [18,19], following methods described in detail elsewhere [4]. In short, echocardiographic images were obtained with a Vivid7 Pro device (GE Vingmed, Horten, Norway) interfaced with a 2.5- to 3.5-MHz phased-array probe. For off-line analysis, we used EchoPac software, version 4.0.4 (GE Vingmed, Horten, Norway) and averaged measurements over three heart cycles. We determined the peak early (E) and peak late (A) diastolic velocities of the transmitral blood flow from the pulsed Doppler signal and peak early (e’) and peak late (a’) velocities of the mitral annular movement by tissue Doppler imaging (TDI) with velocities averaged over four acquisition sites (septal, lateral, inferior, and posterior). The intra-observer reproducibility coefficient for the single observer in FLEMENGHO was 4.8% for e’ and 4.2% for a’ [6]. In SKIPOGH (two observers), the intra- and inter-observer reproducibility coeffi- cients were 4.4% and 13.8% for e’ and 5.0% and 12.0% for a’, respectively. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Population studies The P value is for departure of the actually observed logarithmically transformed distribution (kernel distribution; dotted line) from normality (full line). The geometric mean concentration was 3.94 μg/L in Flemish and 4.20 μg/L in Swiss (P = 0.056). https://doi.org/10.1371/journal.pone.0193967.g001 https://doi.org/10.1371/journal.pone.0193967.g001 relevance have been described in previous publications [4–6] and included sex, age, body mass index, mean arterial pressure, pulse pressure, heart rate, serum total cholesterol, plasma glucose, left ventricular mass indexed to body surface area, alcohol intake and use of antihypertensive drugs. Analyses including both Flemish and Swiss were additionally adjusted for center. Population studies To stage diastolic LV function, we combined the velocities of the transmitral blood flow and the mitral annular movement [4]. Patients with diastolic LV dysfunction had an abnormally low age-specific transmitral E/A ratio indicative of impaired relaxation, a mildly-to-moderately elevated LV filling pressure (E/e’ >8.5) with normal or decreased age-specific E/A ratio. These age-specific criteria in a healthy reference sample drawn from FLEMENGHO [4] were replicated in an independent European population study [5]. Blood pressure was the average of five consecutive auscultatory readings obtained with a standard mercury sphygmomanometer in FLEMENGHO [13,15] and with a non-mercury manual auscultatory sphygmomanometer (A&D UM-101; A&D Company Ltd, Toshima Ku, Tokyo, Japan [20]) in SKIPOGH [8]. Nurses administered questionnaires inquiring into each participant’s medical history, smoking and drinking habits, and intake of medications. Fasting blood samples were analyzed for glucose and total cholesterol, using automated methods in certified laboratories. In both cohorts, dp-ucMGP was measured by VitaK (Maastricht Univer- sity, The Netherlands) on citrated plasma by pre-commercial ELISA kits [21]. For database management and statistical analysis, we used the SAS system, version 9.4 (SAS Institute Inc., Cary, NC). Significance was a two-tailed α-level of 0.05 or less. Means were com- pared using the large-sample z-test and proportions by Fisher’s exact test. We normalized the distribution of dp-ucMGP by a logarithmic transformation. We compared cohort-, sex- and age-standardized echocardiographic measurements across halves of the dp-ucMGP distribu- tion. We applied mixed models and logistic regression to model the multivariable-adjusted associations of continuous or categorical LV traits with dp-ucMGP, while accounting for clus- tering within families and cohort as random effects, as appropriate. Covariables with potential PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 3 / 20 Diastolic LV function and matrix Gla protein Fig 1. Frequency distributions of dp-ucMGP in Flemish and Swiss. S and K are the coefficients of skewness and kurtosis. The P value is for departure of the actually observed logarithmically transformed distribution (kernel distribution; dotted line) from normality (full line). The geometric mean concentration was 3.94 μg/L in Flemish and 4.20 μg/L in Swiss (P = 0.056). htt //d i /10 1371/j l 0193967 001 Fig 1. Frequency distributions of dp-ucMGP in Flemish and Swiss. S and K are the coefficients of skewness and Fig 1. Frequency distributions of dp-ucMGP in Flemish and Swiss. S and K are the coefficients of skewness and kurtosis. Diastolic LV function and matrix Gla protein Table 1. Characteristics of participants. Characteristic FLEMENGHO SKIPOGH All dp-ucMGP Threshold (μg/L) 4.16 >4.16 4.51 >4.51 4.29 >4.29 Number (%) with characteristics 334 334 194 192 527 527 Women (%) 163 (48.8) 172 (51.5) 101 (52.1) 109 (56.8) 265 (50.3) 280 (53.1) Current smoker (%) 88 (26.4) 49 (14.7)‡ 67 (34.5) 40 (20.8)† 153 (29.0) 91 (17.3)‡ Drinking alcohol (%) 243 (72.8) 219 (65.6) 135 (69.6) 124 (64.6) 375 (71.2) 346 (65.6) Diabetes mellitus (%) 6 (1.8) 14 (4.2) 1 (0.52) 6 (3.1) 8 (1.5) 19 (3.6) Hypertension (%) 92 (27.5) 162 (48.5)‡ 24 (12.4) 58 (30.2)‡ 119 (22.6) 217 (41.2)‡ Treated hypertension (%) 51 (15.3) 99 (29.6)‡ 10 (5.2) 29 (15.1)† 64 (12.1) 125 (23.7)‡ Mean (±SD) of characteristics Age (years) 44.6±14.1 53.5±15.1‡ 43.8±15.6 56.8±15.1‡ 44.3±14.7 54.7±15.1‡ Body mass index (kg/m2) 25.1±4.1 27.4±4.4‡ 24.2±4.0 26.7±4.6‡ 24.8±4.0 27.2±4.5‡ Systolic pressure (mm Hg) 124.4±14.0 131.1±16.1‡ 114.6±16.1 122.5±15.5‡ 121.0±15.6 127.8±16.4‡ Diastolic pressure (mm Hg) 78.4±9.1 81.1±9.4‡ 73.1±8.4 77.2±8.9‡ 76.5±9.3 79.6±9.4‡ Pulse pressure (mm Hg) 46.0±11.2 50.1±14.1‡ 41.5±12.7 45.4±12.2‡ 44.6±11.9 48.2±13.7‡ Heart rate (beats/min) 62.0±9.1 64.4±9.4‡ 65.0±9.5 67.1±11.2 63.2±9.4 65.3±10.1‡ Total cholesterol (mmol/L) 5.07±0.90 5.29±0.96‡ 4.78±0.98 5.26±1.10‡ 4.97±0.94 5.27±1.02‡ Plasma glucose (mmol/L) 4.82±0.64 4.95±0.74 4.82±0.57 5.21±1.17‡ 4.82±0.62 5.04±0.93‡ Serum creatinine (μmol/L) 80.8±13.4 84.7±15.4‡ 73.9±11.7 78.6±16.4‡ 78.2±13.1 82.4±16.1‡ Table 1. Characteristics of participants. Values are mean (SD) or number of subjects (%). Diabetes was a fasting plasma glucose 7.0 mmol/L or use of antidiabetic drugs. Hypertension was a blood pressure of 140 mm Hg systolic or 90 mm Hg diastolic or use of antihypertensive drugs. Pulse pressure is the difference between systolic and diastolic blood pressure. P values indicate the significance of the differences across median of the matrix Gla protein distribution https://doi.org/10.1371/journal.pone.0193967.t001 ethical screening by the European Research Council Executive Agency (ERCEA). Patients undergoing transplantation provided informed written consent. Donors of healthy hearts dis- carded for transplantation were unconscious and could not provide consent. However, their next of kin did provide consent for removal of the donor heart. The identity of the donors remained unknown to the transplant team in Leuven. As outlined above, the Ethics Commit- tee of the University Hospitals Leuven approved the use of discarded donor hearts for research. Tissue samples were paraffin embedded and stained after antigen retrieval (Dako, Glostrup, Denmark) and blocking steps with methanol + 1% H2O2 and goat serum (1:5). Immunofluo- rescence staining was performed on 6-μm paraffin sections. Histological studies Cardiac tissue samples were obtained from patients with end-stage ischemic (ICM, n = 4) or dilated (DCM, n = 4) cardiomyopathy undergoing cardiac transplantation at the University Hospitals Leuven, Belgium. Control tissue samples were obtained from unused donor hearts (HD, n = 4). The Ethics Committee of the University Hospitals Leuven approved the study (approval numbers B322201421186 [S56384] and B322201421045 [S56472]) and it passes PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 4 / 20 Values are mean (SD) or number of subjects (%). Diabetes was a fasting plasma glucose 7.0 mmol/L or use of antidiabetic drugs. Hypertension was a blood pressure of 140 mm Hg systolic or 90 mm Hg diastolic or use of antihypertensive drugs. Pulse pressure is the difference between systolic and diastolic blood pressure. P values indicate the significance of the differences across median of the matrix Gla protein distribution P0 05 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 https://doi.org/10.1371/journal.pone.0193967.t001 Images were recorded on a Carl Zeiss LSM700 confocal microscope and analysed with ZEN 2012 software (Jena, Germany). For staining, we used primary mouse anti-human conformation-specific antibodies (VitaK, Maastricht, The Netherlands) specifically directed against uncarboxylated MGP (ucMGP; 1:100), carboxylated MGP (cMGP; 1:250), desphospho-MGP (dpMGP; 1:100) and phosphory- lated MGP (pMGP; 1:100). The secondary biotinylated goat anti-mouse antibody Dako, E0433, 1:300) was amplified with Streptavidin and Cy3 (Perkin Elmer, TSA Cy3 system NEL704A001KT; 1:100 and 1:50). An Alexa Fluor1 488 conjugate of wheat germ agglutinin (WGA, W11261, 1:100, Thermo-Fisher Scientific, Waltham, MA) was used as lectin to stain cell membranes and enhance morphological characterization. Endothelium was stained with mouse anti-human CD31 (Dako M0823; 1:50) and smooth muscle cells with a monoclonal mouse anti-human SM actin (M0851, Dako, 1:500). Nuclei were counterstained with TO-PRO-3 (Thermo-fisher, T3605, PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 5 / 20 Values are regression coefficients (95% confidence interval). Estimates accounted for family cluster were adjusted sex, age, body mass index, mean arterial pressure, pulse pressure, heart rate, total cholesterol, plasma glucose, left ventricular mass index, alcohol intake and antihypertensive drug treatment. Estimates combining Flemish and Swiss were additionally adjusted for center. Values are mean (SD). Ejection fraction was available in 631 Flemish, 379 Swiss and 1010 all participants combined. Significance of the differences across halves of the matrix Gla protein distribution Diastolic LV function and matrix Gla protein Table 2. Sex- and age-standardized echocardiographic measurements by median of dp-ucMGP distribution. Measurement FLEMENGHO SKIPOGH All dp-ucMGP threshold (μg/L) 4.16 >4.16 4.51 >4.51 4.29 >4.29 Left ventricular structure 334 334 194 192 527 527 Internal diameter (cm) 5.01±0.40 5.05±0.41 4.52±0.49 4.57±0.51 4.84±0.49 4.87±0.50 Interventricular septum (cm) 0.98±0.14 0.97±0.15 0.88±0.35 0.90±0.38 0.93±0.24 0.95±0.26 Posterior wall (cm) 0.89±0.12 0.90±0.12 0.80±0.28 0.81±0.13 0.86±0.19 0.86±0.13 Relative wall thickness 0.37±0.06 0.37±0.06 0.40±0.43 0.41±0.47 0.38±0.27 0.38±0.29 Mass index (g/m2) 89.4±21.1 92.6±20.5 67.9±18.4 69.1±18.3 81.5±22.5 84.1±22.7 Ejection fraction (%) 68.3±7.1 68.7±7.4 63.3±4.9 64.8±6.0† 66.5±6.8 67.2±7.1 Doppler data A peak (cm/s) 61.1±12.6 64.0±13.6† 58.6±13.6 63.8±15.4 60.1±13.2 64.2±14.1† E peak (cm/s) 75.6±14.3 75.5±13.1 73.2±14.9 71.2±15.6 75.2±14.6 73.4±14.0 E/A ratio 1.34±0.41 1.27±0.37 1.36±0.48 1.23±0.35 1.36±0.43 1.25±0.36† e’ peak (cm/s) 12.1±2.7 11.6±2.6† 12.0±2.3 10.7±2.4† 12.1±2.6 11.2±2.6‡ a’ peak (cm/s) 9.77±1.87 9.95±1.86 9.36±1.72 9.59±1.74 9.60±1.85 9.87±1.82 e’/a’ ratio 1.38±0.56 1.28±0.49† 1.45±0.49 1.25±0.45 1.41±0.54 1.26±0.47‡ E/e’ ratio 6.60±1.47 6.94±1.50† 6.30±1.45 7.16±1.94† 6.50±1.49 7.02±1.69‡ Table 2. Sex- and age-standardized echocardiographic measurements by median of dp-ucMGP distribution. https://doi.org/10.1371/journal.pone.0193967.t002 1:500). Paraffin sections were stained with hematoxylin and eosin for routine histological analysis and Sirius Red for detection of fibrillar collagen. Staining highlighted uncarboxylated MGP (ucMGP), carboxylated MGP (cMGP), unphosphorylated MGP (dpMGP) and phosphorylated MGP (pMGP). pMGP and cMGP include the secreted and active MGP conformation. Statistical analyses of the histological data were performed with Graphpad Prism 7.0b (La Jolla, CA), using the non-parametric Mann-Whitney test. https://doi.org/10.1371/journal.pone.0193967.t003 Population studies The geometric mean concentration of dp-ucMGP was 3.94 μg/L (interquartile range, 2.89– 5.76 μg/L; 5-95th percentile interval, 1.50–8.08 μg/L) in Flemish and 4.20 μg/L (interquartile Table 3. Multivariable-adjusted association of left ventricular diastolic function with dp-ucMGP. Table 3. Multivariable-adjusted association of left ventricular diastolic function with dp-ucMGP. Doppler Indices FLEMENGHO (n = 668) SKIPOGH (n = 386) ALL (n = 1054) Estimate (95%CI) P Value Estimate (95%CI) P Value Estimate (95%CI) P Value A peak, cm/s 1.26 (0.10 to 2.42) 0.034 1.07 (–0.88 to 3.03) 0.279 1.28 (0.26 to 2.31) 0.014 E peak, cm/s 0.87 (–0.51 to 2.25) 0.217 –0.09 (–2.23 to 2.05) 0.933 0.62 (–0.55 to 1.80) 0.298 e’ peak, cm/s –0.21 (–0.41 to –0.0003) 0.049 –0.43 (–0.68 to –0.18) 0.001 –0.31 (–0.47 to –0.15) <0.001 e’/a’ ratio –0.029 (–0.068 to 0.010) 0.146 –0.044 (–0.092 to 0.005) 0.079 –0.034 (–0.065 to –0.004) 0.002 E/e’ ratio 0.26 (0.12 to 0.40) <0.001 0.33 (0.10 to 0.55) 0.004 0.31 (0.18 to 0.43) 0.026 Values are regression coefficients (95% confidence interval) Estimates accounted for family cluster were adjusted sex age body mass index mean arterial pressure PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 6 / 20 Diastolic LV function and matrix Gla protein Fig 2. Odds ratios relating diastolic LV dysfunction to dp-ucMGP. For definition of the age-specific criteria of impaired relaxation and increased filling pressure, see Methods. The analyses accounted for cohort and family cluster and were adjusted for sex, age, body mass index, mean arterial pressure, pulse pressure, heart rate, total cholesterol, plasma glucose, left ventricular mass index, alcohol intake and antihypertensive drug treatment. Horizontal bars denote the 95% confidence interval. For each entry, the number of people with diastolic LV dysfunction vs. normal function are given. https://doi org/10 1371/journal pone 0193967 g002 Fig 2. Odds ratios relating diastolic LV dysfunction to dp-ucMGP. For definition of the age-specific criteria of impaired relaxation and increased filling pressure, see Methods. The analyses accounted for cohort and family cluster and were adjusted for sex, age, body mass index, mean arterial pressure, pulse pressure, heart rate, total cholesterol, plasma glucose, left ventricular mass index, alcohol intake and antihypertensive drug treatment. Horizontal bars denote the 95% confidence interval. For each entry, the number of people with diastolic LV dysfunction vs. normal function are given. https://doi.org/10.1371/journal.pone.0193967.g002 https://doi.org/10.1371/journal.pone.0193967.g002 range, 3.09–5.88 μg/L; 5-95th percentile interval, 1.65–8.74 μg/L) in Swiss (Fig 1). PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Population studies Table 1 lists the characteristics of participants by the cohort-specific median of the dp-ucMGP distribution. Age, body mass index, systolic and diastolic blood pressure, heart rate, the prevalence of hyper- tension and use of antihypertensive drugs, total cholesterol, plasma glucose, and serum creati- nine all increased (P0.023) with higher dp-ucMGP category, whereas the opposite was the case for smoking (P<0.001). The distribution of these cardiovascular risk factors across cate- gories of dp-ucMGP was consistent in Flemish, Swiss and all participants combined (Table 1). Table 2 shows the sex- and age-standardized echocardiographic measurements by the cohort-specific median of the dp-ucMGP distributions. In Flemish and in all participants com- bined, the A peak velocity and the E/e’ ratio increased (P0.026) with higher dp-ucMGP cate- gory, whereas the opposite was the case (P0.012) for the e’ peak velocities and the ratios E/A and e’/a’. In Swiss, trends were similar, but significance was only reached for the decrease in e’ (P = 0.002) and the increase in E/e’ (P = 0.007) with higher dp-ucMGP category (Table 2). Single-model tests demonstrated that the multivariable-adjusted regression lines relating the echocardiographic indexes to dp-ucMGP were all coincident in Flemish and Swiss. In Flemish, Swiss and all participants combined, each doubling of dp-ucMGP was associated PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 7 / 20 For each doubling of dp-ucMGP, the transmitral A peak increased by 1.26 cm/s (P = 0.034) in Flemish and by 1.28 cm/s (P = 0.014) in both cohorts combined with a similar trend in Swiss (1.07 cm/s; P = 0.279); the e’/a’ ratio decreased by 0.034 (P = 0.002) in both cohorts combined with a similar trend in Flemish (–0.029; P = 0.146) and in Swiss (–0.044; P = 0.079). The E/A ratio (P0.211) and the a’ peak (P0.582) were not associated with dp-ucMGP in any cohort. l bl l l d l d h d with increases in E/e’ by 0.26 (P<0.001), 0.33 (P = 0.004) and 0.31 (P = 0.026), respectively (Table 3). Considering the e’ denominator of the E/e’ ratio, the association sizes were –0.21 cm/s (P = 0.049) in Flemish, –0.43 cm/s (P = 0.001) in Swiss and –0.31 cm/s (P<0.001) in all participants. The E numerator was not associated with dp-ucMGP in any cohort (P0.217). For each doubling of dp-ucMGP, the transmitral A peak increased by 1.26 cm/s (P = 0.034) in Flemish and by 1.28 cm/s (P = 0.014) in both cohorts combined with a similar trend in Swiss (1.07 cm/s; P = 0.279); the e’/a’ ratio decreased by 0.034 (P = 0.002) in both cohorts combined with a similar trend in Flemish (–0.029; P = 0.146) and in Swiss (–0.044; P = 0.079). The E/A ratio (P0.211) and the a’ peak (P0.582) were not associated with dp-ucMGP in any cohort. In multivariable categorical analyses, impaired LV relaxation was not associated with dp- ucMGP in any cohort (0.96 odds ratio 1.27; P0.455; Fig 2). Increased LV filling pressure was associated with dp-ucMGP in all cohorts with odds ratios for a doubling of dp-ucMGP amounting to 1.99 (P = 0.017) in Flemish, 3.29 (P<0.001) in Swiss and 2.36 (P<0.001) in all participants (Fig 2). For impaired relaxation combined with increased LV filling pressure, the odds ratios were 1.40; P = 0.061) in Flemish, 1.72 (P = 0.032) in Swiss and 1.54 (P = 0.004) in all participants (Fig 2). y In multivariable categorical analyses, impaired LV relaxation was not associated with dp- ucMGP in any cohort (0.96 odds ratio 1.27; P0.455; Fig 2). Increased LV filling pressure was associated with dp-ucMGP in all cohorts with odds ratios for a doubling of dp-ucMGP amounting to 1.99 (P = 0.017) in Flemish, 3.29 (P<0.001) in Swiss and 2.36 (P<0.001) in all participants (Fig 2). For impaired relaxation combined with increased LV filling pressure, the odds ratios were 1.40; P = 0.061) in Flemish, 1.72 (P = 0.032) in Swiss and 1.54 (P = 0.004) in all participants (Fig 2). Diastolic LV function and matrix Gla protein Table 4. Characteristics of patients with cardiomyopathy and donors. Subjects Age (years) Sex Device LVEF (%) LVDD (mm) IVS (mm) LVPW (mm) ucMGP area (μm2/ field) pMGP Number Mean SEM Mean SEM DCM 66 Man CRT-D 30 68.0 9.9 4.1 84.00 25.32 37.33 3.28 DCM 65 Man ICD 15 58.1 8.3 10.2 115.00 40.09 32.00 1.00 DCM 61 Man CRT-D 22 54.7 13.6 10.7 91.50 24.78 36.67 1.76 DCM 20 Man Heartmate II 15 70.2 8.4 7.9 129.75 34.83 42.00 2.08 ICM 63 Man CRT-D 10 78.1 7.5 11.5 36.25 6.76 16.67 0.88 ICM 61 Man CRT-D 34 51.4 11.2 11.4 24.75 12.64 30.67 2.73 ICM 59 Man Heartware, CRT-D 15 76.5 4.4 7.9 252.25 73.42 15.00 2.00 ICM 61 Man Heartmate II 10 66.5 7.4 8.0 901.75 161.59 40.33 2.31 Donor 71 Man None . . . . . . . . . . . . 11.50 2.25 2.67 0.88 Donor 63 Man None . . . . . . . . . . . . 5.50 2.36 8.67 1.33 Donor 82 Woman None . . . . . . . . . . . . 16.50 4.57 10.00 1.73 Donor 75 Man None . . . . . . . . . . . . 1.00 1.00 5.00 1.15 Table 4. Characteristics of patients with cardiomyopathy and donors. Abbreviations: DCM, dilated cardiomyopathy; ICM, ischaemic cardiomyopathy; CRT-D, cardiac resynchronisation therapy defibrillator; ICD, implantable cardioverter defibrillator; LVEF, left ventricular ejection fraction; LVDD, left ventricular end-diastolic diameter; IVS, Thickness of the interventricular septum at end- diastole; LVPW, thickness of the left ventricular posterior wall at end-diastole; ucMGP, uncarboxylated matrix Gla protein; pMGP, phosphorylated matrix Gla protein. Heartmate II (Thoratec Corporation, California, USA) is a left ventricular assist device implanted as bridge to heart transplantation. An ellipsis indicates data not available https://doi.org/10.1371/journal.pone.0193967.t004 https://doi.org/10.1371/journal.pone.0193967.t004 with increases in E/e’ by 0.26 (P<0.001), 0.33 (P = 0.004) and 0.31 (P = 0.026), respectively (Table 3). Considering the e’ denominator of the E/e’ ratio, the association sizes were –0.21 cm/s (P = 0.049) in Flemish, –0.43 cm/s (P = 0.001) in Swiss and –0.31 cm/s (P<0.001) in all participants. The E numerator was not associated with dp-ucMGP in any cohort (P0.217). Abbreviations: DCM, dilated cardiomyopathy; ICM, ischaemic cardiomyopathy; CRT-D, cardiac resynchronisation therapy defibrillator; ICD, implantable cardioverter defibrillator; LVEF, left ventricular ejection fraction; LVDD, left ventricular end-diastolic diameter; IVS, Thickness of the interventricular septum at end- diastole; LVPW, thickness of the left ventricular posterior wall at end-diastole; ucMGP, uncarboxylated matrix Gla protein; pMGP, phosphorylated matrix Gla protein. Heartmate II (Thoratec Corporation, California, USA) is a left ventricular assist device implanted as bridge to heart transplantation. An ellipsis indicates data not available diomyopathy; ICM, ischaemic cardiomyopathy; CRT-D, cardiac resynchronisation therapy defibrillator; ICD, implantable https://doi.org/10.1371/journal.pone.0193967.t004 Histological studies All patients with heart disease and donors were white. Age at transplantation ranged from 20 to 66 years and from 59 to 63 years in patients with dilated or ischemic cardiomyopathy, respectively (Table 4). Age at the death among the donors ranged from 63 to 82 years. The causes of death in donors were hemorrhagic stroke in three and ischemic stroke in one. Age was the main criterion why donor hearts were not implanted. All patients with cardiomyopa- thy wore a cardiac resynchronization, defibrillator or left ventricular assist device before being transplanted (Table 4). 8 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Diastolic LV function and matrix Gla protein Fig 3. Immunofluorescent localization of MGP species in consecutive sections of the left ventricular myocardium. Rows are labelled by the MGP conformation (red) stained for: from top to bottom uncarboxylated MGP (ucMGP), carboxylated MGP (cMGP), unphosphorylated MGP (dpMGP) and phosphorylated MGP (pMGP). pMGP and cMGP include the secreted and active MGP conformation. Columns refer to exemplary tissue samples of male patients aged 61–63 years: from left to right unused donor heart (.h [HD]), dilated cardiomyopathy (.d [DCM] and ischemic cardiomyopathy representative section (.i [ICM]) and fibrotic area of the same ischemic heart (.f). WGA (green) and TO-PRO3 (blue) stain cell membranes and nuclei, respectively. The dilution of the conformation specific antibodies was 1:100 for ucMGP, dpMGP, pMGP and the inset in panel B.co and 1:250 for cMGP. The scale bar corresponds to 25 μm. The findings are described in a qualitative way in the results section. The quantitative analysis, comparing abundance Fig 3. Immunofluorescent localization of MGP species in consecutive sections of the left ventricular myocardium. Rows are labelled by the MGP conformation (red) stained for: from top to bottom uncarboxylated MGP (ucMGP), carboxylated MGP (cMGP), unphosphorylated MGP (dpMGP) and phosphorylated MGP (pMGP). pMGP and cMGP include the secreted and active MGP conformation. Columns refer to exemplary tissue samples of mal patients aged 61–63 years: from left to right unused donor heart (.h [HD]), dilated cardiomyopathy (.d [DCM] and ischemic cardiomyopathy representative section (.i [ICM]) and fibrotic area of the same ischemic heart (.f). WGA (green) and TO-PRO3 (blue) stain cell membranes and nuclei, respectively. The dilution of the conformation specific antibodies was 1:100 for ucMGP, dpMGP, pMGP and the inset in panel B.co and 1:250 for cMGP The scale bar corresponds to 25 μm. Diastolic LV function and matrix Gla protein of ucMGP and pMGP among HD, DCM and ICM appear in panels E and F, respectively. n refers to the number of tissue samples included in the quantitative analysis. Labels: asterisks indicate cardiomyocytes; arrows point to interstitial cells with perinuclear MGP deposition; L lumen of muscularized microvessels. https://doi.org/10.1371/journal.pone.0193967.g003 https://doi.org/10.1371/journal.pone.0193967.g003 The active MGP moieties, cMGP (Fig 3B) and pMGP (Fig 3D), were predominantly distrib- uted in the media and intima of muscular left ventricular microvessels in healthy and diseased hearts. Inactive ucMGP was abundant in fibrotic areas of diseased hearts, around the nuclei of interstitial cells and in the perivascular matrix (Fig 3A and Fig 4). ucMGP was more abundant in DCM and ICM than HD hearts (mean±SEM, 105.1±10.6 and 303.8±206.1 vs. 8.6±3.4 μm2/ field; P = 0.029; Fig 3E). In ICM hearts, ucMGP was particularly present in fibrotic areas and, depending on the degree of fibrosis, showed large variability between patients (Table 4). ucMGP staining was absent in cardiomyocytes (Fig 3A). Furthermore, DCM and ICM myo- cardium showed more pMGP positive capillaries and interstitial cells than HD hearts (37.0 ±2.0 and 25.7±6.0 vs. 6.6±1.7; P = 0.029; Fig 3F and Table 4). Finally, staining for inactive dpMGP was almost absent in the vessel wall and in fibrotic areas, but was abundant in cardio- myocytes of all hearts and co-localized with active cMGP (Fig 3C). There were no differences in cMGP (Figs 5 and 6) and pMGP (Figs 7 and 8) staining between an older (61 years) and young (20 years) patient with DCM. Histological studies The findings are described in a qualitative way in the results section. The quantitative analysis, comparing abundanc ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 9 / Fig 3. Immunofluorescent localization of MGP species in consecutive sections of the left ventricular myocardium. Rows are labelled by the MGP conformation (red) stained for: from top to bottom uncarboxylated MGP (ucMGP), carboxylated MGP (cMGP), unphosphorylated MGP (dpMGP) and phosphorylated MGP (pMGP). pMGP and cMGP include the secreted and active MGP conformation. Columns refer to exemplary tissue samples of male patients aged 61–63 years: from left to right unused donor heart (.h [HD]), dilated cardiomyopathy (.d [DCM] and ischemic cardiomyopathy representative section (.i [ICM]) and fibrotic area of the same ischemic heart (.f). WGA (green) and TO-PRO3 (blue) stain cell membranes and nuclei, respectively. The dilution of the conformation specific antibodies was 1:100 for ucMGP, dpMGP, pMGP and the inset in panel B.co and 1:250 for cMGP. The scale bar corresponds to 25 μm. The findings are described in a qualitative way in the results section. The quantitative analysis, comparing abundance PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 9 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Discussion Paulus and Tscho¨pe proposed a novel paradigm that shifts emphasis in diastolic LV dysfunc- tion from LV afterload to inflammation of the coronary microcirculation [2]. This paradigm [2] along with the premise that MGP is promoting the integrity of the microcirculation [11] and might be a local inhibitor of soft tissue calcification [12,13] provided the rationale for the hypothesis that echocardiographically assessed diastolic LV function might be inversely associ- ated with circulating inactive dp-ucMGP. We confirmed our hypothesis in two population cohorts, while subsequent tissue staining studies revealed the localization of the active and inactive MGP moieties in the myocardium, thereby supporting a role of active MGP in main- taining a healthy myocardium. In diseased compared with normal hearts, uncarboxylated MGP was more prevalent in the perivascular matrix and interstitium and phosphorylated MGP in and around capillaries and interstitial cells. Cardiomyocytes showed no staining for uncarboxylated MGP. Unphosphory- lated MGP was almost absent in vessel walls and fibrotic areas, but was present in cardiomyo- cytes of all hearts and co-localized with carboxylated MGP. The ischemia-induced fibrotic remodeling in ICM patients seemed to be accompanied by an even higher deposition of ucMGP (Fig 3A and 3F). Fig 9 illustrates the hypothetical sequence of events possibly explain- ing our histological observations. Once secreted into the extracellular matrix, carboxylated and phosphorylated MGP protects against calcium deposition [22,23] and inhibits trans-differenti- ation of vascular smooth muscle cells [24] and signaling via the bone morphogenetic protein (BMP) pathway [25–27]. Taken together with the literature [22–27], our current findings sup- port the idea that activated MGP is an ubiquitous locally acting agent protecting the microcir- culation and the perivascular matrix. Several mechanisms might explain how activated MGP might promote preservation of dia- stolic LV function. First, MGP binds to calcium ions as well as to hydroxyapatite crystals and may thereby inhibit crystal growth [22]. Depending on the amino-acid sequence, varying phosphorylated and γ-carboxylated sequences of human MGP inhibited either nucleation or growth, or both, of hydroxyapatite and calcium oxalate monohydrate crystals [23]. Relevant to the current study, is the strong well-documented protein-protein interaction between MGP PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 10 / 20 Diastolic LV function and matrix Gla protein Fig 4. Fluorescence, H&E, Sirius Red staining and magnified ucMGP images of panels A.d (dilated cardiomyopathy [DCM]), A.i (ischemic cardiomyopathy [ICM]) and A.f (ICM fibrotic area) taken from Fig 3. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Discussion T scale bar in A.d, A.i, A.f, H&E and Sirius Red images corresponds to 50 μm. Squares indicate areas that are magnifie (scale bar 10 μm). X and Y label two magnified areas of DCM panel A.d. L indicates vessel lumen. White arrows poin to ucMGP staining (red). https://doi.org/10.1371/journal.pone.0193967.g004 Fig 4. Fluorescence, H&E, Sirius Red staining and magnified ucMGP images of panels A.d (dilated cardiomyopathy [DCM]), A.i (ischemic cardiomyopathy [ICM]) and A.f (ICM fibrotic area) taken from Fig 3. The scale bar in A.d, A.i, A.f, H&E and Sirius Red images corresponds to 50 μm. Squares indicate areas that are magnified (scale bar 10 μm). X and Y label two magnified areas of DCM panel A.d. L indicates vessel lumen. White arrows point to ucMGP staining (red). https://doi.org/10.1371/journal.pone.0193967.g004 11 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Diastolic LV function and matrix Gla protein Fig 5. cMGP immunofluorescent localization in left ventricular myocardium in an older patient aged 61 years with dilated cardiomyopathy. H&E staining (A), Sirius Red staining (B), cMGP (red [D]), CD31 (green [F]), α- smooth muscle actin (αSMA, green [G]). Panels are multiple stains highlighting: (i) cMGP (red), cell membranes (WGA, green) and nuclei (TO-PRO3, blue) in panel C; (ii) cMGP (red) and endothelium (CD31, green) in panel H; (iii) cMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel I; and (iv) a negative control without cMGP and CD31 primary antibodies in panel E. Two sets of magnifications visualize the microvascular endothelium. Images J, K and L correspond to the square in panel H, while images M, N and O are taken from another area of panel H. cMGP is abundant in the small muscularized vessels (panels D, E, F, G, H and I), in microvascular endothelium (panels J, K, L, M, N and O) and in (panel C). The scale bar represents 100 μm in panels A, B and C; 50 μm in panels D, E, F, G, H and I; and 10 μm in panels J, K, L, M, N and O. L indicates vascular lumen. Fig 5. cMGP immunofluorescent localization in left ventricular myocardium in an older patient aged 61 years with dilated cardiomyopathy. H&E staining (A), Sirius Red staining (B), cMGP (red [D]), CD31 (green [F]), α- smooth muscle actin (αSMA, green [G]). PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Discussion Panels are multiple stains highlighting: (i) cMGP (red), cell membranes (WGA, green) and nuclei (TO-PRO3, blue) in panel C; (ii) cMGP (red) and endothelium (CD31, green) in panel H; (iii) cMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel I; and (iv) a negative control without cMGP and CD31 primary antibodies in panel E. Two sets of magnifications visualize the microvascular endothelium. Images J, K and L correspond to the square in panel H, while images M, N and O are taken from another area of panel H. cMGP is abundant in the small muscularized vessels (panels D, E, F, G, H and I), in microvascular endothelium (panels J, K, L, M, N and O) and in (panel C). The scale bar represents 100 μm in panels A, B and C; 50 μm in panels D, E, F, G, H and I; and 10 μm in panels J, K, L, M, N and O. L indicates vascular lumen. https://doi.org/10.1371/journal.pone.0193967.g005 https://doi.org/10.1371/journal.pone.0193967.g005 and BMP, including BMP-2 [25,26] and BMP-4 [27], whereby bound MGP reduces BMP sig- naling. In the heart, BMP pathways play a pivotal role in the embryogenesis of the LV chamber [28], the differentiation of cardiac progenitor cells into functional cardiomyocytes [29], main- tenance of the balance between LV growth and apoptosis [30], initiating fibrosis [30], and Ca2 + channel remodeling [31]. In line with the paradigm that the microcirculation is involved in PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 12 / 20 Diastolic LV function and matrix Gla protein Fig 6. cMGP immunofluorescent localization in left ventricular myocardium in younger patient aged 20 years with dilated cardiomyopathy. Sirius Red staining (A), α-smooth muscle actin (αSMA, green [B]), CD31 (green [C]), cMGP (red [E]). Panels are triple stains highlighting: (i) cMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel F; (ii) cMGP (red), endothelium (CD31, green) and nuclei (TO-PRO3, blue) in panel G; and (iii) a negative control without cMGP and CD31 primary antibodies in panel D. The squares in panel G identify a muscularized vessel magnified in panel K and microvascular endothelium magnified in panels L, M and N. White arrows point to capillaries. cMGP deposition is comparable to the older patient in Fig 5. The scale bar represents 50 μm in panels A, B, C, D, E, F and G and 10 μm in panels K, L, M and N. L indicates vascular lumen. Fig 6. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Discussion cMGP immunofluorescent localization in left ventricular myocardium in younger patient aged 20 years with dilated cardiomyopathy. Sirius Red staining (A), α-smooth muscle actin (αSMA, green [B]), CD31 (green [C]), cMGP (red [E]). Panels are triple stains highlighting: (i) cMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel F; (ii) cMGP (red), endothelium (CD31, green) and nuclei (TO-PRO3, blue) in panel G; and (iii) a negative control without cMGP and CD31 primary antibodies in panel D. The squares in panel G identify a muscularized vessel magnified in panel K and microvascular endothelium magnified in panels L, M and N. White arrows point to capillaries. cMGP deposition is comparable to the older patient in Fig 5. The scale bar represents 50 μm in panels A, B, C, D, E, F and G and 10 μm in panels K, L, M and N. L indicates vascular lumen. https://doi.org/10.1371/journal.pone.0193967.g006 the pathogenesis of diastolic LV function [2], BMP signaling also promotes endothelial apo- ptosis [32], increases endothelial permeability [32], and is involved in the expression of vascu- lar endothelial growth factor and angiogenesis [27]. Circulating dp-ucMGP is a biomarker reflecting vitamin K deficiency [33]. Indeed, among 60 middle-aged healthy volunteers randomized in a placebo-controlled double-blind trial, plasma dp-ucMGP dropped dose-dependently by 31% and 46% in response to daily supplem- entation for 12 weeks with 180 μg and 360 μg of menaquinone-7 (vitamin K2) [33]. Selectively re-introducing MGP expression in the liver of MGP knockout mice, produced circulating MGP levels 6- to 10-fold higher than in wild type animals [34]. The MGP originating from the transgene conserved its biological activity in vitro, but did not inhibit arterial calcification [34], confirming that MGP is a locally acting paracrine protein. We did not measure circulat- ing levels of vitamin K, which is rarely done in clinical practice, because of the complexity of the assay and the lack of a high-throughput method [35] and because plasma levels only reflect dietary intake (vitamin K1; phylloquinone) and production by the gut microflora (vitamin K2; PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 13 / 20 Diastolic LV function and matrix Gla protein Fig 7. pMGP immunofluorescent localization in left ventricular myocardium in an older patient aged 61 years with dilated cardiomyopathy. H&E staining appears in panel A, Sirius Red staining in panel B, and staining for pMGP (red) and WGA (green) and TO-PRO3 (blue) in panel C. https://doi.org/10.1371/journal.pone.0193967.g007 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Discussion Magnifications of staining of capillaries appear in panel D for pMGP, in panel E for CD31 and in panel F for pMGP, CD31 and TO-PRO3. The scale bar represents 100 μm in panels A, B and C and 10 μm in panels D, E and F. Diastolic LV function and matrix Gla protein Fig 7. pMGP immunofluorescent localization in left ventricular myocardium in an in panel A, Sirius Red staining in panel B, and staining for pMGP (red) and WGA (gree in panel D for pMGP, in panel E for CD31 and in panel F for pMGP, CD31 and TO-PR and F Fig 7. pMGP immunofluorescent localization in left ventricular myocardium in an older patient aged 61 years with dilated cardiomyopathy. H&E staining appears in panel A, Sirius Red staining in panel B, and staining for pMGP (red) and WGA (green) and TO-PRO3 (blue) in panel C. Magnifications of staining of capillaries appear in panel D for pMGP, in panel E for CD31 and in panel F for pMGP, CD31 and TO-PRO3. The scale bar represents 100 μm in panels A, B and C and 10 μm in panels D, E and F. Fig 7. pMGP immunofluorescent localization in left ventricular myocardium in an older patient aged 61 years with dilated cardiomyopathy. H&E staining appears in panel A, Sirius Red staining in panel B, and staining for pMGP (red) and WGA (green) and TO-PRO3 (blue) in panel C. Magnifications of staining of capillaries appear in panel D for pMGP, in panel E for CD31 and in panel F for pMGP, CD31 and TO-PRO3. The scale bar represents 100 μm in panels A, B and C and 10 μm in panels D, E and F. https://doi.org/10.1371/journal.pone.0193967.g007 14 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Fig 8. pMGP immunofluorescent localization in the left ventricular myocardium in younger patient aged 20 years with dilated cardiomyopathy. Staining for pMGP (red) appears in panels A and G, for CD31 (green) in panels B and H, for pMGP (red) and TO-PRO3 (blue) in panel C, and for α-smooth muscle actin (αSMA, green) in panel E. Triple stains highlight: (i) pMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel F; (ii) pMGP (red), endothelium (CD31, green) and nuclei (TO-PRO3, blue) in panel I; and (iii) a negative control without pMGP and CD31 primary antibodies in panel D. Discussion Hypothetical pathways aligning with the histological findings. Cardiomyocytes, interstitial, endothelial and vascular smooth muscle cells of the human heart express MGP. Step 1: After translation in the endoplasmatic reticulum (ER), vitamin K activates MGP by stimulating γ-carboxylation. Step 2: Carboxylated desphospho- MGP (dp-cMGP) can sequesters intracellular calcium, thereby providing protection against injury caused by calcium deposition. Step 3: A Golgi-associated casein kinase phosphorylates the serine residues of dp-cMGP to p-cMGP, thereby facilitating secretion. Step 4: p-cMGP is secreted into the extracellular matrix or the circulation to inhibit soft tissue calcification, vascular smooth muscle cell (VSMC) trans-differentiation into osteochondrogenic progenitor cells (OPC) and signaling via the BMP (bone morphogenetic protein) pathway. Step 5: Inactive desphospho-uncarboxylated MGP (dp-ucMGP), a biomarker reflecting poor vitamin K status, escapes from cells into the blood stream, but does not inhibit calcification. https://doi org/10 1371/journal pone 0193967 g009 Fig 9. Hypothetical pathways aligning with the histological findings. Cardiomyocytes, interstitial, endothelial and vascular smooth muscle cells of the human heart express MGP. Step 1: After translation in the endoplasmatic reticulum (ER), vitamin K activates MGP by stimulating γ-carboxylation. Step 2: Carboxylated desphospho- MGP (dp-cMGP) can sequesters intracellular calcium, thereby providing protection against injury caused by calcium deposition. Step 3: A Golgi-associated casein kinase phosphorylates the serine residues of dp-cMGP to p-cMGP, thereby facilitating secretion. Step 4: p-cMGP is secreted into the extracellular matrix or the circulation to inhibit soft tissue calcification, vascular smooth muscle cell (VSMC) trans-differentiation into osteochondrogenic progenitor cells (OPC) and signaling via the BMP (bone morphogenetic protein) pathway. Step 5: Inactive desphospho-uncarboxylated MGP (dp-ucMGP), a biomarker reflecting poor vitamin K status, escapes from cells into the blood stream, but does not inhibit calcification. https://doi.org/10.1371/journal.pone.0193967.g009 https://doi.org/10.1371/journal.pone.0193967.g009 menaquinones) without giving any indication of functionality, i.e. the amount of MGP under- going carboxylation [33]. The clinical implications of our current findings rest on the prognostic significance of the E/e’ ratio [36–39]. Among 816 hypertensive patients with echocardiographic data randomized in the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT), 56 cardiac events occurred over 4.2 years. The E/e’ ratio was the strongest predictor of a first cardiac event. Following adjust- ment for covariables, a one unit rise in the E/e’ ratio was associated with 17% increment in risk of a cardiac event (95% confidence interval, 5% to 29%; P = 0.003) [38]. Discussion pMGP is abundant in the endothelium and vessel wall of muscularized vessels (panels A, B, C, D, E and F) and in capillaries (panels G, H and I). pMGP deposition is comparable to the old patient in Fig 7. White arrows point to endothelial layer. The scale bar represents 10 μm. L indicates the vessel lumen. https://doi.org/10.1371/journal.pone.0193967.g008 Diastolic LV function and matrix Gla protein Diastolic LV function and matrix Gla protein Fig 8. pMGP immunofluorescent localization in the left ventricular myocardium in younger patient aged 20 years with dilated cardiomyopathy. Staining for pMGP (red) appears in panels A and G, for CD31 (green) in panels B and H, for pMGP (red) and TO-PRO3 (blue) in panel C, and for α-smooth muscle actin (αSMA, green) in panel E. Triple stains highlight: (i) pMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel F; (ii) pMGP (red), endothelium (CD31, green) and nuclei (TO-PRO3, blue) in panel I; and (iii) a negative control without pMGP and CD31 primary antibodies in panel D. pMGP is abundant in the endothelium and vessel wall of muscularized vessels (panels A, B, C, D, E and F) and in capillaries (panels G, H and I). pMGP deposition is comparable to the old patient in Fig 7. White arrows point to endothelial layer. The scale bar represents 10 μm. L indicates the vessel lumen. Fig 8. pMGP immunofluorescent localization in the left ventricular myocardium in younger patient aged 20 years with dilated cardiomyopathy. Staining for pMGP (red) appears in panels A and G, for CD31 (green) in panels B and H, for pMGP (red) and TO-PRO3 (blue) in panel C, and for α-smooth muscle actin (αSMA, green) in panel E. Triple stains highlight: (i) pMGP (red), αSMA (green) and nuclei (TO-PRO3, blue) in panel F; (ii) pMGP (red), endothelium (CD31, green) and nuclei (TO-PRO3, blue) in panel I; and (iii) a negative control without pMGP and CD31 primary antibodies in panel D. pMGP is abundant in the endothelium and vessel wall of muscularized vessels (panels A, B, C, D, E and F) and in capillaries (panels G, H and I). pMGP deposition is comparable to the old patient in Fig 7. White arrows point to endothelial layer. The scale bar represents 10 μm. L indicates the vessel lumen. https://doi.org/10.1371/journal.pone.0193967.g008 https://doi.org/10.1371/journal.pone.0193967.g008 15 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Diastolic LV function and matrix Gla protein Fig 9. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 Acknowledgments The authors gratefully acknowledge the contribution of the staff and participants of the FLE- MENGHO and SKIPOGH studies and the laboratory technicians who contributed to the his- tological studies. Discussion Similarly, among 406 patients with type-2 diabetes mellitus, the E/e’ ratio was an independent predictor of cardio- vascular events occurring over 5.6 years, including 12 myocardial infarctions and 7 strokes. In analyses adjusted for sex and age, a one unit rise in the E/e’ ratio was associated with 8% incre- ment in risk (1% to 16%; P = 0.021), whereas the association with global LV strain was not sig- nificant (P = 0.913) [39]. Along similar lines, e’ predicted mortality in a study including 174 hypertensive patients and 78 age-matched controls [40]. Our current findings focusing on dia- stolic LV function, are in line with our previous report showing that high dp-ucMGP predicted adverse health outcomes, including total and cardiovascular mortality in 2318 FLEMENGHO participants followed-up for a median of 14.1 years [9]. Strong points of our current study are the consistency of the association of the E/e’ ratio with dp-ucMGP in two population cohorts and the histological evidence supporting the epide- miological observations. However, our current study must also be interpreted within the con- text of its potential limitations. First, our population studies had a cross-sectional design, which precludes direct causal inference. Second, we did not measure circulating dp-ucMGP in the PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 16 / 20 Diastolic LV function and matrix Gla protein patients enrolled in the histological studies and we cannot conclude to what extent the patients with end-stage DCM or ICM are representative for early-stage diastolic LV dysfunction as observed in the general population. Third, the sample size of histological studies was relatively small, but of the same order of magnitude as in other studies of the same nature [41,42]. Fourth, the location of MGP species was not confirmed by Western blots, because the antibodies avail- able to us had never before been validated for use in such studies. Finally, for the histological studies, we used conformation-specific MGP antibodies, but we did not apply measurement of protein expression or mRNA. Conclusions In the general population, E/e’ increased with higher dp-ucMGP, a marker of vitamin K defi- ciency. The combined epidemiological and histological findings suggest that poor vitamin K status, as exemplified by higher plasma dp-ucMGP, represents a risk factor for diastolic LV dysfunction. Whether our current findings open new avenues to the prevention or treatment of diastolic LV dysfunction or its progression to HF remains to be established in randomized clinical trials of vitamin-K supplementation. References 1. Bui AL, Horwich TB, Fonarow GC. Epidemiology and risk profile of heart failure. Nat Rev Cardiol. 2011; 8: 30–41. https://doi.org/10.1038/nrcardio.2010.165 PMID: 21060326 2. Paulus WJ, Tscho¨pe C. A novel paradigm for heart failure with preserved ejection fraction. J Am Coll Cardiol. 2013; 62: 263–71. https://doi.org/10.1016/j.jacc.2013.02.092 PMID: 23684677 3. Owan TE, Hodge DO, Herges RM, Jacobsen SJ, Roger VL, Redfield MM. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med. 2006; 355: 251–9. https://doi. org/10.1056/NEJMoa052256 PMID: 16855265 4. Kuznetsova T, Herbots L, Lo´pez B, Jin Y, Richart T, Thijs L, et al. Prevalence of left ventricular diastolic dysfunction in a general population. Circ Heart Fail. 2009; 2: 105–12. https://doi.org/10.1161/ CIRCHEARTFAILURE.108.822627 PMID: 19808325 5. Kloch-Badelek M, Kuznetsova T, Sakiewicz W, Tikhonoff V, Ryabikov A, Gonza´lez A, et al. Prevalence of diastolic left ventricular dysfunction in European populations based on cross-validated diagnostic thresholds. Cardiovascular Ultrasound. 2012; 10: 10. https://doi.org/10.1186/1476-7120-10-10 PMID: 22429658 6. Kuznetsova T, Thijs L, Knez J, Cauwenberghs N, Petit T, Gu YM, et al. Longitudinal changes in left ven- tricular diastolic dysfunction in a general population. Circ Cardiovasc Imaging. 2015; 8: e002882. https://doi.org/10.1161/CIRCIMAGING.114.002882 PMID: 25873723 7. Schurgers LJ, Cranenburg ECM, Vermeer C, VitaK and Cardiovascular Research Institute (CARIM) MUMTN. Matrix gla-protein: the calcification inhibitor in need of vitamin K. Thromb Haemost. 2008; 100: 593–603. PMID: 18841280 8. Pivin E, Ponte B, Pruijm M, Ackermann D, Guessous I, Ehret G, et al. Inactive matrix gla-protein is asso- ciated with arterial stiffness in an adult population-based study. Hypertension. 2015; 66: 85–92. https:// doi.org/10.1161/HYPERTENSIONAHA.115.05177 PMID: 25987667 9. Liu YP, Gu YM, Thijs L, Knapen MHJ, Salvi E, Citterio L, et al. Inactive matrix gla protein is causally related to adverse health outcomes: a Mendelian randomization study in a Flemish population. Hyper- tension. 2015; 65: 463–70. https://doi.org/10.1161/HYPERTENSIONAHA.114.04494 PMID: 25421980 10. Knapen MHJ, Braam LAJLM, Drummen NEA, Bekers O, Hoeks APG, Theuwissen E, et al. Low-dose menaquinone-7 supplementation improves vascular properties in healthy postmenopausal women. J Thromb Haemost. 2015; 113: 1135–44. 11. Wei FF, Drummen NEA, Schutte AE, Thijs L, Jacobs L, Petit T, et al. Vitamin K dependent protection of renal function in multi-ethnic population studies. EBioMed. 2016; 4: 162–9. 12. Wei FF, Drummen NEA, Thijs L, Jacobs L, Herfs M, van’t Hoofd C, et al. Vitamin-K-dependent protec- tion of the renal microvasculature: histopathological studies in normal and diseased kidneys. Pulse. 2016; 4: 85–91. Author Contributions Conceptualization: Cees Vermeer, Murielle Bochud, Jan A. Staessen. Conceptualization: Cees Vermeer, Murielle Bochud, Jan A. Staessen. Data curation: Fang-Fei Wei, Pierre Monney, Wen-Yi Yang, Menno Pruijm, Zhen-Yu Zhang, Yassine Bouatou, Qi-Fang Huang, Belen Ponte, Pierre-Yves Martin, Lutgarde Thijs, Tatiana Kuznetsova, Georg Ehret. Data curation: Fang-Fei Wei, Pierre Monney, Wen-Yi Yang, Menno Pruijm, Zhen-Yu Zhang, Yassine Bouatou, Qi-Fang Huang, Belen Ponte, Pierre-Yves Martin, Lutgarde Thijs, Tatiana Kuznetsova, Georg Ehret. Formal analysis: Fang-Fei Wei, Lutgarde Thijs, Jan A. Staessen. Funding acquisition: Murielle Bochud, Jan A. Staessen. Investigation: Fang-Fei Wei, Sander Trenson, Pierre Monney, Tatiana Kuznetsova, Georg Ehret. Methodology: Sander Trenson, Cees Vermeer, Murielle Bochud, Georg Ehret, Jan A. Staessen. Methodology: Sander Trenson, Cees Vermeer, Murielle Bochud, Georg Ehret, Jan A. Staessen. Project administration: Jan A. Staessen. gy , , , g , J Project administration: Jan A. Staessen. Project administration: Jan A. Staessen. Resources: Cees Vermeer, Murielle Bochud, Jan A. Staessen. Software: Lutgarde Thijs. Software: Lutgarde Thijs. Supervision: Karel Allegaert, Stefan Janssens, Peter Verhamme, Michel Burnier, Murielle Bochud, Jan A. Staessen. Validation: Jan A. Staessen. Validation: Jan A. Staessen. Visualization: Fang-Fei Wei, Sander Trenson. Writing – original draft: Fang-Fei Wei, Sander Trenson, Jan A. Staessen. Writing – review & editing: Fang-Fei Wei, Sander Trenson, Pierre Monney, Wen-Yi Yang, Menno Pruijm, Zhen-Yu Zhang, Yassine Bouatou, Qi-Fang Huang, Belen Ponte, Pierre- Yves Martin, Lutgarde Thijs, Tatiana Kuznetsova, Karel Allegaert, Stefan Janssens, Cees PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 17 / 20 Diastolic LV function and matrix Gla protein Vermeer, Peter Verhamme, Michel Burnier, Murielle Bochud, Georg Ehret, Jan A. Staessen. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 References https://doi.org/10.1159/000448008 PMID: 27752480 13. Wei FF, Thijs L, Zhang ZY, Jacobs L, Yang WY, Salvi E, et al. The risk of nephrolithiasis is causally related to inactive matrix Gla protein, a marker of vitamin K status: a Mendelian randomization study in a Flemish population. Nephrol Dial Transplant. 2017; https://doi.org/10.1093/ndt/gfx014 14. Bers DM. Calcium fluxes involved in control of cardiac myocyte contraction. Circ Res. 2000; 87: 275– 81. PMID: 10948060 15. Zhang ZY, Staessen JA, Thijs L, Gu Y, Liu Y, Jacobs L, et al. Left ventricular diastolic function in relation to the urinary proteome: a proof-of-concept study in a general population. Int J Cardiol. 2014; 176: 158– 65. https://doi.org/10.1016/j.ijcard.2014.07.014 PMID: 25065337 16. Fraser JD, Price PA. Lung, heart, and kidney express high levels of mRNA for the vitamin K-dependent matrix Gla protein. Implications for the possible functions of matrix Gla protein and for the tissue distribu- tion of the γ-carboxylase. J Biol Chem. 1988; 263: 11033–6. PMID: 3042764 17. World Medical Association. World Medical Association Declaration of Helsinki. Ethical principles for medical research involving human subjects. JAMA. 2013; 310: 2191–4. https://doi.org/10.1001/jama. 2013.281053 PMID: 24141714 18. Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM, Derumeaux G, et al. Current and evolving echocardiographic techniques for the quantitative evaluation of cardiac mechanics: ASE/EAE concen- sus statement on methodology and indications. Endorsed by Japanese Society of Echocardardiogra- phy. J Am Soc Echocardiogr. 2011; 24: 277–313. https://doi.org/10.1016/j.echo.2011.01.015 PMID: 21338865 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 18 / 20 Diastolic LV function and matrix Gla protein 19. Gottdiener JS, Bednarz J, Devereux R, Gardin J, Klein A, Manning WJ, et al. American Society of Echo- cardiography recommendations for use of echocardiography in clinical trials. A report from the Ameri- can Society of Echocardiography’s Guidelines and Standard Committee and the Task Force on Echocardiography in Clinical Trials. J Am Soc Echocardiogr. 2004; 17: 1086–119. 20. Pruijm MT, Wuerzner G, Glatz N, Alwan H, Ponte B, Ackermann D, et al. A new technique for simulta- neous validation of two manual nonmercury auscultatory sphygmomanometers (A&D UM-101 and Accoson Greenlight 300) based on the international protocol. Blood Press Monit. 2010; 15: 322–5. https://doi.org/10.1097/MBP.0b013e32833f56a8 PMID: 20827175 21. Cranenburg ECM, Koos R, Schurgers LJ, Magdeleyns EJ, Schoonbrood TH, Landewe RB, et al. Char- acterisation and potential diagnostic value of circulating matrix Gla protein (MGP) species. Thromb Haemost. 2010; 104: 811–22. https://doi.org/10.1160/TH09-11-0786 PMID: 20694284 22. References Schurgers LJ, Uitto J, Reutelingsperger CP. Vitamin K-dependent carboxylation of matrix GLA-protein: a crucial switch to control ectopic mineralization. Trends Mol Med. 2013; 19: 217–26. https://doi.org/10. 1016/j.molmed.2012.12.008 PMID: 23375872 23. Goiko M, Dierolf J, Gleberzon JS, Liao Y, Grohe B, Goldberg HA, et al. Peptides of matrix gla protein inhibit nucleation and growth of hydroxyapatite and calcium oxalate monohydrate crystals. PLoS One. 2013; 8: e80344. https://doi.org/10.1371/journal.pone.0080344 PMID: 24265810 24. Speer MY, Yang HY, Brabb T, Leaf E, Look A, Lin WL, et al. Smooth muscle cells give rise to osteo- chondrogenic precursors and chondrocytes in calcifying arteries. Circ Res. 2009; 104: 733–41. https:// doi.org/10.1161/CIRCRESAHA.108.183053 PMID: 19197075 25. Wallin R, Cain D, Hutson SM, Sane DC, Loeser R. Modulation of the binding of matrix Gla protein (MGP) to bone morphogenetic protein-2 (BMP-2). Thromb Haemost. 2000; 84: 1039–44. PMID: 11154111 26. Zebboudj AF, Imura M, Bostro¨m K. Matrix GLA protein, a regulatory protein for bone morphogenetic protein-2. J Biol Chem. 2002; 277: 4388–94. https://doi.org/10.1074/jbc.M109683200 PMID: 11741887 27. Yao Y, Zebboudj AF, Shao E, Perez M, Bostro¨m K. Regulation of bone morphogenetic protein-4 by matrix GLA protein in vascular endothelial cells involves activin-like kinase receptor 1. J Biol Chem. 2006; 281: 33921–30. https://doi.org/10.1074/jbc.M604239200 PMID: 16950789 28. Han P, Bloomekatz J, Ren J, Zhang R, Grinstein JD, Zhao L, et al. Coordinating cardiomyocyte interac- tions to direct ventricular chamber morphogenesis. Nature. 2016; 534: 700–4. https://doi.org/10.1038/ nature18310 PMID: 27357797 29. Cagavi E, Bartulos O, Suh CY, Sun B, Yue Z, Jiang Z, et al. Functional cardiomyocytes derived from Isl1 cardiac progenitors via BMP4 stimulation. PLoS One. 2014; 9: e110752. https://doi.org/10.1371/ journal.pone.0110752 PMID: 25522363 30. Sun B, Huo R, Sheng Y, Li Y, Xie X, Chen C, et al. Bone morphogenetic protein-4 mediates cardiac hypertrophy, apoptosis, and fibrosis in experimentally pathological cardiac hypertrophy. Hypertension. 2013; 61: 352–60. https://doi.org/10.1161/HYPERTENSIONAHA.111.00562 PMID: 23248151 31. Hu CW, Li Q, Zhang Y, Li YH, Jiang HC, Liu MY, et al. Bone morphogenetic protein-4 induces upregula- tion of Cav3.1 Ca2+ channels in HL-1 atrial myocytes. Pflu¨gers Arch—Eur J Physiol. 2014; 466: 2049– 57. 32. Wiley DM, Jin SW. Bone morphogenetic protein functions as a context-dependent angiogenic cue in vertebrates. Semin Cell Dev Biol. 2011; 22: 1012–8. https://doi.org/10.1016/j.semcdb.2011.10.005 PMID: 22008724 33. Dalmeijer GW, van der Schouw YT, Magdeleyns E, Ahmed N, Vermeer C, Beulens JW. The effect of menaquinone-7 supplementation on circulating species of matrix Gla protein. Atherosclerosis. 2012; 225: 397–402. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 References https://doi.org/10.1016/j.atherosclerosis.2012.09.019 PMID: 23062766 34. Murshed M, Schinke T, McKee MD, Karsenty G. Extracellular matrix mineralization is regulated locally; different roles of two gla-containing proteins. J Cell Biol. 2004; 165: 625–30. https://doi.org/10.1083/ jcb.200402046 PMID: 15184399 35. Riphagen IJ, van der Molen JC, van Faassen M, Navis G, de Borst MH, Muskiet FAJ, et al. Measure- ment of plasma vitamin K1 (phylloquinone) and K2 (menaquinones-4 and -7) using HPLC-tandem mass spectrometry. Clin Chem Lab Med. 2016; 54: 1201–10. https://doi.org/10.1515/cclm-2015-0864 PMID: 26630696 36. Okura H, Takada Y, Kubo T, Iwata K, Mizoguchi S, Taguchi H et al. Tissue Doppler-derived index of left ventricular filling pressure, E/ E’, predicts survival of patients with non-valvular atrial fibrillation. Heart. 2006; 92: 1248–52. https://doi.org/10.1136/hrt.2005.082594 PMID: 16449507 37. Hillis GS, Møller JE, Pelikka PA, Gersh BJ, Wright RS, Omnen SR, et al. Noninvasive estimation of left ventricular filling pressure by E/e’ is a powerful predictor of survival after myocardial infarction. J Am Coll Cardiol. 2004; 43: 360–7. https://doi.org/10.1016/j.jacc.2003.07.044 PMID: 15013115 PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 19 / 20 Diastolic LV function and matrix Gla protein 38. Sharp AS, Tapp RJ, Thom SA, Francis DP, Hughes AD, Stanton AV, et al. Tissue Doppler E/E’ ratio is a powerful predictor of primary cardiac events in a hypertensive population: an ASCOT substudy. Eur Heart J. 2010; 31: 747–52. https://doi.org/10.1093/eurheartj/ehp498 PMID: 19942604 38. Sharp AS, Tapp RJ, Thom SA, Francis DP, Hughes AD, Stanton AV, et al. Tissue Doppler E/E’ ratio is a powerful predictor of primary cardiac events in a hypertensive population: an ASCOT substudy. Eur Heart J. 2010; 31: 747–52. https://doi.org/10.1093/eurheartj/ehp498 PMID: 19942604 39. Blomstrand P, Engvall M, Festin K, Lindstrõm T, La¨nne T, Maret E, et al. Left ventricular diastolic func- tion, assessed by echocardiography and tissue Doppler imaging, is a strong predictor of cardiovascular events, superior to global left ventricular longitudinal strain, in patients with type 2 diabetes. European Heart Journal—Cardiovascular Imaging. 2015; 16: 1000–7. https://doi.org/10.1093/ehjci/jev027 PMID: 25750201 40. Wang M, Yip GW, Wang AY, Zhang Y, Ho PY, Tse MK, et al. Tissue Doppler imaging provides incre- mental prognostic value in patients with systemic hypertension and left ventricular hypertrophy. J Hypertens. 2005; 23: 183–91. PMID: 15643141 41. Venardos N, Bennett D, Weyant MJ, Reece TB, Meng X, Fullerton DA. Matrix Gla protein regulates cal- cification of the aortic valve. J Surg Res. 2015; 199: 1–6. https://doi.org/10.1016/j.jss.2015.04.076 PMID: 25990696 42. PLOS ONE \| https://doi.org/10.1371/journal.pone.0193967 March 12, 2018 42. Schurgers LJ, Teunissen KJF, Knapen MHJ, Kwaijtaal M, van Diest R, Appels A, et al. Novel conforma- tion-specific antibodies against matrix γ-carboxyglutamic acid (Gla) protein: Undercarboxylated matrix Gla protein as marker for vascular calcification. Arterioscler Thromb Vasc Biol. 2005; 25: 1629–33. https://doi.org/10.1161/01.ATV.0000173313.46222.43 PMID: 15961706 References Schurgers LJ, Teunissen KJF, Knapen MHJ, Kwaijtaal M, van Diest R, Appels A, et al. Novel conforma- tion-specific antibodies against matrix γ-carboxyglutamic acid (Gla) protein: Undercarboxylated matrix Gla protein as marker for vascular calcification. Arterioscler Thromb Vasc Biol. 2005; 25: 1629–33. https://doi.org/10.1161/01.ATV.0000173313.46222.43 PMID: 15961706 20 / 20
https://openalex.org/W4243828031	https://www.qeios.com/read/M2MZ59/pdf	English	null	Silmitasertib Sodium	Definitions	2,020	cc-by	152	Qeios · Definition, February 7, 2020 Open Peer Review on Qeios Open Peer Review on Qeios Silmitasertib Sodium National Cancer Institute National Cancer Institute Qeios ID: M2MZ59 · https://doi.org/10.32388/M2MZ59 Source National Cancer Institute. Silmitasertib Sodium. NCI Thesaurus. Code C85460. The sodium salt of silmitasertib, an orally bioavailable small-molecule inhibitor of CK2 with potential antineoplastic activity. Silmitasertib selectively binds to and inhibits the enzyme casein kinase II (CK2), which may lead to an inhibition of cellular proliferation. CK2, a protein kinase often overexpressed in a variety of cancer cell types, appears to be correlated with malignant transformation, tumor growth and survival. CK2 regulates a diverse array of pro-survival cellular processes including epidermal growth factor receptor (EGFR) signaling, PI3K/AKT/mTOR signaling, hedgehog (Hh) signaling, Hsp90 machinery, hypoxia, and interleukin (IL)-6 expression. CK2 also regulates the activity of XRCC1 and MDC1, two mediator/adaptor proteins that are essential for DNA repair. Qeios ID: M2MZ59 · https://doi.org/10.32388/M2MZ59 1/1
https://openalex.org/W2945615087	https://jurnal.unimed.ac.id/2012/index.php/jgkp/article/download/10463/9394	Indonesian	null	MENINGKATKAN KETERAMPILAN BERTANYA DENGAN MENGGUNAKAN METODE TANYA JAWAB PADA MATA PELAJARAN PKN SISWA KELAS V SDN 106 AEK GALOGA TAHUN AJARAN 2017/2018	JGK (Jurnal Guru Kita)	2,018	cc-by-sa	2,831	Jurnal Guru Kita (JGK). Vol 2 (1) Desember 2017, hlm. 123-129 Jurnal Guru Kita (JGK). Vol 2 (1) Desember 2017, hlm. 123-129 Kata Kunci : Keterampilan bertanya, Metode Tanya Jawab, Hasil Belajar 1945, sedang fungsinya yaitu mengembangkan kemampuan dan membentuk watak serta peradaban bangsa yang bermartabat dalam rangka mencerdaskan kehidupan bangsa. Tujuan pendidikan nasional adalah mengembangkan potensi peserta didik agar menjadi manusia yang beriman dan bertaqwa kepada Tuhan Yang Maha Esa, berakhlak mulia, sehat, berilmu, cakap, MENINGKATKAN KETERAMPILAN BERTANYA DENGAN MENGGUNAKAN METODE TANYA JAWAB PADA MATA PELAJARAN PKN SISWA KELAS V SDN 106 AEK GALOGA TAHUN AJARAN 2017/2018 Ahmat Subuhi Guru SDN 106 Aek Galoga Surel : ahmat_subuhi@gmail.com Abstract : Improving Questioning Skills By Using Question Answer Methods In The Subject Of Civics Student Class V SDN 106 Aek Galoga Lesson Year 2017/2018. This study was conducted with the aim to know whether by using question and answer method in learning Civics can improve the skills to ask students in class V SDN 106 Aek Galoga on the subject of conducting the meeting results Research results before the cycle (prasiklus) shows students are still less skilled in asking that there are 16 people (88.9%) are less skilled and 2 people (11.1%) are quite skilled. After cycles I using question and answer method, there were 3 skillful students (22,2%), 7 (33,3%) skilled and 8 people (44,5% ) are still less skilled. Furthermore, after the cycle II, there appears to be a better increase that is until the end of cycle II there are 3 people (16.7%) who are highly skilled ask, 12 people (66.6%) skilled, 3 people (16.7%) who are already skilled enough to ask questions and no one is classified as less skilled in asking. Keywords : Questioning Skills, Q & A Methods, Learning Outcomes Keywords : Questioning Skills, Q & A Methods, Learning Outcomes Abstrak : Meningkatkan Keterampilan Bertanya Dengan Menggunakan Metode Tanya Jawab Pada Mata Pelajaran PKn Siswa Kelas V SDN 106 Aek Galoga Tahun Pelajaran 2017/2018. Penelitian ini dilakukan dengan tujuan untuk mengetahui apakah dengan menggunakan metode tanya jawab dalam pembelajaran PKn dapat meningkatkan keterampilan bertanya siswa di kelas V SDN 106 Aek Galoga pada pokok bahasan melaksanakan hasil rapat Hasil penelitian sebelum siklus (prasiklus) menunjukkan siswa masih kurang terampil dalam bertanya yaitu terdapat 16 orang (88,9%) yang kurang terampil dan 2 orang (11,1%) yang cukup terampil. Setelah dilakukan siklus I menggunakan metode tanya jawab terjadi peningkatan keterampilan bertanya siswa yaitu hingga akhir siklus I terdapat 3 orang (22,2%) yang terampil, 7 orang (33,3%) yang cukup terampil, dan 8 orang (44,5%) yang masih kurang terampil. Selanjutnya setelah dilakukan siklus II, tampak adanya peningkatan yang lebih baik yaitu hingga akhir siklus II terdapat 3 orang (16,7%) yang sangat terampil bertanya, 12 orang (66,6%) yang terampil, 3 orang (16,7%) yang sudah cukup terampil bertanya dan tidak seorangpun yang tergolong kurang terampil bertanya. Kata Kunci : Keterampilan bertanya, Metode Tanya Jawab, Hasil Belajar PENDAHULUAN Undang-Undang Nomor 20 Tahun 2003 tentang Sistem Pendidikan Nasional telah merumuskan secara tegas mengenai dasar, fungsi, dan tujuan Pendidikan Nasional. Pasal 2 Undang- Undang Nomor 20 Tahun 2003 tentang Sistem Pendidikan Nasional yang memuat dasar pendidikan nasional, yaitu berdasar Pancasila dan Undang-Undang Dasar p-ISSN :2548-883X e-ISSN : 2549-1288 123 Ahmad Subuhi, Meningkatkan Keterampilan Bertanya.... kreatif, dan menjadi warga Negara yang demokratis serta bertanggung jawab. Bertitik tolak dari dasar, fungsi, dan tujuan pendidikan nasional tersebut tampak jelas bahwa manusia Indonesia akan dibentuk melalui proses pendidikan bukan sekedar manusia yang berilmu pengetahuan semata tetapi sekaligus membentuk manusia Indonesia yang berkepribadian sebagai warga Negara Indonesia yang demokratis dan bertanggung jawab. sehingga tujuan pembelajaran tidak dapat tercapai sesuai dengan yang diharapkan. Kondisi semacam ini tentu tidak sejalan dengan semangat untuk menciptakan pembelajaran yang bermakna bagi siswa. Pembelajaran yang bersifat monoton dan masih menerapkan strategi maupun pendekatan pembelajaran konvensional yang memandang siswa sebagai objek, komunikasi lebih banyak berlangsung searah, dan penilaian lebih menekankan aspek kognitif, maka tujuan pembelajaran tidak akan tercapai. Oleh karena itu guru perlu menciptakan suasana pembelajaran yang interaktif dan menyenangkan. Dalam kaitannya dengan pembentukan warga negara Indonesia yang demokratis dan bertanggung jawab, mata pelajaran Pendidikan Kewarganegaraan (PKn) memiliki peranan yang strategis dan penting, yaitu dalam membentuk pribadi siswa maupun sikap dalam berperilaku keseharian, sehingga diharapkan setiap individu mampu menjadi pribadi yang baik. Menciptakan situasi kelas yang inspiratif, interaktif, dan menyenangkan dalam pembelajaran PKn tidaklah mudah, karena sebagian besar siswa masih menganggap PKn sebagai pelajaran yang mementingkan hafalan. Pengetahuan yang diberikan guru dianggap kurang mendayagunakan potensi kognitif, afektif, dan psikomotorik siswa secara optimal. Melalui mata pelajaran PKn ini, diharapkan siswa sebagai warga negara dapat mengkaji dan memahami hak, kewajiban dan tanggung jawabnya sebagai warga negara. Berkaitan dengan tujuan pendidikan nasional, pembangunan dalam dunia pendidikan perlu ditingkatkan. Melalui pembelajaran PKn akan ditanamkan moral yang baik pada diri siswa dari sejak dini. Namun kenyataannya, sebagian siswa memandang mata pelajaran Pendidikan Kewarganegaraan sebagai mata pelajaran yang bersifat konseptual dan teoritis. Akibatnya siswa ketika mengikuti pembelajaran PKn merasa cukup mencatat dan menghafal konsep-konsep dan teori-teori yang dijelaskan oleh guru, tugas-tugas terstruktur yang diberikan dikerjakan secara tidak serius dan bila dikerjakan pun sekedar memenuhi formalitas. Untuk mengubah anggapan tersebut, guru dituntut untuk dapat menciptkan suasana belajar yang lebih menekankan pada aktivitas belajar siswa dalam belajar. Salah satu aktivitas belajar itu adalah keterampilan bertanya siswa. Siswa Kelas V SDN 106 Aek Galoga Tahun Pelajaran 2017/2018” Siswa Kelas V SDN 106 Aek Galoga Tahun Pelajaran 2017/2018” kesempatan kepada siswa untuk bertanya, 2) guru kurang memberikan motivasi kepada siswa untuk bertanya, 3) banyaknya siswa yang enggan dan takut bertanya kepada guru, 4) suasana atau iklim kelas yang kurang kondusif, dan 5) banyaknya siswa yang kurang memahami materi yang disampaikan oleh guru. Berdasarkan uraian latar belakang masalah di atas, dapat diidentifikasi beberapa permasalahan yang berhubungan dengan pelajaran PKn, diantaranya : 1. Sebagian siswa memandang mata pelajaran PKn sebagai mata pelajaran yang bersifat konseptual dan teoritis atau hanya berupa hafalan. Dalam proses pembelajaran, guru tidak menyampaikan informasi begitu saja, akan tetapi memancing agar siswa dapat menemukan sendiri. Oleh sebab itu peran bertanya sangat penting, sebab melalui pertanyaan-pertanyaan guru dapat membimbing dan mengarahkan siswa untuk menemukan setiap materi yang dipelajari. Menurut Nurhadi dan Senduk (2003), “bertanya adalah suatu strategi yang digunakan secara aktif oleh siswa untuk menganalisis dan mengeksplorasi gagasan-gagasan”. 2. Pembelajaran PKn masih bersifat monoton kurang menarik, dan terpusat pada buku (teks book) sehingga setiap pelajaran berlangsung siswa jadi kurang tertarik dan kurang berminat dalam mengikuti pelajarannya. 3. Selama pembelajaran PKn di kelas guru kurang memberikan kesempatan kepada siswa untuk bertanya, dan kurang memberikan motivasi kepada siswa untuk bertanya. Dengan demikian keterampilan siswa dalam bertanya perlu untuk ditingkatkan. Salah satu cara untuk meningkatkan keterlibatan siswa dalam belajar terutama meningkatkan keterampilan bertanya siswa dapat digunakan metode tanya jawab dalam proses pembelajaran di kelas. Metode tanya jawab adalah cara penyampaian suatu pelajaran melalui interaksi dua arah dari guru kepada siswa atau dari siswa kepada guru agar diperoleh jawaban kepastian materi melalui jawaban lisan guru atau siswa. Dalam metode tanya jawab, guru dan siswa sama-sama aktif. Siswa dituntut untuk aktif agar mereka tidak tergantung pada keaktifan guru. 4. Banyak siswa yang kurang memahami materi yang disampaikan oleh guru namun siswa enggan dan takut bertanya kepada guru. Dari beberapa indentifikasi masalah di atas, yang menjadi batasan masalah dalam penelitian ini adalah meningkatkan keterampilan bertanya siswa dalam proses pembelajaran PKn dengan menggunakan metode tanya jawab pada siswa kelas V di SDN 106 Aek Galoga . Materi bahasan PKn yang diteliti dibatasi pada pokok bahasan melaksanakan hasil rapat. Bertitik tolak dari latar belakang masalah dan pembatasan masalah di atas, maka masalah yang akan dibahas dalam penelitian ini dirumuskan: “Apakah dengan menggunakan metode tanya jawab dapat meningkatkan keterampilan bertanya siswa di kelas V di SDN 106 Aek Galoga pada pokok bahasan melaksanakan hasil rapat?”. PENDAHULUAN Keterampilan bertanya merupakan bagian yang tidak terpisahkan dalam rangka meningkatkan kualitas proses dan hasil pembelajaran, yang sekaligus merupakan bagian dari keberhasilan dalam pengelolaan instruksional dan pengelolaan kelas. Melalui keterampilan bertanya guru mampu mendeteksi hambatan proses berpikir di kalangan siswa dan sekaligus dapat memperbaiki dan meningkatkan proses belajar siswa. Namun kenyataannya, dalam proses pembelajaran yang berlangsung dalam kelas menunjukkan keterampilan bertanya siswa masih kurang dikembangkan dan masih kurang optimal, hal ini dapat disebabkan antara lain karena: 1) guru jarang memberikan Berdasarkan observasi awal, di SDN 106 Aek Galoga pembelajaran PKn masih cenderung monoton, kurang menarik, terpusat pada buku (teks book), p-ISSN :2548-883X e-ISSN : 2549-1288 124 Jurnal Guru Kita (JGK). Vol 2 (1) Desember 2017, hlm. 123-129 Siswa Kelas V SDN 106 Aek Galoga Tahun Pelajaran 2017/2018” Berdasarkan uraian di atas, dapat dipahami bahwa bertanya atau pertanyaan dalam pembelajaran merupakan suatu hal yang sangat penting, oleh karena itu peneliti merasa perlu untuk melakukan suatu penelitian tindakan kelas dengan judul “Meningkatkan Keterampilan Bertanya Dengan Menggunakan Metode Tanya Jawab Pada Mata Pelajaran PKn Adapun tujuan yang hendak dicapai dalam penelitian ini adalah untuk mengetahui penggunaan metode tanya p-ISSN :2548-883X e-ISSN : 2549-1288 125 Ahmad Subuhi, Meningkatkan Keterampilan Bertanya.... Panyabungan, Kabupatem Mandailing Natal. jawab dalam pembelajaran PKn dapat meningkatkan keterampilan bertanya siswa di kelas V di SDN 106 Aek Galoga pada pokok bahasan melaksanakan hasil rapat. Subjek dalam penelitian ini adalah siswa kelas V SDN 106 Aek Galoga sebanyak satu kelas yang berjumlah 18 orang. Sedangkan objek dalam penelitian tindakan kelas ini adalah penggunaan metode tanya jawab dalam meningkatkan keterampilan bertanya siswa pada pembelajaran PKn pokok bahasan melaksanakan hasil rapat. Manfaat yang diharapkan dari Manfaat yang diharapkan dari hasil penelitian ini antara lain : hasil penelitian ini antara lain : 1. Bagi siswa a. Menjadikan siswa untuk aktif dalam proses pembelajaran melalui metode tanya jawab. b. Melatih dan meningkatkan keterampilan siswa dalam mengajukan pertanyaan maupun pendapatnya. Intrumen yang digunakan untuk pengumpulan data dalam penelitian ini, adalah lembar observasi. Pengumpulan data dengan lembar observasi dilakukan selama proses pembelajaran berlangsung di tempat penelitian. Observasi dimaksudkan untuk mengetahui kesesuaian tindakan dengan rencana yang telah disusun dan untuk mengetahui bagaimana pelaksanaan tindakan dapat menghasilkan perubahan yang sesuai dengan yang diharapkan. 2. Bagi guru 2. Bagi guru a. Memberikan masukan kepada guru dalam mengajar dan mengembangkan keterampilan bertanya b. Meningkatkan kemampuan mengajar guru. c. Umpan balik bagi guru untuk mengukur keberhasilan dalam pelaksanaan kegiatan belajar mengajar melalui Penelitian Tindakan Kelas. Indikator keterampilan bertanya siswa yang diamati antara lain: (1) berani bertanya atau menyampaikan pertanyaan; (2) memperhatikan dan menyimak pertanyaan guru atau teman; (3) bertanya sesuai dengan topik atau materi yang sedang dipelajari; (4) mengungkapkan pertanyaan dengan jelas dan singkat; dan (5) kelancaran dalam bertanya. 3. Bagi Sekolah a. Meningkatkan kualitas dan mutu sekolah khususnya keterampilan bertanyan siswa dengan menggunakan metode tanya jawab. Data hasil observasi dianalisis bersama-sama dengan guru kelas sebagai mitra kolaborasi, kemudian ditafsirkan berdasarkan kajian pustaka dan pengalaman guru kelas, menggunakan statistik sederhana dengan tabel frekuensi yang menguraikan persentase, dengan rumus : Hasil penelitian ini diharapkan dapat sebagai umpan balik untuk meningkatkan efektivitas dan efisiensi pembelajaran. PEMBAHASAN Hingga pertemuan kedua siklus I masih terdapat 8 orang (44,4%) yang kurang terampil bertanya sehingga secara kelas dinyatakan siswa masih belum terampil dalam bertanya. Hal ini menunjukkan bahwa setelah diterapkan metode tanya jawab pada siklus I, siswa masih mengalami kesulitan dalam mengembangkan keterampilan bertanyanya dan belum siap melaksanakan tanya jawab. Berdasarkan hasil pengamatan awal sebelum dilakukan pembelajaran dengan metode tanya jawab, dari hasil pengamatan peneliti menunjukkan siswa masih kurang memiliki keterampilan dalam bertanya, baik pada aspek keberanian bertanya, memperhatikan dan menyimak pertanyaan guru atau teman, bertanya sesuai materi, mengungkapkan pertanyaan dengan jelas dan singkat serta kelancaran dalam bertanya tampak masih kurang. Hasil analisis pengamatan awal sebelum diberikan tindakan terdapat 16 orang (88,9%) kurang terampil bertanya dan 2 orang (11,1%) yang cukup terampil. Hal ini berarti sebelum pembelajaran menggunakan metode tanya jawab siswa masih kurang terampil dalam bertanya. Siklus II dilakukan dengan lebih menekankan kepada pemberian motivasi kepada siswa untuk lebih berani dan terampil bertanya dengan membimbing dan melatih siswa mengungkapkan pertanyaan sesuai dengan materi yang diajarkan serta memberikan giliran kepada siswa yang jarang atau tidak pernah bertanya pada siklus I agar mau dan tidak takut bertanya, menghargai dan memberikan respon atas pertanyaan atau jawaban yang diberikan siswa. Selama pembelajaran siklus II menggunakan metode tanya jawab, guru kelas tetap melakukan pengamatan sesuai format observasi yang tersedia. Hasil pengamatan menunjukkan bahwa kegiatan pelaksanaan pembelajaran yang dilakukan telah berjalan dengan sangat baik dan berdasarkan pengamatan keterampilan siswa dalam bertanya juga menunjukkan adanya peningkatan. Selanjutnya dilakukan tindakan siklus I dengan menggunakan metode tanya jawab, siklus I dilakukan sebanyak 2 kali pertemuan pada pokok bahasan melaksanakan hasil rapat. Selama pembelajaran berlangsung, guru kelas selaku mitra kolaborasi melakukan pengamatan tentang kegiatan pembelajaran yang dilakukan dan keterampilan siswa dalam bertanya. Hasil pengamatan guru kelas menunjukkan adanya peningkatan keterampilan bertanya siswa dibandingkan sebelum dilakukan siklus I. Hingga pertemuan kedua siklus I, pada aspek memperhatikan dan menyimak pertanyaan guru atau teman, aspek bertanya sesuai dengan topik atau materi serta aspek kelancaran dalam bertanya sudah tergolong cukup. Hasil analisis yang dilakukan menunjukkan bahwa pada pertemuan pertama siklus I terdapat 1 orang (5,6%) yang terampil bertanya, 7 orang (38,9%) yang cukup terampil, dan 10 orang (55,5%) yang kurang terampil. Pada pertemuan kedua siklus I terdapat 3 orang (22,2%) yang terampil, 7 orang (33,3%) yang cukup terampil, dan 8 orang (44,5%) yang masih kurang terampil. METODE Jenis penelitian ini adalah penelitian tindakan kelas (classroom action research). Sesuai dengan jenis penelitian ini maka penelitian ini memiliki tahap-tahap penelitian berupa siklus yang terdiri dari perencanaan, pelaksanaan, ovservasi dan refleksi. 100% N f P   Keterangan P = Persentase subjek yang diamati f = Jumlah subjek yang diamati N = Jumlah subjek keseluruhan Penelitian dilaksanakan di kelas V SDN 106 Aek Galoga Kecamatan 126 Jurnal Guru Kita (JGK). Vol 2 (1) Desember 2017, hlm. 123-129 DAFTAR PUSTAKA Dengan demikian berdasarkan hasil temuan dan analisis data penelitian maka dapat disimpulkan bahwa penggunaan metode tanya jawab dapat meningkatkan keterampilan bertanya siswa Kelas V SDN 106 Aek Galoga Kabupaten Mandailing Natal dalam proses pembelajaran PKn pokok bahasan melaksanakan hasil rapat. Ahmadi, A., dan Prasetya, J.T. 1997. Strategi Belajar Mengajar Untuk Fakultas Tarbiyah Komponen MKDK. Bandung: Pustaka Setia. Aqib, Z. 2006. Peneltian Tindakan Kelas Untuk Guru. Bandung: CV. Yrama Widya. Conny, S., Tangyong, A.F., Belen, S., Matahelemual, Y., dan Suseloardjo, W. 1987. Pendekatan Keterampilan Proses, Bagaimana Mengaktifkan Siswa dalam Belajar. Jakarta: Gramedia. PEMBAHASAN Selama siklus II siswa tampak aktif dan berani bertanya, aktif menyimak pertanyaan guru atau teman dan bertanya sesuai dengan materi yang dipelajari sudah. Para siswa juga sudah sangat baik dalam mengungkapkan pertanyaan dengan jelas dan singkat dan cukup lancar dalam bertanya. Berdasarkan hasil analisis hingga pertemuan keempat siklus II terdapat 3 orang (16,7%) yang sangat terampil dalam bertanya, 12 orang (66,6%) yang terampil, dan 3 orang (16,7%) yang cukup terampil dan tidak seorangpun siswa yang termasuk kurang terampil dalam bertanya. Hal ini p-ISSN :2548-883X e-ISSN : 2549-1288 p-ISSN :2548-883X e-ISSN : 2549-1288 127 Ahmad Subuhi, Meningkatkan Keterampilan Bertanya.... guru lebih terampil menggunakan berbagai metode terutama menggunakan metode tanya jawab. menunjukkan bahwa siswa sudah memiliki keterampilan bertanya yang baik sekaligus berarti siswa telah terampil dalam bertanya. KESIMPULAN Berdasarkan hasil penelitian dan pembahasan, maka diperoleh kesimpulan sebagai berikut: Penggunaan metode tanya jawab dapat meningkatkan keterampilan bertanya siswa Kelas V SDN 106 Aek Galoga Kabupaten Mandailing Natal dalam proses pembelajaran PKn pokok bahasan melaksanakan hasil rapat. Djamarah, S.B., dan Zain, A. 2002. Strategi Belajar Mengajar. Jakarta: Rineka Cipta Djamarah, S.B., dan Zain, A. 2002. Strategi Belajar Mengajar. Jakarta: Rineka Cipta Berdasarkan kesimpulan di atas, sebagai tindak lanjut dari hasil penelitian ini maka diajukan beberapa saran sebagai berikut: Mahfouz, N. 2008. Bertanya, Membuka Pikiran Kita, “You can tell whether a man is clever by his answers. You can tell whether a man is wise by his questions”. http://fransnadeak.blogspot.com, Diakses Desember 2009. 1. Kepada guru hendaknya dalam melaksanakan pembelajaran guru menerapkan metode tanya jawab dan terus memotivasi siswa guna meningkatkan keberanian siswa menanyakan hal-hal yang kurang dimengerti tentang materi yang dipelajari. Nurhadi, dan Senduk, A.G. 2003. Pembelajaran Kontekstual (Contextual Teaching and Learning/ CTL) dan Penerapannya Dalam KBK,. Malang: Universitas Negeri Malang. 2. Bagi siswa diharapkan untuk lebih giat dalam belajar dan tidak malu atau takut bertanya kepada guru tentang materi yang masih kurang dipahami atau kurang jelas, dan disarankan untuk dapat menghargai pertanyaan atau jawaban yang diajukan teman. Poerwadarminta, W.J.S. 1984. Kamus Bahasa Indonesia. Jakarta: Balai Pustaka. 3. Kepada kepala sekolah, dalam upaya meningkatkan proses pembelajaran hendaknya kepala sekolah mengikut sertakan guru-guru dalam pelatihan- pelatihan atau seminar-seminar agar Ribowo, B. 2006. Upaya Meningkatkan Hasil Belajar Siswa Kelas II A SMP Negeri 2 Bajarharjo Brebes dalam Pokok Bahasan Segiempat 128 Jurnal Guru Kita (JGK). Vol 2 (1) Desember 2017, hlm. 123-129 Melalui Model Pembelajaran Tutor Sebaya dalam Kelompok Kecil Tahun Pembelajaran 2005/2006. Skripsi: FMIPA Universitas Negeri Semarang, http://digilib.unnes.ac.id. Diakses Desember 2009. Melalui Model Pembelajaran Tutor Sebaya dalam Kelompok Kecil Tahun Pembelajaran 2005/2006. Skripsi: FMIPA Universitas Negeri Semarang, http://digilib.unnes.ac.id. Diakses Desember 2009. Russeffendi, E.T. 1991. Pengantar Kepada Membantu Guru Mengembangkan Kompetensi Dalam Pengajaran Matematika Untuk Meningkatkan CBSA. Bandung: Tarsito. Sanjaya, W. 2005. Pembelajaran Dalam Implementasi Kurikulum Berbasis Kompetensi. Jakarta: Kencana. Sudjana, N. 1988. Dasar-Dasar Proses Belajar Mengajar. Bandung: Sinar Baru. Suryosubroto, B. 2002. Proses Belajar Mengajar di Sekolah. Jakarta: Rineka Cipta. Wardhani, IGAK.. 2007. Materi Pokok Penelitian Tindakan Kelas. Jakarta: Penerbit Universitas Terbuka. 129
https://openalex.org/W2783751059	https://europepmc.org/articles/pmc5765704?pdf=render	English	null	Persistent mammalian orthoreovirus, coxsackievirus and adenovirus co-infection in a child with a primary immunodeficiency detected by metagenomic sequencing: a case report	BMC infectious diseases	2,018	cc-by	3,631	© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lewandowska et al. BMC Infectious Diseases (2018) 18:33 DOI 10.1186/s12879-018-2946-7 Lewandowska et al. BMC Infectious Diseases (2018) 18:33 DOI 10.1186/s12879-018-2946-7 Open Access Persistent mammalian orthoreovirus, coxsackievirus and adenovirus co-infection in a child with a primary immunodeficiency detected by metagenomic sequencing: a case report Dagmara W. Lewandowska1†, Riccarda Capaul1†, Seraina Prader2, Osvaldo Zagordi1, Fabienne-Desirée Geissberger1, Martin Kügler1,5, Marcus Knorr1, Christoph Berger3, Tayfun Güngör4, Janine Reichenbach2, Cyril Shah1, Jürg Böni1, Andrea Zbinden1, Alexandra Trkola1, Jana Pachlopnik Schmid2† and Michael Huber1† Abstract Background: We report a rare case of Mammalian orthoreovirus (MRV) infection in a child with a primary immunodeficiency (PID). Infections with Mammalian orthoreovirus are very rare and probably of zoonotic origin. Only a few cases have been described so far, including one with similar pathogenesis as in our case. Case presentation: The patient, age 11, presented with flu-like symptoms and persistent severe diarrhea. Enterovirus has been detected over several months, however, exact typing of a positive cell culture remained inconclusive. Unbiased metagenomic sequencing then detected MRV in stool samples from several time points. The sequencing approach further revealed co-infection with a recombinant Coxsackievirus and Adenovirus. MRV-specific antibodies detected by immunofluorescence proved that the patient seroconverted. Conclusion: This case highlights the potential of unbiased metagenomic sequencing in supplementing routine diagnostic methods, especially in situations of chronic infection with multiple viruses as seen here in an immunocompromised host. The origin, transmission routes and implications of MRV infection in humans merit further investigation. Keywords: Orthoreovirus, Coxsackievirus, Adenovirus, Primary immunodeficiency, Metagenomic se Rapid diagnosis of viral infections is crucial in immuno- compromised patients. While a range of established mo- lecular tests can detect specific viruses, high-throughput metagenomic sequencing is based on virus-sequence inde- pendent amplification of nucleic acids isolated directly from clinical samples; as such, it has the potential to identify any virus in an “open diagnostics” approach [2, 3]. Hence, this approach can detect rare viruses that are not included in routine diagnostic panels and viruses with sequence variations that would otherwise be missed [4]. Here we used metagenomic sequencing to complement routine diagnostic methods in a child with a PID suffering from persistent diarrhea. Case presentation sequenced both the Caco-2 culture supernatant and the original stool suspension. We detected many reads of Mammalian orthoreovirus 3 (MRV-3) in the cell culture supernatant and reads of Coxsackievirus A (CV-A) in the original stool suspension (Table 1). No virus reads were detected in the supernatant of a non-infected Caco-2 cell culture used as a negative control. p We report on a female child living in Switzerland with a combined B- and T-cell immunodeficiency, hypogamma- globulinaemia and autoimmunity (diabetes mellitus) under immunoglobulin replacement therapy. In early February 2014, the patient, aged 11 at that time, pre- sented with flu-like symptoms with cough, headache, fever and severe diarrhea. Although the other symptoms resolved, the diarrhea persisted. In view of low initial calprotectin levels (100–400 μg/g), a fecal marker for in- testinal inflammation, the diarrhea was considered as caused by the underlying PID and not by an infectious agent. However, a rise in calprotectin level to 4000 μg/g in March 2015 prompted us to initiate virological investi- gations. Enterovirus had been detected over several months using a specific RT-PCR (Additional file 1: Table S1) and was thus considered as causative agent although other viruses were not tested for during this time period. To trace the timing of these two virus infections we con- ducted a retrospective analysis of the available stool sam- ples by both cell culture and metagenomic sequencing (Table 1). In cell cultures, a CPE was visible after about 14 days of culturing. By sequencing the cell culture superna- tants, we found numerous reads for Human adenovirus C (HAdV-C) in these earlier time points, but none for MRV- 3. By sequencing original stool suspensions, we identified reads corresponding to the previously identified CV-A. We verified all the metagenomics analyses by specific PCRs designed for the isolates in this study (Table 1). In addition to confirming MRV-3 in the three cell cultures supernatants with MRV-3 reads, we also detected MRV- 3 in two corresponding stool suspensions by specific PCR. CV-A was confirmed in all stool suspensions and in one of two colon biopsies; HAdV-C was confirmed in all tested cell cultures and stool suspensions (Table 1). As symptoms continued, additional testing was per- formed. A stool sample (November 3rd 2015) was positive in Caco-2 cell culture showing a cytopathic effect (CPE) after 9 days of incubation. Background Individuals suffering from primary immunodeficiencies (PIDs) are prone to a variety of infections, and some types of PIDs can predispose the affected individuals to particu- lar pathogens. Infections that are usually controlled and asymptomatic in immunocompetent individuals often cause chronic, active disease in immunocompromised individuals [1]. Correspondence: huber.michael@virology.uzh.ch †Equal contributors 1Institute of Medical Virology, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland Full list of author information is available at the end of the article * Correspondence: huber.michael@virology.uzh.ch †Equal contributors 1Institute of Medical Virology, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland Full list of author information is available at the end of the article Page 2 of 5 Lewandowska et al. BMC Infectious Diseases (2018) 18:33 Page 2 of 5 na not done or not available, undet undetermined threshold cycle (> 45), cell culture (CPE) was tested on Caco-2 cells Case presentation The culture sample stained positive with a pool of anti-enterovirus monoclonal anti- bodies, but we did not succeed to further subtype the sus- pected enterovirus using routine immunostaining methods. In order to identify the virus amplified in the Caco-2 cell culture, we applied unbiased metagenomic sequencing as previously described [4]. For three sampling time points in September and November 2015, we To define the MRV-3 infection in more detail, we combined sequencing reads from all time points and re- constructed a consensus sequence that covered the full coding sequences in all 10 segments of the MRV-3 iso- late (GenBank KX932029 - KX932038). Phylogenetic Table 1 Retrospective viral diagnostics using cell culture, metagenomic sequencing and specific PCRs Cell culture (CPE) Metagenomic sequencing (number of reads) Specific PCR (threshold cycle) Date Sample Original material Original material Cell culture supernatant MRV-3 original material MRV-3 cell culture supernatant CV-A original material HAdV original material HAdV cell culture supernatant 25.09.2014 colon biopsy na na na undet na 37.3 undet na 02.01.2015 stool negative CV-A (9′618) na undet na 20.1 33.0 na 29.04.2015 colon biopsy na na na undet na undet undet na 30.04.2015 stool positive CV-A (555) HAdV-C (328′173) undet undet 23.6 35.3 12.0 26.05.2015 stool positive CV-A (746′862) HAdV-C (6′131) HAdV-C (228′718) undet undet 28.0 40.5 13.8 30.06.2015 stool positive na HAdV-C (439′345) undet undet 24.0 37.0 14.6 14.07.2015 stool positive na HAdV-C (452′131) undet undet 23.5 33.3 12.2 10.09.2015 stool positive CV-A (195′249) MRV-3 (6′580) undet 21.6 20.1 undet na 3.11.2015 stool positive CV-A (414) MRV-3 (2′742) 35.8 22.2 27.3 undet na 25.11.2015 stool positive CV-A (1′549’977) HAdV-C (102) MRV-3 (850) 37.3 24.0 21.2 undet undet 09.02.2016 stool na na na undet na 14.5 undet na 18.03.2016 stool negative na na undet na undet undet na 04.05.2016 stool negative na na undet na 23.5 38.24 na na not done or not available, undet undetermined threshold cycle (> 45), cell culture (CPE) was tested on Caco-2 cells Lewandowska et al. BMC Infectious Diseases (2018) 18:33 Page 3 of 5 Page 3 of 5 The most unexpected finding was the infection with MRV-3, proven by metagenomic sequencing and ser- ology of samples from several time points. MRVs belong to the Reoviridae family, a group of non-enveloped dsRNA viruses with 10 genome segments and have been isolated from a wide range of mammalian hosts and in a variety of clinical contexts [5–7]. Case presentation To date, the diseases associated with MRV infections in humans include respiratory disease [8], meningitis [9–11], acute necrotiz- ing encephalopathy [12] and (in a similar case involving a child living in Slovenia) acute gastroenteritis [5]. Zoo- notic transmission is often suspected [6, 13, 14]. Our patient lives close to a farm and might have come in contact with pets, farm animals and animal feces making a zoonotic infection conceivable, however, due to the lack of data on MRV distribution and appropriate samples we can only hypothesize on the transmission route [7]. Stool samples from 16 children with suspicion of gastrointes- tinal infection treated at the same hospital during 2015 all tested negative with qPCR (data not shown). analyses showed that the MRV-3 sequence isolated in this study was most related in all 10 segments to an iso- late from a child in Slovenia [5] and further clustered with isolates from bats in Germany [6] and pigs in Italy [7] (Fig. 1a and Additional file 1: Figures S1-S9). In order to proof active replication and diagnosis of MRV infection, we performed immunofluorescence for the detection of MRV-specific antibodies of the patient. Patient plasma from time points after MRV detection were positive for anti-MRV IgG antibodies showing that the patient seroconverted (Fig. 1b). We were able to reconstruct the full-length coxsackie- virus genome present in this patient using additional sequence information from sequencing with CV-A22 serotype consensus primers. Genotyping and bootscan analysis revealed that the isolate (GenBank KX932039) probably resulted from recombination between CV-A19 and CV-A22 (Additional file 1: Figures S10 and S11). The detected HAdV-C was type 2, strain human/ARG/ A15932/2002/2[P2H2F2] (JX173079.1). The virus growing in the initial Caco-2 cell culture was therefore MRV-3 and not an enterovirus as suggested by the serotyping assay. In fact, the manufac- turer’s datasheet for the reagent used states that there is potential for cross-reaction with Hepatitis A, Reovirus 3, and some Rhinovirus and Astrovirus strains. The latter highlights the genuine difficulty in accurate detection of highly diverse virus families with numerous genotypes such as Enteroviruses where typing by specific PCR can be challenging due to the high susceptibility to recom- bination and the emergence of novel strains [15]. Competing interests h h d l h Competing interests The authors declare that they have no competing interests. Abbreviations CPE: Cytopathic effect; CV-A: Coxsackievirus A; HAdV-C: Human adenovirus C; MRV: Mammalian orthoreovirus; PCR: Polymerase chain reaction; PID: Primary immunodeficiency 6. Kohl C, Lesnik R, Brinkmann A, Ebinger A, Radonic A, Nitsche A, Muhldorfer K, Wibbelt G, Kurth A. Isolation and characterization of three mammalian orthoreoviruses from European bats. PLoS One. 2012;7(8):e43106. 6. Kohl C, Lesnik R, Brinkmann A, Ebinger A, Radonic A, Nitsche A, Muhldorfer K, Wibbelt G, Kurth A. Isolation and characterization of three mammalian orthoreoviruses from European bats. PLoS One. 2012;7(8):e43106. References 1. Dropulic L, Cohen J. Severe viral infections and primary immunodeficiencies. Clin Infect Dis. 2011;53(9):897–909. 1. Dropulic L, Cohen J. Severe viral infections and primary immunodeficiencies. Clin Infect Dis. 2011;53(9):897–909. Funding Funding was provided by the Clinical Research Priority Program “Viral Infectious Diseases” of the University of Zurich. The funding body did not have any role in the design of the study, in the collection, analysis, and interpretation of data and in writing the manuscript. 8. Chua KB, Crameri G, Hyatt A, Yu M, Tompang MR, Rosli J, McEachern J, Crameri S, Kumarasamy V, Eaton BT, et al. A previously unknown reovirus of bat origin is associated with an acute respiratory disease in humans. Proc Natl Acad Sci U S A. 2007;104(27):11424–9. have any role in the design of the study, in the collection, analysis, and interpretation of data and in writing the manuscript. 9. Johansson PJ, Sveger T, Ahlfors K, Ekstrand J, Svensson L. Reovirus type 1 associated with meningitis. Scand J Infect Dis. 1996;28(2):117–20. 10. Hermann L, Embree J, Hazelton P, Wells B, Coombs RTK. Reovirus type 2 isolated from cerebrospinal fluid. Pediatr Infect Dis J. 2004;23(4):373–5. 9. Johansson PJ, Sveger T, Ahlfors K, Ekstrand J, Svensson L. Reovirus type 1 associated with meningitis. Scand J Infect Dis. 1996;28(2):117–20. Received: 12 April 2017 Accepted: 4 January 2018 Received: 12 April 2017 Accepted: 4 January 2018 Received: 12 April 2017 Accepted: 4 January 2018 Additional file 2. Delwart EL. Viral metagenomics. Rev Med Virol. 2007;17(2):115–31. 3. Mokili JL, Rohwer F, Dutilh BE. Metagenomics and future perspectives in virus discovery. Current Opinion in Virology. 2012;2(1):63–77. Additional file 1: Portable Document Format. Table S1. Routine screening results for Enterovirus. Figures S1 to S9. Phylogenetic analyses of all Mammalian orthoreovirus segments isolated in this study. Figure S10. Coxsackievirus BLAST genotyping. Figure S11. Coxsackievirus Bootscan Analysis. Supplementary Methods. Supplementary References. (PDF 456 kb) 4. Lewandowska DW, Zagordi O, Zbinden A, Schuurmans MM, Schreiber P, Geissberger FD, Huder JB, Böni J, Benden C, Mueller NJ, et al. Unbiased metagenomic sequencing complements specific routine diagnostic methods and increases chances to detect rare viral strains. Diagn Microbio Infect Dis. 2015;83(2):133–8. 4. Lewandowska DW, Zagordi O, Zbinden A, Schuurmans MM, Schreiber P, Geissberger FD, Huder JB, Böni J, Benden C, Mueller NJ, et al. Unbiased metagenomic sequencing complements specific routine diagnostic methods and increases chances to detect rare viral strains. Diagn Microbiol Infect Dis. 2015;83(2):133–8. 5. Steyer A, Gutierrez-Aguire I, Kolenc M, Koren S, Kutnjak D, Pokorn M, Poljsak- Prijatelj M, Racki N, Ravnikar M, Sagadin M, et al. High similarity of novel orthoreovirus detected in a child hospitalized with acute gastroenteritis to mammalian orthoreoviruses found in bats in Europe. J Clin Microbiol. 2013; 51(11):3818–25. 5. Steyer A, Gutierrez-Aguire I, Kolenc M, Koren S, Kutnjak D, Pokorn M, Poljsak- Prijatelj M, Racki N, Ravnikar M, Sagadin M, et al. High similarity of novel orthoreovirus detected in a child hospitalized with acute gastroenteritis to mammalian orthoreoviruses found in bats in Europe. J Clin Microbiol. 2013; 51(11):3818–25. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. 10. Hermann L, Embree J, Hazelton P, Wells B, Coombs RTK. Reovirus type 2 isolated from cerebrospinal fluid. Pediatr Infect Dis J. 2004;23(4):373–5. 11. Tyler KL, Barton ES, Ibach ML, Robinson C, Campbell JA, O’Donnell SM, Valyi- Nagy T, Clarke P, Wetzel JD, Dermody TS. Isolation and molecular characterization of a novel type 3 reovirus from a child with meningitis. J Infect Dis 2004;189(9):1664-1675. Publisher’s Note In summary, this case highlights the complexity of infec- tions in immunocompromised hosts and reveals limita- tions of routine diagnostic methods. A combination of traditional cell culture, metagenomic sequencing, verifica- tion by specific PCRs and serology proved key to detect a persistent co-infection with three clinically relevant viruses. While the presence of these viruses in stool specimen suggests a link with the observed pathogenesis, it cannot be defined to what extent each of the viruses contributed to the initial flu-like symptoms and prolonged diarrhea. Based on the timing of virus detection, the earlier detected CV-A and HAdV-C are more likely to be involved in the child’s condition than the later emerging MRV-3. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgements Not applicable 7. Lelli D, Beato MS, Cavicchio L, Lavazza A, Chiapponi C, Leopardi S, Baioni L, De Benedictis P, Moreno A. First identification of mammalian orthoreovirus type 3 in diarrheic pigs in Europe. Virol J. 2016;13(1):139. 7. Lelli D, Beato MS, Cavicchio L, Lavazza A, Chiapponi C, Leopardi S, Baioni L, De Benedictis P, Moreno A. First identification of mammalian orthoreovirus type 3 in diarrheic pigs in Europe. Virol J. 2016;13(1):139. Authors’ contributions SP, CB, TG, JR and JPS cared for the patient and provided clinical data and materials. RC, JB and AZ analyzed and interpreted routine diagnostic results. DWL, FDG and CS performed sequencing experiments and PCR assays. DWL, OZ and MH analyzed the sequencing data. MK and MK performed cell culture and immunostaining. DWL, AT and MH wrote the manuscript. All authors read and approved the final manuscript. 12. Ouattara LA, Barin F, Barthez MA, Bonnaud B, Roingeard P, Goudeau A, Castelnau P, Vernet G, Paranhos-Baccala G, Komurian-Pradel F. Novel human reovirus isolated from children with acute necrotizing encephalopathy. Emerg Infect Dis. 2011;17(8):1436–44. 13. Lelli D, Moreno A, Lavazza A, Bresaola M, Canelli E, Boniotti MB, Cordioli P. Identification of Mammalian orthoreovirus type 3 in Italian bats. Zoonoses Public Health. 2013;60(1):84–92. Discussion and conclusions Using a metagenomic sequencing approach, we detected multiple virus infections in a child with PID and persis- tant diarrhea. Although routine PCR methods detected Enterovirus in stool samples for a prolonged period of time, an attempt to subtype the virus after cell culturing was not successful. In order to resolve this, we per- formed metagenomic sequencing of cell culture superna- tants and stool suspensions. Fig. 1 Mammalian orthoreovirus infection confirmed by phylogenetic analysis and immunofluorescence staining. a Phylogenetic analysis of Mammalian orthoreovirus segment S1 isolated in this study (circle, mew716_S1_type_3_human) reveals a close relationship with previously described isolates identified in a child in Slovenia (triangle, KF154730), bats in Germany (JQ412761), and pigs in Italy (KX343206). Phylogenetic trees were constructed in MEGA7 using the Maximum Likelihood method based on the Kimura 2-parameter model. Bootstrap values from 1000 tries are shown. MRV-type and host species are depicted if available. b Anti-MRV-3 immunofluorescence of patient plasma before and after seroconversion on MRV-3-infected and uninfected Caco-2 cells Fig. 1 Mammalian orthoreovirus infection confirmed by phylogenetic analysis and immunofluorescence staining. a Phylogenetic analysis of Mammalian orthoreovirus segment S1 isolated in this study (circle, mew716_S1_type_3_human) reveals a close relationship with previously described isolates identified in a child in Slovenia (triangle, KF154730), bats in Germany (JQ412761), and pigs in Italy (KX343206). Phylogenetic trees were constructed in MEGA7 using the Maximum Likelihood method based on the Kimura 2-parameter model. Bootstrap values from 1000 tries are shown. MRV-type and host species are depicted if available. b Anti-MRV-3 immunofluorescence of patient plasma before and after seroconversion on MRV-3-infected and uninfected Caco-2 cells Page 4 of 5 Lewandowska et al. BMC Infectious Diseases (2018) 18:33 Page 4 of 5 Notably, CV-A was detected in stool suspensions but not in cell culture supernatants. While cell culturing was critical for MRV-3 detection, CV-A, in line with a general difficulty to culture coxsackieviruses [16], did not infect Caco-2 cells. The reason that MRV-3 was not detected by metagenomic sequencing in the original stool suspension, but only after amplification in cell culture, is likely because it was at levels too low to be detected with the applied depth of sequencing. canton of Zurich approved the study and written informed consent was obtained. canton of Zurich approved the study and written informed consent was obtained. Author details 1Institute of Medical Virology, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland. 2Division of Immunology, University Children’s Hospital Zurich, Steinwiesstrasse 75, 8032 Zurich, Switzerland. 3Division of Infectious Diseases and Hospital Epidemiology, University Children’s Hospital Zurich, Steinwiesstrasse 75, 8032 Zurich, Switzerland. 4Division of Stem Cell Transplantation, University Children’s Hospital Zurich, Rämistrasse 100, 8091 Zurich, Switzerland. 5Present address: Unilabs, Ringstrasse 12, 8600 Dübendorf, Switzerland. Consent for publication Written informed consent for the publication of this case report was obtained from both parents of the child. Competing interests The authors declare that they have no competing interests. Lewandowska et al. BMC Infectious Diseases (2018) 18:33 associated with acute influenza-like illness in an adult patient. PLoS One. 2011;6(10):e25434. 15. Lukashev AN. Role of recombination in evolution of enteroviruses. Rev Med Virol. 2005;15(3):157–67. 16. Wenner HA, Lenahan MF. Propagation of group a Coxsackie viruses in tissue cultures. II. Some interactions between virus and mammalian cells. Yale J Biol Med. 1961;34:421–38. 16. Wenner HA, Lenahan MF. Propagation of group a Coxsackie viruses in tissue cultures. II. Some interactions between virus and mammalian cells. Yale J Biol Med. 1961;34:421–38. Ethics approval and consent to participate Non-invasive samples were obtained from patient mew716 in the frame of the Viral Metagenome Study of the Clinical Research Priority Program ‘Viral Infectious Diseases’ of the University of Zurich. The ethics committee of the 14. Chua KB, Voon K, Yu M, Keniscope C, Abdul Rasid K, Wang L-F. Investigation of a potential zoonotic transmission of orthoreovirus 14. Chua KB, Voon K, Yu M, Keniscope C, Abdul Rasid K, Wang L-F. Investigation of a potential zoonotic transmission of orthoreovirus Page 5 of 5 Page 5 of 5 Lewandowska et al. BMC Infectious Diseases (2018) 18:33 • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step:
https://openalex.org/W4320727042	https://www.nature.com/articles/s41598-023-29768-6.pdf	English	null	Author Correction: Association between weekend catch-up sleep and dyslipidemia among Korean workers	Scientific reports	2,023	cc-by	422	www.nature.com/scientificreports www.nature.com/scientificreports The original version of this Article contained an error in the Acknowledgements section. “This research did not receive any specific grant from funding agencies in the public, commercial, or not-for- profit sectors.” Author Correction: Association between weekend catch‑up sleep and dyslipidemia among Korean workers OPEN Ye Seul Jang , Yu Shin Park , Kyungduk Hurh , Eun‑Cheol Park & Sung‑In Jang g, , y g , g g Correction to: Scientific Reports https://doi.org/10.1038/s41598-023-28142-w, published online 17 January 2023 Correction to: Scientific Reports https://doi.org/10.1038/s41598-023-28142-w, published online 17 January 2023 The original version of this Article contained an error in the Acknowledgements section. now reads: “This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2022R1F1A1062794).” “This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2022R1F1A1062794).” The original Article has been corrected. 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/. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2023 \| https://doi.org/10.1038/s41598-023-29768-6 Scientific Reports \| (2023) 13:2615
https://openalex.org/W2951872997	https://europepmc.org/articles/pmc6627339?pdf=render	English	null	Attenuation of UVB-Induced Photo-Aging by Polyphenolic-Rich Spatholobus Suberectus Stem Extract Via Modulation of MAPK/AP-1/MMPs Signaling in Human Keratinocytes	Nutrients	2,019	cc-by	12,324	Received: 25 April 2019; Accepted: 7 June 2019; Published: 14 June 2019 Abstract: It is well known that ultraviolet light activates mitogen-activated protein (MAP) kinase by increasing the reactive oxygen species (ROS) in the body, enhancing activating protein 1(AP-1) complexes (c-Jun and c-Fos), increasing matrix metalloproteinases (MMPs) and degrading collagen and elastin. In this study, we conﬁrmed that polyphenolic rich Spatholobus suberectus (SS) stem extracts suppressed ultraviolet (UV)-induced photo-aging. The major active components of SS stem extracts were identiﬁed as gallic acid, catechin, vanillic acid, syringic acid and epicatechin. The aqueous and ethanolic extracts of the stem of SS (SSW and SSE, respectively) signiﬁcantly reduced the elastase enzyme activity. Moreover, both extracts were suppressed the ROS generation and cellular damage induced by UVB in HaCaT cells. Our results also revealed that SSE could regulate the expression of MMPs, tissue inhibitor of matrix metalloproteinase (TIMP)-1, collagen type I alpha 1 (COL1A1), elastin (ELN) and hyaluronan synthase 2 (HAS2) at their transcriptional and translational level. Furthermore, SSE was blocked the UVB-induced phosphorylation of mitogen-activated protein kinases (MAPKs), nuclear factor-kappa B (NF-κB) and c-Jun. Moreover, combination of syringic acid, epicatechin and vanillic acid showed strong synergistic eﬀects on elastase inhibition activity, in which the combination index (CI) was 0.28. Overall, these results strongly suggest that the polyphenolics of SSE exert anti-ageing potential as a natural biomaterial to inhibit UVB-induced photo-aging. Keywords: anti-aging; Spatholobus suberectus; matrix metalloproteinases (MMPs); collagen type I alpha 1 (COL1A1); elastin (ELN); mitogen-activated protein kinase (MAPK) Received: 25 April 2019; Accepted: 7 June 2019; Published: 14 June 2019 Attenuation of UVB-Induced Photo-Aging by Polyphenolic-Rich Spatholobus Suberectus Stem Extract Via Modulation of MAPK/AP-1/MMPs Signaling in Human Keratinocytes Kyoo-Ri Kwon 1,†, Md Badrul Alam 1,2,† , Ji-Hyun Park 1, Tae-Ho Kim 3 and Sang-Han Lee 1,2,* 1 Department of Food Science & Biotechnology, Kyungpook National University, Daegu 41566, Korea; rbﬂ0116@knu.ac.kr (K.-R.K.); mbalam@knu.ac.kr (M.B.A.); wlgus6744@knu.ac.kr (J.-H.P.) 2 Food and Bio-Industry Research Institute, Inner Beauty/Anti-ageing Center, Kyungpook National University, Daegu 41566, Korea 3 Biomedical Research Institute Kyungpook National University Hospital Daegu 41944 Korea; Kyoo-Ri Kwon 1,†, Md Badrul Alam 1,2,† , Ji-Hyun Park 1, Tae-Ho Kim 3 and Sang-Han Lee 1,2,* 1 Department of Food Science & Biotechnology, Kyungpook National University, Daegu 41566, Korea; rbﬂ0116@knu.ac.kr (K.-R.K.); mbalam@knu.ac.kr (M.B.A.); wlgus6744@knu.ac.kr (J.-H.P.) 2 Food and Bio-Industry Research Institute, Inner Beauty/Anti-ageing Center, Kyungpook National University, Daegu 41566, Korea 3 Kyoo-Ri Kwon 1,†, Md Badrul Alam 1,2,† , Ji-Hyun Park 1, Tae-Ho Kim 3 and Sang-Han Lee 1,2,* 1 Department of Food Science & Biotechnology, Kyungpook National University, Daegu 41566, Korea; rbﬂ0116@knu.ac.kr (K.-R.K.); mbalam@knu.ac.kr (M.B.A.); wlgus6744@knu.ac.kr (J.-H.P.) 2 Food and Bio-Industry Research Institute, Inner Beauty/Anti-ageing Center, Kyungpook National University, Daegu 41566, Korea 3 Biomedical Research Institute, Kyungpook National University Hospital, Daegu 41944, Korea; kimth0929@ynu.ac.kr 3 Biomedical Research Institute, Kyungpook National University Hospital, Daegu 41944, Korea; kimth0929@ynu.ac.kr 3 Biomedical Research Institute, Kyungpook National University Hospital, Daegu 41944, Korea; kimth0929@ynu.ac.kr * Correspondence: sang@knu.ac.kr; Tel.: +82-53-950-7754 † These authors contributed equally to this work. nutrients nutrients 1. Introduction Skin aging is aﬀected by many factors, such as ultraviolet radiation (UVR), oxidative stress, and inheritance. Among them, photoaging accounts for about 80% of skin aging [1]. UVR is divided into three wavelengths, including UVA (320–400 nm), UVB (290–320 nm), and UVC (200–290 nm). While UVC is absorbed by the ozone layer, UVA and UVB can reach the surface of the earth [2]. While most biomolecules cannot absorb UVA, UVB is predominantly injurious to living organisms. UVB contributes to detrimental eﬀects directly through the production of reactive oxygen species (ROS) which is associated with DNA damage and inﬂection of gene expression. UVB initiates a Nutrients 2019, 11, 1341; doi:10.3390/nu11061341 www.mdpi.com/journal/nutrients www.mdpi.com/journal/nutrients 2 of 14 Nutrients 2019, 11, 1341 photo-oxidation reaction primarily on the epidermis of the skin via augmentation of the cellular ROS level that causes imbalance of the skin antioxidant, sequentially accelerating photoaging [3,4]. UVB not only hinders collagen synthesis and promotes its breakdown, but also boosts the degradation of elastin in ﬁbroblasts, resulting in the loss of skin elasticity which make deep wrinkle formation [5]. Collagen constitutes about 70% of the dermal layer and both matrix metalloproteinases (MMPs), and tissue inhibitors of matrix metalloproteinases (TIMPs) play a pivotal role in collagen degradation in epidermal keratinocytes and dermal ﬁbroblasts. Up to now, 28 MMPs and 4 TIMPs have been identiﬁed [6]. MMPs are a family of zinc-dependent enzymes which are responsible for the degradation of collagen and extracellular matrix (ECM) components [7]. Type I collagen, which constitutes about 80 to 90% of the skin, is mainly controlled by MMP-1 (type I collagenase) and MMP-3 (stromelysin-1), whereas MMP-2 (gelatinase-A) and MMP-9 (gelatinase-B) degrade gelatin; moreover MMP-2, MMP-7 (matrilysin-1), and MMP-12 (metalloelastase) degrade elastin [8]. Not only in response to UVB, UVA and blue light, the level of MMPs in the skin is highly up-regulated [3,4,9,10], therefore, the prevention of UVB-induced up-regulation of MMPs is one of the target pathways to inhibit wrinkle formation and prevent skin aging. In living cells, ultraviolet (UV)-irradiation activates mitogen-activated protein kinase (MAPK) and regulates nuclear factor-kappa B (NF-κB) expression by generating reactive oxygen species (ROS) [3,4]. MAPK, is an enzyme family, consisting of three types: extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 kinase, which are involved in cell proliferation, diﬀerentiation, apoptosis, and inﬂammation. 1. Introduction The MAPK pathways also regulate the transcription factor activating protein 1 (AP-1), a heterodimer comprised of c-Fos and c-Jun, which, in turn, up-regulates MMPs in the skin [11,12]. Various synthetic materials have been developed as skin anti-aging agents and retinoid is the most popular among them. Retinoid is a synthetic analog of vitamin A consisting of retinoic acid and retinol (vitamin A) which promotes collagen synthesis in photoaged skin and inhibits the expression of MMPs [13]. Retinoic acid also promotes elastin and collagen synthesis [14]. However, retinoid is also reported to cause skin and liver toxicity and diseases, such as paronychia [14]. Owing to safety and lack of side eﬀects, natural materials are being considered for the development of eﬀective and safe anti-photoaging agents in the ﬁeld of cosmetic research and development. Spatholobus suberectus (SS) is a climbing shrub plant containing a red resin, belonging to the Leguminosae family, and mainly grows in China (Figure 1A). The stem of SS is known as “Gye-Hyeol-Deung” in Korea and “Ji Xue Teng” in China because it produces a red juice like chicken’s blood [15]. Traditionally, the stem of SS has been applied to treat inﬂammation-induced thrombosis and peripheral blood vessels [16]. Numerous scientiﬁc reports have revealed that SS has anti-hepatitis C virus activity [17], antiplatelet [18], anti-breast cancer [19], antioxidant [20], chondrogenesis stimulating [21], antiviral [22], and protective eﬀects against cerebral ischemia [23]. It has also been reported that SS has the potential to regulate cartilage-related MMPs and TIMPs [21] and anti-inﬂammatory activity [23]. p g g y y Based on previous reports, it was hypothesized that the aqueous and ethanolic extracts of Spatholobus suberectus stem (SSW and SSE, respectively) would play a pivotal role in human healthy skin homeostasis by mediating functionality and nutritional balance in body. However, there have been no reports regarding the potential dermatological application of the SS stem. SS extracts were considered in this study to evaluate its eﬀects against photo-aging in human keratinocyte cultures and the underlying mechanism by which SS extract mitigates the appearance of wrinkles. 1. Introduction (E) Molecular structure of the 5 compounds identified, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). Figure 1. Classical feature of Sapatholobus suberectus (SS) and identiﬁcation of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) proﬁle of the stem of Spatholobus suberectus. (E) Molecular structure of the 5 compounds identiﬁed, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). 2 1 Plant Materials and Preparation of Plant Extracts 2.1. Plant Materials and Preparation of Plant Extracts 2.1. Plant Materials and Preparation of Plant Extracts The stem of Spatholobus suberectus (SS) was purchased from a Chinese medicinal herb shop in Zhengzhou, China (Figure 1A). The sample was dried at 30 °C for two days and then pulverized into a fine powder. Ten grams of coarsely dried powder was extracted three times using 100% ethanol and distilled water under reflux for 3 h. The extract was decanted using filter paper (Whatman No. 1, Schleicher & Schuell, Keene, NH, USA). Then, the solvent was removed and dried using a rotary vacuum evaporator (Tokyo Rikakikai Co. Ltd., Tokyo, Japan) and finally pulverized after freeze- drying (Ilshin Biobase, Goyang, Korea) the aqueous and ethanolic extract of SS (SSW and SSE, respectively) (Figure 1B,C, respectively). Powdered extracts were dissolved in distilled water and stored at 4 °C until testing. The stem of Spatholobus suberectus (SS) was purchased from a Chinese medicinal herb shop in Zhengzhou, China (Figure 1A). The sample was dried at 30 ◦C for two days and then pulverized into a ﬁne powder. Ten grams of coarsely dried powder was extracted three times using 100% ethanol and distilled water under reﬂux for 3 h. The extract was decanted using ﬁlter paper (Whatman No. 1, Schleicher & Schuell, Keene, NH, USA). Then, the solvent was removed and dried using a rotary vacuum evaporator (Tokyo Rikakikai Co. Ltd., Tokyo, Japan) and ﬁnally pulverized after freeze-drying (Ilshin Biobase, Goyang, Korea) the aqueous and ethanolic extract of SS (SSW and SSE, respectively) (Figure 1B,C, respectively). Powdered extracts were dissolved in distilled water and stored at 4 ◦C until testing. 1. Introduction Therefore, in this study, in order to assess what kinds of nutritional food ingredients are involved in the plant and how the nutrients can be applied for skin care for the better healthy life, we determined its eﬀect on skin wrinkles and elasticity, as well as the mechanism to assess whether it has suﬃcient value as a natural material to suppress aging. This can contribute to the development of novel and useful cosmetic agents, supplements, and functional foods. 3 of 14 of 15 Nutrients 2019, 11, 1341 Nutrients 2019, 11, x FO Figure 1. Classical feature of Sapatholobus suberectus (SS) and identification of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) profile of the stem of Spatholobus suberectus. (E) Molecular structure of the 5 compounds identified, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). Figure 1. Classical feature of Sapatholobus suberectus (SS) and identiﬁcation of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) proﬁle of the stem of Spatholobus suberectus. (E) Molecular structure of the 5 compounds identiﬁed, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). Figure 1. Classical feature of Sapatholobus suberectus (SS) and identification of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) profile of the stem of Spatholobus suberectus. (E) Molecular structure of the 5 compounds identified, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). Figure 1. Classical feature of Sapatholobus suberectus (SS) and identiﬁcation of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) proﬁle of the stem of Spatholobus suberectus. (E) Molecular structure of the 5 compounds identiﬁed, gallic acid (1); catechin (2); vanillic acid (3); syringic acid (4); and epicatechin (5). Figure 1. Classical feature of Sapatholobus suberectus (SS) and identification of ingredients. (A) Stem of Sapatholobus suberectus. (B,C) Powder of Sapatholobus suberectus stem’s aqueous (SSW) and ethanolic (SSE) extracts. (D) high-performance liquid chromatography (HPLC) profile of the stem of Spatholobus suberectus. 2.4. Cell Culture, UVB-Irradiation and Cell Viability Assay Human keratinocyte (HaCaT) cells were purchased from AddexBio Technologies (San Diego, CA, USA). Cells were grown in Dulbecco’s modiﬁed Eagle’s medium (DMEM) (Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS) (Hyclone, Mordialloc, Victoria, Australia) and 1% penicillin-streptomycin (P/S) (Sigma-Aldrich, St. Louis, MO, USA) at 37 ◦C in a humidiﬁed atmosphere containing 5% CO2. Then, sub-conﬂuent cells were treated with indicated concentration of SSW and SSE for 24 h. Subsequently, the cells were exposed to UVB at a dose of 40 mJ/cm2 using a UVB source (Bio-Link Crosslinker, Vilber Lourmat, Cedex, France) set at a spectral peak of 312 nm for 20 s. Rigel et al. reported healthy high school volunteers daily received the UVB irradiation by 8.01 mJ/cm2/day [27]. In this experiment, we used UVB radiation at 40 mJ/cm2, which is equivalent to approximately 5 days of sun exposure. After UVB irradiation, the cells were cultured in serum-free medium for 24 h. Cell viability was determined using the 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) colorimetric assay as described previously [28]. 2.2. High-Performance Liquid Chromatography (HPLC) Analysis 2.2. High-Performance Liquid Chromatography (HPLC) Analysis The phytochemical characteristics of SSW and SSE were analyzed by high-performance liquid chromatography (HPLC) using standard compounds such as catechin, (-)-epigallocatechin gallate (EGCG, E4268, Sigma-Aldrich), epicatechin, gallic acid, syringic acid, and vanillic acid. The HPLC analysis was performed using a Shimadzu Prominence Auto Sampler (SIL-20A) HPLC system (Shimadzu, Kyoto, Japan), equipped with an SPD-M20A diode array detector (PDA) and LC solution 1.22 SP1 software. Before analysis, the samples were filtered through a 0.2 µm syringe filter (Pall Life Sciences, Ann Arbor, MI, USA). The reverse-phase chromatographic analysis was carried out using a Phenomenex C18 column (4.6 mm × 250 mm) packed with 5 µm diameter particles and maintained at 25 °C. A stepwise gradient of solvent A to B was used (A: 2% acetic acid and B: 50% acetonitrile (ACN) in 0.5% acetic acid) according to a previous report [24] with a slight modification. The flow rate was 1 mL/min, and 10 µL was injected. The phytochemical characteristics of SSW and SSE were analyzed by high-performance liquid chromatography (HPLC) using standard compounds such as catechin, (−)-epigallocatechin gallate (EGCG, E4268, Sigma-Aldrich), epicatechin, gallic acid, syringic acid, and vanillic acid. The HPLC analysis was performed using a Shimadzu Prominence Auto Sampler (SIL-20A) HPLC system (Shimadzu, Kyoto, Japan), equipped with an SPD-M20A diode array detector (PDA) and LC solution 1.22 SP1 software. Before analysis, the samples were ﬁltered through a 0.2 µm syringe ﬁlter (Pall Life Sciences, Ann Arbor, MI, USA). The reverse-phase chromatographic analysis was carried out using a Phenomenex C18 column (4.6 mm × 250 mm) packed with 5 µm diameter particles and maintained at 25 ◦C. A stepwise gradient of solvent A to B was used (A: 2% acetic acid and B: 50% acetonitrile (ACN) in 0.5% acetic acid) according to a previous report [24] with a slight modiﬁcation. The ﬂow rate was 1 mL/min, and 10 µL was injected. Nutrients 2019, 11, 1341 4 of 14 2.3. Elastase Inhibition Assay and Combination Index 2.3. Elastase Inhibition Assay and Combination Index The elastase inhibition assay was performed according to a previously reported method with minor modiﬁcations [25]. Brieﬂy, the reaction was carried out in a 0.1 M Tris-HCl buﬀer (pH 8.0) containing 0.78 mM N-Succinyl-(Ala)-3-p-nitroanilide (Sigma-Aldrich, St. Louis, MO, USA) as a substrate and 0.04 U elastase (Sigma-Aldrich, St. Louis, MO, USA). EGCG was used as a positive control. A predetermined concentration (10, 30, 100, or 300 µg/mL) of each sample was mixed thoroughly with 100 µL of the substrate solution and consequently 100 µL of the enzyme solution was added, and the absorbance was measured at 405 nm in a microplate reader (Wallac Victor3 1420 Multilabel Counter, Perkin Elmer, Waltham, MA, USA). The mode of interaction between syringic acid (SA), epicatechin (EP) and vanillic acid (VA) in inhibiting elastase was further analyzed by the CompuSyn program (ComboSyn Inc, Paramus, NJ, USA), which applies median eﬀect equation methods. The combined drug eﬀect is expressed as combination index (CI) versus fraction aﬀected (Fa), with CI < 1 indicating synergism, CI = 1 indicating an additive eﬀect, and CI > 1 indicating antagonism [26]. The CI values were calculated by the Chou–Talalay method based on the median-eﬀect equation and the classic isobologram equation [26] using CompuSyn software (ComboSyn Inc, Paramus, NJ, USA). 3.1. HPLC Analysis of Stem of Spatholobus Suberectus (SS) To gain insight into the phytochemicals present in SS stem extract, HPLC analysis was performed with standard phenolic and ﬂavonoid compounds. Interestingly, the aqueous and ethanolic extracts of SS stem (SSW and SSE, respectively) (Figure 1B,C) demonstrated several peaks, with the retention times to be harmonized in the following standard compounds: gallic acid (6.380 min), catechin (20.074 min), vanillic acid (23.601 min), syringic acid (25.352 min) and epicatechin (26.809 min) (Figure 1D). By using the peak areas of known concentrations of standards, the amounts of these polyphenols in Spatholobus suberectus stem extract were determined. As shown in Figure 1D, SSW contains gallic acid (6.90 µg/mL), syringic acid (27.08 µg/mL) and epicatechin (105.59 µg/mL) while SSE contains gallic acid (1.48 µg/mL), vanillic acid (21.07 µg/mL), syringic acid (28.23 µg/mL), catechin (1.86 µg/mL), and epicatechin (142.17 µg/mL) (Figure 1D). Both extracts commonly found to contain gallic acid, catechin, syringic acid and epicatechin. 2.5. ROS Generation Assay Intracellular ROS production in cells was detected using 2′,7′-Dichloroﬂuorescein diacetate (DCFH-DA) (Sigma-Aldrich, St. Louis, MO, USA) according to the methods described previously [26]. Cells were grown to 70–80% conﬂuence then cultured (1 × 105 cells/mL) with indicated concentration of SSW and SSE in 96-well black clear bottomed plates (Corning Inc., Corning, NY, USA) for 24 h and then exposed to UVB-irradiation (40 mJ/cm2), and further incubated for 24 h. Then, the cells were washed with phosphate-buﬀered saline (PBS) twice and treated with DCFH-DA (25 µM) for 30 min. Finally, ﬂuorescence intensity was measured at excitation and emission wavelengths of 485 and 528 nm, respectively, by a ﬂuorescence microplate reader (Victor3, PerkinElmer, Waltham, MA, USA). 2.6. RNA Extraction and Reverse Transcription-Polymerase Chain Reaction (RT-PCR) HaCaT cells (1 × 105 cells/mL) were cultured with indicated concentration of SSW and SSE (f.c. 3, 10, or 30 µg/mL) in 6-well plates for 24 h. Total RNA was isolated using TRIzol (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. 2 µg of total RNA was used to prepare the complementary DNA (cDNA) using RT & GO Mastermix (MP Biomedicals, Seoul, Korea). A polymerase chain reaction (PCR) Thermal Cycler Dice TP600 (Takara Bio Inc., Otsu, Japan) was used to carry out RT-PCR using the various primer sequences (Table S1), according to the methods described earlier [28]. 5 of 14 Nutrients 2019, 11, 1341 2.7. Preparation of Protein Lysates and Western Blotting 2.7. Preparation of Protein Lysates and Western Blotting 2.7. Preparation of Protein Lysates and Western Blotting The lysates of HaCaT cells (2 × 105 cells/mL) were prepared using radioimmunoprecipitation assay (RIPA) buﬀer with a phosphatase and protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA) and the bicinchoninic acid (BCA) method was applied to quantify the protein content. A nuclear/cytosolic fractionation kit (Sigma-Aldrich, St. Louis, MO, USA) was used for the extraction of nuclear proteins. Aliquots of 30 µg of total proteins were used to carry out the Western blot analysis using various antibodies (Table S2) according to our previously described methods [28]. 3.2. Inhibition of Elastase Activity by SS Stem The elastase inhibition assay revealed both SSW and SSE were suppressed the elastase activity in a concentration-dependent manner (Figure 2A). At 100 µg/mL, SSE showed 79% inhibition of elastase activity whereas EGCG (used as positive control) and SSW had similar inhibitory activity as 59%. The IC50 values of the elastase inhibition activity were in the order of SSE > SSW > EGCG. These results suggest that SSW and SSE were found to suppress the breakdown of elastin, which in combination with collagen contributes to skin wrinkles. 2.8. Statistical Analysis The experimental data are presented as the mean ± standard deviation (SD). Enzyme, cell viability data were analyzed using a paired Student’s t-test. The statistical analysis of the rest of the data was carried out by one-way analysis of variance (ANOVA), followed by Dennett’s test using the SigmaPlot 12.5 (Systat Software Inc., San Jose, CA, USA). Diﬀerences were considered signiﬁcant when ** p ≤0.05 or * p ≤0.01. 3.3. The Eﬀet of SS on Cell Viability of Human Keratinocytes (HaCaT) Cells Cell viability was determined using the MTT assay in HaCaT cells (Figure 2B). The cell viability of approximately 80% was considered to be non-toxic. SSW did not show any toxicity up to a concentration of 30 µg/mL, but the survival rate was decreased to 68% at a concentration of 100 µg/mL. On the other hand, SSE showed no cell toxicity up to 100 µg/mL, but the survival rate was approximately 68% at a concentration of 300 µg/mL. Therefore, 30 µg/mL was considered the highest concentration for the subsequent experiments for both samples. To determine the UVB intensity to be used in the analysis, the survival rate of the cells was examined by irradiating them with UVB. Cells were irradiated with UVB at 0, 30, 40, and 50 mJ/cm2 for cytotoxicity according to UVB irradiation intensity. The survival rate of the HaCaT cells was about 86% at 30 mJ/cm2 and 75% at 40 mJ/cm2 (Figure S1). As shown in Figure 2C, UVB irradiation 6 of 14 Nutrients 2019, 11, 1341 signiﬁcantly reduced the cell viability, while pretreatment of both SSW and SSE alleviated the harmful eﬀect of radiation to the cells. Upon UVB irradiation, the cell viability was reduced by 18% whereas pretreatment with both SSW and SSE restored the cell survivability. Surprisingly, at a concentration of 30 µg/mL of SSW and SSE, both extracts completely abolished the eﬀects of UV-induced cell damage (Figure 2C). Nutrients 2019, 11, x FOR PEER REVIEW 6 of 15 Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory effects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. 3.3. The Eﬀet of SS on Cell Viability of Human Keratinocytes (HaCaT) Cells Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. The data are denoted as the mean ± SD from triplicate results. (# p ≤ 0.01, compared with the non treated control (NT); ** p ≤ 0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in materials and methods EGCG: (-)-epigallocatechin gallate; GA: gallic acid Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory eﬀects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤0.01, * p ≤0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. The data are denoted as the mean ± SD from triplicate results. (# p ≤0.01, compared with the non treated control (NT); ** p ≤0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in materials and methods. EGCG: (−)-epigallocatechin gallate; GA: gallic acid. Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory effects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. 3.3. The Eﬀet of SS on Cell Viability of Human Keratinocytes (HaCaT) Cells Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. The data are denoted as the mean ± SD from triplicate results. (# p ≤ 0.01, compared with the non treated control (NT); ** p ≤ 0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in materials and methods EGCG: (-)-epigallocatechin gallate; GA: gallic acid Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory eﬀects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤0.01, * p ≤0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. The data are denoted as the mean ± SD from triplicate results. (# p ≤0.01, compared with the non treated control (NT); ** p ≤0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in materials and methods. EGCG: (−)-epigallocatechin gallate; GA: gallic acid. 3.3. The Eﬀet of SS on Cell Viability of Human Keratinocytes (HaCaT) Cells The data are denoted as the mean ± SD from triplicate results. (# p ≤ 0.01, compared with the non treated control (NT); ** p ≤ 0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in l d h d EGCG ( ) ll h ll GA ll d Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory eﬀects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤0.01, * p ≤0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were treated with SSW and SSE at 3, 10, or 30 µg/mL. After 24 h, UVB (40 mJ/cm2) was irradiated and the cells were cultured in serum-free medium for 24 h. The cell viability was measured using the MTT assay. The data are denoted as the mean ± SD from triplicate results. (# p ≤0.01, compared with the non treated control (NT); ** p ≤0.01, * p ≤ 0.05 compared with the UV control group). (D) Reactive oxygen species (ROS) generation activity was measured using HaCaT cells according to the method described in materials and methods. EGCG: (−)-epigallocatechin gallate; GA: gallic acid. Figure 2. Comparison of anti-aging potential by SS. (A) Inhibitory effects of SS stem extracts on elastase activity. The results are denoted as the mean ± standard deviation (SD) from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). (B) Cytotoxicity of SS stem extracts in HaCaT cells. Cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with aqueous and ethanolic extracts of SS stem (SSW and SSE) at 1, 3, 10, 30, 100, and 300 µg/mL. The cell viability was determined using 3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT) assay. (C) Phototoxicity of SS stem extracts in HaCaT cells. ( ) p g g g 3 3 Th Eff t f SS C ll Vi bilit f H K ti t (H C T) C ll 3.4. Suppression of UVB-Stimulated Reactive Oxygen Species (ROS) Generation by SS ( ) p g g g 3 3 Th Eff t f SS C ll Vi bilit f H K ti t (H C T) C ll 3.4. Suppression of UVB-Stimulated Reactive Oxygen Species (ROS) Generation by SS 3.3. The Effet of SS on Cell Viability of Human Keratinocytes (HaCaT) Cells Cell viability was determined using the MTT assay in HaCaT cells (Figure 2B). The cell viability of approximately 80% was considered to be non-toxic. SSW did not show any toxicity up to a concentration of 30 µg/mL, but the survival rate was decreased to 68% at a concentration of 100 µg/mL. On the other hand, SSE showed no cell toxicity up to 100 µg/mL, but the survival rate was approximately 68% at a concentration of 300 µg/mL. Therefore, 30 µg/mL was considered the highest i f h b i f b h l The goal was to investigate whether SSW and SSE suppress the UVB-stimulated ROS production. As described in Figure 2D, UVB irradiation signiﬁcantly enhanced ROS generation compared with the non-irradiated cells. Pretreatment of SSW and SSE signiﬁcantly lessened ROS generation compared with the UV-irradiated control. At a concentration of 30 µg/mL, both SSW and SSE decreased ROS production about 11% and 18%, respectively. On the other hand, both gallic acid and EGCG suppressed almost 50% of UVB stimulated ROS generation at 50 µg/mL. Nutrients 2019, 11, 1341 3.5. Regulation of In order to 7 of 14 aT 7 of 14 aT 3.5. Regulation of UVB-Induced Matrix Metalloproteinases (MMPs) Expression by SS e s e e p e ea e y SSE o o e y su je e o e pose i U ( 0 J/ ), a of MMPs was determined by RT-PCR and immunoblotting assay. As described in F 3.5. Regulation of UVB-Induced Matrix Metalloproteinases (MMPs) Expression by SS p y y j p ( J ) of MMPs was determined by RT-PCR and immunoblotting assay. As described in F 3.5. Regulation of UVB-Induced Matrix Metalloproteinases (MMPs) Expression by SS p y y j p ( ) of MMPs was determined by RT-PCR and immunoblotting assay. As described in F In order to conﬁrm the capability of SSE to regulate UVB-stimulated MMPs expression, HaCaT cells were pretreated by SSE followed by subjected to expose in UVB (40 mJ/cm2), and the expression of MMPs was determined by RT-PCR and immunoblotting assay. ( ) p g g g 3 3 Th Eff t f SS C ll Vi bilit f H K ti t (H C T) C ll 3.4. Suppression of UVB-Stimulated Reactive Oxygen Species (ROS) Generation by SS As described in Figure 3A,B, pretreatment of SSE signiﬁcantly abolished the UVB-stimulated MMPs mRNA level in a dose-dependent fashion. Interestingly, the suppressive eﬀect of SSE on MMP-1 and -2 expression was higher than that of EGCG used as a positive control (Figure 3B). Furthermore, the results show that pretreatment with both SSW and SSE was signiﬁcantly reduced in UV-induced MMP-12 transcriptional levels in a concentration-dependent fashion (Figure 3A). pretreatment of SSE significantly abolished the UVB-stimulated MMPs mRNA level in a dose- dependent fashion. Interestingly, the suppressive effect of SSE on MMP-1 and -2 expression was higher than that of EGCG used as a positive control (Figure 3B). Furthermore, the results show that pretreatment with both SSW and SSE was significantly reduced in UV-induced MMP-12 transcriptional levels in a concentration-dependent fashion (Figure 3A). Furthermore, to confirm whether SSE could regulate the TIMPs expression in UVB-stimulated HaCaT cells, both RT-PCR and western blotting assay were carried out. As described in Figure 3C and D, UVB treatment downregulated the TIMP expression compared to non-treated cells, while SSE pretreatment further halted this action. Figure 3. Effects of aging-related biomarkers by SSE. The effects of SSE on the regulation of (A) the mRNA expression and (B) the protein level of MMPs; (C) the mRNA expression and (D) the protein level of tissue inhibitors of matrix metalloproteinases (TIMPs); (E) the mRNA expression and (F) the protein level of elastin (ELN), type I collagen (COL1A1) and hyaluronan synthase 2 (HAS2) induced by UVB (40 mJ/cm2) in HaCaT cells. Cells were cultured for 24 h, and then with the indicated concentration of SSW and SSE for further 12 h. Then the cells were irradiated with 40 mJ/cm2 UVB Figure 3. Eﬀects of aging-related biomarkers by SSE. The eﬀects of SSE on the regulation of (A) the mRNA expression and (B) the protein level of MMPs; (C) the mRNA expression and (D) the protein level of tissue inhibitors of matrix metalloproteinases (TIMPs); (E) the mRNA expression and (F) the protein level of elastin (ELN), type I collagen (COL1A1) and hyaluronan synthase 2 (HAS2) induced by UVB (40 mJ/cm2) in HaCaT cells. Cells were cultured for 24 h, and then with the indicated concentration of SSW and SSE for further 12 h. Then the cells were irradiated with 40 mJ/cm2 UVB and cultured for additional 12 h. ( ) p g g g 3 3 Th Eff t f SS C ll Vi bilit f H K ti t (H C T) C ll 3.4. Suppression of UVB-Stimulated Reactive Oxygen Species (ROS) Generation by SS Reverse transcription-polymerase chain reaction (RT-PCR) and western blot was carried out according to the methods described in materials and methods. The data are denoted as the mean ± SD from triplicate results. (# p ≤0.01, compared with the NT; p ≤0.05, compared with the ultraviolet (UV) control group). EGCG: (−)-epigallocatechin gallate used as a positive control. Figure 3. Effects of aging-related biomarkers by SSE. The effects of SSE on the regulation of (A) the mRNA expression and (B) the protein level of MMPs; (C) the mRNA expression and (D) the protein level of tissue inhibitors of matrix metalloproteinases (TIMPs); (E) the mRNA expression and (F) the protein level of elastin (ELN), type I collagen (COL1A1) and hyaluronan synthase 2 (HAS2) induced by UVB (40 mJ/cm2) in HaCaT cells. Cells were cultured for 24 h, and then with the indicated concentration of SSW and SSE for further 12 h. Then the cells were irradiated with 40 mJ/cm2 UVB Figure 3. Eﬀects of aging-related biomarkers by SSE. The eﬀects of SSE on the regulation of (A) the mRNA expression and (B) the protein level of MMPs; (C) the mRNA expression and (D) the protein level of tissue inhibitors of matrix metalloproteinases (TIMPs); (E) the mRNA expression and (F) the protein level of elastin (ELN), type I collagen (COL1A1) and hyaluronan synthase 2 (HAS2) induced by UVB (40 mJ/cm2) in HaCaT cells. Cells were cultured for 24 h, and then with the indicated concentration of SSW and SSE for further 12 h. Then the cells were irradiated with 40 mJ/cm2 UVB and cultured for additional 12 h. Reverse transcription-polymerase chain reaction (RT-PCR) and western blot was carried out according to the methods described in materials and methods. The data are denoted as the mean ± SD from triplicate results. (# p ≤0.01, compared with the NT; p ≤0.05, compared with the ultraviolet (UV) control group). EGCG: (−)-epigallocatechin gallate used as a positive control. Furthermore, to conﬁrm whether SSE could regulate the TIMPs expression in UVB-stimulated HaCaT cells, both RT-PCR and western blotting assay were carried out. As described in Figure 3C,D, UVB treatment downregulated the TIMP expression compared to non-treated cells, while SSE pretreatment further halted this action. 3.7. Downregulation of NF-κB and AP-1 by SSE p 3 7 Downregulation of NF-κB and AP-1 by SSE A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was confirmed by western blot. (C) Specific extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the effects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above- mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤ 0.05, compared with the NT; p ≤ 0.05, compared with the ultraviolet (UV) control group). EGCG: (-)- epigallocatechin gallate. g g p ( )/ pp ( ) AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The eﬀect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was conﬁrmed by western blot as described in materials and methods. (B) The eﬀects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was conﬁrmed by western blot. (C) Speciﬁc extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the eﬀects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above-mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤0.05, compared with the NT; p ≤0.05, compared with the ultraviolet (UV) control group). EGCG: (−)-epigallocatechin gallate. p g g 3.8. Eﬀects of SSE on the Phosphorylation of Mitogen-Activated Protein Kinase (MAPK) Proteins 3.8. Eﬀects of SSE on the Phosphorylation of Mitogen-Activated Protein Kinase (MAPK) Proteins Next, we investigated the pathway through which SSE exhibits its anti-photoaging eﬀects. Generally, UVB induced augmented ROS production leads to the activation of MAPK proteins including extracellular signal–regulated kinases (ERK), p38 mitogen-activated protein kinases (p38) and c-Jun N-terminal kinases (JNK). MAPK induces NF-κB and AP-1 which consequently boosts the expression of MMPs, lead to decrease of collagen and other ECM in aged skin [30]. To investigate the eﬀects of SSE on UVB-induced photoaging, the phosphorylation of MAPKs was assessed. 3.7. Downregulation of NF-κB and AP-1 by SSE p 3 7 Downregulation of NF-κB and AP-1 by SSE 3.7. Downregulation of NF-κB and AP-1 by SSE 3 7 Downregulation of NF κB and AP 1 by SSE It is well known that the transcriptional levels of various MMPs are strongly modulated by NF-κB and AP-1. They also play signiﬁcant roles in maintaining the ECM composition as well as in cytokine expression [29]. Because SSE stiﬂed the expression of MMPs, we next examined whether SSE was able to regulate the transcription factors of NF-κB and AP-1 which is responsible for MMPs expression. To estimate the modiﬁcations of the expressions of NF-κB and AP-1 transcription factors induced by SSE in UVB-treated HaCaT cells, we turned to immunoblotting. The protein levels of NF-κB family members, p65, were signiﬁcantly raised by UVB but lessened by SSE in a concentration dependent fashion (Figure 4A, upper layer). Likewise, SSE also considerably reduced the protein levels of p-c-Jun, which are the components of AP-1 (Figure 4A, lower layer). g f y It is well known that the transcriptional levels of various MMPs are strongly modulated by NF- κB and AP-1. They also play significant roles in maintaining the ECM composition as well as in cytokine expression [29]. Because SSE stifled the expression of MMPs, we next examined whether SSE was able to regulate the transcription factors of NF-κB and AP-1 which is responsible for MMPs expression. To estimate the modifications of the expressions of NF-κB and AP-1 transcription factors induced by SSE in UVB-treated HaCaT cells, we turned to immunoblotting. The protein levels of NF- κB family members, p65, were significantly raised by UVB but lessened by SSE in a concentration dependent fashion (Figure 4A, upper layer). Likewise, SSE also considerably reduced the protein levels of p-c-Jun, which are the components of AP-1 (Figure 4A, lower layer). Figure 4 Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The effect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was confirmed by western blot as described in materials and methods. (B) The effects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. 3.6. Eﬀects of SSE on the Expression of COL1A1, ELN and HAS2 Skin wrinkles and elasticity are mainly inﬂuenced by collagen, elastin, and hyaluronic acid. Therefore, to examine whether the SS stem extracts had the ability to regulate these biomarkers, the transcriptional and translational expression of elastin (ELN), type I collagen (COL1A1) and hyaluronan synthase 2 (HAS2) was conﬁrmed by RT-PCR and immunoblotting, respectively (Figure 3E,F). The Nutrients 2019, 11, 1341 Therefore, to ex transcriptional 8 of 14 he nd results show that UVB treatment drastically downregulated the transcriptional and translational level of elastin, type I collagen, and hyaluronan synthase 2, while pretreatment of SSE ﬁxed their expression. y y ( ) y g p y ( g 3E,F). The results show that UVB treatment drastically downregulated the transcriptional and translational level of elastin, type I collagen, and hyaluronan synthase 2, while pretreatment of SSE fixed their expression. 3.7. Downregulation of NF-κB and AP-1 by SSE p 3 7 Downregulation of NF-κB and AP-1 by SSE (B) The effects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was confirmed by western blot. (C) Specific extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the effects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above- mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤ 0.05, compared with the NT; p ≤ 0.05, compared with the ultraviolet (UV) control group). EGCG: (-)- epigallocatechin gallate. Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The eﬀect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was conﬁrmed by western blot as described in materials and methods. (B) The eﬀects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was conﬁrmed by western blot. (C) Speciﬁc extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the eﬀects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above-mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤0.05, compared with the NT; p ≤0.05, compared with the ultraviolet (UV) control group). EGCG: (−)-epigallocatechin gallate. Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The effect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was confirmed by western blot as described in materials and methods. (B) The effects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. 3.7. Downregulation of NF-κB and AP-1 by SSE p 3 7 Downregulation of NF-κB and AP-1 by SSE A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was confirmed by western blot. (C) Specific extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the effects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above- mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤ 0.05, compared with the NT; p ≤ 0.05, compared with the ultraviolet (UV) control group). EGCG: (-)- epigallocatechin gallate. Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The eﬀect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was conﬁrmed by western blot as described in materials and methods. (B) The eﬀects of SSE on the phosphorylation of mitogen-activated protein (MAP) kinase activated by UVB. A 12 h pretreated cells by SSE, UVB was irradiated at 40 mJ/cm2, and the protein was extracted after 30 min. Protein expression was conﬁrmed by western blot. (C) Speciﬁc extracellular signal–regulated kinases (ERK) and p38 inhibitors (U0126 and SB239063, respectively) manifest the eﬀects of SSE on the regulation of NF-κB and AP-1 expression in UVB-irradiated HaCaT cells. Cell were treated according to the above-mentioned protocol and protein expression was evaluated by immunoblotting. (# p ≤0.05, compared with the NT; p ≤0.05, compared with the ultraviolet (UV) control group). EGCG: (−)-epigallocatechin gallate. Figure 4 Modulation of mitogen activated protein kinase (MAPK)/nuclear factor kappa B (NF κB) and Figure 4 Modulation of mitogen activated protein kinase (MAPK)/nuclear factor kappa B (NF κB) Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and Figure 4. Modulation of mitogen-activated protein kinase (MAPK)/nuclear factor-kappa B (NF-κB) and AP-1 signaling by SSE in UVB-stimulated HaCaT cells. (A) The effect of SSE on the phosphorylation of NF-κB, p-c-Jun and c-Jun modulated by UVB. Cells were cultured for 24 h, and then SSE were added for 12 h. After that, UVB was irradiated at 40 mJ/cm2, and the protein expression was confirmed by western blot as described in materials and methods. 3.9. Combination Studies of the Major Components of SSE on Elastase Inhibition Activity The major compounds of SSE identiﬁed were syringic acid (SA), epicatechin (EP) and vanillic acid (VA), and all of them showed dose-dependent moderate elastase inhibitory activity (Figure 5A). These results suggest that SA, EP and VA could inhibit elastase activity by diﬀerent pathways and that combination of these agents might result in boosted inhibitory eﬀects. Thus, we compared the inhibitory potencies of SA, EP and VA and their combinations (SA and EP; VA and EP, SA and VA at 1:9, 1:4, 1:2 and 1:1 ratio as well as SA, VA and EP at 1:1:1, 2:1:1, 1:2:1 and 1:1:2 ratio) on elastase inhibition activity. Combinations of equimolar doses of SA and EP showed nearly additive eﬀects (CI = 0.97) while other all combination doses showed slight antagonism (Figure 5B). As described in Figure 5C, VA and EP combinations also showed nearly additive eﬀects in all combinations (CI = 0.99, 0.91 and 0.99 at 1:4, 1:2 and 1:1 ratio, respectively) except 1:9 (VA:EP) ratio. These results suggested that the lower concentration of EP showed additive eﬀects by combining with SA and/or VA. Moreover, combination of SA and VA at 1:2, 1:1, and 2:1 ratio showed very good synergistic eﬀects (CI = 0.41, 0.22 and 0.37, respectively) on elastase inhibition (Figure 5D). Furthermore, strong synergism was observed at 2:1:1 molar ratio of SA, VA and EP (CI = 0.28), while other combinations of these three compounds also gave a moderate synergism eﬀect (Figure 5E). Our experiments demonstrated that only lower concentrations of EP with SA and/or VA had additive eﬀects, while combination of SA, VA and EP at lower concentrations was potently synergistic. Nutrients 2019, 11, x FOR PEER REVIEW 10 of 15 Figure 5. Individual and combination effects of elastase inhibition by the major identified compounds of SSE. (A) Elastase inhibition activity of the major identified compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). (B–E) Combination effects of elastase inhibition by the major identified compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive effect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Figure 5. 3.9. Combination Studies of the Major Components of SSE on Elastase Inhibition Activity Individual and combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. (A) Elastase inhibition activity of the major identiﬁed compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments. (B–E) Combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive eﬀect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Figure 5 Individual and combination effects of elastase inhibition by the major identified compounds Figure 5. Individual and combination eﬀects of elastase inhibition by the major identiﬁed compounds Figure 5. Individual and combination effects of elastase inhibition by the major identified compounds of SSE. (A) Elastase inhibition activity of the major identified compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). (B–E) Combination effects of elastase inhibition by the major identified compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive effect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Figure 5. Individual and combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. (A) Elastase inhibition activity of the major identiﬁed compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments. (B–E) Combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive eﬀect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Figure 5. Individual and combination effects of elastase inhibition by the major identified compounds of SSE. (A) Elastase inhibition activity of the major identified compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments (** p ≤ 0.01, * p ≤ 0.05 compared with the control group). 3.7. Downregulation of NF-κB and AP-1 by SSE p 3 7 Downregulation of NF-κB and AP-1 by SSE UVB-irradiation signiﬁcantly ampliﬁed the phosphorylation of ERK, and p38, compared with non-irradiated cells and peak at 30 min after UVB exposure (Figure S2), while the phosphorylation of JNK was minimal. Our data are consistent with studies by Chouinard et al. where it was shown that UVB failed to activate JNK, albeit the induction was more pronounced for the phosphorylation of ERK1/2 and p38 [31]. Treatment with SSE at 30 µg/mL provided the most inhibition on phosphorylated ERK1/2 and p38 (Figure 4B) whereas the inhibition eﬀects of SSE on the phosphorylation of JNK was absent. Thus, to validate whether SSE-modulated downregulation of NF-κB and AP-1 is associated with the MAPK 9 of 14 Nutrients 2019, 11, 1341 signaling cascade, cells were treated with speciﬁc p38 and ERK1/2 inhibitors, such as SB239063 and U0126, respectively, before being treated with SSE. Both p38 and ERK1/2 inhibition, along with SSE, exhibited the potential to attenuate the protein expression of NF-κB and AP-1 (Figure 4C). These data show that the suppression of UVB-stimulated ERK and p38 phosphorylation by SSE are required in the attenuation of NF-κB and AP-1 in HaCaT cells. 4. Discussion Epidemiological studies revealed that overexposure of UV irradiation signiﬁcantly increases the magnitude of patients with skin damage. Chronical exposure of UV irradiation to skin causes oxidative stress, inﬂammation, ROS-mediated DNA damage, and disorder of cellular signaling pathways, resulting in accelerated skin photo-aging [32,33]. Botanicals possessing antioxidant, anti-inﬂammatory and immunomodulatory properties are promising to be exploited as therapeutic agents for a variety of skin disorders, including photo-aging [34,35]. Spatholobus suberectus (SS) stem was found to contain various polyphenolics, such as gallic acid, vanillic acid (VA), epicatechin (EP), and syringic acid (SA) (Figure 1D). Among them, gallic acid has been reported to be eﬀective for anti-aging in human ﬁbroblasts through wound healing [36], and syringic acid has been reported to be eﬀective for elastase inhibition [37]. Indeed, various photochemopreventive agents derived from dietary origin have shown the potential eﬃcacy on skin delivery through oral systemic administration, an emerging concept referred to as ‘nutritional’ or ‘systemic photoprotection’. The systemic administration of an apocarotenoid bixin, an FDA-approved natural food colorant from the seeds of the achiote tree (Bixa orellana), protect the solar UV-induced skin damage in SKH-1 mice through the activation of cutaneous Nrf-2 signaling [38]. In this study, the aqueous and ethanolic extracts of SS stem’s (SSW and SSE, respectively) signiﬁcantly inhibited elastase activity (Figure 2A), possibly due to the presence of SA, VA and EP, which may cause synergism on elastase inhibition (Figure 5). Furthermore, HaCaT cells, which are human epidermal keratinocytes, were considered to prove the underlining mechanism of the anti-photoaging activity of SSE. Before experimentation, the toxicity of the extracts to the cells was conﬁrmed. Both SSW and SSE were found to be non-toxic in HaCaT cells up to a concentration of 30 µg/mL (Figure 2B). Therefore, the following concentrations were used in subsequent experiments. MMPs play an important role in the physiological mechanisms of skin photo-aging. UV irradiation alters the connective tissues of the skin by up-regulating the expression of MMPs, which degrade collagen and other ECM proteins. SSE has been shown to suppress UVB-induced upregulation of MMP-1 and -2 (Figure 3A,B). MMP-1 is a major collagenase that causes collagen degradation in skin severely damaged by UV. MMP-12 is the most important elastin-degrading enzyme and is involved with elastin and many other substrates, such as ECM collagen [39]. 3.9. Combination Studies of the Major Components of SSE on Elastase Inhibition Activity (B–E) Combination effects of elastase inhibition by the major identified compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive effect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Figure 5. Individual and combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. (A) Elastase inhibition activity of the major identiﬁed compounds of SSE. The results are denoted as the mean ± SD from triplicate experiments. (B–E) Combination eﬀects of elastase inhibition by the major identiﬁed compounds of SSE. Combination index (CI) values of <1, =1, and >1 indicate synergism, additive eﬀect, and antagonism, respectively. CI values shown are mean ± SE with a minimum of three experiments. SA: syringic acid; EP: epicatechin and VA: vanillic acid. The raw compuSyn report are given in supplementary data. Nutrients 2019, 11, 1341 10 of 14 10 of 14 4. Discussion The tissue inhibitors of metalloproteinases or TIMPs suppress MMP activity critical for extracellular matrix turnover associated with both physiologic and pathologic tissue remodeling. TIMP concentrations generally far exceed the concentration of MMPs in tissue and extracellular ﬂuids, thereby limiting their proteolytic activity to focal pericellular sites [40]. Treatment of SSE signiﬁcantly boosted the TIMP-1 and -2 expression than that of UVB-stimulated cells (Figure 3C,D) Notably, collagen is the principal protein that connects skin tissue, and the decomposition of collagen produces wrinkles in the skin. COL1A1 is one of the genes that constitute type I collagen and decreases upon exposure of UV [5,6]. Elastin is an extracellular matrix protein that binds to collagen and provides elasticity to the skin or elastic tissue. It is reduced by the activation of elastase upon UVB exposure [41]. When collagen and elastin are combined, hyaluronic acid plays a supporting role. Hyaluronan synthase 2 (HAS2) is an enzyme that synthesizes hyaluronic acid and plays a role in supporting the structure of the skin. In addition, HAS2 is abolished upon UVB skin irradiation [42]. SSE treatment signiﬁcantly hindered the UVB-induced downregulation of type I collagen, elastin and HAS2 expression in concentration dependent manner (Figure 3E,F), which also supports the result that treatment with SSE increases cell proliferation. UV irradiation stimulates cell surface growth factor receptors, cytokines, and MAPKs, which in turn regulate AP-1. Increased AP-1 activity down-regulates type I pro-collagen and up-regulates MMP-1. MMP-1 gene expression is regulated by c-Jun and c-Fos, which are components of the AP-1 heterodimer complex [33]. Moreover, it has been well established that a transcription factor, nuclear factor κB (NF-κB), is activated in skin keratinocytes by UV irradiation and leads to the augmentation of the levels of MMP-1, thus suppression of NF-κB pathway would freeze the UVB-induced skin photoaging process [43]. Interestingly, SSE treatment signiﬁcantly suppressed UV-induced c-Jun 11 of 14 Nutrients 2019, 11, 1341 expression (Figure 4B), which may inhibit AP-1 activity as well as NF-κB signaling cascade resulting in suppression of MMPs expression. Our results are also in accordance with the previous studies [44], in which the active parthenolide, a sesquiterpene lactone compound of Tanacetum parthenium, completely blocked the NF-κB signaling pathway and hindered the UVB-mediated cutaneous alteration in both cultured cell and animal model. Furthermore, Terminalia catappa water extract protected skin from photodamage by inhibiting the MAPK/AP-1/MMP pathway [45]. 4. Discussion Cumulating evidence has shown that exposure of the skin to UV induces ROS generation triggers MAPKs (ERK, JNK, and p38) phosphorylation, NF-κB, AP-1 activation, and regulates the expression of genes and proteins such as MMPs, leading to collagen degradation, resulting in photodamage and photo-carcinogenesis [41,43]. In the present study, UVB stimulation upregulated the phosphorylation of MAPKs, AP-1, NF-κB and MMPs, while SSE signiﬁcantly suppressed these eﬀects (Figure 4). Pharmacological inhibition of these signaling cascades abolished SSE-induced nuclear accumulation of AP-1, and NF-κB (Figure 4C). This ﬁnding suggests that SSE activity is dependent on this signaling transduction. The possible mechanisms of action of the Spatholobus suberectus stem extract against skin photoaging are summarized in Figure 6. Nutrients 2019, 11, x FOR PEER REVIEW 12 of 15 Figure 6. A proposed mechanism of SSE against UVB-induced photo-aging. Figure 6. A proposed mechanism of SSE against UVB-induced photo-aging. Figure 6. A proposed mechanism of SSE against UVB-induced photo-aging. Figure 6. A proposed mechanism of SSE against UVB-induced photo-aging. References 1. Kammeyer, A.; Luiten, R.M. Oxidation events and skin aging. Ageing Res. Rev. 2015, 21, 16–29. [CrossRef] [PubMed] 1. Kammeyer, A.; Luiten, R.M. Oxidation events and skin aging. Ageing Res. Rev. 2015, 21, 16–29. [CrossRef] [PubMed] 2. Cela, E.M.; Friedrich, A.; Paz, M.L.; Vanzulli, S.I.; Leoni, J.; Gonzalez Maglio, D.H. Time-course study of diﬀerent innate immune mediators produced by UV-irradiated skin: Comparative eﬀects of short and daily versus a single harmful UV exposure. Immunology 2015, 145, 82–93. [CrossRef] [PubMed] 2. Cela, E.M.; Friedrich, A.; Paz, M.L.; Vanzulli, S.I.; Leoni, J.; Gonzalez Maglio, D.H. Time-course study of diﬀerent innate immune mediators produced by UV-irradiated skin: Comparative eﬀects of short and daily versus a single harmful UV exposure. Immunology 2015, 145, 82–93. [CrossRef] [PubMed] 3. Lawrence, K.P.; Douki, T.; Sarkany, R.P.E.; Acker, S.; Herzog, B.; Young, A.R. The UV/visible radiation boundary region (385–405 nm) damages skin cells and induces “dark” cyclobutane pyrimidine dimers in human skin in vivo. Sci. Rep. 2018, 8, 12722. [CrossRef] [PubMed] 3. Lawrence, K.P.; Douki, T.; Sarkany, R.P.E.; Acker, S.; Herzog, B.; Young, A.R. The UV/visible radiation boundary region (385–405 nm) damages skin cells and induces “dark” cyclobutane pyrimidine dimers in human skin in vivo. Sci. Rep. 2018, 8, 12722. [CrossRef] [PubMed] 4. Tewari, A.; Grys, K.; Kollet, J.; Sarkany, R.; Young, A.R. Upregulation of MMP12 and its activity by UVA1 in human skin: Potential implications for photoaging. J. Investig. Dermatol. 2014, 134, 2598–2609. [CrossRef] [PubMed] 5. Varani, J.; Spearman, D.; Perone, P.; Fligiel, S.E.; Datta, S.C.; Wang, Z.Q.; Shao, Y.; Kang, S.; Fisher, G.J.; Voorhees, J.J. Inhibition of type I procollagen synthesis by damaged collagen in photoaged skin and by collagenase-degraded collagen in vitro. Am. J. Pathol. 2001, 158, 931–942. [CrossRef] 6. Catania, J.M.; Chen, G.; Parrish, A.R. Role of matrix metalloproteinases in renal pathophysiologies. Am. J. Physiol. Ren. Physiol. 2007, 292, F905–F911. [CrossRef] 7. Woessner, J.F. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 1991, 5, 2145–2154. [CrossRef] 8. Wlaschek, M.; Tantcheva-Poor, I.; Naderi, L.; Ma, W.; Schneider, L.A.; Razi-Wolf, Z.; Schuller, J.; Scharﬀetter-Kochanek, K. Solar UV irradiation and dermal photoaging. J. Photochem. Photobiol. B Biol. 2001, 63, 41–51. [CrossRef] 9. Fisher, G.J.; Datta, S.C.; Talwar, H.S.; Wang, Z.Q.; Varani, J.; Kang, S.; Voorhees, J.J. Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 1996, 379, 335–339. [CrossRef] 10. Kim, H.S.; Park, W.S.; Baek, J.I.; Lee, B.S.; Yoo, D.S.; Park, S.J. 5. Conclusions 5. Conclusions Acknowledgments: Md Badrul Alam and Ji-Hyun Park are supported by BK21Plus Creative Innovative Group for Leading Future Functional Food Industry, Kyungpook National University. Acknowledgments: Md Badrul Alam and Ji-Hyun Park are supported by BK21Plus Creative Innovative Group for Leading Future Functional Food Industry, Kyungpook National University. Acknowledgments: Md Badrul Alam and Ji-Hyun Park are supported by BK21Plus Creative Innovative Group for Leading Future Functional Food Industry, Kyungpook National University. Conﬂicts of Interest: The authors declare that there was no conﬂict of interest. Conﬂicts of Interest: The authors declare that there was no conﬂict of interest. Conﬂicts of Interest: The authors declare that there was no conﬂict of interest. Conﬂicts of Interest: The authors declare that there was no conﬂict of interest. 5. Conclusions 5. Conclusions In this study, SS stem extract inhibited the ROS production by UV as well as repressing UV- activated MAP kinase, and ultimately enhanced the expression of COL1A1, ELN and HAS2. The active components, which were identified by HPLC as gallic acid, catechin, vanillic acid, syringic acid, and epicatechin, not only affect protection from ROS and cellular damage, but also inhibition of MMPs, activation of COL1A1, ELN, and HAS2, eventually blocking the phosphorylation of MAPKs and its downstream transcription factor such as AP-1, and NF-κB. These results collectively suggest that Spatholobus suberectus stem extract seems to be valuable as a natural biomaterial that can In this study, SS stem extract inhibited the ROS production by UV as well as repressing UV-activated MAP kinase, and ultimately enhanced the expression of COL1A1, ELN and HAS2. The active components, which were identiﬁed by HPLC as gallic acid, catechin, vanillic acid, syringic acid, and epicatechin, not only aﬀect protection from ROS and cellular damage, but also inhibition of MMPs, activation of COL1A1, ELN, and HAS2, eventually blocking the phosphorylation of MAPKs and its downstream transcription factor such as AP-1, and NF-κB. These results collectively suggest that 12 of 14 Nutrients 2019, 11, 1341 Spatholobus suberectus stem extract seems to be valuable as a natural biomaterial that can inhibit UVB-stimulated photo-aging. Spatholobus suberectus stem extract seems to be valuable as a natural biomaterial that can inhibit UVB-stimulated photo-aging. Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/11/6/1341/s1, Figure S1: Cell viability UVB-irradiation. HaCaT cells (1 × 105 cells/mL) were seeded in a 96-well plate and treated with UVB-irradiation at 20, 30, 40 and 50 mJ/cm2. The cell viability was determined using the MTT assay, Figure S2: The eﬀects of UVB-irradiation on the phosphorylation of MAP kinase in HaCaT cells. HaCaT cells (1 × 105 cells/mL) were seeded in a 6-well plate and treated with UVBirradiation at 40 mJ/cm2 and the protein was extracted after 15, 30, 60 and 120 min Protein expression was conﬁrmed by western blot, Table S1: List of the primer sets used in this study, Table S2: List of antibodies used in this study. Author Contributions: K.-R.K. and J.-H.P. performed the experiments. K.-R.K., M.B.A., J.-H.P., T.-H.K. and S.-H.L. designed the research and analyzed the data. K.-R.K., M.B.A. and S.-H.L. wrote the paper. M.B.A. and S.-H.L. revised the paper. Funding: This research received no external funding. References 2014, 42, 1123–1138. [CrossRef] [PubMed] 22. Pang, J.; Guo, J.P.; Jin, M.; Chen, Z.Q.; Wang, X.W.; Li, J.W. Antiviral eﬀects of aqueous extract from Spatholobus suberectus Dunn. Against coxsackievirus B3 in mice. Chin. J. Integr. Med. 2011, 17, 764–769. [CrossRef] 23. Zhang, R.; Liu, C.; Liu, X.; Guo, Y. Protective eﬀect of Spatholobus suberectus on brain tissues in cerebral ischemia. Am. J. Transl. Res. 2016, 8, 3963–3969. 24. Brito, S.M.; Coutinho, H.D.; Talvani, A.; Coronel, C.; Barbosa, A.G.; Vega, C.; Figueredo, F.G.; Tintino, S.R.; Lima, L.F.; Boligon, A.A.; et al. Analysis of bioactivities and chemical composition of Ziziphus joazeiro Mart. using HPLC-DAD. Food Chem. 2015, 186, 185–191. [CrossRef] [PubMed] 25. Kraunsoe, J.A.; Claridge, T.D.; Lowe, G. Inhibition of human leukocyte and porcine pancreatic elastase by homologues of bovine pancreatic trypsin inhibitor. Biochemistry 1996, 35, 9090–9096. [CrossRef] [PubMed] 26. Chou, T.C. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism 25. Kraunsoe, J.A.; Claridge, T.D.; Lowe, G. Inhibition of human leukocyte and porcine pancreatic elastase by homologues of bovine pancreatic trypsin inhibitor. Biochemistry 1996, 35, 9090–9096. [CrossRef] [PubMed] homologues of bovine pancreatic trypsin inhibitor. Biochemistry 1996, 35, 9090–9096. [CrossRef] [PubMed] 26. Chou, T.C. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol. Rev. 2006, 58, 621–681. [CrossRef] [PubMed] 26. Chou, T.C. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol. Rev. 2006, 58, 621–681. [CrossRef] [PubMed] 27. Rigel, E.G.; Lebwohl, M.; Rigel, A.C.; Rigel, D.S. Daily UVB exposure levels in high-school students measured with digital dosimeters. J. Am. Acad. Dermatol. 2003, 49, 1112–1114. [CrossRef] 28. Zhao, P.; Alam, M.B.; Lee, S.H. Protection of UVB-induced photoaging by Fuzhuan-Brick tea aqueous extract via MAPKs/Nrf2-mediated down-regulation of MMP-1. Nutrients 2018, 11, 60. [CrossRef] 29. Bohm, M.; Schulte, U.; Kalden, H.; Luger, T.A. Alpha-melanocyte-stimulating hormone modulates activation of NF-κB and AP-1 and secretion of interleukin-8 in human dermal ﬁbroblasts. Ann. NY. Acad. Sci. 1999, 885, 277–286. [CrossRef] 30. Son, Y.; Cheong, Y.-K.; Kim, N.-H.; Chung, H.-T.; Kang, D.G.; Pae, H.-O. Mitogen-activated protein kinases and reactive oxygen species: How can ROS activate MAPK pathways? J. Signal Transduct. 2011, 2011, 792639. [CrossRef] 31. Chouinard, N.; Valerie, K.; Rouabhia, M.; Huot, J. UVB-mediated activation of p38 mitogen-activated protein kinase enhances resistance of normal human keratinocytes to apoptosis by stabilizing cytoplasmic p53. Biochem. J. 2002, 365, 133–145. [CrossRef] [PubMed] 32. References Continuous irradiation with a 633-nm light-emitting diode exerts an anti-aging eﬀect on human skin cells. Int. J. Mol. Med. 2015, 35, 383–390. [CrossRef] 11. Weston, C.R.; Davis, R.J. The JNK signal transduction pathway. Curr. Opin. Genet. Dev. 2002, 12, 14–21. [CrossRef] 12. Anggakusuma; Yanti; Hwang, J.K. Eﬀects of macelignan isolated from Myristica fragrans Houtt. on UVB-induced matrix metalloproteinase-9 and cyclooxygenase-2 in HaCaT cells. J. Dermatol. Sci. 2010, 57, 114–122. [CrossRef] [PubMed] 13. Kong, R.; Cui, Y.; Fisher, G.J.; Wang, X.; Chen, Y.; Schneider, L.M.; Majmudar, G. A comparative study of the eﬀects of retinol and retinoic acid on histological, molecular, and clinical properties of human skin. J. Cosmet. Dermatol. 2016, 15, 49–57. [CrossRef] [PubMed] 13 of 14 13 of 14 Nutrients 2019, 11, 1341 14. Varani, J.; Warner, R.L.; Gharaee-Kermani, M.; Phan, S.H.; Kang, S.; Chung, J.H.; Wang, Z.Q.; Datta, S.C.; Fisher, G.J.; Voorhees, J.J. Vitamin A antagonizes decreased cell growth and elevated collagen-degrading matrix metalloproteinases and stimulates collagen accumulation in naturally aged human skin. J. Investig. Dermatol. 2000, 114, 480–486. [CrossRef] [PubMed] 15. Zhang, S.; Xuan, L. New phenolic constituents from the stems of Spatholobus suberectus. Helv. Chim. Acta 2006, 89, 1241–1245. [CrossRef] 16. Lee, M.H.; Lin, Y.P.; Hsu, F.L.; Zhan, G.R.; Yen, K.Y. Bioactive constituents of Spatholobus suberectus in regulating tyrosinase-related proteins and mRNA in HEMn cells. Phytochemistry 2006, 67, 1262–1270. [CrossRef] 17. Chen, S.R.; Wang, A.Q.; Lin, L.G.; Qiu, H.C.; Wang, Y.T.; Wang, Y. In vitro study on anti-hepatitis C virus activity of Spatholobus suberectus Dunn. Molecules 2016, 21, 1367. [CrossRef] 18. Lee, B.J.; Jo, I.Y.; Bu, Y.; Park, J.W.; Maeng, S.; Kang, H.; Jang, W.; Hwang, D.S.; Lee, W.; Min, K.; et al. Antiplatelet eﬀects of Spatholobus suberectus via inhibition of the glycoprotein IIb/IIIa receptor. J. Ethnopharmacol. 2011, 134, 460–467. [CrossRef] 19. Peng, F.; Meng, C.W.; Zhou, Q.M.; Chen, J.P.; Xiong, L. Cytotoxic evaluation against breast cancer cells of isoliquiritigenin analogues from Spatholobus suberectus and their synthetic derivatives. J. Nat. Prod. 2016, 79, 248–251. [CrossRef] 20. Toyama, T.; Wada-Takahashi, S.; Takamichi, M.; Watanabe, K.; Yoshida, A.; Yoshino, F.; Miyamoto, C.; Maehata, Y.; Sugiyama, S.; Takahashi, S.S.; et al. Reactive oxygen species scavenging activity of Jixueteng evaluated by electron spin resonance (ESR) and photon emission. Nat. Prod. Commun. 2014, 9, 1755–1759. [CrossRef] 21. Im, N.K.; Lee, S.G.; Lee, D.S.; Park, P.H.; Lee, I.S.; Jeong, G.S. Spatholobus suberectus inhibits osteoclastogenesis and stimulates chondrogenesis. Am. J. Chin. Med. References Rabe, J.H.; Mamelak, A.J.; McElgunn, P.J.; Morison, W.L.; Sauder, D.N. Photoaging: Mechanisms and repair. J. Am. Acad. Dermatol. 2006, 55, 1–19. [CrossRef] [PubMed] 33. Quan, T.; Qin, Z.; Xia, W.; Shao, Y.; Voorhees, J.J.; Fisher, G.J. Matrix-degrading metalloproteinases in photoaging. J. Investig. Dermatol. Symp. Proc. 2009, 14, 20–24. [CrossRef] [PubMed] 34. Niu, T.; Tian, Y.; Ren, Q.; Wei, L.; Li, X.; Cai, Q. Red light interferes in UVA-induced photoaging of human skin ﬁbroblast cells. Photochem. Photobiol. 2014, 90, 1349–1358. [CrossRef] [PubMed] 14 of 14 Nutrients 2019, 11, 1341 14 of 14 35. Fisher, G.J.; Kang, S.; Varani, J.; Bata-Csorgo, Z.; Wan, Y.; Datta, S.; Voorhees, J.J. Mechanisms of photoaging and chronological skin aging. Arch. Dermatol. 2002, 138, 1462–1470. [CrossRef] [PubMed] 36. Yang, D.J.; Moh, S.H.; Son, D.H.; You, S.; Kinyua, A.W.; Ko, C.M.; Song, M.; Yeo, J.; Choi, Y.H.; Kim, K.W. Gallic acid promotes wound healing in normal and hyperglucidic conditions. Molecules 2016, 21, 899. [CrossRef] [PubMed] 37. Maity, N.; Nema, N.K.; Abedy, M.K.; Sarkar, B.K.; Mukherjee, P.K. Exploring Tagetes erecta Linn ﬂower for the elastase, hyaluronidase and MMP-1 inhibitory activity. J. Ethnopharmacol. 2011, 137, 1300–1305. [CrossRef] 38. Rojo de la Vega, M.; Krajisnik, A.; Zhang, D.D.; Wondrak, G.T. Targeting NRF2 for improved skin barrier function and photoprotection: Focus on the achiote-derived apocarotenoid bixin. Nutrients 2017, 9, 1371. [CrossRef] 39. Pittayapruek, P.; Meephansan, J.; Prapapan, O.; Komine, M.; Ohtsuki, M. Role of Matrix metalloproteinases in photoaging and photocarcinogenesis. Int. J. Mol. Sci. 2016, 17, 868. [CrossRef] 40. Sun, J. Matrix metalloproteinases and tissue inhibitor of metalloproteinases are essential for the inﬂammatory response in cancer cells. J. Signal Transduct. 2010, 2010, 985132. [CrossRef] 1. Ryu, J.; Park, S.J.; Kim, I.H.; Choi, Y.H.; Nam, T.J. Protective eﬀect of porphyra-334 on UVA-indu photoaging in human skin ﬁbroblasts. Int. J. Mol. Med. 2014, 34, 796–803. [CrossRef] [PubMed] 42. Wang, Y.; Lauer, M.E.; Anand, S.; Mack, J.A.; Maytin, E.V. Hyaluronan synthase 2 protects skin ﬁbroblasts against apoptosis induced by environmental stress. J. Biol. Chem. 2014, 289, 32253–32265. [CrossRef] [PubMed] 43. Tanaka, K.; Asamitsu, K.; Uranishi, H.; Iddamalgoda, A.; Ito, K.; Kojima, H.; Okamoto, T. Protecting skin photoaging by NF-κB inhibitor. Curr. Drug Metab. 2010, 11, 431–435. [CrossRef] [PubMed] 44. Kwok, B.H.; Koh, B.; Ndubuisi, M.I.; Elofsson, M.; Crews, C.M. The anti-inﬂammatory natural product parthenolide from the medicinal herb Feverfew directly binds to and inhibits I-κB kinase. Chem. Biol. 2001, 8, 759–766. [CrossRef] 45. References Huang, Y.-H.; Wu, P.-Y.; Wen, K.-C.; Lin, C.-Y.; Chiang, H.-M. Protective eﬀects and mechanisms of Terminalia catappa L. methenolic extract on hydrogen-peroxide-induced oxidative stress in human skin ﬁbroblasts. BMC Complement. Altern. Med. 2018, 18, 266. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4390665828	https://www.frontiersin.org/articles/10.3389/fonc.2023.1320867/pdf?isPublishedV2=False	English	null	Axillary management in patients with clinical node-negative early breast cancer and positive sentinel lymph node: a systematic review and meta-analysis	Frontiers in oncology	2,024	cc-by	9,073	OPEN ACCESS Changzai Li 1, Pan Zhang 2, Jie Lv 1, Wei Dong 1, Baoshan Hu 1, Jinji Zhang 1 and Hongcheng Zhu 1 1Department of Oncological Surgery, North China University of Science and Technology Afﬁliated Hospital, Tangshan, Hebei, China, 2College of Nursing and Rehabilitation, North China University of Science and Technology, Tangshan, Hebei, China CITATION Li C, Zhang P, Lv J, Dong W, Hu B, Zhang J and Zhu H (2024) Axillary management in patients with clinical node-negative early breast cancer and positive sentinel lymph node: a systematic review and meta-analysis. Front. Oncol. 13:1320867. d i 10 3389/f 2023 1320867 Background: The omission of axillary lymph node dissection (ALND) or axillary radiation (AxRT) remains controversial in patients with clinical node-negative early breast cancer and a positive sentinel lymph node. COPYRIGHT © 2024 Li, Zhang, Lv, Dong, Hu, Zhang and Zhu. 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. Methods: We conducted a comprehensive review by searching PubMed, Embase, Web of Science, and Cochrane databases (up to November 2023). Our primary outcomes were overall survival (OS), disease-free survival (DFS), locoregional recurrence (LRR), and axillary recurrence (AR). Results: We included 26 studies encompassing 145,548 women with clinical node- negative early breast cancer and positive sentinel lymph node. Pooled data revealed no signiﬁcant differences between ALND and sentinel lymph node biopsy (SLNB) alone in terms of OS (hazard ratio [HR]0.99, 95% conﬁdence interval [CI] 0.91-1.08, p=0.84), DFS (HR 1.04, 95% CI 0.90-1.19, p=0.61), LRR (HR 0.76, 95% CI 0.45-1.20, p=0.31), and AR (HR 1.01, 95% CI 0.99-1.03, p=0.35). Similarly, no signiﬁcant differences were observed between AxRT and SLNB alone for OS (HR 0.57, 95% CI 0.32-1.02, p=0.06) and DFS (HR 0.52, 95% CI 0.26-1.05, p=0.07). When comparing AxRT and ALND, a trend towards higher OS was observed the AxRT group (HR 0.08, 95% CI 0.67-1.15), but the difference did not reach statistical signiﬁcance (p=0.35, I2 = 0%). Additionally, no signiﬁcant differences signiﬁcance observed for DFS or AR (p=0.13 and p=0.73, respectively) between the AxRT and ALND groups. TYPE Systematic Review PUBLISHED 08 January 2024 DOI 10.3389/fonc.2023.1320867 TYPE Systematic Review PUBLISHED 08 January 2024 DOI 10.3389/fonc.2023.1320867 TYPE Systematic Review PUBLISHED 08 January 2024 DOI 10.3389/fonc.2023.1320867 breast cancer, sentinel lymph node biopsy, axillary lymph node dissection, axillary radiation, axillary management Axillary management in patients with clinical node-negative early breast cancer and positive sentinel lymph node: a systematic review and meta-analysis OPEN ACCESS EDITED BY Kevin Ni, St George Hospital Cancer Care Centre, Australia REVIEWED BY Laurence Gluch, The Strathﬁeld Breast Centre, Australia Ibrahim Mohamed Elzayat, Aswan University, Egypt CORRESPONDENCE Pan Zhang zhangpan0901@163.com RECEIVED 13 October 2023 ACCEPTED 13 December 2023 PUBLISHED 08 January 2024 CITATION Li C, Zhang P, Lv J, Dong W, Hu B, Zhang J and Zhu H (2024) Axillary management in patients with clinical node-negative early breast cancer and positive sentinel lymph node: a systematic review and meta-analysis. Front. Oncol. 13:1320867. doi: 10.3389/fonc.2023.1320867 Frontiers in Oncology KEYWORDS breast cancer, sentinel lymph node biopsy, axillary lymph node dissection, axillary radiation, axillary management OPEN ACCESS Conclusion: Our ﬁndings suggest that survival and recurrence rates are not inferior in patients with clinical node-negative early breast cancer and a positive sentinel lymph node who receive SLNB alone compared to those undergoing ALND or AxRT. breast cancer, sentinel lymph node biopsy, axillary lymph node dissection, axillary radiation, axillary management 01 Frontiers in Oncology frontiersin.org Li et al. 10.3389/fonc.2023.1320867 Study inclusion and exclusion criteria The inclusion criterial were as follows: (1) Design: randomized controlled trials (RCTs) and retrospective studies. (2) Patient eligibility: Studies enrolling patients with clinical node-negative early breast cancer and positive sentinel lymph node (3) Comparative interventions: SLNB alone versus ALND, ALND versus AxRT, and SLNB alone versus AxRT. (4) Outcomes: Studies reporting on overall survival, disease-free survival, axillary recurrence, and locoregional recurrence. The exclusion criterial included abstracts, reviews, case reports, and articles deemed irrelevant or containing missing data. Data extraction and management Following the Cochrane Handbook guidelines, two authors independently extracted data from the included studies. Recorded information included the authors’ names, publication year, number of participants, study design, intervention type, tumor stage, micometastasis or macrometastsis count, adjuvant radiation therapy, follow-up duration (years), outcomes, and the quality of evidence in each study. Any discrepancies were addressed and resolved through discussion or with the assistance of a third author. The AMAROS trial aimed to evaluate whether axillary radiation (AxRT) achieved better regional control and fewer side effects compared to ALND. Finding demonstrated that AxRT offered similar axillary control for patients with T1/T2 breast cancer and positive sentinel lymph nodes, while signiﬁcantly reducing. the occurrence of lymphedema (10). A retrospective cohort study comparing patients with T1/T2 and less than two macrometastases (>2 mm) who underwent either AxRT or non-AxRT also observed similar overall and disease-free survival rates. There was no statistically signiﬁcant difference in the ﬁve-year outcomes between the two groups (11). Motivated by these ﬁnding, we conducted a systematic review to compare outcomes in clinical node-negative early breast cancer patients with sentinel lymph node metastasis who Study selection We conducted a systematic search of English literature in PubMed, Embase and Cochrane Library databases up to November 2023. Our search encompassed published data only. Search terms included keywords and MeSH terms such as “breast cancer/breast carcinoma”, “sentinel lymph node biopsy,” “axillary lymph node dissection”, and “axillary radiation,” Two authors (C.Z.L and P.Z) independently reviewed the available literature based on the inclusion criteria. Subsequently, potentially relevant references with sufﬁcient information in their titles and abstracts were retrieved full-text article assessment. If the included studies were based on the same data, we selected the latest published version. Any disagreements were resolved through discussion and consensus among the authors. The study selection process adhered to the PRISMA guidelines. However, Galimberti et al. suggest that ALND may be overtreatment for early-stage breast cancer patients, particularly when the tumor burden in the sentinel lymph nodes is minimal or moderate (5). NCCN also suggests that patients who have T1/T2 tumors, one to two positive sentinel lymph nodes, and plan to undergo whole-breast radiotherapy (RT) following breast- conserving therapy are not recommended for ALND (6). The ACOSOG Z0011 (American College of Surgeons Oncology Group) trial demonstrated that patients with clinical T1/T2 tumors and fewer than three positive sentinel lymph nodes undergoing lumpectomy and whole-breast radiation therapy could avoid ALND without negatively impacting local recurrence, disease-free survival, and overall survival (7).Additionally, the IBCSG 23-01 trial, designed to compare outcomes in patients with one or more sentinel micrometastases (≤2 mm) treated with ALND versus no ALND, showed no signiﬁcant differences in ﬁve- year overall survival and ﬁve-year disease-free survival (8). A previous retrospective study also indicated that ALND did not improve either post-mastectomy overall survival or disease-free survival among breast cancer patients with one to three positive sentinel lymph nodes (9). Frontiers in Oncology Introduction underwent mastectomy or breast-conserving surgery. This meta- analysis aims to assess overall survival, disease-free survival, locoregional recurrence and axillary recurrence according to the type of axillary management(SLNB alone, ALND, or AxRT). Axillary lymph node dissection (ALND) has been the standard therapeutic approach for breast cancer patients with positive sentinel lymph nodes. However, ALND is associated with various complications, including lymphedema, paresthesia, infections, axillary seromas, and other signiﬁcant morbidities (1). Currently, sentinel lymph node biopsy (SLNB) is recommended for assessing axillary nodal lymph node status in early breast cancer patients who are clinically node-negative. The National Comprehensive Cancer Network (NCCN) suggests that ALND is not required for breast cancer patients with a negative sentinel node. The role of ALND for early-stage breast cancer patients with a limited number of metastatic sentinel lymph nodes remains controversial. According to the American Society of Clinical Oncology Clinical Practice Guideline, ALND should be considered for women with early breast cancer and one to two positive sentinel lymph nodes who are planning to undergo mastectomy (2). Pepels et al. indicated that ALND was recommended in patients with sentinel micrometastases and unfavorable tumor characteristics (3), while no ALND for patients with sentinel lymph nodes micrometastases resulted in a higher ﬁve -year regional recurrence rate compared to ALND (4). However, Galimberti et al. suggest that ALND may be overtreatment for early-stage breast cancer patients, particularly when the tumor burden in the sentinel lymph nodes is minimal or moderate (5). NCCN also suggests that patients who have T1/T2 tumors, one to two positive sentinel lymph nodes, and plan to undergo whole-breast radiotherapy (RT) following breast- conserving therapy are not recommended for ALND (6). The ACOSOG Z0011 (American College of Surgeons Oncology Group) trial demonstrated that patients with clinical T1/T2 tumors and fewer than three positive sentinel lymph nodes undergoing lumpectomy and whole-breast radiation therapy could avoid ALND without negatively impacting local recurrence, disease-free survival, and overall survival (7).Additionally, the IBCSG 23-01 trial, designed to compare outcomes in patients with one or more sentinel micrometastases (≤2 mm) treated with ALND versus no ALND, showed no signiﬁcant differences in ﬁve- year overall survival and ﬁve-year disease-free survival (8). A previous retrospective study also indicated that ALND did not improve either post-mastectomy overall survival or disease-free survival among breast cancer patients with one to three positive sentinel lymph nodes (9). Introduction The AMAROS trial aimed to evaluate whether axillary radiation (AxRT) achieved better regional control and fewer side effects Axillary lymph node dissection (ALND) has been the standard therapeutic approach for breast cancer patients with positive sentinel lymph nodes. However, ALND is associated with various complications, including lymphedema, paresthesia, infections, axillary seromas, and other signiﬁcant morbidities (1). Currently, sentinel lymph node biopsy (SLNB) is recommended for assessing axillary nodal lymph node status in early breast cancer patients who are clinically node-negative. The National Comprehensive Cancer Network (NCCN) suggests that ALND is not required for breast cancer patients with a negative sentinel node. The role of ALND for early-stage breast cancer patients with a limited number of metastatic sentinel lymph nodes remains controversial. According to the American Society of Clinical Oncology Clinical Practice Guideline, ALND should be considered for women with early breast cancer and one to two positive sentinel lymph nodes who are planning to undergo mastectomy (2). Pepels et al. indicated that ALND was recommended in patients with sentinel micrometastases and unfavorable tumor characteristics (3), while no ALND for patients with sentinel lymph nodes micrometastases resulted in a higher ﬁve -year regional recurrence rate compared to ALND (4). frontiersin.org Quality assessment The NOS was used to assess the quality of evidence in non-RCT trials. Five studies received a rating of seven or more stars, indicating high quality (15, 16, 19, 21, 35).Seven studies received a rating of six stars (11, 20, 22, 23, 26, 34, 36), and six studies were assessed as having four to ﬁve stars (See Table 1). The GRADEpro GDT tool was used to classify the evidence of RCTs comparing ALND to SLNB alone for patients with clinical node-negative early- stage breast cancer and positive sentinel lymph nodes. The quality of evidence was high for disease-free survival and locoregional recurrence, and moderate in overall survival (See Table 2). Statistical analysis We used the Review Manager software (version 5.4), update by the Cochrane Library for Systematic Review, to perform the analysis. The summary statistic of generic inverse variance (overall survival, disease-free survival, axillary recurrence, and locoregional recurrence) was assessed by hazard ratios (HRs). 95% conﬁdence intervals (CIs) were calculated using the ﬁxed-effect model. The statistical heterogeneity of the included studies was quantiﬁed and examined using the I2 statistics. An I2 value of 0% to 25% indicates low heterogeneity, 25% to50% indicates moderate heterogeneity, 50% to75% indicates large heterogeneity, and 75% to100% indicates huge heterogeneity (14). When heterogeneity was observed, we employed the random-effects model. We conducted the subgroup analysis based on the type of axillary management (SLNB alone, ALND, and AxRT). Sensitivity analysis was performed to identify sources of heterogeneity. A funnel plot was used to assess publication bias in the included studies. A p value less than 0.05 was considered statistically signiﬁcant. Study characteristics Table 1 outlines the characteristics of the included studies. Eighteen studies were retrospective cohort studies (11, 18–27, 34– 36), while the remaining eight were RCTs (5, 7, 28–33). The studies were published between April 2009 to July 2023.Sample sizes range from 121 to 97,314 patients, with a total of 145,548 patients analyzed. Among these patients, 40,156 received SLNB alone, while 105,418 underwent either ALND or AxRT. All included studies were published in English. Twenty-two studies reported overall survival data (5, 7, 11, 15–27, 29, 30, 32, 33, 35). Eleven studies reported disease-free survival data (5, 7, 11, 15, 23, 30–35). Six studies analyzed locoregional recurrence (21, 26, 28, 35, 36), and eight studies reported axillary recurrence (11, 22, 24, 26, 30, 32, 33, 35). Effect of ALND versus SLNB alone Eighteen studies (5, 7, 15–27, 29, 32, 35) reported the overall survival data for patients with SLNB alone versus ALND. The pooled results revealed no statistically signiﬁcant difference between the groups (HR0.99, 95% CI0.91-1.08; p=0.84), and no signiﬁcant heterogeneity was observed (I2 = 30%, p=0.11) (Figure 2A). A potential publication bias was suggested by the the asymmetry of the funnel plot (Figure 3A), Subgroup analysis showed no statistically signiﬁcant difference in overall survival between SLNB alone and ALND in either RCTs (HR 1.09, 95% CI:0.92-1.28; p=0.33) or retrospective studies (HR 0.96, 95% CI:0.86-1.06; p=0.40). Data pooled from the eight studies (5, 7, 15, 20, 29, 31, 32, 35) demonstrated no substantial difference in disease-free survival between the SLNB alone group and ALND groups (HR 1.04, 95%CI:0.90-1.19; p=0.61). Additionally, the studies showed no signiﬁcant heterogeneity (I2 = 0%, p=0.73) (Figure 2B). The funnel plot suggested the presence of publication bias (Figure 3B). Subgroup analysis revealed no signiﬁcant difference in disease- free survival between RCTs (HR 1.03, 95%CI:0.89-1.19; p=0.72) or retrospective studies (HR 1.09, 95%CI:0.77-1.54; p=0.64). Quality assessment in individual studies The risk of bias in all included studies was evaluated using guidelines in the Cochrane Handbook for Systematic Reviews of 02 frontiersin.org Li et al. 10.3389/fonc.2023.1320867 Interventions (https://training.cochrane.org/handbooks) (12). Two authors independently assessed the potential risk of bias, including selection bias, performance bias, detection bias, attrition bias, reporting bias, and confounding bias, and other sources of bias. For RCTs, the GRADEpro GDT (Grading of Recommendation Assessment Development and Evaluation Proﬁler Guide line Development Tool)was used to assess evidence quality. This online too, available at https://www.gradepro.org, evaluates ﬁve factors: risk of bias, inconsistency, indirectness, imprecision, and other considerations. Based on these factors, evidence quality is classiﬁed into four levels: high (⊕⊕⊕⊕), moderate (⊕⊕⊕⊖), low (⊕⊕⊖⊖) or very low (⊕⊖⊖⊖). For non-RCTs, the Newcastle- Ottawa Scale (NOS) served as the quality assessment tool (13). The NOS awards stars based on three domains: quality of patient selection (up to four stars), comparability between cases and controls (up to two stars), and adequate ascertainment of exposure (up to three stars). Studies with more than seven stars were considered to have a high level of evidence. Two authors independently assessed the quality of evidence in the included studies, With any disagreements resolved through discussion. Interventions (https://training.cochrane.org/handbooks) (12). Two authors independently assessed the potential risk of bias, including selection bias, performance bias, detection bias, attrition bias, reporting bias, and confounding bias, and other sources of bias. For RCTs, the GRADEpro GDT (Grading of Recommendation Assessment Development and Evaluation Proﬁler Guide line Development Tool)was used to assess evidence quality. This online too, available at https://www.gradepro.org, evaluates ﬁve factors: risk of bias, inconsistency, indirectness, imprecision, and other considerations. Based on these factors, evidence quality is classiﬁed into four levels: high (⊕⊕⊕⊕), moderate (⊕⊕⊕⊖), low (⊕⊕⊖⊖) or very low (⊕⊖⊖⊖). For non-RCTs, the Newcastle- Ottawa Scale (NOS) served as the quality assessment tool (13). The NOS awards stars based on three domains: quality of patient selection (up to four stars), comparability between cases and controls (up to two stars), and adequate ascertainment of exposure (up to three stars). Studies with more than seven stars were considered to have a high level of evidence. Two authors independently assessed the quality of evidence in the included studies, With any disagreements resolved through discussion. Frontiers in Oncology frontiersin.org Effect of ALND versus AxRT Four studies (25, 30, 33, 37) compared overall survival between ALND and AxRT. Pooled data analysis revealed no signiﬁcant difference between the two groups (HR 0.88, 95% CI: 0.67-1.15; p=0.35) (Figure 4A). Additionally, pooling data from three studies (30, 33, 34) assessing disease-free survival also showed no signiﬁcant difference between ALND and AxRT (HR 0.85, 95% CI: 0.68-1.05; p=0.13). Furthermore, these studies exhibited no signiﬁcant heterogeneity(I2 = 0%, p=0.71) (Figure 4B). Two studies (30, 33) reported axillary recurrence. While the axillary recurrence rate was higher in the AxRT group compared to the ALND group, the difference was not statistically signiﬁcant (HR Effect of AxRT versus SLNB alone Four studies (11, 25, 35, 37) assessed overall survival by comparing the AxRT group to the SLNB alone group. The results revealed no statistical difference in overall survival between the groups (HR 0.57, 95% CI: 0.32-1.02; p=0.25), with moderate heterogeneity between studies (c2 = 4.12, I2 = 27%, p=0.27) (Figure 5A). The asymmetry of the funnel plot suggests publication bias. Three studies (11, 25, 35) reported disease-free survival. Patients who received AxRT had a higher rate of disease-free survival than those who underwent SLNB alone (HR 0.52, 95% CI: 0.26-1.05).However, no statistically signiﬁcant difference was observed between groups (p=0.07). There was moderate heterogeneity among studies (c2 = 3.79, I2 = 47%, p=0.15) as shown in Figure 5B. Study selection Our initial database search yielded a total of 4,714 studies. After removing 504 duplicate studies, we screened the titles and abstracts of 4,210remaining studies. A total of 4,124 studies were excluded due to irrelevance (non-related studies, review articles, case reports, meta-analysis, or lack of data). At the full-text level, 86 potentially eligible studies were assessed, t of which 60 were ultimately excluded after a thorough review. This left 26 studies involving 145,548 patients for inclusion in the meta-analysis (5, 7, 11, 15–36) The PRISMA ﬂowchart in Figure 1 illustrates this selection process. Five studies evaluated locoregional recurrence (5, 21, 28, 35, 36). Pooled data indicated no statistically signiﬁcant difference between patients who received SLNB alone and those who underwent ALND (HR 0.76, 95%CI 0.45-1.29; p = 0.31) (Figure 2C).However, the 03 frontiersin.org Li et al. 10.3389/fonc.2023.1320867 Records identified through databases searchingg (n=4714) Records after duplicates removed p (n=4210) Records screened (n=4210) Records excluded (n=4124) Case reports p Reviews and editorials Conference abstract onlyy Not on SLNB, ALND or AxRT Full-text articles assessed for eligibility (n=86) Studies included in this meta-analysis (n=26) Full-text articles are excluded (n= 60) ( Data were overlapping (n=3) pp g ( ) Does not meet inclusion criteria (n=57) Identification Screening Addition records identified through other sources (n=0) Eligibility Included FIGURE 1 Flow diagram of study selection. Addition records identified through other sources (n=0) Records after duplicates removed p (n=4210) Records excluded (n=4124) Case reports p Reviews and editorials Conference abstract onlyy Not on SLNB, ALND or AxRT Full-text articles are excluded (n= 60) ( Data were overlapping (n=3) pp g ( ) Does not meet inclusion criteria (n=57) ull-text articles assessed for eligibility (n=86) 0.94, 95% CI: 0.68-1.31; p=0.73) (Figure 4C). There was also no signiﬁcant between-study heterogeneity(I2 = 21%, p=0.26). asymmetry of the funnel plot (Figure 3C) suggests potential publication bias. Four studies reported the ﬁve-year cumulative incidence of axillary recurrence (22, 24, 32, 35). Although the axillary recurrence rate was higher in the SLNB alone group compared to the ALND group, but this difference was not statistically signiﬁcant (HR1.01, 95% CI 0.99-1.03; p = 0.35) (Figure 2D).Additionally, the studies showed no signiﬁcant heterogeneity (I2 = 41%, p = 0.17). However, only one study reported the 10-year axillary recurrence outcome (32). This study found that axillary recurrence was more frequent among those who received SLNB alone compared to ALND (HR 5.47, 95% CI 1.21-24.63; p=0.013). Frontiers in Oncology Discussion This meta-analysis aimed to compare the effects of SLNB alone, ALND, and AxRT on various outcomes, including overall survival, disease-free survival, locoregional recurrence, and axillary 04 Frontiers in Oncology frontiersin.org Li et al. 10.3389/fonc.2023.1320867 TABLE 1 Summary of characteristics of included studies. frontiersin.org Discussion 10.3389/fonc.2023.1320867 TABLE 1 Continued Reference Type of study SLNB alone/ ALND or AxRT T stage (T1/T2) Micro/ Macro Adjuvant RT (Yes/No) Follow up (years) Outcomes Quality (NOSa) SLNB alone ALND SLNB alone ALND SLNB alone ALND Bilimoria 2009 (22) Retrospective 20217/77097 NA NA 3674/16543 6585/70512 NA NA 7.9 OS,AR 6 RCT, randomized controlled trial; SLNB, sentinel lymph nodes biopsy; ALND, axillary lymph node dissection; OS, overall survival; LLR, locoregional recurrence; DFS, disease-free survival; AR, axillary recurrence; NOS, Newcastlee-Ottawa Scale; NA, not available; Micro, micrometasis (<0.2-2.0 mm); Macro, macromeataiss (>2.0 mm). RT, radiation therapy. a Quality assessment of the observational studies was assessed using the Newcastlee-Ottawa Scale. The quality of the evidence is classiﬁed as three levels: high (more than seven stars), moderate (four to six stars), poor (less than four stars). TABLE 1 Continued RCT, randomized controlled trial; SLNB, sentinel lymph nodes biopsy; ALND, axillary lymph node dissection; OS, overall survival; LLR, locoregional recurrence; DFS, disease-free survival; AR, axillary recurrence; NOS, Newcastlee-Ottawa Scale; NA, not available; Micro, micrometasis (<0.2-2.0 mm); Macro, macromeataiss (>2.0 mm). RT, radiation therapy. a Quality assessment of the observational studies was assessed using the Newcastlee-Ottawa Scale. The quality of the evidence is classiﬁed as three levels: high (more than seven stars), moderate (four to six stars), poor (less than four stars). recurrence, in 145,548 patients with early-stage breast cancer, clinical negative axillary lymph nodes, and positive sentinel lymph nodes. The collected data revealed no signiﬁcant differences in overall survival, disease-free survival, locoregional recurrence, and axillary recurrence between the SLNB alone group and the ALND or AxRT groups. While the AxRT group showed a higher overall survival rate compared to the ALND group, this difference was not statistically signiﬁcant. Additionally, no signiﬁcant disparities were observed in terms of overall survival and disease-free survival between patients who received AxRT and those who received SLNB alone. the impact of ALND remains controversial. Peristeri et al. (38) performed a meta-analysis comparing the effects of SLNB/RT and ALND in ﬁve RCTs. Their pooled data showed that the SLNB/RT group had better overall and disease-free survival than the ALND group, with a statistically signiﬁcant difference in axillary recurrence favoring the ALND group. However, our previous meta-analysis comparing the two approaches in early-stage breast cancer with sentinel lymph node metastasis (43), found no signiﬁcant differences in overall survival, disease-free survival and locoregional recurrence between the SLNB alone and the ALND group. Discussion Reference Type of study SLNB alone/ ALND or AxRT T stage (T1/T2) Micro/ Macro Adjuvant RT (Yes/No) Follow up (years) Outcomes Quality (NOSa) SLNB alone ALND SLNB alone ALND SLNB alone ALND Tinterri 2023 (29) RCT 107/218 53/51 47/56 NA NA NA NA 2.8 OS RCT Houvenaeghel 2023 (23) Retrospective 185/1266 NA NA NA NA 1123/32 174/9 5.8 OS,DFS 6 Campbell 2023 (32) RCT 544/544 NA NA NA NA 482/62 466/73 10 OS,DFS,AR RCT Bartels 2023 (30) RCT 681/744 533/143 612/132 195/419 215/442 681/0 703/41 10 OS,DFS,AR RCT Zhou 2022 (19) Retrospective 1883/1883 740/878 725/862 1883/0 1883/0 596/1287 621/1262 4 OS 7 Kantor 2022 (34) Retrospective 79/42 48/31 22/20 23/56 9/33 61/18 27/15 2.0 DFS, LRR 6 Sun 2021 (21) Retrospective 128/201 62/58 82/101 NA NA 68/60 108/93 4.2 OS,LRR 7 Lim 2021 (35) Retrospective 92/168 41/51 70/28 92/0 168/0 31/61 46/122 5.1 OS, DFS, LRR, AR 7 Ortega 2021 (11) Retrospective 167/93 NA NA 0/167 0/93 95/72 NA/NA 4.5 OS, DFS, AR 6 Kim 2020 (16) Retrospective 179/704 83/96 326/378 NA NA NA NA 4.5 OS 8 Arisio 2019 (26) Retrospective 211/406 118/82 211/189 155/95 84/322 169/42 335/71 7.0 OS, AR 6 Lee 2018 (20) Retrospective 1268/3174 NA NA NA NA NA NA 3.9 OS 6 Galimberti 2018 (5) RCT 469/465 322/140 316/142 NA NA NA NA 9.7 OS, DFS, LRR RCT Wu 2018 (37) Retrospective 11368/2651 4617/6751 1444/1207 NA NA NA NA 1.9 OS 4 Savolt 2017 (33) RCT 230/244 157/73 152/92 NA NA 230/0 0/244 8.1 OS, DFS, AR RCT Mamtani 2017 (36) Retrospective 162/190 NA NA NA NA NA NA 6.0 LRR 6 Giuliano 2017 (7) RCT 436/420 303/126 284/134 NA NA NA NA 9.3 OS, DFS RCT Giuliano 2016 (28) RCT 436/420 303/126 284/134 NA NA NA NA 9.3 LRR RCT Tvedskov 2015 (24) Retrospective 240/1834 NA NA NA NA NA NA 6.3 OS, AR 5 Snow 2015 (17) Retrospective 60/258 NA NA NA NA 36/24 147/111 6.3 OS 5 Bonneau 2015 (18) Retrospective 402/9119 174/228 3665/5454 NA NA 192/210 5426/3677 2.6 OS 4 Park 2014 (27) Retrospective 197/2384 130/67 1171/1177 NA NA 4/55 439/757 3.5 OS 5 Fu 2014 (25) Retrospective 106/108 49/47/7 25/53/26 NA NA 59/46 65/28 3.6 OS 4 Yi 2013 (15) Retrospective 188/673 152/36 445/228 136/52 158/515 NA NA 5.8 OS, DFS 7 Solá 2013 (31) RCT 121/112 NA NA NA NA NA NA 5.0 DFS RCT (Continued) 05 Frontiers in Oncology frontiersin.org Li et al. a.I value is 42% as moderate heterogeneity. ALND, axillary lymph node dissection; SLNB, sentinel lymph nodes biopsy; OS, overall survival; LLR, locoregional recurrence; DFS, disease-free survival; CI, conﬁdence interval; HR, hazard Ratio. The evidence quality was classiﬁed into 4 levels: high (⊕⊕⊕⊕), moderate (⊕⊕⊕⊖), low (⊕⊕⊖⊖) or very low (⊕⊖⊖⊖). Discussion Similarly, a meta-analysis of Real-World Evidence in the Post-ACOSOG Z0011 trial (39), which included one RCT and six retrospective studies with 8,864 early-stage breast cancer patients with one or two SLN metastases, found no differences between SLNB alone and ALND groups in overall survival, disease-free Several meta-analyses (38–47) have been conducted to compare the differences in overall survival, disease-free survival, and recurrence rates between ALND and SLNB alone in early-stage breast cancer patients with positive sentinel lymph nodes. However, Patient or population: patients with clinical node-negative early breast cancer and positive sentinel lymph node Setting: Hospital Intervention: ALND Comparison: SLNB alone Outcomes Anticipated absolute effects* (95% CI) Relative effect (95% CI) №of participants (studies) Certainty of the evidence (GRADE) Comments Risk with SLNB Risk with ALND OS – Randomized control trials 514 per 1,000 545 per 1,000 (485 to 603) HR 1.09 (0.92 to 1.28) 6406 (4 RCTs) ⊕⊕⊕◯ Moderatea DFS - Randomized control trials 512 per 1,000 522 per 1,000 (472 to 574) HR 1.03 (0.89 to 1.19) 6872 (5 RCTs) ⊕⊕⊕⊕ High LRR - Randomized control trials 494 per 1,000 400 per 1,000 (239 to 615) HR 0.75 (0.40 to 1.40) 3580 (2 RCTs) ⊕⊕⊕⊕ High h k h ( d ﬁd l) b d h d k h d h l ff f h ( ) ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; HR, hazard Ratio; GRADEpro GDT: Grading of Recommendation Assessment Development and Evaluation Pproﬁler Guide- line Development Tool. High certainty: we are very conﬁdent that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately conﬁdent in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. y Low certainty: our conﬁdence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little conﬁdence in the effect estimate: the true effect is likely to be substantially different from the estimate of effec ainty: our conﬁdence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. certainty: we have very little conﬁdence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. g y y Moderate certainty: we are moderately conﬁdent in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. TABLE 2 Evaluating the quality of evidence in randomized controlled trials by GRADEpro GDT ALND compared to SLNB alone for patients with clinical node-negative early breast cancer and positive sentinel lymph node. king Group grades of evidence y: we are very conﬁdent that the true effect lies close to that of the estimate of the effect. ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; HR, hazard Ratio; GRADEpro GDT: Grading of Recommendation Assessment Development and Evaluation Pproﬁler Guide- line Development Tool. Low certainty: our conﬁdence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little conﬁdence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. y ainty: our conﬁdence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Frontiers in Oncology Discussion 06 Frontiers in Oncology frontiersin.org frontiersin.org Li et al. 10.3389/fonc.2023.1320867 survival and recurrence rate However the incidence rate of positive sentinel lymph nodes Three studies reported the ﬁve year A B D C FIGURE 2 Forest plot of ALND versus SLNB alone (A) overall survival (B) disease-free survival (C) locoregional recurrence (D) axillary recurrence. ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. i l d t H th i id t f iti ti l l h d Th t di t d th ﬁ A B D C FIGURE 2 Forest plot of ALND versus SLNB alone (A) overall survival (B) disease-free survival (C) locoregional recurrence (D) axillary recurrence. ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. A B D C FIGURE 2 Forest plot of ALND versus SLNB alone (A) overall survival (B) disease-free survival (C) locore lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, stand A B C D FIGURE 2 Forest plot of ALND versus SLNB alone (A) overall survival (B) disease-free survival (C) locoregional recurrence (D) axillary recurrence. ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. FIGURE 2 Forest plot of ALND versus SLNB alone (A) overall survival (B) disease-free survival (C) locoregional recurrence (D) axillary recurrence. ALND, axillary lymph node dissection; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. positive sentinel lymph nodes. Three studies reported the ﬁve-year cumulative incidence of axillary recurrence. The rate was higher in the SLNB alone group compared to the ALND group, but this difference was not statistically signiﬁcant (p = 0.36). However, when the follow-up period was extended to 10 years, Campbell et al. (32) found that that axillary recurrence was more frequent in the SLNB alone group compared to ALND group. Therefore, longer follow-up survival, and recurrence rate. However, the incidence rate of lymphedema was signiﬁcantly lower in SLNB alone group Our systematic review and meta-analysis included eight RCTs and eighteen retrospective cohort studies. The results indicated no statistically signiﬁcant difference in disease-free survival, overall survival, and locoregional recurrence between ALND and SLNB alone in clinical node-negative early breast cancer patients with 07 Frontiers in Oncology frontiersin.org Li et al. Frontiers in Oncology frontiersin.org Discussion 10.3389/fonc.2023.1320867 A B C FIGURE 3 Funnel plot in ALND versus SLNB alone (A) Funnel plot for overall survival (B) Funnel plot for disease-free survival (C) Funnel plot for locoregional recurrence. A B A C Funnel plot in ALND versus SLNB alone (A) Funnel plot for overall survival (B) Funnel plot for disease-free survival (C) Funnel plot f locoregional recurrence. times are required to deﬁnitively compare the axillary recurrence rates between the two groups in this patients population. After ten year of follow-up, the Randomized Controlled EORTC AMAROS trials (30) showed that both AxRT and ALND groups achieved excellent locoregional control and survival in cT1-T2 breast cancer patients with positive sentinel lymph nodes. Additionally, the AxRT group had a lower rate of lymphedema and no difference in quality of life compared to the ALND group. This meta-analysis also conﬁrms no signiﬁcant differences in disease-free survival, overall survival, and axillary recurrence between patients treated with ALND versus AxRT. Our review diverges from previous studies in several key aspects. First, by incorporating eight RCTs and 18 retrospective studies involving 145,548 patients, our meta-analysis signiﬁcantly increase the sample size, leading to more precise and reliable results. Second, we employed the GRADEpro GTD tool to assess evidence 08 Frontiers in Oncology frontiersin.org frontiersin.org frontiersin.org Li et al. 10.3389/fonc.2023.1320867 A B C FIGURE 4 Forest plot of ALND versus AxRT (A) overall survival (B) disease-free survival (C) axillary recurrence. ALND, axillary lymph node dissection; AxRT, axillary radiation; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. A B C FIGURE 4 Forest plot of ALND versus AxRT (A) overall survival (B) disease-free survival (C) axillary recurrence. ALND, axillary lymph node dissection; AxRT, axillary radiation; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. A FIGURE 4 Forest plot of ALND versus AxRT (A) overall survival (B) disease-free survival (C) axillary recurrence. ALND, axillary lymph node dissection; AxRT, axillary radiation; CI, conﬁdence interval; SE, standard error; IV, Inverse Variance. evidence. Third, both overall and disease-free survival analyses showed evidence of publication bias. Fourth, moderate heterogeneity was observed among studies regarding disease-free and overall survival when comparing the AxRT and SLNB alone groups. A study by Ortega et al. (11) identiﬁed potential sources of this heterogeneity. While removing their study resulted in a signiﬁcant decrease in heterogeneity for both outcomes, the overall and disease-free survival data also changed substantially (Supplementary Figures 1, 2). Discussion Therefore, further careful evaluation of the effect of AxRT versus SLNB alone is required, and these results necessitate conﬁrmation through well-designed prospective studies. Finally, because the lack of sufﬁcient information within the included studies precluded further subgroup analyses based on quality within RCTs, revealing high quality for disease-free survival and locoregional recurrence, and moderate quality for overall survival. Furthermore, subgroup analyses based on study type (RCT vs. retrospective) demonstrated that the pooled analysis results remained unchanged, indicating the stability of our ﬁnding. Third, and uniquely, our review evaluated the effects of AxRT and SLNB alone in patients with clinical node-negative early breast cancer and positive sentinel lymph nodes, an aspect lacking in prior meta-analysis. However, limitations exist within our meta- analysis. First, our inclusion criteria restricted us to English studies only, potentially introducing publication bias by excluding unpublished data. Second, the NOS tool revealed six non-RCT studies with a, four to ﬁve-star rating, including lower-quality A B FIGURE 5 Forest plot of AxRT versus SLNB lone (A) overall survival (B) disease-free survival. AxRT, axillary radiation; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV,Inverse Variance. A B FIGURE 5 Forest plot of AxRT versus SLNB lone (A) overall survival (B) disease-free survival. AxRT, axillary radiation; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV,Inverse Variance. A FIGURE 5 Forest plot of AxRT versus SLNB lone (A) overall survival (B) disease-free survival. AxRT, axillary radiation; SLNB, sentinel lymph node biopsy; CI, conﬁdence interval; SE, standard error; IV,Inverse Variance. 09 Frontiers in Oncology frontiersin.org Li et al. Li et al. 10.3389/fonc.2023.1320867 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. Funding factors like the T1/T2 stage, number of positive sentinel lymph nodes, micrometastasis or macrometastasis, molecular subtype, and age. The author(s) declare ﬁnancial support was received for the research, authorship, and/or publication of this article. This work was supported by grants from Hebei Province University Basic Research Business Fee Project (Project number: JYG2020003). Conclusion This study demonstrates that patients with early-stage breast cancer and positive sentinel lymph nodes who undergo SLNB alone achieve comparable locoregional control and survival to those who receive ALND or AxRT. Our ﬁndings suggest that omitting ALND or AxRT may be safe for these patients, although further veriﬁcation is needed through rigorously designed prospective studies. SUPPLEMENTARY FIGURE 1 SUPPLEMENTARY FIGURE 1 Forest plot of overall survival after heterogeneity analysis in AxRT versus SLNB lone. Forest plot of overall survival after heterogeneity analysis in AxRT versus SLNB lone. Data availability statement 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. The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author. SUPPLEMENTARY FIGURE 2 Forest plot of disease-free survival after heterogeneity analysis in AxRT versus SLNB lone. References 1. Lucci A, McCall LM, Beitsch PD, Whitworth PW, Reintgen DS, Blumencranz PW, et al. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol (2007) 25:3657–63. doi: 10.1200/JCO.2006.07.4062 6. National Comprehensive Cancer Network (NCCN) clinical practice guidelines in oncology: breast. (2021). https://www.nccn.org/professionals/physician_gls/pdf/breast. pdf. Accessed on 20 December 2021. 6. National Comprehensive Cancer Network (NCCN) clinical practice guidelines in oncology: breast. (2021). https://www.nccn.org/professionals/physician_gls/pdf/breast. pdf. Accessed on 20 December 2021. 7. Giuliano AE, Ballman KV, McCall L, Beitsch PD, Brennan MB, Kelemen PR, et al. Effect of axillary dissection vs no axillary dissection on 10-year overall survival among women with invasive breast cancer and sentinel node metastasis: the ACOSOG Z0011 (Alliance) randomized clinical trial. JAMA (2017) 318:918–26. doi: 10.1001/ jama.2017.11470 2. Lyman GH, Somerﬁeld MR, Bosserman LD, Perkins CL, Weaver DL, Giuliano AE. Sentinel lymph node biopsy for patients with early-stage breast cancer: american society of clinical oncology clinical practice guideline update. J Clin Oncol (2017) 35:561–4. doi: 10.1200/JCO.2016.71.0947 8. Galimberti V, Cole BF, Zurrida S, Viale G, Luini A, Veronesi P, et al. Axillary dissection versus no axillary dissection in patients with sentinel-node micrometastases (IBCSG 23-01): a phase 3 randomised controlled trial. Lancet Oncol (2013) 14:297–305. doi: 10.1016/S1470-2045(13)70035-4 3. Pepels MJ, de Boer M, Bult P, van Dijck JA, van Deurzen CH, Menke-Pluymers MB, et al. Regional recurrence in breast cancer patients with sentinel node micrometastases and isolated tumor cells. Ann Surg (2012) 255:116–21. doi: 10.1097/ SLA.0b013e31823dc616 9. Joo JH, Kim SS, Son BH, Ahn SD, Jung JH, Choi EK, et al. Axillary lymph node dissection does not improve post-mastectomy overall or disease-free survival among breast cancer patients with 1-3 positive nodes. Cancer Res Treat (2019) 51:1011–21. doi: 10.4143/crt.2018.438 4. Pepels MJ, Vestjens JH, de Boer M, Smidt M, van Diest PJ, Borm GF, et al. Safety of avoiding routine use of axillary dissection in early stage breast cancer: a systematic review. Breast Cancer Res Treat (2011) 125:301–13. doi: 10.1007/s10549-010-1210-7 10. Donker M, van Tienhoven G, Straver ME, Meijnen P, van de Velde CJ, Mansel RE, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981-22023 AMAROS): a randomised, multicentre, open-label, phase 3 non- inferiority trial. Lancet Oncol (2014) 15:1303–10. doi: 10.1016/S1470-2045(14)70460-7 5. Supplementary material PZ: Data curation, Formal Analysis, Methodology, Writing – review & editing. CL: Data curation, Investigation, Methodology, Writing – original draft, Writing – review & editing. JL: Data curation, Investigation, Project administration, Writing – review & editing. WD: Formal Analysis, Methodology, Writing – review & editing. BH: Investigation, Methodology, Project administration, Software, Supervision, Writing – review & editing. JZ: Data curation, Investigation, Methodology, Writing – review & editing. HZ: Investigation, Methodology, Project administration, Software, Supervision, Writing – review & editing. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fonc.2023.1320867/ full#supplementary-material Frontiers in Oncology 6. National Comprehensive Cancer Network (NCCN) clinical practice guidelines in oncology: breast. (2021). https://www.nccn.org/professionals/physician_gls/pdf/breast. pdf. Accessed on 20 December 2021. 6. National Comprehensive Cancer Network (NCCN) clinical practice guidelines in oncology: breast. (2021). https://www.nccn.org/professionals/physician_gls/pdf/breast. pdf. Accessed on 20 December 2021. 7. Giuliano AE, Ballman KV, McCall L, Beitsch PD, Brennan MB, Kelemen PR, et al. Effect of axillary dissection vs no axillary dissection on 10-year overall survival among women with invasive breast cancer and sentinel node metastasis: the ACOSOG Z0011 (Alliance) randomized clinical trial. JAMA (2017) 318:918–26. doi: 10.1001/ jama.2017.11470 8. Galimberti V, Cole BF, Zurrida S, Viale G, Luini A, Veronesi P, et al. Axillary dissection versus no axillary dissection in patients with sentinel-node micrometastases (IBCSG 23-01): a phase 3 randomised controlled trial. Lancet Oncol (2013) 14:297–305. doi: 10.1016/S1470-2045(13)70035-4 9. Joo JH, Kim SS, Son BH, Ahn SD, Jung JH, Choi EK, et al. Axillary lymph node dissection does not improve post-mastectomy overall or disease-free survival among breast cancer patients with 1-3 positive nodes. Cancer Res Treat (2019) 51:1011–21. doi: 10.4143/crt.2018.438 10. Donker M, van Tienhoven G, Straver ME, Meijnen P, van de Velde CJ, Mansel RE, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981-22023 AMAROS): a randomised, multicentre, open-label, phase 3 non- inferiority trial. Lancet Oncol (2014) 15:1303–10. doi: 10.1016/S1470-2045(14)70460-7 4. Pepels MJ, Vestjens JH, de Boer M, Smidt M, van Diest PJ, Borm GF, et al. Safety of avoiding routine use of axillary dissection in early stage breast cancer: a systematic review. Breast Cancer Res Treat (2011) 125:301–13. doi: 10.1007/s10549-010-1210-7 References Hatala R, Keitz S, Wyer P, Guyatt G. Tips for learners of evidence-based medicine: 4. Assessing heterogeneity of primary studies in systematic reviews and whether to combine their results. CMAJ (2005) 172:661–5. doi: 10.1503/cmaj.1031920 15. Yi M, Kuerer HM, Mittendorf EA, Hwang RF, Caudle AS, Bedrosian I, et al. Impact of the american college of surgeons oncology group Z0011 criteria applied to a contemporary patient population. J Am Coll Surg (2013) 216:105–13. doi: 10.1016/ j.jamcollsurg.2012.09.005 33. Savolt A, Peley G, Polgar C, Udvarhelyi N, Rubovszky G, Kovacs E, et al. Eight- year follow up result of the OTOASOR trial: The Optimal Treatment Of the Axilla - Surgery Or Radiotherapy after positive sentinel lymph node biopsy in early-stage breast cancer: A randomized, single centre, phase III, non-inferiority trial. Eur J Surg Oncol (2017) 43:672–9. doi: 10.1016/j.ejso.2016.12.011 33. Savolt A, Peley G, Polgar C, Udvarhelyi N, Rubovszky G, Kovacs E, et al. Eight- year follow up result of the OTOASOR trial: The Optimal Treatment Of the Axilla - Surgery Or Radiotherapy after positive sentinel lymph node biopsy in early-stage breast cancer: A randomized, single centre, phase III, non-inferiority trial. Eur J Surg Oncol (2017) 43:672–9. doi: 10.1016/j.ejso.2016.12.011 16. Kim BK, Park BW, Hur MH, Lee HB, Park MH, Jeong J, et al. Omission of axillary lymph node dissection in patients who underwent total mastectomy with 1 or 2 metastatic lymph nodes. Ann Surg Treat Res (2020) 98:283–90. doi: 10.4174/ astr.2020.98.6.283 34. Kantor O, Means J, Grossmith S, Dey T, Bellon JR, Mittendorf EA, et al. Optimizing axillary management in clinical T1-2N0 mastectomy patients with positive sentinel lymph nodes. Ann Surg Oncol (2022) 29:972–80. doi: 10.1245/s10434-021- 10726-3 34. Kantor O, Means J, Grossmith S, Dey T, Bellon JR, Mittendorf EA, et al. Optimizing axillary management in clinical T1-2N0 mastectomy patients with positive sentinel lymph nodes. Ann Surg Oncol (2022) 29:972–80. doi: 10.1245/s10434-021- 10726-3 17. Snow R, Reyna C, Johns C, Lee MC, Sun W, Fulp WJ, et al. Outcomes with and without axillary node dissection for node-positive lumpectomy and mastectomy patients. Am J Surg (2015) 210:685–93. doi: 10.1016/j.amjsurg.2015.05.004 35. Lim SZ, Kusumawidjaja G, Mohd Ishak HM, Tan BKT, Tan SY, Hamzah JL, et al. Outcomes of stage I and II breast cancer with nodal micrometastases treated with mastectomy without axillary therapy. Breast Cancer Res Treat (2021) 189:837–43. doi: 10.1007/s10549-021-06341-1 18. Bonneau C, Hequet D, Estevez JP, Pouget N, Rouzier R. References Galimberti V, Cole BF, Viale G, Veronesi P, Vicini E, Intra M, et al. Axillary dissection versus no axillary dissection in patients with breast cancer and sentinel-node micrometastases (IBCSG 23-01): 10-year follow-up of a randomised, controlled phase 3 trial. Lancet Oncol (2018) 19:1385–93. doi: 10.1016/S1470-2045(18)30380-2 10 frontiersin.org Li et al. 10.3389/fonc.2023.1320867 analysis of the SINODAR-ONE multicentre randomized clinical trial and reopening of enrolment. Br J Surg (2023) 110:1143–52. doi: 10.1093/bjs/znad215 11. Ortega Expósito C, Falo C, Pernas S, Pérez Carton S, Gil Gil M, Ortega R, et al. The effect of omitting axillary dissection and the impact of radiotherapy on patients with breast cancer sentinel node macrometastases: a cohort study following the ACOSOG Z0011 and AMAROS trials. Breast Cancer Res Treat (2021) 189:111–20. doi: 10.1007/s10549-021-06274-9 analysis of the SINODAR-ONE multicentre randomized clinical trial and reopening of enrolment. Br J Surg (2023) 110:1143–52. doi: 10.1093/bjs/znad215 30. Bartels SAL, Donker M, Poncet C, Sauve N, Straver ME, van de Velde CJH, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer: 10- year results of the randomized controlled EORTC 10981-22023 AMAROS trial. J Clin Oncol (2023) 41:2159–65. doi: 10.1200/JCO.22.01565 30. Bartels SAL, Donker M, Poncet C, Sauve N, Straver ME, van de Velde CJH, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer: 10- year results of the randomized controlled EORTC 10981-22023 AMAROS trial. J Clin Oncol (2023) 41:2159–65. doi: 10.1200/JCO.22.01565 12. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ (2011) 343:d5928. doi: 10.1136/bmj.d5928 31. Solá M, Alberro JA, Fraile M, Santesteban P, Ramos M, Fabregas R, et al. Complete axillary lymph node dissection versus clinical follow-up in breast cancer patients with sentinel node micrometastasis: Final results from the multicenter clinical trial AATRM 048/13/2000. Ann Surg Oncol (2013) 20:120–7. doi: 10.1245/s10434-012-2569-y 13. Wells GA SB OCD, Peterson J, Welch V, Peterson J, Welch V, Losos M, et al. The NewcastleOttawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. 32. Campbell I, Wetzig N, Ung O, Espinoza D, Farshid G, Collins J, et al. 10-Year axillary recurrence in the RACS SNAC1 randomised trial of sentinel lymph node-based management versus routine axillary lymph node dissection. Breast (2023) 70:70–5. doi: 10.1016/j.breast.2023.06.009 14. References Impact of axillary dissection in women with invasive breast cancer who do not ﬁt the Z0011 ACOSOG trial because of three or more metastatic sentinel lymph nodes. Eur J Surg Oncol (2015) 41:998–1004. doi: 10.1016/j.ejso.2015.04.003 36. Mamtani A, Patil S, Stempel M, Morrow M. Axillary micrometastases and isolated tumor cells are not an indication for post-mastectomy radiotherapy in stage 1 and 2 breast cancer. Ann Surg Oncol (2017) 24:2182–8. doi: 10.1245/s10434-017-5866-7 19. Zhou Y, Pu S, Jiang S, Li D, Li S, Liu Y, et al. The prognostic signiﬁcance of further axillary dissection for sentinel lymph node micrometastases in female breast cancer: A competing risk analysis using the SEER database. Front Oncol (2022) 12:1012646. doi: 10.3389/fonc.2022.1012646 37. Wu SP, Tam M, Shaikh F, Lee A, Chun J, Schnabel F, et al. Post-mastectomy radiation therapy in breast cancer patients with nodal micrometastases. Ann Surg Oncol (2018) 25:2620–31. doi: 10.1245/s10434-018-6632-1 37. Wu SP, Tam M, Shaikh F, Lee A, Chun J, Schnabel F, et al. Post-mastectomy radiation therapy in breast cancer patients with nodal micrometastases. Ann Surg Oncol (2018) 25:2620–31. doi: 10.1245/s10434-018-6632-1 38. Peristeri DV, Harissis HV. Axillary lymph node dissection vs sentinel biopsy only among women with early-stage breast cancer and sentinel node metastasis: A systematic review and meta-analysis. Breast J (2021) 27:158–64. doi: 10.1111/tbj.14140 20. Lee J, Choi JE, Kim SJ, Lee SB, Seong MK, Jeong J, et al. Comparative study between sentinel lymph node biopsy and axillary dissection in patients with one or two lymph node metastases. J Breast Cancer (2018) 21:306–14. doi: 10.4048/jbc.2018.21.e44 38. Peristeri DV, Harissis HV. Axillary lymph node dissection vs sentinel biopsy only among women with early-stage breast cancer and sentinel node metastasis: A systematic review and meta-analysis. Breast J (2021) 27:158–64. doi: 10.1111/tbj.14140 21. Sun J, Mathias BJ, Laronga C, Sun W, Zhou JM, Fulp WJ, et al. Impact of axillary dissection among patients with sentinel node-positive breast cancer undergoing mastectomy. J Natl Compr Canc Netw (2021) 19:40–7. doi: 10.6004/jnccn.2020.7597 21. Sun J, Mathias BJ, Laronga C, Sun W, Zhou JM, Fulp WJ, et al. Impact of axillary dissection among patients with sentinel node-positive breast cancer undergoing mastectomy. J Natl Compr Canc Netw (2021) 19:40–7. doi: 10.6004/jnccn.2020.7597 39. Huang TW, Su CM, Tam KW. Axillary management in women with early breast cancer and limited sentinel node metastasis: A systematic review and metaanalysis of real-world evidence in the post-ACOSOG Z0011 era. References Arisio R, Borella F, Porpiglia M, Durando A, Bellino R, Bau MG, et al. Axillary Dissection vs. no Axillary Dissection in Breast Cancer Patients With Positive Sentinel Lymph Node: A Single Institution Experience. In Vivo (2019) 33:1941–7. doi: 10.21873/ invivo.11689 26. Arisio R, Borella F, Porpiglia M, Durando A, Bellino R, Bau MG, et al. Axillary Dissection vs. no Axillary Dissection in Breast Cancer Patients With Positive Sentinel Lymph Node: A Single Institution Experience. In Vivo (2019) 33:1941–7. doi: 10.21873/ invivo.11689 44. Joyce DP, Manning A, Carter M, Hill AD, Kell MR, Barry M. Meta-analysis to determine the clinical impact of axillary lymph node dissection in the treatment of invasive breast cancer. Breast Cancer Res Treat (2015) 153:235–40. doi: 10.1007/ s10549-015-3549-2 27. Park HS, Chae BJ, Song BJ, Jung SS, Han W, Nam SJ, et al. Effect of axillary lymph node dissection after sentinel lymph node biopsy on overall survival in patients with T1 or T2 node-positive breast cancer: Report from the Korean Breast Cancer Society. Ann Surg Oncol (2014) 21:1231–6. doi: 10.1245/s10434-013-3350-6 27. Park HS, Chae BJ, Song BJ, Jung SS, Han W, Nam SJ, et al. Effect of axillary lymph node dissection after sentinel lymph node biopsy on overall survival in patients with T1 or T2 node-positive breast cancer: Report from the Korean Breast Cancer Society. Ann Surg Oncol (2014) 21:1231–6. doi: 10.1245/s10434-013-3350-6 45. Glechner A, Wöckel A, Gartlehner G, Thaler K, Strobelberger M, Griebler U, et al. Sentinel lymph node dissection only versus complete axillary lymph node dissection in early invasive breast cancer: a systematic review and meta-analysis. Eur J Cancer (2013) 49:812–25. doi: 10.1016/j.ejca.2012.09.010 45. Glechner A, Wöckel A, Gartlehner G, Thaler K, Strobelberger M, Griebler U, et al. Sentinel lymph node dissection only versus complete axillary lymph node dissection in early invasive breast cancer: a systematic review and meta-analysis. Eur J Cancer (2013) 49:812–25. doi: 10.1016/j.ejca.2012.09.010 28. Giuliano AE, Ballman K, McCall L, Beitsch P, Whitworth PW, Blumencranz P, et al. Locoregional recurrence after sentinel lymph node dissection with or without axillary dissection in patients with sentinel lymph node metastases: long-term follow- up from the american college of surgeons oncology group (Alliance) ACOSOG Z0011 randomized trial. Ann Surg (2016) 264:413–20. doi: 10.1097/SLA.0000000000001863 46. Wang Z, Wu LC, Chen JQ. Sentinel lymph node biopsy compared with axillary lymph node dissection in early breast cancer: a meta-analysis. Breast Cancer Res Treat (2011) 129:675–89. doi: 10.1007/s10549-011-1665-1 46. References Ann Surg Oncol (2021) 28:920–9. doi: 10.1245/s10434-020-08923-7 22. Bilimoria KY, Bentrem DJ, Hansen NM, Bethke KP, Rademaker AW, Ko CY, et al. Comparison of sentinel lymph node biopsy alone and completion axillary lymph node dissection for node-positive breast cancer. J Clin Oncol (2009) 27:2946–53. doi: 10.1200/JCO.2008.19.5750 40. Cao S, Liu X, Cui J, Liu X, Zhong J, Yang Z, et al. Feasibility and reliability of sentinel lymph node biopsy after neoadjuvant chemotherapy in breast cancer patients with positive axillary nodes at initial diagnosis: An up-to-date meta-analysis of 3,578 patients. Breast (2021) 59:256–69. doi: 10.1016/j.breast.2021.07.015 23. Houvenaeghel G, de Nonneville A, Chopin N, Classe JM, Mazouni C, Chauvet M, et al. The need to tailor the omission of axillary lymph node dissection to patients with good prognosis and sentinel node micro-metastases. Cancer Med (2023) 12:4023– 32. doi: 10.1002/cam4.5257 23. Houvenaeghel G, de Nonneville A, Chopin N, Classe JM, Mazouni C, Chauvet M, et al. The need to tailor the omission of axillary lymph node dissection to patients with good prognosis and sentinel node micro-metastases. Cancer Med (2023) 12:4023– 32. doi: 10.1002/cam4.5257 41. Liang S, Hallet J, Simpson JS, Tricco AC, Scheer AS. Omission of axillary staging in elderly patients with early stage breast cancer impacts regional control but not survival: A systematic review and meta-analysis. J Geriatr Oncol (2017) 8:140–7. doi: 10.1016/j.jgo.2016.12.003 24. Tvedskov TF, Jensen MB, Ejlertsen B, Christiansen P, Balslev E, Kroman N. Prognostic signiﬁcance of axillary dissection in breast cancer patients with micrometastases or isolated tumor cells in sentinel nodes: a nationwide study. Breast Cancer Res Treat (2015) 153:599–606. doi: 10.1007/s10549-015-3560-7 42. Huang TW, Kuo KN, Chen KH, Chen C, Hou WH, Lee WH, et al. Recommendation for axillary lymph node dissection in women with early breast cancer and sentinel node metastasis: A systematic review and meta-analysis of randomized controlled trials using the GRADE system. Int J Surg (2016) 34:73–80. doi: 10.1016/j.ijsu.2016.08.022 25. Fu Y, Chung D, Cao MA, Apple S, Chang H. Is axillary lymph node dissection necessary after sentinel lymph node biopsy in patients with mastectomy and pathological N1 breast cancer? Ann Surg Oncol (2014) 21:4109–23. doi: 10.1245/ s10434-014-3814-3 43. Li CZ, Zhang P, Li RW, Wu CT, Zhang XP, Zhu HC. Axillary lymph node dissection versus sentinel lymph node biopsy alone for early breast cancer with sentinel node metastasis: A meta-analysis. Eur J Surg Oncol (2015) 41:958–66. doi: 10.1016/ j.ejso.2015.05.007 26. References Wang Z, Wu LC, Chen JQ. Sentinel lymph node biopsy compared with axillary lymph node dissection in early breast cancer: a meta-analysis. Breast Cancer Res Treat (2011) 129:675–89. doi: 10.1007/s10549-011-1665-1 47. Sanghani M, Balk EM, Cady B. Impact of axillary lymph node dissection on breast cancer outcome in clinically node negative patients: a systematic review and meta-analysis. Cancer (2009) 115:1613–20. doi: 10.1002/cncr.24174 47. Sanghani M, Balk EM, Cady B. Impact of axillary lymph node dissection on breast cancer outcome in clinically node negative patients: a systematic review and meta-analysis. Cancer (2009) 115:1613–20. doi: 10.1002/cncr.24174 29. Tinterri C, Canavese G, Gatzemeier W, Barbieri E, Bottini A, Sagona A, et al. Sentinel lymph node biopsy versus axillary lymph node dissection in breast cancer patients undergoing mastectomy with one to two metastatic sentinel lymph nodes: sub- 11 Frontiers in Oncology frontiersin.org
https://openalex.org/W2038092672	https://link.springer.com/content/pdf/10.1007%2Fs11606-013-2473-6.pdf	English	null	Women’s Health During Health Care Transformation	Journal of general internal medicine	2,013	cc-by	2,979	1Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health & Human Services, Rockville, MD, USA; 2George Washington University School of Medicine, Washington, DC, USA. 1Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health & Human Services, Rockvi Washington University School of Medicine, Washington, DC, USA. absolutely necessary to improve health and health care, but it is not sufficient. Our current system is highly fragmented as well as costly, so improving care delivery and enhancing coordination across settings are important to assure better access that result in improved health. The Affordable Care Act thus includes numerous provisions focused on improving quality of healthcare services. One model that has captured the attention of the primary care community is the patient-centered medical home (PCMH). This approach, with its strong focus on coordination and integration of services has enormous potential for women’s health since historically, even primary care for women has too often represented a patchwork quilt with gaps.1 PCMH is particularly relevant to women’s role as health care coordinators for families. Effective, meaningful use of health information technology (IT) is vital to translating this exciting model into reality. Developing effective PCMH models will require significant workforce development, teamwork enhancement and fundamental payment reform. J Gen Intern Med 28(Suppl 2):S500–3 DOI: 10.1007/s11606-013-2473-6 © Society of General Internal Medicine 2013 I n the complex dynamic of today’s health care transfor- mation, an essential component is the promise and potential to improve women’s health and health care. Moreover, women’s vital roles as decision makers, pro- viders, and caregivers mean that the transformation in progress is certain to enhance access to and quality of care for individual women, as well as their families and extended social networks. I The driving force for health care transformation is the Affordable Care Act of 2010 (ACA). Many physicians are already familiar with provisions in the ACA to expand health care insurance coverage to women and their families. In 2014, it will be illegal for insurance companies to deny coverage to anyone with a pre-existing condition, including pregnancy. In addition, sex will no longer be a pre-existing condition: insurers will not be able to charge women higher premiums than they charge men. Some provisions have already been enacted, such as providing coverage for dependent children through age 26 and no longer allowing coverage of children to be denied because of a pre-existing condition. Of the key features of PCMH, providing “comprehen- sive” care is probably the most challenging, particularly the urgency of addressing medical and behavioral health needs. WHAT CAN THE U.S. LEARN FROM THE VA IN GENERAL AND IN SUPPORT OF WOMEN’S HEALTH? & Particular problem areas include cancer screening and management of diabetes. & Quality of care varies not only across types of care but also across parts of the country. While we face enormous challenges ahead as we embark on health care transformation as a country, the Veterans Health Administration (VA) healthcare system—the only national integrated healthcare system in the US—has made many of the advances over the past 20 years that we seek to accomplish now.4 Veterans enrolled in the VA “health plan” are not turned away for pre-existing conditions (though they may have a co-pay for non-service-connected care). The VA has launched medical homes and already serves as an accountable care organization, given its integration across hospitals, nursing homes, and outpatient primary, specialty, surgical and mental health care programs, in addition to home-based primary care and major initiatives to increase non-face-to-face care. The VA’s electronic medical record already spans all of those care settings, and leverages its enormous data resources with decision support functions that support practice improvement and evidence-based policymaking.5 The VA has also a decade-long primary care-mental health integration initiative, which has already co-located mental health professionals and/or implemented collaborative care in virtually every VA primary care practice nationwide.6 In terms of women’s health, measures show a mixed picture of health care quality and disparities. A few examples3: & In all years, females were more likely than males to be unable to get or delayed in getting needed medical care, dental care or prescription medicines. & In 2009, adult females with a major depressive episode were more likely than their male counterparts to receive any treatment for depression in the last 12 months (67.4 % vs. 59 %). However, since 2008, the rate for females has decreased, while the rate for males has increased. & From 2000 to 2008, the percentage of women ages 50– 74 who reported they had a mammogram in the past 2 years did not change significantly. & From 2000 to 2007, the rate of advanced stage breast cancer in women ages 50–64 decreased from 106 to 96 per 100,000 women. Rates among women ages 40–49 and age 65 and over did not change significantly. 1Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health & Human Services, Rockville, MD, USA; 2George Washington University School of Medicine, Washington, DC, USA. However, though there are a few areas for optimism with respect to reducing disparities associated with race, ethnic- ity, gender, income and education, disparities remain pervasive across numerous domains. Specifically: Another significant initiative called for in ACA is the development of a national strategy to improve healthcare quality. Recognizing multiple efforts under way by the public and private sectors to improve quality, the act calls on the U.S. Department of Health & Human Services (DHHS) to identify priorities to improve the delivery of healthcare services, patient health outcomes, and population health. The National Quality Strategy also should include provisions to align efforts of public and private payers and reflect the “meaningful use” requirements for health IT (www.ahrq.gov/workingforquality). & Overall, improvement in the quality of care remains suboptimal and access to care is not improving (yet!). & Few disparities in quality are getting smaller and almost no disparities in access are getting smaller. 1Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health & Human Services, Rockville, MD, USA; 2George Washington University School of Medicine, Washington, DC, USA. Traditional primary care settings are comfortable with care coordination for healthy patients or those with a mild or common illness such as osteoarthritis or simple hyperten- sion. However, treatment options have increased in both number and complexity. Patients increasingly have multiple morbidities, each needing more “specialized” or compre- hensive care. Often, the specialized care relates to a behavioral-health morbidity. AHRQ-supported researchers have outlined programmatic and policy actions that are needed to facilitate integration of behavioral health care within a PCMH. These include: develop a strategy to normalize mental health into mainstream medical practice, integrate reimbursement mechanisms, create a roadmap for implementation, determine mechanisms to address the needs of those with complex mental health problems, and disseminate the tools needed by primary care providers.2 It is especially exciting that expanded coverage for women’s health explicitly includes preventive services. The law requires new health plans to cover recommended preventive services, including vaccinations, cost-free. Regular well-baby and well- child visits are also covered from birth through age 21. These services do not require a co-pay or co-insurance when offered by providers in the insurer’s network. Preventive services include evidence-based services rated “A” or “B” by the U.S. Preventive Services Task Force; for example, depression screening, flu shots and colorectal cancer screening for adults over 50. In addition, there are over 20 services targeted to women such as annual well-woman visits to obtain the recommended preventive services, domestic and interpersonal violence screening and counseling, contraception and contra- ceptive counseling, and chlamydia infection screening for younger women and other women at higher risk. Promoting mental health of women and their families is an important strategy to address access and health dispar- ities issues. Achieving equity in health requires two things. First, addressing the root causes of health inequities and second, resource allocation and interventions to specifically When coverage increases, how will access and coordi- nated care follow? Expanding access and coverage is S500 S501 Clancy and Collins Sharp: Women’s Health During Health Care Transformation JGIM address the unique needs of disadvantaged populations. Disparities in care must be described and considered. Every year since 2003, AHRQ reports to the Congress on the state of healthcare quality and healthcare disparities. During that period, we have documented steady, albeit slow, improve- ments in quality and safety, especially for acute conditions. WHAT CAN THE U.S. LEARN FROM THE VA IN GENERAL AND IN SUPPORT OF WOMEN’S HEALTH? VA initiatives in women’s health care delivery have capitalized on these system-level advantages and map to ACA priorities.7 For example, the VA reports quality indicators by gender, making reductions in documented gender disparities a formal part of network-level incentiv- ized performance plans, which have made important in- roads. VA already covers women’s preventive services and have achieved much higher rates than outside the VA.8 To address the rapid influx of women Veterans into VA care, the VA also developed women’s health mini-residencies, training over 1,500 providers to ensure proficiency, established national policy requiring comprehensive care delivery to women Veterans in “one-stop shopping” models, and recently instituted after-hours care policies to accom- modate working Veterans (women and men). VA also Accelerating Quality Improvement Is Imperative. High quality, affordable, patient-centered care is an outcome that has coalesced across settings, providers and insurers to become a shared goal with action-based focus. The ACA offers many opportunities to improve the quality of healthcare delivered to Americans. These opportunities include: & the creation of an Interagency Working Group on Healthcare Quality; & increased quality measure development; and & support for the expansion of quality-focused entities such as accountable care organizations and patient centered medical homes. Clancy and Collins Sharp: Women’s Health During Health Care Transformation S502 JGIM covers maternity care and recently added seven days of routine medical care for the newborn. paths, promote multidisciplinary research and adjust promotion/tenure criteria to reward novel and patient- centered research on health services. Specific implications of ACA for women Veterans are not clear as yet. Like other women in America, women Veterans will have more choices and may elect care delivery options outside the VA. Non-VA options may provide access through geographically closer facilities in environ- ments with more women, but especially for women Veterans with service-connected disabilities, dual use may involve challenges ensuring continuity and coordination, as well as impair access to other needed services for which VA has special expertise, such as VA mental health services. The VA also serves the Veteran (and active duty military under selected arrangements) and not family members (except under CHAMPVA), including support for their transition from military service to civilian life, and is in contrast to most other health plans. Helping women Veterans bridge care needs that may go beyond the VA’s typical boundaries may need to become an even higher priority. WHAT CAN THE U.S. LEARN FROM THE VA IN GENERAL AND IN SUPPORT OF WOMEN’S HEALTH? The American healthcare system is as diverse as it is vast—ranging from solo physician practices to large integrated delivery networks spanning many states. Yet wherever they practice, no matter the setting, healthcare practitioners are united in their goal of providing high- quality, affordable care for their patients. Registries and similar approaches to collecting clinical data systematically can generate important hypotheses, facilitate longitudinal follow up, and help assure that resources used for clinical trials are invested as wisely as possible. Women need to see quality, equitable and coordinated care on the other side of where the barrier used to be. Research that improves health care has to be firmly grounded in patients’ concerns and questions to assure relevance. We now know that research that is responsive to patients’ needs and preferences can make a huge impact and result in: & Better translation, dissemination and use of research results & Better evidence to inform guidelines & Better evidence to inform guidelines FINAL THOUGHTS & Targeted quality improvement (QI) for women to improve quality and safety Making the most of ACA-directed initiatives and responding to the social and economic urgency inherent in the scale of healthcare improvements represents an unprec- edented leap forward for women’s health, but will also involve changes in virtually all domains of general internal medicine: education, clinical care, research, patient engage- ment and advocacy. It is not yet routine that all health care professionals learn about the importance of teamwork or acquire the skills to participate as team members. Implicit in the ACA’s promise of fair, affordable coverage is the need for individual women to be partners in their care, and to have clinicians who have the capacity and skill to respond. Transparency regarding clinical performance, with a strong focus on patients’ experience of care, is a reality now—but clinicians desperately need support to improve and oppor- tunities to learn from colleagues striving toward the same goals. In addition, dramatic changes in the capacity of systems to respond to individual women’s needs and preferences will require substantial changes to business as usual, e.g., after-hours care, and schedules that fit with the time demands of women’s daily lives. & Using research to address concerns of diverse patients and clinicians Taking advantage of the opportunities that currently exist for advancing women’s health research will lead to improved health services. The VA has already leveraged these opportunities and stands at the forefront of women’s health research and quality improvement. Women’s health efforts within VA, some of which are described in this supplement, can serve as models for providers and patients in the health care system interested in promoting health care for women, their families and communities. Transparency regarding clinical performance, with a strong focus on patients’ experience of care, is a reality now—but clinicians desperately need support to improve and oppor- tunities to learn from colleagues striving toward the same goals. In addition, dramatic changes in the capacity of systems to respond to individual women’s needs and preferences will require substantial changes to business as usual, e.g., after-hours care, and schedules that fit with the time demands of women’s daily lives. Corresponding Author: Carolyn M. Clancy, MD; Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health Human Services, 540 Gaither Road, Rockville, MD 20850, USA (e-mail: Carolyn.clancy@ahrq.hhs.gov). 9. Bindman AB. The evolution of health services research. Health Serv Res. 2013;48(2 Pt 1):349–53. 5. Evans DC, Nichol WP, Perlin JB. Effect of the implementation of an enterprise-wide electronic health record on productivity in the Veterans Health Administration. Health Econ Policy Law. 2006;1(Pt 2):163–9. 4. Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med. 2003;348(22):2218–27. REFERENCES These changes represent an incredible opportunity for health care researchers to apply their skills and methods to rapid learning and evaluation, but this will require a participatory model that is quite different from research of the past.9 Research that involves ongoing consultation with stakeholders, including pa- tients, may be perceived as less efficient, but the potential for impact and engagement is enormous. Enlightened institutions will create alternative career 1. Clancy CM, Massion C. American women’s health care: a patchwork quilt with gaps. JAMA. 1992;268(14):1918–20. 2. Croghan TW, Brown JD. Integrating Mental Health Treatment Into the Patient Centered Medical Home. (Prepared by Mathematica Policy Re- search under Contract No. HHSA290200900019I TO2.) AHRQ Publication No. 10-0084-EF. Rockville, MD: Agency for Healthcare Research and Quality. June 2010. 3. Healthcare Quality and Disparities in Women: Selected Findings from the 2010 National Healthcare Quality and Disparities Reports. Fact Sheet. AHRQ Publication No. 11-0005-1-EF, March 2011. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/qual/ nhqrwomen/nhqrwomen.htm. Accessed April 10 2013. 3. Healthcare Quality and Disparities in Women: Selected Findings from the 2010 National Healthcare Quality and Disparities Reports. Fact Sheet. AHRQ Publication No. 11-0005-1-EF, March 2011. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/qual/ nhqrwomen/nhqrwomen.htm. Accessed April 10 2013. Research and Quality, Rockville, MD. http://www.ahrq.gov/qual/ nhqrwomen/nhqrwomen.htm. Accessed April 10 2013. S503 JGIM Clancy and Collins Sharp: Women’s Health During Health Care Transformation 7. Hayes PM. Serving women veterans: improving health of veterans through research collaborations. J Gen Intern Med. 2013. doi:10.1007/s11606- 013-2471-8. 4. Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med. 2003;348(22):2218–27. 8. Yano EM, Hayes P, Wright S, Schnurr PP, Lipson L, Bean-Mayberry B, Washington DL. Integration of women veterans into VA quality improve- ment research efforts: what researchers need to know. J Gen Intern Med. 2010;1:56–61. 5. Evans DC, Nichol WP, Perlin JB. Effect of the implementation of an enterprise-wide electronic health record on productivity in the Veterans Health Administration. Health Econ Policy Law. 2006;1(Pt 2):163–9. 8. Yano EM, Hayes P, Wright S, Schnurr PP, Lipson L, Bean-Mayberry B, Washington DL. Integration of women veterans into VA quality improve- ment research efforts: what researchers need to know. J Gen Intern Med. 2010;1:56–61. 6. Post EP, Metzger M, Dumas P, Lehmann L. Integrating mental health into primary care within the Veterans Health Administration. 6. Post EP, Metzger M, Dumas P, Lehmann L. Integrating mental health into primary care within the Veterans Health Administration. Fam Syst Health. 2010;28(2):83–90. 4. Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med. 2003;348(22):2218–27. 5. Evans DC, Nichol WP, Perlin JB. Effect of the implementation of an enterprise-wide electronic health record on productivity in the Veterans Health Administration. Health Econ Policy Law. 2006;1(Pt 2):163–9. 6. Post EP, Metzger M, Dumas P, Lehmann L. Integrating mental health into primary care within the Veterans Health Administration. Fam Syst Health. 2010;28(2):83–90. 7. Hayes PM. Serving women veterans: improving health of veterans through research collaborations. J Gen Intern Med. 2013. doi:10.1007/s11606- 013-2471-8. REFERENCES Fam Syst Health. 2010;28(2):83–90. 9. Bindman AB. The evolution of health services research. Health Serv Res. 2013;48(2 Pt 1):349–53.
https://openalex.org/W4368246541	https://bajangjournal.com/index.php/J-ABDI/article/download/4832/3592	Indonesian	null	PELATIHAN PEMBUATAN ES KRIM MANGROVE PADA KELOMPOK MITRA USAHA MINA SENTOSA PATUGURAN REJOSO PASURUAN	Jurnal Pengabdian Kepada Masyarakat	2,023	cc-by	2,924	Oleh Oleh Bambang Sutikno1, Sri hastari2, Yufenti Oktavia3 1,2,3Universitas Merdeka Pasuruan E-mail: 3oktaviavnty@gmail.com Abstract: The use of mangrove fruit as a basic ingredient for drinks has not been carried out optimally, especially for coastal communities which are rich in mangrove forests. Socialization and training activities related to beverage products from mangrove fruit which have high nutritional content need to be held. Socialization and training activities are packaged in the form of community service in Patuguran village, Rejoso sub-district, Pasuruan district. The training activities were carried out together with home industry (micro-enterprise) women who are members of the Mina Sentosa group, Tambakrejo hamlet, Patuguran village to make processed mangrove es krim with 3 flavors, namely chocolate flavor, strawberry flavor and melon flavor made from mangrove raw materials. The method used in training is lecture, practice and mentoring methods. Lecture method by providing basic theory in processing mangrove fruit into es krim, so that it has added value in terms of taste and nutrition. The practical method is training that emphasizes personal skills to be able to make frozen drinks in the form of es krim. Assistance is provided for business continuity in terms of marketing with the aim of creating self-sufficiency to seek sustainable profits. The output of this community service activity is to produce frozen beverage products in the form of es krim with three flavors as an innovative and nutritious product, so that they can increase the income of women who are members of the Mina Sentosa group. Article History: Received: 02-12-2022 Revised: 13-12-2022 Accepted: 23-12-2022 Keywords: Training, Outreach, Mangroves 6157 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 6157 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 PELATIHAN PEMBUATAN ES KRIM MANGROVE PADA KELOMPOK MITRA USAHA MINA SENTOSA PATUGURAN REJOSO PASURUAN Oleh Bambang Sutikno1, Sri hastari2, Yufenti Oktavia3 1,2,3Universitas Merdeka Pasuruan E-mail: 3oktaviavnty@gmail.com PENDAHULUAN Masyarakat pesisir di desa Patuguran, kecamatan Rejoso, kabupaten Pasuruan terdapat kelompok ibu-ibu yang mengolah hasil tambak dan keripik mangrove yang tergabung dalam usaha kelompok Mina Sentosa yang di ketua oleh Ibu Hj.Ilyas. Pada kelompok Mina Sentosa ini sering diberi pelatihan tentang pengolahan ikan bandeng dan buah mangrove oleh dinas perikanan kabupaten Pasuruan dan Dinas koperasi kabupaten Pasuruan. Pelaksanaan pengabdian pada tanggal 2 September 2022 dengan survey lokasi di dusun Tambakrejo desa Patuguran kecamatan Rejoso kabupaten Pasuruan. Setelah survey dilakukan proses pemanenan buah mangrove jenis bogem dan lindur untuk dijadikan bahan utama dalam http://bajangjournal.com/index.php/J-ABDI ISSN: 2797-9210 (Print) \| 2798-2912(Online) 6158 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 6158 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 pembuatan olahan es krim mangrove. Pada tanggal 3 September 2022 bertempat di rumah tokoh masyarakat (Ibu. H. Ilyas) bersama-sama ibu-ibu yang lain sebagai pelaku home industry (usaha mikro) yang tergabung dalam kelompok Mina Sentosa dari dusun Tambakrejo desa Patuguran untuk membuat olahan es krim mangrove dengan 3 rasa, yaitu rasa coklat, rasa strawberry dan rasa melon. Semua anggota yang tergabung dalam kelompok Mina Sentosa Tambakrejo bersama- sama dengan tim untuk membuat es krim mangrove, kemudian dilanjutkan dengan memberikan pelatihan cara pengemasan es krim pada cup agar lebih awet dan mudah dipasarkan. Berdasarkan hasil survey pendahuluan dalam kelompok Mina Sentosa menyatakan bahwa pemanfaatan buah mangrove oleh masyarakat sudah dilakukan khususnya dalam pembuatan aneka olahan keripik dengan berbagai rasa dan diproduksi untuk dipasarkan pada masyarakat umum khususnya kepada para pengunjung wisata mangrove, tetapi kondisi sampai saat ini kurang bisa berkembang dengan baik. Hal ini disebabkan karena pengetahuan yang minim dalam melakukan inovasi produk dari mangrove, adanya keterbatasan peralatan yang dimiliki usaha mikro yang tergabung dalam Mina Sentosa. Buah mangrove jenis pedada hanya diketahui oleh kelompok tertentu saja, sehingga buah mangrove yang ada tidak termanfaatkan secara optimal. j gg g y g p Melalui program pengabdian kepada masyarakat ini diharapkan masyarakat dapat mengembangkan produk minuman dari buah mangrove, supaya dapat meningkatkan ketrampilan, pengetahuan dan penghasilan kelompok Mina Sentosa dalam membuat aneka produk olahan buah mangrove. Hal ini sangat penting mengingat buah mangrove baik untuk kesehatan, seperti misalnya buah mangrove jenis pedada (Sonneratia caseolaris) dan lindur (Bruguiera ghymnorhiza) yang banyak tumbuh di daerah pesisir desa Patuguran sebagai area wisata hutan mangrove, serta jenis buah mangrove lainnya. METODE PELAKSANAAN KEGIATAN Metode Pendekatan Metode Pendampingan dilakukan dalam rangka untuk menjaga keberlanjutannya dari kegiatan usaha dan terhadap pemasaran hasil. Kegiatan pelatihan ini metode yang digunakan dalam memfasilitasi peningkatan ketrampilan kelompok Mina Sentosa dalam pengolahan minuman yang memanfaatkan sumber bahan baku lokal berupa buah mangrove yang ada di Patuguran adalah dengan metode pelatihannya adalah: 1. Metode Ceramah yaitu dengan memberikan pemahan tentang teori dasar dari pembuatan produk minuman beku berupa es krim dengan bahan baku tepung buah mangrove dan memberikan tambahan bahan dasar berupa coklat, strawberry dan buah melon, sehingga mempunyai nilai dari segi jenis, rasa, maupun nutrisi dari produk es krim tersebut. 2. Metode praktik yaitu dilakukan dengan memberikan pelatihan yang ditekankan pada kemampuan ketrampilan individu kelompok dalam mengolah produk minuman es krim yang lebih kreatif dan inovatif, sehingga mempunyai aneka rasa khas dalam satu produk es krim, dengan kandungan nutrisi yang bervariasi dan sangat bermanfaat bagi kesehatan, dan memiliki nilai lebih (added value) dengan metode praktik pembuatan secara langsung. Pelatihan ini menekankan pada rasa kebersamaan dengan mengedepankan partisipasi aktif dari peserta pelatihan yang tergabung dalam kelompok usaha Mina Sentosa. PENDAHULUAN Menurut Bandararanayake (2002) menyatakan bahwa kandungan antioksidan yang terdapat pada buah mangrove dapat mencegah adanya radikal bebas yang sangat berbahaya yang ada dalam tubuh, sedangkan kandungan serat pangan pada buah mangrove bersifat mengenyangkan dan mampu menurunkan kadar gula darah dan kolesterol, dengan demikian buah mangrove sangat baik bagi kesehatan. Manulu (2011) bahwa buah mangrove juga kaya akan kandungan provitamin A, vitamin B5, vitamin B2, vitamin C dan juga banyak mengandung beberapa jenis mineral yang sangat baik untuk Kesehatan. Berdasarkan dari nilai potensi gizi yang terkandung pada buah mangrove akan sangat bermanfaat apa bila ditambahkan beberapa bahan dasar dalam pembuatan produk olahannya, dengan demikian akan menambah nilai gizi yang lebih baik dalam produk olahannya serta memberikan aneka rasa yang akan membuat masyarakat luas semakin tertarik untuk mengkonsumsinya dan mendorong dari para usaha mirkro untuk melakukan diversifikasi olahan buah mangrove sebagai bahan pangan. Selain itu keinginan masyarakat terus menuntut adanya inovasi terutama yang berkaitan dengan produk olahan minuman cair maupun beku (es krim). Beraneka ragam produk minuman yang berbahan baku buah mangrove sebagai produk unggulan desa wisata. Memanfaatkan potensi sumber daya alam yang ada secara optimal yang akan menambah potensi sumber ekonomi masyarakat secara berkelanjutan dan berkesinambungan, selain itu juga dapat menambah variasi jenis minuman olahan sehat berbahan baku buah mangrove. Beberapa jenis produk olahan yang dapat dibuat dengan ISSN: 2797-9210 (Print) \| 2798-2912(Online) http://bajangjournal.com/index.php/J-ABDI 6159 J-Abdi 6159 bd 6159 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 bahan baku buah mangrove yaitu membuat sirup, selai dan dodol, es puter, es krim dan keripik dari buah mangrove jenis pedada (Sonneratia caseolaris). Hasil analisis lingkungan internal dan eksternal terkait potensi desa Patuguran dan berdasarkan hasil wawancara dengan usaha kelompok Mina Sentosa, maka pada kegiatan pengabdian masyarakat ini akan dilakukan pelatihan dan pendampingan pengolahan buah mangrove menjadi produk minuman yaitu es krim yang diambil dari buah mangrove jenis bogem atau pedada (Sonneratia caseolaris). j g p ( ) Kegiatan pelatihan ini diharapkan dapat memperkaya aneka olahan yang mampu dibuat oleh masyarakat desa Patuguran yang berbahan baku buah mangrove menjadi produk minuman dan akan berkembang menjadi produk unggulan di desa Patuguran dalam menunjang wisata edukasi lingkungan berbasis hutan mangrove. Buah mangrove jenis pidada bisa digunakan sebagai dasar masakan es krim melalui proses yaitu mula-mula buah mangrove pidada direbus sekitar tiga jam. Setelah empuk, buah dibuka dan dibuang bagian dalamnya yang mengandung rasa pahit. Buah yang empuk lantas diblender dan diberi istilah bubur mangrove. http://bajangjournal.com/index.php/J-ABDI Persiapan produksi Persiapan produksi Persiapan produksi Pelaksanaan pelatihan pembuatan es krim dari buah mangrove dengan bahan dan alat sebagai berikut: Persiapan produksi Pelaksanaan pelatihan pembuatan es krim dari buah mangrove dengan bahan dan alat b i b ik p p Pelaksanaan pelatihan pembuatan es krim dari buah mangrove dengan bahan dan ala sebagai berikut: Bahan-bahan yang digunakan: 6. Garam secukupnya atau 25 gram HASIL DAN PEMBAHASAN Metode Pelaksanaan Sosialisasi dilakukan di Pendopo Balai Desa Patuguran, sedangkan tempat praktik pelatihan pembuatan es krim dilakukan di rumah ketua kelompok Mina Sentosa yaitu, Ibu Hj. Ilyas di dusun Tambakrejo Desa Patuguran Kecamatan Rejoso Kabupaten Pasuruan dengan jumlah peserta dari mitra sebanyak 15 orang. Kegiatan pelatihan dari praktik pembuatan produk es krim, maka ibu-ibu peserta pelatihan (Kelompok Mina Sentosa) diberi penjelasan dasar teori tentang manfaat buah mangrove sebagai bahan baku http://bajangjournal.com/index.php/J-ABDI ISSN: 2797-9210 (Print) \| 2798-2912(Online) ISSN: 2797-9210 (Print) \| 2798-2912(Online) ISSN: 2797-9210 (Print) \| 2798-2912(Online) 6160 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 6160 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 pangan yang sangat bermanfaat bagi Kesehatan. Sosialisasi ini diberikan oleh tim pendamping dengan tujuan agar para peserta pelatihan dapat mempunyai gambaran mengenai tatacara saat praktik berlangsung nantinya. Sebelum pelaksanaan pelatihan, kegiatan diawali dengan melakukan koordinasi dengan mitra Mina Sentosa yang beranggotakan Ibu-ibu kelompok usaha dalam rangka untuk menentukan hari dan jam pelaksanaan kegiatan. Setelah ada kata kesepakatan antara tim pelaksana kegiatan dengan mitra Mina Sentosa maka dilanjutkan dengan melakukan praktik untuk pembuatan produk minuman beku dari buah mangrove yaitu es krim dengan tiga rasa. Sebelum pelaksanaan praktik pelatihan yang dilakukan pertama kali adalah pengumpulan buah mangrove jenis pidada setelah itu buah mangrove direbus sekitar tiga jam sampai kondisi buahnya lunak. Setelah lunak, buah dibuka untuk membuang bagian dalamnya yang memiliki rasa yang pahit. Buah yang sudah lunak tadi kemudian diblender sampai menjadi bubur dan istilah bubur mangrove. Bubur inilah yang menjadi bahan dasar olahan mangrove (sebagai bahan dasar menjadi tepung mangrove). j p g g ) Gambar 1 Foto bersama pada acara sosialisasi Pembuatan Es Krim Mangrove Persiapan produksi Pelaksanaan pelatihan pembuatan es krim dari buah mangrove dengan bahan dan alat sebagai berikut: Gambar 1 Foto bersama pada acara sosialisasi Pembuatan Es Krim Mangrove d k Foto bersama pada acara sosialisasi Pembuatan Es Krim Mangrove http://bajangjournal.com/index.php/J-ABDI 6161 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 Prosedur pembuatan es krim mangrove 1. Semua bahan dicampur menjadi satu (kecuali emulsifier) 2. Campuran direbus dan diaduk sampai mengental p p g 3. Campuran yang mau mengental diangkat dari kompor dan biarkan sampai kondisi dingin, kemudian dimasukkan kedalam Freezer 3. Campuran yang mau mengental diangkat dari kompor dan biarkan sampai kondisi dingin, kemudian dimasukkan kedalam Freezer kemudian dimasukkan kedalam Freezer 4. Campuran yang ada dalam freezer ditunggu sampai hampir membeku 5. Kemudian campuran di keluarkan dari freezer dan ditambahkan emulsifier 6. Campuran di aduk dengan kecepatan tinggi menggunakan mixer 7. Campuran yang dimixer akan mengembang menjadi 2 kali lipat 8. Tambahkan perasa dan proses pembuatan es krim sudah selesai. 9 E k i i di kk k t d t di jik 8. Tambahkan perasa dan proses pembuatan es krim sudah selesai. 9. Es krim siap dimasukkan kemasan untu dapat disajikan. Bahan baku dan alat serta proses pembuatan produk es krim tersebut di atas, tentunya sangat cocok untuk diproduksi dan dikembangkan oleh para pelaku usaha mikro yang tergabung dalam mitra Mina Sentosa, hal ini karena proses pembuatan es krim sangat sederhana dan bahan bakunya mudah didapat termasuk peralatannya, sehingga kalau dilakukan dengan sungguh-sungguh dengan pembinaan yang intensif akan dapat dijadikan sebagai penambah penghasilan dari kelompok Mina Sentosa dengan sistem usaha yang efektif dan efisien. Proses pembuatan es krim buah mangrove dan hasil produk es krim buah mangrove disajikan pada Gambar 2 di bawah ini. Proses dan hasil pembuatan es krim dengan bahan baku buah mangrove dengan rasa coklat, strowbery dan rasa melon. Gambar 2 Proses Dan Hasil Pembuatan Es Krim Mangrove dengan Penambah Rasa Coklat, Strowbery dan Melon. Proses Dan Hasil Pembuatan Es Krim Mangrove dengan Penambah Rasa Coklat, Strowbery dan Melon. Produksi es krim tersebut merupakan hasil yang dipraktikan pada saat pelatihan dengan anggota kelompok Mina Sentosa desa Patuguran dan kegiatan dan hasilnya mendapat respon dan persepsi tentang prospek es krim yang sangat positif di kalangan peserta pelatihan, semua peserta dapat mengikuti dengan tekun dan seksama dalam mempraktikan proses pembuatan es krim buah mangrove. http://bajangjournal.com/index.php/J-ABDI Berdasarkan hasil interview di saat awal pelaksanaan tentang pengetahuan dari ibu-ibu peserta pelatihan mengenai produk es krim dan cara membuat es krim dari buah mangrove, yang rata-rata banyak yang belum tahu dan belum paham mengenai pokok materi pelatihan, tetapi setelah adanya sosialisasi dan praktik langsung pembuatan es krim mangrove maka hasilnya dapat berupa pengetahuan dan pemahaman dari semua peserta dan peserta pelatihan sudah memahami dengan persepsi yang sama dalam proses pembuatan es krim mangrove dan para peserta juga merasa puas dalam mengikuti proses pelatihan pembuatan es krim mangrove, hal ini juga akan memberikan harapan yang positif untuk usaha kedepannya, karena ditunjang oleh bahan dan peralatan yang digunakan semua mudah di dapat dan tersedia. http://bajangjournal.com/index.php/J-ABDI ISSN: 2797-9210 (Print) \| 2798-2912(Online) ISSN: 2797-9210 (Print) \| 2798-2912(Online) ISSN: 2797-9210 (Print) \| 2798-2912(Online) 6162 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 6162 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 Menurut m i t r a d a r i kelompok usaha Mina Sentosa, pelatihan yang diberikan dapat memberikan manfaat yang besar dalam meningkatkan pendapatan kelompok, diantaranya yaitu menambah wawasan ilmu yang dapat menunjang pengembangan ide kewirausahaan. Pengetahuan mitra kelompok usaha Mina Sentosa dan masyarakat pesisir secara umum di desa Patuguran dapat dinyatakan masih banyak yang belum mengetahui seputar es krim, hal ini dapat disebabkan karena wilayah mereka yang agak jauh dari pusat kota Pasuruan. Hal tersebut dimungkinkan karena situasi dan kondisi mereka yang kesehariannya hanya berprofesi sebagai petani tambak dan nelayan, walaupun disekitar mereka terdapat banyak tanaman mangrove, akan tetapi mereka masih belum bisa memanfaatkan buah mangrove tersebut secara maksimal. Pentingnya transfer ilmu dan teknologi mengenai pengolahan hasil buah mangrove khususnya pada kali ini yaitu buah pidada yang banyak terdapat pada kawasan pesisir Desa Patuguran. Beberapa pendapat mitra tentang kegiatan pelatihan ini yaitu menambah pengetahuan, menambah ide jualan, menambah penghasilan, cara berjualan, menambah wawasan dan mereka senang ada pengalaman baru, ada produk baru dalam bentuk es krim, senang karena dapat ilmu tambahan buat usaha, menambah wawasan, menambah pengalaman bagi masyarakat pesisir, kegiatan ini sangat positif dan sangat bermanfaat, senang karena juga bisa ilmu baru tentang pemasaran produk yang selama ini dilakukan hanya menggunakan perasaan/insting saja. Harapannya program selanjutnya perlu adanya pengembangan dan diversifikasi produk dari buah mangrove agar dapat menjadikan masyarakat peisisir Desa Patuguran menjadi masyarakat yang mandiri khususnya dari mitra yang merupakan perkumpulan ibu-ibu yang memiliki usaha yang tergabung sebagai anggota kelompok Mina Sentosa. Kesimpulan Kegiatan pengabdian pada bulan September 2022 di Desa Patuguran, Kecamatan Rejoso, Kabupaten Pasuruan, khususnya pada kegiatan pelatihan pembuatan es krim dari buah mangrove dirasakan sangat bermanfaat, hal ini dapat ditunjukkan dengan 87% dari kondisi awal dari mitra yang belum mengetahui proses dan bahan pembuatan es krim dari buah mangrove dengan tiga rasa (Coklat, Strowbery dan Melon). Pelatihan ini diharapkan, para pelaku usaha mikro yang tergabung sebagai anggota kelompok Mina Sentosa merasa paham tentang pembuatan es krim mangrove dan merasa bertambah ilmu pengetahuannya tentang manfaat dari buah mangrove sebagai sumber bahan baku pangan yang kaya akan gizi sehingga tergerak untuk menambah produk usahanya khususnya produk minuman beku dan cair yang berbahan baku buah mangrove dan dapat dijadikan sebagai produk unggulan desa Patuguran yang akan berdampak memberikan tambahan penghasilan bagi pelaku usaha mikro secara berkelanjutan. Saran Perlu adanya program pemberdayaan dan pendampingan secara keberlanjutan yang bersifat lintas sektoral (Tri Helik) sehingga dapat menjadikan masyarakat menjadi mandiri dalam berwirausaha dan berdaya dalam memanfaatkan potensi sumberdaya alam pesisir secara maksimal dan berkesinambungan. DAFTAR REFERENSI [1] Bengen, D.G. 2001. Pengelolaan Sumberdaya Wilayah Pesisir Secara Terpadu, Berkelanjutan dan Berbasis Masyarakat. Makalah pada Sosialisasi 22 ) ( Pemberdayaan Perempuan Pesisir Melalui Pengembangan Volume 1, Number 1, November 2018 [2] Hamsah. 2013. Karakteristik Sifat Fisikokimia Tepung Buah Pedada (Sonneratia Caseolaris). (http://repository.unhas.ac.id/bitstream/handle/123456789/5977/HAMSAH%20% 28G311%2009%20991%29.pdf?sequence=1. diakses 10 Maret 2014). [3] Ariyah, Widjanarko, S.B. , Estiasih, T. and Yunianta. 2014. Hypoglycemic effect of Pedada (Sonneratia caseolaris) Fruit Flour (PFF). in alloxan-induced diabetic rats. International Journal of PharmTech Research. Vol.7, No.1, pp 31-40. pp [4] Jariyah, Azkiyah, L., Widjanarko, S.B., Estiasih, T., Yuwono, S.S and Yunianta. 2013. Hypocholesterolemic Effect of Pedada (Sonneratia caseolaris) Fruit Flour in Wistar Rats. International Journal of Pharm Tech Research. vol.5, No.4, pp.1619-1627. J pp [5] Jariyah, S. B. Widjanarko, Yunianta and T. Estiasih. 2014. Hypoglycemic effect of Pedada (Sonneratiacaseolaris) Fruit Flour (PFF) inalloxan-induced diabetic rats. International Journal of PharmTech Research. 7(1):31-40. [6] Jariyah, S. B. Widjanarko, Yunianta and T. Estiasih. 2015. Phytochemical and acute toxicity studies of ethanol extract from Pedada (Sonneratia caseoalris) fruit flour (PFF). International Journal on Advanced Science Engineering Information Technology. 5(2):39-42. gy ( ) [7] Manalu, R. D. E. 2011. Kadar beberapa vitamin pada buah pedada (Sonneratia caseolaris) dan hasil olahannya. (Skripsi) Institut Pertanian Bogor. Bogor. [8] Padaga, Masdian dan Sawitri, Manik Eirry. 2005. Membuat Es Krim yang Sehat. Surabaya : Trubus Agrisarana. [8] Padaga, Masdian dan Sawitri, Manik Eirry. 2005. Membuat Es Krim yang Sehat. Surabaya : Trubus Agrisarana. [9] Priyono, Ilminingtyas, Mohsoni, Sri mulyani, Hakim. 2010. Beragam Produk Olahan Berbahan Dasar Mangrove. Data Modul g [10] Sugiyono. 2010. Metode Penelitian Kuantitatif, Kualitatif dan R & B. Ban [10] Sugiyono. 2010. Metode Penelitian Kuantitatif, Kualitatif dan R & B. Bandung : Alfabeta. [11] Undang-Undang Nomor 6 Tahun 2014 tentang Desa. [ ] g y g [11] Undang-Undang Nomor 6 Tahun 2014 tentang Desa. [11] Undang-Undang Nomor 6 Tahun 2014 tentang Desa. ISSN: 2797-9210 (Print) \| 2798-2912(Online) http://bajangjournal.com/index.php/J-ABDI 6163 J-Abdi 6163 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 6163 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 http://bajangjournal.com/index.php/J-ABDI ISSN: 2797-9210 (Print) \| 2798-2912(Online) 6164 J-Abdi Jurnal Pengabdian Kepada Masyarakat Vol.2, No.9 Februari 2023 HALAMAN INI SENGAJA DIKOSONGKAN http://bajangjournal.com/index.php/J-ABDI ISSN: 2797-9210 (Print) \| 2798-2912(Online) http://bajangjournal.com/index.php/J-ABDI
https://openalex.org/W4317513973	https://kclpure.kcl.ac.uk/ws/files/198371348/1_s2.0_S2352152X23000828_main.pdf	English	null	Domestic thermal energy storage applications: What parameters should they focus on?	Journal of energy storage	2,023	cc-by	11,792	Citation for published version (APA): Ryland, M., & He, W. (2023). Domestic thermal energy storage applications: What parameters should they focus on? Journal of Energy Storage, 60, Article 106685. https://doi.org/10.1016/j.est.2023.106685 Citing this paper Pl h C t g t s pape Please note that where the full-text provided on King's Research Portal is the Author Accepted Manuscript or Post-Print version this may differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version for pagination, volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are again advised to check the publisher's website for any subsequent corrections. Citation for published version (APA): Ryland, M., & He, W. (2023). Domestic thermal energy storage applications: What parameters should they focus on? Journal of Energy Storage, 60, Article 106685. https://doi.org/10.1016/j.est.2023.106685 A R T I C L E I N F O Keywords: Thermal energy storage Domestic Heating Decarbonisation Thermal energy storage (TES) is required to allow low-carbon heating to meet the mismatch in supply and demand from renewable generation, yet domestic TES has received low levels of adoption, mainly limited to hot water tanks. Current reviews and studies primarily focus on the comparison of storage materials neglecting the performances at a system level and analysis studies tend to solely look at hot water tanks, missing the key technology developments in thermal storage systems which are under development. Therefore, this paper in vestigates performance and cost variations of TES from material-level to system-level analysis and assesses im pacts of emerging heat storage technologies. By simulating different types of TES materials and varied system integration options, a significant reduction in energy densities and increase in specific costs of TES systems were found compared to the material-level analysis. Direct electrical heating has much greater potential to integrate with TES from its high operating temperature with TES compared to heat pumps or solar thermal which are constrained to lower temperatures. TES properties are simulated in various scenarios in a domestic heating techno-economic framework. It was found that for heat pumps there is economically-limited potential for TES, even if very high energy densities are possible. In addition, the priority for TES coupling with heat pumps is low capital cost, although current high tariff rates due to the energy-crisis do improve economic viability of TES. On the other hand, with direct electrical heating, high energy density is the most valuable parameter for TES, as it allows significant quantity of demand to be shifted to very low-tariff times, in particular for low demand dwellings where negligible amounts of peak electricity could be required for heating. Domestic thermal energy storage applications: What parameters should they focus on? Michael Ryland a, Wei He a,b,* a School of Engineering, University of Warwick, Coventry CV4 7AL, United Kingdom b School of Engineering, King’s College London, London WC2R 2LS, United Kingdom Michael Ryland a, Wei He a,b,* a School of Engineering, University of Warwick, Coventry CV4 7AL, United Kingdom b School of Engineering, King’s College London, London WC2R 2LS, United Kingdom Michael Ryland a, Wei He a * Corresponding author at: School of Engineering, King’s College London, London WC2R 2LS, United Kingdom. E-mail addresses: michael.ryland@warwick.ac.uk (M. Ryland), wei.he.2@warwick.ac.uk (W. He). General rights General rights Copyright and moral rights for the publications made accessible in the Research Portal are retained by the authors and owners and it is a condition of accessing publications that users recognize and abide by the legal requirements associ ral rights for the publications made accessible in the Research Portal are retained by the authors and/or other copyright condition of accessing publications that users recognize and abide by the legal requirements associated with these right •Users may download and print one copy of any publication from the Research Portal for the purpose of private study or research. •You may not further distribute the material or use it for any profit-making activity or commercial gain •You may freely distribute the URL identifying the publication in the Research Portal •Users may download and print one copy of any publication from the Research Portal for the purpose of private study •You may not further distribute the material or use it for any profit-making activity or commercial gain y y p g y •You may freely distribute the URL identifying the publication in the Research Portal Take down policy If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. p y If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we w the work immediately and investigate your claim. Download date: 24. Oct. 2024 Journal of Energy Storage 60 (2023) 106685 Available online 19 January 2023 2352-152X/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1016/j.est.2023.106685 Received 31 October 2022; Received in revised form 23 December 2022; Accepted 9 January 2023 2352-152X/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Available online 19 January 2023 2352-152X/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). * Corresponding author at: School of Engineering, King’s College London, London WC2R 2LS, United Kingdom. E-mail addresses: michael.ryland@warwick.ac.uk (M. Ryland), wei.he.2@warwick.ac.uk (W. He). https://doi.org/10.1016/j.est.2023.106685 Received 31 October 2022; Received in revised form 23 December 2022; Accepted 9 January 2023 2.2. TES integration into domestic heating framework The study will then go on to look at how TES can integrate into a domestic heating application with hourly simulations across a year for a dwellings demand met by TES that is heated with DEH or ASHP, comparing economics and environmental factors. This study will then look at how changing parameters of TES alters the system viability, to demonstrate which parameters are most valuable to reducing costs of TES. A consumer centric mathematical model to simulate domestic heating across the year at an hourly resolution is used as introduced in previous work [10,16]. In this paper, we cover the range of technology readiness level from existing TES technologies of hot water storage tanks and electric storage heaters, emerging TES technologies such as high temperature SHS and LHS which are just starting the commercialisation process, and potential future technologies from theoretical improvements to determine what parameters are most valuable for TES. A hot water tank TES is added into a domestic heating simulation framework, where 0.1 to 0.5m3 sizes are analysed. Additionally theoretical changes to TES parameters of energy densities, CapEx, storage temperature and insulation value are investi gated. This enables an understanding of which aspects are useful for TES rather than examining specific materials/systems, which has already been done in existing TES studies. A default temperature of 51 ◦C is used for the TES storage temperature, but higher temperatures of up to 500 ◦C are considered in the simulations, and up to 1500 ◦C in initial material and system comparisons. The framework holistically considers each combination of heating technologies, ancillary solar and TES sizes, and tariffs. The CapEx and the single years’ operating expenditure (OpEx) are then used to calculate the 20-year Net Present Cost (NPC). A time period, t, of one year and a discount rate, rd, of 0.035 are used for the NPC Eq. (1). The OpEx, CapEx, NPC, and emissions values are used for technology comparisons. Emis sions include operational emissions and embodied emissions, to give an equivalent annual emission. Inputs to the model are: dwelling location; number of occupants; desired thermostat temperature; dwelling floor area, and annual space heating demand, the latter two can be found on the dwelling’s energy performance certificate in many countries. To demonstrate the sensitivity of TES parameters a case study is complete in the UK for multiple scenarios. 1. Background pumps or direct electrical heating (DEH), and solar thermal is required. DEH makes use of resistance heaters which have close to 100 % effi ciency and can be flexible and fast responding with no or minimal moving parts and low capital expenditure (CapEx). Heat pumps use electrical energy as an input to make use of outdoor low-grade heat, allowing much higher efficiencies than DEH, but with higher CapEx requirements. Solar thermal collectors directly use the irradiation from the sun and therefore have no flexibility or control on when they generate heat. As electrified heating, heat pumps or DEH, needs to use high amounts of renewable generated electricity to reduce emissions, both electrified heating and solar thermal face the challenge of a mismatch in the supply of variable renewable energy against heating demands, from seasonal to diurnal timescales. Heating currently accounts for 37 % of worldwide greenhouse gas emissions mostly through burning fossil fuels, which needs to be reduced to meet countries’ net zero targets [1]. Fossil fuel fired heating is also the major source of health damaging air pollution due to insuf ficient combustion. Globally, residential and commercial building en ergy use is the largest source that contributes nearly one-third of deaths related to air pollution [2]. In the UK, where burning wood is a very small proportion of the total heat source, domestic combustion emis sions from heating has been increasing between 2005 and 2020 and is responsible for 20 % of the country’s PM 2.5 and PM 10 (particulate matter less than or equal to an aerodynamic diameter of 2.5 or 10 μm respectively), larger than the transportation sector [3]. Thermal energy storage (TES) can help to play a key role in meeting this mismatch, by storing the energy at the time of generation and allowing it to be used at a time of demand. However, the transition to To reduce these heating emissions a shift towards alternative environmental-friendly heating such as electrified heating, of heat Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He identified in this study likely will be of interest to homeowners, manu facturers of heating systems, and policymakers in energy sectors. identified in this study likely will be of interest to homeowners, manu facturers of heating systems, and policymakers in energy sectors. 1. Background low-carbon heating technologies and TES has been slow in many regions due to various technical and economic challenges [4]. Both TES and low- carbon heaters face challenges of high investment cost and space constraint challenges in retrofitting [4]. Currently, water tanks are the most used domestic TES technology, but water storage suffers from low energy density, so the storage usually only provides domestic hot water that is about 15 % of domestic heating demand [4]. The only other TES that has seen reasonable uptake is in lower demand dwellings that have historically been a popular choice for electrical storage heaters which use ceramic materials and so can reach elevated temperatures, although this technology differs to other TES as it is also acts as the radiator [5]. 2. Methodology In order to decarbonise heating, the source of the energy needs to come from low-carbon sources. As this is generally considered to be from variable renewable energy generation, TES is required to meet this mismatch. TES needs to be able to store sufficient capacity of energy to allow a shift in demand, it also aims to do this in the most efficient way as possible to maximise use of renewable generation. There are various aspects to a TES round trip efficiency, as this includes charging effi ciency, storage efficiency and discharging efficiencies, all of which need to be considered. gy Previous studies of TES technologies typically focus on the material level analysis of the storage materials [6–9], which may not create feasible solutions to scale up to commercialisation and does not include economic parameters of investment and operational costs, and associ ated emissions, for consumers using these TES concepts. Analysis studies of TES alongside low-carbon heating technologies that have been completed typically focuses only on water tanks as TES storage material and give little comparison of TES technologies and parameters other than variation in sizes [10–12]. Yet new TES designs are emerging and entering the marketplace. Emerging TES technologies include the use of higher temperature sensible heat storage (SHS) which are smartly charged at off-peak times using modern tariffs and, have significantly higher storage capacities than the traditional electrical storage heaters [13,14]. In addition to SHS, latent heat storage (LHS) technologies have begun commercialisation by storing thermal energy from the phase change of salt hydrates without the need for higher temperatures [15]. 2.1. Analysis of TES applications: from materials to systems First an analysis of existing and emerging TES materials and TES systems will be undertaken from a comprehensive review of the litera ture, covering the full spectrum SHS, LHS, and TCS. Comparisons of the TES will be made at a material level and system level to compare how this alters the position of the technologies, instead of purely focusing on the material level as current studies tends to. The system level analysis will include manufacturers data on traditional hot water tanks and electrical storage heaters as current TES technologies, as well as emerging commercial products that target high efficiency and storage densities that are using SHS at higher temperatures with high quality insulation [13,14], and LHS systems using salt as the phase change material designed for domestic heating application with melt tempera tures setup to efficiently store energy around the domestic demand temperatures [15]. Analysis will also be broken down showing how the TES technologies alter in their performance depending on if they are coupled with the prominent low-carbon heating technology of an air source heat pump (ASHP) with its limited operating temperatures, or with the more flexible and lower CapEx DEH which is better suited to lower-demand dwellings and can achieve higher temperatures. Here, we examine emerging domestic TES technologies and concepts and their integration with renewable and electrification heater options while also exploring the impact of power system decarbonisation on emissions of future domestic heating. In particular, we focus on the trade-offs between costs and emissions that consumers face in selecting a low-carbon alternative. Although cost is not the sole influence on con sumer purchasing decisions, low-carbon heating and TES will only achieve a dominant market share if they are affordable to most of the population. We address these issues by comprehensively examining TES technologies and concepts, covering all prominent options: SHS, LHS, and thermo-chemical storage (TCS). We take a step back from the existing thorough material analysis which focuses on how specific ma terials differ, instead focusing on the integration of the technologies at a system level and applying them to domestic heating applications from the consumer’s perspective. Table 3 Table 3 Technology capital expenditure equations. Source Formula Equation number [25] TES Capex = 2068.3 × Vtes0.553 (11) [26] DEH Capex = 1100 (12) [23] ASHP Capex = ( 200 + 4750/Php,th 1.25) × Php,th + 1500 (13) [26,27] Gas Boiler Capex = { 15000 + Qsh,epc/25; If Qsh,epc ≤25000 2500; If Qsh,epc > 25000 (14) Hot water daily volumetric demand, Vt in litres, is determined from BREDEM assumptions relative to the amount of residents in the dwelling [18]. Geographical monthly cold water temperatures Tcw, m in ◦C and hourly ratios of daily demand Rdhw, h i from Energy Saving Trust are used alongside BREDEM monthly ratios in daily hot water demands Rdhw, m [18,20]. Combining these parameters gives the hourly hot water de mand Qdhw i in kWh. 0.5m3 considered. All dwellings use an average thermal efficiency of 1.85 W/m2 K, then dwelling size is adjusted to match different percen tiles of UK homes heating demands. Resulting in the average demand dwelling set to 87m2, lower 10th percentile demand to 31m2, lower 25th percentile demand to 52m2, and high 75th percentile demand at 114m2 as determined in previous work [10]. Emissions are also taken from previous work [10]. Heating demands are met by either the baseline natural gas boiler using an efficiency of 90 %, DEH with an efficiency of 100 % or ASHP which has varying Coefficient of Performance (COP) as shown in Eq. (7). TES is simulated as stratified hot water tanks at sizes from 0.1 to 0.5m3 in 0.1m3 steps, but with the ability for its parameters to be changed to test their sensitivities on the TES integrated systems performance. The TES maximum thermal energy capacity Qtes i in kWh is calculated depending on the TES volume in that simulation Vtes in m3, and the TES specific heat capacity Cp in kJ/kg ◦C which is nominally set to 4.18 for water unless stated otherwise, and relative to a minimum useful tem perature of 40 ◦C. The TES is simulated as stratified hot water tanks with upper and lower heat losses Qtes, up i and Qtes, lower i in kWh, respectively above and below the thermocline height ht, c i at that time, which is a ratio of the TES height htes in m. Table 3 TES losses are also relative to its temperatures Ttes, up i and Ttes, low i in ◦C above and below the thermocline respectively, the radius of the TES rTES in m, and the thermal efficiency of the TES Utes in W/m2 ◦C that is nominally set to 1.3 unless stated otherwise (Table 2). The CapEx for each of the technologies are shown in Table 3. p To promote the shifting of energy to consumers, which is the main function of TES and other energy storage, variable rate tariffs are used as these create low electricity rates at times of lower demand. A range of different tariffs are considered, the flat rate tariff is the only tariff which does not promote shifting demands due to a constant rate across the day. Night off-peak tariff is a traditional two rate tariff, with 7 h of low-cost electricity at typical low demand times in the night, but a higher day rate than the flat tariff. A more modern version of this is the EV off-peak tariff, which has a shorter four-hour window of very low rates. Finally, a variable, day ahead, time of use tariff is also considered, which has a different rate for each hour of the day and changes every day depending on supply and demands. As the tariffs used in this study are existing tariffs available to customers, they all include sufficient costs to cover demand/distribution charges, energy generators and supplier costs and profits, and any other associated costs with purchasing elec tricity from a national electrical network. This paper then also considers the comparison of these tariffs at the pre-energy crisis low rates from 2020 to the current high costs tariffs from 2022 to determine how the tariff changes alter the position of TES, as shown in Table 4. The tariffs selected are the lowest rate tariffs of that structure available across the year 2020 and 2022, as any consumer is likely to select the lowest rate tariff available to them. Apart from the variable time of use tariff which uses data with the changing rate for each half hour across the two years. The CapEx for each of the technologies are shown in Table 3. Table 3 The framework created offers the ability to be easily adopted for any home and set of personal demands, by adjusting the dwelling inputs and locations for heating demands and in scaling the typical baseload elec tricity and transport demands. Case studies are used in the paper to demonstrate the functionality of the framework using an average, low and high demand UK dwellings are considered at a central England location of Coventry (3717 heating degree days), with two occupants, using a thermostat temperature of 20.0 ◦C, and a maximum TES size of Table 1 Key heating demand equations. Source Formula Equation number [21] Qhl i = Ad × Ud × (Tin i −Tout i )/1000 (2) [17] Cd = (250 × Ad)/3600 (3) Tin i = Tin i−1 + (Qhl i + Gs i + Gm)/Cd (4) Qi sh = { ( Ti d −Ti in ) × Cd; If Ti d > Ti in 0; If Ti d ≤Ti in (5) [22] Qdhw i = (Vt × 4.18 × (Thw −Tcw, m)/3600) × Rdhw, m × Rdhw, h i (6) Table 1 Key heating demand equations. 2.2. TES integration into domestic heating framework We evaluate cost and emission impacts of TES on domestic heating decarbonisation to answer the overarching questions: why TES has not seen more widespread adoption beyond basic hot water tanks, and what parameters are the most valuable for TES concepts to improve their economic viability? In addition to addressing these questions, we also find the use of TES in domestic heating systems can ensure heating se curity and provide benefits to wider energy system operations and decarbonisation, while changes in carbon emissions and costs vary greatly depending on heater and TES choices. The insights and choices NPC = ∑ n t=0 OpEx (1 + rd)t + CapEx (1) (1) The heating model is described in detail in previous work [10], The heating model is described in detail in previous work [10], The heating model is described in detail in previous work [10], The heating model is described in detail in previous work [10], 2 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He where the thermal efficiency of the dwelling is back-calculated from the input data, then a higher resolution space heating demand can be calculated using calculations and assumptions from SAP [17] and BREDEM [18] and using location specific reanalysis weather dataset from Renewable Ninja for the year 2019 [19]. Heat pumps are set to operate at a constant indoor temperature throughout the day, due to their low thermal power, and other heating devices set the target ther mostat temperature from 07:00 to 22:00. The key heating demand equations are shown in Table 1. Table 2 Key technology equations. Key technology equations. Source Formula Equation number [23] COPASHP = 6.81 −0.121 × (Thw −Tout i ) + 0.00063 × (Thw −Tout i )2 (7) [22] Qtes i = Vtes × 1000 × Cp × (Thw −40)/3600 (8) [24] Qtes, up i = (Ttes, up i −Tin i ) × Utes × (π × 2r × (ht, c i × htes) + π × rtes 2 )/1000 (9) [24] Qtes, lower i = (Ttes, low i −Tin i ) × Utes × (π × 2r × ((1 −ht, c i ) × htes) + π × rtes 2 )/1000 (10) Source Formula Equation number [23] COPASHP = 6.81 −0.121 × (Thw −Tout i ) + 0.00063 × (Thw −Tout i )2 (7) [22] Qtes i = Vtes × 1000 × Cp × (Thw −40)/3600 (8) [24] Qtes, up i = (Ttes, up i −Tin i ) × Utes × (π × 2r × (ht, c i × htes) + π × rtes 2 )/1000 (9) [24] Qtes, lower i = (Ttes, low i −Tin i ) × Utes × (π × 2r × ((1 −ht, c i ) × htes) + π × rtes 2 )/1000 (10) Where Qhl i is the heat loss from the dwelling at that time interval in kWh, Ad is the heated floor area of the dwelling in m2, Ud is the overall dwelling thermal efficiency relative to its floor area in W/m2 ◦C, and Tin i and Tout i are the inside and outside temperature of the dwelling in ◦C. Cd is the heat capacity of the dwelling in kWh/◦C, using the average specific heat capacity of UK dwellings at 250 kJ/m2 ◦C [18]. The previous hour’s indoor temperature Tin i−1 in ◦C is used to calculate the current hour’s temperature with the dwelling’s heat loss, solar gains Gs i and metabolic gains Gm both in kWh, and the dwelling’s heat capacity. The space heating demand Qsh i in kWh is then calculated relative to the Td i desired indoor temperature in ◦C (thermostat setpoint for that time). 3. Thermal energy storage materials and heat store designs – from commercial products to technology concepts Many current studies on TES focus on the potential of the materials, at their upper temperature limits showing what is possible for different TES materials and how the different categories of SHS, LHS and TCS typically differ, as shown in Fig. 1(a), however by considering the lim itations of the low-carbon heaters gives a more complete picture. This temperature constraint from heat pumps and solar thermal collectors restrains the performance of the TES, making it an important aspect to consider when analysing TES applied to domestic heating. A survey has been completed of the literature [6–9,15,32–43] to gather data on existing and under development SHS, LHS and TCS ma terials that could potentially be used in a domestic TES application, to show the full technology landscape of TES. Fig. 1(a) shows a range of materials down to a temperature of 30 ◦C, as less than this yields little value for domestic heating, comparing two important parameters: en ergy density and specific cost. However, as found in previous studies, many other factors need to be considered when selecting TES materials including: discharging rate; charging rate; discharging efficiency; charging efficiency; storage efficiency; corrosivity; acidity; toxicity; life duration, and technology readiness level [6–9,15,32–43]. By adapting the data in Fig. 1(a) to show how TES materials can perform when restricted to sensible upper values for ASHP and solar thermal of 70 ◦C Fig. 1(c) is created. SHS now has significantly lower values as the upper temperatures are limited, reducing the capacity and energy densities of these systems. LHS and TCS removes any material that melts/reacts above the 70 ◦C limit, significantly reducing the number of options, leaving much lower energy densities for LHS and with the remaining TCS now significantly higher energy densities than other thermal storage materials. With this limit also imposed, Li-ion batteries become much more favourable forms of energy storage, espe cially if considering being coupled with ASHP that operate at higher efficiencies. For SHS the upper material temperature limit is used to calculate the values, importantly these high potential temperatures allow relatively good densities and specific costs for SHS, but do not emphasise that with the higher temperatures comes more storage losses (with the same insulation). SHS materials are also grouped by their sub-categories and some of the best performing materials for energy density and specific costs of key sub-categories are labelled. 3. Thermal energy storage materials and heat store designs – from commercial products to technology concepts For SHS oils and salts, there are only small differences in densities within their groups, cost also remain similar, with the exception of vegetable oils which give comparable costs to the salt group but at lower densities. The metals and earth materials groups have varied performance across their groups, with some materials able to withstand very high temperatures and high densities. This results in some of the best energy densities, and, due to their low cost, they are the clear preferred TES materials for these parameters. i As prior studies tend to focus on TES materials, they can miss the performance changes that occur when considering the full thermal storage system. Fig. 2 shows how different TES systems can perform against each other using thermal store data from manufacturers, with (a) using the maximum TES system operating temperature and (b) limiting temperatures to 70 ◦C. The system level data analysis is taken from manufacturers, shown in Appendix A, when including their whole installation, not simply just the storage material. This higher level differs from the material level as the upper temperatures of the system may be limited lower than the maximum possible temperature achievable by the material. It also includes the increase in volume and cost requirements from the heat exchangers, insulation, and other ancillary parts, which lowers the energy densities and increases the specific costs compared to at the material level. i LHS data focuses on potential energy storage available from phase changing: the materials can achieve reasonable densities just relying on the latent heat from the phase change. Although LHS can allow further improvements in energy densities if the materials continue to be elevated up to their maximum workable temperatures, this also comes with the downside of more heat loss and lower storage efficiencies as with SHS. In addition, the larger temperature change alongside the phase change may cause further degradation of the LHS materials. For the LHS sub-categories, the paraffins, fatty acids, alcohols, and salt hy drates all have comparable energy densities, but with vast ranges in costs depending on the abundance of the material. Hydroxides in LHS improve on energy densities, but not as significantly as some metals and other salt materials. These are important factors as this is not the same fixed values for all types of TES i.e., higher storage temperatures require more advanced insulation materials. Table 4 Electricity tariffs used in the analysis. Tariff [28,29] Peak cost (p/ kWh) Off-peak cost (p/ kWh) Off-peak times (h) Export tariff (p/ kWh) Standing charge (p/ day) Flat rate (low tariff) 13.35 N/A N/A 5.5 20.06 Night off-peak (low tariff) 15.33 8.91 23–06 5.5 20.06 EV off-peak (low tariff) 13.45 5.0 0–4 3.0 25.0 Variable time of use (low tariff) Variable day ahead tariff. Average cost 9.3p/kWh, min −10.4p/kWh, max capped at 35p/kWh. Off-peak considered as anything <9.0p/ kWh. 5.5 21.0 Flat rate (high tariff) 32.42 N/A N/A 7.5 23.76 Night off-peak (high tariff) 35.93 21.63 23–06 7.5 23.85 EV off-peak (high tariff) 34.43 7.5 0–4 4.1 44.48 Variable time of use (high tariff) Variable day ahead tariff. Average cost 31.3p/kWh, min −8.6p/kWh, max capped at 35p/kWh. Off-peak considered as anything <31p/kWh. 7.5 21.0 In addition to higher operating temperatures of TES materials mak ing high storage efficiencies more challenging, they also restrict the type of heater used, which is pertinent for decarbonising heating. Fig. 1(b) shows how the potential low-carbon heaters perform depending on their output temperature, with thermal power on the left axis with solid lines and efficiency on the right axis with dashed lines. Solar thermal and ASHP show decreases in efficiency and therefore thermal power output with higher sink temperatures. Although DEH has a relative low effi ciency at lower sink temperatures compared to the heat pumps, it re mains nearly 100 % efficient at higher temperatures. Making DEH the only suitable method to couple with the higher temperatures required by some TES. Table 1 Key heating demand equations. For this study variable GB electricity grid emissions are used from the year 2020, which averaged 181gCO2e/kWh across the year [30]. Where reduced grid emissions scenarios were input into the framework this was completed by subtracting or adding a fixed value, in 25gCO2e/kWh increments, to each hourly value. Values were limited to a minimum value of zero emissions at each time step. The electrified technologies coupled with TES are compared against the baseline of natural gas boilers where natural gas emissions are 181gCO2e/kWh and costs use low 2020 tariffs of 2.1 p/kWh and 17.85p/day and high 2022 tariffs of 8.34p/kWh and 26.1p/day [28,31]. Table 1 Key heating demand equations. Source Formula Equation number [21] Qhl i = Ad × Ud × (Tin i −Tout i )/1000 (2) [17] Cd = (250 × Ad)/3600 (3) Tin i = Tin i−1 + (Qhl i + Gs i + Gm)/Cd (4) Qi sh = { ( Ti d −Ti in ) × Cd; If Ti d > Ti in 0; If Ti d ≤Ti in (5) [22] Qdhw i = (Vt × 4.18 × (Thw −Tcw, m)/3600) × Rdhw, m × Rdhw, h i (6) 3 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He Table 4 Electricity tariffs used in the analysis. As TCS is at the development stages, mass production costs are un clear, the potential range of production costs and energy densities of TCS are shown by the transparent green box using data gathered from the thorough literature review, and TCS technologies are positioned at the upper cost value of this due to not being commercially available. The energy density for TCS is shown using the energy from the material’s chemical reaction. The materials under consideration from the literature are predominantly using adsorption reactions instead of absorption re actions. This has very varied results for TCS but does have many options with very high energy densities. The additional benefit of LHS and TCS, not show, being lower temperatures and therefore higher storage effi ciencies than some of the high SHS. With temperature constraints, and disadvantages, removed it becomes clear that SHS generally is the most cost and space effective compromise. 3. Thermal energy storage materials and heat store designs – from commercial products to technology concepts As discussed, getting the maximum performance from many of the TES systems as shown in Fig. 2(a) is only possible with DEH. At the maximum TES system temperature scenario, as found in the material level, SHS remains the most advantageous technology, but with 4 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He Journal of Energy Storage 60 (2023) 106685 a reduced cost benefit compared to at the material level. At the system level the cost benefit of water for storage is limited as it is comparative to energy density over seven times better than any TES. Fig. 2(b) using the limited temperatures causes a few key changes to Fig. 1. (a) and (c) TES materials energy densities against specific material costs for SHS, LHS, and TCS materials. Energy storage using (a) the maximum material temperature and (c) using an upper temperature of 70 ◦C, both down to 30 ◦C. The shaded green box shows the range of potential commercial costs and energy densities of TSC materials due to many uncertainties at mass production, the TCS technologies are positioned on the upper cost value of this estimate as they are currently not in production. Lead and lithium-ion electrical battery materials shown for comparison. (b) Shows low-carbon heater thermal power output and effi ciencies against increasing output temperature. With heater power in solid lines on the left axis and efficiencies in dashed lines on the right axis of the same colours. Using fixed electrical input, 10 ◦C ambient conditions and 1000 W/m2 solar irradiance. ASHP COP calculated from [23]. Solar thermal collectors efficiencies calculated from [17,44]. Data in Appendix A. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) M. Ryland and W. He Fig. 1. (a) and (c) TES materials energy densities against specific material costs for SHS, LHS, and TCS materials. Energy storage using (a) the maximum material temperature and (c) using an upper temperature of 70 ◦C, both down to 30 ◦C. The shaded green box shows the range of potential commercial costs and energy densities of TSC materials due to many uncertainties at mass production, the TCS technologies are positioned on the upper cost value of this estimate as they are currently not in production. Lead and lithium-ion electrical battery materials shown for comparison. 3. Thermal energy storage materials and heat store designs – from commercial products to technology concepts (b) Shows low-carbon heater thermal power output and effi ciencies against increasing output temperature. With heater power in solid lines on the left axis and efficiencies in dashed lines on the right axis of the same colours. Using fixed electrical input, 10 ◦C ambient conditions and 1000 W/m2 solar irradiance. ASHP COP calculated from [23]. Solar thermal collectors efficiencies calculated from [17,44]. Data in Appendix A. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) energy density over seven times better than any TES. a reduced cost benefit compared to at the material level. At the system level the cost benefit of water for storage is limited as it is comparative to storage radiators and is only slightly lower specific cost than new high temperature SHS technologies which all come with higher energy den sities. LHS has acceptable energy densities put at higher costs, but not as high as batteries, as shown in Fig. 2. However, batteries are positioned using their stored electrical energy to calculate both energy density and specific costs, unlike the TES technologies which are used in thermal energy. The thermal equivalent of energy storage for batteries depends on which heater it is coupled with: if this is coupled with DEH this is near identical to the electrical values shown as DEH efficiency is close to 100 %. If electrical batteries are coupled with ASHP the battery performance would be 3.5 times better with a typical ASHP COP of 3.5, making it the second lowest specific cost in Fig. 2(b) after hot water tanks but with an Fig. 2(b) using the limited temperatures causes a few key changes to energy storage system landscape. SHS energy densities and costs are reduced, so much so that water tanks are the only feasible option as the high temperature SHS technologies are limited when coupled with ASHP or solar thermal. The reduction in performance of SHS makes LHS and batteries much more competitive, with LHS only slightly higher specific costs but with good energy density improvements over water tanks. These new insights show how important it is to consider the system level for domestic TES, where consumers have limited space, capital to invest, and TES may be coupled with different types of heaters. 4. Integration to domestic heating analysis framework To better understand how TES operates in domestic applications and which TES parameters are most effective to improving its performance, various TES and heating demand scenarios are simulated at an hourly resolution across the year in framework introduced in previous work and explained in the methodology [10,16]. With multiple tariffs avail able for consumers considered in the framework it is worth stating that, other than flat tariffs, the tariffs that have varying costs across the day generally have lower costs at the times when the associated emissions from grid generated electricity is also lower, as demonstrated in Fig. 3. This highlights that if TES is targeting reducing consumers OpEx by shifting demand to off-peak times of the day with electrified heating this has the added benefit of reducing emissions and improving utilisation of variable renewable energy generation. The positive effect of TES to reduce emission will be enhanced with the progress of the power system i With the electrified heaters different TES scenarios are presented, a minimal 0.1m3 TES, a large 0.5m3 TES, a 0.5m3 TES with high (×10) and low (÷10) specific heat capacities/energy densities to simulate how significant improvements to energy densities would alter TES viability, compared to the current technologies which can approach 2.5 times hot water tank energy densities shown in Fig. 2. Scenarios are also included for a 0.5m3 TES with high (×10) and low (÷10) TES CapEx, where high CapEx is comparable with more recent TES technologies and low CapEx could be an ideal scenario. These hypothetical changes to key parame ters can help identify what direction domestic TES should develop. Firstly, looking at Fig. 4(a) for the average UK dwelling heating de mand, the ASHP is more viable across the 20-years over DEH, although neither can compete with the gas boiler economically. With the ASHP the addition of larger TES and changing TES parameters has a small impact over an ASHP with a minimal TES, other than high CapEx TES which significantly increases the NPC of the system. This is primarily due to the lower OpEx from using an ASHP, meaning any improvements in reducing its OpEx using TES make a small absolute different and therefore do not payback the increased CapEx of TES over its lifetime. decarbonisation. ASHP and DEH are simulated alongside the commonly used domestic TES of hot water tanks and variations for representing other TES sce narios, the heater parameters are detailed in Appendix A. Then, the results are compared to the fossil fuel baseline of natural gas boilers in Fig. 4 in their systems OpEx and 20-year NPC. Tariffs considered are from 2020 (pre-energy crisis) and results shown use the EV style tariff, with 4 h of very low cost electricity in the night, as this tariff structure is shown to be the most beneficial for energy storage viability [16]. i 3. Thermal energy storage materials and heat store designs – from commercial products to technology concepts As the system level shows the significant increase in specific costs and decrease in energy densities compared to the material level. It highlights that if the commonly favoured low-carbon heater of ASHP is to be used, that 5 5 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He Fig. 2. Manufacturers data for TES system energy densities against specific system energy costs (a) using their maximum system operating temperatures (b) only heating up to 70 ◦C to remain suitable for heat pumps and solar thermal technologies. Electrical wall battery system shown for comparison. Data provided in Appendix A [13,15,33,42,45–50]. Fig. 2. Manufacturers data for TES system energy densities against specific system energy costs (a) using their maximum system operating temperatures (b) only heating up to 70 ◦C to remain suitable for heat pumps and solar thermal technologies. Electrical wall battery system shown for comparison. Data provided in Appendix A [13,15,33,42,45–50]. the simple hot water tank remains competitive although LHS can have a strong future in this market if costs can reduce, therefore integrating the right technologies for the specific application is important. 4. Integration to domestic heating analysis framework Increasing the energy density and therefore capacity of the TES allows a decrease in emissions, however this is less than the OpEx reduction as the variation in the EV tariff peak vs off-peak rates is greater than the variation in electrical grid emissions. A key factor to domestic energy technologies viability is the tariff rates, to understand how the increased cost of energy from the energy crisis has altered the position of the heating systems, Fig. 5 uses the same tariff styles but from 2022 for (a) 10th percentile and (b) 25th percentile dwellings again. As found with the higher demand dwellings, the higher OpEx, now from higher tariff rates, benefits the more efficient ASHP over DEH. However, as the demand remains low a high percentage of the electricity used can be from off-peak times for the high energy density TES with the fast-charging DEH, keeping the high energy density TES with DEH as the overall optimum NPC solution. difference due to the nominally higher OpEx of DEH compared to ASHP. Although this best-case scenario for DEH can result in similar peak electricity usage and NPC to ASHP, the lower efficiency of DEH still results in over double the equivalent annual emissions to the ASHP system. In comparing the emissions of the key combinations of heaters and TES parameters in Fig. 4 reveals similar trends to OpEx, as Fig. 3 high lights the link between tariff rates and electrical grid emissions. Emis sions fall in-line with demand, so reducing heating demands is one of the largest ways to reduce associated emissions. The difference between ASHP and DEH efficiencies (typically 3.5) is greater than the variation in electrical grid emissions and therefore concludes ASHP has lower asso ciated emissions than DEH in all TES scenarios, even if ASHP use elec tricity at times of higher emissions. Increasing the energy density and therefore capacity of the TES allows a decrease in emissions, however this is less than the OpEx reduction as the variation in the EV tariff peak vs off-peak rates is greater than the variation in electrical grid emissions. Fig. 4(b) introduces a higher demand dwelling which has been correlated to the upper 75th percentile UK dwelling heating demand. 4. Integration to domestic heating analysis framework In addition, the most cost-effective charging time for TES is using the night- time off-peak times, which is accompanied by cooler temperatures and therefore lower ASHP efficiencies, which typically give ASHP lower charging power than DEH. Showing TES struggles to improve the eco nomics of the relatively lower OpEx ASHP, and that lower CapEx TES is the most useful direction for TES coupled with and ASHP. Fig. 3. Variation in British electrical grid emissions compared to a variable time of use tariff, for the first four days of 2020 [29,30]. However, alongside DEH the potential for TES is much greater. Although low CapEx TES has similar overall advantages as with ASHP, the higher energy density has a much stronger impact. The high energy storage capacity of the high energy densities scenarios with the large 0.5m3 TES coupled with the faster charging DEH, can better take advantage of off-peak electricity rates, and make a larger absolute Fig. 3. Variation in British electrical grid emissions compared to a variable time of use tariff, for the first four days of 2020 [29,30]. 6 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He g. 4. Operational and 20-year costs for DEH and ASHP with various TES scenarios for (a) 50th percentile, average dwelling (b) upper 75th percentile dwelling, wer 10th percentile dwelling, and (d) lower 25th percentile dwelling. Text labels for key results show peak electricity used over a year and equivale nual emissions. Fig. 4. Operational and 20-year costs for DEH and ASHP with various TES scenarios for (a) 50th percentile, average dwelling (b) upper 75th percentile dwelling, (c) lower 10th percentile dwelling, and (d) lower 25th percentile dwelling. Text labels for key results show peak electricity used over a year and equivalent annual emissions. In comparing the emissions of the key combinations of heaters and TES parameters in Fig. 4 reveals similar trends to OpEx, as Fig. 3 high lights the link between tariff rates and electrical grid emissions. Emis sions fall in-line with demand, so reducing heating demands is one of the largest ways to reduce associated emissions. The difference between ASHP and DEH efficiencies (typically 3.5) is greater than the variation in electrical grid emissions and therefore concludes ASHP has lower asso ciated emissions than DEH in all TES scenarios, even if ASHP use elec tricity at times of higher emissions. 4. Integration to domestic heating analysis framework The scene remains similar for the high demand dwelling as with the average demand dwelling, but with further increase in DEH relative to ASHP due to the higher OpEx representing more of the overall NPC than the CapEx, where ASHP CapEx is higher than DEH. Lower heating demand dwellings are shown in Fig. 4(c) and (d) representing the lower 10th and 25th percentile respectively. In lower demand, with lower OpEx, DEH becomes more favourable, and with the high energy density TES, DEH is even the optimum cost overall low- carbon solution. However, although this is now the optimum solution the benefit of high energy density reduces compared to low CapEx TES alongside reduced demand. The ability of DEH with high-capacity TES in the low demand dwelling shows great flexibility potential for being able to shift demands, as only £2 of the £331 annual bill is from peak elec tricity usage (98 % reduction compared to £91 a year was from the fixed daily standing charge). A key factor to domestic energy technologies viability is the tariff rates, to understand how the increased cost of energy from the energy crisis has altered the position of the heating systems, Fig. 5 uses the same tariff styles but from 2022 for (a) 10th percentile and (b) 25th percentile dwellings again. As found with the higher demand dwellings, the higher OpEx, now from higher tariff rates, benefits the more efficient ASHP over DEH. However, as the demand remains low a high percentage of the electricity used can be from off-peak times for the high energy density TES with the fast-charging DEH, keeping the high energy density TES with DEH as the overall optimum NPC solution. 7 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He f gy g Fig. 5. Operational and 20-year costs for DEH and ASHP with various TES scenarios with 2022 high tariffs rates for (a) lower 10th percentile dwelling and (b) lower 25th percentile dwelling. y Fig. 5. Operational and 20-year costs for DEH and ASHP with various TES scenarios with 2022 high tariffs rates for (a) low 25th percentile dwelling. Fig. 5. Operational and 20-year costs for DEH and ASHP with various TES scenarios with 2022 high tariffs rates for (a) lower 10th percentile dwelling and (b) lower 25th percentile dwelling. 4. Integration to domestic heating analysis framework Although gas and electricity costs have both increased, the ratio between gas and electricity has reduced. On top of this, the off-peak electricity rate for the EV tariff only has a small absolute increase of 2.5p/kWh. Combining these points gives a significant economic improvement in electrified technologies relative to gas boilers, but gas remains the lowest NPC and very competitive on OpEx. ASHP having similar OpEx to gas and DEH with high density TES a very strong po tential using large quantities of the off-peak electricity. i value), which is of particular importance when operating at the higher temperatures to retain the stored energy, alters the system performance. Now shifting the focus to the environmental aspects of the system for an average demand dwelling, showing how the input variables can impact the equivalent annual emissions, even if both OpEx and emissions benefit in the same way of being able to shift demands from peak times to off-peak times efficiently. This plots the equivalent annual emissions from the system at different TES U values and different average grid emissions, all grid emissions used the varying hourly profile based on 2020 data as explained in the methodology section. The default value in the framework had average 2020 electrical grid emissions at 181gCO2e/ kWh and a TES U value of 1.3 W/m2 K. Fig. 6 uses a large, 0.5m3, hot water tank with (a) using DEH and an upper (hypothetical) TES limit of 500 ◦C to simulate values close to new commercial SHS technologies considered in Section 3 which are well suited to low-demand dwellings As discussed, the high energy density can bring benefits to the OpEx of electrified heating systems, so far in framework this has been considered by using a theoretical material with higher specific heat capacities. As introduced in Section 3, high energy densities can also be achieved by SHS operating at higher temperatures. Although this makes storage efficiency more challenging as heat is lost through from the TES. i gi y g g g Fig. 6 looks at how the thermal storage efficiency of the TES (the U Fig. 6. 5. Conclusion The data and simulations shown in this study demonstrate the ben efits of TES and importantly which areas of improvement can result in improved economic and environmental viability of TES in domestic applications, and therefore increase its adoption by consumers. TES can bring wider system benefits from shifting of demand, leading to reduc tion in peak electricity use which eases the burden along the electrical network and generation demands. This flexibility can also allow the increased utilisation of variable renewable energy generation, whether regionally or decentralised at the dwelling, allowing further reduction in emissions. For DEH although the high temperature allows high energy storage densities, at the higher U values there are more losses and so there is little benefit until around 0.8 W/m2 K, below this point the heat can better be retained to more efficiently use off-peak low emissions grid electricity. As the average grid emissions reduce although the percent age of heating emissions still reduces with the U value, the absolute difference decreases, however this is very dependent to how future grid emissions vary on an hour-by-hour basis. On the other hand, TES coupled with ASHP shows that the U value makes very little difference to heating emissions, due to the lower temperature limits of ASHP making TES capacity low and losses remain very low even at higher U values compared to overall heating demands. i An important comparison of TES at a system level is considered as well as at a material level, which emphasis why system approach needs to be considered as the addition of ancillary parts alongside the storage material significantly alters the specific costs and energy densities. The study also quantifies how TES requirements change from DEH to ASHP. Giving clear conclusions that ASHP and solar thermal low-carbon heating technologies with limited output temperatures can benefit well alongside LHS that have a suitable melting temperature around 50 ◦C to provide useful heat to the home while maintaining high heater efficiencies. On the other hand, DEH, which maintains its efficiency at higher temperatures and has high charging power couples best with high temperature SHS. To better visualise the greater system benefits of TES coupled with electrified heating systems Fig. 4. Integration to domestic heating analysis framework Specific heating emissions in gCO2e/kWh for the average demand dwelling with varying levels of average grid emissions in the y-axis against TES thermal efficiency improvements on the x-axis with a 0.5m3 thermal store for (a) DEH with an elevated maximum TES temperature of 500 ◦C and (b) an ASHP at a maximum TES temperature of 51 ◦C. Fig. 6. Specific heating emissions in gCO2e/kWh for the average demand dwelling with varying levels of average grid emissions in the y-axis against TES thermal efficiency improvements on the x-axis with a 0.5m3 thermal store for (a) DEH with an elevated maximum TES temperature of 500 ◦C and (b) an ASHP at a maximum TES temperature of 51 ◦C. Fig. 6. Specific heating emissions in gCO2e/kWh for the average demand dwelling with varying levels of average grid emissions in the y-axis against TES thermal efficiency improvements on the x-axis with a 0.5m3 thermal store for (a) DEH with an elevated maximum TES temperature of 500 ◦C and (b) an ASHP at a maximum TES temperature of 51 ◦C. Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He and (b) remains at the 51 ◦C temperature and is coupled with an ASHP which is economically a preferred solution for the average demand dwelling. TES studies which complete material level analysis. However, the study has its limitations as it only investigates in detail some of the important TES parameters. Future work needs to be done at a system level to include variations in charging and discharging efficiencies alongside other TES parameter changings which occur within the material sub- categories. Unsurprisingly, as grid emissions are reduced the equivalent heating emissions from electrified technologies reduce proportionately. Because the difference of energy efficiency for heating between DEH and ASHP, heating emissions vary significantly as shown in Fig. 6. For example, with the average grid emission of 150 gCO2e/kWh, the heating emission of DEH can be three times of the emission of TES coupled ASHP. How ever, if electrical grid is deeply decarbonised (average emission lower than 25 gCO2e/kWh), both DEH and ASHP (coupled with TES) can deliver low-carbon heating at a similar emission level (i.e., <30 gCO2e/ kWh). 5. Conclusion 7 shows the first two days of the simu lation for DEH in the lower 25th percentile home, which has previously been found to have a good potential for shifting demand and reducing peak electricity usage. Fig. 7(a) without any TES and (b) with a 0.5m3 hot water tank TES. Without the TES the heating supply must instan taneously follow the heating demand to balance out building heating losses, domestic hot water demands, and increases in desired thermostat temperatures. Any following of off-peak electricity use here is purely coincidental as the system has little control. In reality DEH is likely to struggle to meet instantaneous heating demands without some TES. When then including TES, although similar trends occur of following increasing thermostat set points, the TES can significantly reduce peak electricity use and recharge when prices reduce again while still meeting the household demands. This affect is stronger with the lower demand dwelling as even with current TES technologies the capacity is signifi cant enough to have an impact for low demand dwellings. Various scenarios simulated in the heating framework allow a clear clarification of what parameters are the most valuable for TES concepts to improve their economic viability, finding low CapEx is the most important factor for domestic TES, as also found in a review by Alva et al. [6]. The study specifies low CapEx TES is a much more dominant factor for ASHP compared to DEH. Our work further concludes that at the pre-energy crisis low tariffs, larger TES does not reduce its OpEx sufficiently enough over its lifetime to payback the additional CapEx alongside ASHP, but with the current high tariffs larger TES has now become cost effective. The low CapEx requirement for TES viability is why the simpler hot water tanks, with their competitive costs are the dominate technology, as other technologies are more expensive at a system level in £/kWh. This study focuses on the system-level integration of different TES technologies into domestic heating, in contradiction to most previous Fig. 7. Heater demands against variable tariff costs for the lower 25th percentile dwelling demand using direct electrical heating with (a) no TES, and (b) with a 0.5m3 TES. Fig. 7. Heater demands against variable tariff costs for the lower 25th percentile dwelling demand using direct electrical heating with (a) no TES, and (b) with a 0.5m3 TES. Data availability Data will be made available on request. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. M. Ryland and W. He Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization. Wei He: Conceptualization, Writing – review & editing, Supervision, Funding acquisition. For DEH, although low CapEx is found to be valuable for TES, the most valuable parameter for TES coupled with DEH is high energy densities, allowing greater use of off-peak electricity. This can be ach ieved at low CapEx by using low-cost materials that are capable of withstanding higher temperatures. It is found that for high temperature TES alongside DEH that the U value needs to be <0.8 W/m2 K to reap the advantageous of the higher temperatures. Space availability for TES in homes can be a restricting factor, hence why this study focuses on en ergy densities as high energy densities can allow sufficient energy storage capacities in smaller spaces. With smaller dwellings there is likely a smaller area available for TES and ASHP, alongside this the smaller the dwelling the lower the demand (with a fixed dwelling thermal efficiency), therefore in lower demand dwellings it was found high-capacity TES has more potential, these points together highlight a real benefit of high energy density TES with DEH in low demand dwellings. Acknowledgements The authors wish to thank the School of Engineering at the Univer sity of Warwick who funded this research. Wei He acknowledges the financial support from the Royal Academy of Engineering (RF\201819 \18\89). CRediT authorship contribution statement Michael Ryland: Conceptualization, Methodology, Software, 5. Conclusion ble tariff costs for the lower 25th percentile dwelling demand using direct electrical heating with (a) no TES, and (b) with a Fig. 7. Heater demands against variable tariff costs for the lower 25th percentile dwelling demand using direct electrical heating with (a) no TES, and (b) with a 0.5m3 TES. Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He Appendix A Appendix A Table 5 TES material specific cost and energy density, using maximum temperatures down to 30 ◦C. Material category Material Specific cost (£/kWh) Energy density (kWh/m3) SHS, H2O Water 1 75 Steam (5 bar) 3 1 SHS, oils Therminol VP-1 408 228 Syltherm XLT 504 116 Vegetable oil 11 171 SHS, salts HITEC 5 279 HITEC XL 6 304 Solar salt 5 301 SHS, metals Aluminium (500 ◦C) 12 317 Cast iron (1000 ◦C) 0.56 962 Cast steel (1000 ◦C) 0.867 1051 Sodium 5 295 Sodium-potassium eutectic 7 146 Lead-bismuth eutectic 173 564 SHS, earth materials Basalt 498 4604 Concrete 1 340 Silica fire bricks 2 339 Magnesia fire bricks 1 1121 LHS, paraffins n-Hexadecane 4611 51 n-Heptadecane 9455 46 n-Octadecane 4094 53 n-Nonadecane 13,611 48 Rubitherm RT-25 80 44 Rubitherm RT-50 110 36 Rubitherm RT-82 103 39 PRS paraffin wax 23 44 LHS, fatty acids Lauric acid 1152 41 Palmitic acid 341 341 Stearic acid 352 352 LHS, alcohols Xylitol 1248 118 D-sorbitol 4276 46 Meso-erythritol 3285 139 LHS, salt hydrates CaCl2 6H2O 289 83 NaCH3COO 3H2O 55 51 MgCl2 6H2O 86 73 Mg(NO3)2 6H2O 168 74 Ba(OH)2 8H2O 56 153 Na2S2O3 5H2O 66 97 Na2HPO4 12H2O 47 118 LHS, other salts NaNO3 92 108 KNO3 69 156 Na2CO3 22 194 K2CO3 78 150 CaCO3 148 116 Li2CO3 573 298 (continued on next page) Table 5 TES material specific cost and energy density, using maximum temperatures down to 30 ◦C. 10 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. References References [1] Visual Capitalist, Race to Net Zero: Carbon Neutral Goals by Country, 2021 accessed Sep. 28, https://www.visualcapitalist.com/race-to-net-zero-carbon-ne utral-goals-by-country/. [2] J. Lelieveld, J.S. Evans, M. Fnais, D. Giannadaki, A. Pozzer, The contribution of outdoor air pollution sources to premature mortality on a global scale, 5257569, Nature 525 (7569) (2015) 367–371, https://doi.org/10.1038/nature15371. Sep. 2015. [21] The Engineering ToolBox, Overall Heat Transfer Coefficient [Online]. Available, https://www.engineeringtoolbox.com/overall-heat-transfer-coefficient-d_434. html. (Accessed 3 September 2021). [3] National Statistics, Emissions of AirPollutants in the UK – Particulate Matter (PM10 and PM2.5), 5257569, 2022. Sep. 2015, https://www.gov.uk/government/statis tics/emissions-of-air-pollutants/emissions-of-air-pollutants-in-the-uk-particulate-m atter-pm10-and-pm25. [22] The Engineering ToolBox, Energy transfer equation. https://www.engineeringt oolbox.com/energy-transfer-equation-d_1051.html. (Accessed 3 September 2021). p p [4] J. Barnes, S.M. Bhagavathy, The economics of heat pumps and the (un)intended consequences of government policy, Energy Policy 138 (Mar. 2020), 111198, https://doi.org/10.1016/j.enpol.2019.111198. [23] I. Staffell, D. Brett, N. Brandon, A. Hawkes, A review of domestic heat pumps, The Royal Society of Chemistry, Energy and Environmental Science 5 (11) (2012) 9291–9306, https://doi.org/10.1039/c2ee22653g. Nov. 18. g j [5] S.J. Darby, Smart electric storage heating and potential for residential demand response, Energy Effic. 11 (1) (Jan. 2018) 67–77, https://doi.org/10.1007/s12053- 017-9550-3. [24] D. Steen, M. Stadler, G. Cardoso, M. Groissb¨ock, N. DeForest, C. Marnay, Modeling of thermal storage systems in MILP distributed energy resource models, Appl. Energy 137 (Jan. 2015) 782–792, https://doi.org/10.1016/J. APENERGY.2014.07.036. [6] G. Alva, Y. Lin, G. Fang, An overview of thermal energy storage systems, Elsevier Ltd, Energy 144 (2018) 341–378, https://doi.org/10.1016/j.energy.2017.12.037. Feb. 01. [25] Department for Business Energy &amp, Industrial Strategy, Thermal Energy Storage (TES) technologies report [Online]. Available, https://www.gov.uk/gove rnment/publications/evidence-gathering-thermal-energy-storage, 2016. (Accessed 16 August 2021). [7] I. Sarbu, C. Sebarchievici, A comprehensive review of thermal energy storage, Sustainabiblity 10 (1) (2018), https://doi.org/10.3390/su10010191. [8] L.F. Cabeza, E. Or´o, Thermal energy storage for renewable heating and cooling systems, in: Renewable Heating and Cooling: Technologies and Applications, Elsevier Inc., 2016, pp. 139–179. [26] A.V. Olympios, M. Mersch, P. Sapin, A.M. Pantaleo, C.N. Markides, Library of price and performance data of domestic and commercial technologies for low-carbon energy systems, 2021, https://doi.org/10.5281/ZENODO.4692649. Apr. [9] IRENA, Innovation Outlook: Thermal Energy Storage, /publications/2020/Nov/ Innovation-outlook-Thermal-energy-storage, Accessed: Jan. 27, 2021 [Online]. Available: /publications/2020/Nov/Innovation-outlook-Thermal-energy-storage. [27] J. Speirs, P. Balcombe, E. Johnson, J. Martin, N. Brandon, A. Hawkes, A greener gas grid: what are the options, Energy Policy 118 (Jul. 2018) 291–297, https://doi. org/10.1016/j.enpol.2018.03.069. p gy g [10] M. Ryland, W. Appendix A He Table 5 (continued) Material category Material Specific cost (£/kWh) Energy density (kWh/m3) ZnCl2 491 61 NaCl 18 252 KCl 44 194 MgCl2 232 291 LiCl 238 254 CaCl2 53 151 Na2SO4 54 123 Li2SO4 5080 52 K2SO4 102 157 MgSO4 295 90 CaSO4 350 131 LiF 380 766 NaF 19 564 KF 919 308 CaF2 46 345 LHS, hydroxides NaOH 65 96 KOH 88 85 LiOH 497 354 LHS, metals Copper 950 518 Zinc 233 224 TCS MgSO4 7H2O 917 MgSO4 6H2O 580 Mg(OH)2 778 FeCO3 722 Fe(OH)2 611 Ca(OH)2 806 CaCl2 2H2O 306 CaCl2 NH3 230 Al2(SO4) 6H2O 528 CaSO4 2H2O 389 MgCl2 6H2O 694 Na2S 5H2O 989 SrBr2 6H2O 639 Li2SO4 H2O 256 CuSO4 5H2O 575 SiO2 H2O 220 Zeolith H2O 230 NiCl2NH3 280 CH4 H2O 9 NH3 H2O 1 2H2 O2 600 Calcium looping 1200 Batteries Lead batteries 73 80 Li-ion batteries 146 500 For the analysis of material suitability and properties for low-carbon heaters, sensible heat storage materials are limited to 70 ◦C, whereas latent heat and thermochemical materials use the same data but only those which have suitable melt/reaction temperatures between 70 and 30 ◦C. Sources [6–9,15,32–43]. Table 5 (continued) Material category For the analysis of material suitability and properties for low-carbon heaters, sensible heat storage materials are limited to 70 ◦C, whereas latent heat and thermochemical materials use the same data but only those which have suitable melt/reaction temperatures between 70 and 30 ◦C. Sources [6–9,15,32–43]. Table 6 TES materials specific costs and energy density using maximum temperatures of 70 ◦C down to 30 ◦C. Table 6 TES materials specific costs and energy density using maximum temperatures of 70 ◦C down to 30 ◦C. Material category Material Specific cost (£/kWh) Energy density (kWh/m3) SHS, H2O Water 1.6 46 SHS, oils Therminol VP-1 3777 25 Syltherm XLT 2899 20 Vegetable oil 70 26 SHS, metals Aluminium 144 27 Cast iron 14 40 Cast steel 21 43 Sodium-potassium eutectic 148 8 SHS, earth materials Basalt 4604 44 Concrete 13 26 Silica fire bricks 33 20 Magnesia fire bricks 29 38 Sources [6–9,32,39,43]. able 6 ES materials specific costs and energy density using maximum temperatures of 70 ◦C down to 30 ◦C. 11 Journal of Energy Storage 60 (2023) 106685 M. Ryland and W. He Table 7 Table 7 Thermal energy storage system level data [13–15,33,42,45–48]. Appendix A TES system Dimensions (m) Cost (£) Energy capacity (kWh) Hot water tanks ø0.579 × 1.739 1063 18.6 Ceramic storage radiator 0.78 × 0.25 × 0.75 836 7.7 0.78 × 0.25 × 0.99 983 15.54 0.78 × 0.25 × 1.23 1162 23.1 Commercial salt LHS 0.429 × 0.365 × 0.575 1800 3.5 0.64 × 0.365 × 0.575 2075 7 0.87 × 0.365 × 0.575 2469 10.5 1.05 × 0.365 × 0.575 3230 15 New commercial SHS 0.98 × 0.6 × 0.66 5000 40 ø1.0 × 1.7 10,000 100 Table 8 ASHP and DEH parameters used in the simulations. Heater Dwelling percentile Thermal power (kW) COP 2–4 for ASHP Heater CapEx (£) DEH All 7 1100 ASHP 10th 2.0–4.1 5659 25th 2.5–5.0 5671 50th 3.9–7.9 5894 75th 5.1–10.1 6162 References He, Heating economics evaluated against emissions: an analysis of low-carbon heating systems with spatiotemporal and dwelling variations, Energy Build. 277 (2022), https://doi.org/10.1016/j.enbuild.2022.112561. [28] Octopus Energy, All our tariffs. https://octopus.energy/tariffs/. (Accessed 28 April 2022). [29] Energy-Stats, Octopus agile - energy stats UK. https://www.energy-stats.uk/octop us-agile/. (Accessed 1 June 2022). [11] Z. Ma, H. Bao, A.P. Roskilly, Feasibility study of seasonal solar thermal energy storage in domestic dwellings in the UK, Sol. Energy 162 (Mar. 2018) 489–499, https://doi.org/10.1016/j.solener.2018.01.013. [30] National Grid, Carbon intensity. https://carbonintensity.org.uk/. (Accessed 5 August 2022). [12] R. Renaldi, A. Kiprakis, D. Friedrich, An optimisation framework for thermal energy storage integration in a residential heat pump heating system, Appl. Energy 186 (Jan. 2017) 520–529, https://doi.org/10.1016/j.apenergy.2016.02.067. [31] Department for Business Energy &amp, Industrial Strategy, Greenhouse gas reporting: conversion factors 2021 [Online]. Available, 2021, https://www.gov. uk/government/publications/greenhouse-gas-reporting-conversion-factors-2021. p g j p gy [13] Tepeo, The ZEB, 2022 accessed Aug. 11, https://tepeo.com/thezeb. [13] Tepeo, The ZEB, 2022 accessed Aug. 11, h [32] H.A. Behabtu, et al., A review of energy storage technologies’ application potentials in renewable energy sources grid integration, 2020, Vol. 12, Page 10511, Sustain. 12 (24) (2020) 10511, https://doi.org/10.3390/SU122410511. Dec. [14] Caldera, Caldera Heat Batteries, 2022 accessed Oct. 11, https://www.caldera.co. uk/. [15] Sunamp, Sunamp Heat Batteries & Thermal Energy Storage, 2021 accessed Jan. 29, https://sunamp.com/. [33] Midsummer, Midsummer Wholesale. https://midsummerwholesale.co.uk/. (Accessed 11 August 2022). [16] M. Ryland, W. He, Holistic analysis of consumer energy decarbonisation options and tariff effects [Online]. Available, https://ssrn.com/abstract=4254958, 2022. [34] T. Bauer D. Laing R. Tamme n.d. “Overview of PCMs for Concentrated Solar Power in the Temperature Range 200 to 350 ◦C,” doi: 10.4028/www.scientific.net/ AST.74.272. [17] BRE, The Governments Standard Assessment Procedure for Energy Rating of Dwellings [Online]. Available, 2014, https://www.bre.co.uk/filelibrary/SAP/2012 /SAP-2012_9-92.pdf. (Accessed 28 January 2021). 12 M. Ryland and W. He Journal of Energy Storage 60 (2023) 106685 [35] S. Nagihara P. Ngo K. Zacny n.d. “Thermal Diffusivity-conductivity Measurements on Ice Samples of Magnesium Sulfate and Sodium Sulfate Solutions: Implications for Europa’s Ice Shell.”. [43] F. Nadeem, Comparative Review of Energy Storage Systems, Their Roles, and Impacts on Future Power System [Online]. Available, https://ieeexplore.ieee.org/ stamp/stamp.jsp?arnumber=8580457. (Accessed 8 February 2021). p [36] F.L. Guan, C.X. Gui, H. Bin Zhang, Z.G. Jiang, Y. Jiang, Z.Z. Yu, Enhanced thermal conductivity and satisfactory flame retardancy of epoxy/alumina composites by combination with graphene nanoplatelets and magnesium hydroxide, Compos. Part B Eng. 98 (Aug. 2016) 134–140, https://doi.org/10.1016/J. [44] W. References Streicher, Solar thermal technologies for domestic hot water preparation and space heating, in: Renewable Heating and Cooling: Technologies and Applications, Elsevier Inc., 2016, pp. 9–39. l [45] DirectHeatingSuppliers, Heatrae Megaflo Eco Dd250 250l Direct Unvented Cylinder. https://directheatingsupplies.co.uk/products/heatrae-megaflo-eco-dd 250-250-litre-direct-unvented-cylinder. (Accessed 11 August 2022). [37] Y. Sakamoto, H. Yamamoto, Y. Sakamoto, H. Yamamoto, Measurement of thermophysical property of energy storage system (CaCl2⋅NH3 system), Nat. Resour. 5 (12) (2014) 687–697, https://doi.org/10.4236/NR.2014.512060. Sep. 250-250-litre-direct-unvented-cylinder. (Accessed 11 August 202 [46] Dimplex, Electric Heating & Air Treatment For The Home. https://www.dimplex. co.uk/. (Accessed 11 August 2022). [38] J.N. Sweet, J.E. McCreight, Thermal conductivity of rocksalt and other geologic materials from the site of the proposed waste isolation pilot plant, Therm. Conduct 16 (1983) 61–78, https://doi.org/10.1007/978-1-4684-4265-6_7. [47] Energy Systems Catapult, Caldera heat batteries, Elsevier Ltd, https://es.catapult. org.uk/project/caldera/. (Accessed 11 August 2022). Feb. 01. [48] Osborne Clarke, Ofgem consultation on capacity market rules, Caldera launch domestic heat battery. https://www.osborneclarke.com/insights/energy-trans ition-ofgem-consultation-capacity-market-rules-new-energy-storage-site-develope d-yorkshire-caldera-launch-domestic-heat-battery-europes-hydrogen-strategy. (Accessed 11 August 2022). [39] The Engineering ToolBox, Threads. https://www.engineeringtoolbox.com/thre ads-t_73.html. (Accessed 11 August 2022). [39] The Engineering ToolBox, Threads. https://w ads-t_73.html. (Accessed 11 August 2022). [40] Jinhe, 江苏金合能源科技有限公司/全球热能存储专家. http://www.jinhe-energy. com/#/. (Accessed 8 February 2021). [40] Jinhe, 江苏金合能源科技有限公司/全球热能 [40] Jinhe, 江苏金合能源科技有限公司/全球 com/#/. (Accessed 8 February 2021) [40] Jinhe, 江苏金合能源科技有限公司/全球热能存储专家. http com/#/. (Accessed 8 February 2021). com/#/. (Accessed 8 February 2021). g [49] Naked Solar, Solar Batteries & Storage. https://nakedsolar.co.uk/storage/. (Accessed 10 November 2021). [41] PCM Products n.d. “Phase Change Material Products Limited Thin Ice Phase Change Material TM.”. [42] F. Ghani, R. Waser, T.S. O’donovan, P. Schuetz, M. Zaglio, J. Wortischek, Accepted manuscript non-linear system identification of a latent heat thermal energy storage system [Online]. Available, Appl. Therm. Eng. (2018), https://doi.org/10.1016/j. applthermaleng.2018.02.035. [50] GivEnergy, Giv AC 3.0. https://www.givenergy.co.uk/products#givbat82. (Accessed 11 August 2022). 13 13
https://openalex.org/W4297450288	https://zenodo.org/record/7120805/files/A%20CONCILIA%C3%87%C3%83O%20NA%20REFORMA%20DO%20C%C3%93DIGO%20DE%20PROCESSO%20CIVIL%20COMO%20CONDI%C3%87%C3%83O%20PR%C3%89-PROCESSUAL%20NA%20COMARCA%20DE%20BONFIM_MG%20%E2%80%93%20ISSN%201678-0817.pdf	Portuguese	null	A CONCILIAÇÃO NA REFORMA DO CÓDIGO DE PROCESSO CIVIL COMO CONDIÇÃO PRÉ-PROCESSUAL NA COMARCA DE BONFIM/MG	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	9,341	Orientador: Orientador: João Paulo Jucatelli3 João Paulo Jucatelli3 A CONCILIAÇÃO NA REFORMA DO CÓDIGO DE PROCESSO CIVIL COMO CONDIÇÃO PRÉ- PROCESSUAL NA COMARCA DE BONFIM/MG Ciências Jurídicas, Edição 114 SET/22 / 28/09/2022 REGISTRO DOI: 10.5281/zenodo.7120805 ABSTRACT This article sought to examine the alternative methods of conflict resolution brought by the new Civil Procedure Code (NCPC). To this end, an expository content will be presented addressing various nuances of the theme. First, it will analyze the main techniques for resolving existing disputes in our legal system. In a second moment, it will address the main aspects of Resolution n. 125 of the CNJ. In addition, a bibliographic study was carried out on doctrines, laws, secondary data and documentary sources, as it refers to a methodological procedure, whose purpose is to expose the understandings on the subject. In addition, it will examine the changes brought by the new procedural law with regard to Conciliation, in particular its definition and main characteristics. Finally, after the development of the theme, the practical consequences that the changes brought by the new Code will provide to the progress of the demands will be addressed, and whether these will be positive and will in fact meet the objectives pursued by the new procedural legislation. And to give an exact idea of what was exposed, a case study carried out in the Comarca de Bonfim/MG on the subject in question will be presented. Keywords: Alternative conflict resolution methods; conciliation; autocomposition; new Civil Procedure Code; Bonfim/MG. RESUMO O presente artigo buscou examinar os métodos alternativos de solução de conflitos trazidos pelo novo Código de Processo Civil (NCPC). Para tanto, será apresentado um conteúdo expositivo abordando diversas nuances do tema. Primeiramente, analisará as principais técnicas para resolução de lides existentes em nosso ordenamento jurídico. Em um segundo momento, abordará os principais aspectos da Resolução n. 125 do CNJ. Ademais empreendeu-se um estudo bibliográfico em doutrinas, leis, dados secundários e fontes documentais, por referir-se a um procedimento metodológico, cuja finalidade é a exposição dos entendimentos sobre o assunto. Além disso, examinará as mudanças trazidas pelo novo diploma processual no que diz respeito a Conciliação, em especial sua definição e principais características. Por fim, após o desenvolvimento do tema, O presente artigo buscou examinar os métodos alternativos de solução de conflitos trazidos pelo novo Código de Processo Civil (NCPC). Para tanto, será apresentado um conteúdo expositivo abordando diversas nuances do tema. será abordado as consequências práticas que as mudanças trazidas pelo novo Código irão proporcionar ao andamento das demandas, e se estas serão positivas e de fato atenderão os objetivos visados pela nova legislação processual. E para conferir uma noção exata do que fora exposto, será apresentado um estudo de caso realizado na Comarca de Bonfim/MG sobre o assunto em questão. Palavras-chave: Métodos alternativos de solução de conflitos; conciliação; autocomposição; novo Código de Processo Civil; Comarca de Bonfim/MG. 1. INTRODUÇÃO Com o passar dos anos, percebemos que o Direito brasileiro passou por modificações emblemáticas. Entre as transformações ocorridas no Código de Processo Civil, destacamos a ampliação dos métodos de solução de conflitos. Os métodos de solução de conflitos são a exteriorização de uma nova concepção para a defesa dos direitos. Ao invés de se ter uma só solução, a qual possibilita o ingresso de todos e a qualquer momento, sem discriminações objetivas, individuais ou teleológicas, o ordenamento jurídico apresenta vários métodos de resolução, o qual possuem o mesmo objetivo: a tutela dos direitos, de modo satisfatório, acertado e eficiente. A essência desses procedimentos excepcionais só será significativa enquanto for também, concomitantemente constitucionalmente devida. O Direito pertinente aos meios de solução de conflitos acolhe demandas jurídicas individuais, coletivas, disponíveis e indisponíveis, entre particulares e até mesmo entre o poder público. Nessa perspectiva, percebemos que o novo CPC trouxe em seu bojo o artigo 3°, §3°, o qual traz como preceito o incentivo à solução pacífica de pleitos. Tal dispositivo legal, além de ser um princípio atinente ao Código de Ritos, também é uma regra indispensável, aplicável tanto na parte geral do código, quanto a demandas e procedimentos não codificados. Desse modo, notamos que o Direito brasileiro vem sendo conduzido para a resolução de litígios de maneira extrajudicial, sejam os denominados métodos autocompositivos – conciliação, mediação ou negociação – ou heterocompositivos, como a arbitragem. Como vimos acima, as formas de solução de conflitos surgem no Código Instrumental por meio de seus institutos mais reputados, como a arbitragem, conciliação e mediação. A mediação é aplicada de modo mais contido, em áreas específicas, como no Direito de Família. Já a A arbitragem é regulada pela Lei n. 9.307/96, a qual possui as regras atinentes ao tema. Em relação a conciliação, verificamos que ela já estava sendo estimulada e utilizada pelo Judiciário e, por isso, abordaremos mais a fundo o presente instituto. A possibilidade em solucionar o conflito por outro modo que não seja através da via judicial, proporciona certos benefícios, sendo o principal deles, o contentamento das partes e o reestabelecimento do convívio entre os envolvidos. Outra consequência do uso de tais meios é a diminuição das ações judiciais litigiosas, bem como o cumprimento voluntário daquilo que foi convecionado. 1. INTRODUÇÃO Assim, o uso de tais métodos tem se mostrado como um avanço e melhoria para os jurisdicionados, os quais almejam resultados descomplicados, práticos, diretos, eficazes e menos custosos. Entre os meios alternativos de solução de conflitos, a conciliação tem demonstrado benefícios, visto que o novo CPC ampliou sua aplicação, pois foi ao encontro da Resolução nº 125 de 2010 do Conselho Nacional de Justiça – que dispõe sobre a Política Judiciária Nacional de tratamento adequado dos conflitos de interesses no âmbito do Poder Judiciário – ao determinar a criação, pelos tribunais, de centros de solução de conflitos e regulamentar o cadastro dos conciliadores. Com isso, diante de várias mudanças significativas, tem-se o instituto da conciliação como condição pré-processual para a resolução das lides, fazendo com que, antes de uma possível demanda judicial, as partes, podem juntas, resolverem suas querelas. O presente trabalho pretende demonstrar e examinar a conciliação, suas principais características, em especial a sua aplicabilidade e eficiência como condição pré-processual na comarca de Bonfim/MG, local este escolhido para ser realizado um breve estudo de caso. 2. OS PRINCIPAIS MÉTODOS ALTERNATIVOS DE RESOLUÇÃO DE CONFLITOS NO ORDENAMENTO JURÍDICO BRASILEIRO Ao longo dos anos, o Direito Processual Civil lidou com a concepção de que a viabilização tão somente do processo judicial – o qual é constituído por quatro princípios essenciais de ação, jurisdição, defesa e procedimento – seria hábil para conferir uma tutela satisfatória e devida ao direito pleiteado. A esse respeito, as ideias conceituais implícitas no revogado CPC de 1973 apresentavam um significativo padrão sobre: possuir o direito de ação consistia em desfrutar do direito ao processo e a decisão de mérito, da qual o magistrado, ao julgar desempenhava a função jurisdicional. Ou seja, todas as maneiras para se resolver a lide eram realizadas pela via judicial. (MARINONI, 2020). Com o passar do tempo, esse cenário começou a ser transformado. Preliminarmente, segundo os métodos incomuns de solução de conflitos, diferentes modelos heterocompositivos (tal como a arbitragem) e autocompositivas (como a conciliação e mediação) obtém lugar como formas capazes para a resolução das adversidades de cooperação na execução do direito material. Apesar disso, existia certa censura frequentemente voltada a esses meios de solução de embates. (MARINONI, 2020). É válido que essas acabaram se corroborando não exclusivamente como técnicas facultativas, mas como procedimentos sérios. Por esse motivo, a definição de “Justiça Multiportas” prevista no artigo 3° do Código de Ritos, cujo objeto está em explorar uma definitiva adequação dos modelos de resolução das lides às espécies de litígios que precisam ser assentados. (MARINONI, 2020). Vejamos o que dispõe o dispositivo citado. “Art. 3º Não se excluirá da apreciação jurisdicional ameaça ou lesão a direito. § 1º É permitida a arbitragem, na forma da lei. § 2º O Estado promoverá, sempre que possível, a solução consensual dos conflitos. § 3º A conciliação, a mediação e outros métodos de solução consensual de conflitos deverão ser estimulados por juízes, advogados, defensores públicos e membros do Ministério Público, inclusive no curso do processo judicial”. § 3º A conciliação, a mediação e outros métodos de solução consensual de conflitos deverão ser estimulados por juízes, advogados, defensores públicos e membros do Ministério Público, inclusive no curso do processo judicial”. Para melhor elucidação, tem-se os dizeres do Ilustre Marinoni: “Tudo isso significa, ademais, que o processo reage ao direito. Vale dizer: que o plano do direito processual reage ao plano do direito material. 2. OS PRINCIPAIS MÉTODOS ALTERNATIVOS DE RESOLUÇÃO DE CONFLITOS NO ORDENAMENTO JURÍDICO BRASILEIRO Na realidade, a nova codificação estabelece como uma de suas principais premissas o incentivo a utilização dos métodos adequados de solução consensual de conflitos, conforme se vê do artigo 3°, § 3°, inserido no capítulo inicial que trata das normas fundamentais do processo civil. Não obstante, o CPC/2015 menciona a conciliação, a mediação e a arbitragem em diversas passagens, deixando clara a intenção do legislador de incentivar a utilização de variados métodos de resolução de controvérsias. Além disso, o novo Código trata dos mediadores e conciliadores judiciais, atribuindo-lhes a qualidade de auxiliares da justiça (art. 149), estando sujeitos, inclusive, aos motivos de impedimento e suspeição (art. 148, II). (CNJ, 2015, p. 45)”. Antes de abordamos a fundo o instituto da conciliação, torna-se indispensável conhecermos, ainda que breve, os métodos de solução de conflitos de um modo geral. Entre as técnicas eletivas para se resolver as lides aponta-se aqui as mais utilizadas, como por exemplo: mediação, a negociação, a transação, arbitragem e a conciliação, sendo está o objeto do presente estudo. Vejamos: 2. OS PRINCIPAIS MÉTODOS ALTERNATIVOS DE RESOLUÇÃO DE CONFLITOS NO ORDENAMENTO JURÍDICO BRASILEIRO Além de procurar resolver as mais diferentes crises ou ameaças de crises de colaboração na realização do direito material e tratá-las de forma adequada, o processo civil procura fazê-lo com o emprego de diferentes técnicas processuais visando à promoção de uma tutela que consiga ao mesmo tempo ser adequada, efetiva e tempestiva para as partes e a menos dispendiosa possível – em termos de recursos econômicos e pessoais – para a administração da Justiça Civil. (MARINONI, 2020, p.35)”. Com isso, tornou-se inapropriado cogitar a jurisdição como o único método de solução do conflito por uma sentença. De fato, o litígio necessita ser analisado com o meio procedimental mais adequado às suas particularidades – o que pode, até mesmo, apontar a solução pela via da jurisdição como a última opção. (MARINONI, 2020). À vista disso, o Novo Código de Processo Civil categoricamente trouxe em seu bojo, além da jurisdição – presente desde o CPC de 1973 – como um dos prováveis meios para se resolver a controvérsia. Ao estabelecê-lo, o Código Instrumental cria a Justiça Civil proporcionando não só um único método para sanar o conflito – não apenas uma única “porta”. Longe disso, o novo CPC legitima um complexo de “Justiça Multiportas” o qual proporciona diversas ferramentas para resolver as lides – destacando-se a conciliação e a mediação. (MARINONI, 2020). Sobre o tema, tem-se o disposto no Guia de Conciliação e Mediação Judicial, editado pelo Conselho Nacional de Justiça: “O CPC/2015 fortalece, em boa hora, a conciliação, a mediação e a arbitragem como mecanismos hábeis a pacificação social. Na realidade, a nova codificação estabelece como uma de suas principais premissas o incentivo a utilização dos métodos adequados de solução consensual de conflitos, conforme se vê do artigo 3°, § 3°, inserido no capítulo inicial que trata das normas fundamentais do processo civil. Não obstante, o CPC/2015 menciona a conciliação, a mediação e a arbitragem em diversas passagens, deixando clara a intenção do legislador de incentivar a utilização de variados métodos de resolução de controvérsias. Além disso, o novo Código trata dos mediadores e conciliadores judiciais, atribuindo-lhes a qualidade de auxiliares da justiça (art. 149), estando sujeitos, inclusive, aos motivos de impedimento e suspeição (art. 148, II). (CNJ, 2015, p. 45)”. “O CPC/2015 fortalece, em boa hora, a conciliação, a mediação e a arbitragem como mecanismos hábeis a pacificação social. 2.3 Transação A transação na área do Direito Civil acontece quando há o compromisso, por meio de consentimentos mútuos em que os envolvidos evitam ou dão fim aos embates. (BUZZI, 2013). 2.2 Negociação A negociação é uma maneira de se resolver a querela entre os envolvidos e consiste no deleite simultâneo das partes na disputa. (BUZZI, 2013, p. 7. apud AZEVEDO, 2012, p. 91). 2.1 Mediação A Mediação é uma palavra latina mediare que quer dizer, apreciar, deliberar, colocar-se entre; e, desse modo, no campo dos métodos de resolução de conflitos, o vocábulo refere-se a uma técnica sem adversários, a qual deriva da coparticipação dos litigantes, para que juntos cheguem a um resultado satisfatório, visando o reestabelecimento das relações e a harmonia. (BUZZI, 2013, p. 5, apud SERPA, 1995). Segundo o Ilustre Professor, Fredie Didier Jr, sobre a temática, disserta que: “Cabe ao mediador servir como veículo de comunicação entre os interessados, um facilitador do diálogo entre eles, auxiliando-os a compreender as questões e os interesses em conflito, de modo que eles possam identificar, por si mesmos, soluções consensuais que gerem benefícios mútuos. Na técnica da mediação, o mediador não propõe soluções aos interessados. Ela é por isso mais indicada nos casos de conflitos societários e familiares. A mediação será exitosa quando os envolvidos conseguirem construir a solução negociada do conflito”. (DIDIER JR. 2016, p. 274). 2.4 Arbitragem A arbitragem é um método heterocompositivo de solução do conflito e se dá quando os demandantes procuram em um terceiro, os quais confiam, a resposta amistosa e justa – visto que, a tratativa não é realizada pelas partes envolvidas diretamente na lide, mas sim ao árbitro (DIDIER JR, 2016). Nos dizeres de Fredie Didier Jr: 2.5 Conciliação A Conciliação é o ajuste dos anseios decorrente de consentimentos recíprocos, em que um conciliador de modo imparcial auxilia, aconselha e oportuniza a transação. O conciliador, além de direcionar as partes, pode propor a saída para o conflito, desempenhando a sua atribuição de forma neutra, ocasião em que ele vai examinar o contexto litigioso e indicar a sua resolução, expondo os benefícios e malefícios que o referido instituto possa causar as partes. (DINAMARCO, 2005). Percebe-se que a conciliação contribui efetivamente para a solução da lide, tendo em vista que a sua finalidade é a realização do acerto entre as partes, as quais, ainda que “rivais”, obtêm um acordo e impedem a propositura de uma ação judicial. (BUZZI, 2013). Ressalta-se que, as práticas que implicam a busca da resolução de contendas por intermédio da conciliação podem ser realizadas tanto na fase em que o embate ainda não fora judicializado, em outras palavras, na etapa pré-processual ou extrajudicial, ou durante o processo. (BUZZI, 2013). Sobre o tema, tem-se os dizeres do professor Kazuo Watanabe: Nos dizeres de Fredie Didier Jr: “Trata-se de uma opção conferida a pessoas capazes para solucionar problemas relacionados a direitos disponíveis. Não se admite em causas penais. Ademais, a Emenda Constitucional n. 45/2004 consagra a arbitragem em nível constitucional, no âmbito trabalhista (art. 114, §§ 1° e 2°, CF/1988). […] a arbitragem no Brasil, é regulamentada pela Lei n. 9.0307/1996. Pode ser constituída por meio de um negócio jurídico denominado convenção de arbitragem que, na forma do art. 3° da Lei n. 9.307/1996, compreende tanto a cláusula compromissória quanto o compromisso arbitral”. (DIDIER JR, 2016, p. 171). jurisdicionados o acesso à ordem jurídica justa […].” (BUZZI, 2013, p. 10, apud WATANABE, 2013). O instituto da conciliação tem demonstrado o real objetivo da justiça, em que as partes resolvem suas demandas, restabelecem as relações e reduzem os inúmeros processos judiciais, contribuindo, assim, para o fim da crise do judiciário. Como visto acima, o tema em exame é de extrema relevância, pois demonstra um novo caminho, uma nova porta na busca do efetivo direito e da verdadeira justiça. Sobre o tema, tem-se os dizeres do professor Kazuo Watanabe: “[…] o objetivo primordial que se busca com a instituição de semelhante política pública, é a solução mais adequada dos conflitos de interesses, pela participação decisiva de ambas as partes na busca do resultado que satisfaça seus interesses, o que preservará o relacionamento delas, propiciando a justiça coexistencial. A redução do volume de serviços do Judiciário é uma consequência importante desse resultado social, mas não seu escopo fundamental. Por meio dessa política pública judiciária, que proporciona aos jurisdicionados uma solução mais adequada dos conflitos, o Judiciário Nacional estará adotando um importante filtro da litigiosidade, que ao contrário de barrar o acesso à justiça, assegurará aos 3. A RESOLUÇÃO N. 125 DO CONSELHO NACIONAL DE JUSTIÇA Da mesma forma que o Código instrumental, outras orientações regulamentares desenvolveram a temática dos métodos alternativos de conflitos, expondo consideráveis normas e aprimoraram o campo de aplicabilidade dos meios excepcionais de resolução de lides, como por exemplo: a conciliação, mediação e arbitragem. O presente capítulo apresentará uma breve síntese sobre as fundamentais informações constantes da Resolução n. 125 do Conselho Nacional de Justiça – CNJ – em especial, sobre a conciliação. No dia 29 de novembro de 2010 foi publicada a Resolução n. 125 do Conselho Nacional de Justiça – CNJ a qual preceitua sobre a Política Judiciária Nacional de Tratamento apropriado dos conflitos de interesses na esfera do Judiciário. A referida Resolução simbolizou um marco no ordenamento jurídico brasileiro, por ser a ferramenta mais importante em relação a conciliação. Sua finalidade é estruturar o conjunto de solução pacífica de lides, concedendo a técnica pertinente a cada demanda, para esse objetivo, tem-se o disposto no artigo 1°, parágrafo único que estabelece: “Art. 1º Fica instituída a Política Judiciária Nacional de tratamento dos conflitos de interesses, tendente a assegurar a todos o direito à solução dos conflitos por meios adequados à sua natureza e peculiaridade. Parágrafo único. Aos órgãos judiciários incumbe, nos termos do art. 334 do Novo Código de Processo Civil combinado com o art. 27 da Lei de Mediação, antes da solução adjudicada mediante sentença, oferecer outros mecanismos de soluções de controvérsias, em especial os chamados meios consensuais, como a mediação e a conciliação, bem assim prestar atendimento e orientação ao cidadão”. Em harmonia com o I. Fredie Didider Jr. (2015), a citada Resolução implementou a Política Pública de procedimento próprio de conflitos de interesses, além disso, determinou o CNJ como estruturador desta política. Ademais, determinou a constituição, pelos tribunais, de centros de resolução de embates, assim como regulou a atividade dos conciliadores, atribuindo aos tribunais o encargo de conferir publicidade aos bancos de estatísticas de seus centros. O artigo 2° da Resolução é possível visualizar alguns modos para se chegar a tal finalidade, ou sob um olhar doutrinário: “três metas fundamentais para a disseminação da cultura de pacificação social: 1) centralização das estruturas judiciárias; 2) adequada formação e treinamento de servidores, conciliadores e mediadores; 3) acompanhamento estatístico específico”. (IWAKURA, 2014, p.76). 3. A RESOLUÇÃO N. 125 DO CONSELHO NACIONAL DE JUSTIÇA Em consenso com a Resolução está o CPC, o qual inseriu essas regras e recomendou a aplicação das mesmas, demonstrando os métodos alternativos de solução das lides, para resolver cada embate do modo mais acertado, observando suas singularidades. (CUNHA; NETO, 2014). 4. A CONCILIAÇÃO Realizadas as breves considerações sobre os meios alternativos de resolução consensual de conflitos, passaremos a analisar o instituto da Conciliação, objeto principal do presente trabalho. 4.1 Conceito e principais características A conciliação, como vimos alhures, é uma técnica de solução de conflito, pela qual o terceiro denominado conciliador, atuará em certo procedimento, com o objetivo de amparar os envolvidos a alcançarem a autocomposição. (DIDIER JR, 2016). Na conciliação a querela é debatida de maneira perfunctória, na qual procura- se, basilarmente, autocomposição, com a consequente conclusão da lide. (CABRAL, 2017). O referido instituto é altamente propagado em nosso ordenamento jurídico, além de simbolizar expressivo encargo na resolução consensual do combate, mesmo que, em alguns casos, não consiga a diminuição das demandas judiciais com o consequente desafogamento do Poder Judiciário. (CABRAL, 2017). A conciliação obteve destaque com o estabelecimento dos Juízados Especiais Cíveis, sendo uma fase imprescindível à demanda. Apesar de ter sofrido certa objeção preliminar, os efeitos positivos gozavam de confiabilidade a esta medida, e na atualidade a maioria das lides são resolvidas na audiência de conciliação, isto é, sem passar por pronunciamento judicial do Magistrado. (CABRAL, 2017). Por seu turno, a conciliação possui previsão normativa no Código de Ritos, assim como em determinadas legislações específicas. (CABRAL, 2017). Vale ressaltar que o mencionado instituto pode ser enquadrado entre os métodos permissíveis de ingresso à justiça, consoante expresso no artigo 5°, inciso XXXV da Constituição Federal de 1988, posto que, soluciona o litígio de modo satisfatório e, consequentemente, mais legítimo. Refere-se, então a ferramenta apta a resolver os embates de modo consentâneo, a fim de diminuir as demandas judiciais, além de opor-se a descaracterização da atribuição jurisdicional, concedendo, destarte, uma interpretação moderna do acesso a justiça. (CABRAL, 2017). 4.2 A Conciliação no revogado Código de Processo Civil de 1973 A principal diferença diz respeito à amplitude do seu conteúdo, na medida em que ocorre o recebimento da resposta do réu, além de se realizar antes do encerramento da fase postulatória”. (NETO, Armando Ghedini. 2015, p. 37). Vale dizer que, a finalidade do legislador ao disciplinar o rito sumário foi de oportunizar um veredicto mais rápido a fundamentos específicos, por essa razão é o procedimento mais ágil que o ordinário. (THEODORO, 2012). Para que o procedimento citado cumpra o seu propósito de prosseguir com mais rapidez, é necessário que seja designada audiência de conciliação e, esta não acontecendo, já apresenta-se a resposta do réu, de acordo com os artigos 277 e 278, ambos do CPC/73). Ressalta-se que, aqui a atuação do conciliador é permitida, pois possui a atribuição de ajudar o Magistrado na missão de lograr êxito na demanda conciliatória. (THEODORO, 2012). Assim, a audiência de instrução não era instaurada sem que fosse tentada a conciliação, ocasião em que, caso ocorresse, o deliberado seria reduzido a termo e homologado por sentença. No entanto, se a conciliação fosse negativa, a AIJ prosseguiria com a proposição da resposta do réu, segundo o artigo 278 do CPC/73. (THEODORO, 2012). Nos dizeres do professor Humberto Theodoro Jr (2012, p. 365): “Somente ocorrerá a segunda audiência, destinada à instrução e julgamento, se, após a frustração da tentativa de conciliação, houver necessidade de colher prova oral para dirimir a lide”. Após a breve síntese sobre a conciliação prevista no CPC/73, torna-se crucial aprofundarmos nas consideráveis alterações trazidas pelo Novo CPC ao tema em exame. 4.2 A Conciliação no revogado Código de Processo Civil de 1973 A conciliação encontrava-se expressamente prevista na Lei n. 5.869 de 11 de janeiro de 1973 – Revogado CPC/73 – mais precisamente a partir do artigo 277 e seguintes. Entretanto, o Código Instrumental não tinha tanta preocupação com o citado instituto, sendo mera formalidade o uso de tal técnica, a qual só era utilizada se houvessem indícios de possível acordo entre as partes. Na sistematização do CPC passado haviam três classes processuais: i) conhecimento, ii) execução e, iii) cautelar. Relativamente ao processo de conhecimento, existiam os procedimentos comuns e especiais. O procedimento comum era utilizado de modo residual, tão somente para aquelas ações que não tinham rito próprio e, consoante o artigo 272 do CPC/73: “O procedimento comum é ordinário ou sumário”. O rito ordinário era ajustado de modo integral e pormenorizado pelo CPC obsoleto, ao passo que os outros eram prescritos tão somente naquilo que se divergiam deste, visto que, o procedimento ordinário aplicava-se de forma subsidiária a eles. (THEODORO, 2012). O procedimento ordinário, conforme o artigo 331 do CPC/73, realizava-se da seguinte forma,: o autor ofertava a exordial, após o julgador determinava a citação do réu, para que oferecesse sua defesa e, posteriormente, realizava-se uma audiência preambular onde eram feitas as tentativas de conciliação. Havendo possibilidade de consenso entre as partes acerca da lide, o ajuste seria homologado por sentença pelo juiz. Todavia, caso restasse frustrada a tratativa, delimitava-se o objeto contencioso e admitia-se as provas que seriam produzidas e, caso fosse relevante, era designada audiência de instrução e julgamento, a qual ao início desta, o magistrado faria a seguinte pergunta: Há possibilidade de acordo? Em sendo a resposta negativa, o processo prosseguiria.(THEODORO, 2012). O procedimento ordinário, conforme o artigo 331 do CPC/73, realizava-se da seguinte forma,: o autor ofertava a exordial, após o julgador determinava a citação do réu, para que oferecesse sua defesa e, posteriormente, realizava-se uma audiência preambular onde eram feitas as tentativas de conciliação. O procedimento sumário, por sua vez, era delimitado conforme parâmetros relacionados à matéria e ao valor da causa, os quais eram previstos no artigo 275 do código anterior. (THEODORO, 2012). Nesse sentido: Nesse sentido: “A audiência de conciliação prevista no procedimento sumário, embora tenha algumas diferenças significativas, possui conteúdo semelhante à audiência preliminar prevista no artigo 331 do Código de Processo Civil de 1973. 4.3 A Conciliação no novo CPC A Lei n. 13.105 de 16 de março de 2015, instituiu o Novo Código de Processo Civil, o qual estabeleceu como um dos seus basilares intentos o fomento ao emprego dos meios pertinentes de resolução consensual de conflitos, em conformidade com o artigo 3°, §3°, introduzido na parte introdutória que versa sobre as regras elementares do processo civil. (CABRAL, 2017). dos meios pertinentes de resolução consensual de conflitos, em conformidade com o artigo 3°, §3°, introduzido na parte introdutória que versa sobre as regras elementares do processo civil. (CABRAL, 2017). Evidencia-se que o novo CPC refere-se a conciliação e aos outros métodos de solução de conflitos em vários pontos, deixando nítido o intento do legislador em estimular e impulsionar a utilização dos meios existentes. Outrossim, o novo Código de ritos descreve os conciliares como colaboradores da justiça, segundo o artigo 149 do CPC, os quais estão submetidos, ainda, as causas de impedimento e suspeição, conforme artigo 148, II do CPC. Além disso, o Código dedicou a Seção V, do Capítulo III, para normatizar os deveres dos conciliadores, e entre outros temas dispôs sobre: “a) a criação de centros judiciários de solução consensual de conflitos pelos tribunais, destinados à realização de audiências e pelo desenvolvimento de programas para auxiliar, orientar e estimular a autocomposição (art. 165); b) os princípios que informam a conciliação e a mediação (art. 166); c) o cadastro e a capacitação de conciliadores e mediadores (art. 167); d) a possibilidade de as partes escolherem, de comum acordo, o conciliador ou mediador (art. 168); e) as formas de remuneração dos conciliadores e mediadores (art. 169); f) os casos de impedimento (art. 170); g) a impossibilidade temporária do exercício da função (art. 171); g) o prazo de impedimento de um ano para o conciliador e mediador assessorar, representar ou patrocinar as partes (art. 172); h) as hipóteses de exclusão do cadastro (art. 173); i) a criação de câmaras de mediação e conciliação para a solução de controvérsias no âmbito da administração pública (art. 174); j) a possibilidade de outras formas de conciliação e mediação extrajudiciais (art. 175)”. (CABRAL, 2017. pgs. 361/362). À vista disso, deve o Judiciário desfrutar oportunamente de tais instrumentos em benefícios aos envolvidos. 4.3 A Conciliação no novo CPC Vale dizer que o NCPC colocou em nosso Poder Judiciário avultada esperança de transformação de condutas dos conflitantes, para que estes reflitam e reconsiderem os prováveis resultados em relação a lide, passando a utilizar ferramentas mais apropriadas para a solução do embate, por intermédio de um alicerce hábil a tal objetivo. (CABRAL, 2017). reconsiderem os prováveis resultados em relação a lide, passando a utilizar ferramentas mais apropriadas para a solução do embate, por intermédio de um alicerce hábil a tal objetivo. (CABRAL, 2017). Vale destacar a inovação mais surpreendente na estruturação do Judiciário: a instituição, como regra, da audiência de conciliação e/ou mediação, como prática preliminar do procedimento comum, isto é, antes do oferecimento da contestação pelo demandado. (CABRAL, 2017). De acordo com o Código Instrumental, o reclamado será citado para ir a audiência de conciliação e/ou mediação – segundo o disposto no artigo 334 do NCPC – sendo que, apenas com o término do feito e em sendo frustrada a convenção, será aberto o prazo para, querendo, apresentar contestação – consoante artigo 335, I do NCPC. Salienta-se que a citada audiência não será realizada, conforme o exposto no artigo 334, §4° do CPC quando: “i) se ambas as partes manifestarem, expressamente, desinteresse na composição consensual ou II – quando não se admitir a autocomposição”. Percebe-se que, ao Magistrado é proibido rejeitar a realização da audiência de conciliação, ainda que o consenso seja impossível. Além disso, a norma inadmite a renúncia de apenas um dos litigantes, de modo que, o não comparecimento infundado ao procedimento será reputado como ato atentatório à dignidade da justiça, passível de multa de até dois por cento do valor da causa ou do benefício financeiro esperado, a qual será convertido em proveito da União ou do Estado. (CABRAL, 2017). Após os breves apontamentos sobre a conciliação prevista no CPC/2015, torna- se essencial analisarmos o preparo material técnico dos operadores do direito para implementação da conciliação os quais serão tratadas no capítulo a seguir. 5. O PREPARO MATERIAL TÉCNICO DOS OPERADORES DO DIREITO PARA IMPLEMENTAÇÃO DA CONCILIAÇÃO O preparo material técnico dos operadores do direito para implementação da conciliação é estimulado e definido pelo legislador, o qual determinou as regras de atuação, instituição e capacitação. Senão Vejamos: Como vimos, a atual sistemática processualista é idealizada para incentivar a utilização e o estabelecimento dos métodos alternativos de resolução de conflitos. 4.3 A Conciliação no novo CPC Na listagem das regras primordiais do NCPC estão as regras do artigo 3°, §2° e §3°, in verbis: “Art. 3º Não se excluirá da apreciação jurisdicional ameaça ou lesão a direito. […] § 2º O Estado promoverá, sempre que possível, a solução consensual dos conflitos. § 3º A conciliação, a mediação e outros métodos de solução consensual de conflitos deverão ser estimulados por juízes, advogados, defensores públicos e membros do Ministério Público, inclusive no curso do processo judicial.” Nesse sentido, ressalta-se os ensinos dos professores Leonardo Carneiro da Cunha e João Luiz Lessa de Azevedo Neto (2014, p. 199): “O Estado deverá promover o uso dos ADR39 e os profissionais da área jurídica deverão estimular o seu uso. Isso inclui um esforço de capacitação de pessoal, criação de estrutura física, esclarecimento da população e treinamento dos servidores e dos profissionais do meio jurídico em geral. Não apenas estimula o uso dos ADR em âmbito judicial, mas o projeto também estabelece que a União, os Estados, e o Distrito Federal e os Municípios deverão criar câmaras de mediação e conciliação, com atribuições relacionadas à solução consensual de conflitos no âmbito administrativo. Assim, há a construção de um verdadeiro sistema de resolução de disputas, composto pelo Poder Judiciário e por instituições públicas e privadas dedicadas ao desenvolvimento de mediação, conciliação e arbitragem”. O CPC dispõe no caput do artigo 166 que: “A conciliação e a mediação são informadas pelos princípios da independência, da imparcialidade, da autonomia da vontade, da confidencialidade, da oralidade, da informalidade e da decisão informada”. A independência deve guiar a ação do conciliador, para que ele detenha certa autonomia e não se sinta coagido, no exercício de sua atribuição, assim como, lhe é autorizado rejeitar, cancelar ou suspender a sessão se verificar a ausência dos requisitos primordiais para o prosseguimento. (DIDIER JR, 2015). Tocante a imparcialidade, o conciliador não pode possuir qualquer interesse na querela em exame. Esse atributo retrata o princípio da imparcialidade, o qual conduz a Administração Pública e está previsto no artigo 37, caput da CF/88. (DIDIER JR, 2015). Vale enfatizar que: “a aplicação de técnicas negociais pelo conciliador ou mediador, com o objetivo de proporcionar ambiente favorável à autocomposição, não ofende o dever de imparcialidade” (DA CUNHA; NETO, 2014, p. 201). 4.3 A Conciliação no novo CPC Tocante a autonomia da vontade, verificar-se que ela está expressamente inserida no §4° do artigo 166 do CPC, que estabelece: “A mediação e a conciliação serão regidas conforme a livre autonomia dos interessados, inclusive no que diz respeito à definição das regras procedimentais”. Ressalta-se que, referida norma encontra-se em conformidade com todo ordenamento, citando caso análogo, tem-se os artigos 168 e 190, ambos do CPC, os quais permitem aos litigantes a viabilidade de definirem alterações no procedimento para adaptá-lo as peculiaridades da causa, bem como de indicarem, desde que em comum acordo, o conciliador ou a câmara privada para a realização da conciliação. A confidencialidade, prevista nos parágrafos 1° e 2° do artigo 166 do CPC, aduz sobre a incumbência do conciliador de manter as informações em segredo. Tal atributo requer certa relevância, visto que, os litigantes necessitam estar à vontade para que possam revelar seus dissabores em busca da autocomposição. (CUNHA; NETO. 2014). Concernente a oralidade e informalidade no campo da conciliação: Concernente a oralidade e informalidade no campo da conciliação: “A oralidade e a informalidade orientam a mediação e a conciliação. Ambas dão a este processo mais “leveza”, sem o ritual e a simbologia próprios da atuação jurisdicional. Mediador e conciliador devem comunicar-se em linguagem simples e acessível e não devem usar nenhum tipo de roupa solene (veste talar, toga, etc.). É conveniente que a negociação realize-se em ambiente tranquilo, se possível sem barulho, em mesa redonda e com paredes pintadas com cor clara. Todos são aspectos cênicos importantes, pois permitem um diálogo mais franco, reforçando a oralidade e a informalidade”. (DIDIER, 2015, p. 278). Ainda sobre o tema: Ainda sobre o tema: “a conciliação ou a mediação não precisa sequer ocorrer no ambiente judiciário, podendo, se as partes preferirem ou caso se sentirem mais à vontade, ser realizado no escritório de um dos advogados ou em outro ambiente”. (CUNHA; NETO. 2014. p. 278). Atinente ao principio da decisão informada, o acordo entre os envolvidos precisa ser alcançado, logo após o entendimento da contrariedade, como também, dos efeitos do consenso. (DIDIER, 2015). Nessa perspectiva: “é necessário, enfim, que os interessados sejam bem informados para que não sejam surpreendidos por qualquer consequência inesperada da solução pela qual venham a optar”. (CUNHA; NETO. 2014. p. 202). Cabe apontar que, além das regras acima, existem deliberações na seção V, acerca dos Centros judiciários de resolução consensual de lides. O artigo 165, caput e §1° do CPC define que: “Art. 165. Os tribunais criarão centros judiciários de solução consensual de conflitos, responsáveis pela realização de sessões e audiências de conciliação e mediação e pelo desenvolvimento de programas destinados a auxiliar, orientar e estimular a autocomposição. § 1º A composição e a organização dos centros serão definidas pelo respectivo tribunal, observadas as normas do Conselho Nacional de Justiça”. “Art. 165. Os tribunais criarão centros judiciários de solução consensual de conflitos, responsáveis pela realização de sessões e audiências de conciliação e mediação e pelo desenvolvimento de programas destinados a auxiliar, orientar e estimular a autocomposição. § 1º A composição e a organização dos centros serão definidas pelo respectivo tribunal, observadas as normas do Conselho Nacional de Justiça”. Assim: “tal disposição vai ao encontro da política pública de tratamento adequado dos conflitos de interesses no âmbito do Poder Judiciário desenvolvida pelo CNJ por meio da Resolução 125” (MEIRELLES; MARQUES, 2014, p. 294). A criação dos citados Centros Judiciários é inescusável, os quais serão administrados por um magistrado, e se for preciso, terão um auxiliar, sendo que, estes possuirão a responsabilidade de coordenar e supervisionar os conciliadores. (DIDIER, 2015). Além disso, os conciliadores passarão por cursos de capacitação e treinamento conferidos pelos próprios Tribunais. 6. CONCILIAÇÃO: APLICABILIDADE E EFICIÊNCIA A conciliação possui presentemente um aparato legislativo qualificado para certificar ao citado instituto a segurança jurídica basilar para propagar a cultura da pacificação social e para aprimorar a sua funcionalidade, tanto na seara extrajudicial quanto na judicial. E essa alteração de padrão já pode ser verificada entre os profissionais do Direito, em especial na comarca de Bonfim/MG, em que, após uma pesquisa realizada, observou-se certas mudanças, as quais serão apresentadas no próximo capítulo. 6.1 Estudo de Caso: A Conciliação na Reforma do CPC como condição pré- processual na Comarca de Bonfim/MG Nessa direção: acertado ao seu litigio, visto que, apesar da recomendação da técnica cabia ao julgador ou auxiliar da justiça, a decisão caberá as partes. Nessa direção: Nessa direção: “A capacitação é indispensável, pois o sucesso do mesmo depende da correta explicação em relação aos métodos de solução de conflitos disponíveis (judicial e extrajudiciais: conciliação e mediação), o que possibilitará a escolha do mais adequado pelas partes. Para tanto, a pessoa responsável pela triagem dos casos deve conhecer profundamente todos os métodos de solução de conflitos disponíveis e seus respectivos procedimentos, pois apenas assim poderá passar as informações necessárias para o devido esclarecimento das partes, que devem fazer uma opção consciente. “A capacitação é indispensável, pois o sucesso do mesmo depende da correta explicação em relação aos métodos de solução de conflitos disponíveis (judicial e extrajudiciais: conciliação e mediação), o que possibilitará a escolha do mais adequado pelas partes. Para tanto, a pessoa responsável pela triagem dos casos deve conhecer profundamente todos os métodos de solução de conflitos disponíveis e seus respectivos procedimentos, pois apenas assim poderá passar as informações necessárias para o devido esclarecimento das partes, que devem fazer uma opção consciente. […] na fase inicial, deve o juiz, serventuário da justiça ou técnico, devidamente treinado, e conhecedor dos diversos métodos de solução de conflitos existentes, fornecer as informações necessárias sobre esses métodos (apresentando as vantagens e desvantagens dos mesmos) e indicar a parte o mais adequado para o caso concreto, verificando as características, não só do conflito, mas das partes nele envolvidas e dos próprios procedimentos disponíveis, esclarecendo como funcionará o procedimento escolhido”. (CNJ. Guia da Conciliação e Mediação. 2015. P. 17). Assim, diante de profissionais capacitados, os envolvidos disporão de na fase inicial, deve o juiz, serventuário da justiça ou técnico, devidamente treinado, e conhecedor dos diversos métodos de solução de conflitos existentes, fornecer as informações necessárias sobre esses métodos (apresentando as vantagens e desvantagens dos mesmos) e indicar a parte o mais adequado para o caso concreto, verificando as características, não só do conflito, mas das partes nele envolvidas e dos próprios procedimentos disponíveis, esclarecendo como funcionará o procedimento escolhido”. (CNJ. Guia da Conciliação e Mediação. 2015. P. 17). Assim, diante de profissionais capacitados, os envolvidos disporão de informações aptas para realizarem a escolha consciente do método mais acertado ao seu litigio, visto que, apesar da recomendação da técnica cabia ao julgador ou auxiliar da justiça, a decisão caberá as partes. 6.1 Estudo de Caso: A Conciliação na Reforma do CPC como condição pré- processual na Comarca de Bonfim/MG A princípio foi realizado um levantamento de dados, a partir do mês de novembro de 2021, ocasião em que havia sido instalado o CEJUSC – Centro Judiciário de Solução de Conflitos e Cidadania – e os resultados obtidos foram: ◉ Demandas judicias – processos eletrônicos: foram realizados 21 acordos contra 234 acordos frustrados. ◉ Demandas judicias – processos eletrônicos: foram realizados 21 acordos contra 234 acordos frustrados. ◉ Demandas extrajudiciais: foram realizados 21 acordos e contra 13 acordos frustrados. Analisando os dados acima, constatou-se que, as demandas extrajudiciais via CEJUSC, novidade em nosso ordenamento, demonstrou resultados satisfatórios. Analisando os dados acima, constatou-se que, as demandas extrajudiciais via CEJUSC, novidade em nosso ordenamento, demonstrou resultados satisfatórios. Além de tais dados, foram realizadas entrevistas aos operadores do Direito da Comarca de Bonfim/MG, as quais possuíam as seguintes indagações: 1) O(a) Senhor(a) percebe alguma diferença na realização da audiência de conciliação antes e depois da reforma do CPC? Se sim, qual? 2) Na prática, é perceptível a mudança acerca da audiência de conciliação no judiciário? 3) Na sua concepção, falta preparo dos profissionais para condução e realização da audiência de conciliação? 4) O que pode ser feito para que as partes desconstruam essa ideia de que somente um terceiro poderá resolver o conflito? conciliação? 4) O que pode ser feito para que as partes desconstruam essa ideia de que somente um terceiro poderá resolver o conflito? Em relação ao primeiro questionamento tem-se os dizeres do Dr. Túlio Augusto Geraldo Parreiras – Assesssor de Juiz: 1) O Senhor percebe alguma diferença na realização da audiência de conciliação antes e depois da reforma do CPC? Se sim, qual? “No CPC/73 não tinha tanta preocupação com acordo como hoje, antes podia marcar audiência de conciliação e tentar conciliar, mas não tinha obrigação de marcar a audiência, era mais embrionário e não tinha tanto essa visão. O TJ/MG até antes da reforma do CPC, criou uma resolução para fazer audiências de conciliação nos processos de família, sendo que o juiz fazia o despacho inicial e já marcava audiência, era inconstitucional, pois era totalmente o contrário do texto do CPC, porém, foi aplicado e tinha resultados bons, já que era na área de família e se trata da área que mais tem acordo. No CPC antigo, não tinha muita preocupação com acordo, era maias a questão de litigar para ver quem “chegava no final, quem ganhava e perdia.” Após isso, começou o movimento de conscientização pelo CNJ, em que precisava fazer acordo. Depois disso, veio o CPC/15, possuindo uma visão bem melhor do acordo, possuindo uma parte geral que o antigo não tinha, abordando sobre a obrigatoriedade de promover a conciliação. Sendo assim, a partir da reforma, a audiência só não acontece se as partes não quiserem, então, a chance de ter a conciliação é muito maior, e isso tem tido, aparentemente, resultados positivos. Nas ações com empresas, não tem surtido muito efeito, mas nos casos de pessoas físicas, principalmente em processos de família, o índice de acordo tem sido bom. A crítica em relação ao CPC/2015 no que tange à conciliação, é que ele perdeu a chance de talvez fazer o procedimento ficar mais rápido, porque a parte tem 15 dias a partir da audiência para contestar, e isso poderia ter sido concentrado em uma única audiência, fazendo tudo naquele ato e prosseguir para a decisão do juiz. Além disso, ainda tem a preocupação do CNJ com a justiça multiportas, e agora está caminhando para uma tentativa de maior conciliação.” “No CPC/73 não tinha tanta preocupação com acordo como hoje, antes podia marcar audiência de conciliação e tentar conciliar, mas não tinha obrigação de marcar a audiência, era mais embrionário e não tinha tanto essa visão. 1) O Senhor percebe alguma diferença na realização da audiência de conciliação antes e depois da reforma do CPC? Se sim, qual? O TJ/MG até antes da reforma do CPC, criou uma resolução para fazer audiências de conciliação nos processos de família, sendo que o juiz fazia o despacho inicial e já marcava audiência, era inconstitucional, pois era totalmente o contrário do texto do CPC, porém, foi aplicado e tinha resultados bons, já que era na área de família e se trata da área que mais tem acordo. No CPC antigo, não tinha muita preocupação com acordo, era maias a questão de litigar para ver quem “chegava no final, quem ganhava e perdia.” Após isso, começou o movimento de conscientização pelo CNJ, em que precisava fazer acordo. Depois disso, veio o CPC/15, possuindo uma visão bem melhor do acordo, possuindo uma parte geral que o antigo não tinha, abordando sobre a obrigatoriedade de promover a conciliação. Sendo assim, a partir da reforma, a audiência só não acontece se as partes não quiserem, então, a chance de ter a conciliação é muito maior, e isso tem tido, aparentemente, resultados positivos. Nas ações com empresas, não tem surtido muito efeito, mas nos casos de pessoas físicas, principalmente em processos de família, o índice de acordo tem sido bom A crítica em relação ao CPC/2015 no que tange à conciliação, é que ele perdeu a chance de talvez fazer o procedimento ficar mais rápido, porque a parte tem 15 dias a partir da audiência para contestar, e isso poderia ter sido concentrado em uma única audiência, fazendo tudo naquele ato e prosseguir para a decisão do juiz. Além disso, ainda tem a preocupação do CNJ com a justiça multiportas, e agora está caminhando para uma tentativa de maior conciliação.” Já em relação aos questionamentos dois e três tem-se os dizeres do Dr. Robert Lopes de Almeida – Juiz Direitor da Comarca e Coordenador do CEJUSC: 2) Na prática, é perceptível a mudança acerca da audiência de conciliação no judiciário? 2) Na prática, é perceptível a mudança acerca da audiência de conciliação no judiciário? O Código de processo civil é muito recente, então para uma lei processual é muito nova ainda, demandaria assim, uma mudança de mentalidade. Existe um grupo de advogados, de juízes e juristas, que foram educados na escola do contencioso, vez que as pessoas tendem a litigar, então seria realmente uma mudança de cultura. Dessa forma, é perceptível que tivemos mudança, uma vez que os advogados estão mais abertos a conciliação, os juízes estão mais cientes da necessidade dela, inclusive o ministério público tem aderido. No ambiente forense, os operadores do direito tiveram mudança de mentalidade acerca da audiência de conciliação. Quanto a sua forma, a audiência não teve muita alteração, mas sim na mentalidade quanto a conciliação dentro do judiciário. O Código de processo civil é muito recente, então para uma lei processual é muito nova ainda, demandaria assim, uma mudança de mentalidade. Existe um grupo de advogados, de juízes e juristas, que foram educados na escola do contencioso, vez que as pessoas tendem a litigar, então seria realmente uma mudança de cultura. Dessa forma, é perceptível que tivemos mudança, uma vez que os advogados estão mais abertos a conciliação, os juízes estão mais cientes da necessidade dela, inclusive o ministério público tem aderido. No ambiente forense, os operadores do direito tiveram mudança de mentalidade acerca da audiência de conciliação. Quanto a sua forma, a audiência não teve muita alteração, mas sim na mentalidade quanto a conciliação dentro do judiciário. 3) Na sua concepção, falta preparo dos profissionais para condução e realização da audiência de conciliação? Penso que nós temos profissionais preparados para condução da audiência de conciliação, inclusive eles são capacitados pelo Tribunal de Justiça, onde é fornecido curso para as pessoas que vão conduzir essas audiências, informando, ensinando e colocando-os a par da situação. Por vezes, essas pessoas apresentam dificuldades para se fazer entender quando não têm muito costume com essa audiência. Diante disso, penso que estão sim capacitados, porém, como em toda área do Direito, existem àqueles que estão mais preparados que outros, mas isso é, muitas vezes, atinente a questão pessoal, porém, de modo geral, eles são bem preparados. Concernente ao quarto questionamento tem-se os dizeres do Dr. Neider Chaves Ribeiro – Defensor Público: 4) “O que pode ser feito para que as partes desconstruam essa ideia de que somente um terceiro poderá resolver o conflito? 2) Na prática, é perceptível a mudança acerca da audiência de conciliação no judiciário? “Visando esclarecer e conscientizar os interessados de que a construção da solução do conflito deve partir dos próprios envolvidos, campanhas educativas, difundidas através dos meios de comunicação, seriam de fundamental importância. Operadores do direito devem, sempre que possível, conscientizar as partes acerca da necessidade de elas próprias elaborarem a solução para o conflito, através do consenso.” Sobre o questionamento acima, vale destacar os dizeres do Dr. Spencer dos Santos Ferreira Júnior – Promotor de Justiça: 4) “O que pode ser feito para que as partes desconstruam essa ideia de que somente um terceiro poderá resolver o conflito? 4) “O que pode ser feito para que as partes desconstruam essa ideia de que somente um terceiro poderá resolver o conflito? “Por meio da educação, consegue-se alcançar uma relação social mais consensual, com menos conflitos de direitos, pois é por meio da educação que o indivíduo passa a ter conhecimento dos seus direitos e deveres e a partir daí diminuir a quantidade de violação de direitos, por inobservância de deveres. Então, é uma questão que passa, necessariamente, pela educação, de forma geral e com enfoque na educação que se propõe a trazer o mínimo essencial na temática de direitos na nossa sociedade, sejam eles relacionados a consumo ou civis obrigacionais, por exemplo. Dessa forma, diminuiria a carga e a concepção de que só um terceiro ajudaria, pois conseguiria alcançar um consenso sem se socorrer ao poder judiciário.” Ao analisar as informações percebemos que na comarca de Bonfim/MG a conciliação tem obtido resultados satisfatórios, em especial nas demandas pré- processuais, em que as partes, magistrado, servidores e auxiliares da justiça tem se empenhado para que haja a resolução da lide e o reestabelecimento da harmonia entre os envolvidos. Vale dizer, o Judiciário deveria ser o último meio a ser buscado para a resolução do conflito. Entretanto, é necessário uma mudança cultural, deixando claro que, os métodos alternativos de lides são os instrumentos mais adequados para que as partes possam utilizar para resolverem suas controvérsias. os métodos alternativos de lides são os instrumentos mais adequados para que as partes possam utilizar para resolverem suas controvérsias. 7. CONCLUSÃO O presente trabalho buscou demonstrar que os métodos alternativos de solução de conflitos são de substancial relevância para o Direito Processual Civil, para os envolvidos, para o Judiciário e para a sociedade como um todo, visto que, apontam incontáveis vantagens se forem bem empregados. O intento do legislador ao elaborar o novo Código de Ritos foi de avultar os meios consensuais de resolução de lides. Além de incentivar a resposta autocompositiva entre os litigantes, objetivando a execução de uma audiência com este propósito antes mesmo da exibição da resposta do demandado, realmente, pode eliminar vários procedimentos judiciais, desafogando o Judiciário e, assim sendo, ofertar uma solução mais célere igualmente aos processos em que serão obrigatórios um pronunciamento do juiz ao final. Percebe-se que as benfeitorias vão adiante, uma vez que, visando implantar a cultura da pacificação dos embates por intermédio das técnicas autocompositivas, as alterações postas pelo novo CPC cooperam para que as partes detenham conhecimento de que uma solução concebida por eles mesmos é mais proveitosa do que aquela determinada por um terceiro alheio ao problema. Dentre essas técnicas, a conciliação foi objeto do presente estudo, na qual percebemos consideráveis mudanças na utilização do referido instituto. Vimos que no CPC revogado não havia tanta preocupação com a conciliação das partes. Entretanto, notou-se certa movimentação e conscientação pelo CNJ de que o uso dos meios autocompositivos eram necessários. Assim, surge o CPC/2015, o qual trouxe um incremento no uso de tais técnicas, cuja finalidade é que as próprias partes contribuam de modo efetivo para a construção de soluções mais justas e céleres. A propósito, a solução consensual de conflitos deve ser estimulada por todos operadores do Direito. A propósito, a solução consensual de conflitos deve ser estimulada por todos operadores do Direito. Ao realizarmos o estudo de caso na comarca de Bonfim/MG, vislumbramos um exemplo positivo de que o uso dos métodos, sobretudo a conciliação, realmente funciona, pois todos tem se empenhado para que haja a resolução da lide e o reestabelecimento da harmonia entre os envolvidos. REFERÊNCIAS BIBLIOGRÁFICAS BRASIL. Conselho Nacional de Justiça 2015. Guia de Conciliação e Mediação Judicial: orientação para instalação de CEJUSC. (Brasília/DF: Conselho Nacional de Justiça). BRASIL. Lei n. 5.869, de 11 de janeiro de 1973. Institui o Código de Processo Civil. Diário Oficial da União, Brasília, DF, 11 de janeiro de 1973. BRASIL. Lei n. 13.105, de 16 de março de 2015. Institui o Código de Processo Civil. Diário Oficial da União, Brasília, DF, 17 de março de 2015. BUZZI, Marco Aurélio Gastaldi. A Mudança de Cultura Pela Composição de Litígios. Lisboa/Portugal. 2013. BUZZI, Marco Aurélio Gastaldi. A Mudança de Cultura Pela Composição de Litígios. Lisboa/Portugal. 2013. BUZZI, Marco Aurélio Gastaldi. A Mudança de Cultura Pela Composição de Litígios. Lisboa/Portugal. 2013 apud AZEVEDO, André Gomma [org.]. Manual de Mediação Judicial. 3.ª ed. Brasília: Ministério da Justiça, 2012. MARINONI, Luiz Guilherme. Curso de processo civil: teoria do processo civil, volume 1 [livro eletrônico] / Luiz Guilherme Marinoni, Sérgio Cruz Arenhart, Daniel Mitidiero São Paulo: Thomson Reuters Brasil – 5. ed. rev., atual. e ampl: 2020. MARINONI, Luiz Guilherme. Curso de processo civil: teoria do processo civil, volume 1 [livro eletrônico] / Luiz Guilherme Marinoni, Sérgio Cruz Arenhart, Daniel Mitidiero São Paulo: Thomson Reuters Brasil – 5. ed. rev., atual. e ampl: 2020. DIDIER JR, Fredie. Curso de Direito Processual Civil: Introdução ao Direito Processual Civil, parte geral e processo de conhecimento/ Fredie Didier Jr. – 17. ed. – Salvador: Ed. JusPodivm, 2015. v. I. DIDIER JR, Fredie. Curso de Direito Processual Civil: Introdução ao Direito Processual Civil, parte geral e processo de conhecimento/ Fredie Didier Jr. – 17. ed. – Salvador: Ed. JusPodivm, 2015. v. I. DIDIER JR, Fredie. Curso de Direito Processual Civil: Introdução ao Direito Processual Civil, parte geral e processo de conhecimento/ Fredie Didier Jr. – 17. ed. – Salvador: Ed. JusPodivm, 2015. v. I. DIDIER JR, Fredie. Curso de Direito Processual Civil: Introdução ao Direito Processual Civil, parte geral e processo de conhecimento/ Fredie Didier Jr. – 17. ed. – Salvador: Ed. JusPodivm, 2015. v. I. MATIAS, José Pereira. Manual de Metodologia da Pesquisa Científica. Disponível em: Minha Biblioteca, (4th edição). Grupo GEN, 2016. MEIRELLES, Delton Ricardo Soares; MARQUES, Giselle Picorelli Yacoub. A mediação no projeto do novo Código de Processo Civil: solução para o judiciário? In: Novas tendências do processo civil: estudos sobre o projeto do novo Código de Processo Civil. v. 2, Salvador: JusPodivum, 2014, p. 285-303. NETO, Armando Ghedini. A audiência de conciliação no novo Código de Processo Civil. Revista Eletrônica de Direito Processual – REDP. Rio de Janeiro: Universidade do Estado do Rio de Janeiro, vol. 16, n. 16, p. 29-57, 2015. THEODORO JÚNIOR, Humberto. Curso de Direito Processual Civil – Teoria geral do direito processual civil e processo de conhecimento – vol. 1 – Humberto Theodoro Júnior – Rio de Janeiro: Forense, 2012. DINAMARCO, Cândido. Instituições de Direito Processual Civil – vol. I. São Paulo: Malheiros Editores, Paulo, 2005. IWAKURA, Cristiane. Mediadores: disposições gerais. Revista Eletrônica de Direito Processual – REDP. Rio de Janeiro: Universidade do Estado do Rio de Janeiro, vol. Especial, Ano 8, p. 62-78, 2014. IWAKURA, Cristiane. Mediadores: disposições gerais. Revista Eletrônica de Direito Processual – REDP. Rio de Janeiro: Universidade do Estado do Rio de Janeiro, vol. Especial, Ano 8, p. 62-78, 2014. DIER JR, Fredie. Curso de Direito Processual Civil: Introdução ao Dire DINAMARCO, Cândido. Instituições de Direito Processual Civil – vol. I. São Paulo: Malheiros Editores, Paulo, 2005. CABRAL, Trícia Navarro Xavier. A evolução da conciliação e mediação no Brasil. Revista FONAMEC – Rio de Janeiro, v.1, n. 1, p. 354 -369, maio 2017. CABRAL, Trícia Navarro Xavier. A evolução da conciliação e mediação no Brasil. Revista FONAMEC – Rio de Janeiro, v.1, n. 1, p. 354 -369, maio 2017. CUNHA, Leonardo Carneiro da; NETO, João Luiz Lessa de Azevedo. A mediação e a conciliação no projeto do novo CPC: Meios integrados de resolução. In: DIDIER JUNIOR, Fredie et al (org.). Novas Tendências do Processo Civil: Estudos Sobre o Projeto do Novo Código de Processo Civil. Salvador: JusPodivm, 2014. THEODORO JÚNIOR, Humberto. Curso de Direito Processual Civil – Teoria geral do direito processual civil e processo de conhecimento – vol. 1 – Humberto Theodoro Júnior – Rio de Janeiro: Forense, 2012. Acadêmica de Direito no Centro Universitário Una – Betim/MG. E-mail: grasyparreiras@yahoo.com.br 1 Acadêmica de Direito no Centro Universitário Una – Betim/MG. E-mail: larissaandreza14@gmail.com 2 Graduado em Direito. Pós-graduado em Direito Notarial e Registral. Mestre em Direitos Coletivos e Cidadania. Professor no Centro Universitário Una das seguintes disciplinas: Direitos Reais. Contratos e Direito Internacional. E-mail: jucatelli7@hotmail.com 3 ← Post anterior ← Post anterior RevistaFT A RevistaFT é uma Revista Científica Eletrônica Multidisciplinar Indexada de Alto Impacto e Qualis “B”. Periodicidade mensal e de acesso livre. Leia gratuitamente todos os artigos e publique o seu também clicando aqui. Contato Queremos te ouvir. WhatsApp: 11 98597-3405 e-Mail: contato@revistaft.com.br ISSN: 1678-0817 CNPJ: 45.773.558/0001-48 Copyright © Editora Oston Ltda. 1996 - 2022 Rua José Linhares, 134 - Leblon \| Rio de Janeiro-RJ \| Brasil Copyright © Editora Oston Ltda. 1996 - 2022 Rua José Linhares, 134 - Leblon \| Rio de Janeiro-RJ \| Brasil
https://openalex.org/W4361266496	https://aacr.figshare.com/articles/journal_contribution/Supplementary_Figure_2_from_Molecular_Staging_of_Cervical_Lymph_Nodes_in_Squamous_Cell_Carcinoma_of_the_Head_and_Neck/22364321/1/files/39808583.pdf	unk	null	Supplementary Figure 2 from Molecular Staging of Cervical Lymph Nodes in Squamous Cell Carcinoma of the Head and Neck	null	2,023	cc-by	3	D D D
https://openalex.org/W4362637534	https://nottingham-repository.worktribe.com/file/19296533/1/Experimental%20Physiology%20-%202023%20-%20Jones%20-%20Acute%20adaptation%20of%20central%20and%20peripheral%20motor%20unit%20features%20to%20exercise%E2%80%90induced	English	null	Acute adaptation of central and peripheral motor unit features to exercise‐induced fatigue differs with concentric and eccentric loading	Experimental physiology	2,023	cc-by	9,584	R E S E A R C H A R T I C L E R E S E A R C H A R T I C L E Abstract Abstract Force output of muscle is partly mediated by the adjustment of motor unit (MU) firing rate (FR). Disparities in MU features in response to fatigue may be influenced by contraction type, as concentric (CON) and eccentric (ECC) contractions demand variable amounts of neural input, which alters the response to fatigue. This study aimed to determine the effects of fatigue following CON and ECC loading on MU features of the vastus lateralis (VL). High-density surface (HD-sEMG) and intramuscular (iEMG) electromyography were used to record MU potentials (MUPs) from bilateral VLs of 12 young volunteers (six females) during sustained isometric contractions at 25% and 40% of the maximum voluntary contraction (MVC), before and after completing CON and ECC weighted stepping exercise. Multi-level mixed effects linear regression models were performed with significance assumed as P < 0.05. MVC decreased in both CON and ECC legs post-exercise (P < 0.0001), as did force steadiness at both 25% and 40% MVC (P < 0.004). MU FR increased in ECC at both contraction levels (P < 0.001) but did not change in CON. FR variability increased in both legs at 25% and 40% MVC following fatigue (P < 0.01). From iEMG measures at 25% MVC, MUP shape did not change (P > 0.1) but neuromuscular junction transmission instability increased in both legs (P < 0.04), and markers of fibre membrane excitability increased following CON only (P = 0.018). These data demonstrate that central and peripheral MU features are altered following exercise-induced fatigue and differ according to exercise modality. This is important when considering interventional strategies targeting MU function. 2Centre of Precision Rehabilitation for Spinal Pain (CPR Spine), School of Sport, Exercise and Rehabilitation Sciences, College of Life and Environmental Sciences, University of Birmingham, Birmingham, UK 3Department of Clinical and Experimental Sciences, Università degli Studi di Brescia, Brescia, Italy 4Department of Systems Design Engineering, University of Waterloo, Waterloo, Ontario, Canada wileyonlinelibrary.com/journal/eph 1 Experimental Physiology. 2023;1–11. Correspondence Mathew Piasecki, Centre Of Metabolism, Ageing & Physiology (COMAP) Academic Unit of Injury, Inflammation & Recovery Sciences, School of Medicine, Faculty of Medicine & Health Sciences, University of Nottingham, Royal Derby Hospital Centre (Room 3011), Derby DE22 3DT, UK. Eleanor J. Jones1 Yuxiao Guo1 Eduardo Martinez-Valdes2 Francesco Negro3 Daniel W. Stashuk4 Philip J. Atherton1 Bethan E. Phillips1 Mathew Piasecki1 Eleanor J. Jones1 Yuxiao Guo1 Eduardo Martinez-Valdes2 Francesco Negro3 Daniel W. Stashuk4 Philip J. Atherton1 Bethan E. Phillips1 Mathew Piasecki1 1Centre of Metabolism, Ageing and Physiology (COMAP), MRC-Versus Arthritis Centre for Musculoskeletal Ageing Research, National Institute for Health Research (NIHR) Nottingham Biomedical Research Centre, University of Nottingham, Nottingham, UK Received: 1 December 2022 Accepted: 14 March 2023 Received: 1 December 2022 Accepted: 14 March 2023 DOI: 10.1113/EP091058 Received: 1 December 2022 Accepted: 14 March 2023 DOI: 10.1113/EP091058 R E S E A R C H A R T I C L E New Findings The motor unit (MU) is a key component of the motor system and control of muscle force is regulated by the rate coding and recruitment of MUs (Enoka & Duchateau, 2017). The concept of fatigue is vast and encompasses several definitions; however, that related to neuro- muscular decrement is most accurately described as performance fatigue: the loss of force and/or power output from a muscle as a result of impaired contractile function and/or muscle activation (Enoka & Duchateau, 2016). Performance fatigue is a feature often seen acutely post-exercise and is commonly the cause of task failure (Hunter et al., 2004). It is also observed in chronic clinical conditions (Burtin et al., 2012; Chaudhuri & Behan, 2004; Prinsen et al., 2015) and ageing (Christie et al., 2011; Merletti et al., 2002). Irrespective of the setting, the involvement of multiple physiological mechanisms and experimental triggers presents challenges in determining the cause (Enoka & Duchateau, 2008). ∙What is the central question of this study? Conflicting evidence exists on motor unit (MU) firing rate in response to exercise-induced fatigue, possibly due to the contraction modality used: Do MU properties adapt similarly following concentric and eccentric loading? ∙What is the central question of this study? Conflicting evidence exists on motor unit (MU) firing rate in response to exercise-induced fatigue, possibly due to the contraction modality used: Do MU properties adapt similarly following concentric and eccentric loading? ∙What is the main finding and its importance? MU firing rate increased following eccentric loading only despite a decline in absolute force. Force steadiness deteriorated following both loading methods. Central and peripheral MU features are altered in a contraction type-dependant manner, which is an important consideration for training interventions. ∙What is the main finding and its importance? MU firing rate increased following eccentric loading only despite a decline in absolute force. Force steadiness deteriorated following both loading methods. Central and peripheral MU features are altered in a contraction type-dependant manner, which is an important consideration for training interventions. The muscle response to fatigue has been shown to be affected by the muscle contraction modality. For example, fatigue caused by both concentric (CON) and eccentric (ECC) contractions elicits reductions in muscle strength (Linnamo et al., 2000) and muscle activation (Peñailillo et al., 2013) yet the recovery profiles relating to damage are different (Souron et al., 2018). KEY WO RDS KEY WO RDS fatigue, high density electromyography, intramuscular electromyography, motor unit 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. © 2023 The Authors. Experimental Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. Experimental Physiology. 2023;1–11. wileyonlinelibrary.com/journal/eph 1 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. © 2023 The Authors. Experimental Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 023 The Authors. Experimental Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. New Findings Of the three commonly applied contraction types (isometric, CON and ECC), ECC ‘lengthening’ contractions generate greater voluntary forces and appear less strenuous due to a lower metabolic cost and cardiovascular stress (Hody et al., 2019; Over- end et al., 2000; Webber & Kriellaars, 1997). These features make eccentric training programmes favourable for improving muscle mass and strength in both healthy adults and those with compromised musculoskeletal health (Cook et al., 2013; Roig et al., 2009). the limitations of signal attenuation through skin and subcutaneous tissue. iEMG recordings enable estimation of neuromuscular junction (NMJ) transmission instability and temporal dispersion across MU propagating action potentials (Piasecki et al., 2021) as well as MUP gradients reflecting ion exchange and membrane excitability. Thus, the combination of these techniques allows an overall assessment of central and peripheral MU adaptations to a given stimulus. The aim of this study was to determine the muscle-specific response of vastus lateralis (VL) MU features following a functional stepping task employing bilateral CON and ECC movements performed to failure. It was hypothesised that strength and force steadiness would decline in both legs as a result of performance fatigue. Due to the higher metabolic demand of CON movements, it was hypothesised MU function would be more greatly affected by CON movements than ECC. nlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License The neural mechanisms influencing performance fatigue, including MU recruitment thresholds, firing rate (FR) and voluntary activation, appear to differ across exercise modalities (Duchateau & Baudry, 2014; Kay et al., 2000) with some data showing an increase in MU FR following fatiguing exercise (Dartnall et al., 2009; Piitulainen et al., 2012), while others report a decrease (Adam & De Luca, 2005; Contessa et al., 2016; Kuchinad et al., 2004; Rubinstein & Kamen, 2005; Stock et al., 2012). More recently, a direct comparison of contraction modalities showed a greater increase in MU FR following CON, when compared to ECC contractions (Hirono et al., 2022). These equivocal findings may be explained by a muscle specific effect or by differences in the way fatigue was induced, with reductions in maximal contraction following prolonged low-intensity activity occurring as a result of reduced muscle activation, and reductions following high- intensity activity attributable to impaired contractile function (Enoka & Duchateau, 2016). 2 1 INTRODUCTION JONES ET AL. 2 New Findings or neuromuscular disorders. All participants were asked to refrain from strenuous exercise 48 h prior to assessment. All participants were medically screened prior to the study via a medical history questionnaire and a resting electrocardiogram. Body mass and height were measured using calibrated scales and stadiometry, respectively, and BMI was calculated. real-time using Spike2 software (v9.06, Cambridge Electronic Design, Cambridge, UK) and data collected simultaneously from the force transducer for offline analysis. EMG data was collected during each of these contractions as detailed below. 2.2.4 Intramuscular EMG A 25 mm concentric needle electrode (74025-45/25 Neuroline; Ambu, Baltorpbakken, Denmark) was inserted directly above the HD-sEMG electrode. A voluntary, low force contraction was performed by the participant while the needle position was adjusted to ensure its tip was close to fibres belonging to active MUs (Piasecki et al., 2019; Stashuk, 1999a). The participant then performed isometric voluntary contra- ctions lasting 12−15 s, aiming to hold a target line set at 25% and 40% MVC as described above. The needle electrode was repositioned between contractions by combinations of rotating the bevel 180◦and withdrawing it by 10–25 mm to sample MUs at a range of depths (Jones et al., 2021). The procedure of needle positioning, voluntary contraction and signal recording was repeated until a minimum of six recordings at 25% from varying depths had been obtained to ensure sampling from a representative set of MUs. iEMG signals were amplified (D440, Digitimer, Welwyn Garden City, UK), acquired and bandpass filtered from 10 Hz to 10 kHz and sampled at 50 kHz (1401, Cambridge Electronic Design). The force and EMG signals were displayed in real-time using Spike2 software (v9.06), and data were stored for off-line analysis. The needle insertion site was recorded and matched from pre- to post-intervention. 2.2.1 Fatigue protocol To induce simultaneous CON and ECC fatigue in all participants, we employed a step up, step down protocol (Kostek et al., 2007). Participants concentrically contracted the quadriceps by stepping up onto a 43.5 cm-high bench with one leg, and stepping down with the opposing leg eccentrically, with each stepping contraction timed to a 3 s metronome. Participants wore a weighted vest (initially 25% of body weight but increasing up to 40% depending on tolerance) during the intervention. The task was performed until self-ascribed exhaustion, which was indicated as 10 using a modified Borg scale. The average time to exhaustion was 53 ± 12 min, with 60 s timed rest stop permitted when requested, for a maximum of three rest stops. All participants reported they were right-leg dominant, and the leg assigned to CON or ECC was randomised, as was the order in which the post-stepping assessments were conducted. All assessment procedureswereperformedimmediatelybeforethefatiguingprotocol, and post-testing began on the first limb within 3 min following completion of the protocol. All electrodes remained in place secured by tape between pre- and post-testing, so reapplication was not required. Assessment of the second limb began approximately 12 min later. 2.2.3 HD-sEMG A semi-disposable adhesive monopolar HD-sEMG matrix (GR08MM1305, OT Bioelettronica, Torino, Italy) consisting of 64 × 8 mm spaced electrodes was positioned in the presumed direction of the fascicles on both (right and left) VLs and secured to the skin before the pre-fatigue assessments. This remained in situ until after the post-assessments. A reference cable (CPAT1, OT Bioelettronica) was attached around the ankle of the recording limb. The signal was amplified, sampled at 2000 Hz, band-pass filtered (10–500 Hz) and converted to digital data (16-bit analog–digital converter, 3 dB bandwidth) using a Sessantaquatro (OT Bioelettronica) multi-channel amplifier. Data were visualised in real-time in OTBioLab+ software (v1.3.2, OT Bioelettronica) and stored for offline analysis. 2.2.2 Strength assessment Knee extensor strength was assessed with participants sitting with hips and knees flexed at 90◦and the leg securely fastened to a force transducer just above the ankle. To familiarise with the equipment and ‘warm-up’ the muscle, participants performed a series of sub- maximal contractions. They were then instructed to perform a maximal isometric contraction, accompanied by verbal encouragement and visual feedback of force. This was repeated three times, with 60 s rest intervals. The best effort was taken as maximum voluntary iso- metric contraction force (MVC). After determining MVC, participants performed isometric voluntary contractions each lasting 12−15 s, aiming to hold a target line set at 25% and 40% MVC displayed on a monitor. Following a familiarisation trial, HD-sEMG and iEMG were recorded simultaneously, with rest for ∼30 s between contra- ctions. Voluntary contraction and signal recording was repeated until a minimum of six recordings at 25% and three at 40% had been obtained. Contractions were performed alternating between 25% and 40% MVC. Post-intervention forces were normalised to the post- intervention MVC. The coefficient of variation of the force trace (CoV force)representedforcesteadiness. Theforcesignalsweredisplayedin 2.1 Ethical approval Seventeen healthy recreationally active volunteers (nine female, eight male) gave written informed consent to take part in this study, which was approved by the University of Nottingham Faculty of Medicine & Health Sciences Research Ethics Committee (186-1812). The study was conducted in accordance with the Declaration of Helsinki except for registration in a database. Data are available for 12 participants (six female, six male) for iEMG, and 10 participants (five female, five male) for HD-sEMG due to low yield of MUs at one or more time points and/or contraction levels. Exclusion criteria included a body mass index (BMI) <18 or >35 kg/m2, currently competing in sports at regional level or above, or a history of cardiovascular, respiratory High density surface EMG (HD-sEMG) records multiple MU potentials (MUPs) from a relatively large volume of muscle (Martinez- Valdes et al., 2016) and is well placed to assess FR of multiple MUs during a single contraction (Negro et al., 2016; Oliveira & Negro, 2021) and at larger contraction intensities (Del Vecchio et al., 2018). Additionally, the use of intramuscular EMG (iEMG) with concentric needles enables a more detailed view of MUPs via sampling without 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Cre JONES ET AL. 3 2.3.2 iEMG analysis Two-way repeated measures analysis of variance (ANOVA) with Šidák’s post hoc analysis were performed in GraphPad Prism (v9.2, GraphPad Sofware, San Diego, CA, USA) to test for the effect of time and contraction modality on MVC and CoV force. Where no significant contraction modality × time interaction was found, main effects are reported. As multiple MUPs were recorded from each participant, multi-level mixed effects linear regression models were performed in StataSE (v15.0, StataCorp LLC, College Station, TX, USA) with contraction modality and time as factors and contraction × time interactions included in each model. In each model the first level (individual MUs) was clustered according to each participant to form the second level. This modelling framework is suitable for data of this nature as it incorporates all sampled MUs as opposed to only the mean values obtained from each participant, which pre- serves variability to a greater extent within and across participants simultaneously. Regression coefficients (β) and 95% confidence inter- vals (CI) are reported that indicate the magnitude and direction of the effects of interest. Significance was assumed if P < 0.05. The procedures for recording and analysing individual MUPs have been described in detail previously (Piasecki, Ireland, Coulson et al., 2016; Piasecki, Ireland, Stashuk et al., 2016). Intramuscular signals from 25% contractions only were analysed using decomposition-based quantitative electromyography (DQEMG) (Stashuk, 1999b). DQEMG was used to automatically extract MU potential trains (MUPTs) of separate MUs from an iEMG signal and calculate a MUP template for each extracted MUPT. MUPTs composed of MUPs from more than one MU or with fewer than 30 MUPs were excluded. MUP templates of included MUPTs were visually inspected and markers corresponding to the onset, end, and positive and negative peaks adjusted where required. MUP duration, in ms, is the time interval between the MUP template onset and end, MUP area, in μVms, is the MUP template area between its onset and end, MUP amplitude, in mV, is the difference between the MUP template positive and negative peak values, and MUP thickness is MUP area divided by MUP amplitude and describes the shape of the MUP template (Abdelmaseeh et al., 2014). MUP complexity was assessed using the number of turns in the MUP template. 2.3.1 HD-sEMG analysis 2.3.1 HD-sEMG analysis HD-sEMG data was analysed in MATLAB (v2019a, IBM) using custom written scripts to decompose and identify MUs using an extensively validated method (Martinez-Valdes et al., 2017; Negro et al., 2016). HD-sEMG data from the two middle contractions (contraction 3 and JONES ET AL. 4 TABLE 1 Participant characteristics (n = 12). Mean (SD) Age (years) 21 (0.6) Height (m) 1.74 (0.11) Weight (kg) 71.9 (13.5) BMI (kg/m2) 23.7 (3.4) TABLE 1 Participant characteristics (n = 12). 4 out of six recorded) at 25% and from contraction 2 and 3 at 40% MVC were analysed. The HD-sEMG recording electrodes remained in place during the intervention to improve the probability of sampling the same MUs. The accuracy of the decomposition for each MU was tested with the silhouette (SIL) measure, which is a normalised accuracy index for sEMG decomposition (Negro et al., 2016). Only MUPs with a SIL greater than 0.90 were included for further analysis. Subsequently, the decomposition accuracy was improved by manual editing of consecutive firings (Afsharipour et al., 2020; Boccia et al., 2019). Mean FR and the coefficient of variation for the interspike interval (FR variability) from recruited MUs were calculated from the HD-sEMG signals from the sustained force section of the contractions. Following decomposition, individual MUs were tracked by a previously validated technique based on cross-correlation of single-differential 2D MUPs (Martinez-Valdes et al., 2017). In this procedure, matched MUPsbetweenpre-and post-fatigue trialswere visuallyinspected, and the two identified MU were regarded as the same when they had a cross-correlation coefficient >0.80. of the needle electrode, ensuring only the nearest fibres significantly contribute to the NFM and reducing interference from distant active fibres of other MUs (Piasecki et al., 2021; Stashuk, 1999a). All NFMs werevisuallyinspectedandthosecontainingcontaminationfromother NFMs were removed. NFM jiggle is a measure of the shape variability of consecutive NFMs of an MUPT expressed as a percentage of the NFM template total area (Piasecki et al., 2021) and is representative of NMJ transmission instability (Figure 1c). 2.3.2 iEMG analysis A ‘turn’ was defined as a change in direction of the MUP template of at least 25 μV and indicates the level of temporal dispersion across individual muscle fibre contributions to a single MUP. MUP negative peak slope ratio was calculated as the absolute value of the rise of the MUP template negative peak, across the 500 μs interval before the negative peak, divided by the fall of MUP template negative peak, across the 500 μs interval after the peak (Figure 1b). While the MUP template negative peak rise and fall slopes will each be similarly influenced by the relative location of the recording electrode, their ratio can represent relative rates of ion exchange during the depolarisation and repolarisation phases of an action potential, respectively. A near fibre MUP (NFM) is calculated by applying a low pass, second-order differential filter to its corresponding MUP, which effectively reduces the recording area 3.1 Participant characteristics Participant characteristics (6 M/6 F) are presented in Table 1. Participant characteristics (6 M/6 F) are presented in Table 1. 3.2.1 Functional properties There was no significant contraction modality × time interaction (P = 0.742) for MVC but there was a main effect of time (P < 0.0001; Figure 2a), with MVC decreasing approximately 15.8% with CON 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/20 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules JONES ET AL. 5 FIGURE 1 (a) Example force trace at 25% MVC (red) with corresponding iEMG (green) and subset of 4 HD-sEMG channels (blue). (b) Representative MUP template indicating negative peak slope ratio. (c) Representative MUP templates indicating MUP turns (T) and corresponding near-fibre MUPs (NFMs) shown in raster and shimmer plots. Abbreviations: MUP, motor unit potential; NFM, near fibre motor unit potential. oc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/ nelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditio y.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Te FIGURE 1 (a) Example force trace at 25% MVC (red) with corresponding iEMG (green) and subset of 4 HD-sEMG channels (blue). (b) Representative MUP template indicating negative peak slope ratio. (c) Representative MUP templates indicating MUP turns (T) and corresponding near-fibre MUPs (NFMs) shown in raster and shimmer plots. Abbreviations: MUP, motor unit potential; NFM, near fibre motor potential FIGURE 2 Individual maximum voluntary contraction (MVC) (a) and force steadiness at 25% (b) and 40% MVC (c) before (PRE, blue) and after (POST, red) concentric (CON) and eccentric (ECC) loading. Interaction (Int) and effect of time (TE) from two-way ANOVAs are shown. FIGURE 2 Individual maximum voluntary contraction (MVC) (a) and force steadiness at 25% (b) and 40% MVC (c) before (PRE, blue) and after (POST, red) concentric (CON) and eccentric (ECC) loading. Interaction (Int) and effect of time (TE) from two-way ANOVAs are shown. (Pre vs. Post: 473 ± 143 N vs. 386 ± 126 N) and 20.6% with ECC (463 ± 142 N vs. 363 ± 109 N). TABLE 2 Group means, regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features (n = 10). Additionally, in these tracked MUs, there was no significant contraction modality × time interaction for FR variability at 25% (P = 0.663) with FR variability not significantly changing in CON (β = 1.96, P = 0.08) or ECC (β = 1.21, P = 0.352) following fatigue (Figure 4c). At 40% there was no significant contraction modality × time interaction (P = 0.203) and FR variability increased in the CON leg (β = 2.62, P = 0.011) but did not change in the ECC leg (β = 0.64, P = 0.580; Table 3, Figure 4d, f). significant contraction modality × time interaction for FR variability at 25% MVC (P = 0.526), although it was higher post-exercise at 25% MVC (CON; β = 2.32, P < 0.001, ECC; β = 1.79, P = 0.004). At 40% MVC there was also no significant contraction modality × time inter- action (P = 0.902) for FR variability, with an increase in both the CON leg (β = 2.27, P = 0.002) and the ECC leg (β = 2.35, P = 0.006) (Table 2, Figure 3b). significant contraction modality × time interaction for FR variability at 25% MVC (P = 0.526), although it was higher post-exercise at 25% MVC (CON; β = 2.32, P < 0.001, ECC; β = 1.79, P = 0.004). At 40% MVC there was also no significant contraction modality × time inter- action (P = 0.902) for FR variability, with an increase in both the CON leg (β = 2.27, P = 0.002) and the ECC leg (β = 2.35, P = 0.006) (Table 2, Figure 3b). TABLE 2 Group means, regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features (n = 10). TABLE 2 Group means, regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features (n = 10). TABLE 2 Group means, regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features (n = 10). Mean (SD) PRE POST β 95% CI P FR (Hz) – 25% CON 7.73 (1.23) 8.00 (1.53) 0.34 −0.08, 0.75 0.112 ECC 7.71 (1.02) 8.78 (2.33) 1.44 0.98, 1.90 <0.001 FR (Hz) – 40% CON 8.99 (1.84) 8.82 (2.02) −0.55 −1.15, 0.05 0.072 ECC 8.81 (1.41) 10.1 (4.08) 1.62 0.91, 2.33 <0.001 FR variability (%) – 25% CON 12.0 (2.02) 13.6 (2.20) 2.32 1.22, 3.41 <0.001 ECC 12.7 (2.36) 14.5 (3.20) 1.79 0.589, 2.99 0.004 FR variability (%) – 40% CON 13.8 (2.16) 16.1 (3.18) 2.27 0.49, 4.05 0.002 ECC 13.8 (2.49) 15.5 (5.02) 2.35 0.67, 4.03 0.006 Significant values shown in bold. CON, concentric; ECC, eccentric; FR, firing rate. β 95% CI P FR (Hz) – 25% CON −0.15 −0.98, 0.68 0.725 ECC 1.12 0.14, 2.09 0.025 FR (Hz) – 40% CON −0.27 −1.01, 0.47 0.471 ECC 1.17 0.34, 2.00 0.006 FR variability (%) – 25% CON 1.96 −0.23, 4.15 0.080 ECC 1.21 −1.34, 3.76 0.352 FR variability (%) – 40% CON 2.62 0.588, 4.65 0.011 ECC 0.64 −1.63, 2.91 0.580 Significant values shown in bold. CON, concentric; ECC, eccentric; FR, firing rate. Significant values shown in bold. CON, concentric; ECC, eccentric; FR, firing rate. Significant values shown in bold. CON, concentric; ECC, eccentric; FR, firing rate. with ECC FR increased at 40% MVC (β = 1.17, P = 0.006; Figure 4b, e). Additionally, in these tracked MUs, there was no significant contraction modality × time interaction for FR variability at 25% (P = 0.663) with FR variability not significantly changing in CON (β = 1.96, P = 0.08) or ECC (β = 1.21, P = 0.352) following fatigue (Figure 4c). At 40% there was no significant contraction modality × time interaction (P = 0.203) and FR variability increased in the CON leg (β = 2.62, P = 0.011) but did not change in the ECC leg (β = 0.64, P = 0.580; Table 3, Figure 4d, f). with ECC FR increased at 40% MVC (β = 1.17, P = 0.006; Figure 4b, e). 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Comm 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/20 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of JONES ET AL. 6 6 3.2.2 MU firing A total of 881 MUs were sampled with HD-sEMG, with a mean of 7 ± 2 per 25% contraction per leg, and a mean of 6 ± 2 per 40% contraction per leg. At 25% MVC there was a significant contraction modality × time interaction for FR (P < 0.001), with no difference following CON (β = 0.34, P = 0.112) but a significant increase following ECC (β = 1.44, P < 0.001). Similarly, at 40% MVC there was a significant contraction modality × time interaction (P < 0.001) for FR, with no significant change in the CON limb (β = −0.55, P = 0.072), but an increase in the ECC limb (β = 1.62, P < 0.001) (Table 2, Figure 3a). There was no No significant contraction modality × time interaction for force steadiness was observed at 25% MVC but there was a main effect of time (P = 0.0016; Figure 2b) with a 21.7% increase with CON (3.51 ± 0.95% vs. 4.09 ± 0.95%) and a 48.8% increase with ECC (3.34 ± 0.99% vs. 5.04 ± 1.96%). No significant contraction modality × time interaction for force steadiness was observed at 40% MVC but a main effect of time was observed (P = 0.0038; Figure 2c) with a 20.9% increase with CON (3.71 ± 1.26% vs. 4.22 ± 1.51%) and a 56.1% increase with ECC (3.65 ± 1.34% vs. 5.45 ± 2.38%). FIGURE 3 Motor unit firing rate (FR) (a) and FR variability (b) forest plots showing β and 95% confidence intervals at 25% (top) and 40% maximum voluntary contraction (MVC) (bottom) following concentric (CON) and eccentric (ECC) exercise. Statistical analyses are based on multilevel mixed effects linear models. FIGURE 3 Motor unit firing rate (FR) (a) and FR variability (b) forest plots showing β and 95% confidence intervals at 25% (top) and 40% maximum voluntary contraction (MVC) (bottom) following concentric (CON) and eccentric (ECC) exercise. Statistical analyses are based on multilevel mixed effects linear models. TABLE 3 Regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features in tracked units (n = 10). TABLE 3 Regression coefficient (β) and 95% confidence interval (CI) for all HD-sEMG derived MU features in tracked units (n = 10). 3.2.3 Tracked MUs See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley On nlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License FIGURE 4 (a–d) Motor unit firing rate at 25% maximal voluntary contraction (MVC) (a) and 40% MVC (b) of individual tracked motor units for before and after concentric (CON) and eccentric (ECC) exercise, and motor unit firing rate variability at 25% MVC (c) and at 40% MVC (d). Points are colour coded for individual participants. (e, f) Motor unit firing rate (e) and firing rate variability (f) forest plots showing β and 95% confidence intervals for the change in CON and ECC exercise at both 25% (top) and 40% MVC (bottom) for the tracked MUs, from pre- to post-fatiguing exercise. Statistical analyses are based on multi-level mixed effects linear models. responses to a targeted fatiguing intervention and demonstrate that individual MU features are mediated by contraction type. Both ECC and CON resulted in similar reductions in isometric strength and force steadiness which were matched by an increase in FR variability. The neural response differed across limbs within an individual at sub-maximal levels, with only ECC loading resulting in an increased FR during normalised force levels, and CON loading altering fibre membrane properties to a greater extent than ECC. These findings are important in understanding neuromuscular function following exercise-induced performance fatigue and highlight the influence of contraction type on these parameters. negative peak slope ratio significantly increased (β = 0.411, P = 0.018), but there was no change with ECC (β = 0.159, P = 0.392; Table 4, Figure 5c). Although there was no significant contraction modality × time interaction for NMJ transmission instability (P = 0.244), it increased in both the CON (β = 1.66, P = 0.037) and ECC legs (β = 3.08, P = 0.001) (Table 4, Figure 5d). 3.2.3 Tracked MUs From HD-sEMG, 129 MUs (approx. 15%) were tracked from pre- to post-fatiguing exercise across 10 participants. At 25% MVC there was no significant contraction modality × time interaction for FR (P = 0.053), and no difference in FR was observed in the CON leg following fatigue (β = −0.15, P = 0.725; Figure 4a). However, with ECC FR increased at 25% (β = 1.12, P = 0.025, Table 3, Figure 4a). There was a significant contraction modality × time interaction in FR at 40% MVC (P = 0.011). There was no difference in FR in the CON leg following fatigue (β = −0.27, P = 0.471; Table 3; Figure 4b); however, as at 25%, Multi-level mixed effects regression models showed there was no significant contraction modality × time interaction for MUP thickness (P = 0.059) which did not differ in the CON (β = 0.0611, P = 0.130) or ECC leg (β = −0.0511, P = 0.244) (Table 4, Figure 5a). No significant contraction modality × time interaction was observed for MUP complexity (P = 0.58) with no significant change with either CON (β = −0.216, P = 0.168) or ECC (β = −0.088, P = 0.605; Table 4, Figure 5b). There was no significant contraction modality × time inter- action for MUP negative peak slope ratio (P = 0.319). With CON, MUP 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of JONES ET AL. 7 FIGURE 4 (a–d) Motor unit firing rate at 25% maximal voluntary contraction (MVC) (a) and 40% MVC (b) of individual tracked motor units for before and after concentric (CON) and eccentric (ECC) exercise, and motor unit firing rate variability at 25% MVC (c) and at 40% MVC (d). Points are colour coded for individual participants. (e, f) Motor unit firing rate (e) and firing rate variability (f) forest plots showing β and 95% confidence intervals for the change in CON and ECC exercise at both 25% (top) and 40% MVC (bottom) for the tracked MUs, from pre- to post-fatiguing exercise. Statistical analyses are based on multi-level mixed effects linear models. m/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. 4 DISCUSSION Here we present data using intramuscular needle electrodes in combination with HD-sEMG to investigate muscle specific 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Cre JONES ET AL. 8 8 JONES ET AL. FIGURE 5 Motor unit potential (MUP) thickness (a), MUP turns (b), MUP negative peak slope ratio (c) and near fibre motor unit potential (NFM) features of neuromuscular junction (NMJ) transmission instability (d) forest plots showing β and 95% confidence intervals for the change after concentric (CON) and eccentric (ECC) exercise. Statistical analyses are based on multi-level mixed effects linear models. FIGURE 5 Motor unit potential (MUP) thickness (a), MUP turns (b), MUP negative peak slope ratio (c) and near fibre motor unit potential (NFM) features of neuromuscular junction (NMJ) transmission instability (d) forest plots showing β and 95% confidence intervals for the change after concentric (CON) and eccentric (ECC) exercise. Statistical analyses are based on multi-level mixed effects linear models. TABLE 4 Group means, regression coefficient (β) and 95% confidence interval (CI) for all iEMG derived motor unit and near fibre features (n = 12). TABLE 4 Group means, regression coefficient (β) and 95% confidence interval (CI) for all iEMG derived motor unit and near fibre features (n = 12). also increased following both ECC and CON, at both contraction intensities. MU FR increased following ECC but did not change following CON, when assessed at 25% and 40% of MVC. Contrasting findings of FR have been observed during ECC and CON contractions in previous studies (Altenburg et al., 2008; Linnamo et al., 2003), but here we demonstrate different FR responses during an identical iso- metric task, after an ECC and CON fatiguing intervention. In a recent unilateral crossover study, tracked MUs increased their FR to a greater degree following CON when compared to ECC, again occurring in a contraction-level specific manner (Hirono et al., 2022). However, these MU FRs were calculated across a broad range of contraction levels during ramped contractions, making the studies difficult to compare, and fatigue was induced by machine-based exercise as opposed to the functional task applied here. Previous findings also show muscle twitch force reduced to a greater extent following ECC, when compared to CON contractions (Piitulainen et al., 2011), and the marked increase in FR following ECC observed in the current study likely occurs as a central mechanism to compensate for declines in twitch force (Pincheira et al., 2021). These findings are also supported by the observations of Piitulainen et al. (2012), where FR increased during sustained contractions following ECC exercise. features (n = 12). Mean (SD) PRE POST β 95% CI P MUP thickness CON 1.34 (0.41) 1.34 (0.46) 0.0611 −0.018, 0.140 0.130 ECC 1.50 (0.44) 1.48 (0.57) −0.0511 −0.137, 0.035 0.244 MUP complexity CON 4.1 (1.0) 4.1 (0.8) −0.216 −0.523, 0.091 0.168 ECC 4.0 (1.1) 3.9 (1.0) −0.088 −0.421, 0.245 0.605 MUP neg peak slope ratio CON 1.87 (0.28) 2.14 (0.53) 0.411 0.072, 0.751 0.018 ECC 2.07 (0.72) 1.97 (0.28) 0.159 −0.204, 0.521 0.392 NMJ transmission instability (%) CON 14.2 (3.3) 16.0 (4.7) 1.66 0.10, 3.23 0.037 ECC 14.8 (1.9) 17.2 (3.3) 3.08 1.29, 4.88 0.001 Significant values shown in bold. CON, concentric; ECC, eccentric; MUP, motor unit potential; NMJ, neuromuscular junction. 4.1 Limitations The fatiguing exercise protocol employed here required CON and ECC contractions performed to self-reported exhaustion, but it is not possible to directly quantify the contribution of each contraction type to total exhaustion. We explored MU features up to 40% of MVC only, and further contraction-level specific adaptations may be present at higher contractions where MUs with MU fibre type proportions may differ. Here, we were not investigating possible muscle damage effects, and therefore changes to contractile properties were not assessed. The possible influence of sex on these parameters is unclear and we did not control for hormonal status in females. However, whilst this may be viewed as a mechanistic limitation, we believe the full exclusion of females in studies of this nature would be more limiting and our pre- vious work has found no differences in neuromuscular recruitment strategies between sexes (Guo et al., 2022). fatiguing exercise (Hirono et al., 2022), the relatively low proportion of MUs able to be tracked prevents robust reporting of this in the current study. exercise tolerance due to muscle fatigue and weakness present a challenge in strength training (Gault & Willems, 2013). exercise tolerance due to muscle fatigue and weakness present a challenge in strength training (Gault & Willems, 2013). The simultaneous use of iEMG enabled the collection of more detailed information on peripheral MU parameters following differing loading strategies. MUP thickness describes the shape and size of a MUP and no changes were observed following either exercise modality. The number of turns of a MUP is an assessment of MUP complexity and reflects increased temporal dispersion across MU propagating action potentials. It is applied in clinical settings to investigate myopathic and neuropathic conditions alongside other MU parameters (Allen et al., 2015; De Carvalho et al., 2014), and increases acutely following limb immobilisation (Inns et al., 2022; Sarto et al., 2022). However, similar to MUP thickness, complexity did not change following CON or ECC exercise. NF jiggle is a measure of the variability of temporal dispersion across MU propagating action potentials occurring primarily due to NMJ transmission variability (Piasecki et al., 2021), and this increased following both contraction modes. Although identifying pre- and post- synaptic NMJ dysregulation in response to fatigue is extremely difficult in humans, physiological plausibility exists relative to both sites; data from animal models has shown a depletion of synaptic vesicles following prolonged stimulation mimicking fatigue (Wu & Betz, 1998), and increased exposure to acetylcholine (ACh) can result in ACh receptor desensitisation (Magleby & Pallotta, 1981). These distinct differences in NF jiggle have also been observed across age (Hourigan et al., 2015; Piasecki et al., 2020) and in disease (Allen et al., 2015), but as a result of MU remodelling functionally affecting the NMJ (Jones et al., 2022), which is unlikely to have occurred acutely in the current study. Notably, NMJ transmission instability increased at a lower absolute force (normalised to 25% post-fatigue MVC), which is the opposite of the expected trend of increasing with larger forces (Guo et al., 2022). 4.2 Conclusion The current findings highlight how MU features respond to fatigue in a contraction-dependent manner, and how despite similar reductions in strength after fatiguing exercise, MU FR responded opposingly following concentric and eccentric contractions. The instability of NMJ transmission increased following both exercise modes, but markers of fibre membrane excitability were altered following the more metabolically demanding CON loading only, which likely reflects an accumulation of extracellular K+ and a slowing of fibre repolarisation. Differences in the MU response to concentric and eccentric induced fatigue are important when considering exercise training protocols, particularly in populations with musculoskeletal limitations. The MUP negative peak slope ratio quantifies the relationship between the rise and fall slopes of the negative peak of the MUP template, reflecting the ratio of ionic exchange rates across the depolarisation and repolarisation phases of the action potential, respectively. This ratio increased following CON exercise only, which was explained by a shorter rise (depolarisation) time relative to the fall time (repolarisation), the latter of which is achieved via the efflux of intracellular K+. Repeated activation of skeletal muscle results in a net efflux of K+, and thus an increase in extracellular K+ concentration in fatigued muscle (Allen et al., 2008; Fortune & Lowery, 2009). In this fatigued state, K+ is required to move up its concentration gradient and may explain the slowing of repolarisation (relative to depolarisation) occurring in the limb with higher metabolic cost (the CON leg). 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and C JONES ET AL. 4/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the appl .com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licens From the smaller sample of MUs tracked following fatigue, the differences in firing properties generally followed a similar pattern to the population results, showing an increase in FR in ECC but no significant change in CON. However, FR variability increased in CON at 40% MVC only following fatigue, but no statistical differences were observed at the lower intensity or after ECC in these tracked units. The percentage of MUs tracked is lower than previously reported and may reflect the methodological approach (e.g., HD electrodes remaining in place during intervention), an alteration of MUP duration as a result of fatigue (McManus et al., 2017) or differences in MU populations recruited immediately post-fatigue (Martinez-Valdes et al., 2017). Although previous research has found moderate to high level recruitment thresholds to decrease following concentric and eccentric The functional endurance-based intervention utilised in this study was selected to induce performance fatigue via limiting muscle activation as opposed to contractile function (Enoka & Duchateau, 2016), and the form of movement mimics that of ascending and descending stairs, a highly functional task with translational implications to ageing and disease. Both forms of contraction elicited similar decreases in maximal isometric strength, which supports similar previous findings in this muscle group (Molinari et al., 2006). Neuromuscular control, as assessed by force steadiness at normalised force levels across limbs, deteriorated in both legs. The ability to hold a steady contraction with minimal fluctuation requires effective neuro- modulation of motor output (Enoka & Farina, 2021), and FR variability REFERENCES Abdelmaseeh, M., Smith, B., & Stashuk, D. (2014). Feature selection for motor unit potential train characterization. Muscle and Nerve, 49(5), 680–690. Gault, M. L., & Willems, M. E. T. (2013). Aging, functional capacity and eccentric exercise training. Aging and Disease, 4(6), 351–363. Adam, A., & De Luca, C. J. (2005). Firing rates of motor units in human vastus lateralis muscle during fatiguing isometric contractions. Journal of Applied Physiology, 99(1), 268–280. Guo, Y., Jones, E. J., Inns, T. B., Ely, I. A., Stashuk, D. W., Wilkinson, D. J., Smith, K., Piasecki, J., Phillips, B. E., Atherton, P. J., & Piasecki, M. (2022). Neuro- muscular recruitment strategies of the vastus lateralis according to sex. Acta Physiologica, 235(2), e13803. Afsharipour, B., Manzur, N., Duchcherer, J., Fenrich, K. F., Thompson, C. K., Negro, F., Quinlan, K. A., Bennett, D. J., & Gorassini, M. A. (2020). Estimation of self-sustained activity produced by persistent inward currents using firing rate profiles of multiple motor units in humans. Journal of Neurophysiology, 124(1), 63–85. Hirono, T., Kunugi, S., Yoshimura, A., Holobar, A., & Watanabe, K. (2022). Acute changes in motor unit discharge property after concentric versus eccentric contraction exercise in knee extensor. Journal of Electro- myography and Kinesiology, 67, 102704. Allen, D. G., Lamb, G. D., & Westerblad, H. (2008). Skeletal muscle fatigue: Cellular mechanisms. Physiological Reviews, 88(1), 287–332. Hody, S., Croisier, J. L., Bury, T., Rogister, B., & Leprince, P. (2019). Eccentric muscle contractions: Risks and benefits. Frontiers in Physiology, 1, 1–18. Allen, M. D., Stashuk, D. W., Kimpinski, K., Doherty, T. J., Hourigan, M. L., & Rice, C. L. (2015). Increased neuromuscular transmission instability and motor unit remodelling with diabetic neuropathy as assessed using novel near fibre motor unit potential parameters. Clinical Neurophysiology, 126(4), 794–802. Hourigan, M. L., McKinnon, N. B., Johnson, M., Rice, C. L., Stashuk, D. W., & Doherty, T. J. (2015). Increased motor unit potential shape variability across consecutive motor unit discharges in the tibialis anterior and vastus medialis muscles of healthy older subjects. Clinical Neurophysiology, 126(12), 2381–2389. Altenburg, T. M., De Ruiter, C. J., Verdijk, P. W. L., Van Mechelen, W., & De Haan, A. (2008). Vastus lateralis surface and single motor unit EMG following submaximal shortening and lengthening contractions. Applied Physiology, Nutrition and Metabolism, 33(6), 1086–1095. Hunter, S. K., Duchateau, J., & Enoka, R. M. (2004). Muscle fatigue and the mechanisms of task failure. Exercise and Sport Sciences Reviews, 32(2), 44– 49. AUTHOR CONTRIBUTIONS Eleanor J. Jones, Philip J. Atherton, Bethan E. Phillips and Mathew Piasecki contributed to the conception and design of the work. Eleanor J. Jones and Yuxiao Guo acquired the data. Eleanor J. Jones, Yuxiao Guo,EduardoMartinez-ValdesandMathewPiaseckianalysedthedata. Eleanor J. Jones and Mathew Piasecki drafted the manuscript and prepared the figures. Eleanor J. Jones, Yuxiao Guo, Eduardo Martinez- Valdes, Francesco Negro, Daniel W. Stashuk and Mathew Piasecki contributed to the interpretation of the results. All authors have read and approved the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. Collectively, these findings of adaptation of MU features following fatiguing contractions as a direct result of contraction modalities has implications for interventional strategies intended to target neural input and MU function. This knowledge may influence the contraction type employed for training depending on the requirements of the targeted population. This may also have translational relevance in understanding conditions such as sarcopenia and cancer where JONES ET AL. 10 Mathew Piasecki https://orcid.org/0000-0002-7804-4631 Enoka, R. M., & Farina, D. (2021). Force steadiness: From motor units to voluntary actions. Physiology, 36(2), 114–130. Fortune, E., & Lowery, M. M. (2009). Effect of extracellular potassium accumulation on muscle fiber conduction velocity: A simulation study. Annals of Biomedical Engineering, 37(10), 2105–2117. FUNDING INFORMATION M.P., P.J.A., B.E.P. and E.J.J. are supported by the Medical Research Council (grant number MR/P021220/1) as part of the MRC-Versus Arthritis Centre for Musculoskeletal Ageing Research awarded to the Universities of Nottingham and Birmingham, and by the NIHR Nottingham Biomedical Research Centre. Duchateau, J., & Baudry, S. (2014). Insights into the neural control of eccentric contractions. Journal of Applied Physiology, 116(11), 1418– 1425. Enoka, R. M., & Duchateau, J. (2008). Muscle fatigue: What, why and how it influences muscle function. The Journal of Physiology, 586(1), 11–23. Enoka, R. M., & Duchateau, J. (2016). Translating fatigue to human performance. Medicine and Science in Sports and Exercise, 48(11), 2228– 2238. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. Del Vecchio, A., Negro, F., Felici, F., & Farina, D. (2018). Distribution of muscle fibre conduction velocity for representative samples of motor units in the full recruitment range of the tibialis anterior muscle. Acta Physiologica, 222(2), e12930. ACKNOWLEDGEMENTS Cook, C. J., Beaven, C. M., & Kilduff, L. P. (2013). Three weeks of eccentric training combined with overspeed exercises enhances power and running speed performance gains in trained athletes. Journal of Strength and Conditioning Research, 27(5), 1280–1286. We are grateful to Dr Andrea Venn from the University of Nottingham for assistance with statistical analysis. Strength and Conditioning Research, 27(5), 1280–1286. Dartnall, T. J., Rogasch, N. C., Nordstrom, M. A., & Semmler, J. G. (2009). Eccentric muscle damage has variable effects on motor unit recruitment thresholds and discharge patterns in elbow flexor muscles. Journal of Neurophysiology, 102(1), 413–423. CONFLICT OF INTEREST The authors declare no conflicts of interest. De Carvalho, M., Turkman, A., & Swash, M. (2014). Sensitivity of MUP parameters in detecting change in early ALS. Clinical Neurophysiology, 125(1), 166–169. ORCID Enoka, R. M., & Duchateau, J. (2017). Rate coding and the control of muscle force. Cold Spring Harbor Perspectives in Medicine, 7(10), a029702. Eleanor J. Jones https://orcid.org/0000-0003-3261-3787 Eleanor J. Jones https://orcid.org/0000-0003-3261-3787 Mathew Piasecki https://orcid.org/0000-0002-7804-4631 Mathew Piasecki https://orcid.org/0000-0002-7804-4631 The effects of eccentric versus concentric resistance training on muscle strength and mass in healthy adults: A systematic review with meta-analysis. British Journal of Sports Medicine, 43(8), 556–568. Martinez-Valdes, E., Negro, F., Laine, C. M., Falla, D., Mayer, F., & Farina, D. (2017). Tracking motor units longitudinally across experimental sessions with high-density surface electromyography. The Journal of Physiology, 595(5), 1479–1496. Rubinstein, S., & Kamen, G. (2005). Decreases in motor unit firing rate during sustained maximal-effort contractions in young and older adults. Journal of Electromyography and Kinesiology, 15(6), 536–543. McManus, L., Hu, X., Rymer, W. Z., Suresh, N. L., & Lowery, M. M. (2017). Motor unit activity during fatiguing isometric muscle contraction in hemispheric stroke survivors. Frontiers in Human Neuroscience, 11, 569. Sarto, F., Stashuk, D. W., Franchi, M. V., Monti, E., Zampieri, S., Valli, G., Sirago, G., Candia, J., Hartnell, L. M., Paganini, M., McPhee, J. S., De Vito, G., Ferrucci, L., Reggiani, C., & Narici, M. V. (2022). Effects of short-term unloading and active recovery on human motor unit properties, neuro- muscular junction transmission and transcriptomic profile. The Journal of Physiology, 600(21), 4731–4751. Merletti, R., Farina, D., Gazzoni, M., & Schieroni, M. P. (2002). Effect of age on muscle functions investigated with surface electromyography. Muscle & Nerve, 25(1), 65–76. Molinari, F., Knaflitz, M., Bonato, P., & Actis, M. V. (2006). Electrical manifestations of muscle fatigue during concentric and eccentric iso- kinetic knee flexion-extension movements. IEEE Transactions on Bio- Medical Engineering, 53(7), 1309–1316. Souron, R., Nosaka, K., & Jubeau, M. (2018). Changes in central and peri- pheral neuromuscular fatigue indices after concentric versus eccentric contractions of the knee extensors. European Journal of Applied Physio- logy, 118(4), 805–816. Negro, F., Muceli, S., Castronovo, A. M., Holobar, A., & Farina, D. (2016). Multi-channel intramuscular and surface EMG decomposition by convolutive blind source separation. Journal of Neural Engineering, 13(2), 026027. Stashuk, D. W. (1999a). Detecting single fiber contributions to motor unit action potentials. Muscle and Nerve, 22(2), 218–229. Oliveira, A. S., & Negro, F. (2021). Neural control of matched motor units during muscle shortening and lengthening at increasing velocities. Journal of Applied Physiology, 130(6), 1798–1813. Stashuk, D. W. (1999b). Decomposition and quantitative analysis of clinical electromyographic signals. Medical Engineering & Physics, 21(6–7), 389– 404. Overend, T. J., Versteegh, T. H., Thompson, E., Birmingham, T. B., & Vandervoort, A. A. (2000). Kostek, M. C., Chen, Y. W., Cuthbertson, D. J., Shi, R., Fedele, M. J., Esser, K. A., & Rennie, M. J. (2007). Gene expression responses over 24 h to lengthening and shortening contractions in human muscle: Major changes in CSRP3, MUSTN1, SIX1, and FBXO32. Physiological Genomics, 31(1), 42–52. Piasecki, M., Ireland, A., Stashuk, D., Hamilton-Wright, A., Jones, D. A., & McPhee, J. S. (2016). Age-related neuromuscular changes affecting human vastus lateralis. The Journal of Physiology, 594(16), 4525–4536. Piitulainen, H., Botter, A., Merletti, R., & Avela, J. (2011). Muscle fiber conduction velocity is more affected after eccentric than concentric exercise. European Journal of Applied Physiology, 111(2), 261–273. Kuchinad, R., Ivanova, T., & Garland, S. J. (2004). Modulation of motor unit discharge rate and H-reflex amplitude during submaximal fatigue of the human soleus muscle. Experimental Brain Research, 158(3), 345–355. Piitulainen, H., Holobar, A., & Avela, J. (2012). Changes in motor unit characteristicsaftereccentricelbowflexorexercise. ScandinavianJournal of Medicine & Science in Sports, 22(3), 418–429. Linnamo, V., Bottas, R., & Komi, P. V. (2000). Force and EMG power spectrum during and after eccentric and concentric fatigue. Journal of Electromyography and Kinesiology, 10(5), 293–300. Pincheira, P. A., Martinez-Valdes, E., Guzman-Venegas, R., Falla, D., Garrido, M. I., Cresswell, A. G., & Lichtwark, G. A. (2021). Regional changes in muscle activity do not underlie the repeated bout effect in the human gastrocnemius muscle. Scandinavian Journal of Medicine & Science in Sports, 31(4), 799–812. Linnamo, V., Moritani, T., Nicol, C., & Komi, P. V. (2003). Motor unit activation patterns during isometric, concentric and eccentric actions at different force levels. Journal of Electromyography and Kinesiology, 13(1), 93–101. Prinsen, H., Van Dijk, J. P., Zwarts, M. J., Leer, J. W. H., Bleijenberg, G., & Van Laarhoven, H. W. M. (2015). The role of central and peripheral muscle fatigue in postcancer fatigue: A randomized controlled trial. Journal of Pain and Symptom Management, 49(2), 173–182. Magleby, K. L., & Pallotta, B. S. (1981). A study of desensitization of acetyl- choline receptors using nerve-released transmitter in the frog. The Journal of Physiology, 316(1), 225–250. Martinez-Valdes, E., Laine, C. M., Falla, D., Mayer, F., & Farina, D. (2016). High-density surface electromyography provides reliable estimates of motor unit behavior. Clinical Neurophysiology, 127(6), 2534–2541. Roig, M., O’Brien, K., Kirk, G., Murray, R., McKinnon, P., Shadgan, B., & Reid, W. D. (2009). Cardiovascular stress associated with concentric and eccentric isokinetic exercise in young and older adults. Journals Gerontol - Series A, Biological Sciences and Medical Sciences, 55(4), B177–B182. Stock, M. S., Beck, T. W., & Defreitas, J. M. (2012). Effects of fatigue on motor unit firing rate versus recruitment threshold relationships. Muscle & Nerve, 45(1), 100–109. Webber, S., & Kriellaars, D. (1997). Neuromuscular factors contributing to in vivo eccentric moment generation. Journal of Applied Physiology, 83(1), 40–45. Peñailillo, L., Blazevich, A., Numazawa, H., & Nosaka, K. (2013). Metabolic and muscle damage profiles of concentric versus repeated eccentric cycling. Medicine and Science in Sports and Exercise, 45(9), 1773–1781. Wu, L.-G., & Betz, W. J. (1998). Kinetics of synaptic depression and vesicle recycling after tetanic stimulation of frog motor nerve terminals. Biophysical Journal, 74(6), 3003–3009. Piasecki, J., Inns, T. B., Bass, J. J., Scott, R., Stashuk, D. W., Phillips, B. E., Atherton, P. J., & Piasecki, M. (2020). Influence of sex on the age-related adaptations of neuromuscular function and motor unit properties in elite masters athletes. The Journal of Physiology, 599(1), 193–205. 1469445x, 0, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/EP091058 by Test, Wiley Online Library on [20/04/2023]. See the Terms and C JONES ET AL. 11 REFERENCES Boccia, G., Martinez-Valdes, E., Negro, F., Rainoldi, A., & Falla, D. (2019). Motor unit discharge rate and the estimated synaptic input to the vasti muscles is higher in open compared with closed kinetic chain exercise. Journal of Applied Physiology, 127(4), 950–958. Inns, T. B., Bass, J. J., Hardy, E. J. O., Wilkinson, D. J., Stashuk, D. W., Atherton, P. J., Phillips, B. E., & Piasecki, M. (2022). Motor unit dysregulation following 15 days of unilateral lower limb immobilisation. The Journal of Physiology, 600(21), 4753–4769. Burtin, C., Saey, D., Saglam, M., Langer, D., Gosselink, R., Janssens, W., Decramer, M., Maltais, F., & Troosters, T. (2012). Effectiveness of exercise training in patients with COPD: The role of muscle fatigue. European Respiratory Journal, 40(2), 338–344. Jones, E. J., Chiou, S.-Y., Atherton, P. J., Phillips, B. E., & Piasecki, M. (2022). Ageing and exercise-induced motor unit remodelling. The Journal of Physiology, 600(8), 1839–1849. Jones, E. J., Piasecki, J., Ireland, A., Stashuk, D. W., Atherton, P. J., Phillips, B. E., McPhee, J. S., & Piasecki, M. (2021). Lifelong exercise is associated with more homogeneous motor unit potential features across deep and superficial areas of vastus lateralis. GeroScience, 43(4), 1555– 1565. Chaudhuri, A., & Behan, P. O. (2004). Fatigue in neurological disorders. Lancet, 363(9413), 978–988. Christie, A., Snook, E. M., & Kent-Braun, J. A. (2011). Systematic review and meta-analysis of skeletal muscle fatigue in old age. Medicine and Science in Sports and Exercise, 43(4), 568–577. Kay, D., St Clair Gibson, A., Mitchell, M. J., Lambert, M. I., & Noakes, T. D. (2000). Different neuromuscular recruitment patterns during eccentric, concentric and isometric contractions. In Journal of Electromyography and Kinesiology. pp. 425–431. Elsevier. Contessa, P., De Luca, C. J., & Kline, J. C. (2016). The compensatory inter- action between motor unit firing behavior and muscle force during fatigue. Journal of Neurophysiology, 116(4), 1579–1585. SUPPORTING INFORMATION Additional supporting information can be found online in the Piasecki, M., Garnés-Camarena, O., & Stashuk, D. W. (2021). Near-fiber electromyography. Clinical Neurophysiology, 132(5), 1089–1104. Additional supporting information can be found online in the Supporting Information section at the end of this article. Supporting Information section at the end of this article. A articles are governed by the applicable Creative Commons License Piasecki, M., Ireland, A., Coulson, J., Stashuk, D. W., Hamilton-Wright, A., Swiecicka, A., Rutter, M. K., McPhee, J. S., & Jones, D. A. (2016). Motor unit number estimates and neuromuscular transmission in the tibialis anterior of master athletes: Evidence that athletic older people are not spared from age-related motor unit remodeling. Physiological Reports, 4(19), e12987. How to cite this article: Jones, E. J., Guo, Y., Martinez-Valdes, E., Negro, F., Stashuk, D. W., Atherton, P. J., Phillips, B. E., & Piasecki, M. (2023). Acute adaptation of central and peripheral motor unit features to exercise-induced fatigue differs with concentric and eccentric loading. Experimental Physiology, 1–11. https://doi.org/10.1113/EP091058 Piasecki, M., Ireland, A., Piasecki, J., Degens, H., Stashuk, D. W., Swiecicka, A., Rutter, M. K., Jones, D. A., & McPhee, J. S. (2019). Long-term end- urance and power training may facilitate motor unit size expansion to compensate for declining motor unit numbers in older age. Frontiers in Physiology, 10, 449.
https://openalex.org/W4212996365	https://hal.inrae.fr/hal-03615662/file/coatings-12-00239.pdf	English	null	Waste Biopolymers for Eco-Friendly Agriculture and Safe Food Production	Coatings	2,022	cc-by	8,702	To cite this version: Elio Padoan, Enzo Montoneri, Giorgio Bordiglia, Valter Boero, Marco Ginepro, et al.. Waste Biopolymers for Eco-Friendly Agriculture and Safe Food Production. Coatings, 2022, 12 (2), pp.239. 10.3390/coatings12020239. hal-03615662 Distributed under a Creative Commons Attribution 4.0 International License Article Elio Padoan 1 , Enzo Montoneri 2,, Giorgio Bordiglia 1, Valter Boero 1, Marco Ginepro 3 , Philippe Evon 4, , Carlos Vaca-Garcia 4 , Giancarlo Fascella 5 and Michéle Negre 1 1 Dipartimento di Scienze Agrarie, Forestali e Alimentari, Università di Torino, Largo P. Braccini 2, 10095 Grugliasco, Italy; elio.padoan@unito.it (E.P.); giorgio.bordiglia@unito.it (G.B.); valter.boero@unito.it (V.B.); negre.michele@gmail.com (M.N.) 1 Dipartimento di Scienze Agrarie, Forestali e Alimentari, Università di Torino, Largo P. Braccini 2, 10095 Grugliasco, Italy; elio.padoan@unito.it (E.P.); giorgio.bordiglia@unito.it (G.B.); lt b @ it it (VB ) i h l @ il (M N ) 2 Dipartimento di Scienze delle Produzioni Agrarie e Alimentari, Università di Catania, Via S. Soﬁa 98, 95123 Catania, Italy y 3 Dipartimento di Chimica, Università di Torino, Via P. Giuria 7, 10125 Torino, Italy; marco.ginepro@unito.it 4 Dipartimento di Chimica, Università di Torino, Via P. Giuria 7, 10125 Torino, Italy; marco.ginepro@unito.it 4 Laboratoire de Chimie Agro-Industrielle (LCA), Université de Toulouse, ENSIACET, INRAE, Toulouse INP, 4 Allée Emile Monso, 31030 Toulouse, France; carlos.vacagarcia@toulouse-inp.fr , , ; g p 5 CREA Research Centre for Plant Protection and Certiﬁcation, 90011 Bagheria, Italy; giancarlo.fascella@crea.gov.it g g * Correspondence: enzo.montoneri@gmail.com (E.M.); philippe.evon@toulouse-inp.fr (P.E.) Abstract: This work addresses environmental problems connected with biowaste management, the chemical industry, and agriculture. These sectors of human activity cause greenhouse gas (GHG) emissions in the air, climate change, leaching of excess mineral fertilizers applied to soil into ground water, and eutrophication. To mitigate this problem in agriculture, controlled release fertilizers (CRFs) are made by coating mineral fertilizers granules with synthetic polymers produced from the fossil-based chemical industry. This strategy aggravates GHG emission. In the present work, six formulations containing sunﬂower protein concentrate (SPC) and a new biopolymer (BP) obtained from sunﬂower oil cake and by hydrolysis of municipal biowaste, respectively, and commercial urea were tested as CRFs for spinach cultivation against the control growing substrate Evergreen TS and commercial Osmocote®. The results show large differences in plants’ nitrate concentration due to the different treatments, although the same nitrogen amount is added to the substrate in all trials. BP is the key component mitigating nitrate accumulation in plants. HAL Id: hal-03615662 https://hal.inrae.fr/hal-03615662v1 Submitted on 21 Mar 2022 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 coatings Citation: Padoan, E.; Montoneri, E.; Bordiglia, G.; Boero, V.; Ginepro, M.; Evon, P.; Vaca-Garcia, C.; Fascella, G.; Negre, M. Waste Biopolymers for Eco-Friendly Agriculture and Safe Food Production. Coatings 2022, 12, 239. https://doi.org/10.3390/ coatings12020239 Keywords: biopolymers; municipal bio-waste; spinach; agriculture; nitrates Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Article The plants grown in the substrates containing BP together with SPC and/or urea, although exhibiting relatively high total N uptake (47–52 g kg−1), have signiﬁcantly lower nitric to total N ratio (9.6–12.0) than that (15.3–16.5) shown by the plants grown in the substrates containing SPC and/or urea, but no BP. The data conﬁrm that all composites containing BP yield the safest crop coupled with high biomass production. Replication of BP effects for the cultivation of different plants will contribute to the development of a biobased chemical industry exploiting biowastes as feedstock. Citation: Padoan, E.; Montoneri, E.; Bordiglia, G.; Boero, V.; Ginepro, M.; Evon, P.; Vaca-Garcia, C.; Fascella, G.; Negre, M. Waste Biopolymers for Eco-Friendly Agriculture and Safe Food Production. Coatings 2022, 12, 239. https://doi.org/10.3390/ coatings12020239 1. Introduction The augmentation of human population, along with its concentration in cities and increasing consumption habits, is causing several problems connected with biowaste management, the chemical industry, and agriculture. For example, municipal biowaste (MBW) production in Europe is 100 Mt yr−1 [1], with more than half still landﬁlled releasing 25,000 Mm3 yr−1 GHG [2]. To satisfy the food demand, crop production is boosted by applying mineral fertilizer doses higher than those adsorbed by soil and plants. Excess nutrients accumulate in soil, leach into ground water, and cause eutrophication [3]. In this context, for a long time the Universities of Torino and Toulouse focused their research on the valorization of biowaste from urban (MBW) and agro-industrial sources as feedstock 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/coatings Coatings 2022, 12, 239. https://doi.org/10.3390/coatings12020239 Coatings 2022, 12, 239 2 of 12 for the production of value added biobased chemical specialties and materials to use in place of commercial products obtained from fossil sources [4]. for the production of value added biobased chemical specialties and materials to use in place of commercial products obtained from fossil sources [4]. With speciﬁc reference to the agriculture sector, very recently the authors of the present work have reported the manufacturing and thermomechanical properties of a composite material made by twin-screw extrusion followed by injection-molding for use as controlled release fertilizer (CRF) [5]. This material contains urea and two polymers, herein after named sunﬂower proteins concentrate (SPC) and biopolymer (BP). These are obtained from sunﬂower oil cake and by hydrolysis of the anaerobic fermentation digestate of municipal unsorted food wastes, respectively. All three components contain nitrogen, have soil fertilizing power, have different solubility in water and, therefore, can release nitrogen in soil. Their fertilizing properties depend on the nitrogen release rate. In addition to the fertilizing property, SPC is well known for its good processability and is suitable to be used as a thermoplastic matrix for manufacturing the CRF composite pellets by twin-screw extrusion followed by injection-molding [5]. 1. Introduction To test and assess the performance of the SPC and BP biopolymers, the experimental plant reported in the present work included three key elements, i.e., urea as widely used fertilizer, spinach as a sensitive probe to measure SPC/BP effects, and Osmocote® as a reference of widely used commercial CRF, representing a typical material made by conventional coating technology. Urea is one of the most important N-fertilizers. Its world consumption is 51 Mt yr−1 [6]. Urea has environmental drawbacks deriving from the release of excess nitrogen over the plant uptake rate. Urea is highly soluble in water. In soil, it is hydrolyzed to ammonia [7,8] and then, transformed into nitrates. These are adsorbed by the plant roots and transferred to the leaves. Nitrates in leaves are reduced to ammonia and then converted to proteins. In the presence of excess ammonia, proteins’ production is slowed down. Nitrates accumulate in leaves and soil. They may impact the environment negatively, leaching from soil into ground water and causing eutrophication [9]. Nitrates in food plants may also have carcinogenic effects for humans [10]. In the human body, nitrate can be reduced to nitrite, which may cause methemoglobinemia, and the possibility of gastric cancer and other diseases [11,12]. Leaves of spinach are a typical food that can exhibit the nitrates’ accumulation ef- fects [13]. Spinach is a high value crop that requires sufﬁcient N fertilizer to ensure optimal growth and to meet high quality criteria [14]. It absorbs NO3−from the soil efﬁciently but is known to be relatively inefﬁcient in NO3−reduction [15]. The European Commission issued the regulation No. 1258/2011, stating the maximum acceptable concentration of nitrates in spinach [16]. These are 3.5 g kg−1 in fresh spinach, 2 g kg−1 in preserved, frozen spinach, and 0.2 g kg−1 in baby food. Spinach leaves should have high nutritional value from protein content and no toxicity from nitrates. p y Commercial Osmocote® is a typical material belonging to the family of CRFs con- taining urea granules coated with synthetic polymers [17,18]. These are used to mitigate nitrate accumulation and effects in the environment [19]. The drawbacks of these CRFs are those typical of the fossil-based chemical industry. To substitute synthetic organic materials derived from fossil sources with natural or biobased materials [20], other CRFs have been investigated, based on natural polymers derived from plants [21]. 1. Introduction No CRF is known to contain biopolymers derived from biowastes as SPC and BP. In the previous work [5], the mechanical properties’ advantages of the injection- molding technology, compared to conventional coating processes, were discussed. The inclusion of Osmocote® in the experimental plan of the present work has allowed the vis-à-vis direct comparison of the twin-screw extruded/injection-molded SPC composite and the coated Osmocote® composite for their performance as CRF for the cultivation of spinach. 3 of 12 Coatings 2022, 12, 239 2.2. Fabrication and Composition of SPC Composites The SPC composites were fabricated and characterized for their elemental composition (Table 1) in the previous work [5]. Granules containing urea and/or BP dispersed into the SPC matrix were made by twin-screw extrusion. These were converted into dense dark pellets, all having the same (10 mm × 10 mm × 4 mm) dimension, by injection-molding. The tested SPC composites had the following compositions: neat SPC, SPC-U containing 10% urea, SPC-BP containing 10% BP, and SPC-BP-U containing 5% BP and 5% urea. For the neat components and the composite materials, Table 1 reports the C and soil/plant N, P, and K nutrient contents. Table 1. Total C, N, P, and K concentrations (%, w w−1) in neat SPC, BP, urea (U), and in the SPC pellets’ samples 1. Table 1. Total C, N, P, and K concentrations (%, w w−1) in neat SPC, BP, urea (U), and in the SPC pellets’ samples 1. Formulation C N P2O5 K2O BP 39.6 6.6 1.1 5.5 U 20.0 46.6 - - SPC 38.7 ± 5.2 a 6.9 ± 0.9 a 2.5 ± 0.1 a 1.5 ± 0.1 a SPC-U 37.7 ± 0.2 a 10.5 ± 0.1 c 2.3 ± 0.1 a 1.3 ± 0.1 a SPC-BP 39.5 ± 2.6 a 7.0 ± 0.6 a 2.6 ± 0.3 a 1.8 ± 0.1 b SPC-BP-U 38.6 ± 0.4 a 8.5 ± 0.0 b 2.1 ± 0.1 a 1.5 ± 0.1 a 1 Values in the same column followed by different letters are signiﬁcantly different (p < 0.05). 2.3. Plant Growth Trials 2. Materials and Methods 2.1. Materials The components to fabricate the SPC composites were available from previous work [5]. SPC was obtained by sieving a sunﬂower oil cake (SOC) using a Ritec (Signes, France) 600 vibrating sieve shaker ﬁtted with a 1 mm grid. The undersize (SPC) was enriched with the smaller particles coming from the kernel of the seed, thus having a high protein content (51%, in proportion to the SPC dry weight). The rest (oversize) was constituted mainly by particles from the seed hull, and it was therefore rich in lignocellulosic ﬁbers and minerals. BP was obtained by hydrolysis at pH 13 and 60 ◦C of the anaerobic digestate of MBW [4]. The anaerobic digestate was collected from the MBW waste treatment plant owned by the Acea Pinerolese Industriale SpA (ACEA) company in Pinerolo (TO). This plant processes, by fermentation, urban food wastes from separate source collection to produce biogas and solid anaerobic digestate containing the recalcitrant lignocellulosic fraction of the pristine MBW. To prepare the BP, the ACEA anaerobic digestate was taken up in water to yield a slurry at 4 water/digestate (w/w) ratio. Potassium hydroxide was added to the slurry until it reached pH 13. The alkaline slurry was heated up to 60 ◦C for 4 h. Afterwards, the slurry was allowed to settle until the liquid hydrolysate separated from the insoluble phase. The liquid hydrolysate was ﬁltered through a 5 kDa polysulphone membrane to separate the retentate containing BP and the permeate containing the excess unreacted alkali reagent. The retentate was ﬁnally dried at 60 ◦C to yield the solid BP, which was used in the present work. Major components of BP were 45% protein, 13% lignin, and 15% minerals. Urea was a common commercial product. 2.3. Plant Growth Trials Spinacia oleracea L. “Gigante d’inverno” seedlings, purchased from Beltrame Roberto nursery, strada Sanda, Moncalieri, Italy, were transplanted on the commercial growing sub- strate Evergreen TS (Turco S.a.s, Moncalieri, Italy) in 2 L pots with 13 cm × 13 cm × 13 cm size (5 plants per pot, and 5 pots per trial). The plants were grown during November and December 2019 in an unheated greenhouse at an average temperature of 18 ◦C and 90% relative humidity under natural lighting conditions. Surface irrigation was performed manually every three days. After three days, the neat components and the composites were added to the substrate in the following amounts: 4 of 12 Coatings 2022, 12, 239 • Trial 0: control (growing substrate) containing 25 mg organic N. • Trial 1: growing substrate + SPC (4.1 g per pot, 283 mg N). • Trial 2: growing substrate + SPC-U (2.7 g per pot, 284 mg N). • Trial 3: growing substrate + SPC-BP (4.0 g per pot, 280 mg N). • Trial 4: growing substrate + SPC-BP-U (3.3 g per pot, 281 mg N). • Trial 5: growing substrate + urea (0.6 g per pot, 280 mg N). • Trial 6: growing substrate + BP (4.3 g per pot, 284 mg N). • Trial 7: growing substrate + urea (0.3 g per pot, 140 mg N) + BP (2.2 g per pot, 145 mg N). l ® ( ) g g g p p g • Trial 6: growing substrate + BP (4.3 g per pot, 284 mg N). g g g p p g • Trial 7: growing substrate + urea (0.3 g per pot, 140 mg N) + BP (2.2 g per pot 145 mg N). g g g p p g • Trial 7: growing substrate + urea (0.3 g per pot, 140 mg N) + BP (2.2 g per pot, 145 mg N). • Trial 8: Osmocote® (1.8 g per pot, 282 mg N). The trials’ plan included the control (trial 0) without added N fertilization, trials 1–7 in the presence of added SPC, BP, and urea products, and trial 8 in the presence of commercial slow-release N, P, K fertilizer Osmocote® [22], which was used as reference material. 2.4. Analyses Total N was determined by the ﬂash combustion method (i.e., “Dumas method”) with thermal conductivity detection using the UNICUBE elementar analyzer (Elementar, Langenselbold, Germany) according to the CSN EN 13654-2 standard method. Nitrate concentrations in leaves and roots at the end of the trial have been analyzed on the fresh material. For each plant, 100 mg of fresh tissue was ground in liquid nitrogen and sus- pended in 10 mL of deionized water. Suspensions were incubated for 1 h at 45 ◦C and then centrifuged at 5000 rpm for 15 min. The extract was ﬁltered, and the nitrate N concentration was determined spectrophotometrically by the Griess reaction [23]. The determination of chlorophyll a and b and of carotenoids was performed on each plant by extraction of 300 mg fresh foliar tissue ground in liquid nitrogen with 10 mL 96% (v/v) ethanol. The samples were kept in the dark for 2 days at 4 ◦C, and the extracts were ﬁltered and then analyzed by spectrophotometry using a Hitachi (Tokyo, Japan) U-2000 spectrophotometer. The ab- sorbance readings were performed at 665 nm for chlorophyll a, at 649 nm for chlorophyll b, and at 470 nm for total carotene. Chlorophyll a, b, and total carotenoid concentrations were calculated according to the literature [24]. 2.5. Statistical Treatment of Data The data were evaluated by one-way ANOVA (p < 0.05 or 0.01) followed by the Tukey’s test for multiple comparison procedures. 2.3. Plant Growth Trials The amount of SPC materials (trial 1–4) and of urea and BP (trial 5–7) added in each pot was calculated to provide, in each pot, nearly the same total N amount (280–285 mg N) as in trial 8. After 55 days, the shoots and roots were collected, and their respective fresh weights were measured. Two plants per pot were dried at 60 ◦C for 3 days for the determination of the dry weight of leaves and roots. 3. Results Fresh and dry weight of spinach leaves and roots (mean value ± standard deviation) 1 cultivated in trials 0–8. Table 2. Fresh and dry weight of spinach leaves and roots (mean value ± standard deviation) 1 cultivated in trials 0–8. Table 3. Chlorophyll and carotenoids content in spinach leaves expressed as mg kg−1 fresh weight (mean value ± standard deviation) 1 cultivated in trials 0–8. Trial Chlorophyll a Chlorophyll b Carotenoids 0 242 ± 49.0 a 123 ± 28.9 a 19 ± 0.9 a 1 330 ± 32.3 b 176 ± 11.0 b 8 ± 0.3 a 2 323 ± 46.6 ab 172 ± 15.0 b 8 ± 0.4 a 3 258 ± 41.0 ab 125 ± 26.9 a 16 ± 0.8 a 4 321 ± 29.4 ab 158 ± 11.0 ab 7 ± 0.3 a 5 341 ± 29.9 b 172 ± 34.1 ab 18 ± 0.6 a 6 241 ± 55.4 a 138 ± 23.3 ab 24 ± 0.7 a 7 309 ± 22.1 ab 158 ± 19.4 ab 11 ± 0.4 a 8 289 ± 42.2 ab 168 ± 34.5 ab 17 ± 0.6 a 1 Within columns, mean values followed by the same letter are not signiﬁcantly different (p < 0.01). Table 4 reports the total N concentration in the plants. The experimental data in Table 4 show that the nitrogen uptake per plant is much lower in roots than in leaves. This is expected since most of the nitrate absorbed by the roots is transported in the xylem to the leaves for further reduction to ammonia, which is then used to synthesize the leaf proteins. The results also point out some signiﬁcant differences between the trials. In trials 1–5, and 7 and 8, the total N concentration in both leaves and roots is higher than that for the control (trial 0). The highest total N concentration values are shown in the leaves of the plants grown on SPC-based fertilizers (trials 1–4) and urea (trial 5). g The data in Table 5 point out large differences in nitrate concentration among the trials, although the same nitrogen amount was added to the substrate in all trials. In all trials, the nitrate concentration in fresh spinach is well below the 3.5 g kg−1 safe limit recommended by the European Commission [18]. 3. Results For the spinach cultivated in the different trials, Table 2 reports the fresh and dry weights of leaves and roots. At the end of the experiment, the plants were healthy and reached the marketable size without apparent differences between the trials, except for the fresh weight of the leaves as higher values were recorded in trials 2 and 3, compared to trial 6. Overall, the experimental values suggest that the control substrate provides a sufﬁcient amount of nutrients to allow a regular and constant growth of the plants. Table 3 reports the chlorophyll a, chlorophyll b, and carotenoids concentrations in the fresh spinach leaves. It shows that the chlorophyll a content in the plants fertilized with SPC and urea (trials 1 and 5) is signiﬁcantly higher than that of the control (trial 0). The plants fertilized with SPC and SPC-U (trials 1 and 2) exhibit signiﬁcantly higher concentrations of chlorophyll b than the control (trial 0). The treatments appear not to signiﬁcantly affect the carotenoid concentration compared to the control (trial 0). Coatings 2022, 12, 239 5 of 12 Table 2. Fresh and dry weight of spinach leaves and roots (mean value ± standard deviation) 1 cultivated in trials 0–8. Trial Leaves Roots Fresh Weight (g) Dry Weight (g) Fresh Weight (g) Dry Weight (g) 0 18.7 ± 3.2 ab 1.6 ± 0.55 a 0.86 ± 0.29 a 0.11 ± 0.04 a 1 25.6 ± 5.3 ab 2.3 ± 0.56 a 1.16 ± 0.22 a 0.13 ± 0.02 a 2 27.3 ± 4.5 a 2.4 ± 0.39 a 1.08 ± 0.23 a 0.12 ± 0.03 a 3 27.7 ± 6.1 a 2.3 ± 0.61 a 0.99 ± 0.26 a 0.10 ± 0.03 a 4 23.5 ± 5.1 ab 2.2 ± 0.54 a 0.99 ± 0.18 a 0.11 ± 0.03 a 5 24.5 ± 5.0 ab 2.3 ± 0.60 a 0.98 ± 0.16 a 0.12 ± 0.03 a 6 18.1 ± 1.9 b 1.8 ± 0.20 a 0.89 ± 0.10 a 0.12 ± 0.02 a 7 23.2 ± 3.4 ab 2.3 ± 0.54 a 1.08 ± 0.14 a 0.13 ± 0.03 a 8 22.8 ± 2.7 ab 2.2 ± 0.27 a 1.04 ± 0.08 a 0.13 ± 0.01 a 1 Within columns, mean values followed by the same letter are not signiﬁcantly different (p < 0.01). Table 2. 3. Results The highest concentration is found in the spinach fertilized with urea (2816 mg kg−1 in trial 5) and in the spinach fertilized with the sunﬂower proteins-based fertilizer (1890–2559 mg kg−1 in trials 1, 2, and 4). The lowest nitrate concentration is found in the plants grown on the control substrate (101 mg kg−1 in trial 0) and in the plants fertilized with BP (247 mg kg−1 in trial 6). However, these plants exhibit a rather low total nitrogen concentration (Table 4), probably due to the low level of nitrogen mineralization in the substrate. This is conﬁrmed by the low nitric to total N ratio (1.0 and 2.2 in trial 0 and trial 6, respectively), which suggests that most of the absorbed inorganic nitrogen is promptly transformed into amino acids and proteinaceous matter. In all other cases, the nitric to total N ratio ranges from 9.6 to 16.5. Coatings 2022, 12, 239 6 of 12 Table 4. Total nitrogen concentration and N uptake for spinach plants (mean value ± standard deviation) 1 cultivated in trials 0–8. Trial Total N (g kg−1 Dry Matter) N Uptake (mg Plant−1) Leaves Roots Leaves Roots 0 34.0 ± 1.82 a 20.9 ± 2.15 a 54.4 ± 21.6 a 2.3 ± 1.01 a 1 51.9 ± 1.32 bcd 29.8 ± 1.72 bc 117.6 ± 32.0 b 3.7 ± 0.88 a 2 52.7 ± 1.31 bd 30.4 ± 2.09 bc 128.3 ± 23.6 bc 3.6 ± 1.09 a 3 48.5 ± 2.04 bce 28.2 ± 0.81 bcd 113.0 ± 34.3 ab 2.9 ± 1.01 a 4 52.1 ± 2.99 bcd 29.1 ± 1.77 bc 114.5 ± 34.5 b 3.3 ± 1.12 a 5 55.5 ± 1.71 d 30.7 ± 2.36 b 129.7 ± 37.4 bc 3.6 ± 1.20 a 6 34.3 ± 3.44 a 20.7 ± 3.44 a 63.1 ± 13.3 ab 2.6 ± 0.87 a 7 47.2 ± 2.55 ce 26.9 ± 2.14 bc 106.2 ± 31.1 ab 3.5 ± 0.96 a 8 45.1 ± 3.47 e 26.5 ± 2.11 c 99.8 ± 19.8 ab 3.3 ± 0.57 a 1 Within columns, mean values with the same letter are not signiﬁcantly different (p < 0.01). Table 5. Nitric nitrogen (mean value ± standard deviation) 1 in spinach leaves cultivated in trials 0–8. Table 5. Nitric nitrogen (mean value ± standard deviation) 1 in spinach leaves cultivated in trials 0–8. 3. Results Trial NO3−N (g kg−1 Dry Matter) NO3−/Total N (w w−1) NO3−N (mg kg−1 Fresh Matter) 0 1.1 ± 0.60 a 1.0 ± 0.58 a 101 ± 53.7 a 1 26.0 ± 4.9 bcd 15.3 ± 3.3 c 2281 ± 410 bcd 2 28.6 ± 2.8 bc 16.5 ± 2.0 c 2559 ± 297 bc 3 15.4 ± 7.5 e 9.6 ± 5.1 b 1373 ± 575 e 4 20.5 ± 7.0 bde 12.0 ± 4.8 bc 1890 ± 593 bde 5 29.8 ± 1.9 c 16.3 ± 1.5 c 2816 ± 125 c 6 2.5 ± 1.4 a 2.2 ± 1.5 a 247 ± 122 a 7 16.4 ± 5.1 de 10.5 ± 3.9 b 1541 ± 398 de 8 14.9 ± 4.0 e 10.1 ± 3.5 b 1438 ± 323 e 1 Within columns, mean values followed the same letter are not signiﬁcantly different (p < 0.01). 4. Discussion 4.1. Effect of BP on Plant Performances 4.1. Effect of BP on Plant Performances The data obtained from the present experiment show signiﬁcant differences only between trials 2, 3, and 6 for the leaf fresh biomass, but not the dry one (Table 2). The average dry weight of the leaves (2.2 g per plant) is consistent with literature data for cultivated spinach under different conditions: 0.6–2.0 g per plant for hydroponic cultures [15,25], 0.2–1.2 g per plant in pot experiments, depending on the amount of added nitrogen [26], and 4.3–5.4 g per plant in trials where different composts were used as fertilizers [27]. g p p p The chlorophyll content of spinach leaves was also affected by BP application (Table 3). The concentration values reported in Table 3 fall in the range of values reported by other workers for spinach cultivated under various operational conditions. For example, for chlorophyll a, the following values are reported: 300–400 mg kg−1 in spinach cultivated in ﬁeld experiments at different N and water applied levels [28], and 860 mg kg−1 in spinach cultivated under hydroponic conditions at 105 mg L−1 N applied doses [29]. The data for the group of trials 1–7, in which SPC, BP, and/or urea were used, show that trials 3 and 6 exhibit lower contents of chlorophyll and higher contents of carotenoids than the other 5 trials. The statistical analysis of these groups of data does not prove signiﬁcant differences between most of the trials. Yet, speciﬁcally for chlorophyll, the apparent lower chlorophyll concentration recorded in leaves from trials 3 and 6 may be correlated to the lower total N content and uptake (Table 4) measured in leaves. Indeed, it is well known that leaf chlorophyll concentration is linked to leaf N content due to the presence of N atoms in chlorophyll molecules [30]. Some authors observed that BP application to plants enhanced the availability of N requested for chlorophyll formation [31], with positive effects on photosynthetic activity. The apparent higher carotenoid content and lower chlorophyll content in trials 3 and 6 can be related to a trend that plants show in Coatings 2022, 12, 239 7 of 12 7 of 12 some growing conditions (particularly under stress), i.e., when chlorophyll values decrease, carotenoids tend to increase, and vice versa [32,33]. some growing conditions (particularly under stress), i.e., when chlorophyll values decrease, carotenoids tend to increase, and vice versa [32,33]. 4.1. Effect of BP on Plant Performances The BP application also inﬂuenced the total nitrogen content in spinach leaves and roots (Table 4). To be assimilated by plants, nitrogen must be in nitric or ammonia forms. Therefore, nitrogen present in organic forms in soil or substrate must be made available through mineralization by microorganisms. The prevailing form is the nitric one because ammonium ion is promptly oxidized and made less available by adsorption on the soil surfaces. Once absorbed by the roots, nitrates are transferred to the shoots where they are reduced to ammonium and used to ﬁrst synthesize glutamate and glutamine through the enzymatic nitroreductase and GOGAT systems, and then other amino acids and N- containing compounds. The total N concentration values are consistent with those reported in literature for spinach leaves (22.7–51.5 g kg−1) [15,27]. The low nitrogen concentration found in the plants grown on the BP-added substrate (trial 6) suggests that BP is recalcitrant to the biochemical attack by microorganisms, in accordance with the high lignin-like chemical moieties present in its macromolecular structure [4]. On the other hand, the low nitrogen uptake of the plants grown in trial 6 does not seem to negatively affect the leaf biomass yield as shown in Table 2. This is consistent with previous results reported by other authors, who used BPs in ﬂoriculture trials [30,31,34]. These authors agreed that the positive effects exhibited by the investigated biopolymers were due to the biopolymers’ chemical structure interacting with the microorganisms and stimulating the plant metabolism, more than to their fertilizing power stemming from their contribution of organic nitrogen as soil fertilizer. On the other hand, although a high N content in the leaves, as shown in Table 4, can be considered a positive result because it is correlated to a high protein concentration, it should not be accompanied with a high nitrate concentration. Accumulation of nitrates in plants generally occurs when the plant uptakes more NO3−than the NO3−assimilable in protein form [35]. Most absorbed nitrate is stored in the vacuole until release for reduction in the cytosol [36]. In the present work, this issue is addressed by the collected data shown in Table 5. Actually, BP application also affected nitric nitrogen content in spinach leaves (Table 5). Accumulation of nitrates in the vacuoles of the cells occurs when the enzymatic systems leading to the reduction of nitrates and further synthesis of protein matter are inhibited by excessive nitrate uptake. 4.1. Effect of BP on Plant Performances In this scenario, it is noteworthy that the plants grown in trials 3, 4, and 7 substrates, all containing BP together with SPC and/or urea, although exhibiting a high total N uptake (Table 4), have signiﬁcantly lower nitric to total N ratio (9.6–12.0) than that (15.3−16.5) shown in the other trials (1, 2, and 5) containing SPC and/or urea but no BP. The best fertilizers in terms of high total N content and low nitrates accumulation were the SPC-BP (trial 3) and SPC-BP-U (trial 4) composites, the mix of urea and BP in trial 7, and Osmocote® (trial 8). In all these trials, the nitrate concentration in the spinach leaves is even below the limit of 2 g kg−1 recommended for preserved frozen spinach by the European Commission [16]. The data conﬁrm that all composites containing BPs yield the safest crop coupled with high biomass production. 4.2. Plausible Explanation of the BP Effects The experimental data point out that BP can mitigate nitrates’ accumulation in spinach plants. The fertilizer used in trial 3 was SPC loaded with 10% BP. About 90% of the total nitrogen present in this specimen comes from SPC. Therefore, most of the N uptake of the plants in trial 3 is to be attributed to SPC nitrogen. Indeed, trial 6 has demonstrated that BP nitrogen is relatively less available for the plant to take up. In trial 7, in which urea was the nitrogen source and the same amount of BP was added, high nitrogen uptake is also accompanied by a relatively low nitrate content. The results suggest that BP, although not supplying to the plant as much nitrogen as SPC and urea, strongly affects the pathways responsible for the mineralization of organic nitrogen. A similar effect was observed in the previous work [5] reporting the kinetics of ammonia and organic N release rates of the SPC composites in water. BP was shown to retard the formation of ammonia from urea Coatings 2022, 12, 239 8 of 12 hydrolysis and enhance the release of organic nitrogen from SPC. In this case, the effect might be due to a plausible interaction of BP functional groups with urea and SPC. g p g p BP belongs to a family of biopolymers obtained by the same hydrolysis process from fermented lignocellulose biowastes of different sources [4]. All these biopolymers keep the memory of the sourcing lignocellulose materials, are constituted by a mix of macromolecules containing the same carbon types and various acid and basic functional groups capable of interacting with other molecules by protonation and donor-acceptor complexing reactions. The relative ratios of the chemical functionalities in these families of biopolymers depend on the sourcing materials. As a consequence, they have the same multiple properties and performances, although at different levels depending on the sourcing materials. Other BPs, obtained from composted green wastes or composted mixes of green wastes and MBW anaerobic digestates, have been investigated as auxiliaries in the anaerobic fermentation of MBW [37,38] carried out in bioreactors dedicated to the production of biogas. These biopolymers can reduce the ammonia and/or the nitrate level in the process digestate. Other workers have used BPs as animal diet supplements [39]. They have shown that BPs reduce the proteolysis occurring in the caecum intestine of pigs with consequent reduction of ammonia formations. 4.2. Plausible Explanation of the BP Effects Biagini and coworkers [40] tested the biopolymer obtained from composted mixes of green wastes and MBW anaerobic digestates as a supplement for rabbits’ protein diet. They demonstrated that the rabbits fed with the biopolymer-supplemented diet produced manure with signiﬁcantly lower ammonia and GHG emissions compared to the animals fed with the control diet. While the conﬁrmation and replicability of the effects of the above biopolymers under different operational conditions and of the BP used in the present work are unquestionable, the role of these biopolymers is not yet clear. So far, it cannot be established deﬁnitely whether the biopolymer effects involve pure chemical reactions or biochemical processes with participation of a microorganism. The second hypothesis is the most likely, according to Baglieri and coworkers [41]. These researchers cultivated bean plants using a biopolymer obtained from the hydrolysis of exhausted tomato plants as fertilizers. The biopolymer sourced from the agriculture biowastes bears strong chemical similarities with BP sourced from MBW. The former was found [41] to signiﬁcantly enhance nitrate reductase, glutamine synthetase, and glutamate synthase activities, and to increase soluble proteins’ concentra- tion in shoots and roots, compared to the control. Based on the lack of differences between the concentrations of mineral nitrogen in the control and treated cultivation substrate, as opposed to the signiﬁcant differences observed for enzymatic activity and soluble proteins’ concentration in the plants, Baglieri and coworkers [41] concluded that the biopolymer acts as plant biostimulant with a possible auxin-like effect, more than as soil fertilizer. On the other hand, a possible action of BP as chemical catalyst must be considered. This is also in view of previous work reporting the property of BPs to catalyze oxidation reactions in the absence of any microorganism [42]. According to the current view of the behavior of urea in soil and the fate of the produced ammonia [7,8,35] for the system investigated in the present work, the following reaction scheme may help clarify the BPs effects: CO(NH2)2 + H2O ⇆2 NH3 + CO2 (1) NH3 + 2 O2 ⇆HNO3 + H2O (2) R-CH(OH)-COOH + CH3-COOH + NH3 ⇆R-CH[NH-(C = O)-CH3]-COOH + 2 H2O (3) (2) (3) The scheme shows that urea is hydrolyzed to ammonia (reaction (1)) and then, trans- formed into nitrates (reaction (2) forward). These are adsorbed by the plant roots and transferred to the leaves. 4.2. Plausible Explanation of the BP Effects Nitrates in leaves are reduced to ammonia (reaction (2) backward) and then converted to proteins (reaction (3)). A recent work [38] investigated the BPs’ assisted anaerobic fermentation of MBW in 150 mL shake ﬂasks. It demonstrated that BPs, Coatings 2022, 12, 239 9 of 12 in the investigated MBW fermentation system comprising organic, nitrate and ammonia N, catalyze the following chemical redox reaction: in the investigated MBW fermentation system comprising organic, nitrate and ammonia N, catalyze the following chemical redox reaction: (4) HNO3 + NH3 ⇆N2 + 2 H2O + 1/2 O2 (4) HNO3 + NH3 ⇆N2 + 2 H2O + 1/2 O2 The calculated Gibbs free energy from literature data [43] for this reaction, i.e., −360 kJ/N2 mole, shows that reaction (4) is thermodynamically favored. Occurrence of the chemical catal- ysis by BP in the system investigated in the present work, and the consequent reduction of nitrates by reaction (4), may be a plausible explanation for the lower nitric to total N ratio reported in Table 4 for the composite materials containing BPs (trials 3, 4 and 7), compared to the other materials in trials 1, 2, and 5 that contain SPC and/or urea but no BP. 4.3. Perspectives for a New Biowaste-Based Chemical Industry The spinach case study reported in the present work shows that, for use in the agricul- ture sector, materials obtained from urban and agro-industrial biowastes are competitive with materials obtained from fossil sources. Particularly, the BP material obtained from MBW exhibits unique properties that allow modulating the N release rate and fate in soil and in the plant crop. These properties are very important to safeguard the quality of soil, water, and crops. More ambitiously, the present article has relevance also for the sectors of waste management, pollution, and the chemical industry. Montoneri [4] and Tabasso and coworkers [44] have reviewed the sustainability of the hydrolysis process to produce BPs, as well as the BPs’ multipurpose performance and related economic, environmental, and social beneﬁts for several sectors of the chemical industry and agriculture. The engineered composite materials tested in the present work disclose a further beneﬁt offered by the tested BP for developing safe agriculture and food. They prove a new property of BP capable of reducing fertilizer nitrate leaching through soil and eutrophication effects, and at the same time diminish nitrate accumulation in crops. They add further important incentives for valorizing MBW as feedstock for the production of BPs and their use at commercial scale. They also prospect the feasibility of substituting products from fossil sources with products from biowastes. Particularly, because of their origin and special properties, BPs have high potential for developing a sustainable, waste-based industry integrating chemical and biochemical processes. MBW treatment plants are the ideal settings to this end. At present, they are service providers for citizens, as they perform the collection, disposal, and recycling of urban biowastes. To reduce landﬁll disposal, the most advanced plants process MBW by anaer- obic fermentation yielding biogas and digestate, and by aerobic fermentation producing compost. The value of these products is not enough to cover the plant operational costs [4]. The missing revenue is covered by citizens’ taxes. No MBW plant applies chemical pro- cesses. Yet, MBWs are a potential source of valuable renewable organic C, which could be recycled in the form of valued-added, biobased products for consumer use. 6. Patent The French patent application submitted 16 October 2020 under number FR2010597 to Institut National de la Propriété Industrielle (INPI) in France and entitled “Produit pour l’agriculture, et procédé de préparation” results from both [5] and the work reported in the present manuscript. The inventors of this patent are Philippe Evon, Carlos Vaca-Garcia, Laurent Labonne, and Antoine Rouilly. The owners of this patent are Institut National Polytechnique de Toulouse (INPT) and Institut National pour l’Agriculture, l’Alimentation et l’Environnement (INRAE). Author Contributions: E.P., E.M., G.B., V.B., M.G., P.E., C.V.-G., G.F. and M.N. contributed equally to this research article. All authors have read and agreed to the published version of the manuscript. Funding: This research was supported partly by endowed funds of the authors’ institutions, and partly funded by the European Commission in order to support the implementation of the actions pursued in the LIFE16 ENV/IT/000179-LIFECAB and the LIFE19 ENV/IT/000004-LIFEEBP projects. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data is contained within the article. Data Availability Statement: Data is contained within the article. 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. 5. Conclusions It has been demonstrated that composite controlled release fertilizers made by twin- screw extrusion followed by injection-molding can achieve the same performance in terms of spinach growth, nitrogen uptake, and nitrate accumulation as the commercial Osmocote® controlled release fertilizer, which contains urea coated with synthetic polymers. This result is achieved thanks to the municipal biowaste derived biopolymer. This component can control the nitrogen release rate of urea and from the sunﬂower protein concentrate, and reduce nitrogen accumulation by the plant while maintaining the same biomass growth and nitrogen uptake of Osmocote®. This ﬁnding poses a worthwhile scope for testing the twin-screw extruded/injection-molded biopolymer composites in the cultivation of other plant species in order to contribute to the replacement of current commercial materials coated with synthetic polymers from fossil sources. However, the ambition of the authors of the present work is far beyond developing controlled release fertilizers. Coatings 2022, 12, 239 10 of 12 10 of 12 Previous work [4] has demonstrated that the hydrolysis of municipal biowaste from different sources is a cost-effective, feasible process that yields a family of biopolymers for sustainable use in different sectors of agriculture and the chemical industry. The demonstration of the BP biopolymer properties in the present work is a new ﬁnding. It widens the ﬁelds of application and the potential beneﬁts of BPs. It adds a further argument for developing a waste-based chemical industry, which exploits biowaste as feedstock for the production of value-added chemical specialties and materials rather than plants speciﬁcally cultivated for this purpose. This approach, extended to the exploitation of biowastes from urban, agriculture, and agro-industrial sources, would allow not only the replacement of chemicals and materials from fossil sources, but would also dismiss environmentally unfriendly waste disposal practices and keeping agriculture soil for food production rather than for the production of chemicals. g g 8. Beig, B.; Niazi, M.B.K.; Jahan, Z.; Kakar, S.J.; Shah, G.A.; Shahi, M.; Zia, M.; Haq, M.U.; Rashid, M.I. Biodegradable polymer coated granular urea slows down N release kinetics and improves spinach productivity. Polymers 2020, 12, 2623. [CrossRef] [PubMed] References 1. European Commission. Biodegradable Waste. Available online: https://ec.europa.eu/environment/topics/waste-and-recycling/ biodegradable-waste_en (accessed on 30 December 2021). 1. European Commission. Biodegradable Waste. Available online: https://ec.europa.eu/environment/topics/waste-and-recycling/ biodegradable-waste_en (accessed on 30 December 2021). g 2. Eurostat. Municipal Waste Statistics. Available online: https://ec.europa.eu/eurostat/statistics-explained/index.php?title= Municipal_waste_statistics (accessed on 30 December 2021). urostat. Municipal Waste Statistics. Available online 2. Eurostat. Municipal Waste Statistics. Available online: https Municipal_waste_statistics (accessed on 30 December 2021). p p Municipal_waste_statistics (accessed on 30 December 2021). 3. IFA. Fertilizers, for Productive and Sustainable Agriculture Systems. Available online: https://www 30 December 2021). 4. Montoneri, E. Municipal waste treatment, technological scale up and commercial exploitation: The case of bio-waste lignin to soluble lignin-like polymers. In Food Waste Reduction and Valorisation; Morone, P., Papendiek, F., Tartiu, V.E., Eds.; Springer: Berlin/Heidelberg, Germany, 2017; Chapter 6. 4. Montoneri, E. Municipal waste treatment, technological scale up and commercial exploitation: The case of bio-waste lignin to soluble lignin-like polymers. In Food Waste Reduction and Valorisation; Morone, P., Papendiek, F., Tartiu, V.E., Eds.; Springer: Berlin/Heidelberg, Germany, 2017; Chapter 6. g y p 5. Evon, P.; Labonne, L.; Padoan, E.; Vaca-Garcia, C.; Montoneri, E.; Boero, V.; Negre, M. A new composite biomaterial made from sunﬂower proteins, urea, and soluble polymers obtained from industrial and municipal biowastes to perform as slow release fertiliser. Coatings 2021, 11, 43. [CrossRef] g y p 5. Evon, P.; Labonne, L.; Padoan, E.; Vaca-Garcia, C.; Montoneri, E.; Boero, V.; Negre, M. A new composite biomaterial made from sunﬂower proteins, urea, and soluble polymers obtained from industrial and municipal biowastes to perform as slow release fertiliser. Coatings 2021, 11, 43. [CrossRef] g 6. IFA. Consumption Urea World. Available online: https://www.ifastat.org/databases/plant-nutrition (accessed on 30 December 2021). g 6. IFA. Consumption Urea World. Available online: https://www.ifastat.org/databases/plant-nutrition (accessed on 30 December 2021). ) 7. Sigurdarson, J.J.; Svane, S.; Karring, H. The molecular processes of urea hydrolysis in relation t agriculture. Agric. Rev. Environ. Sci. Biotechnol. 2018, 17, 241–258. [CrossRef] 8. Beig, B.; Niazi, M.B.K.; Jahan, Z.; Kakar, S.J.; Shah, G.A.; Shahi, M.; Zia, M.; Haq, M.U.; Rashid, M.I. Biodegradable polymer coated granular urea slows down N release kinetics and improves spinach productivity. Polymers 2020, 12, 2623. [CrossRef] [PubMed] 11 of 12 11 of 12 Coatings 2022, 12, 239 9. Leip, A.; Billen, G.; Garnier, J.; Grizzetti, B.; Lassaletta, L.; Reis, S.; Simpson, D.; Sutton, M.A.; Vries, W.; Weiss, F.; et al. References Naz, M.Y.; Sulaiman, S.A. Slow release coating remedy for nitrogen loss from conventional urea: A review. J. Control. Release 2016, 225, 109–120. [CrossRef] 20 European Commission Bio Based Products A ailable online http //ec europa eu/growth/sectors/biotechnolog /bio based 225, 109–120. [CrossRef] 20. European Commission. Bio-Based Products. Available online: http://ec.europa.eu/growth/sectors/biotechnology/bio-based- 20. European Commission. Bio-Based Products. Available online: http://ec.europa.eu/growth/sectors/biotechnology/bio-based- products_it (accessed on 30 December 2021). p 21. Khalid, N.N.A.; Ashaari, Z.; Mohd, A.H.; Mohamed, H.A.; Lee, S.H. Nitrogen deposition and release pattern of slow release fertiliser made from urea-impregnated oil palm frond and rubberwood chips. J. For. Res. 2019, 30, 208762094. ® 22. Osmocote®. Product Information. Available online: https://icl-sf.com/uploads/ITALY/Ita_PI/Ita_PI_OH/8756_PI%20 OsmocotePro_12-14M%20new.pdf (accessed on 30 December 2021). 23. Miranda, K.M.; Espey, M.G.; Wink, D.A. A rapid, simple spectrophotometric method for simultaneous detection of nitrate and nitrite. Nitric Oxide 2001, 5, 61–71. [CrossRef] 24. Wellburn, A.R.; Lichtenthaler, H. Formulae and program determine carotenoids and chlorophyll a and b of leaf extracts inferent solvents. Adv. Photosyn. Res. 1984, 2, 272–284. y K.; Tamaki, M. Effect of microbubbles in deep ﬂow hydroponic culture on spinach growth. J. Plant Nutr. 2017 sRef] 25. Ikeura, H.; Tsukad, K.; Tamaki, M. Effect of microbubbles in deep ﬂow hydroponic culture on spinac 40, 2358–2364. [CrossRef] 26. Liu, I.J.; Tong, Y.; Zhu, Y.; Ding, H.; Smith, F.A. Leaf chlorophyll readings as an indicator for spinach yield and nutritional quality with different nitrogen fertilizer applications. J. Plant Nutr. 2006, 29, 1207–1217. [CrossRef] g pp J 27. Ebid, E.A.; Ueno, H.; Ghoneim, A.; Asagi, N. Nitrogen uptake by radish, spinach and “chingensai” from composted tea leaves, coffee waste and kitchen garbage. Compost. Sci. Util. 2008, 16, 152–158. [CrossRef] 28. Zhang, J.; Liang, Z.; Jiao, D.; Tian, X.; Wang, C. Different water and nitrogen fertilizer rates effects on growth and development of spinach. Comm. Soil Sci. Plant Anal. 2018, 49, 1922–1933. [CrossRef] 29. Lefsrud, M.; Kopsell, D.; Sams, C.; Wills, J.; Both, A.J. Dry matter content and stability of carotenoids in kale and spinach during drying. Hort. Sci. 2008, 43, 1731–1736. [CrossRef] 30. Fascella, G.; Montoneri, E.; Francavilla, M. Biowaste versus fossil sourced auxiliaries for plant cultiv J. Clean. Prod. 2018, 185, 322–330. [CrossRef] 30. Fascella, G.; Montoneri, E.; Francavilla, M. Biowaste versus fossil sourced auxiliaries for plant cultivation: The lantana case study. J. Clean. Prod. 2018, 185, 322–330. References Impacts of European live-stock production: Nitrogen, sulphur, phosphorus and greenhouse gas emissions, land-use, water eutrophication and biodiversity. Environ. Res. Lett. 2015, 10, 115004. [CrossRef] .; Iqbal, M. Nitrate accumulation in plants, factors affecting the process, and human health implications. A in. Dev. 2007, 27, 45–77. [CrossRef] y 10. Anjana, A.; Umar, S.; Iqbal, M. Nitrate accumulation in plants, factors affecting the process, and hu review. Agron. Sustain. Dev. 2007, 27, 45–77. [CrossRef] g 11. Breimer, T. Environmental Factors and Cultural Measures Affecting the Nitrate Content in Spinachs. Doctoral Dissertation, Wageningen University & Research (WUR), Wageningen, The Netherlands, 1982. Available online: https://library.wur.nl/ WebQuery/wurpubs/fulltext/201755 (accessed on 30 December 2021). 12. Chen, B.; Wang, Z.; Li, S.; Wang, G.; Song, H.; Wang, X. Effects of nitrate supply on plant growth, nitrate accumulation, metabolic nitrate concentration and nitrate reductase activity in three leafy vegetables. Plant Sci. 2004, 167, 635–643. [CrossRef] 13. Shoji, S. Innovative use of controlled availability fertilizers with high performance for intensive agricu conservation. Sci. China Life Sci. 2005, 48, 912–920. f 14. Heinrich, A.; Smith, R.; Cahn, M. Nutrient and water use of fresh market spinach. Horttechnology 2013, 23, 325–333. [CrossRef] 14. Heinrich, A.; Smith, R.; Cahn, M. Nutrient and water use of fresh market spinach. Horttechnology 2013, 23, 325–333. [CrossRef] 15. Chan-Navarrete, R.; Kawai, A.; Dolstra, O.; Lammerts van Bueren, E.T.; van der Linden, C.G. Genetic diversity for nitrogen use efﬁciency in spinach (Spinacia oleracea L.) cultivars using the ingestad model on hydroponics. Euphytica 2014, 199, 155–166. [CrossRef] 16. European Commission. Regulation n◦1258/2011 of 2 December 2011 Amending Regulation (EC) n◦1881/2006 as Regards Maximum Levels for Nitrates in Foodstuffs. Available online: https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L: 2011:320:0015:0017:EN:PDF (accessed on 30 December 2021). 17. Trenkel, M.E. Slow- and Controlled-Release and Stabilized Fertilizers: An Option for Enhancing Nutrient Use Efﬁciency in Agriculture; International Fertilizer Industry Association (IFA): Paris, France, 2010. Available online: https://www.fertilizer.org/images/ Library_Downloads/2010_Trenkel_slow%20release%20book.pdf (accessed on 30 December 2021). 18. Incrocci, L.; Maggini, R.; Cei, T.; Carmassi, G.; Botrini, L.; Filippi, F.; Clemens, R.; Terrones, C.; Pardossi, A. Innovative controlled- release polyurethane-coated urea could reduce n leaching in tomato crop in comparison to conventional and stabilized fertilizers. Agronomy 2020, 10, 1827. [CrossRef] g y [ ] Naz, M.Y.; Sulaiman, S.A. Slow release coating remedy for nitrogen loss from conventional urea: A review. J 225, 109–120. [CrossRef] , S.A. Slow release coating remedy for nitrogen loss from conventional urea: A review. J. Control. Release 2016 Ref] 19. References [CrossRef] ll G h l d d h l k b h d l f O 31. Fascella, G.; Montoneri, E.; Rouphael, Y. Biowaste-derived humic-like substances improve growth an (Murraya paniculata L. Jacq.) plants in soilless potted culture. Resources 2021, 10, 80. [CrossRef] 32. Fascella, G.; Mammano, M.M.; D’Angiolillo, F.; Rouphael, Y. Effects of conifers wood biochar as substrate component on ornamental performance, photosynthetic activity and mineral composition of potted Rosa rugosa. J. Hort. Sci. Biotech. 2017, 93, 519–528. [CrossRef] 33. Fascella, G.; Mammano, M.; D’Angiolillo, F.; Pannico, A.; Rouphael, Y. Coniferous wood biochar as substrate component of two containerized Lavender species: Effects on morpho- physiological traits and nutrients partitioning. Sci. Hort. 2020, 267, 109356. [CrossRef] 33. Fascella, G.; Mammano, M.; D’Angiolillo, F.; Pannico, A.; Rouphael, Y. Coniferous wood biochar as substrate component of two containerized Lavender species: Effects on morpho- physiological traits and nutrients partitioning. Sci. Hort. 2020, 267, 109356. [CrossRef] 34. Massa, D.; Lenzi, A.; Montoneri, E.; Ginepro, M.; Prisa, D.; Burchi, G. Plant response to biowaste soluble hydrolysates in hibiscus grown under limiting nutrient availability. J. Plant Nutr. 2018, 41, 396–409. [CrossRef] 34. Massa, D.; Lenzi, A.; Montoneri, E.; Ginepro, M.; Prisa, D.; Burchi, G. Plant response to biowaste soluble hydrolysates in hibiscus grown under limiting nutrient availability. J. Plant Nutr. 2018, 41, 396–409. [CrossRef] 12 of 12 12 of 12 Coatings 2022, 12, 239 35. Gülser, F. Effects of ammonium sulphate and urea on NO3−and NO2−accumulation, nutrient contents and yield criteria in spinach. Sci. Hortic. 2005, 106, 330–340. [CrossRef] p 36. Stagnari, F.; Di Bitetto, V.; Pisante, M. Effects of N fertilizers and rates on yield, safety and nutrients in processing spinach genotypes. Sci. Hortic. 2007, 114, 225–233. [CrossRef] 37. Francavilla, M.; Beneduce, L.; Gatta, G.; Montoneri, E.; Monteleone, M.; Mainero, D. Biochemical and chemical technology for a virtuous bio-waste cycle to produce biogas without ammonia and speciality bio-based chemicals with reduced entrepreneurial risk. J. Chem. Technol. Biotechnol. 2016, 91, 2679–2687. [CrossRef] J 38. Photiou, P.; Kallis, M.; Samanides, C.; Vyrides, I.; Padoan, E.; Montoneri, E.; Koutinas, M. Integrated chemical biochemical technology to reduce ammonia emission from fermented municipal biowaste. ACS Sustain. Chem. Eng. 2021, 9, 8402–8413. [CrossRef] 39. Montoneri, C.; Montoneri, E.; Tomasso, L.; Piva, A. Compost derived substances decrease feed prote cecal fermentation. J. Agric. Sci. 2013, 13, 31–44. [CrossRef] g 40. Biagini, D.; Montoneri, E.; Rosato, R.; Lazzaroni, C.; Dinuccio, E. References Reducing ammonia and GHG emissions from rabbit rearing through a feed additive produced from green urban residues. Sustain. Prod. Consum. 2021, 27, 1–9. [CrossRef] 41. Baglieri, A.; Cadilia, V.; Mozzetti Monterumici, C.; Gennari, M.; Tabasso, S.; Montoneri, E.; Nardi, S.; Negre, M. Fertilization of bean plants with tomato plants hydrolysates. Effect on biomass production, chlorophyll content and N assimilation. Sci. Hortic. 2016, 176, 194–199. 42. Gomis, J.; Bianco Prevot, A.; Montoneri, E.; González, M.C.; Amat, A.M.; Mártire, D.O.; Arque, A.; Carlos, L. Waste sourced biobased substances for solar-driven wastewater remediation: Photodegradation of emerging pollutants. Chem. Eng. J. 2014, 235, 236–243. [CrossRef] 43. Rossini, F.D.; Wagman, D.D.; Evans, W.H.; Levine, S.; Jaffe, I. Circular of the Bureau of Standards n◦500: Thermodynamic Properties; Nat. Bureau of Standards, Circ.; U.S. Government, Printing Ofﬁce: Washingt g g 44. Tabasso, S.; Ginepro, M.; Tomasso, L.; Nisticò, R.; Francavilla, M. Integrated biochemical and chemical processing of municipal bio-waste to obtain bio based products for multiple uses. The case of soil remediation. J. Clean. Prod. 2020, 245, 119191. [CrossRef]
https://openalex.org/W3215560296	https://pure.tue.nl/ws/files/217800298/s13638_022_02168_6.pdf	English	null	ELIoT: Enhancing LiFi for a Next Generation Internet of Things	Research Square (Research Square)	2,021	cc-by	17,348	Document status and date: Published: 22/09/2022 Document status and date: Published: 22/09/2022 Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. p • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights C i ht d General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study Y t f th di t ib t th t i l it f fit ki ti it i l i • You may not further distribute the material or use it for any profit-making activity o • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: Document license: CC BY Document license: CC BY DOI: 10.1186/s13638-022-02168-6 Document status and date: Published: 22/09/2022 Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There ca important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or vis DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication DOI: 10.1186/s13638-022-02168-6 Download date: 24. Oct. 2024 ELIoT: enhancing LiFi for next-generation Internet of things Citation for published version (APA): Linnartz, J. P., Ribeiro Barbio Correa, C., Bitencourt Cunha, T., Tangdiongga, E., Koonen, A. M. J., Deng, X., Abbo, A. A., Polak, P., Müller, M., Behnke, D., Vicent Colonques, S., Metin, T., Emmelmann, M., Kouhini, S. M., Bober, K. L., Kottke, C., & Jugnickel, V. (2022). ELIoT: enhancing LiFi for next-generation Internet of things. EURASIP Journal on Wireless Communications and Networking, 2022(1), Article 89. https://doi.org/10.1186/s13638-022-02168-6 Citation for published version (APA): Linnartz, J. P., Ribeiro Barbio Correa, C., Bitencourt Cunha, T., Tangdiongga, E., Koonen, A. M. J., Deng, X., Abbo, A. A., Polak, P., Müller, M., Behnke, D., Vicent Colonques, S., Metin, T., Emmelmann, M., Kouhini, S. M., Bober, K. L., Kottke, C., & Jugnickel, V. (2022). ELIoT: enhancing LiFi for next-generation Internet of things. EURASIP Journal on Wireless Communications and Networking, 2022(1), Article 89. https://doi.org/10.1186/s13638-022-02168-6 © 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 mate- rial. 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/. Abstract Communication for the Internet of things (IoT) currently is predominantly narrowband and cannot always guarantee low latency and high reliability. Future IoT applications such as flexible manufacturing, augmented reality and self-driving vehicles rely on sophisticated real-time processing in the cloud to which mobile IoT devices are con- nected. High-capacity links that meet the requirements of the upcoming 6G systems cannot easily be provided by the current radio-based communication infrastructure. Light communication, which is also denoted as LiFi, offers huge amounts of spectrum, extra security and low-latency transmission free of interference even in dense reuse settings. We present the current state-of-the-art of LiFi systems and introduce new fea- tures needed for future IoT applications. We discuss results from a distributed multiple- input multiple-output topology with a fronthaul using plastic optical fibre. We evaluate seamless mobility between the light access points and also handovers to 5G, besides low-power transmission and integrated positioning. Future LiFi development, imple- mentation and efforts towards standardization are addressed in the EU ELIoT project which is presented here. 1 TU Eindhoven, 5612AZ Eindhoven, The Netherlands 2 Signify, 5656AE Eindhoven, The Netherlands 3 Weidmüller Group, 32758 Detmold, Germany 4 MaxLinear, 46980 Paterna, Valencia, Spain 5 Fraunhofer FOKUS, 10589 Berlin, Germany 6 Fraunhofer HHI, 10587 Berlin, Germany Keywords: Future IoT, LiFi, Optical wireless communication, Light communication, IEEE 802.11bb, ITU-T G.vlc ELIoT: enhancing LiFi for next‑generation Internet of things J. P. M. G. Linnartz1,2, C. R. B. Corrêa1* , T. E. B. Cunha1, E. Tangdiongga1, T. Koonen1, X. Deng1, M. Wendt2, A. A. Abbo2, P. J. Stobbelaar2, P. Polak2, M. Müller3, D. Behnke3, M. Martínez4, S. Vicent4, T. Metin5, M. Emmelmann5, S. M. Kouhini6, K. L. Bober6, C. Kottke6 and V. Jungnickel6 Correspondence: c.ribeiro.barbio.correa@tue.nl 1 TU Eindhoven, 5612AZ Eindhoven, The Netherlands 2 Signify, 5656AE Eindhoven, The Netherlands 3 Weidmüller Group, 32758 Detmold, Germany 4 MaxLinear, 46980 Paterna, Valencia, Spain 5 Fraunhofer FOKUS, 10589 Berlin, Germany 6 Fraunhofer HHI, 10587 Berlin, Germany Correspondence: c.ribeiro.barbio.correa@tue.nl www.tue.nl/taverne Take down policy If you believe that this document breaches copyright please contact us at: openaccess@tue.nl providing details and we will investigate your claim. Download date: 24. Oct. 2024 Linnartz et al. J Wireless Com Network (2022) 2022:89 https://doi.org/10.1186/s13638-022-02168-6 EURASIP Journal on Wireless Communications and Networking Open Access *Correspondence: c.ribeiro.barbio.correa@tue.nl 1 TU Eindhoven, 5612AZ Eindhoven, The Netherlands 2 Signify, 5656AE Eindhoven, The Netherlands 3 Weidmüller Group, 32758 Detmold, Germany 4 MaxLinear, 46980 Paterna, Valencia, Spain 5 Fraunhofer FOKUS, 10589 Berlin, Germany 6 Fraunhofer HHI, 10587 Berlin, Germany 1 Introductionh The amount of wireless data traffic and the number of devices continues to grow at an exponential rate. This puts high pressure on the radio spectrum. Over the past decades, we have seen waves of innovation to enhance the bit rates (bit/s) and the density (bit/s/ m2) that can be provided by wireless radio networks, but the wireless technology needed to support this also becomes increasingly complex. For instance, massive multiple-input and multiple-output (MIMO) and beam-steering in 5G push the radio technology fron- tiers but may run into limits of complexity and power consumption also for signal pro- cessing and for conversion between analogue and digital domains. Reaching limits with RF motivates the industry into exploring new directions including optical wireless communications, which is also denoted as LiFi [1]. For light waves, walls, ceiling and floor are natural boundaries between the wireless cells that Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 2 of 33 allow very dense reuse of a vast amount of optical spectrum. Light sources such as light emitting diodes (LEDs) can offer gigabits per second transmission with simple emitters and receivers, with the potential for a very low cost. The concept of communication via light is older than via radio. However, when local area networking (LAN) went wireless in the 1990s, the demand for achieving cover- age across multiple rooms was larger than the desire to very densely reuse the radio spectrum, as at that time, not many devices were using it. That favoured radio solu- tions. Meanwhile, since the 1990s, Wi-Fi and cellular technologies became ubiqui- tous. The majority of the increased capacity is due to steadily reduced cell sizes as the need to serve many more users in dense areas grew. Extrapolating these trends explains the increasing interest in LiFi, which can cover small “personal” cells with very high data rates. Moreover, such a dedicated beam experiences and causes little interference from other users, thus can guarantee undisturbed, low-latency traffic to its destined user. Particularly, the Internet of things (IoT) can be seen as a driving force behind fur- ther densification. The IoT is often characterized by a vast multitude of many devices that each generate only limited traffic, but collectively cause a substantial increase in traffic. However, as Fig. 1 also illustrates, we foresee numerous future IoT applications that demand higher rates, lower latency, and increased link reliability. 2 Method This paper addresses results achieved in the ELIoT research to create technical improve- ments to make LiFi more mature, in particular via MIMO, multicasting, LiFi-5G inte- gration, POF front-hauls, and positioning technology, which are to be contributed to standardization. We followed the method of initially identifying use cases and their specific demands for improvements. Secondly, on initial visible light communication (VLC) test networks at our partners’ we landscaped the state-of-the-art using available advanced pro- totypes. In this way, the shortcomings of current solutions have been identified and based on this input, ELIoT conceived, developed, implemented and tested identified potentially innovative ceiling architectures, components, protocols and algorithms, which were vali- dated independently and in parallel by multiple partners. Results from building experimen- tal set-ups were captured in new models and translated into transferable building blocks. Yet, this also required us to develop an overarching reference architectural view. These have been reviewed in the following stage for consistency and compatibility with standardized protocols and algorithms. This paper follows an outline that starts with this architectural view in Section III. Specific results on security and mobility support can be seen as more or less independent protocol functionalities and are presented and discussed in Sections IV and V, respectively. Further architectural results are presented and discussed in Section VI for the cell layout, Section VII for the fronthaul, and Section VIII for the MIMO PHY. The step to ensure valorization via standardization is discussed in Section IX. A further enhancement by adding position- ing, beyond the primary communication functionality is described in Section X. 1 Introductionh Examples are in factories with Industry 4.0 machines, industrial devices or smart glasses [2, 3]. In par- ticular, we see an increasing need for low-latency, guaranteed bandwidth combined with positioning not only in augmented reality but also in real-time control of indus- trial processes. LiFi is a promising approach to address these future needs, possibly combining the wider through-the-wall coverage of RF networks with high-density very small cell high quality of service (QoS) LiFi, as seamless handovers and a com- mon security approach appear to be feasible, as work in this paper demonstrates. For instance, RF communication may be well suited for many devices each with a modest bit rate, while in addition LiFi can fill in QoS demanding hotspots. i The first LiFi systems are now deployed commercially, and further innovation and new features are needed to exploit their full potential in an increasing number of use cases and applications. This paper highlights how these challenges are addressed in the EU H2020 project ELIoT (Enhance Lighting for the Internet of things). As this is an overview paper, extending our EuCNC paper [4] we refer the reader to multiple Fig. 1 Some IoT use cases targeted by the ELIoT project Fig. 1 Some IoT use cases targeted by the ELIoT project Fig. 1 Some IoT use cases targeted by the ELIoT project Fig. 1 Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 3 of 33 publicly available project papers for more details. In particular, the main contribu- tions reported are: publicly available project papers for more details. In particular, the main contribu- tions reported are: • Providing and testing interworking with other systems such as the radio and wired infrastructure, in particular with 3GPP and 5G. • Providing a viable approach to integrating positioning within the communication stand- ard • Allowing mobility to other LiFi access points via MIMO link adaptation (horizon- tal mobility) and to other technologies, in particular with Wi-Fi and with 5G (vertical mobility) while protecting against outages from light beam blockage. • Proposals for introducing MIMO and experimental performance verification by related G.hn home networking standards. • Developing cost-effective and easy-to-install in-building backbone infrastructure net- works, for instance, plastic optical fibre (POF). i • Demonstrating end-to-end security concepts and comparing 5G and IEEE 802.1 × approaches to security. 3 LiFi concept There are many interesting applications for LiFi in different environments, such as office, industrial, in-home or outdoor [5]. Each of these poses different requirements, so a rigid single-use system concept may not be appropriate. However, cost and scalability Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 4 of 33 considerations dictate that solutions for the various use cases can be served flexibly by functional components, exploiting commonalities and reusability of hardware and soft- ware. In fact, network integration of LiFi as a layer 2 LAN, equivalent to a classical Ether- net connection, is needed and provides good versatility by supporting various protocols such as the internet protocol (IP) or industrial automation protocols, see Fig. 2. A com- mon physical layer (PHY) and medium access (MAC) convergence format are important for harmonization. It addresses the key commonality of use cases and solutions. The most widely used LiFi PHY makes use of direct current (DC)-biased orthogonal frequency division multiplexing (OFDM) with adaptive bit loading to allow scalabil- ity and exploit the full capacity of LiFi. Distributed MIMO (D-MIMO) is an attractive extension as it supports spatial multiplexing and diversity [6, 7]. In fact, for industrial- grade QoS, MIMO appears key to avoiding link outages if a line of sight (LoS) is blocked [8]. As further elaborated in Section VI.B, D-MIMO using spatially separated optical front ends (OFEs) can avoid frequent handovers associated with the small optical cell size, thus ensuring consistent QoS and reliability for high mobility and high user densities. The MIMO PHY preferably is based on subcarrier-wise channel estimation. It can be based on the feedback of the channel state information (CSI) but can also exploit generic low-pass properties of LEDs [9]. To support battery-constrained devices, a low-power PHY can be very useful. Pulse amplitude modulation (PAM) at a low rate, e.g. below 3 MHz bandwidth, proved to be effective in the uplink or in a VLC downlink. OFDM signals in higher parts of the modulation spectrum, e.g. beyond 5 MHz, can coexist with such PAM signals. To ensure QoS with guaranteed throughput and latency, the channel access mecha- nism is reservation-based, using spatial time-division multiple access (spatial TDMA), similar to Space–Time Reservation Multiple Access [10]. Power saving through sched- uled sleep periods yields longer battery lifetimes. Fig. 3 LiFi concept 2 Indoor system architecture overview comprising LiFi and Wi-Fi: each LiFi access point (AP) has a LiFi central unit (CU) that performs the baseband (BB) PHY and MAC layers and connects to multiple distributed units (DUs) which are optical front ends (OFEs) in the ceiling. The user terminal is also referred to as mobile unit (MU) Fig. 2 Indoor system architecture overview comprising LiFi and Wi-Fi: each LiFi access point (AP) has a LiFi central unit (CU) that performs the baseband (BB) PHY and MAC layers and connects to multiple distributed units (DUs) which are optical front ends (OFEs) in the ceiling. The user terminal is also referred to as mobile unit (MU) Linnartz et al. J Wireless Com Network (2022) 2022:89 Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 5 of 33 The backhaul connects the LiFi infrastructure to a fixed access network with a com- mon interface to the higher layer services. This backhaul network may be realized over different media, such as Ethernet, power line communications (PLC) and POF, each hav- ing its own capacity and latency characteristics. Whether or not a separate power infra- structure is needed for active OFEs can also be a key design consideration if the cost of the backbone is critical. A transparent (“analogue, jitter free”) fronthaul network allows versatile integration and upgradability. The identified main functional components of LiFi connectivity can be combined in order to build practical solutions for different use cases. Extended functionality for specific solutions can be realized through software modules in the CU (only). Further advanced functions can be link monitoring to facil- itate vertical handovers, remote configuration based on standard protocols, as well as QoS management and service metering. Extending Fig. 2, a LiFi system concept for an industrial scenario has been investigated in ELIoT [3, 11], and the concept may also be used in other situations. The concept is based on large-scale distributed MIMO to cope with the line-of-sight characteristics of light and the required QoS. A key aspect of our approach is to scale up the number of OFEs that are controlled by a single AP to cover larger areas. This AP can execute the PHY and MAC processing in a synchronous manner for all OFEs. As a result, moving users stay connected as the AP dynamically selects the appropriate set of OFEs. 3 LiFi concept This enables virtually seamless connectivity without the need for handover protocol proce- dures. Moreover, centralized signal processing for the distributed OFEs facilitates the use of synchronous MIMO schemes to increase link robustness and throughput further. Figure 3 applies this to an industrial context. Depending on the context, such APs are in the literature and in standardization referred to as central units (CUs) and distributed ceiling nodes, possibly luminaires, that are equipped with OFEs are denoted as distributed units (DUs), see Fig. 3. Each DU can reach terminals or end points like moving users in a certain area with its light cone. The cones of neighbouring DUs can overlap to provide homogenous coverage with adequate spatial diversity opportunities. There are multiple ways to split functionality between the CU and DUs [12, 13], but for indoor LiFi, we see an attractive approach in creat- ing the waveforms in the CU and feed these over a transparent linear channel without digitization. A transparent and synchronous fronthaul network connects all DUs with their common CU. In fact, DUs are understood as the optical antennas of the CU, which Fig. 3 Distributed multi-user MIMO system architecture, depicting a CU connected with DUs via a fronthaul network. Mobile and stationary devices communicate via the LiFi infrastructure at the same time. Several key aspects are indicated Fig. 3 Distributed multi-user MIMO system architecture, depicting a CU connected with DUs via a fronthaul network. Mobile and stationary devices communicate via the LiFi infrastructure at the same time. Several key aspects are indicated Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 6 of 33 receive analogue waveforms ready for transmission to and from the MUs. In the CU, the PHY defines the transmitted waveform, including error coding and modulation, and aids the MAC through measuring the channel. As we will elaborate in Section V, because the CU controls all MIMO processing for its DUs and MUs in the whole coverage area, no exchange of PHY-layer information between multiple APs (CUs) is necessary. Mobil- ity is adaptively supported by adapting the signals to DUs for each MU based on the channel state. This can be done in various ways, such as by approaches that are similar to antenna combining or selection in radio systems, or power loading based on a total power constraint or per-DU power constraint [14]. 3 LiFi concept IEEE 802.15.13 already makes use of different PHYs for downlink and uplink transmissions which may be attractive for other standards, for instance in ITU G.9991, also known as G.vlc [15]. Both are chosen such that they support the different requirements for downlink and uplink in an optimal way. Because MUs may be battery-powered, the uplink PHY should be more energy-efficient and able to operate at a low signal-to-noise ratio (SNR) [16]. To select and track the best DUs based on the MU’s mobility, an IEEE 802.15.13 sched- uler in the CU considers the latest channel state. Moreover, the modulation and coding scheme (MCS) is selected carefully to optimize the rate while frame losses and increased latencies through retransmission are avoided. To obtain the necessary information for this scheduling, the CU assesses the channel between all its DUs and the MUs periodi- cally. Moreover, features such as multi-user access and conflict-free scheduling are sup- ported [11, 17]. Accurate positioning is considered an enabling feature for wireless communication in factories, e.g. to locate automated guided vehicles. For integrated positioning, the high- speed OFDM PHY can perform sub-sample accurate timing measurements based on a conventional ranging (also denoted as timing advance) aided by additional phase estima- tion in the frequency domain. If this approach is combined with distributed MIMO, the MAC can triangulate the position of the terminal and reach centimetre precision [18]. A detailed description of our approach can be found in Section X. 4 Indoor positioning Besides communication, smart manufacturing calls for positioning to facilitate various new Industry 4.0 applications. For example, reliable indoor positioning is necessary for using intelligent transport systems (ITS) that transport parts on pallets from predefined locations to other predefined locations. It is usual that transport systems in factories and warehouses travel along predefined paths, defined by inductive loops or optical mark- ers on the floor. But these systems are not flexible, and modifications need effort. LiFi enables localization beyond predefined paths and allows extended positioning use cases due to more degrees of freedom. For example, a transport system can determine a new path on-demand to drive around an obstacle that occurred on the planned path. In addi- tion, there are new opportunities when production resources can be located on-demand. For example, its position can be displayed on a digital factory map. This makes it pos- sible to find a tool, a container with raw material or (semi-) finished products needed in a few seconds, which supports the work of machine operators and production plan- ners. In addition, real-time positioning of mobile devices can support technical main- tenance staff or production managers. For example, tablets can support them by using Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 7 of 33 the position information to display position-based information such as dashboards with machine data of the nearest machine(s). In the ELIoT project, we propose to realize positioning based on a multi-lateration algorithm, which measures the wireless propagation times between a mobile device and multiple LiFi APs at the ceiling [19, 20]. The ranging or timing advance algorithms for positioning are well understood and used in fixed and mobile access systems [18]. The accuracy of the result depends primarily on the quality of estimating the time-of-flight. To address the inherent synchronization challenges between mobile and ceiling units, an active ping-pong protocol is used to measure the round-trip time. This allows, in com- bination with standard techniques for clock offset estimation, an accurate estimation of the time-of-flight. The overall principle of LiFi-based positioning is shown in Fig. 4 [18]. Figure 4a shows the ranging by multiple transmitter units, Fig. 4b the signal structure based on the G.hn standard and Fig. 4c the ping-pong protocol for the estimation of the round-trip time. To confirm this concept, a series of simulations and measurements have been carried out in ELIoT. 4 Indoor positioning First results [18] showed the feasibility of the concept through simulations in a 3D environment as well as by ranging experiments of LiFi point-to- point links. Further investigation addressing 3D positioning scenario is presented in this section. The set-up is shown in Fig. 5 and consists of 4 Tx units at fixed positions and a single Rx unit. The signal progressing, as shown in Fig. 4, is performed in MATLAB and the signal conversion by DACs and ADCs. For the measurements, the Rx units were at multiple positions and the performance at each of them was evaluated. Figure 6 shows the estimated (blue circle) versus the actual positions of the receiver for 40 measurements (iterations) at each of the 12 positions. The transmitter positions are indicated by the triangles. The difference between estimated and real receiver posi- tions is overall very small, with a higher deviation towards the edges of the room. Fig- ure 7a shows the resulting average mean-square errors (MSE), for each the x-,y- and z-axis and for each Rx location taking 40 independent measurements into account. The x-axis shows generally higher MSEs and the errors of the z-axis are the smallest, with one exception. The smaller error for the z-axis can be attributed to the alignment of Fig. 4 a + c Principle of LiFi-based positioning, b signal structure Fig. 4 a + c Principle of LiFi-based positioning, b signal structure Page 8 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 Fig. 5 Measurement set-up for 3D evaluation Fig. 6 Estimated (denoted by blue circles) versus real position (denoted by red stars) for multiple receiver positions. Transmitter positions are denoted by triangles Fig. 5 Measurement set-up for 3D evaluation Fig. 5 Measurement set-up for 3D evaluation Fig. 6 Estimated (denoted by blue circles) versus real position (denoted by red stars) for multiple receiver positions. Transmitter positions are denoted by triangles Fig. 6 Estimated (denoted by blue circles) versus real position (denoted by red stars) for multiple receiver positions. Transmitter positions are denoted by triangles the set-up. In Fig. 7b, the combined x, y, z error is shown for three Rx positions over 40 measurements. There are two observations: First, there is a different offset error for each position, with Rx(1,0.72,0) the highest and Rx(0.67,0.725,0) the lowest. 4.1 Possible roadmap for adding positioning to G.vlc Enabling real-time positioning requires integration into a LiFi chipset. Today’s chipsets only partly support the proposed concept. The ELIoT project is working with a chipset from MaxLinear that implements the ITU-T G.9991 recommendation [21]. With this chipset, a coarse round-trip time estimation can already be achieved by using the time tagging capabilities. More precise timing information can be extracted from the channel frequency response (CFR) as demonstrated by the ELIoT experiments. For a next-gener- ation chipset, new APIs have to be developed in the PHY to provide information on the SNR and CFR to the higher layers. The high-level architecture of the digital baseband G.9991 chipset is shown in [21] (page 9). The baseband chip decodes the frames coming from the channel and injects frames into the channel through an analogue front-end chip that performs the signal adaptation to the medium. The positioning techniques explained here and in [18] would run on application programming interfaces (API) and can already be partially imple- mented using commercially available chipsets. The chipset can make use of some of the functionalities and framing of the standard that allow refining the procedure explained in [21]. 4 Indoor positioning The offset error can be attributed to non-ideal calibration, which was only performed in 1D for one location only. Note that the offset error is smaller in the centre position between the four transmitters (blue curve) and is higher at the edges (red curve). Second, there is a second type of error caused by the signal noise. This random error is indicated by the variation around the offset error, and its magnitude is directly related to the SNR and to the distance between Tx and Rx. The offset error can be minimized by a more careful initial calibration of the system before the actual measurements and the second error the set-up. In Fig. 7b, the combined x, y, z error is shown for three Rx positions over 40 measurements. There are two observations: First, there is a different offset error for each position, with Rx(1,0.72,0) the highest and Rx(0.67,0.725,0) the lowest. The offset error can be attributed to non-ideal calibration, which was only performed in 1D for one location only. Note that the offset error is smaller in the centre position between the four transmitters (blue curve) and is higher at the edges (red curve). Second, there is a second type of error caused by the signal noise. This random error is indicated by the variation around the offset error, and its magnitude is directly related to the SNR and to the distance between Tx and Rx. The offset error can be minimized by a more careful initial calibration of the system before the actual measurements and the second error Page 9 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 Fig. 7 Mean square error (MSE) of x, y, z for selected Rx position (a) and combined error or three receiver positions over 40 measurements (b) Fig. 7 Mean square error (MSE) of x, y, z for selected Rx position (a) and combined error or three receiver positions over 40 measurements (b) by applying a stronger averaging, i.e. taking more measurements for each position into account. 5 Security (results and discussion) LiFi is said to be inherently more secure than radio. Light can be easily kept inside a room and signal levels outside a main light beam are inadequate to eavesdrop on the signal. The inherent protection against a jamming attack on a large industrial installation or factory hall becomes increasingly important in Industry 4.0 settings with autonomous devices. Nonetheless, this view disregards many types of potential attacks. So, we rather phrase the property that light stays inside the room as adding an additional layer of security. However, this does not obviate the need for proper encryption, authentication, access control, key management and hardware device security. If LiFi is used in security-critical settings, the “digital” security needs to be Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 10 of 33 at least as good as is common practice for radio-based infrastructures. For instance, bringing a hacked LiFi device into a secured area should not compromise the com- plete LiFi network security. IT departments or operators prefer to rely on security mechanisms that are compatible with commonly used industry standards, such as Wi-Fi-compatible or 5G-based approaches, respectively. The operator-focussed approach taken in ELIoT sees LiFi as a new access technol- ogy for non3GPP access to 5G network systems. The secured connection for the non- 3GPP access to 5G is accomplished through the encapsulation and encryption of the transferred packets. A 3GPP technical specification [22] has introduced a new com- ponent called the Non-3GPP Interworking Function (N3IWF) which is responsible for the access and session operations between the user equipment (UE) and the core network (CN). It realizes IPSec tunnelling from the UE to the N3IWF to control the data security in both 5G control and user planes. Internet key exchange (IKE) and extensible authentication protocols authorization and key agreement (EAP-AKA) generate the key pairs for packet encryption and decryption. In this way, encapsula- tion and protection of LiFi communications comply with common security standards. A second approach, which is appropriate for enterprise networks, provides compat- ibility with the IEEE 802.1X standard [23]. This approach makes any LiFi device oper- ates in a similar way to a Wi-Fi device. The IEEE 802.1X standard defines a port-based network access control and authentication protocol that prevents an unauthorized client device from connecting to a LAN through publicly accessible ports unless they are properly authenticated. As shown in Fig. 5 Security (results and discussion) 8, IEEE 802.1X systems use a standard client/server architecture including the following three components: •A Supplicant, which is a software module running on the terminal device to be authenticated and providing credentials to the authenticator using EAPOL frames [23]. The credentials are provisioned in advance by the network administrator, and could include a user name/password, shared key or a cryptographic certificate [24]. i •An Authenticator, which is a software module running on the access point that controls the access of the terminal device to the network, and that relays the commu- nications between the authenticating device (using EAPOL frames towards the AP) and the authentication server (using RADIUS protocol towards the AP) [25]. Fig. 8 LiFi enterprise network deployment using IEEE 802.1 × including the Supplicant, Authenticator and Authentication Server Fig. 8 LiFi enterprise network deployment using IEEE 802.1 × including the Supplicant, Authenticator and Authentication Server Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 11 of 33 •An Authentication Server, which maintains the trust relationships between terminal devices and the access network that can receive and respond to requests for network access originating from a terminal device running the supplicant. The server can evalu- ate the access requests and inform the authenticator if the connection is to be allowed for the device requesting access. The authentication server runs software supporting the RADIUS and EAP authentication protocols [26]. The authentication exchange is carried out between the supplicant and the authenti- cation server with the authenticator acting as a relay for the EAP messages, see Fig. 8. The EAP messages carrying the EAP method-specific data are transported using EAPOL frames between the supplicant and the authenticator and RADIUS protocol between the authenticator and the authentication server, as shown in Fig. 9. The authenticator facilitating EAP between the supplicant and the authentication server ensures that no user data can be transferred through the access point before the device is granted access and the secure session establishment involving device authenti- cation and key derivation is finalized. i The following steps are performed for the secure session establishment: • Device and network authentication through EAP method handshake between the supplicant and authentication server using pre-provisioned credentials, which results in both, the supplicant and the authentication server, in a common master key: pair- wise master key (PMK). • Transfer of the PMK from the authentication server to the access point • Transfer of the PMK from the authentication server to the access point • Key derivation handshake like a 4-way handshake for Wi-Fi defined by [27] per- formed between the terminal device and the access point by using the master key, resulting in a set of specific transient keys: pairwise transient key (PTK) and group transient key (GTK) keys. • Key derivation handshake like a 4-way handshake for Wi-Fi defined by [27] per- formed between the terminal device and the access point by using the master key, resulting in a set of specific transient keys: pairwise transient key (PTK) and group transient key (GTK) keys. The PTK and GTK keys resulting from the key derivation handshake are used to derive session keys used to secure the data exchanged between the LiFi endpoint node in the client device and the LiFi access point over LiFi link. Once the session keys are derived, the device is granted access to the network and a secure channel is established between the client device and the access point allowing user data to be securely exchanged. This concludes the session establishment exchange. Fig. 9 IEEE 802.1 × port-based network access control protocol stack Fig. 9 IEEE 802.1 × port-based network access control protocol stack Page 12 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 Fig. 10 Handover scenarios between LiFi, Wi-Fi and 5G Fig. 10 Handover scenarios between LiFi, Wi-Fi and 5G Table 1 Handover definitions between LiFi, Wi-Fi and 5G Term Description/definition MIMO link adaptation A single AP optimizes the use of multiple OFEs to aid end point mobility. Association remains in the AP Horizontal handover between non-overlapping and overlapping areas Transfers of association state between two APs of overlapping LiFi cells. This is independent of the interference coordination between the APs L2 vertical handover Moves layer 2 connectivity between a Wi-Fi network and a LiFi network. The terminal’s MAC address is preserved L3 vertical handover Moves IP-level connectivity of a terminal between different access tech- nologies. The terminal’s IP address might be preserved. Integration into 5G core network Fig. 10 Handover scenarios between LiFi, Wi-Fi and 5G Fig. 10 Handover scenarios between LiFi, Wi-Fi and 5G Fig. • Transfer of the PMK from the authentication server to the access point 10 Handover scenarios between LiFi, Wi-Fi and 5G Table 1 Handover definitions between LiFi, Wi-Fi and 5G Term Description/definition MIMO link adaptation A single AP optimizes the use of multiple OFEs to aid end point mobility. Association remains in the AP Horizontal handover between non-overlapping and overlapping areas Transfers of association state between two APs of overlapping LiFi cells. This is independent of the interference coordination between the APs L2 vertical handover Moves layer 2 connectivity between a Wi-Fi network and a LiFi network. The terminal’s MAC address is preserved L3 vertical handover Moves IP-level connectivity of a terminal between different access tech- nologies. The terminal’s IP address might be preserved. Integration into 5G core network Table 1 Handover definitions between LiFi, Wi-Fi and 5G This approach provides not only full control over who is joining the enterprise net- work but also the flexibility of supporting anyone of the standardized EAP authenti- cation methods within the same infrastructure. A standardized set of commonly used protocols ensures that LiFi connectivity can easily be incorporated into any enterprise IT infrastructure. 6 Mobility support (results and discussion) Seamless large-area coverage involving multiple LiFi APs and the integration of radio- based networks like 5G and Wi-Fi are important topics in ELIoT. Different handover scenarios are investigated, each with different characteristics and requirements. Moreo- ver, security and access aspects are addressed to allow the integration of LiFi into exist- ing Wi-Fi and 5G radio networks [13]. Figure 10 shows the handovers studied, with the definition in Table 1 and a detailed description of various mobility scenarios below. 6.3 Horizontal handover: moving between APs with overlapping coverage areasfi In this scenario, a large area such as an open office space is served by multiple LiFi APs, each covering part of the total area. Now the coverage of APs intentionally overlaps to prevent dead zones. A terminal in an overlap area connects to one AP, while the interfer- ence has to be managed, e.g. by coordinating APs. The motion of terminals across ser- vice areas is similar to Wi-Fi: a terminal can initiate a handover. This could be a “break before make” handover, whereby the LiFi link is lost, but most preferably a “make before break” handover is supported to keep breaks of the LiFi link short. LiFi will have much smaller cells than Wi-Fi and more abrupt cell fringes. This demands a fast handover. A terminal therefore preferably anticipates a handover by pre- registering to a neighbouring AP and pre-establishing security keys via the currently active connection and the backhaul. The line-of-sight propagation characteristics of LiFi require adequate handling of interference and hidden-terminal problems. A terminal typically sees the APs, but not any of the other terminals. An AP sees terminals but not the neighbouring APs. Hence, carrier-sense multiple access (CSMA) may not prevent APs (or terminals) from transmitting at the same time. LiFi preferably uses coordinated medium access. This also strengthens the ability to guarantee QoS, by organizing the medium access so that the link is not hampered by interference. 6.1 MIMO link adaptation within one LiFi AP In this scenario, multiple OFEs provide LiFi access in a service area (e.g. a single room) via a common AP that adapts its MIMO PHY layer if mobility demands so. Through other OFEs, the LiFi AP keeps the connection to the terminal alive even if the line of sight towards one OFE is accidentally blocked. Moreover, if the terminal is moved or rotated, the connection is maintained. The LiFi AP and terminal may optimize link Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 13 of 33 performance by taking advantage of the light traveling via different signal paths. This is realized by using the D-MIMO technology. If a terminal moves within the coverage area of the LiFi AP, the latter can adapt the connection to the terminal by changing the selection of an active subset of OFEs for this terminal and by adapting the physical layer parameters (e.g. bit loading) for optimal link quality. The LiFi AP can trade off between robustness and power consumption. For robustness, the AP may activate all its OFEs for a terminal. To reduce power consumption, the AP may activate only the best OFE. 6.2 Horizontal handover: moving between APs with non‑overlapping coverage areas 6.2 Horizontal handover: moving between APs with non‑overlapping coverage areas In this scenario, multiple areas are each served by a LiFi AP, whereby these areas are optically separated. This is typically the case for the situation of multiple (small) rooms in a building. No interference occurs between APs. If a terminal moves from one area to another, it typically enters an intermediate area (e.g. a corridor) without LiFi coverage and loses LiFi connectivity temporarily. As soon as the terminal re-enters an area with LiFi coverage, it re-connects (rapidly) to LiFi. 6.4 L2 vertical handover: moving in/out of LiFi area within a Wi‑Fi area In this scenario, a small area is served by LiFi (e.g. a single room), while a larger area is covered via Wi-Fi (e.g. in a hallway or in less intensively used rooms). For a terminal that moves in or out of the LiFi area, a vertical L2 handover between LiFi and Wi-Fi takes place. A terminal in the LiFi area can offload the traffic from Wi-Fi and increases its QoS. A terminal that moves out of the LiFi area keeps a connection through Wi-Fi, as shown in Fig. 10. Ideally, by aggregating two wireless techniques, their throughput is added. In the IEEE 802.11 standard, L2 handover is defined by fast basic service set (BSS) transition and fast session transfer. As LiFi is not yet tightly integrated into 802.11, alter- natively, integration of both techniques can be realized via the IEEE. 1905.1 standard. In Page 14 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 fact, 1905.1 existed already for some time but gained popularity as the Wi-Fi mesh tech- nology reuses clauses from it. 6.5 L3 vertical handover in 5G context Integration of LiFi into radio-based 5G connections requires an L3 vertical handover [13]. This enables a user to freely move between both 3GPP-trusted and untrusted access systems. The implementation in ELIoT of such a handover is shown in Fig. 11. Building blocks on the left-hand side of Fig. 11 consist of a user terminal, e.g. a PC, which has parallel connections to the application server, by LiFi and by 5G new radio. The application server runs the 5G Core Network using software-defined networking (SDN). The radio connection, i.e. the air interface, takes place on the 3GPP radio access network (RAN). As a mediator, gNodeB is used, which represents the 5G RAN, and the core network, which is responsible to investigate the access operation. A second connec- tion is created over an Ethernet interface ETH to the LiFi OFE, then through the actual LiFi wireless link to the infrastructure LiFi OFE, and finally via Ethernet to N3IWF cop- ing with the integration of non-3GPP access technologies into the 5G packet core. In principle, the LiFi link is untrusted connectivity. Thus, the integration of LiFi into a 5G system mandates to encapsulate and encrypt packets between the user terminal and the N3IWF before these are transferred over the air; regardless of any link encryption. This operation for non3GPP access has been implemented in ELIoT. The responsible 5G net- work function is named N3IWF and it makes sure that the IPSec tunnelling is estab- lished without flaws to match the security standards [13]. In the case of link failures, the integrated system is capable of switching between these two access networks with very low latency and keeping user entities preserved. 7 Inter and intra‑cell interference To avoid that interference degrades performance inside cells or in overlap zones, one can divide the channel resources over the contending nodes. Multiple access is possible by time division (TDMA), modulation frequency division (FDMA), code division (CDMA) or wavelength division (WDMA), provided that a flexible MAC protocol can handle spatially conflicting demands. Radio systems, particularly those designed for unlicensed bands as used for IEEE 802.11, use carrier-sense multiple access—collision avoidance (CSMA/CA), that is, a node “listens before it talks”. CSMA can flexibly handle random arrivals but may not guarantee QoS for ongoing sessions as it lacks the ability to reserve resources. Fig. 11 LiFi-5G integration, as experimentally implemented at test set-up in ELIoT Fig. 11 LiFi-5G integration, as experimentally implemented at test set-up in ELIoT Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 15 of 33 As indoor LiFi networks use line-of-sight propagation characteristics, CSMA faces hidden node problems. An upward-looking MU or end point (EP) typically sees the downwards-faced OFE of the APs, but not any of the other EPs. An AP sees termi- nals but not the neighbouring APs. Hence, CSMA may not prevent APs (or terminals) from transmitting at the same time. LiFi preferably uses coordinated medium access within each cell. This also strengthens the ability to guarantee QoS, by organizing the medium access so that the link is not hampered by interference. In ELIoT, we focus on the ITU G.9991 standard that adopts TDMA for high QoS. Continuous coverage implies that cells will overlap and potentially interfere with each other. So while moving between APs with overlapping coverage areas, not only a handover but also interference coordination of conflicting transmissions needs to be addressed spatially. A distinction from RF is that LiFi communicates via a directional LoS. This keeps the interference mostly localized to the overlap zone of the directly adjacent illuminated areas. This makes it possible to densely reuse the optical spec- trum, so that the full high data rates become available at every machine location in a factory hall. In most cases, leakage into remote cells is negligible, which contrasts with rich multipath propagation in indoor radio systems. Our reference system con- tains ceiling-mounted LiFi APs while LiFi end points (EP) are spread in the area. Fig- ure 12a depicts the coverage areas of AP1 and AP2 in solid grey and those of the EP1 and EP2 in dashed lines, respectively. 7.1 Spatial TDMA The ITU-T G.9991 TDMA protocol can run as a specific firmware on ICs available for ITU-T G.9960/G.9961 [21]. ITU-T G.vlc separates the network into “domains”. A LiFi domain can contain multiple cells, thus multiple CAs, and is seen as a part of a larger LiFi network. The domain concept may also fit to IEEE. 802.11 standard, if a LiFi domain is regarded as a BSS. All APs of one domain use a synchronized clock to ensure that TDMA frames are aligned across cells. Whether or not two neighbouring APs can simultaneously use the same time slot depends on the interference levels seen by EPs in the overlap zone [17]. In fact, the idea of dynamically assigning time slots not only in a time domain (TDMA) but also in spatial domain was launched in [28] and allows a vir- tual (cell-free) concept [29]. For optical wireless communications (OWC), ELIoT exploited the opportunity of con- necting every domain via their AP to a backbone which exchanges inter-domain manage- ment messages via unique connections. Figure 12b shows this architecture. A dedicated LiFi controller (LC) handles and manages inter-domain contention, but does not act as a traffic router in the backbone. The LC observes instances of inter-domain interference and coordinates APs by setting constraints to their access schedules to avoid possible collisions, for instance using insights from [17]. The LC can be implemented as a central entity, but its functionality may also be distributed among the APs while the APs can nonetheless converge to a common, mutually coordinated scheduling. A common clock is shared among the domains to coordinate the MAC cycles in multiple domains. Fig- ure 12b shows how APs are synchronized. This can be realized by running the precision time protocol (PTP) according to IEEE 1588 over the backbone. APs and EP terminals advertise their presence over a CC to detect potential inter-domain interference. Every AP tracks the activity of neighbouring nodes and reports interference events (collisions) to the LC, to allow it to set scheduling constraints for adjacent APs to resolve any con- flicts. We illustrate this for the scenario in Fig. 12. In domain 1, EP1 receives advertise- ments of AP2 belonging to domain 2. AP1 flags this as an interference risk to the LC. In response, the LC constraints AP1 by only allowing the transmission to EP1 in a limited set of timeslots. 7.1 Spatial TDMA It also constrains AP2 by prohibiting this set of slots in any communica- tion with its own EPs. The algorithm to set time-division constraints is left as an oppor- tunity for proprietary innovation, as it does not need to be fixed in ITU-T G.9991. A certification authority may be needed to ensure the interoperability of LCs, APs, and EPs from different vendors. 7 Inter and intra‑cell interference EP1 is in an overlap zone for AP1 and AP2, both in up and downlink. Figure 12a shows how AP1 interferes with AP2 transmitting to EP2I. AP2 may not detect transmissions from AP1. Figure 12a also shows that EP2 is a hidden node for EP1. Although LoS propagation limits the coverage to a specific cell with only a small spillover, hidden node problems require mitigation as APs see EPs, but not other APs, while EPs see Aps, but not other EPs. So, in fact, a connection (in red) between AP2 and EP1 would be within coverage, but sees interference from hid- den terminals in both directions. LiFi is often promoted for its ability to provide contention-free QoS for ongoing ses- sions. In fact, TDMA allows coordinated scheduling. The network of APs determines the scheduling of the medium. Competition for access is resolved by coordinating the time schedules within and among adjacent APs, where the latter becomes a spatial exten- sion of TDMA. A common channel (CC) may be established in order to facilitate the exchange of information between domains and facilitate detection of the interference. Fig. 12 LiFi network consisting of fixed APs and mobile user terminals EPs (a) and MAC cycle alignment to common clock and CC (b) Fig. 12 LiFi network consisting of fixed APs and mobile user terminals EPs (a) and MAC cycle alignment to common clock and CC (b) Page 16 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 7.2 Handover‑friendly and interference‑mitigating cell layout In a basic spatially extended LiFi network, every light point (e.g. a ceiling-mounted luminaire) can become an OFE or even a full AP, so cell boundaries and handovers can occur in the middle between two such light points. In RF-based WLAN systems, co-located antennas can offer MIMO gains because of the multipath nature of indoor radio propagation. For a rich scattering RF environment, an antenna separation of just half a wavelength is known to be adequate for creating independent, thus MIMO- separable streams. In LiFi, this mechanism of random multipath wave cancellation is not seen, because the detector itself spans thousands of wavelengths. Hence, even for Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 17 of 33 non-LoS LiFi with a rich scattering of intensity-modulated LC, the detector averages out multipath randomness by non-coherently adding photon flows. To achieve MIMO gains OFE separation needs to be much larger than the wavelength of the light; it needs to be larger than c/f, where f is the modulation frequency (10 to 100 MHz), thus many metres rather than micrometres. Co-location of optical MIMO OFEs in one device does not lead to independent channels, so a distributed architecture of MIMO OFEs is needed. Secondly, in optical propagation, the LoS is typically much stronger than the collective set of reflections, such that multipath does not yield a MIMO gain. Thirdly, and possibly most importantly, the LiFi channel statistics are dominated by blockage effects rather than by multipath wave cancellation. f If we consider these factors and the observation that the signal is most vulner- able in the middle between two light points where propagation distances are larger and where slant angles of arrival are more prone to blockage, we come to a cell lay- out, in which every cell is illuminated from its corners, as in Fig. 13. Inside the area, mobility is supported by MIMO link adaptation. Handover to an adjacent cell occurs right underneath each light point rather than halfway between APs. For the typical 60 × 60 cm ceiling tiles in offices, this calls for 90-degree corner sectors pointing into the cell with cell sizes of an integer multiple of the ceiling grid size, e.g. 1.8 or 2.4 m. The EP can of course rely on angular diversity and thus can be miniaturized. 8 Broadband PHY fronthaul As argued before, LiFi needs D-MIMO to protect against link blockage. This section presents a fronthaul infrastructure and identifies how the architecture of the fronthaul can be (cost-) optimized by leveraging the MIMO capabilities to jointly address wire- less cross talk and possible cross talk in the fronthaul infrastructure. To explain that, we initially model the concatenation of a fronthaul and a wireless channel from a MIMO perspective. The end-to-end channel, including wireless and fronthaul, from NT emitters to NR receivers is described by a NR × NT MIMO matrix H f = Z f G f . That is the prod- uct of the cross talk coefficients from the POF fronthaul link represented by G f and the coefficients of the spatial overlap in the wireless channel given by Z f . That is, we found that the Z f = Z0 is mostly constant over frequency and that the response of G f mainly stems from known component properties. In fact, because of practical con- siderations only the ceiling OFEs, i.e. the DUs, can be spatially distributed, while at the client EP, angular separation of OFE co-located in the same miniaturized MU devices is preferred. Z0 originates from distance-related path losses, emission patterns and collimated receiver opening angles, while multipath reflections are weak. Thus, Z0 is independent of the modulation frequency. This structure in the channel matrix allows efficient implementation and limits protocol overhead. In fact, while we saw that in radio communication over mobile multipath channels, adaptive subcarrier-dependent loading requires prohibitive amounts of signalling overhead, in OWC over LEDs, it is very feasi- ble, and actually proven within the ITU G.9991 to be effective and efficient. ffi To evaluate the throughput of a DC-biased optical OFDM (DCO-OFDM) D-MIMO link, a singular value decomposition (SVD) of the overall channel matrix H f is com- puted and the overall throughput is estimated by (1): (1) R = B NSC n=1 K k=1 log2 1 + SNR NTηHŴ ξ2 k fn (1) where B is the bandwidth occupied by each subcarrier, NSC the number of subcarriers, K the rank of H f , ηH the frequency–average path loss, Ŵ = 10 reflects typical system properties and, peak-to-average ratio (PAPR) DC biasing penalty [9, 19] and ξk fn is the k th singular value of H f at the n th subcarrier. 7.2 Handover‑friendly and interference‑mitigating cell layout The cell overlap zone now falls underneath an OFE and may be abrupt, thus the size of the overlap area will be mainly determined by the transition distance needed to execute a horizontal handover for typical EP speeds. For collimated sectors, thus when the sector boundary is sharply confined in azimuth rather than a gradu- ally increasing lateral path loss, this handover line can be defined much more accu- rately for sector handover within one OFE, than for a transition somewhere halfway between two spatially separated OFEs. Reducing the overlap area also reduces the need for interference protection at the MAC layer, thus reducing the need for inhibit- ing potentially conflicting emissions, and thus the user capacity increases. Fig. 13 LiFi cell layout that places cell boundaries along collimated sectors. Horizontal handovers take place on a well-defined line, while soft MIMO link adaptation takes place in areas in between APs. Left: cross section. Right: top view of MIMO channels x,y in cell 1, 2,…4 Fig. 13 LiFi cell layout that places cell boundaries along collimated sectors. Horizontal handovers take place on a well-defined line, while soft MIMO link adaptation takes place in areas in between APs. Left: cross section. Right: top view of MIMO channels x,y in cell 1, 2,…4 Fig. 13 LiFi cell layout that places cell boundaries along collimated sectors. Horizontal handovers take place on a well-defined line, while soft MIMO link adaptation takes place in areas in between APs. Left: cros section. Right: top view of MIMO channels x,y in cell 1, 2,…4 Page 18 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 8 Broadband PHY fronthaul SNR is the signal-to-noise ratio refer- enced to the transmitter, thus defined as the transmit power PT divided over the receiver noise spectral density N0 times the total bandwidth B fN −f1 . The SNR can be influ- enced by spreading the power of the streams all over the frequency band in the most effective way within the power budget, but we assumed uniform power loading as done in ITU G.9991. A simplified form is to use a single spatial stream with only one non-directional receiver photodiode (force K = 1), but to emit this from multiple ceiling locations. Then, the signal strength becomes the sum of multiple beams, with some phase delay effects if the path length differs. In ELIoT, we evaluated and tested K > 1 in various ways. In the next sections, we initially address the combined challenge of how to build a D-MIMO system and a corresponding ceiling infrastructure and subsequently test to what extent the existing solution (in particular [21]) can already address this or need to be improved. Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 19 of 33 8.1 Results on SDM and WDM‑over‑POF POF can be attractive for the fronthaul of LiFi systems due to its do-it-yourself (DIY) ease of installation and its immunity against electromagnetic interference (EMI) [9]. To accommodate D-MIMO in the wireless link, two techniques can be used for POF-based LiFi fronthaul: space-division multiplexing (SDM) and wavelength-division multiplex- ing (WDM). SDM realizes point-to-point connections at the same wavelength over POF to connect each AP to the LiFi modem, as depicted in Fig. 14a. The main advantage of SDM is that there is no optical cross talk in the optical feeding network. For the WDM approach, a single feeder POF is used to distribute the signals to the various APs by using different wavelengths, as illustrated in Fig. 14b. WDM can simplify the installation and maintenance; however, it has higher complexity. With WDM, the amount of cross talk between channels can be significant, if there is overlap in the optical spectrum. The cross talk can be avoided using narrow optical light sources, such as laser diodes (LDs); however, a lower-cost system can be realized by using LEDs, which have a wider spec- trum, thus higher spectral overlap. Yet, a new insight in ELIoT is that moderate colour cross talk in the POF is not necessarily harmful as it can be mitigated by end-to-end MIMO (matrix inversion) processing that will be applied for the wireless link [9, 30]. We measured SDM-over-POF with D-MIMO considering the channel response for the concatenation of POF and a wireless link. We only had access to commercially available red LEDs that would fit a POF connection, so we characterized the WDM- over-POF using LDs at 520 nm (green) and one at 658 nm (red). The WDM-over-POF Fig. 14 a SDM and b WDM approach using POF as the fronthaul of LiFi Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 20 of 33 link is characterized by measuring the losses due to light absorption in the POF and the cross talk in the wavelength domain. We define cross talk as the leakage between adjacent channels that can occur due to insufficient channel separation in the demul- tiplexer (DeMux). Our DeMux has a cross talk level of − 13 dB and 3.2 dB loss for the green channel and for the red channel a cross talk level of − 25.7 dB and 3.6 dB loss [31]. 8.1 Results on SDM and WDM‑over‑POF J Wireless Com Network (2022) 2022:89 Fig 16 Normalized downlink singular values for SDM ξ1S1 and ξ2S1 for Scenario 1: spatially s Fig. 16 Normalized downlink singular values for SDM,ξ1S1 and ξ2S1 for Scenario 1: spatially separated RX (solid lines) d1 = 100 cm, d2 = 70 cm, d3 = 70 cm, Scenario 2: (dashed lines) and ξ1S2 and ξ2S2 for Scenario 2: co-located RX:d1 = 50 cm, d2 = 5 cm, d3 = 35 cm Fig. 16 Normalized downlink singular values for SDM,ξ1S1 and ξ2S1 for Scenario 1: spatially separated RX (solid lines) d1 = 100 cm, d2 = 70 cm, d3 = 70 cm, Scenario 2: (dashed lines) and ξ1S2 and ξ2S2 for Scenario 2: co-located RX:d1 = 50 cm, d2 = 5 cm, d3 = 35 cm Fig. 16 Normalized downlink singular values for SDM,ξ1S1 and ξ2S1 for Scenario 1: spatially separated RX (solid lines) d1 = 100 cm, d2 = 70 cm, d3 = 70 cm, Scenario 2: (dashed lines) and ξ1S2 and ξ2S2 for Scenario 2: co-located RX:d1 = 50 cm, d2 = 5 cm, d3 = 35 cm Fig. 16 Normalized downlink singular values for SDM,ξ1S1 and ξ2S1 for Scenario 1: spatially separated RX (solid lines) d1 = 100 cm, d2 = 70 cm, d3 = 70 cm, Scenario 2: (dashed lines) and ξ1S2 and ξ2S2 for Scenario 2: co-located RX:d1 = 50 cm, d2 = 5 cm, d3 = 35 cm Table 2 Throughput evaluation of D-MIMO set-up in two different scenarios D-MIMO SDM D-MIMO WDM Scenario 1 586 Mbps Scenario 1 484 Mbps Scenario 2 421 Mbps Scenario 2 369 Mbps Fig. 17 Passive DUs: schematic, where the red path refers to a optical downlink, b uplink Table 2 Throughput evaluation of D-MIMO set-up in two different scenarios Table 2 Throughput evaluation of D-MIMO set-up in two different scenarios Fig. 17 Passive DUs: schematic, where the red path refers to a optical downlink, b uplink 8.1 Results on SDM and WDM‑over‑POF The losses for the green and red channels are asymmetrical due to differences in emitting and received optical power and due to different receiver sensitivities for either wavelength. Considering only the optical link for WDM, a throughput of 2.5 Gbps is achieved for the green wavelength channel and 4.3 Gbps for the red wave- length channel. To evaluate the performance of WDM-over-POF with D-MIMO, the POF link and the wireless link channel response are measured separately and then combined. The experimental set-up is presented in Fig. 15, where d1 is the distance between APS and user receiver, d2 is the distance between receivers and d3 is the distance between the APs. Two measurement scenarios are implemented to repre- sent the downlink of a high bandwidth multi-user MIMO transmission for LiFi. In the first scenario, the access points and users are located at d1 = 100 cm, d2 = 70 cm and d3 = 70 cm. In the second scenario, the receivers are placed closer together, namely at d1 = 50 cm, d2 = 5 cm and d3 = 35 cm. From the singular values of H f , seen in Fig. 16 with size 2 × 2, and considering an SNR of 20 dB, the achievable throughput of the system is calculated using (1). Table 2 reveals that the performance for SDM is, as expected, better than for WDM, but only slightly so. In SDM, the absence of spectral overlap leads to a better-condi- tioned matrix which increases the bit rate. Our WDM system further lacked mar- gins to overcome unequal link budgets, which worsens after cross talk removal due to noise enhancement in the MIMO equalizer. In particular, with available compo- nents, the green channel appeared challenging due to lower optical power and lower responsivity at the receiver. In both scenarios, the red channel can carry more data. We observed that the optical WDM link in isolation provides high performance, but end-to-end throughput reduces when the wireless channel is concatenated. The SDM has been done with LEDs rather than with lasers. Nonetheless, the end-to-end link is mostly limited by the wireless channel rather than by the POF [9]. POF POF USER 2 USER 1 LiFi modem APs Fig. 15 Experimental set-up for LiFi D-MIMO using POF with SDM approach Fig. 15 Experimental set-up for LiFi D-MIMO using POF with SDM approach Page 21 of 33 Linnartz et al. 8.2 Results for passive all‑optical OFE To simplify the ceiling infrastructure for D-MIMO, it is attractive to avoid the need for electrical powering in the DUs (i.e. of the OFEs of the AP). One solution is to remotely feed optical fibres by a broadband LD and directly emit these signals from the fibre end, without any optical–electrical–optical conversion [32]. In Fig. 17a, b the system diagram Page 22 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 for the downlink and uplink is presented, respectively. The transmitter is composed of a 1 × 2 power splitter and one distributed feedback (DFB) LD. The LD used emits red light at 658 nm, and is directly modulated in its linear region and then butt-coupled into the POF. The red LD emits an optical power of + 2 dBm and it is biased at 80 mA. The light beam is transmitted through 5 m of POF, split and then transmitted through 1 m POF. In the DU where APs are placed, a lens is put in front of the POF end to reduce the beam divergence. The standard polymethyl methacrylate (PMMA) POF has a numerical aper- ture of NA = 0.5; thus, if no lens is used, the light exiting the POF would be launched over an angular range of − 30°–30°. To create a wireless cell of 45 cm, as shown in Fig. 18, a lens is placed in a defocused position in the POF-end face. On the receiver side, the beam is received by another lens, coupled into a piece of POF and detected by an optical receiver composed of a silicon photodiode (PD) and a transimpedance amplifier (TIA). The PD + TIA has a detection bandwidth of 1.2 GHz. h The system schematic is presented in Fig. 17, and the implementation in the labora- tory is presented in Fig. 18. The horizontal distance d1 between the two POF ends is set to 30 cm, while the vertical distance d2 between the POF outlets and receiver is set to 1.2 m. Measurements were performed by moving the receiver along the x-axis, to simulate motion across cells. The x-axis position 0 represents the middle between both POFs-end transmitters and the positions + 15 and − 15, represent the position in front of AP1 and AP2, respectively. To evaluate the achievable link performance considering user movement, transmis- sions using discrete multitone (DMT) modulation were realized. 8.2 Results for passive all‑optical OFE DMT, which is a (real- valued) baseband variant of OFDM, were used both for downlink and uplink. DMT was optimized by adaptive bit and power loading over 128 subcarriers, clipped at 9 dB, thus at 2 √ 2 times the rms signal strength. A bit error rate (BER) below the FEC level 1E-3 is achieved for all the presented results. An arbitrary waveform generator (AWG) works as a digital-to-analogue converter (DAC) and generates the DMT signal. At the receiver, the signal is captured by a digital phosphor oscilloscope (DPO) that works as an ana- logue-to-digital converter (ADC) sampling at 50 GSa/s. Offline signal processing is per- formed to obtain the throughput, SNR and BER counting for different positions of the user with respect to the POF outputs. Figure 19a presents the throughput for various receiver positions for the downlink and uplink, respectively. The maximum throughput is obtained at the centre, position 0, Fig. 18 Experimental set-up with passive DUs Fig. 18 Experimental set-up with passive DUs Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 23 of 33 Fig. 19 Link performance of DMT incl. 1.2 m VLC transmission for the downlink and uplink (a) and POF throughput without wireless for various optical power (b) Fig. 19 Link performance of DMT incl. 1.2 m VLC transmission for the downlink and uplink (a) and POF throughput without wireless for various optical power (b) where the receiver is positioned in the middle of the overlapping area, achieving around 3.3 Gbps (downlink) and 2.6 Gbps (uplink) using DMT. At position 0, the receiver obtains signal contributions from both transmitters, which increases its SNR and, con- sequently, the throughput. When a user moves among cells, a throughput variation of 1.3–1.4 Gbps was measured. Figure 19b presents the performance of the POF output, measured directly at AP. Including wireless at x = 0, the received power is − 13.5 dBm. Using Fig. 19b for − 13.5 dBm we can see a difference of 0.7 Gbps, and for x = − 10, where the received power is − 16.5 dBm, the difference becomes 1.2 Gbps. So, the user position with respect to APs has a considerable impact on the link performance. 9.1 Review of ITU‑based LiFi technology in the market The availability of chipsets [21] allowed the commercial release of practical LiFi systems and built confidence in the standard. For example, we can mention Signify’s 2018 prod- ucts offered a PHY rate up to 350 Mbps for downlink and 250 Mbps for uplink [33]. Meanwhile higher bit rates have also been tested. Also, Fraunhofer HHI developed an advanced combiner prototype, based on POF which is more robust against electro- magnetic interference in industrial scenarios also considered in ELIoT. OFEs generally comprise a LED driver (modulator), an IR-LED for the downlink transmitter and a pho- todiode with transimpedance amplifier for the uplink receiver. Fraunhofer HHI manu- factures OFE prototypes with higher power and improved receiver sensitivity intended for larger coverage areas in industrial scenarios. Fraunhofer HHI released a USB LiFi prototype (LiFi NEON) to reach 1 Gbit/s in the downlink [34]. This is intended for dense user scenarios, e.g. in conference and classrooms. Moreover, there is an advanced out- door LiFi prototype manufactured which allows 1 Gbit/s over 100 m for fixed wireless access scenarios investigated in ELIoT. These various products and prototypes show that early LiFi technology is flexible enough to cover a great variety of different use cases. Office use cases demand a moderate, but guaranteed throughput in a wide coverage area. According to feedback from the professional market, fairly uniform coverage with several meters range is considered more important than world record throughput results in a tiny spot. Reliable, guaranteed low-latency QoS at some 100 Mbit/s/device satisfies the expected user experience. Gigabit performance is of interest in dense scenarios when aggregating the traffic of multiple users inside the coverage area. Up to 16 terminals can be served by the Signify LiFi AP in a managed TDMA scheme, within the coverage zone of six OFEs. OFDM allows reliable detection of signals from multiple OFEs. While the downlink uses 850 nm IR wavelength, the uplink uses 940 nm which allows full duplex communication in the future. The optical power emitted in up- and downlink is well below the eye safety limit by a 40% margin. The Signify modem combines the signals and sends the waveforms via analogue wiring to up to six ceiling-mounted OFEs. The diversity gain of the channel highly depends on the layout and the distances between all transmitters and orientation of receivers. 9.1 Review of ITU‑based LiFi technology in the market Path loss can change quickly when the user moves around the room or rotates its device. Meanwhile, other vendors, e.g. OLEDCOMM, offer G.9991 LiFi solutions. In the ELIoT project, various experiments were conducted to verify the performance of G.9991 in different environments both for phoneline and powerline topologies. The results of these experiments are shared in this section. 9 MIMO performance with G.vlc and G.hn profilesh p pi The LiFi G.9991 standard [15] adopts many technical features which are important for high-speed data transmission such as bit loading, channel estimation from a previous standard (G.9960 to G.9964) also known as G.hn and includes some LiFi-specific fea- tures like handover. As these features were already defined in G.hn, these are present in G.hn chipsets. Chipsets and development kits with (sometimes minor) circuit deviations for each wired medium are readily available [21]. This accelerates LiFi developments and market introduction. Secondly, G.hn offers a robust and stable backbone for LiFi, with gigabit connections over any wire including powerline cables, coaxial cables, copper phonelines and POF. The powerline and phoneline profiles are able to run MIMO over a 100 MHz bandwidth and SISO over a 200 MHz bandwidth. The coax profile runs only in SISO mode up to 200 MHz bandwidth. Its PHY achieves a theoretical throughput of 2 Gbps over phoneline and coax, and of 1.5 Gbps over power lines. OFDM modula- tion uses a bandwidth of 50, 100 or 200 MHz with a maximum of 512 subcarriers. Each subcarrier carries quadrature amplitude modulation (QAM) levels with up to 12 bits per symbol, according to an adaptive bit loading scheme. In contrast to radio standards designed for carrier-based transmission, G.hn works well over wired baseband chan- nels and thus is well suited for LiFi channels. However, LEDs and large photodiodes are low-pass and may exhibit distortion due to nonlinearity of the LED modules; addition- ally, LiFi applications, as any other wireless application, can have mobile users leading Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 24 of 33 to time-varying channels. This poses the question whether G.vlc as it stands and as a derived technology from G.hn now solves all major LiFi problems. 9.2 Measured performance under static conditions The experimental set-up consists of a 2 × 2 MIMO LiFi system as shown in Fig. 20: On the one side of the system, there is a ceiling node representing a LiFi access point and on the other side a user node. Both user and ceiling nodes have a Maxlinear G.hn MIMO evaluation kit connected to two Trulifi optical frontends from Signify [35]. Each Trulifi transceiver has one IR-LED and one PD for bidirectional optical/electrical and Page 25 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 Fig. 20 Test set-up for G.vlc MIMO LiFi with phoneline or powerline profiles Fig. 20 Test set-up for G.vlc MIMO LiFi with phoneline or powerline profiles electrical/optical conversion. The transceivers are connected to the evaluation kit PCBs with CAT5 cables. Connected to each evaluation kit PCB, there is one laptop running iPerf software to measure the network throughput while operating under G.hn MIMO profile. 9.3 G.hn Coax Mode: SISO with Multiple Emitters (rank K = 1)h In Coax mode, no MIMO features are available. That does not preclude the creation of diversity channel, even if we only use one photodiode in the AP (NR = 1). This mimics MISO, but without the ability to apply for a transmitter phase compensation. Having calibrated our SISO model in [22], we now rely on such model: using L OFEs in the ceil- ing, having a pathloss hl1, for the lth OFE TX to the common 1st receiver, the rate expres- sion (1) for a distortion-free LED reduces to (2) R = B  N n=1 log2  1 + PT NTηHŴN0 L l=1 hl1 exp −2πj dl c fn 2   (2) Here, we modelled unequal cable lengths dl causing an extra latency, dependent on the propagation speed c in the cable or POF. In fact, MIMO channel estimation could facili- tate a frequency-dependent phase correction for dl, but the Coax mode does not support that. The use of an integer (or even) number of bits per QAM constellation is reflected in the rounding brackets ⌊⌋. Figure 21 shows a frequency response for one OFE-OFE link measured with a spec- trum analyser for a typical OFE implementation, extended to a two-ray path originating from the sample signal travelling via two parallel cables. The TIA and detector produce non-white noise as also plotted in Fig. 21a. Theoretical throughputs are somewhat larger than measured throughput, firstly because of different Signal Power to Noise ratios and secondly because we have neglected distortion here. We conclude that the adaptive bit loading can track the channel and its combined interference pattern. Small differences in cabling reduce the throughput significantly, from 221 Mbit/s for equal cable lengths to 131 Mbit/s for a 3-m difference which creates one notch around 33 MHz. At very long distances, the channel response exhibits many notches and the throughput converges to 148 Mbit/s, surprisingly a bit higher than for smaller cable length differences. This throughput at unequal feeder lengths is still better than what would be achieved without Page 26 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 adaptive bit loading, just by error correction coding across the notches. The latter coding Fig. 21 a Received signal strength and noise floor for a two-ray SISO transmission in the middle of two emitters connected with a cable length difference of 0, 3 and 20 m. 9.3 G.hn Coax Mode: SISO with Multiple Emitters (rank K = 1)h b Resulting estimated bit loading Fig. 21 a Received signal strength and noise floor for a two-ray SISO transmission in the middle of two emitters connected with a cable length difference of 0, 3 and 20 m. b Resulting estimated bit loading Fig. 21 a Received signal strength and noise floor for a two-ray SISO transmission in the middle of two emitters connected with a cable length difference of 0, 3 and 20 m. b Resulting estimated bit loading adaptive bit loading, just by error correction coding across the notches. The latter coding solution is for instance used in single-frequency networks for digital audio broadcast- ing (DAB) where user-specific bit loading is not possible. Our bit loading is plotted in Fig. 21b. adaptive bit loading, just by error correction coding across the notches. The latter coding solution is for instance used in single-frequency networks for digital audio broadcast- ing (DAB) where user-specific bit loading is not possible. Our bit loading is plotted in Fig. 21b. Page 27 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 9.4 G.hn Phoneline mode In the first experiment, we investigate the performance of a 2 × 2 MIMO LiFi system using G.hn MIMO 100 MHz phoneline profile. Four different scenarios are consid- ered and the measured throughput for each of them is shown in Table 3: In Scenario 1, the OFEs are 1 m apart from each other, i.e. d1 = d2 = d3 = 1m . In this case, the cross talk channels have low gain, and spatial multiplexing performs well. The system achieves ~ 500 Mbps for both downlink and uplink. In the second scenario, we placed the luminaires at the user side very close to each other, i.e. d3 ∼2cm , just separated by a small angle. The user node was then placed in the middle of the coverage area of the ceil- ing node OFEs at a distance of 1 m. Now, the channel matrix becomes close to singular and the downlink throughput drops by more than one half and the uplink throughput decreased even more. We conclude that the system does not adapt well to handle cross talk and does not switch to a diversity mode in a highly correlated channel. In the third scenario, we kept the OFEs of the user side as in Scenario 2 but, we placed the user in front of OFEC2 . In this case, cross talk is very high, so a change from MIMO spatial mul- tiplexing to spatial diversity would be required. Unfortunately, the current version of the chipset, anticipating a fixed phoneline, was not programmed to switch its transmission mode. Lacking such adaptation, the measured performance was very low. In the fourth scenario, a piece of cardboard is placed in front of OFEU1 to mimic a link blockage. Although one of the MIMO links is unblocked and still available for communication, the phoneline profile did not adapt well and the throughput fully collapsed. 9.5 G.hn powerline profile In the second experiment, we used the G.hn MIMO 100 MHz powerline profile and the measured throughputs are presented in Table 4. In Scenario 5, we have again a low cor- related MIMO LiFi channel, i.e. d1 = d2 = d3 = 1m . In this case, spatial multiplexing performs well, but the achieved throughput is almost half of the achieved throughput with the phoneline profile for a similar setting. This is due to the additional overhead for more robust coding and to address PLC EMI for instance with spectral notches in the powerline profile. In Scenario 6, ceiling OFEC1 and user OFEU1 are blocked to obtain sin- gle-input single-output (SISO) performance for comparison purposes. In contrast to the performance of phoneline profile in Scenario 4, the powerline mode is able to establish communication over the unblocked link and achieve a throughput of 108 Mbps in the downlink channel, which is approximately half of the performance in Scenario 5. In Sce- nario 7, the OFEs at the user side are placed close to each other and placed in the middle of the coverage area—similar to Scenario 2. Although this represents a highly correlated Table 3 Measured throughput of G.hn MIMO 100 MHz with phoneline profile Scenario Description Downlink [Mb/s] Uplink [Mb/s] 1 User-side OFEs kept 1 m apart 511 501 2 User-side OFEs put close together and placed in the middle of the coverage area 200 59.9 3 Same as [2] but user OFEs placed right below ceiling OFEC2 18.5 42.3 4 Same as [1] but user OFEU1 blocked 0.91 3.14 Table 3 Measured throughput of G.hn MIMO 100 MHz with phoneline profile Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 28 of 33 channel, the measured throughput of 115 Mbps for the downlink and 112 Mbps for the uplink are much better than the phoneline case. In Scenario 8, the user node is placed in front of OFEC2 , and a piece of cardboard was placed in front of OFEC1 . The meas- ured throughput was 108 Mb/s for the downlink and 101 Mbps for the uplink. In Sce- nario 9, OFEC1 was unblocked, and the measured throughput increases to 131 Mbps in the downlink channel and 133 Mbps in the uplink channel. 9.5 G.hn powerline profile For both up- and downlink directions, the MIMO options show 12% – 30% more throughput than the multiple- input single-output (MISO) arrangements where one line of sight is blocked. 9.6 Measured performance under dynamic conditions Tests showed that some blockage is addressed as the system adapts from MIMO to MISO mode of operation. However, we also recorded instances in which the throughput collapsed after blockage and did not recover for minutes. The latter demonstrates the need for further improvement in the chipset (hardware/firmware) and the signalling to track changes in channel characteristics rapidly enough. Figure 22 reports an experiment with Scenario 7, achieving 138 Mb/s. At 104 s, the user-side OFEs were placed in front of OFEC2 (highly correlated channels) and the throughput reduced to 133 Mb/s. Thereafter, OFEC1 was blocked at 205 s, further reduc- ing the throughput to 102 Mb/s. When the blockage was removed at around 287 s, the throughput recovered to 130 Mb/s. At 407 s, OFEC2 was blocked which resulted in the collapse of communication for the following 60 s after which the throughput started to toggle between 20 and 50 Mb/s. When the blockage was removed at 611 s, the 130 Mb/s throughput was regained. In summary, although the phoneline profile gives higher throughputs under static well-conditioned spatial multiplexing modes, current implementations do not cope well with channel changes. The powerline profile, despite its lower overall throughput, can handle some dynamism in the communication channel. However, since the PLC modem is developed/optimized for a powerline medium, it lacks features to handle fast changes in channel characteristics arising from mobile users in LiFi. Evidently, a new optimized mode for lighting communication is desirable. LiFi systems based on the coax profile of G.HN modems run either in SISO or MISO modes and do not exhibit the robustness against partial line-of-site blockage that 2 × 2 MIMO profiles have. i In ELIoT, we concluded that preferably a future generation reuses coax mode PHY parameters but extends these to MIMO. 9.6 Measured performance under dynamic conditions Regrettably, for obvious reasons, current IC Table 4 Measured throughput of G.hn MIMO 100 MHz for powerline profile Scenario Description Downlink [Mb/s] Uplink [Mb/s] 5 User-side OFEs kept 1 m apart 191 Not measured 6 Same as [5] but ceiling OFEC1 and user OFEU1 blocked 108 Not measured 7 User-side OFEs put close together and placed in the mid- dle of the coverage area 115 112 8 Same as [7] but user OFEs placed right below ceiling OFEC2 and ceiling OFEC1 blocked 108 101 9 Same as [8] but ceiling OFEC1 is unblocked 131 133 Table 4 Measured throughput of G.hn MIMO 100 MHz for powerline profile Page 29 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 Fig. 22 Link performance when moving user-side OFEs, blocking and unblocking one of the ceiling-side OFEs Fig. 22 Link performance when moving user-side OFEs, blocking and unblocking one of the ceiling-side OFEs hardware makes MIMO available only in the PLC and phoneline profile, but not in the Coax mode. However, the coax PHY settings allow a higher throughput than (the SISO modes of) other profiles, because of coax choice of constellations and error cor- rection coding. These coax settings have been proven at Technology Readiness Level 9 (i.e. in commercial products) to work well over an optical LiFi channel. Lab experiments with waveform generators indicate that these PHY choices with higher throughput per stream also work in MIMO configurations. However, we do recommend faster channel adaptation and mode switching as the LiFi channel can vary rapidly, while a coax cable is a stationary channel. 10 Roadmap for implementation ITU-T approved the ITU-T G.9991 Recommendation for LiFi in 2020. The DCO-OFDM variant of this recommendation is the base of most developments in the ELIoT project. The goal of this recommendation is to provide a fast-to-market LiFi system with ade- quate performance to cover the main industrial, enterprise and residential use cases. By reusing the OFDM engine with its adaptive bit loading, already successfully used in other recommendations (i.e. in the ITU-T G.hn family, aimed to cover in-home connec- tivity over coaxial, phone and power lines), the resulting ITU-T G.9991 document ena- bles LiFi systems by reusing existing chipsets. This way, early LiFi products can benefit from high-volume, low-cost integrated solutions while providing a good adaptation to the specifics of the majority of LiFi use cases. This approach allows integrators to rapidly create an early mass market for LiFi systems with a reasonable investment. In parallel, the ITU-T Q18/15 group within ITU-T is evolving the recommendation to cover remaining and new requirements identified during the first LiFi deployments in industry. In addition, it also develops other scenarios identified in the ELIoT project that were not covered in the first version of the recommendation. Inter-domain handover and interference management is one example that has been added to the G.9991 framework Linnartz et al. J Wireless Com Network (2022) 2022:89 Page 30 of 33 through an amendment to the standard. The work continues to add new features to the standard that will lead in the future to new amendments or a new revision of the recom- mendation. This will allow a new generation of LiFi chipsets that will incorporate specific LiFi hardware macros allowing new applications (e.g. positioning). New evolutions are also expected to address power reduction and better integration with other technologies to fos- ter the adoption of LiFi technology in mobile devices. In this sense, new features have been identified within ELIoT project. Some of them can be incorporated into the current stand- ard and in current generation chipsets through software development while others will necessitate a change of hardware and will take some time to be available since LiFi must first generate a market that is wide enough to justify a new silicon investment or to continue the strategy of piggybacking on existing implementations. 10 Roadmap for implementation Among the new required fea- tures identified by ELIoT project, we may mention faster channel estimations for enhanced mobility support, profile selection to adapt the characteristics of the transmission to the channel specifics and, last but not least, the inclusion of MIMO technologies, currently not in the standard and with limited support in the existing hardware. An evolution of MIMO techniques for LiFi as the ones investigated in the project (multi-user environments) would represent an important step in the performances achieved by LiFi systems. Finally, we mention an integration opportunity of LED front end functionality with the existing general purpose analogue front ends used so far, for instance for PLC, Coax or phonelines. In summary, we can say that the approach used so far allows LiFi vendors to lever- age a broad portfolio of existing ITU-T G.hn compliant chipsets in their early products and facilitate interoperability of early solutions early on, reusing existing ecosystems and lowering the barriers to deploying this new technology. This approach shall be gradually transformed into more specialized LiFi solutions as the market grows and higher vol- umes become possible. 11 Summaryh The ELIoT project addresses LiFi features that enable the next generation of IoT applica- tions for various indoor and outdoor use cases in the industry, office, commercial and consumer sectors. We presented a distributed MIMO wireless topology which can be supported by SDM/WDM fronthaul transport of waveforms via a plastic optical fibre. Besides these, we described security aspects and various concepts for seamless hando- vers between the light-based access points and also to a radio-based infrastructure such as a 5G network. We have shown that the LiFi technology has the potential to provide both precise indoor localization and high-speed data transfer. The results presented indicate that the PHY offers indoor positioning, with an average accuracy of 5 cm, what even the most modern radio-based systems cannot achieve. D-MIMO light communi- cation enhances reliability and delivers throughputs of hundreds of megabits per sec- ond. Concepts to further increase the performance of LiFi systems have already been identified. These features are currently being tested towards use-case demonstrations. We foresee that LiFi systems can complement upcoming 6G systems by offering high QoS link in hotspots. However, research challenges remain, such as making the channel adaptation faster, reduction in power, a further cost-down of the infrastructure that pro- vides signals to all OFEs. Page 31 of 33 Linnartz et al. 11 Summaryh J Wireless Com Network (2022) 2022:89 Abbreviations AP Access point ADC Analogue-to-digital converter API Application programming interfaces AWG Arbitrary waveform generator MSE Average mean square errors BB Baseband BSS Basic service set BER Bit error rate CSMA/CA Carrier-sense multiple access—collision avoidance CSMA Carrier-sense multiple access CU Central unit CFR Channel frequency response CDMA Code division CC Common channel CN Core network DCO-OFDM DC-biased optical orthogonal frequency division multiplexing DAB Digital audio broadcasting DAC Digital-to-analogue converter DC Direct current DMT Discrete multitone D-MIMO Distributed multiple-input multiple-output DUs Distributed units DIY Do-it-yourself EMI Electromagnetic interference EP End points ELIoT Enhance lighting for the Internet of things EAP-AKA Extensible authentication protocols authorization and key agreemen ITS Intelligent transport systems IKE Internet key exchange IoT Internet of things LD Laser diode LC LiFi controller LoS Line of sight LAN Local area network MAC Medium access MU Mobile unit MCS Modulation and coding scheme FDMA Modulation frequency division MIMO Multiple-input multiple-output OFEs Optical front ends OFDM Orthogonal frequency division multiplexing OWC Optical wireless communications PMK Pairwise master key PHY Physical layer POF Plastic optical fibre PMMA Polymethyl methacrylate PLC Power line communication PSD Power spectral density PTP Precision time protocol PAM Pulse amplitude modulation QAM Quadrature amplitude modulation QoS Quality of service RAN Radio access network SNR Signal-to-noise ratio SVD Singular value decomposition SDN Software-defined networking SDM Space division multiplexing SDMA Space division multiple access TDMA Time division TIA Transimpedance amplifier UE User equipment VLC Visible light communication WDMA Wavelength division WDM Wavelength division multiplexing Author contributions The paper is a result of joint work by all authors. JPL and VJ devised and technical technical solutions. JPMGL is the main and coordinating author; CRBC is the corre devised WDM over POF for D-MIMO and the passive all-optical OFE using POF. TE mode. SK, CRBC, AAA and TEBC experimentally verified various MIMO solutions. V References 1. W. Jiang, B. Han, M.A. Habibi, H.D. Schotten, The road towards 6G: a comprehensive survey. IEEE Open J. Commun. Soc. 2, 334–336 (2021) 2. S. Panwar, “Breaking the Milisecond Latency Barrier,” Oct 2020. [Online]. https://spectrum.ieee.org/breaking-the-laten cy-barrier. [Accessed 03 Aug 2021] 3. M. Müller et al., “Leverage LiFi in Smart Manufacturing,” in 2020 IEEE Globecom Workshops , (2020) 4. J.P.M.G. Linnartz et al., “ELIoT: New Features in LiFi for Next-Generation IoT,” in 2021 Joint European Conference 4. J.P.M.G. Linnartz et al., “ELIoT: New Features in LiFi for Next-Generation IoT,” Networks and Communications & 6G Summit (EuCNC/6G Summit), (2021) Networks and Communications & 6G Summit (EuCNC/6G Summit), (2021) 5. K.L. Bober et al., “A Flexible System Concept for LiFi in the Internet of Things,” in ICTON, (2020) 6. T. Fath, H. Haas, Performance comparison of MIMO techniques for optical wireless communications in indoor envi- ronments IEEE Trans Commun 61(2) 733 742 (2013) 6. T. Fath, H. Haas, Performance comparison of MIMO techn ronments. IEEE Trans. Commun. 61(2), 733–742 (2013) 7. L. Zeng et al., High data rate multiple input multiple output (MIMO) optical wireless communications using white led lighting. IEEE J. Sel. Areas Commun. 27(9), 1654–1662 (2009) 7. L. Zeng et al., High data rate multiple input multiple output (MIMO) optical w led lighting. IEEE J. Sel. Areas Commun. 27(9), 1654–1662 (2009) 7. L. Zeng et al., High data rate multiple input multiple out 8. P. Wilke Berenguer et al., Optical wireless MIMO experiments in an industrial environment. IEEE J. Sel. Areas Com- mun. 36(1), 185–193 (2018) 9. S.M. Kouhini et al., “Distributed MIMO Experiment Using LiFi Over Plastic Optical Fiber,” in 2020 IEEE Globecom Work- shops , (2020) 10. C. van den Broek and J.P. M. G. Linnartz, “A simulation study of space and time reservation multiple access,” in 6th I t ti l S i P l I d d M bil R di C i ti (1995) 10. C. van den Broek and J.P. M. G. Linnartz, “A simulation study of space and time reservation multiple access,” in 6th International Symposium on Personal, Indoor and Mobile Radio Communications, (1995) 10. C. van den Broek and J.P. M. G. Linnartz, “A simulation study of space and time reservatio International Symposium on Personal, Indoor and Mobile Radio Communications, (1995) , y p International Symposium on Personal, Indoor and Mobile Radio Communications, (1995) 11. K.L. Availability of data and materials Data sharing is not applicable to this article. Author contributions Author contributions The paper is a result of joint work by all authors. JPL and VJ devised and technically directed the project and co-devised technical solutions. JPMGL is the main and coordinating author; CRBC is the corresponding author. CRBC, ET and TK devised WDM over POF for D-MIMO and the passive all-optical OFE using POF. TEBC and XD measured G.hn in phoneline mode. SK, CRBC, AAA and TEBC experimentally verified various MIMO solutions. VJ, SK, TEBC, JPMGL and XD developed Page 32 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 the theoretical framework. KLB, MR, VJ and MW developed a reference system architecture view. AAA measured LiFi in powerline mode. PvV modelled and analysed the SISO link in coax mode. MR, PS, MW and MM reviewed and improved the presentation and text. PP and TM conceived and described the enterprise and the 5G compatible solution, respectively. MM, DB, CK and KLB devised and described the industrial use case. CK and SK devised and described the positioning solution. MMV, MR, MW and ME verified the compatibility with IC integration and with standardization. All the authors read and approved the final manuscript. Competing interests p g The ELIoT consortium consists of industrial partners who have a common interest in a broad adoption and commercial success of LiFi and its application via standardized and interoperable solutions, and a number of independent academic partners. Received: 17 November 2021 Accepted: 1 September 2022 References Bober et al., Distributed multiuser MIMO for LiFi in industrial wireless applications. J. Lightw. Technol. 39(11), 3420–3433 (2021)l 12. A. Maeder et al., “Towards a flexible functional split for cloud-RAN networks,” in 2014 European Conference on Net- works and Communications (EuCNC), (2014) ( ), ( ) 13. T. Metin, M. Emmelmann, M. Corici, V. Jungnickel, C. Kottke and M. Müller, “Integration of Optical Wireless Communi- cation with 5G Systems,” in 2020 IEEE Globecom Workshops, (2020) 13. T. Metin, M. Emmelmann, M. Corici, V. Jungnickel, C. Kottke and M. Müller, “Integration of Optical Wireless Communi- 13. T. Metin, M. Emmelmann, M. Corici, V. Jungnickel, C. Kottke and M. cation with 5G Systems,” in 2020 IEEE Globecom Workshops, (2020) 13. T. Metin, M. Emmelmann, M. Corici, V. Jungnickel, C. Kottke and M. Müller, Integration of Optical Wireless Communi cation with 5G Systems,” in 2020 IEEE Globecom Workshops, (2020) , , , g , cation with 5G Systems,” in 2020 IEEE Globecom Workshops, (2020) y p 14. T.E.B. Cunha, W.X. Fan, X. Deng, J.P.M.G. Linnartz, “A Space-frequency Power Allocation Algorithm Systems over Low-Pass Channels,” in 2020 1st Optical Wireless Communication Conference, (2020) 14. T.E.B. Cunha, W.X. Fan, X. Deng, J.P.M.G. Linnartz, A Space-frequency Power Allocation Algorithm Systems over Low-Pass Channels,” in 2020 1st Optical Wireless Communication Conference, (2020) Systems over Low-Pass Channels,” in 2020 1st Optical Wireless Communication Conference, (2020) 15. International Telecommunication Union Recommendation G.9991–202104 Amd.2, “High-speed indoo communication transceiver – System architecture, physical layer and data link layer specification,” (202 15. International Telecommunication Union Recommendation G.9991–202104 Amd.2, “High-speed indoor visib communication transceiver – System architecture, physical layer and data link layer specification,” (2021) 16. M. Hinrichs et al., A physical layer for low power optical wireless communications. IEEE Trans. Green Commun. Netw. 5, 4–17 (2021) 17. J. Beysens, J.P.M.G. Linnartz, D. van Wageningen, S. Pollin, TDMA scheduling in spatially extended LiFi networks. IEEE Open J. Commun. Soc. 1, 1524–1538 (2020) 18. S.M. Kouhini et al., LiFi positioning for industry 4.0. IEEE J. Sel Top Quant Electron 27, 1–15 (2021) 19. T.E.B. Cunha, J.P.M.G. Linnartz and X. Deng, “Throughput of Optical WDM with Wide LED Spectra and Imperfect C l d t ti Filt ”i 2020 29th Wi l d O ti l C i ti C f (WOCC) (2020) 19. T.E.B. Cunha, J.P.M.G. Linnartz and X. Funding Funding This work was funded by ELIoT project, which has received funding from the European Union’s Innovation Action Hori- zon 2020 programme under Grant Agreement Number 825651. Availability of data and materials Data sharing is not applicable to this article. Availability of data and materials Data sharing is not applicable to this article. References Linnartz, “Virtual cellular network: a new wireless communications architecture with multiple access ports,” in IEEE Vehicular Technology Conference (VTC), (1994) 29. H.J. Kim and J.P. Linnartz, “Virtual cellular network: a new wireless communications architecture with multiple access ports,” in IEEE Vehicular Technology Conference (VTC), (1994) p gy 30. X. Wu et al., Hybrid LiFi and WiFi networks: a survey. IEEE Commun. Surv. Tutor 23(2), 1398–1420 (2021 30. X. Wu et al., Hybrid LiFi and WiFi networks: a survey. IEEE Com 31. C.R.B. Corrêa et al., “WDM over POF for D-MIMO LiFi system,” in 1st Optical Wireless Communication Conf , y , p , ( ) 32. C.R.B. Corrêa, F.M. Huijskens, E. Tangdiongga and A.M.J. Koonen, “Luminaire-Free Gigabits per second LiFi Transmis- sion employing WDM-over-POF,” in 2020 European Conference on Optical Communications (ECOC), (2020) 32. C.R.B. Corrêa, F.M. Huijskens, E. Tangdiongga and A.M.J. Koonen, “Luminaire-Free Gigabits per second LiFi Transmis- sion employing WDM-over-POF,”in 2020 European Conference on Optical Communications (ECOC), (2020) p y g p p 33. A.M. Khalid et al., “Productization Experiences of G.vlc (ITU) based LiFi System for high Speed Indoor Wireless Access,” in 1st Optical Wireless Communication Conference (OWCC 2020), (2020) 33. A.M. Khalid et al., “Productization Experiences of G.vlc (ITU) based LiFi System for high Speed Indoor Wi in 1st Optical Wireless Communication Conference (OWCC 2020), (2020) 34. LiFi Neon, [Online]. https://lifi-neon.de. [Accessed 05 Septemb pi p 35. Signify Trulifi, [Online]. https://www.assets.signify.com/is/content/PhilipsLighting/Assets/signify/global/20200416- specsheet-trulifi-6002.pdf.. [Accessed 15 April 2021] References Deng, “Throughput of Optical WDM with Wide LED Spectra and Imperfe Color-detecting Filters,” in 2020 29th Wireless and Optical Communications Conference (WOCC), (2020) 19. T.E.B. Cunha, J.P.M.G. Linnartz and X. Deng, Throughput of Optical WDM with Wide LED Spectra and Imperfect Color-detecting Filters,” in 2020 29th Wireless and Optical Communications Conference (WOCC), (2020) 20. S. Mardanikorani, X. Deng, J.P.M.G. Linnartz, A. Khalid, Compensating dynamic nonlinearities in LED photon em to enhance optical wireless communication IEEE Trans Veh Technol 70(2) 1317–1331 (2021) 20. S. Mardanikorani, X. Deng, J.P.M.G. Linnartz, A. Khalid, Compensating dynamic nonlinearities in LED photon emission 20. S. Mardanikorani, X. Deng, J.P.M.G. Linnartz, A. Khalid, Compensating dynamic nonlinearities in L to enhance optical wireless communication. IEEE Trans. Veh. Technol. 70(2), 1317–1331 (2021) to enhance optical wireless communication. IEEE Trans. Veh. Technol. 70(2), 1317–1331 (2021) 21. Maxlinear, “88LX5152, 88LX5153 Wave-2 G.hn Digital Baseband (DBB near.com/ds/88lx515288lx5153:pdf. [Accessed 02 September 2021] 21. Maxlinear, “88LX5152, 88LX5153 Wave-2 G.hn Digital Baseband (DBB) Processor,” 2020. [Online]. https://www.maxli nearcom/ds/88lx515288lx5153:pdf [Accessed 02 September 2021] 22. J.P.M.G. Linnartz, X. Deng and P Communication, August 2021 22. J.P.M.G. Linnartz, X. Deng and P. van Voorthuisen, Impact of Dynamic LED Non-linearity on DCO-OFDM Optical Wireless Communication, August 2021 g 23. IEEE, “IEEE Standard for Local and metropolitan area networks--Port-Based Network Access Control,” [Online]. https:// standards.ieee.org/standard/802_1X-2010.html. [Accessed 18 August 2021] Page 33 of 33 Linnartz et al. J Wireless Com Network (2022) 2022:89 24. IETF, “Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile,” [Online]. https://tools.ietf.org/html/rfc5280. [Accessed 31 July 2021] p g y 25. IETF, “Remote Authentication Dial In User Service (RADIUS),” [Online]. https://tools.ietf.org/html/rfc2865. [Accessed 09 Aug 2021] g ] 26. IETF, “Extensible Authentication Protocol (EAP),” [Online]. https://tools.ietf.org/html/rfc3748. [Accessed 25 Sept 2021] 26. IETF, “Extensible Authentication Protocol (EAP),” [Online]. https://tools.ietf.org/html/rfc3748. [Accessed 25 Sept 2021] 27. IEEE, “IEEE Standard for Information technology--Telecommunications and information exchange between systems Local and metropolitan area networks--Specific requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications,” [Online]. https://standards.ieee.org/standard/802_11-2016.html. [Accessed 12 Oct 2021] g 27. IEEE, “IEEE Standard for Information technology--Telecommunications and information exchange between systems Local and metropolitan area networks--Specific requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications,” [Online]. https://standards.ieee.org/standard/802_11-2016.html. [Accessed 12 Oct 2021] 28. J. Linnartz, On the performance of packet-switched cellular networks for wireless data communications. Wirel. Netw. 1, 129–138 (1995) 29. H.J. Kim and J.P. Publisher’s Note ub s e s ote 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/W4387772432	https://plantmethods.biomedcentral.com/counter/pdf/10.1186/s13007-023-01086-y	English	null	Comparing DNA isolation methods for forest trees: quality, plastic footprint, and time-efficiency	Plant methods	2,023	cc-by	8,945	Plant Methods Plant Methods Plant Methods Guillardín and MacKay Plant Methods (2023) 19:111 https://doi.org/10.1186/s13007-023-01086-y Open Access Abstract Background Genetic and genomic studies are seeing an increase in sample sizes together with a wider range of species investigated in response to environmental change concerns. In turn, these changes may come with chal- lenges including the time and difficulty to isolate nucleic acids (DNA or RNA), the sequencing cost and environmental impacts of the growing amount of plastic waste generated in the process. Pseudotsuga menziesii var. menziesii (Mirbel) Franco (PM), Tsuga heterophylla (Raf.) Sarg. (TH) and Thuja plicata Donn ex D.Don (TP) are conifer species found in diverse woodlands both as natives and naturalized exotics. Our study was carried out whilst investigating their genetics to understand their population structure and potential for adaptation. Results In the present study, we compared two different DNA isolation methods, i.e., spin-column DNeasy plant mini kit (QIAGEN), and temperature-driven enzymatic cocktail Plant DNA Extraction (MicroGEM). The quantity of recovered DNA and the quality of DNA were assessed along with the plastic footprint and time needed for three tree spe- cies. Both methods were optimised and proven to provide enough DNA for each studied species. The yield of DNA for each method depended on the species: QIAGEN showed higher yield in P. menziesii and T. heterophylla, while T. plicata recovered similar amount of DNA for both methods. The DNA quality was investigated using DNA barcoding techniques by confirming species identity and species discrimination. No difference was detected in the PCR amplifi- cation of the two barcoding loci, (rbcL and trnH-psbA), and the recovered sequences between DNA isolation methods. Measurement of the plastic use and the processing time per sample indicated that MicroGEM had a 52.64% lower plastic footprint and was 51.8% faster than QIAGEN. Conclusions QIAGEN gave higher yields in two of the species although both methods showed similar quality results across all species. However, MicroGEM was clearly advantageous to decrease the plastic footprint and improve the time efficiency. Overall, MicroGEM recovers sufficient and reliable DNA to perform common downstream analyses such as PCR and sequencing. Our findings illustrate the benefits of research and efforts towards developing more sustainable methods and techniques to reduce the environmental footprint of molecular analyses. Keywords DNA isolation, DNA quality, Plastic footprint, Time-efficient, Sustainability, Forest Trees, DNA Barcoding © 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/. 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. Laura Guillardín1* and John J. MacKay1 Laura Guillardín1* and John J. MacKay1 Background Genetic and genomic analyses often benefit from a large sample set to reliably detect associations and identify patterns relevant to key biological questions [1] due to higher statistical power and accuracy [2]. However, larger sample sizes require more resources, such as time to collect and process the samples and the volume of Correspondence: Laura Guillardín laura.guillardin@biology.ox.ac.uk 1 Department of Biology, University of Oxford, South Parks Road, Oxford OX1 3RB, United Kingdom Correspondence: Laura Guillardín laura.guillardin@biology.ox.ac.uk 1 Department of Biology, University of Oxford, South Parks Road, Oxford OX1 3RB, United Kingdom Another solution is to choose or develop protocols that use less plastic without compromising the experiment’s outcome [3, 25]. Several methods have been proposed to overcome the difficulties when isolating DNA from tree species. Broadly, there are five main types of DNA isolation sys- tems including organic extraction methods which use organic solvents like phenol and chloroform [26, 27], solid-phase extraction methods that use solid matrices, such as silica to bind and purify the DNA [28], precipi- tation methods which use salts and ethanol to precipi- tate the DNA [29], enzymatic digestion methods that use individual or a combination of enzymes to digest the samples and release the DNA [30] and use of magnetic bead-based coated with an agent which binds with DNA and isolates it from the cellular suspension [31], or mag- netic ionic liquids [32]. These methods are usually com- bined and modified depending on the experimenter’s needs [17, 33]. Molecular studies on non-model tree species may face challenges including obtaining high-quality DNA and RNA, accessing wild populations, the complexity and size of their genome, and the limited available genomic resources [13]. These factors increase the complexity and the resource intensity of genomic studies in trees compared to studies of many other organisms [14]. The challenges of obtaining high-quality DNA and RNA from trees include the difficulty of breaking down the cell walls to release the nucleic acids [15] and the inhibitory effects of secondary metabolites and polysaccharides on downstream applications such as PCR (polymerase chain reaction), sequencing and genotyping. Simple and robust protocols for isolation of high-quality nucleic acids from tree tissues that are suitable for downstream applications would help to overcome these challenges.h In this study, we compared the performance of two different DNA isolation methods in three conifer forest tree species: Pseudotsuga menziesii var. menziesii (Mir- bel) Franco (PM), Tsuga heterophylla (Raf.) Sarg. (TH) and Thuja plicata Donn ex D.Don (TP). They are all large evergreen coniferous trees native to western North America where they appear together in both mixed natu- ral forests and plantations. Coniferous species, includ- ing the study species, are widely found in Europe and North America in woodlands which provide a variety of ecosystem services including timber production, car- bon sequestration, biodiversity and habitat conservation, water regulation, recreation, and tourism [34, 35]. © 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/. 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. Guillardín and MacKay Plant Methods (2023) 19:111 Guillardín and MacKay Plant Methods (2023) 19:111 Page 2 of 11 Page 2 of 11 consumables used in the laboratory [3]. Analyses com- monly used include Genome-wide association studies (GWAS) [4], genomic selection (GS) and Genome envi- ronment association (GEA) to scan genomes for associa- tions with complex traits of interest as in tree breeding programs [5] where the power to detect loci with small effects may be constrained by the sample size [6]. The minimum sample size for each genetic analysis varies depending on the research question, the genetic markers used, the biological system and the available resources. Environmental change [7] and its effects on natural habi- tats as well as in the distribution, population size and genetic diversity [7–11] increases the number of species at risk and in need to be studied [10]. Therefore, both the range and the size of studies involving molecular analyses are increasing [12]. A different approach is to reuse specific plastic items, for example, Grenova Solutions [24] developed an on-bench Pipette Tip Cleaning Machine (TipNovus) that is easy to integrate into laboratory routines. Another solution is to choose or develop protocols that use less plastic without compromising the experiment’s outcome [3, 25]. A different approach is to reuse specific plastic items, for example, Grenova Solutions [24] developed an on-bench Pipette Tip Cleaning Machine (TipNovus) that is easy to integrate into laboratory routines. Addi- tionally, conifer trees are commonly used as pioneer spe- cies in reforestation projects in poorer sites due to their resilience to extreme environmental conditions [36]. At the same time, conifer species are affected by the change in the climate [12] and its subsequent increasing distur- bances such as the increase of forest wildfires and their intensity, longer and higher number of drought periods and pests and disease outbreaks. Therefore, conifers are largely studied organisms to shed light on the genomic mechanisms of adaptation, the association to different environments, the levels of population genetic diversity of different woodlands and the genetic resistance to pests & diseases, among others [12]. The main limitations when designing genomic experi- ments with large sample sizes include the time and difficulty to isolate nucleic acids (DNA or RNA), the sequencing cost and the environmental impact of plas- tic waste generated in the process. Sequencing platforms have reduced nucleic acid sequencing costs over the last two decades [16], making sequencing more acces- sible to researchers studying non-model species and feasible to study larger sample sizes. Some cost-effec- tive pipelines, workflows and strategies have also been developed to reduce costs and process time by reusing consumables [17]. Single-use plastic consumables are a significant source of waste in many scientific fields [18, 19], with an estimated use of 5.5 million tonnes per year worldwide [20]; therefore, although they are essential to perform genomic analyses, they have a large environ- mental impact [21]. Improved laboratory sustainability is encouraged by organizations and labels such as My Green Lab [22] and The Sustainable Laboratory Practices Work- ing Group (SLPWG) [23] which promote green prac- tices in the laboratory to reduce their plastic footprint. We tested the recovered DNA quantity, quality, plas- tic footprint and time of the following methods: (i) Plant DNA Extraction (MicroGEM International PLC, 2019) (MicroGEM) which uses a temperature-controlled enzy- matic cocktail and (ii) DNeasy plant mini kit (Qiagen Guillardín and MacKay Plant Methods (2023) 19:111 Page 3 of 11 USA, Valencia, CA) (QIAGEN), a silica column-based method widely used to isolate DNA from conifer trees. In this context, the aim of this study is to find an efficient DNA isolation method for forest trees by looking at the recovered DNA quality, the plastic waste generated, the energy required to produce it and the total time needed to process the samples using both DNA isolation meth- ods. Improvement of DNA isolation methods and comparison of species We isolated the DNA from four samples in each of the three study species by using MicroGEM and QIA- GEN original and improved protocols (see methods for details). The improvements to DNA isolation meth- ods increased the DNA recovered both for the Micro- GEM and QIAGEN methods in T. plicata, P. menziesii and T. heterophylla (Fig. 1). The data suggest that DNA yields were generally higher with QIAGEN, except for T. plicata where the results were similar across methods (Fig. 1B, D). However, statistical comparison of modi- fied methods and species showed significant effects of species and the interaction of species and methods but not between methods (Table 1). A post hoc Tukey test We verified the quality of the DNA recovered by use of PCR with gene-specific primers followed by DNA sequencing. among species within methods showed that P. menziesii yielded significantly better than the other species when using the original QIAGEN protocol, while T. plicata showed significantly better results when using the origi- nal MicroGEM method (Fig. 1A, C). The modified QIA- GEN showed P. menziesii and T. heterophylla grouping together and yielding significantly higher than T. plicata results while with MicroGEM modified protocol T. pli- cata showed significantly higher yields compared to the other species (Fig. 1B, D).i Overall, the modified procedures yielded more than 1 µg across all the species starting from a minimum of 10 mg when using MicroGEM and 20 mg with QIAGEN of dried tissue (except for TH4 extracted with Micro- GEM, Table 2), which is sufficient for many downstream analyses. Yields above 10 µg were obtained in P. men- ziesii and most of T. heterophylla but only for QIAGEN method (Table 2). The samples extracted with the modi- fied protocols of both DNA isolation methods will be used for the subsequent analyses. DNA recovered qualityh The quality of the DNA recovered from each method was analysed with standard DNA Barcoding techniques with commonly used gene-specific primers, the large subunit of the ribulose-bisphosphate carboxylase gene (rbcL) and the plastid trnH-psbA intergenic spacer. Fig. 1 DNA yield recovered by A using QIAGEN original protocol, B QIAGEN modified protocol, C MicroGEM original protocol and D modified MicroGEM protocol. TP: T. plicata, PM: P. menziesii and TH: T. heterophylla. Letters represent significant differences between species for each of the protocols tested Fig. 1 DNA yield recovered by A using QIAGEN original protocol, B QIAGEN modified protocol, C MicroGEM original protocol and D modified MicroGEM protocol. TP: T. plicata, PM: P. menziesii and TH: T. heterophylla. Letters represent significant differences between species for each of the protocols tested Fig. 1 DNA yield recovered by A using QIAGEN original protocol, B QIAGEN modified protocol, C MicroGEM original protocol and D modified MicroGEM protocol. TP: T. plicata, PM: P. menziesii and TH: T. heterophylla. Letters represent significant differences between species for each of the protocols tested Guillardín and MacKay Plant Methods (2023) 19:111 Page 4 of 11 Table 1 Anova table of factors affecting DNA yield recovered including methods, species and their interaction Sum Sq Sum of squares, Df Degrees of freedom, Pr Probability ’*’ corresponds to a P. value of 0.001 Anova table (Type III tests) Sum Sq Df F value Pr (> F) Significance (Intercept) 124434025.00 1 57.63 5.13E-07 * Method 126253.13 1 0.06 8.12E-01 Species 78713516.67 2 18.23 4.71E-05 * Method:Species 109300794.75 2 25.31 5.88E-06 * Residuals 38865031.50 18 ors affecting DNA yield recovered including methods, species and their interaction Table 2 DNA yield across species with improved methods Sample ID Conc.(ng/µl) MicroGEM Yield (µg) MicroGEM Conc.(ng/µl) QIAGEN Yield (µg) QIAGEN TH1 17.8 1.42 72.3 7.23 TH2 21.2 1.48 133.0 13.30 TH3 19.3 1.54 118.0 11.80 TH4 15.1 0.91 93.7 9.37 TP1 78.0 7.02 51.1 5.11 TP2 75.0 6.38 65.0 6.50 TP3 53.2 4.26 52.8 5.28 TP4 70.8 5.66 54.2 5.42 PM1 65.8 4.61 119.0 11.90 PM2 41.9 2.93 106.0 10.60 PM3 16.8 1.18 132.0 13.20 PM4 44.6 3.12 101.0 10.10 products, but samples extracted with MicroGEM show minor bands in 75% of the rbcL products compared to only 33.3% of QIAGEN samples (Fig. 2A, C). DNA recovered qualityh e 2 DNA yield across species with improved methods pr Table 2 DNA yield across species with improved methods Every trnH-psbA sequence retrieved was of the expected length amplicon size while 25% of the rbcL sequences were unreliable (Table 3). When looking at the differences between the DNA isolation methods, one QIAGEN sample and two MicroGEM samples could not be sequenced satisfactorily for rbcL. We submitted the rbcL and trnH-psbA recovered sequences to BOLD and GENBANK, respectively to verify the species identities of the tested samples. All sequences returned the correct species identification, except for rbcL in samples whose sequence length was shorter than the expected size (Table 4). Two of the problematic sequences were from samples isolated with MicroGEM (T. plicata and P. menziesii) and one from QIAGEN (P. menziesii) (Table 4). Major amplicons of the expected size (Table 8) were detected in all the samples tested for both barcodes rbcL and trnH-psbA (Table 3, Fig. 2). Other minor bands appeared in the rbcL PCR products as displayed in the gel images while none were detected in the trnH- psbA gel (Fig. 2, Table 3). There were no differences between DNA isolation methods in trnH-psbA PCR The sequences obtained of the trnH-psbA ampli- cons were clustered using USEARCH v.11 60 to assess the species discrimination efficacy of both methods. A clustering was determined for the three species and the samples were classified in species-specific clusters with a level of similarity above 99% in every case, without any difference between DNA isolation methods. ummary of PCR and Sequencing results for each DNA isolation method per species and barcode marker Table 3 Summary of PCR and Sequencing results for each DNA isolation method per species and barcode marker PCR Sequencing rbcL trnH-psbA rbcL trnH-psbA DNA isolation Method Species N Major amplicon detected % Minor bands detected % Major amplicon detected % Minor bands detected % N Expected seq. length % Expected seq. length % QIAGEN TP 4 100 50 100 0 2 100 100 PM 100 50 100 0 50 100 TH 100 0 100 0 100 100 MicroGEM TP 100 75 100 0 50 100 PM 100 100 100 0 50 100 TH 100 50 100 0 100 100 Page 5 of 11 Guillardín and MacKay Plant Methods (2023) 19:111 Fig. 2 rbcL PCR product for A QIAGEN B MicroGEM samples and trnH-psbA for C QIAGEN and D MicroGEM samples. Plastic footprint and time needed for DNA isolationsh The plastic consumed was determined for all steps used in each method, including the clean-up step using AMPureXP beads needed after the DNA isolation using MicroGEM (Fig. 3). To extract one sample, 12.51 g of plastic was used with QIAGEN and 5.92 g with Micro- GEM + AMPureXP beads from both tube and tip items (Fig. 3). Overall, MicroGEM required 52.64% less plas- tic than QIAGEN to isolate DNA, per sample. The plas- tic footprint was determined for both CO2 emissions DNA recovered qualityh Sample names: M: 1 Kb molecular-weight size marker, 1:TH1, 2:TH2, 3:TH3, 4:TH4, 5:TP1, 6:TP2, 7:TP3, 8:TP4, 9:PM1, 10:PM2, 11:PM3, 12:PM4, ntc: non-template control, m: 100 bp molecular size marker Fig. 2 rbcL PCR product for A QIAGEN B MicroGEM samples and trnH-psbA for C QIAGEN and D MicroGEM samples. Sample names: M: 1 Kb molecular-weight size marker, 1:TH1, 2:TH2, 3:TH3, 4:TH4, 5:TP1, 6:TP2, 7:TP3, 8:TP4, 9:PM1, 10:PM2, 11:PM3, 12:PM4, ntc: non-template control, m: 100 bp molecular size marker Fig. 2 rbcL PCR product for A QIAGEN B MicroGEM samples and trnH-psbA for C QIAGEN and D MicroGEM samples. Sample names: M: 1 Kb molecular-weight size marker, 1:TH1, 2:TH2, 3:TH3, 4:TH4, 5:TP1, 6:TP2, 7:TP3, 8:TP4, 9:PM1, 10:PM2, 11:PM3, 12:PM4, ntc: non-template control, m: 100 bp molecular size marker was lower and slightly more variable when using Micro- GEM than QIAGEN (Table 5). Table 5 Summary table of the DNA yield recovered in a larger project when using both DNA isolation methods CV Coefficient variation, N Number of samples Method Species Yield Mean (µg) Yield CV % N QIAGEN TP 9.603 59.24 79 PM 7.766 40.37 1146 MicroGEM TP 3.312 66.09 472 Table 5 Summary table of the DNA yield recovered in a larger project when using both DNA isolation methods Table 4 Species identification using GENEBANK and BOLD datasets for both barcodes N Number of samples DNA isolation method Species N BOLD rbcL GENBANK trnH-psbA QIAGEN TP 2 2 2 PM 2 1 2 TH 2 2 2 MicroGEM TP 2 1 2 PM 2 1 2 TH 2 2 2 Table 4 Species identification using GENEBANK and BOLD datasets for both barcodes was lower and slightly more variable when using Micro- GEM than QIAGEN (Table 5). was lower and slightly more variable when using Micro- GEM than QIAGEN (Table 5). Discussionh The central question posed in this study was whether a rapid and plastic-efficient DNA isolation method will recover reliable DNA similarly to a commonly used kit. We compared the DNA yield, DNA quality and effi- ciency on four samples of three different forest tree spe- cies and considered outputs of a separate study in a larger population. Fig. 3 Total plastic required to process one sample by QIAGEN (blue) and MicroGEM (red). The clean-up step plastic use (MicroGEM only) hatched and energy consumption to produce the required plas- tic to process a single sample. The data showed that the plastic footprint was at least 50% lower with Micro- GEM (Table 6). p p We presented data on the use of two methods to iso- late nucleic acids from forest trees including method optimisations to the basic protocols to isolate the DNA compared to the original methods. No significant dif- ference in DNA yield recovered was found between the two methods when using the improved proto- cols, indicating that either method will deliver suf- ficient DNA yield for standard molecular analyses. However, significant differences were found among species, which suggested the need for specific optimi- sation of laboratory protocols when looking at various species. The significant difference found in the inter- action between methods and species suggests that P. menziesii and T. heterophylla recovered more DNA when using QIAGEN, but T. plicata recovered similar amounts in both cases. Several studies looked at the differences in amount of recovered DNA quality and time efficiency of different DNA isolation methods in plant species [37]. Bashalkhanov and Rajora [38] tested several DNA extraction systems suitable for conifers obtaining a mean DNA yield of more than 10 µg when using the QIAGEN DNeasy kit, which aligns with our DNA yield results. They also concluded that the QIA- GEN DNeasy kit is not a preferred method when deal- ing with large numbers of samples due to the long time that is required to complete the isolation. Our initial MicroGEM DNA concentration results were similar to previous research such as described by Ryan et al. [39], which showed a very low recovered DNA concentration (less than 3 ng/µl) when using MIGROGEM on differ- ent plant tissues. After optimisation, we substantially increased the concentration to over 15 ng/µl and up to 78 ng/µl using MicroGEM. DNA obtained on large populations Populations with larger numbers of individuals have been processed in two of the species described above with both methods in a parallel project (Table 5). The QIAGEN method gave higher mean yields for T. pli- cata than P. menziesii with the latter also being more consistent as shown by lower levels of variation (CV) (Table 5). The yield recovered from T. plicata samples Guillardín and MacKay Plant Methods (2023) 19:111 Page 6 of 11 Fig. 3 Total plastic required to process one sample by QIAGEN (blue) and MicroGEM (red). The clean-up step plastic use (MicroGEM only) hatched Table 7 Time required to process one sample by using each of the DNA isolation methods Table 7 Time required to process one sample by using each of the DNA isolation methods Total (min) Total hands-on (min) QIAGEN 119.4 5.4 MicroGEM 57.5 3.5 Discussionh These results confirmed the utility of optimising protocols to obtain sufficient DNA to perform reliable downstream analyses. It is essential to acknowledge the relatively low number of samples We also measured the time required to isolate DNA with each method by determining the time needed to process 24 samples and calculated for a single sample (see methods for details). Overall, the time required to extract a single sample using MicroGEM was 51.8% less than QIAGEN (Table 7). If we only include the hands- on time needed to purify the DNA of one sample by using both methods, then MicroGEM needed 34.6% less time than QIAGEN (Fig. 4).h The plastic consumed to isolate 1146 P. menziesii and 79 T. plicata samples from the population pro- ject described above (see Table 5) by using the DNA isolation QIAGEN method was 15.3 kg, which repre- sented 52.06 kg of CO2 emitted, and 1310 Mj of energy required to produce the plastic used. If MicroGEM was used to isolate the DNA of this project rather than QIAGEN, 8.1 kg less plastic would have been utilised, 27.44 kg of CO2 emission would have been avoided and 686 Mj of energy would have been saved. Based on an 8-h working day, 25.9 days were needed to complete the 1225 isolations using QIAGEN while only 14.7 days would have been needed to isolate the same samples if using MicroGEM. Table 6 Plastic footprint of the DNA isolation methods to process one sample Carbon emissions (Kg CO2) Energy used (Mj) QIAGEN 0.04255 1.07498 MicroGEM 0.02011 0.50809 Table 6 Plastic footprint of the DNA isolation methods to process one sample Guillardín and MacKay Plant Methods (2023) 19:111 Page 7 of 11 in our dataset could limitate and impact our findings. species may provide further validation and enhance the Fig. 4 Diagram of the methodology followed. Illustration created with BioRender.com. The methodology and sequence of major steps implemented across this study, included: 1) plant material collection, 2) DNA isolation, 3) DNA yield quantification and 4) DNA quality assessment by using DNA barcoding for species identification and species discrimination by performing PCR and followed by DNA sequencing Fig. 4 Diagram of the methodology followed. Illustration created with BioRender.com. Discussionh and power of discrimination by PCR amplification and sequencing [42, 43]. For trnH-psbA, both DNA isola- tion methods delivered accurate size amplicons and no minor bands which match with Kress et al. [44] results, where trnH-psbA exhibited the highest PCR success. The sequences from rbcL and trnH-psbA markers were sub- mitted to BOLD and GenBank, respectively, to confirm sample identities and the taxonomical assignment when using trnH-psbA was 100% successful, which relates to what Loera-Sánchez et al. [43] found. Those sequences of unexpected length were the only ones incorrectly iden- tified. In the species discrimination analysis with the trnH-psbA, all samples from the same species clustered together, without any differences between the DNA iso- lation method. Several studies have validated these two loci for efficient DNA barcoding in plants [43, 44], and specifically conifers [41]. In their study, Armenise et al. [41] recovered the same expected fragment size and expected sequence length as our results when analysing P. menziesii using rbcL (710 bp) and trnH-psbA (565 bp) markers, showing consistency in the amplifications. Their findings suggest that combining these two markers may be preferable to perform DNA Barcoding in conifers due to their PCR uniformity and sufficient sequence quality while showing enough variation to perform species iden- tity analyses at the genus level. In contrast, our results suggest that using trnH-psbA alone retrieves sufficient evidence to identify and differentiate species from dif- ferent genera which corresponds with Kress et al. [44] results. plastic used in a molecular laboratory after promoting very specific behavioural and protocol changes to reduce the use of plastic items. They were able to achieve a reduction of more than 10 kg of plastic per week on aver- age, from an initial of 24 kg. The improved protocol for GBS library preparation proposed by Torkamaneh et al. [25] reduces both the time and plastic needed by 75% and 89%, respectively, compared to standard methods. These findings, together with our results, confirm the poten- tial to apply sustainable measures such as using alterna- tive materials, reusing items when possible or developing protocols which use less plastic without compromising the experiment’s outcome. Our study comparing different methods shows the potential to significantly impact on the plastic consump- tion and efficiency of DNA isolation in coniferous for- est tree species. Discussionh Our findings suggest that MicroGEM is a highly suitable method as it provided sufficient DNA yield with good quality while producing the least amount of plastic waste and being the most time-efficient. Our study also shows that selecting the most suitable method will depend on the specific requirements of the pro- ject, the species studied and the resources available. We believe that our results will encourage researchers to select DNA isolation methods based on sustainable labo- ratory practices, although further research is needed to explore the performance of this method on a broader range of species and molecular analyses. Additional research to assess other DNA isolation methods and their potential to reduce plastic use is also needed due to the importance of increasing the studies that evaluate the environmental impact of molecular laboratories. To assess the plastic and time efficiency of each DNA isolation method, we reported the plastic footprint and the time needed to process one sample. The results of this study indicate that a temperature-driven enzy- matic cocktail isolation system reduces plastic footprint by 52.64% compared to a commonly used silica-based nucleic acid isolation method. This enzymatic method also reduces the average time needed to process a sample by 51.8%. When looking at the plastic footprint, we found that there is limited research on the plastic footprint that compares plastic consumption of different isolation nucleic acids methods. Marengo et al. [45] developed a DNA isolation method for plant species where they reduced the sample processing time while maintaining the quality of the DNA recovered. The authors also dis- cussed the reduction of plastic waste achieved although this was not quantified. The lack of previous research on this matter prevents a direct comparison with our DNA isolation plastic footprint results. Nonetheless, there is an increase of studies looking at how to reduce general plastic use in laboratories [3, 19] and proposing proto- cols optimised to decrease the amount of plastic use [25]. Alves et al. [3] developed a 7-week study measuring the Conclusion In conclusion, the main result of this study is that this rapid and plastic-efficient method, Plant DNA Extrac- tion, MicroGEM, recovers sufficient and reliable DNA to perform common downstream analyses such as PCR and sequencing, and performs as well as a commonly used spin-column kit, DNeasy plant mini kit, QIAGEN. Our study highlights the merit of efforts towards developing more sustainable and efficient laboratory practices in the field of molecular biology. A reduction of plastic waste in molecular laboratories deserves further research for the development of techniques and protocols that use less single-use plastic items. Discussionh The methodology and sequence of major steps implemented across this study, included: 1) plant material collection, 2) DNA isolation, 3) DNA yield quantification and 4) DNA quality assessment by using DNA barcoding for species identification and species discrimination by performing PCR and followed by DNA sequencing species may provide further validation and enhance the generalisation of our observations.h in our dataset could limitate and impact our findings. However, similar sample sizes were used in compara- ble studies evaluating a novel DNA isolation method [40]. They found significant trends in different species with datasets of four samples per species. Nonetheless, future tests with larger sample sizes and including more The quality of the DNA was analysed with standard DNA Barcoding techniques with gene-specific primers as done in Armenise et al. [41] using rbcL and trnH-psbA which are commonly used to confirm the species identity Guillardín and MacKay Plant Methods (2023) 19:111 Page 8 of 11 Page 8 of 11 and power of discrimination by PCR amplification and sequencing [42, 43]. For trnH-psbA, both DNA isola- tion methods delivered accurate size amplicons and no minor bands which match with Kress et al. [44] results, where trnH-psbA exhibited the highest PCR success. The sequences from rbcL and trnH-psbA markers were sub- mitted to BOLD and GenBank, respectively, to confirm sample identities and the taxonomical assignment when using trnH-psbA was 100% successful, which relates to what Loera-Sánchez et al. [43] found. Those sequences of unexpected length were the only ones incorrectly iden- tified. In the species discrimination analysis with the trnH-psbA, all samples from the same species clustered together, without any differences between the DNA iso- lation method. Several studies have validated these two loci for efficient DNA barcoding in plants [43, 44], and specifically conifers [41]. In their study, Armenise et al. [41] recovered the same expected fragment size and expected sequence length as our results when analysing P. menziesii using rbcL (710 bp) and trnH-psbA (565 bp) markers, showing consistency in the amplifications. Their findings suggest that combining these two markers may be preferable to perform DNA Barcoding in conifers due to their PCR uniformity and sufficient sequence quality while showing enough variation to perform species iden- tity analyses at the genus level. In contrast, our results suggest that using trnH-psbA alone retrieves sufficient evidence to identify and differentiate species from dif- ferent genera which corresponds with Kress et al. [44] results. DNA analyses Th l f The quality of the DNA was analysed by PCR amplifica- tion (Fig. 4) with the intergenic spacer primers situated on the chloroplast DNA: the large subunit of the ribu- lose-bisphosphate carboxylase gene (rbcL) and the plastid trnH-psbA, and sequencing on each sample. We adjusted the DNA concentrations to 20 ng/µl, when possible (Table 2), for PCR amplification of rbcL and trnH-psbA barcoding loci (Table 8). We used Q5® High-Fidelity 2X Master Mix DNA Polymerase (NewEngland Biolabs®, Ipswich, MA, USA) and added 100 ng of template DNA and 0.5 µM of each primer into a 25 µl final reaction fol- lowing the manufacturer’s instructions. The thermocy- cler conditions were 98 °C for 20 s, followed by 40 cycles starting at 98 °C for 10 s, TA (rbcL: 54 °C and trnH-psbA: 63 °C, Table 8) for 30 s and 72 °C for 30 s, with a final extension at 72 °C for 2 min. The amplified regions were visualized by electrophoresis in a 1.8% Agarose gel with SYBR safe (Invitrogen™, Thermo Fisher Scientific, Cleve- land, OH, USA) as stain. We processed the same 12 samples using the DNA iso- lation method Plant DNA Extraction [48] (MicroGEM) (Fig. 4) following the manufacturer’s original instructions. First, 10 mg of leaf tissue is mixed with the lysis buffer and using a mechanical homogenizer to break the cells, second, the lysate is transferred to a PDQeX DNA-bind- ing cartridge and third, the enzymatic cocktail is added to the cartridge, which is inserted into the PDQeX device which through the combination of changes in tempera- ture and the physical design of the cartridge performs a chemical process that releases the DNA into the recovery tubes. We also modified this procedure to optimise the DNA yield as follows: (1) The tissue was disrupted alone and added with a micro scoop to the pre-sample mix which included the lysate buffer and the enzymatic cock- tail in individual tubes, (2) the sample mix was pipetted into the cartridges, which were inserted into the PDQex device with a modified plant programme with an incuba- tion time increased from 5 to 15 min and extraction time from 5 to 10 min [49]. PCR products of two samples per species for each of the extraction methods were analysed by Sanger sequencing (Source Biosciences). Pre-cleaning and post- quality checks were performed by the sequencing service facilitator. Plant materialh The study species are the conifers Thuja plicata (TP), Pseudotsuga menziesii (PM) and Tsuga heterophylla (TH). They are all part of the Pinales order; T. plicata belongs to the Cupressaceae, whilst P. menziesii and T. Page 9 of 11 Page 9 of 11 Guillardín and MacKay Plant Methods (2023) 19:111 Guillardín and MacKay Plant Methods (2023) 19:111 heterophylla belong to the Pinaceae. Four samples per species (Fig. 4) were retrieved from different planted stands in England using an arborist slingshot following a modified method based on Youngentob et al. [46]. The leaves collected were dried by storing them in silica gel beads after removing them from the tree and until pro- cessing for DNA isolation. DNA recoveryh The DNA yield recovered from all samples was quantified using Qubit® Fluorometer v.4.0 (Invitrogen™, Thermo Fisher Scientific, Cleveland, OH, USA) (Fig. 1). The sam- ples extracted by the modified protocols were used in this study (for details, see Table 2). DNA quality of these sam- ples was visualized by electrophoresis in 0.8% Agarose gels (TAE Buffer), which confirmed the DNA presence (not shown). We performed an ANOVA test (error type iii) using the R [51] package ’ape’ [52] to analyse the dif- ferences in DNA yield recovered between methods and species followed by a post-hoc Tukey’s test [53] for com- paring all possible group pairings. To visualise the data, we used the R package ’ggplot2’ [54]. DNA isolation procedures and improvements We isolated the DNA using 20 mg of leaf tissue from the 12 samples (Fig. 4) by using the manufacturer’s instruc- tions for the DNeasy plant mini kit (Qiagen USA, Valen- cia, CA) (QIAGEN), we also modified the procedure to enhance the quality and yield of the DNA recovered, by adding 20 µl of Protease K in the lysate step and increas- ing the incubation time to a minimum of an hour at 55 °C. [47]. References Independent and Joint-GWAS for growth traits in Eucalyptus by assembling genome-wide data for 3373 individuals across four breeding populations. New Phytol. 2019;221:818–33. 6. Tan B, Ingvarsson PK. Integrating genome-wide association mapping of additive and dominance genetic effects to improve genomic prediction accuracy in Eucalyptus. Plant Genome. 2022;15:e20208. 7. Brown SC, Wigley TML, Otto-Bliesner BL, Rahbek C, Fordham DA. Persistent quaternary climate refugia are hospices for biodiversity in the anthropocene. Nat Clim Chang. 2020;10:244–8. 8. Pauls SU, Nowak C, Bálint M, Pfenninger M. The impact of global climate change on genetic diversity within populations and species. Mol Ecol. 2013;22:925–46. 9. Heuertz M, Carvalho SB, Galindo J, Rinkevich B, Robakowski P, Aavik T, et al. The application gap: genomics for biodiversity and ecosystem Table 9 Weight of plastic items required, and number of items needed per DNA isolation method to process one sample N number of items Plastic item Weight (g) QIAGEN (N) MicroGEM (N) Cartridge 1.183 0.00 1.00 2 ml tube 1.089 1.00 1.00 1.5 mL tube 0.940 1.00 0.00 2 mL no-Lid tube 0.932 3.00 0.00 Column 0.734 2.00 0.00 0.5 mL tube 0.461 0.00 1.00 0.2 mL tube 0.153 0.00 1.00 1000 μl tip 1.033 5.13 0.17 200 μl tip 0.315 2.04 0.08 20 μl tip 0.219 0.04 0.00 10 μl tip 0.275 1.00 0.00 cut tip 0.311 0.00 1.00 Total 15.21 5.25 2. Hong EP, Park JW. Sample size and statistical power calculation in genetic association studies. Genomics Inform. 2012;10:117. Table 9 Weight of plastic items required, and number of items needed per DNA isolation method to process one sample N number of items Plastic item Weight (g) QIAGEN (N) MicroGEM (N) Cartridge 1.183 0.00 1.00 2 ml tube 1.089 1.00 1.00 1.5 mL tube 0.940 1.00 0.00 2 mL no-Lid tube 0.932 3.00 0.00 Column 0.734 2.00 0.00 0.5 mL tube 0.461 0.00 1.00 0.2 mL tube 0.153 0.00 1.00 1000 μl tip 1.033 5.13 0.17 200 μl tip 0.315 2.04 0.08 20 μl tip 0.219 0.04 0.00 10 μl tip 0.275 1.00 0.00 cut tip 0.311 0.00 1.00 Total 15.21 5.25 Table 9 Weight of plastic items required, and number of items needed per DNA isolation method to process one sample 3. Alves J, Sargison FA, Stawarz H, Fox WB, Huete SG, Hassan A, et al. A case report: insights into reducing plastic waste in a microbiology laboratory. Access Microbiol. 2021. https://doi.org/10.1099/acmi.0.000173. 4. Availability of data and materials Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. The carbon dioxide emitted to produce 1 kg of polypro- pylene plastic commonly used to make laboratory sup- plies has been discussed by Harding et al. [61] and 3.4 kg CO2 is generally accepted amount. The total energy required to produce 1 kg of plastic from the extraction of raw materials to the final manufactured product is 85.9 MJ [61]. Following these data, we calculated the kg of CO2 emissions and total energy required to process one sample with each DNA isolation method based on the amount of plastic needed in each case. References 1. Fumagalli M. Assessing the effect of sequencing depth and sample size in population genetics inferences. PLoS ONE. 2013;8:e79667. 2. Hong EP, Park JW. Sample size and statistical power calculation in genetic association studies. Genomics Inform. 2012;10:117. 3. Alves J, Sargison FA, Stawarz H, Fox WB, Huete SG, Hassan A, et al. A case report: insights into reducing plastic waste in a microbiology laboratory. Access Microbiol. 2021. https://doi.org/10.1099/acmi.0.000173. 4. Visscher PM, Wray NR, Zhang Q, Sklar P, McCarthy MI, Brown MA, et al. 10 years of GWAS discovery: biology, function, and translation. Am J Human Genetics. 2017;101:5–22. 5. Müller BSF, de Almeida Filho JE, Lima BM, Garcia CC, Missiaggia A, Aguiar AM, et al. Independent and Joint-GWAS for growth traits in Eucalyptus by assembling genome-wide data for 3373 individuals across four breeding populations. New Phytol. 2019;221:818–33. 6. Tan B, Ingvarsson PK. Integrating genome-wide association mapping of additive and dominance genetic effects to improve genomic prediction accuracy in Eucalyptus. Plant Genome. 2022;15:e20208. 7. Brown SC, Wigley TML, Otto-Bliesner BL, Rahbek C, Fordham DA. Persistent quaternary climate refugia are hospices for biodiversity in the anthropocene. Nat Clim Chang. 2020;10:244–8. 8. Pauls SU, Nowak C, Bálint M, Pfenninger M. The impact of global climate change on genetic diversity within populations and species. Mol Ecol. 2013;22:925–46. 9. Heuertz M, Carvalho SB, Galindo J, Rinkevich B, Robakowski P, Aavik T, et al. The application gap: genomics for biodiversity and ecosystem 1. Fumagalli M. Assessing the effect of sequencing depth and sample size in population genetics inferences. PLoS ONE. 2013;8:e79667. 1. Fumagalli M. Assessing the effect of sequencing depth and sample size in population genetics inferences. PLoS ONE. 2013;8:e79667. 1. Fumagalli M. Assessing the effect of sequencing depth and sample size in population genetics inferences. PLoS ONE. 2013;8:e79667. 2. Hong EP, Park JW. Sample size and statistical power calculation in genetic association studies. Genomics Inform. 2012;10:117. 3. Alves J, Sargison FA, Stawarz H, Fox WB, Huete SG, Hassan A, et al. A case report: insights into reducing plastic waste in a microbiology laboratory. Access Microbiol. 2021. https://doi.org/10.1099/acmi.0.000173. 4. Visscher PM, Wray NR, Zhang Q, Sklar P, McCarthy MI, Brown MA, et al. 10 years of GWAS discovery: biology, function, and translation. Am J Human Genetics. 2017;101:5–22. 5. Müller BSF, de Almeida Filho JE, Lima BM, Garcia CC, Missiaggia A, Aguiar AM, et al. DNA analyses Th l f Retrieved sequences were visualized and the electropherograms manually checked by using SnapGene software [55]. We exported the fasta files from the ab1 files which were trimmed with Trimmomatic [56] and aligned with MUSCLE [57]. After recovering the DNA we performed a clean-up step by using the AMPpureXP beads [50], which use magnetic particles that bind the DNA to allow the clean- up. A concentration of 0.8 × of AMPureXP beads (Beck- man Coulter™, Brea, CA, USA) was used for the size selection needed to retrieve the isolated DNA fragments (0.8 × ratio recommended by NewEngland Biolabs). Table 8 Locus and primer details for PCR. TA: Annealing temperature used Marker name Primer’s sequence TA (°C) Expected product size (bp) Source trnH-psbA CGCGCATGGTGGATTCACAATCC 63 500 [41] GTTATGCATGAACGTAATGCTC rbcL ATGTCACCACAAACAGAAAC 54 750 [41] TCGCATGTACCTGCAGTAGC Guillardín and MacKay Plant Methods (2023) 19:111 Guillardín and MacKay Plant Methods (2023) 19:111 Page 10 of 11 We submitted the rbcL trimmed sequences to The Barcode of Life Data system v4 (BOLD) [58] and the trnH-psbA to the nucleotide dataset from GenBank [59]. BOLD enables a species-level identification of plant taxa by submitting queries of MatK and rbcL sequences to be searched against their reference library through their sequence threshold identification system. We verified the species discrimination efficacy of the trnH-psbA barcode by clustering the aligned sequences together by similar- ity using the software USEARCH v.11 [60]. The UCLUST algorithm divides a set of sequences into clusters provid- ing a high-throughput species-level discrimination. The minimum similarity level was set to 99%. fixed and remain the same for either 1 or 24 samples. We divided the hands-on time by 24 samples and added it to the measured fixed time. Acknowledgements The authors acknowledge the forestry estates where the samples came from: Stourhead (Western), Longleat (Wiltshire) and Bagley Woods. Additionally, we thank Caterina Branca for technical advice on MicroGem methodology. We also thank Barley Rose Collier Harris for sampling support and Tin Hang (Henry) Hung for DNA Barcoding technical advice. Author contributions LG drafted the manuscript and both LG and JM revised the main manuscript and LG conceived and planned the experiments, performed laboratory and statistical analyses and prepared the figures. JM supervised the project. Funding LG received financial support from the Oxford-John Oldacre Graduate Scholar- ship and from Mr Henry Hoare. The project was funded in part by the OxLEP fund for the Oxford Plant Sciences Centre for Innovation. Received: 12 May 2023 Accepted: 29 September 2023 The time required to isolate DNA by using both modi- fied methods was timed for each step in the procedures when processing 24 samples and calculated for a single reaction. Incubation time and centrifuge steps time are Ethics approval and consent to participate Not applicable. Ethics approval and consent to participate Not applicable. Plastic footprint and time needed for DNA isolationsh The plastic used to isolate DNA with both modified methods was calculated for a single reaction by weighing every plastic item with a precision balance (Table 9) not including the packaging.h N number of items Competing interests The authors declare no competing interests. Received: 12 May 2023 Accepted: 29 September 2023 References Visscher PM, Wray NR, Zhang Q, Sklar P, McCarthy MI, Brown MA, et al. 10 years of GWAS discovery: biology, function, and translation. Am J Human Genetics. 2017;101:5–22. 5. Müller BSF, de Almeida Filho JE, Lima BM, Garcia CC, Missiaggia A, Aguiar AM, et al. Independent and Joint-GWAS for growth traits in Eucalyptus by assembling genome-wide data for 3373 individuals across four breeding populations. New Phytol. 2019;221:818–33. 9. Heuertz M, Carvalho SB, Galindo J, Rinkevich B, Robakowski P, Aavik T, et al. The application gap: genomics for biodiversity and ecosystem Guillardín and MacKay Plant Methods (2023) 19:111 Page 11 of 11 Page 11 of 11 service management. Biol Conserv. 2023. https://doi.org/10.1016/j. biocon.2022.109883. service management. Biol Conserv. 2023. https://doi.org/10.1016/j. biocon.2022.109883. 37. Pipan B, Zupančič M, Blatnik E, Dolničar P, Meglič V. Comparison of six genomic DNA extraction methods for molecular downstream applications of apple tree (Malus X domestica ). Cogent Food Agric. 2018;4:1540094. 10. Harley CDG. Climate change, keystone predation, and biodiversity loss. Science. 1979;2011(334):1124–7. 38. Bashalkhanov S, Rajora OP. Protocol: a high-throughput DNA extraction system suitable for conifers. Plant Methods. 2008;4:1–6. 11. Garden JG, O’Donnell T, Catterall CP. Changing habitat areas and static reserves: challenges to species protection under climate change. Landsc Ecol. 2015;30:1959–73. 39. Ryan AL, O’Hern CP, Elkins KM. Evaluation of two new methods for DNA extraction of “Legal High” plant species. J Forensic Sci. 2020;65:1704–8. 12. Alberto FJ, Aitken SN, Alía R, González-Martínez SC, Hänninen H, Kremer A, et al. Potential for evolutionary responses to climate change—evi- dence from tree populations. Glob Chang Biol. 2013;19:1645. 40. Moeller JR, Moehn NR, Waller DM, Givnish TJ. Paramagnetic cellulose DNA isolation improves DNA yield and quality among diverse plant taxa. Appl Plant Sci. 2014;2:10. 13. Holliday JA, Aitken SN, Cooke JEK, Fady B, Gonz Alez-Martinez SC, Heuertz M, et al. Advances in ecological genomics in forest trees and applications to genetic resources conservation and breeding. Mol Ecol. 2017;26:706–17. 41. Armenise L, Simeone MC, Piredda R, Schirone B. Validation of DNA barcoding as an efficient tool for taxon identification and detection of species diversity in Italian conifers. Eur J For Res. 2012;131:1337–53. 14. Abril N, Gion JM, Kerner R, Müller-Starck G, Cerrillo RMN, Plomion C, et al. Proteomics research on forest trees, the most recalcitrant and orphan plant species. Phytochemistry. 2011;72:1219–42. 42. Hollingsworth PM, Forrest LL, Spouge JL, Hajibabaei M, Ratnasingham S, van der Bank M, et al. References A DNA barcode for land plants. Proc Natl Acad Sci U S A. 2009;106:12794–7. y y 15. Rezadoost MH, Kordrostami M, Kumleh HH. An efficient protocol for isolation of inhibitor-free nucleic acids even from recalcitrant plants. 3 Biotech. 2016;6:1–7. 43. Loera-Sánchez M, Studer B, Kölliker R. DNA barcode trnH-psbA is a prom- ising candidate for efficient identification of forage legumes and grasses. BMC Res Notes. 2020;13:1–6. 16. Levy SE, Boone BE. Next-generation sequencing strategies. Cold Spring Harb Perspect Med. 2019. https://doi.org/10.1101/cshperspect.a025791. 44. Kress WJ, Erickson DL. A two-locus global dna barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS ONE. 2007;2:e508. 17. Ze-Yu FU, Jian-Cheng SONG, Jameson PE. A rapid and cost effective pro- tocol for plant genomic DNA isolation using regenerated silica columns in combination with CTAB extraction. J Integr Agric. 2017;16:1682–8. 45. Marengo A, Cagliero C, Sgorbini B, Anderson JL, Emaus MN, Bicchi C, et al. Development of an innovative and sustainable one-step method for rapid plant DNA isolation for targeted PCR using magnetic ionic liquids. Plant Methods. 2019;15:1–11. 18. Sawyer A. The unsustainable lab. Biotechniques. 2019;66:5–7. 19. Howes L. Can laboratories move away from single-use plastic? ACS Cent Sci. 2019;5:1904–6. 46. Youngentob KN, Zdenek C, van Gorsel E. A simple and effective method to collect leaves and seeds from tall trees. Methods Ecol Evol. 2016;7:1119–23. 20. Urbina MA, Watts AJR, Reardon EE. Labs should cut plastic waste too. Nature. 2015;528:479–479. 21. Di Paolo L, Abbate S, Celani E, Di Battista D, Candeloro G. Carbon footprint of single-use plastic items and their substitution. Sustainability. 2022;14:16563. 47. Guillardín L. QIAGEN D Neasy Plant Mini Kit DNA extraction modified protocol—Douglas fir. Website. https://lauguillardin.github.io/2022-07- 12-microGEM-DNAextraction_Thujaplicata/. Accessed 27 Mar 2023. 22. My Green Lab. https://www.mygreenlab.org/. Accessed 7 May 2023. 48. MicroGEM International PLC. Microgem: nucleic acid extraction: extremely fast, extremely simple, extremely effective. Genetic Eng Bio- technol News. 2019;39:S15–S15.i 23. The Sustainable Laboratory Practices Working Group (SLPWG). https:// nems.nih.gov/greening-tools/Pages/Sustainable-Labs.aspx. Accessed 7 May 2023. 49. Guillardin L. microGEM DNA extraction modified protocol—Western redcedar. Website. https://lauguillardin.github.io/2022-07-12-microGEM- DNAextraction_Thujaplicata/. Accessed 27 Mar 2023. y 24. Grenova Solutions. https://grenovasolutions.com/. Accessed 7 May 2023. DNAextraction_Thujaplicata/. Accessed 27 Mar 2023. 25. Torkamaneh D, Boyle B, St-Cyr J, Légaré G, Pomerleau S, Belzile F. NanoGBS: a miniaturized procedure for GBS library preparation. Front Genet. 2020. https://doi.org/10.3389/fgene.2020.00067. 50. Deangelis MM, Wang DG, Hawkins TL. Solid-phase reversible immobiliza- tion for the isolation of PCR products. References Search and clustering orders of magnitude faster than BLAST. Bioinform Appl Note. 2010;26:2460–1. 34. Mina M, Bugmann H, Cordonnier T, Irauschek F, Klopcic M, Pardos M, et al. Future ecosystem services from European mountain forests under climate change. J Appl Ecol. 2017;54:389–401. 61. Harding KG, Dennis JS, von Blottnitz H, Harrison STL. Environmental analysis of plastic production processes: Comparing petroleum-based polypropylene and polyethylene with biologically-based poly-β- hydroxybutyric acid using life cycle analysis. J Biotechnol. 2007;130:57–66. 35. Bastien J-C. Douglas-fir biomass production and carbon sequestration. 4.2. Douglas-fir—an option for Europe. In: European Forest Institute, What Science can tell us. References Nucleic Acids Res. 1995;23:4742. 26. Green MR, Sambrook J. Isolation of high-molecular-weight DNA using organic solvents. Cold Spring Harb Protoc. 2017. https://doi.org/10.1101/ pdb.prot093450. 51. R Core Team. R: A language and environment for statistical computing (4.3.0) [Computer software]. R Foundation for Statistical Computing, Vienna, Austria. 2018 https://www.R-project.org/. 27. Doyle JJ. Isolation of plant DNA from fresh tissue. Focus. 1990;12:13. 52. Paradis E, Schliep K. ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics. 2019;35:526–8. 52. Paradis E, Schliep K. ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics. 2019;35:526–8. 28. Vega-Vela NE, Chacón Sánchez MI. Isolation of high-quality DNA in 16 aromatic and medicinal Colombian species using silica-based extraction columns. Agron Colomb. 2011;29:349–57. 53. Keselman HJ, Rogan JC. The Tukey multiple comparison test: 1953–1976 Psychol Bull. 1977;84:1050–6. g 29. Green MR, Sambrook J. Precipitation of DNA with ethanol. Cold Spring Harb Protoc. 2016. https://doi.org/10.1101/pdb.prot093377. 54. Wickham H. ggplot2: elegant graphics for data analysis. New York: Springer, New York; 2009. 30. Manen J-F, Sinitsyna O, Aeschbach L, Markov AV, Sinitsyn A. A fully autom- atable enzymatic method for DNA extraction from plant tissues. BMC Plant Biol. 2005;5:23. 55. SnapGene Software (https://www.snapgene.com/) 56. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illu- mina sequence data. Bioinformatics. 2014;30:2114–20. 56. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible tr mina sequence data. Bioinformatics. 2014;30:2114–20. mina sequence data. Bioinformatics. 2014;30:2114–20. 57. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–7. 57. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–7. 31. Azimi SM, Nixon G, Ahern J, Balachandran W. A magnetic bead-based DNA extraction and purification microfluidic device. Microfluid Nanoflu- idics. 2011;11:157–65. 58. Ratnasingham S, Hebert PDN. Bold: the barcode of life data system. Mol Ecol Notes. 2007;7:355–64. 32. Emaus MN, Cagliero C, Gostel MR, Johnson G, Anderson JL. Simple and efficient isolation of plant genomic DNA using magnetic ionic liquids. Plant Methods. 2022;18:1–14. 59. Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW. GenBank. Nucleic Acids Res. 2016;44:D67-72. 33. Li J, Wang S, Yu J, Wang L, Zhou S. A modified CTAB protocol for plant DNA extraction. Chinese Bull Botany. 2013;48:72. 60. Edgar RC, Bateman A. Search and clustering orders of magnitude faster than BLAST. Bioinform Appl Note. 2010;26:2460–1. 60. Edgar RC, Bateman A. Publisher’s Note 36. Vadell E, de Miguel S, Pemán J. Large-scale reforestation and afforestation policy in Spain: a historical review of its underlying ecological, socioeco- nomic and political dynamics. Land Use Policy. 2016;55:37–48. 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.
https://openalex.org/W3013225907	https://europepmc.org/articles/pmc7105604?pdf=render	English	null	Combined Transplantation of Olfactory Ensheathing Cells With Rat Neural Stem Cells Enhanced the Therapeutic Effect in the Retina of RCS Rats	Frontiers in cellular neuroscience	2,020	cc-by	12,618	ORIGINAL RESEARCH published: 24 March 2020 doi: 10.3389/fncel.2020.00052 Combined Transplantation of Olfactory Ensheathing Cells With Rat Neural Stem Cells Enhanced the Therapeutic Effect in the Retina of RCS Rats Wei Zhai1,2†, Lixiong Gao1,2,3†, Linghui Qu1,2, Yijian Li1,2, Yuxiao Zeng1,2, Qiyou Li1,2, Haiwei Xu1,2* and Zheng Qin Yin1,2* Keywords: retinal degenerative diseases, neural stem cells, olfactory ensheathing cells, combined transplantation, RCS rats Lankford, Yale University, United States Specialty section: This article was submitted to Cellular Neuropathology, a section of the journal Frontiers in Cellular Neuroscience Received: 14 November 2019 Accepted: 21 February 2020 Published: 24 March 2020 Specialty section: This article was submitted to Cellular Neuropathology, a section of the journal Frontiers in Cellular Neuroscience Received: 14 November 2019 Accepted: 21 February 2020 Published: 24 March 2020 1 Southwest Hospital/Southwest Eye Hospital, Third Military Medical University (Army Medical University), Chongqing, China, 2 Key Lab of Visual Damage and Regeneration & Restoration of Chongqing, Chongqing, China, 3 Department of Ophthalmology, The 6th Medical Center of PLA General Hospital, Beijing, China Edited by: Gary S. L. Peh, Singapore Eye Research Institute (SERI), Singapore Retinal degenerative diseases (RDDs) are the leading causes of blindness and currently lack effective treatment. Cytotherapy has become a promising strategy for RDDs. The transplantation of olfactory ensheathing cells (OECs) or neural stem cells (NSCs) has recently been applied for the experimental treatment of RDDs. However, the long- term outcomes of single-cell transplantation are poor. The combined transplantation of multiple types of cells might achieve better effects. In the present study, OECs [containing olfactory nerve ﬁbroblasts (ONFs)] and NSCs were cotransplanted into the subretinal space of Royal College of Surgeons (RCS) rats. Using electroretinogram (ERG), immunoﬂuorescence, Western blot, and in vitro Transwell system, the differences in the electrophysiological and morphological changes of single and combined transplantation as well as the underlying mechanisms were explored at 4, 8, and 12 weeks postoperation. In addition, using the Transwell system, the inﬂuence of OECs on the stemness of NSCs was discovered. Results showed that, compared to the single transplantation of OECs or NSCs, the combined transplantation of OECs and NSCs produced greater improvements in b-wave amplitudes in ERGs and the thickness of the outer nuclear layer at all three time points. More endogenous stem cells were found within the retina after combined transplantation. Glial ﬁbrillary acidic protein (GFAP) expression decreased signiﬁcantly when NSCs were cotransplanted with OECs. Both the vertical and horizontal migration of grafted cells were enhanced in the combined transplantation group. Meanwhile, the stemness of NSCs was also better maintained after coculture with OECs. Taken together, the results suggested that the combined transplantation of NSCs and OECs enhanced the improvement in retinal protection in RCS rats, providing a new strategy to treat RDDs in the future. Reviewed by: Biju Thomas, University of Southern California, United States Karen L. Lankford, Yale University, United States Correspondence: Haiwei Xu haiweixu2001@163.com; xuhaiwei@tmmu.edu.cn Zheng Qin Yin qinzyin@aliyun.com †These authors have contributed equally to this work Reviewed by: Biju Thomas, University of Southern California, United States Karen L. INTRODUCTION migration of grafted cells was very limited. The combined transplantation of two or more diﬀerent types of cells might improve the therapeutic eﬀect of stem cell transplantation. Our previous work showed that the combined transplantation of mesenchymal stem cells and RPCs can simultaneously generate synergistic eﬀects after subretinal transplantation (Qu et al., 2017). Characterized by the progressive loss or malfunction of retinal cells, retinal degenerative diseases (RDDs) are the leading causes of blindness. One of the most common RDDs is the age- related macular degeneration (AMD) (Veleri et al., 2015). There are two types of AMD, the dry AMD and the wet AMD, whose pathological characteristics are the neovascularization secondary to the stenosis of choroidal vessels and the decrease in phagocytosing function of retina pigment epithelium (RPE) followed by photoreceptor death, respectively (Blasiak, 2020). Data have shown that the general prevalence of all types of AMD is ∼8.7%, and the number of individuals aﬀected by AMD will increase to 196 million in 2020 (Wong et al., 2014; Jonas et al., 2017). Currently, there are no speciﬁc therapeutic methods for AMD, especially for dry AMD. Treatments for wet AMD such as intravitreous injection do not stop the degeneration of the retina in wet AMD patients (Mavija et al., 2014). As retinal cell loss is the ultimate result of AMD, stem cell therapy is becoming a promising method for treating AMD (Nazari et al., 2015). To better understand the diseases, several RDD models have been developed. Royal College of Surgeons (RCS) rat is one of them. Characterized by the inability of RPE to phagocytose photoreceptor outer segments, RCS rat is the ﬁrst known animal with inherited retinal degeneration (Strauss et al., 1998). The mutation in this model is found to be the deletion in the receptor tyrosine kinase gene Mertk, which links to the phagocytosing function of RPE (D’Cruz et al., 2000). The RPE dysfunction will further lead to the deposition of photoreceptor outer segments and consequent photoreceptor degeneration. Since its pathology is similar to RDDs, RCS rat is widely used as an animal model to mimic AMD and retinitis pigmentosa (LaVail, 2001). The progress of retinal degeneration is rapid in RCS rats. The response of electroretinograms (ERGs) in RCS rat begin to decrease at postnatal 21 days, reduce by half at postnatal 50 days, and nearly disappear at postnatal 100 days (Pinilla et al., 2005). INTRODUCTION As a physiological response to lesion to the CNS, gliosis is a double-edged sword. In the retina, the gliotic response from Müller cells regulates the size of the glial scar, which inhibits transplanted cells from exerting their therapeutic eﬀect. Our previous results showed that olfactory ensheathing cells (OECs), a glial cell type originating from the neocortex of the olfactory bulb, inhibits gliosis in the retinas of RCS rats (Xie et al., 2017). Therefore, transplanting OECs with NSCs might improve the restoration of visual function in RCS rats. OECs have been reported to support the continuous growth and regeneration of olfactory axons throughout life (Ramon- Cueto and Avila, 1998). Robust studies have conﬁrmed that transplanted OECs exhibit neuroplastic and neuroregenerative eﬀects via interacting with the glial scar and stimulating angiogenesis, axonal outgrowth, and remyelination in the spinal cord injury (Roet and Verhaagen, 2014; Gomez et al., 2018). OECs have also been found to protect visual function. Our previous work showed that OECs restore retinal function and alleviate retinal degeneration in RDD animal models via reducing the gliotic injury response of Müller cells, phagocytosing retinal outer segments, and inhibiting oxidative stress (Huo et al., 2011, 2012; Xie et al., 2017; Xue et al., 2017). We further conﬁrmed that OECs can promote retinal ganglion cell survival and axonal regeneration after optic nerve injury for 3 months (Wang et al., 2017). However, the protective eﬀect of OECs in the eyes is maintained for a relatively short period (Xue et al., 2017). The capacity of OECs to migrate relies on various factors (Gomez et al., 2018). These limitations restrict research on of the treatment of RDDs with OECs. Because of the potential improvement in the gliotic microenvironment of OECs as well as the therapeutic eﬀect of NSCs, we hypothesized that the cotransplantation of these two cell types might produce a better eﬀect. In the present study, NSCs and OECs were transplanted either singly or in combination into the SRS of RCS rats at early degenerative stage. Using ERGs, immunoﬂuorescence, Western blotting, and an in vitro Transwell system, we discovered the eﬃcacy of combined transplantation and explored the possible underlying mechanisms at 4, 8, and 12 weeks postoperation. These three time points covered the moderate to the severe retinal degeneration of RCS rats. Citation: Zhai W, Gao L, Qu L, Li Y, Zeng Y, Li Q, Xu H and Yin ZQ (2020) Combined Transplantation of Olfactory Ensheathing Cells With Rat Neural Stem Cells Enhanced the Therapeutic Effect in the Retina of RCS Rats. Front. Cell. Neurosci. 14:52. doi: 10.3389/fncel.2020.00052 March 2020 \| Volume 14 \| Article 52 1 Frontiers in Cellular Neuroscience \| www.frontiersin.org Subretinal OEC and NSC Co-transplantation Zhai et al. INTRODUCTION ( , ) As adult stem cells within the central nervous system (CNS), neural stem cells (NSCs) can generate both neurons and glia (Obernier and Alvarez-Buylla, 2019). Several studies have conﬁrmed the therapeutic eﬀect of NSCs after transplantation into the injured CNS (Marsh and Blurton-Jones, 2017). McGill et al. showed that the transplantation of NSCs into the subretinal space (SRS) of RCS rats preserves retinal function and protects photoreceptors from death through the phagocytosis of photoreceptor outer segments (McGill et al., 2012). Lin et al. (2014) found that NSCs have a better proliferative ability than that of retinal progenitor cells (RPCs). Moreover, upon treatment with transforming growth factor beta type III and retinoic acid, NSCs are able to diﬀerentiate into opsin-positive retinal cells (Lin et al., 2014). Our previous work also conﬁrmed the preservative eﬀect of the subretinal transplantation of NSCs in rd1 mice (Li et al., 2016). However, there were several problems associated with transplanting NSCs alone into the retina: evidence of functional improvement was only observed during a small treatment window, transplantation triggered gliosis, and the In vitro Migration and Differentiation Assay of NSCs After LE rats (90 days old) were anesthetized with pentobarbital sodium (10 mg/kg, Sigma-Aldrich), the olfactory bulbs were dissected and removed under a microscope. The glomerular layers of the olfactory bulbs were carefully isolated and cut into small pieces. The tissues were digested in 0.1% trypsin for 15 min at 37◦C, and the reaction was stopped by OEC culture medium containing Dulbecco’s modiﬁed Eagle’s medium/F-12 culture medium (DMEM/F-12, 1:1 mixture, HyClone) supplemented with 10% fetal bovine serum (FBS, Gibco) and a mixture of penicillin and streptomycin (PS, 1%, Gibco). Then, the OEC suspension was centrifuged at 1,500 rpm for 5 min and resuspended in OEC culture medium. Then, OECs were plated on 35-mm dishes coated with 10 µg/ml laminin and incubated in a 5% CO2 saturation- humidity atmosphere at 37◦C. The culture medium was changed every 3 days. Subculturing was performed once the cell density was over 80%. OECs were identiﬁed at passage 3. After being digested by trypsin, plated on laminin- coated coverslips, and cultured for 3 days, OECs were identiﬁed via immunoﬂuorescence. The details are described in section “Immunoﬂuorescence.” y Passage 3 NSCs were used to conduct both migration and diﬀerentiation assays. For the migration assay, Transwell systems and six-well plates were used. In detail, 2 × 105 rhodamine (Sigma-Aldrich)-labeled NSCs alone or 2 × 105 rhodamine-labeled NSCs together with 2 × 105 Hoechst (Beyotime)-labeled OECs (NSCs + OECs) were added to a single Transwell system. The retinas of neonatal LE rats (postnatal day 0) were removed and isolated on ice after anesthetization with pentobarbital sodium (10 mg/kg, Sigma-Aldrich). The retinas were then placed onto the NSCs or NSCs + OECs in the Transwell system with 1 ml of NSC culture medium. After 12 days of coculture, the retinas were ﬁxed with 4% paraformaldehyde (PFA) for 3 h, followed by dehydration with 30% sucrose for 12 h. Then, the retinas were embedded and sectioned into 10- µm slices. 4′,6-Diamidino-2-phenylindole (DAPI) staining was performed on retinal sections from the NSC group. The sections were observed under an immunoﬂuorescence microscope. For the diﬀerentiation assay, Transwell systems and 24-well plates were used. In the control group, 1 × 105 NSCs were plated along the Transwell membrane. In the experimental group, 2 × 105 OECs were ﬁrst plated in the plates, and 1 × 105 NSCs were then plated on the upper Transwell membrane. N Values and Blinding For in vivo study, 18 animals underwent transplantation treatment in each transplantation group at the starting point. On each of the three diﬀerent posttransplantation time points, six animals in each group were killed after recording ERG. Three animals were used for immunoﬂuorescent test and three animals for Western blot test. In summary, the N value in ERG test was 6; in immunoﬂuorescence, 3; and in Western blot, 3. For in vitro study, both immigration and diﬀerentiation tests were repeated three times (N = 3). Cell harvest was repeated three times in both cells, and identiﬁcation of cells in each batch was performed to ensure their characteristics (N = 3). As for the randomization and blinding, all treatments were randomized, and the persons performing the transplantation surgeries and histological analysis were blinded with respect to the treatment condition. For ﬂow cytometry, passage 3 ﬂoating spheres were digested into single cells. After perforation, washing, and blocking, NSCs were divided into blank and experimental groups. The latter group was incubated with Nestin-FITC, Pax6-FITC, and Sox2- FITC primary antibodies. Then, all groups were tested by ﬂow cytometry (BD, United States). In vitro Migration and Differentiation Assay of NSCs All groups were cultured in the NSC culture medium and cultured for 24 h. In the bromodeoxyuridine (BrdU) test, 25 µl of BrdU solution (2 mg/ml) was added to the corresponding wells at 23 h. Then, Transwell membranes containing NSCs were examined via immunoﬂuorescence. The details are described Animals and Ethics The RCS (28 days) and Long Evans (LE) rats were obtained from the Animal Research Center of the Third Military Medical University (TMMU). Rats were raised under a 12-h light/dark cycle in the speciﬁc pathogen-free room of the Animal Care March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org Frontiers in Cellular Neuroscience \| www.frontiersin.org 2 Subretinal OEC and NSC Co-transplantation Zhai et al. were carefully dissected and cut into small pieces under a microscope. After being digested with Accutase (Innovative Cell Technologies, United States) for 5 min at 37◦C and stopped by the NSC culture medium containing DMEM/F- 12 (Hyclone) supplemented with B27 (Gibco), glutamine (Gibco), basic ﬁbroblast growth factor (bFGF) (20 ng/ml, Peprotech), epidermal growth factor (EGF) (10 ng/ml, Peprotech), and a mixture of PS (1%, Gibco), NSC suspension was centrifuged at 4,000 rpm for 5 min and resuspended with the NSC culture medium. Then, NSC suspension was transferred to ﬂoating culture ﬂasks and incubated in a 5% CO2 saturation-humidity atmosphere at 37◦C. The culture medium was changed every 3 days. For identiﬁcation, passage 3 ﬂoating spheres were directly plated on laminin-coated coverslips or plated after being digested into single cells and cultured for 3 days, after which NSCs were identiﬁed via immunoﬂuorescence. The details are described in section “Immunoﬂuorescence.” Center of the First Aﬃliated Hospital of TMMU. The breeding of LE rats was performed to harvest embryos as well as the neonatal LE rats. All tissue collection and experimental procedures were performed according to protocols approved by the Institutional Review Board of the TMMU and conformed to the National Institutes of Health (NIH) guidelines on the ethical use of animals. Center of the First Aﬃliated Hospital of TMMU. The breeding of LE rats was performed to harvest embryos as well as the neonatal LE rats. All tissue collection and experimental procedures were performed according to protocols approved by the Institutional Review Board of the TMMU and conformed to the National Institutes of Health (NIH) guidelines on the ethical use of animals. Morphological Preparation of Retina Subretinal transplantation was performed as previously described (Qu et al., 2017). In brief, cell suspensions containing OECs, NSCs, or their combination were injected into the temporal subretinal space of the left eyes of RCS rats 4 weeks postnatally (5 µl/eye). The cell suspension in the combination group was a mixture containing 2.5 × 105 cell NSCs and 2.5 × 105 cells OECs/olfactory nerve ﬁbroblasts (ONFs) per eye. In the single transplantation groups, 2.5 × 105 cells of NSCs per eye were transplanted in the NSC group, and 2.5 × 105 cells OECs/ONFs per eye were transplanted in the OEC group. An identical volume of 0.01 M phosphate-buﬀered saline (PBS) (5 µl) was injected into the temporal subretinal space of the right eyes of RCS rats. All transplanted cells were labeled with the ﬂuorescent marker CM-DiI (2 mg/ml, Invitrogen). The pupils were dilated with 1% tropicamide (Santen Pharmaceutical Co., Ltd. Osaka, Japan) 30 min before surgery. Once the RCS rats were anesthetized (120 mg/kg ketamine and 20 mg/kg xylazine), a 10-µl Hamilton syringe (30 gauge; Hamilton, NV, United States) containing a cell suspension was tangentially inserted into the subretinal space through the conjunctiva and sclera, which led to a self- sealing wound tunnel. Paracentesis of the anterior chamber was performed to reduce the intraocular pressure and limit the eﬄux of cells from the injection site. Fundus examinations were performed immediately after the operations. The eyes that did not receive any treatment were labeled as the “blank” group. The eyes in which OECs, NSCs, or OECs + NSCs were transplanted were labeled as the “OEC group,” “NSC group,” or “OEC + NSC group,” respectively. The eyes that were injected with an equal amount of PBS were labeled as the “PBS” group. p g p After being anesthetized by 1% pentobarbital (150 mg/kg), RCS rats were perfused with normal saline and 4% PFA via the circulation system on 4, 8, and 12 postoperation weeks as we previously described (Qu et al., 2017). After being enucleated and ﬁxed in 4% PFA for 3 h, eyeballs were incubated in 30% glucose solution overnight. During the embedding of the eyeball, we marked the injection site and placed the embedded eyeball on the slicer. Isolation, Culture, and Identiﬁcation of NSCs Neural stem cells were harvested from the visual cortex of embryonic LE rats at embryonic day 13.5 and cultured. The maternal LE rats were anesthetized with pentobarbital sodium (10 mg/kg, Sigma-Aldrich), and uteruses containing fetal rats were isolated. The visual cortexes of fetal rats March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 3 Subretinal OEC and NSC Co-transplantation Zhai et al. used in clinical and experimental analysis of retinal function (Perlman, 1995), we typically focus on the amplitude of b wave in the present study. in section “Immunoﬂuorescence.” These two assays were repeated three times. Immunoﬂuorescence Immunoﬂuorescence of the identiﬁcation of OECs/ONFs and NSCs as well as NSC diﬀerentiation assay were performed as previously described. In detail, after being rinsed in 0.01 M PBS, blocked in 10% of goat serum, and perforated in 0.1% of Triton X-100, coverlids and Transwell membrane were incubated with primary antibodies overnight at 4◦C. The coverlids with OECs/ONFs were incubated with anti-P75 (1:500, mouse, Santa) antibodies; the coverlids with NSCs with anti-Nestin (1:500, rabbit, Abcam), anti-GFAP (1:500, mouse, Abcam), and anti- Tuj1 (1:1000, mouse, Beyotime) antibodies; and the Transwell membrane with NSCs with anti-GFAP (1:500, mouse, Abcam), anti-Sox2 (1:500, rabbit, Abcam), anti-Pax6 (1:500, rabbit, Santa), and anti-BrdU antibodies (1:500, Cell Signaling). Cy3- or 488-conjugated secondary antibodies, (Invitrogen) were then implemented (1:400, 37◦C, 2 h). Before examination with a confocal laser scanning microscope (Leica, Germany), cells were counterstained with DAPI (Sigma Aldrich). Immunoﬂuorescence of retina sections was also performed as previously described (Gao et al., 2015). In detail, after being washed in 0.01 M PBS, blocked in 10% of goat serum, and perforated in 0.1% of Triton X-100, selected sections were incubated with the primary antibodies, anti-GFAP (1:500, mouse, Abcam), anti-Sox2 (1:500, rabbit, Abcam), and anti-recoverin (1:1000, rabbit, Millipore) antibodies in 1% bovine serum albumin (BSA) at 4◦C overnight. Cy3- or 488-conjugated secondary antibodies (Invitrogen) were then implemented (1:400, 37◦C, 2 h). Before examination with a confocal laser scanning microscope (Leica, Germany), sections were counterstained with DAPI (Sigma Aldrich). Morphological Preparation of Retina This placement was carried out in two disciplines: (1) The eyeball stood vertically in the embedding medium; (2) the plane formed by 3 points (the injection site, optic disk, and the point opposite to the injection site on the eyeball) kept parallel to the blade (Supplementary Figure 1). These disciplines can ensure the consistency in the sections among diﬀerent groups. Then, 10-µm serially frozen sections were carefully made. ERG Recording ERG recording was performed 4, 8, and 12 weeks postoperation to evaluate the retinal functional changes, as previously described (Lin et al., 2014). In brief, after being dark adapted for at least 12 h, RCS rats were anesthetized by an intraperitoneal injection of a solution of ketamine (120 mg/kg) and xylazine (20 mg/kg). Pupils were dilated with 1% tropicamide before testing. A heating pad was used to maintain the body temperature at 37◦C. Two active gold electrodes were placed on each cornea as recording electrodes. The reference and ground electrodes were subcutaneously inserted into the mid-frontal area of the head and tail, respectively. Light stimulations were delivered with a xenon lamp at 3.0 cd s/m2. The b-wave amplitudes were recorded and processed by a RETI-Port device (Roland Consult, Brandenburg, Germany). All procedures were performed in a dark room with dim red safety light. When dealing with the results, a and b waves were marked by the typical type of ERG waves as well as the latent period, which was basically achieved by computer and checked by experimenters. If the computer failed to calculate the proper point, experimenters would manually measure the results. Since b wave represents the transduction of extracellular currents and is considered to be the major component of the human ERG recording as Combined Transplantation Enhanced Electrophysiological Improvement of RCS Rats g y To quantitatively analyze the diﬀerences in the expression of NSC markers after coculture with OECs, three comparable visual ﬁelds from each cell slide were randomly selected. The numbers of BrdU-, Pax6-, Sox2-, and GFAP-positive cells were manually counted and averaged. To quantitatively analyze in vivo cell migration after cell transplantation, six sections that were cut using the same horizontal angle across the optic disk were chosen to conduct the migration measurement. In each section, the photos of retina were taken under a 400× microscope. Integrate image of the whole retina was generated by splicing these photos. The distance that transplanted cells migrated within the SRS was measured by ImageJ (NIH, United States) in each integrate image. To quantitatively analyze the status of endogenous stem cell formation after transplantation, at least three sections across the optic disk were selected from each group after Sox2 staining. The number of Sox2-positive cells within a 150 µm × 150 µm square visual ﬁeld was manually counted and averaged. To semiquantitatively analyze the GFAP expression level, at least three sections across the optic disk were selected from each group after GFAP staining. The density of GFAP-positive cells was recorded by ImageJ (NIH). After transplantation with a single cell type or a combination of the two cell types, ERGs were recorded, and b-wave amplitude was measured to determine electrophysiological improvements at 4, 8, and 12 weeks postoperation. The results showed that, compared to the blank and the PBS group, the OEC, NSC, and OEC + NSC groups presented signiﬁcant improvements in b-wave amplitudes at all time points (P < 0.001; Figures 1A–P). Moreover, there were also diﬀerences among three transplantation groups. In particular, at 4 weeks postoperation, the OEC + NSC group showed signiﬁcantly higher b-wave amplitude than that in either the OEC or the NSC group (P < 0.001, P < 0.001; Figures 1C– E,P). There was no signiﬁcant diﬀerence between the OEC and the NSC group (P > 0.05, Figures 1C–E,P). At 8 weeks postoperation, the OEC + NSC group still presented signiﬁcantly higher b-wave amplitude than that in either the OEC or the NSC group (P < 0.05, P < 0.05; Figures 1H–J,P). There was no signiﬁcant diﬀerence between the OEC and the NSC group at 8 weeks postoperation (P > 0.05; Figures 1H– J,P). However, the retina-function restoration became diﬀerent at 12 weeks postoperation. Outer Nuclear Layer Thickness Analysis Six sections that were cut using the same horizontal angle across the optic disk were chosen to measure the thickness of the outer nuclear layer (ONL). From each section, an average of three areas of the temporal retina area of the optic disk was selected (Supplementary Figure 2). The thickness of the ONL was measured by ImageJ (NIH, United States). The average ONL thickness in the three areas represented the ONL thickness of the section. Identiﬁcation of Primarily Isolated OECs and NSCs The identiﬁcation of both OECs and NSCs was conducted before transplantation. The OECs and NSC were tested at passage 3. The nerve growth factor receptor P75 and Nestin were used as OECs and NSCs markers, respectively (Xie et al., 2017). The OECs were fusiform shaped with elongated processes (Supplementary Figures 3A,B). By immunoﬂuorescence, we found that nearly half of the OEC/ONF mixture expressed P75, and the other half expressed FN (Supplementary Figures 3C–E). For NSCs, the results showed that, when ﬂoating culture was performed, NSCs exhibited a spherical shape and expressed Nestin (Supplementary Figures 4A,B). After being cultured in serum-free media for 2 weeks, NSCs diﬀerentiated into neurons and expressed Tuj1 (Supplementary Figure 4C). Few GFAP-positive glial cells were observed (Supplementary Figure 4D). Flow cytometry showed that a high percentage of NSCs expressed Sox2 (98.86%), Pax6 (98.94%), and Nestin (98.38%) (Supplementary Figure 4E). These results conﬁrmed the characteristics of harvested OECs/ONFs and NSCs. Western Blot Animals were euthanized with CO2 at 4, 8, and 12 weeks postoperation, after which eyeballs were enucleated and retinas were quickly isolated on ice. After being rinsed in 0.01 M PBS and drained, retina tissues were lysed in ice-cold tissue March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org Frontiers in Cellular Neuroscience \| www.frontiersin.org 4 Zhai et al. Subretinal OEC and NSC Co-transplantation lysis buﬀer [10% phenylmethylsulfonyl ﬂuoride (PMSF) + 90% radioimmunoprecipitation assay (RIPA)]. The lysates were then centrifuged at 15,000 rpm for 10 min at 4◦C. Protein concentration was determined using the BCA Protein Assay (Beyotime). After boiling in loading buﬀer for 10 min, total proteins (10 µg per slot) were electrophoresed on a 12% sodium dodecyl sulfate polyacrylamide gel and then transferred onto polyvinylidene ﬂuoride membranes. After being blocked in 5% fat-free milk for 2 h at 37◦C, membranes were incubated with anti-GFAP antibody (1:500, rabbit, Abcam) and anti-glyceraldehyde 3-phosphate dehydrogenase (anti-GAPDH) (1:1,000, mouse, Proteintech Group) antibody overnight at 4◦C. Membranes were then incubated with peroxidase-conjugated immunoglobulin G (1:2,000; Santa Cruz Biotechnology). After being washed in Tris-buﬀered saline with Tween-20 (TBS-T) and developed in developing solution, membranes were scanned using the Bio-Rad exploding system (Bio-Rad, CA, United States) with corresponding software. as the mean ± standard error. P < 0.05 was considered statistically signiﬁcant. as the mean ± standard error. P < 0.05 was considered statistically signiﬁcant. Combined Transplantation Enhanced Electrophysiological Improvement of RCS Rats Compared to the OEC group, the OEC + NSC group showed a signiﬁcant increase in b-wave amplitude (P < 0.001; Figures 1M–O,P), while no statistically signiﬁcant diﬀerence was observed between the Statistical Analysis Using Statistical Product and Service Solutions software V17.0 (SPSS, Chicago, IL, United States), all quantitative results were analyzed by one-way ANOVA followed by Fisher’s protected least-signiﬁcant diﬀerence post hoc tests. The data are presented March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 5 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 1 \| Electrophysiological improvement after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Representative electroretinogram (ERG) results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Representative ERG results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Representative ERG results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (P) Relative statistical analysis of all groups. a and b point in (E,J,O) indicated the a and b waves, which represented the light absorption of photoreceptors and postsynaptic responses of photoreceptors, respectively. ∗∗P < 0.01; ∗∗∗P < 0.001. FIGURE 1 \| Electrophysiological improvement after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Representative electroretinogram (ERG) results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Representative ERG results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Representative ERG results in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (P) Relative statistical analysis of all groups. a and b point in (E,J,O) indicated the a and b waves, which represented the light absorption of photoreceptors and postsynaptic responses of photoreceptors, respectively. ∗∗P < 0.01; ∗∗∗P < 0.001. thickness compared to that of either the OEC or the NSC group (P < 0.001, P < 0.001; Figures 2H–J,P). There was no signiﬁcant diﬀerence between the OEC and the NSC group at this time point (P > 0.05; Figures 2H,I,P). A similar change was found at 12 weeks postoperation. The OEC + NSC group still presented a signiﬁcant higher ONL thickness than that of the two single transplantation groups (OECs, P < 0.001; NSCs, P < 0.05; Figures 2M–O,P). Statistical Analysis No signiﬁcant diﬀerence was found between the two single transplantation groups (P > 0.05; Figures 2M–O,P). Moreover, when retinal sections were stained with recoverin to identify photoreceptors, all three cell transplant groups showed evidence of increased numbers of photoreceptors at each time point examined (Figures 3A– O). Taken together, these results suggested that combined transplantation showed a better protective eﬀect on the ONL than single transplantation. NSC and the OEC + NSC group (P > 0.05; Figures 1N– P). In addition, the NSC group also presented signiﬁcantly higher b-wave amplitude than that of the OEC group at 12 weeks postoperation (P > 0.05; Figures 1M,N,P). These results indicated that, compared with the single transplantation, the combined transplantation of OECs and NSCs enhanced the electrophysiological improvement of RCS. Combined Transplantation Showed Better Photoreceptor Protection in RCS Rats To further evaluate the morphological eﬀects of cell transplantation, the ONL thickness was measured. Compared with the blank and the PBS group, three transplantation groups all presented signiﬁcant protective eﬀects on the ONL (P < 0.001; Figures 2A–P). Among the transplantation groups, both the OEC + NSC and the OEC groups showed signiﬁcantly better protective eﬀects on the ONL than that of the NSC group at 4 weeks postoperation (P < 0.01, P < 0.05; Figures 2C–E,P). No signiﬁcant diﬀerence was found between the OEC + NSC and the OEC group at this time point (P > 0.05; Figures 2C,E,P). At 8 weeks postoperation, the OEC + NSC group displayed a signiﬁcant increase in ONL Endogenous Stem Cell Activation After Combined Transplantation (F–J) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. The mark “#” represented the out nuclear layer. (P) Statistical analysis of the ONL thickness in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. FIGURE 2 \| Protection of ONL after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Retina sections with 4′,6-diamidino-2-phenylindole (DAPI) staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. The mark “#” represented the out nuclear layer. (P) Statistical analysis of the ONL thickness in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. 4F–J,P). However, by 12 weeks post-operation, there was no signiﬁcant diﬀerence among groups (P < 0.05; Figures 4K– O,P). These results suggested that combined transplantation activated more endogenous stem cells at the early stage of transplantation. within the retina were evenly distributed in the inner nuclear layer. At 4 weeks post-operation, three transplantation groups all presented a signiﬁcant higher number of Sox2-positive cells than that of either the Blank group or the PBS group (P < 0.001; Figures 4A–E,P). Among the three transplantation groups, the number of Sox2-positive cells in the OEC + NSC group was signiﬁcantly higher than that of either the OEC group or the NSC group at 4 weeks post-operation (P < 0.05, P < 0.01; Figures 4C–E,P). This signiﬁcant diﬀerence lasted until 8 weeks post-operation (P < 0.01, P < 0.01; Figures 4H– J,P), while no signiﬁcant diﬀerences in Sox2 expression were observed among the Blank group, the PBS, and the two single transplantation groups at this time point (P < 0.05; Figures Endogenous Stem Cell Activation After Combined Transplantation To investigate the possibility that endogenous stem cell responses might underlie the apparent therapeutic actions of cell transplantation, we labeled retinas with Sox2 antibodies (Tian et al., 2011). The results showed that Sox2-positive cells March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org Frontiers in Cellular Neuroscience \| www.frontiersin.org 6 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 2 \| Protection of ONL after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Retina sections with 4′,6-diamidino-2-phenylindole (DAPI) staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. The mark “#” represented the out nuclear layer. (P) Statistical analysis of the ONL thickness in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. within the retina were evenly distributed in the inner nuclear 4F–J,P). However, by 12 weeks post-operation, there was no FIGURE 2 \| Protection of ONL after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Retina sections with 4′,6-diamidino-2-phenylindole (DAPI) staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Retina sections with DAPI staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. The mark “#” represented the out nuclear layer. (P) Statistical analysis of the ONL thickness in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. FIGURE 2 \| Protection of ONL after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Retina sections with 4′,6-diamidino-2-phenylindole (DAPI) staining in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. Inﬂuence on the Reactive Gliosis of Müller Cells After Combined Transplantation The reactive gliosis of Müller cells following retina damage become an obstacle to retinal regeneration. To generally analyze the gliosis of Müller cells, we performed the Western blot March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org Frontiers in Cellular Neuroscience \| www.frontiersin.org 7 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 3 \| Protection of photoreceptors after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. Scale bar: (A–O) 75 µm. FIGURE 3 \| Protection of photoreceptors after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Immunoﬂuorescence of Recoverin in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. Scale bar: (A–O) 75 µm. the OEC and the OEC + NSC group presented signiﬁcantly lower GFAP FIs than those in the blank, the PBS, and the NSC groups, respectively (P < 0.05, P < 0.01, P < 0.01; Figures 5D– H,S). There was no signiﬁcant diﬀerence among the blank, the PBS, and the NSC groups. However, when it came to 8 and 12 weeks postoperation, situation changed dramatically. The NSC and OEC + NSC groups both showed signiﬁcant increases in GFAP FI compared to those in the blank, the PBS, and the OEC groups at these two time points (P < 0.001, P < 0.001, P < 0.001; Figures 5I–S). While the OEC + NSC group displayed a signiﬁcant decrease in GFAP FI compared to that in the NSC group (P < 0.001; Figures 5L,M,Q,R,S). Taken together, transplantation of OECs and NSCs can decrease the gliosis following retinal degeneration. Inﬂuence on the Reactive Gliosis of Müller Cells After Combined Transplantation Moreover, transplantation of NSCs would bring gliosis to the SRS, which could be inhibited by cotransplantation of OECs. analysis and the immunoﬂuorescence of GFAP at 4, 8, and 12 weeks posttransplantation. WB results showed that at 4 weeks postoperation, the blank, the PBS, and the NSC groups presented similar GFAP expression levels, which were signiﬁcantly higher than those in the OEC group and the OEC + NSC group (P < 0.05, P < 0.01; Figures 5A,B). However, at 8 weeks postoperation, GFAP expression level in three transplantation groups all signiﬁcantly decreased compared to that in the blank and the PBS groups (OEC, P < 0.05; NSC, P < 0.001; OEC + NSC, P < 0.001; Figures 5A,B). The situation remained similar at 12 weeks postoperation; the only diﬀerence was that compared with the OEC group, the NSC group presented a signiﬁcantly lower GFAP expression level (P < 0.05, Figures 5A,B). We also divided the retina into upper and lower parts when analyzing the results of immunoﬂuorescence (Figure 5C). The boundary was set to be the outer border of ONL (the inner border of SRS) (solid line in Figure 5C). The ﬂuorescence intensity (FI) within the inner part and the outer part of the retina was measured and labeled as upper and lower, respectively (Figure 5C). Results showed that the trends of GFAP FI in the upper part of the retina were similar to the trends of the GFAP expression level in WB at all three time points (Figures 5B,S). Interestingly, the results of GFAP FI in the lower part of the retina showed distinct diﬀerences. At 4 weeks postoperation, Frontiers in Cellular Neuroscience \| www.frontiersin.org The Possible Rescuing Mechanism of the Combined Transplantation To further investigate the mechanism of rescue following cotransplantation, we evaluated the inﬂuence of migration and the cell state of NSCs upon coculture with OECs. The migration of transplanted cells within the SRS was ﬁrst investigated. The results showed that the OEC + NSC group presented March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 8 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 4 \| Activation of endogenous retinal stem cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (P) Statistical analysis of the numbers of Sox2-positive cells in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. FIGURE 4 \| Activation of endogenous retinal stem cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats (A E) Immunoﬂuorescence of Sox2 in the blank PBS OEC NSC and OEC + NSC groups at 4 weeks FIGURE 4 \| Activation of endogenous retinal stem cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–E) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (F–J) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (K–O) Immunoﬂuorescence of Sox2 in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (P) Statistical analysis of the numbers of Sox2-positive cells in transplantation groups. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. a signiﬁcantly shorter migration distance compared to that of the NSC group at 4 weeks postoperation (P < 0.05; Figures 6B,C,J). No signiﬁcant diﬀerence was detected between the two single transplantation groups (P > 0.05; Figures 6A,B,J). The Possible Rescuing Mechanism of the Combined Transplantation (N–R) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (S) The average t i t it i th bl k PBS OEC NSC d OEC NSC f th d l t f th ti RGCL ti l li ll l IPL i E 5 \| Inﬂuence of the reactive gliosis of Muller cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into bretinal space of Royal College of Surgeons (RCS) rats. (A,B) The Western blot analysis of glial ﬁbrillary acidic protein (GFAP) expression after transplantation. e band of GFAP and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) at 4, 8, and 12 weeks postoperation. (B) Corresponding analysis of the Western sults. (C) Schematic diagram of the division of the retina. Solid line indicated the boundary of the division. (D–H) Immunoﬂuorescence of GFAP in the blank, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (I–M) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at s postoperation. (N–R) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (S) The average cent intensity in the blank, PBS, OEC, NSC, and OEC + NSC groups of the upper and lower parts of the retina. RGCL, retinal ganglion cell layer; IPL, inner m layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. FIGURE 5 \| Inﬂuence of the reactive gliosis of Muller cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A,B) The Western blot analysis of glial ﬁbrillary acidic protein (GFAP) expression after transplantation. (A) The band of GFAP and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) at 4, 8, and 12 weeks postoperation. (B) Corresponding analysis of the Western blot results. (C) Schematic diagram of the division of the retina. Solid line indicated the boundary of the division. (D–H) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (I–M) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. The Possible Rescuing Mechanism of the Combined Transplantation However, by 8 and 12 weeks postoperation, the OEC + NSC group presented a signiﬁcant increase in migration distance compared to that of the single transplantation groups (P < 0.05; Figures 6D–J). There was no signiﬁcant diﬀerence between the two single transplantation groups at these two time points (P > 0.05; Figures 6D–H,J). An in vitro Transwell system was used to investigate the cell migration within the retina tissue (Figure 6K). After 12 days of culture, we found that there was little vertical migration of NSCs in the retinal tissue from the NSCs alone group (Figure 6L), while in the coculture group, there were many more cells that entered the retinal tissue (Figure 6M). Both in vivo and in vitro analyses conﬁrmed the cell-migration-enhancement eﬀect following cotransplantation. To further explore the inﬂuence of the cell state of NSCs following cotransplantation, coculture of OECs and NSCs via Transwell system was performed (Figure 7I). Results showed no signiﬁcant diﬀerence in proliferation between the NSC single culture group and the NSC + OEC coculture group (P > 0.05; Figures 7A,E,J). However, the NSC + OEC coculture group presented signiﬁcant higher Pax6 and Sox2 expressions than those of the NSC single culture group (Pax6, P < 0.001; Sox2, P < 0.001; Figures 7B,C,F,G,J). Meanwhile, a signiﬁcant decrease in GFAP expression was also found in the NSC + OEC coculture group, compared to the NSC single culture group March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 9 Zhai et al. Subretinal OEC and NSC Co-transplantation GURE 5 \| Inﬂuence of the reactive gliosis of Muller cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into e subretinal space of Royal College of Surgeons (RCS) rats. (A,B) The Western blot analysis of glial ﬁbrillary acidic protein (GFAP) expression after transplantation. A) The band of GFAP and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) at 4, 8, and 12 weeks postoperation. (B) Corresponding analysis of the Western ot results. (C) Schematic diagram of the division of the retina. Solid line indicated the boundary of the division. (D–H) Immunoﬂuorescence of GFAP in the blank, BS, OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (I–M) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at weeks postoperation. The Possible Rescuing Mechanism of the Combined Transplantation (K) The schematic diagram of the migration assay performed via Transwell system. (L,M) The results of migration analysis. NSCs were stained by Rhodamine, and OECs were stained by Hoechst. Cell migration with NSCs alone (L) as well as OECs and NSCs mixture (M) were tested. Scale bar: (B,C) 200 µm. Scale bar: (A–I) 2 mm; (A1–I1) 200 µm; (L–M) 200 µm.P < 0.05. FIGURE 6 \| Migration of transplanted cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–C) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (D–F) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (G–I) General microscopy of the retina section in OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (A1–I1) Relative partial enlargements showed the farthest location of transplanted cells. (J) Relative statistical analysis of transplantation groups. (K) The schematic diagram of the migration assay performed via Transwell system. (L,M) The results of migration analysis. NSCs were stained by Rhodamine, and OECs were stained by Hoechst. Cell migration with NSCs alone (L) as well as OECs and NSCs mixture (M) were tested. Scale bar: (B,C) 200 µm. Scale bar: (A–I) 2 mm; (A1–I1) 200 µm; (L–M) 200 µm.P < 0.05. (P < 0.001; Figures 7D,H,J). These results suggested that NSCs exhibited enhanced stemness but reduced gliotic tendency when cocultured with OECs. retinal degeneration in a more eﬀective way than single-cell transplantation in the retinas of RCS rats. Endogenous stem cell activation, enhanced migration of transplanted cells, and stemness maintenance of NSCs following combined transplantation may be possible underlying mechanisms. Comparing to the normal amplitude of ERG b wave in age- corresponding LE rats (about 1,300 ± 200 µV) (Xue et al., 2017), the ratios of the b-wave amplitude between the combined transplantation group and the normal LE rats at 4, 8, and The Possible Rescuing Mechanism of the Combined Transplantation (N–R) Immunoﬂuorescence of GFAP in the blank, PBS, OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (S) The average ﬂuorescent intensity in the blank, PBS, OEC, NSC, and OEC + NSC groups of the upper and lower parts of the retina. RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer. Scale bar: (A–O) 50 µm. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org Frontiers in Cellular Neuroscience \| www.frontiersin.org 10 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 6 \| Migration of transplanted cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–C) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (D–F) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (G–I) General microscopy of the retina section in OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (A1–I1) Relative partial enlargements showed the farthest location of transplanted cells. (J) Relative statistical analysis of transplantation groups. (K) The schematic diagram of the migration assay performed via Transwell system. (L,M) The results of migration analysis. NSCs were stained by Rhodamine, and OECs were stained by Hoechst. Cell migration with NSCs alone (L) as well as OECs and NSCs mixture (M) were tested. Scale bar: (B,C) 200 µm. Scale bar: (A–I) 2 mm; (A1–I1) 200 µm; (L–M) 200 µm.P < 0.05. FIGURE 6 \| Migration of transplanted cells after combined transplantation of rat olfactory ensheathing cells (OECs) and neural stem cells (NSCs) into the subretinal space of Royal College of Surgeons (RCS) rats. (A–C) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 4 weeks postoperation. (D–F) General microscopy of the retina section in the OEC, NSC, and OEC + NSC groups at 8 weeks postoperation. (G–I) General microscopy of the retina section in OEC, NSC, and OEC + NSC groups at 12 weeks postoperation. (A1–I1) Relative partial enlargements showed the farthest location of transplanted cells. (J) Relative statistical analysis of transplantation groups. DISCUSSION In the current study, the combined transplantation of OECs and NSCs produced a better neuroprotective eﬀect and delayed March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 11 Zhai et al. Subretinal OEC and NSC Co-transplantation FIGURE 7 \| The inﬂuence on differentiation of neural stem cells (NSCs) after in vitro coculture with olfactory ensheathing cells (OECs) via Transwell system. (A–H) Immunoﬂuorescence analysis of BrdU, Pax6, Sox2, and glial ﬁbrillary acidic protein (GFAP) in NSCs cultured along group (A–D) and NSCs + OECs coculture group (E–H). (I) The schematic diagram of the differentiation assay performed via Transwell system. (J) Relative statistical analysis of (A–H). Scale bar: (A–H) 25 µm. *P < 0.001. FIGURE 7 \| The inﬂuence on differentiation of neural stem cells (NSCs) after in vitro coculture with olfactory ensheathing cells (OECs) via Transwell system. (A–H) Immunoﬂuorescence analysis of BrdU, Pax6, Sox2, and glial ﬁbrillary acidic protein (GFAP) in NSCs cultured along group (A–D) and NSCs + OECs coculture group (E–H). (I) The schematic diagram of the differentiation assay performed via Transwell system. (J) Relative statistical analysis of (A–H). Scale bar: (A–H) 25 µm. *P < 0.001. 12 weeks postoperation were 10.1, 5.6, and 4%, respectively. Although positive inﬂuence was within a limited extent and dropped obviously with time, the better improvement reﬂected by ERG still lasted for 8 weeks, which is of great signiﬁcance from bench to bedside. These results indicated that the combined transplantation of OECs and NSCs may be an alternative stem cell therapy for patients with RDDs in the future. 12 weeks postoperation were 10.1, 5.6, and 4%, respectively. Although positive inﬂuence was within a limited extent and dropped obviously with time, the better improvement reﬂected by ERG still lasted for 8 weeks, which is of great signiﬁcance from bench to bedside. These results indicated that the combined transplantation of OECs and NSCs may be an alternative stem cell therapy for patients with RDDs in the future. were reported to be reprogrammed as the progenitor cells and repair the degenerated retina (Jorstad et al., 2017; Yao et al., 2018). Under physiological conditions, Müller cells remain quiescent to avoid depletion of stem-cell pool. On the contrary, they are able to exit from latent state to proliferate and diﬀerentiate upon injuries (Madelaine and Mourrain, 2017). DISCUSSION Our previous study found that the dediﬀerentiation of Müller cells was increased after the transplantation of retinal stem cells into the SRS of RCS rats, bringing both morphological and functional improvement to the host (Tian et al., 2011). Besides, Sox2 was identiﬁed to be re-expressed in Müller cells after injury, indicating that Müller cells exited the quiescent state and generated new retinal neurons (Gorsuch et al., 2017). As more Sox2-positive cells within the inner nuclear layer were found after combined transplantation in the present study, stronger endogenous repair of retina following activation of more endogenous stem cells might be an underlying mechanism. However, in the present study, the observation time lasted for only 12 weeks. The continued functional and morphological improvements were not observed. In addition, no direct measure of visual function was conducted in the present study, which could lead to incomplete assessment and functional bias. More importantly, limited observation period was not suﬃcient to discover underlying tumor formation possibility and other safety problems. In the future, studies with longer observation period and functional or behavioral tests need to be conducted. Our previous research conﬁrmed the visual restorative eﬀect of OECs (Huo et al., 2012). The inhibition of gliosis via the downregulation of the Notch signaling pathway in Müller cells might be a possible mechanism (Xie et al., 2017). However, in this study, we found that combined transplantation of OECs and NSCs showed a better preservative eﬀect than that of the OEC single transplantation, and this eﬀect lasted for 8 weeks after transplantation. Retinal gliosis is a non-neoplastic retinal glial proliferation followed by a complex retinal response participated by Müller cells, microglia cells, and alterations of the vasculature (Bringmann et al., 2006). It acts as a double-edged sword in the pathological process of RCS rats. On the one hand, reactive gliosis is a physical process that can be regarded as a cellular response to protect the retina from further damage. In addition, moderate gliosis can promote retinal repair following pathological insult (Graca et al., 2018). On the other hand, however, gliosis after nervous impairment including spinal cord and retina was widely known as an obstacle for the Diﬀerences in the endogenous stem cell activation may be one of the explanations. In the spinal cord injury, the activation of endogenous stem cells was considered a promising method for spinal cord recovery (Qin et al., 2015). Frontiers in Cellular Neuroscience \| www.frontiersin.org DISCUSSION Study had shown that coculture of human NSCs with human mesenchymal stem cells could signiﬁcantly extend the stemness of NSCs via activating Notch- 1 signal transduction (Haragopal et al., 2015). In the present study, with in vitro Transwell system, the stemness of NSCs was enhanced when OECs and NSCs were cocultured. Combined transplantation of OECs and NSCs also repressed the NSC- derived gliosis within SRS, illustrating that the diﬀerentiation of NSCs was inhibited and the stemness of NSCs was maintained by OECs. This gliosis regulating eﬀect of OECs can be attributed to the main function of OECs as supporting cells. However, relatively low retina-protection eﬀect was also observed following OEC single transplantation. As mentioned above, reactive gliosis can also bring about positive eﬀect to retinal repair; inhibition of gliosis might be the cause of the low pro-retina activity. It should be noted that the underlying molecular mechanisms relating to endogenous stem cell activation as well as gliosis inhibition after cotransplantation of OECs and NSCs were not explored in the current study, which should be further explored in the future. p What is more, the fate of transplanted cells should also be taken into consideration. The main functions of NSCs transplanted into the subretinal space in RCS rats were phagocytosis of photoreceptor outer segments, secretion of neurotrophic factors, and inhibition of microglia (McGill et al., 2012; Jung et al., 2013; Li et al., 2016). Although NSCs were observed to be diﬀerentiated into photoreceptors and opsin-positive retinal cells, integration of NSC into ONL was hardly found (Nishida et al., 2000; Lin et al., 2014). As for OECs, our previous study showed that the function of OECs transplantation into the subretinal space of RCS rats mainly consisted of two aspects: (1) the microenvironment regulation eﬀects, including secretion of neurotrophic factors and inhibition of the formation of reactive oxygen species; (2) the suppression of the gliotic injury response of the Müller cells (Huo et al., 2011, 2012; Xue et al., 2017). However, the diﬀerentiation and integration of OECs following retinal transplantation was not observed. Although there was better retina-protection eﬀect after combined transplantation of OECs and NSCs, current study still presented a decreasing trend in both functional and morphological results. Considering the immune response following exotic cell transplantation, we speculate that the major ending of NSCs and OECs is the apoptosis after a certain time period. DISCUSSION In the retina, Müller cells March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 12 Subretinal OEC and NSC Co-transplantation Zhai et al. regeneration of neuron and neural dendrites (Bringmann et al., 2006; Wu et al., 2011). The migration of transplanted cells is essential for the development of their function in the transplanted area. Migration exists in two dimensions: horizontal and vertical. Horizontal migration determines the scope of transplantation (McGill et al., 2012; Peng et al., 2014), while vertical migration determines the function of the inner retinal layer (Santos-Ferreira et al., 2016). Our previous results conﬁrmed that better retinal preservation eﬀects can be derived from combined transplantation due to the enhanced horizontal and vertical migration of transplanted cells (Qu et al., 2017). In the present study, diﬀerences in cell migration were also observed between the combined and the single transplantation groups. Grafted cells reached further following combined transplantation. Moreover, NSCs were able to migrate into the inner layer of the retina in the presence of OECs, although this was rarely observed when NSCs were cultured alone. A study showed that, during physiological development, the gene expression products of OECs, including Nelf and Semaphorin 4, are responsible for neuronal migration within the CNS of mice (Geller et al., 2013), indicating the potential migration-enhancing ability of OECs. In addition, OECs are fundamentally characterized by their ability to promote axonal regeneration (Yang et al., 2015). Our previous work also conﬁrmed the stimulating eﬀect on neuronal survival and the outgrowth of OECs, which was due to the phagocytosis of cell debris from OECs (Li et al., 2017). In this way, the processes and integration of NSCs might be extended and promoted following cotransplantation with OECs. In the present study, apart from the retinal gliosis, there emerged a certain amount of GFAP-positive neural ﬁbers within the SRS following the NSCs transplantation and the combined transplantation. This NSC-derived gliosis might impede the neural regeneration and the repairing eﬀect resulting from transplanted stem cells. As NSCs were able to diﬀerentiate into glial cells, these neural ﬁbers most likely came from the diﬀerentiation of NSCs. Basically, although NSCs were widely used to treat various neurodegenerative diseases, the maintaining of stemness had become an imposing barrier. To compensate the drawback, a promising way was to use another kind of stem cell to inﬂuence NSCs. Frontiers in Cellular Neuroscience \| www.frontiersin.org DISCUSSION Moreover, we could not rule out the possibility that the greater improvements in the cotransplanted condition might be due simply to increased numbers of transplanted cells. The diﬀerences in responses to transplants of each cell type separately suggested that the greater improvements were more likely to be due to the combined eﬀects of the two cell types. p Besides, microglia show close relationship to the gliosis of Müller cells (Gao et al., 2015). As both Müller cells and microglia are responsible for the secretion of neurotrophic factor within the retina, the network formed by microglia–Müller glia– photoreceptors can signiﬁcantly inﬂuence the microenvironment during retinal degeneration (Harada et al., 2002). On the one hand, degenerated photoreceptors induce the activation and migration of microglia from the inner to the outer retina. During this procedure, activated microglia alter the trophic factor production and further cause the gliosis of Müller glia, which can inﬂuence the production of neurotrophic factor in Müller glia (Harada et al., 2002). On the other hand, gliosis of Müller glia triggered by photoreceptor degeneration also leads to the apoptosis of neurons resulting in more severe activation of microglia (Telegina et al., 2018). These aspects trap the microglia–Müller glia–photoreceptor network into a vicious cycle and cause the deﬁciency of trophic factors, which is detrimental to the survival of photoreceptors. However, after combined transplantation of OECs and NSCs, gliosis was found to be reduced in the present study. Besides, our previous work conﬁrmed the inhibition of microglia activation following NSCs transplantation (Li et al., 2016). These situations might improve the neurotrophic factor secretion situation and improve the photoreceptor survival subsequently. March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 13 Zhai et al. Subretinal OEC and NSC Co-transplantation Besides, in our previous study, transplantation of OECs mixed with ONFs had been performed. Results showed that OECs (but not ONFs) phagocytose porcine photoreceptor outer segments. The phagocytosis ability was even stronger than RPE (Huo et al., 2012). However, Li et al. also found that OECs and ONFs had synergistic eﬀects in promoting axon regeneration (Li et al., 2005). In the present study, there is roughly 50/50 mixture of OECs and ﬁbroblasts in the transplanted cells, which accords with the cell ratio in our previous study (Huo et al., 2011). DISCUSSION OECs are known for their properties like interacting with the glial scar, stimulating angiogenesis, and promoting axonal outgrowth (Roet and Verhaagen, 2014; Gomez et al., 2018). Fibroblasts, both from the olfactory bulb and other parts of the body, can aﬀect the function of other cells and present a lot of diﬀerent eﬀects. However, as the purpose of this study is to focus on the function of OECs, we referred the mixture of OECs/ONFs as OECs. transplantation of NSCs and OECs might be a possible alternative for the treatment of RDDs. SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fncel. 2020.00052/full#supplementary-material FUNDING The study was funded by the National Key R&D Program of China (2018YFA0107302), the National Natural Science Foundation of China (Nos. 81873688 and 81800874), and the Foundation of Southwest Hospital (No. SWH2017ZDCX2007). AUTHOR CONTRIBUTIONS Since OEC and NSC transplantation have both entered into the clinical research stage, the advancement of the combined transplantation of OECs and NSCs from bench to bedside in the future is promising. As the main eﬀect of cotransplantation was the activation of endogenous stem cells and improvements in the microenvironment, this therapeutic method could prospectively beneﬁt all types of RDDs as well as other retinal lesions, including glaucoma and ocular trauma. Further research is required to conﬁrm this possibility. HX and ZY contributed to the design of the project. WZ and LG contributed to the in vivo experiments and discussed the results. WZ, YL, QL, and YZ contributed to the in vitro experiments. LG and HX summarized the data and contributed in manuscript preparation. All authors read and approved the manuscript. ETHICS STATEMENT The animal study was reviewed and approved by the Institutional Review Board of the Third Military Medical University. The animal study was reviewed and approved by the Institutional Review Board of the Third Military Medical University. CONCLUSION In summary, the combined transplantation of NSCs and OECs better preserved retina than transplantation with NSCs or OECs alone, and this eﬀect was veriﬁed by improved ERGs and increased ONL thickness in the combined transplantation group. Increased endogenous stem cell activation, better maintenance of NSC stemness, and enhanced migration of transplanted cells following combined transplantation might be potential mechanisms. All of these results illustrated that the combined DATA AVAILABILITY STATEMENT All the data used to support the ﬁndings of this study are available from the corresponding author upon reasonable request. REFERENCES S., Cekic, S., and Stamenkovic, M. (2014). Therapeutic Modalities of Exudative Age-related Macular Degeneration. Med. Arch. 68, 204–208. Xie, J., Huo, S., Li, Y., Dai, J., Xu, H., and Yin, Z. Q. (2017). Olfactory ensheathing cells inhibit Gliosis in retinal degeneration by downregulation of the Muller cell notch signaling pathway. Cell Transplant 26, 967–982. doi: 10.3727/ 096368917X694994 McGill, T. J., Cottam, B., Lu, B., Wang, S., Girman, S., Tian, C., et al. (2012). Transplantation of human central nervous system stem cells - neuroprotection in retinal degeneration. Eur. J. Neurosci. 35, 468–477. doi: 10.1111/j.1460-9568. 2011.07970.x Xue, L., Zeng, Y., Li, Q., Li, Y., Li, Z., Xu, H., et al. (2017). Transplanted olfactory ensheathing cells restore retinal function in a rat model of light-induced retinal damage by inhibiting oxidative stress. Oncotarget 8, 93087–93102. doi: 10. 18632/oncotarget.21857 Nazari, H., Zhang, L., Zhu, D., Chader, G. J., Falabella, P., Stefanini, F., et al. (2015). Stem cell based therapies for age-related macular degeneration: the promises and the challenges. Prog. Retin. Eye Res. 48, 1–39. doi: 10.1016/j.preteyeres.2015. 06.004 Yang, H., He, B. R., and Hao, D. J. (2015). Biological roles of olfactory ensheathing cells in facilitating neural regeneration: a systematic review. Mol. Neurobiol. 51, 168–179. doi: 10.1007/s12035-014-8664-2 Nishida, A., Takahashi, M., Tanihara, H., Nakano, I., Takahashi, J. B., Mizoguchi, A., et al. (2000). Incorporation and diﬀerentiation of hippocampus-derived neural stem cells transplanted in injured adult rat retina. Invest. Ophthalmol. Vis. Sci. 41, 4268–4274. Yao, K., Qiu, S., Wang, Y. V., Park, S. J. H., Mohns, E. J., Mehta, B., et al. (2018). Restoration of vision after de novo genesis of rod photoreceptors in mammalian retinas. Nature 560, 484–488. doi: 10.1038/s41586-018-0425-3 Obernier, K., and Alvarez-Buylla, A. (2019). Neural stem cells: origin, heterogeneity and regulation in the adult mammalian brain. Development 146:dev156059. doi: 10.1242/dev.156059 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. 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. Peng, Y., Zhang, Y., Huang, B., Luo, Y., Zhang, M., Li, K., et al. (2014). Survival and migration of pre-induced adult human peripheral blood mononuclear cells in retinal degeneration slow (rds) mice three months after subretinal transplantation. Curr. Stem Cell Res. Ther. 9, 124–133. REFERENCES Subretinal OEC and NSC Co-transplantation Huo, S. J., Li, Y. C., Xie, J., Li, Y., Raisman, G., Zeng, Y. X., et al. (2012). Transplanted olfactory ensheathing cells reduce retinal degeneration in Royal College of Surgeons rats. Curr. Eye Res. 37, 749–758. doi: 10.3109/02713683. 2012.697972 Pinilla, I., Lund, R. D., and Sauve, Y. (2005). Cone function studied with ﬂicker electroretinogram during progressive retinal degeneration in RCS rats. Exp. Eye Res. 80, 51–59. doi: 10.1016/j.exer.2004.08.012 j Qin, Y., Zhang, W., and Yang, P. (2015). Current states of endogenous stem cells in adult spinal cord. J. Neurosci. Res. 93, 391–398. doi: 10.1002/jnr.23480 Jonas, J. B., Cheung, C. M. G., and Panda-Jonas, S. (2017). Updates on the epidemiology of age-related macular degeneration. Asia Pac. J. Ophthalmol. 6, 493–497. Qu, L., Gao, L., Xu, H., Duan, P., Zeng, Y., Liu, Y., et al. (2017). Combined transplantation of human mesenchymal stem cells and human retinal progenitor cells into the subretinal space of RCS rats. Sci. Rep. 7:199. doi: 10.1038/s41598-017-00241-5 Jorstad, N. L., Wilken, M. S., Grimes, W. N., Wohl, S. G., VandenBosch, L. S., Yoshimatsu, T., et al. (2017). Stimulation of functional neuronal regeneration from Muller glia in adult mice. Nature 548, 103–107. doi: 10.1038/nature 23283 Ramon-Cueto, A., and Avila, J. (1998). Olfactory ensheathing glia: properties and function. Brain Res. Bull. 46, 175–187. doi: 10.1016/s0361-9230(97)00463-2 Jung, G., Sun, J., Petrowitz, B., Riecken, K., Kruszewski, K., Jankowiak, W., et al. (2013). Genetically modiﬁed neural stem cells for a local and sustained delivery of neuroprotective factors to the dystrophic mouse retina. Stem Cells Transl. Med. 2, 1001–1010. doi: 10.5966/sctm.2013-0013 Roet, K. C., and Verhaagen, J. (2014). Understanding the neural repair-promoting properties of olfactory ensheathing cells. Exp. Neurol. 261, 594–609. doi: 10. 1016/j.expneurol.2014.05.007 Santos-Ferreira, T., Llonch, S., Borsch, O., Postel, K., Haas, J., and Ader, M. (2016). Retinal transplantation of photoreceptors results in donor-host cytoplasmic exchange. Nat. Commun. 7:13028. doi: 10.1038/ncomms13028 LaVail, M. M. (2001). Legacy of the RCS rat: impact of a seminal study on retinal cell biology and retinal degenerative diseases. Prog. Brain Res. 131, 617–627. doi: 10.1016/s0079-6123(01)31048-8 Strauss, O., Stumpﬀ, F., Mergler, S., Wienrich, M., and Wiederholt, M. (1998). The Royal College of Surgeons rat: an animal model for inherited retinal degeneration with a still unknown genetic defect. Acta Anat. 162, 101–111. doi: 10.1159/000046474 Li, Y., Field, P. M., and Raisman, G. (2005). REFERENCES Olfactory ensheathing cells and olfactory nerve ﬁbroblasts maintain continuous open channels for regrowth of olfactory nerve ﬁbres. Glia 52, 245–251. doi: 10.1002/glia. 20241 Telegina, D. V., Kozhevnikova, O. S., and Kolosova, N. G. (2018). Changes in retinal glial cells with age and during development of age-related macular degeneration. Biochemistry 83, 1009–1017. doi: 10.1134/S000629791809002X Li, Y., Zou, T., Xue, L., Yin, Z. Q., Huo, S., and Xu, H. (2017). TGF-beta1 enhances phagocytic removal of neuron debris and neuronal survival by olfactory ensheathing cells via integrin/MFG-E8 signaling pathway. Mol. Cell. Neurosci. 85, 45–56. doi: 10.1016/j.mcn.2017.08.006 Tian, C., Zhao, T., Zeng, Y., and Yin, Z. Q. (2011). Increased Muller cell de- diﬀerentiation after grafting of retinal stem cell in the sub-retinal space of Royal College of Surgeons rats. Tissue Eng. Part A 17, 2523–2532. doi: 10.1089/ten. TEA.2010.0649 Li, Z., Zeng, Y., Chen, X., Li, Q., Wu, W., Xue, L., et al. (2016). Neural stem cells transplanted to the subretinal space of rd1 mice delay retinal degeneration by suppressing microglia activation. Cytotherapy 18, 771–784. doi: 10.1016/j.jcyt. 2016.03.001 Veleri, S., Lazar, C. H., Chang, B., Sieving, P. A., Banin, E., and Swaroop, A. (2015). Biology and therapy of inherited retinal degenerative disease: insights from mouse models. Dis. Model. Mech. 8, 109–129. doi: 10.1242/dmm.017913 Lin, T. C., Hsu, C. C., Chien, K. H., Hung, K. H., Peng, C. H., and Chen, S. J. (2014). Retinal stem cells and potential cell transplantation treatments. J. Chin. Med. Assoc. 77, 556–561. doi: 10.1016/j.jcma.2014.08.001 Wang, Y. H., Yin, Z. Q., and Wang, Y. (2017). Synergistic eﬀect of olfactory ensheathing cells and alpha-crystallin on restoration of adult rat optic nerve injury. Neurosci. Lett. 638, 167–174. doi: 10.1016/j.neulet.2016.12.042 Madelaine, R., and Mourrain, P. (2017). Endogenous retinal neural stem cell reprogramming for neuronal regeneration. Neural Regen. Res. 12, 1765–1767. doi: 10.4103/1673-5374.219028 Wong, W. L., Su, X., Li, X., Cheung, C. M. G., Klein, R., Cheng, C.-Y., et al. (2014). Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic review and meta-analysis. Lancet Glob. Health 2, e106–e116. doi: 10.1016/S2214-109X(13)70145-1 Marsh, S. E., and Blurton-Jones, M. (2017). Neural stem cell therapy for neurodegenerative disorders: The role of neurotrophic support. Neurochem. Int. 106, 94–100. doi: 10.1016/j.neuint.2017.02.006 Wu, J., Stoica, B. A., and Faden, A. I. (2011). Cell cycle activation and spinal cord injury. Neurotherapeutics 8, 221–228. doi: 10.1007/s13311-011-0028-2 Mavija, M., Alimanovic, E., Jaksic, V., Kasumovic, S. REFERENCES Gomez, R. M., Sanchez, M. Y., Portela-Lomba, M., Ghotme, K., Barreto, G. E., Sierra, J., et al. (2018). Cell therapy for spinal cord injury with olfactory ensheathing glia cells (OECs). Glia 66, 1267–1301. doi: 10.1002/glia.23282 Blasiak, J. (2020). Senescence in the pathogenesis of age-related macular degeneration. Cell. Mol. Life Sci. doi: 10.1007/s00018-019-03420-x [Epub ahead of print]. g g g Gorsuch, R. A., Lahne, M., Yarka, C. E., Petravick, M. E., Li, J., and Hyde, D. R. (2017). Sox2 regulates Muller glia reprogramming and proliferation in the regenerating zebraﬁsh retina via Lin28 and Ascl1a. Exp. Eye Res. 161, 174–192. doi: 10.1016/j.exer.2017.05.012 Bringmann, A., Pannicke, T., Grosche, J., Francke, M., Wiedemann, P., Skatchkov, S. N., et al. (2006). Muller cells in the healthy and diseased retina. Prog. Retin. Eye Res. 25, 397–424. doi: 10.1016/j.preteyeres.2006. 05.003 Graca, A. B., Hippert, C., and Pearson, R. A. (2018). Muller glia reactivity and development of gliosis in response to pathological conditions. Adv. Exp. Med. Biol. 1074, 303–308. doi: 10.1007/978-3-319-75402-4_37 D’Cruz, P. M., Yasumura, D., Weir, J., Matthes, M. T., Abderrahim, H., LaVail, M. M., et al. (2000). Mutation of the receptor tyrosine kinase gene Mertk in the retinal dystrophic RCS rat. Hum. Mol. Genet. 9, 645–651. doi: 10.1093/hmg/9. 4.645 Harada, T., Harada, C., Kohsaka, S., Wada, E., Yoshida, K., Ohno, S., et al. (2002). Microglia-Muller glia cell interactions control neurotrophic factor production during light-induced retinal degeneration. J. Neurosci. 22, 9228–9236. doi: 10.1523/jneurosci.22-21-09228.2002 Gao, L., Chen, X., Tang, Y., Zhao, J., Li, Q., Fan, X., et al. (2015). Neuroprotective eﬀect of memantine on the retinal ganglion cells of APPswe/PS1DeltaE9 mice and its immunomodulatory mechanisms. Exp. Eye Res. 135, 47–58. doi: 10. 1016/j.exer.2015.04.013 Haragopal, H., Yu, D., Zeng, X., Kim, S.-W., Han, I.-B., Ropper, A. E., et al. (2015). Stemness enhancement of human neural stem cells following bone marrow MSC coculture. Cell Transplant 24, 645–659. doi: 10.3727/096368915X687561 j Geller, S., Kolasa, E., Tillet, Y., Duittoz, A., and Vaudin, P. (2013). Olfactory ensheathing cells form the microenvironment of migrating GnRH-1 neurons during mouse development. Glia 61, 550–566. doi: 10.1002/glia. 22455 Huo, S. J., Li, Y., Raisman, G., and Yin, Z. Q. (2011). Transplanted olfactory ensheathing cells reduce the gliotic injury response of Muller cells in a rat model of retinitis pigmentosa. Brain Res. 1382, 238–244. doi: 10.1016/j.brainres.2010. 12.079 March 2020 \| Volume 14 \| Article 52 14 Frontiers in Cellular Neuroscience \| www.frontiersin.org Zhai et al. REFERENCES doi: 10.2174/ 1574888x09666131219115125 Copyright © 2020 Zhai, Gao, Qu, Li, Zeng, Li, Xu and Yin. 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. Copyright © 2020 Zhai, Gao, Qu, Li, Zeng, Li, Xu and Yin. 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. Perlman, I. (1995). “The Electroretinogram: ERG,” in Webvision: The Organization of the Retina and Visual System, eds H. Kolb, E. Fernandez, and R. Nelson (Salt Lake City, UT: University of Utah Health Sciences Center). March 2020 \| Volume 14 \| Article 52 Frontiers in Cellular Neuroscience \| www.frontiersin.org 15
https://openalex.org/W1990538908	https://epub.ub.uni-muenchen.de/33295/1/10.1371_journal.pone.0115488.pdf	English	null	Interferon Beta and Vitamin D Synergize to Induce Immunoregulatory Receptors on Peripheral Blood Monocytes of Multiple Sclerosis Patients	PloS one	2,014	cc-by	7,636	RESEARCH ARTICLE OPEN ACCESS Citation: Waschbisch A, Sanderson N, Krumbholz M, Vlad G, Theil D, et al. (2014) Interferon Beta and Vitamin D Synergize to Induce Immunoregulatory Receptors on Peripheral Blood Monocytes of Multiple Sclerosis Patients. PLoS ONE 9(12): e115488. doi:10.1371/journal.pone. 0115488 Anne.Waschbisch@uk-erlangen.de ¤a Current address: Novartis Institute of Biomedical Research, Basel, Switzerland ¤b Current address: Dept. of Neurology, Caritas Hospital, Bad Mergentheim, Germany Editor: Steven Jacobson, National Institutes of Health, United States of America Editor: Steven Jacobson, National Institutes of Health, United States of America Received: August 3, 2014 Accepted: November 24, 2014 Published: December 31, 2014 Anne Waschbisch1, Nicholas Sanderson2, Markus Krumbholz3, George Vlad4, Diethilde Theil5¤a, Stefan Schwab1, Mathias Ma¨urer1¤b, Tobias Derfuss2 1. Dept. of Neurology, Friedrich-Alexander University Erlangen-Nu¨rnberg, Erlangen, Germany, 2. Dept. of Neurology and Biomedicine, University Hospital Basel, Basel, Switzerland, 3. Institute of Clinical Neuroimmunology, Klinikum Grosshadern, Ludwig Maximilian University, Munich, Germany, 4. Dept. of Pathology & Cell Biology, Columbia University, New York, New York, United States of America, 5. Dept. of Neurology, Klinikum Grosshadern, Ludwig Maximilian University, Munich, Germany Interferon Beta and Vitamin D Synergize to Induce Immunoregulatory Receptors on Peripheral Blood Monocytes of Multiple Sclerosis Patients Anne Waschbisch1, Nicholas Sanderson2, Markus Krumbholz3, George Vlad4, Diethilde Theil5¤a, Stefan Schwab1, Mathias Ma¨urer1¤b, Tobias Derfuss2 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS honoraria from Bayer, BiogenIdec, Bo¨hringer, Genzyme, Novartis, SanofiAventis, MerckSerono, and Teva. GV has no competing interests. SS has no competing interests. DT is an employee of Novartis Pharma. TD served on advisory boards for Novartis Pharma, Merck-Serono GmbH, Bayer Schering Pharma, Mitsubishi Pharma, Biogen Idec, GeNeuro, Genzyme, and TEVA. He received travel support from Biogen Idec, Bayer Schering Pharma, Genzyme, and Merck Serono GmbH. He received research and/or unrestricted grants from Novartis Pharma, Biogen Idec, Merck Serono GmbH, the German Research Foundation, the Swiss National Foundation, the Swiss MS Society, and the European Union. This does not alter the authors’ adherence to PLOS ONE policies on sharing data and materials. Introduction Immunoglobulin-like transcripts (ILT), also known as leukocyte immunoglobu- lin-like receptors (LILR) belong to a large family of activating and inhibitory receptors that modulate the threshold and amplitude of immune cell activation [1, 2]. Immune inhibitory ILT family members such as ILT3 (CD85k, LILRB4) and ILT4 (CD85d, LILRB2) are characterized by an extracellular immunoglobu- lin-like domain responsible for ligand-binding, and a long cytoplasmatic tail containing immunoreceptor-tyrosine based inhibitory motifs (ITIM) which recruit inhibitory phosphatases and transduce a negative signal into the cell [3]. A high expression of ILT3 and ILT4 on the cell surface of antigen-presenting cells renders them tolerogenic, inhibiting T cell proliferation and favouring the generation of CD8+ T suppressor cells [4, 5, 6, 7]. Multiple sclerosis (MS) is a chronic, inflammatory, demyelinating disease of the central nervous system (CNS). It has been hypothesized that an aberrant activation of autoreactive T cells in the peripheral immune compartment due to failure of peripheral tolerance mechanisms triggers T-cell mediated CNS inflammation in MS [8]. Accordingly, impaired expression of molecules involved in the regulation of T cell activation such as PD-1 and other B7 family members has been found to exacerbate CNS inflammation in animal models of the disease [9, 10, 11]. While the role of ILT3 and ILT4 as regulators of T cell activation and mediators of tolerance is well recognized in transplant and cancer immunology [12, 13, 14], little is known about the influence of these receptors on autoimmune diseases. Here we show that the beneficial effects of IFN beta in multiple sclerosis may in part be mediated by modulation of ILT3 and ILT4 expression on APC.The role of immunoregulatory receptors in CNS inflammation is further highlighted by their enrichment in cerebrospinal fluid and an upregulation of ILT3, ILT4, and B7-H3 in acute MS lesions. Abstract Immunoglobulin-like transcript (ILT) 3 and 4 are inhibitory receptors that modulate immune responses. Their expression has been reported to be affected by interferon, offering a possible mechanism by which this cytokine exerts its therapeutic effect in multiple sclerosis, a condition thought to involve excessive immune activity. To investigate this possibility, we measured expression of ILT3 and ILT4 on immune cells from multiple sclerosis patients, and in post-mortem brain tissue. We also studied the ability of interferon beta, alone or in combination with vitamin D, to induce upregulation of these receptors in vitro, and compared expression levels between interferon-treated and untreated multiple sclerosis patients. In vitro interferon beta treatment led to a robust upregulation of ILT3 and ILT4 on monocytes, and dihydroxyvitamin D3 increased expression of ILT3 but not ILT4. ILT3 was abundant in demyelinating lesions in postmortem brain, and expression on monocytes in the cerebrospinal fluid was higher than in peripheral blood, suggesting that the central nervous system milieu induces ILT3, or that ILT3 positive monocytes preferentially enter the brain. Our data are consistent with involvement of ILT3 and ILT4 in the modulation of immune responsiveness in multiple sclerosis by both interferon and vitamin D. Copyright: 2014 Waschbisch et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and repro- duction 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 in part funded by the Merck Serono GmbH, a division of Merck KGaA, Darmstadt, Germany, and the Else-Kro¨ner Fresenius Stiftung (A_123). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manu- script. Competing Interests: AW has received funding for travel or speaker honoraria from Bayer Schering Pharma, Biogen Idec/Elan Corporation, Genzyme, Sanofi-aventis/Teva Pharmaceutical Industries Ltd., and Merck Serono as well as research support from Biogen Idec/Elan Corporation and Merck Serono GmbH. NS has no competing interests. MK has received support from Novartis Pharma. MM has received speaker 1 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 Additive effects of IFN beta and Vitamin D3 on monocytic ILT3 expression 1a,25 Dihydroxyvitamin D3 (1a,25(OH)2D3) has been reported to induce ILT3 but not ILT4 expression on antigen-presenting cells [16]. Using flow cytometry we assessed the impact of 1a,25(OH)2D3 (100 nM) alone or in combination with IFN beta (1000 IE) on the expression of these immunoregulatory molecules on the surface of CD14+ monocytes derived from a second cohort of RRMS patients (n59, Fig. 2) and healthy controls (n58, S1 Fig.). In line with previous findings 1a,25(OH)2D3 was found to be a potent inducer of ILT3 but not ILT4 expression. Combined stimulation with IFN beta had a strong additive effect, leading to high levels of membrane bound ILT3 protein on CD14+ cells (Fig. 2). In contrast ILT4 levels, which were downregulated in 1a,25(OH)2D3 only treated cells, were restored back to normal levels in the presence of both reagents (Fig. 2). Similar trends were observed in healthy controls; however, the downregulation of ILT4 by 1a,25(OH)2D3 and the restoration of ILT4 surface levels by a combined treatment did not reach the level of significance in controls (S1 Fig.). IFN Beta and Vitamin D Modulate Expression of ILT3 in MS fluorescent index is shown (grey bar: control; black bar: IFN beta). (C) The percentage of CD14+ monocytes expressing the immune inhibitory ILT is strongly increased following incubation with IFN beta as assessed by flow cytometry (n524; grey bar: control; black bar: IFN beta 1000 IE/ml, 16 h). (D) A representative flow cytometry experiment is shown. Please note, for simplification the isotypic control for ILT4 (IgG2a) is not shown in this picture. (E) IFN beta induces ILT3 mRNA in purified CD14+ monocytes derived from healthy donors (n56). Relative expression of ILT3 mRNA was assessed by quantitative PCR (Mann-Whitney U test; p,0,05, ** p,0,005; bars are presented as mean ¡ S.E.M) doi:10.1371/journal.pone.0115488.g001 doi:10.1371/journal.pone.0115488.g001 Furthermore, we could not detect any differences regarding the baseline expression of B7-H1 and B7-H3 (data not shown). In vitro IFN beta treatment of PBMC resulted in a strong upregulation of both ILT3 and ILT4 on CD14+ monocytes (Fig. 1 B, D). The percentage of ILT3+ monocytes was almost doubled from 35% to 68% and the percentage of ILT4 monocytes increased from 47% to 66% (Fig. 1C). The extent of ILT3 and ILT4 induction was comparable to that of B7-H1, a known target protein of IFN beta [15], whereas the expression of the novel B7-homologue B7-H3 remained unaffected (Fig. 1 B). Upregulation of ILT3 by IFN beta was induced in isolated monocytes indicating that lymphocytes do not contribute to this effect (Fig. 1 E). Upregulation of CD86 expression was monitored as a positive control in each experiment (data not shown). ILT3 and ILT4 expression on monocytes is upregulated by in vitro IFN beta treatment CD14+ monocytes derived from treatment-nai¨ve patients with RRMS (n524) or CIS (n56) and healthy controls (n514) were analyzed for surface expression of ILT3 and ILT4 by flow cytometry. Baseline expression of ILT3 and ILT4 as assessed by comparing the specific fluorescent indices (SFI) did not differ significantly between these groups (Fig. 1A). Regarding the percentage of ILT3 positive cells, CIS patients tended to have higher numbers of ILT3+CD14+ cells (mean +/-S.E.M.: 55.03+/25.85; p,0.05) compared to RRMS patients (34.78 +/ 25.73) and healthy controls (25.54 +/24.16). No significant differences concerning the percentages of ILT4+CD14+ cells were observed between these groups (data not shown). PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 2 / 15 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Fig. 1. IFN beta induces ILT3 and ILT4 expression in monocytes. (A) The baseline expression of the immune inhibitory receptors ILT3 and ILT4 on peripheral blood derived CD14+ monocytes assessed by comparing the specific fluorescent index (SFI 5 geo mean ILT/geo mean IgG1) does not differ significantly between RRMS (n524) or CIS patients (n56) and healthy donors (HD, n514) as demonstrated by flow cytometry. (B) In vitro IFN beta (1000IE/ml, 16 h) treatment induces ILT3 and to a lesser extent ILT4 protein expression in CD14+ monocytes from RRMS patients (n524). In addition the monocytic expression of the coinhibitory B7 family member B7-H1 is upregulated, whereas B7-H3 is not significantly affected. The specific Fig. 1. IFN beta induces ILT3 and ILT4 expression in monocytes. (A) The baseline expression of the immune inhibitory receptors ILT3 and ILT4 on peripheral blood derived CD14+ monocytes assessed by comparing the specific fluorescent index (SFI 5 geo mean ILT/geo mean IgG1) does not differ significantly between RRMS (n524) or CIS patients (n56) and healthy donors (HD, n514) as demonstrated by flow cytometry. (B) In vitro IFN beta (1000IE/ml, 16 h) treatment induces ILT3 and to a lesser extent ILT4 protein expression in CD14+ monocytes from RRMS patients (n524). In addition the monocytic expression of the coinhibitory B7 family member B7-H1 is upregulated, whereas B7-H3 is not significantly affected. The specific PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 3 / 15 ILT3 mRNA is upregulated in PBMC from IFN beta treated RRMS patients To corroborate our in vitro findings, we analyzed the expression of ILT3 mRNA in PBMC derived from treatment nai¨ve RRMS patients and patients undergoing long-term treatment with either IFN beta-1b (Betaferon) or IFN beta-1a (Rebif) by quantitative PCR. In line with our in vitro findings, expression of ILT3 mRNA PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 4 / 15 doi:10.1371/journal.pone.0115488.g003 ILT3 expressing monocytes in CSF Flow cytometry of paired blood and CSF specimens demonstrated a higher level of ILT3 surface protein on CSF-derived than on blood-derived CD14+ monocytes, both in patients with and without neuroinflammatory diseases (Fig. 4A, B). Accordingly the percentage of ILT3+ CD14+ monocytes in the CSF ranged from 45% up to 95% demonstrating an enrichment of ILT3+ monocytes in the CSF. There were no statistically significant differences between patients with non- inflammatory neurological diseases (n514) and those with inflammatory neurological diseases (n58) concerning the percentage of ILT3+ monocytes or the level of ILT3 expression. This holds true for both, CSF and blood samples in this cohort (Fig. 4A). IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Soluble ILT3 serum concentrations do not seem to be affected by IFN beta treatment Soluble ILT3 serum concentrations do not seem to be affected by IFN beta treatment In addition to membrane-bound ILT3, soluble ILT3 (sILT3) has been described and presumed to result from proteolytic cleavage from the cell surface or from alternatively spliced ILT3 isoforms that lack the transmembrane domain. Previously, sILT3 was detected in sera from patients with various malignancies, giving rise to the hypothesis that sILT3 may play a role in the immune evasion of tumor cells [12]. We asked whether patients with multiple sclerosis (n515) would show detectable serum levels of sILT3 and whether these levels would be modulated by IFN beta treatment. Serum from melanoma patients about to start (n53) or already undergoing IFN alpha therapy (n58) served as a control. We failed to detect sILT3 in all but 2 patients with RRMS which is in the range of healthy controls [12]. From one of the RRMS patients displaying detectable levels of sILT3 protein two samples from before and 3 months after initiation of IFN beta therapy were available. Interestingly these showed a remarkable consistency over time without any impact of IFN beta therapy (before: 26 ng/ml, after: 28 ng/ ml; S1 Table). Also from 8 patients with malignant melanoma, only 2 had detectable levels of sILT3. An effect of IFN alpha treatment was not detected in the 3 melanoma patients in which pre- and post-treatment samples were available (S1 Table). PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Fig. 2. Effects of 1a,25 Dihydroxyvitamin D3 in combination with IFN beta on monocytic ILT3 and ILT4 expression. PBMC derived from RRMS patients (n59) were stimulated with 1a,25 Dihydroxyvitamin D3 (100 nM) and/or IFN beta (1000 IE) over a 48 h period. ILT3 and ILT4 protein expression on CD14+ cells was assessed by flow cytometry. The fold induction is shown (mean + SEM). A repeated measurement ANOVA with Bonferroni multiple comparison test was performed to assess statistical significance (* p,0,05; *p,0.005). Fig. 2. Effects of 1a,25 Dihydroxyvitamin D3 in combination with IFN beta on monocytic ILT3 and ILT4 expression. PBMC derived from RRMS patients (n59) were stimulated with 1a,25 Dihydroxyvitamin D3 (100 nM) and/or IFN beta (1000 IE) over a 48 h period. ILT3 and ILT4 protein expression on CD14+ cells was assessed by flow cytometry. The fold induction is shown (mean + SEM). A repeated measurement ANOVA with Bonferroni multiple comparison test was performed to assess statistical significance ( p,0,05; p,0.005). doi:10.1371/journal.pone.0115488.g002 was 2.5 fold higher in IFN beta treated compared to untreated RRMS patients (Fig. 3). Of note, there were no differences between the two preparations of IFN beta concerning the expression of ILT3 mRNA. Fig. 3. ILT3 mRNA is upregulated in PBMC derived from IFN beta treated RRMS patients. Relative expression of ILT3 mRNA in PBMC derived from IFN beta treated patients with RRMS (n510) and sex- and age matched controls (n510) as assessed by quantitative PCR (Mann-Whitney U test, p,0,005). Bars are presented as mean ¡ S.E.M. doi:10.1371/journal.pone.0115488.g003 Fig. 3. ILT3 mRNA is upregulated in PBMC derived from IFN beta treated RRMS patients. Relative expression of ILT3 mRNA in PBMC derived from IFN beta treated patients with RRMS (n510) and sex- and age matched controls (n510) as assessed by quantitative PCR (Mann-Whitney U test, p,0,005). Bars are presented as mean ¡ S.E.M. Fig. 3. ILT3 mRNA is upregulated in PBMC derived from IFN beta treated RRMS patients. Relative expression of ILT3 mRNA in PBMC derived from IFN beta treated patients with RRMS (n510) and sex- and age matched controls (n510) as assessed by quantitative PCR (Mann-Whitney U test, p,0,005). Bars are presented as mean ¡ S.E.M. doi:10.1371/journal.pone.0115488.g003 doi:10.1371/journal.pone.0115488.g003 5 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Fig. 4. CSF monocytes express high levels of ILT3. (A) Paired blood and CSF samples were analyzed for expression of ILT3 on CD14+ monocytes by flow cytometry. The specific fluorescent index is shown. Patients were categorized according to their diagnosis as having non-inflammatory neurological diseases (NIND, n514; marked by dot) or inflammatory neurological disease (IND, n58) including 5 MS patients (marked by x) and 3 patients with other inflammatory neurological disease (diamond). A Mann-Whitney U test was performed to assess statistical significance. (B) Histogram of a representative staining in a patient with NIND. Fig. 4. CSF monocytes express high levels of ILT3. (A) Paired blood and CSF samples were analyzed for expression of ILT3 on CD14+ monocytes by flow cytometry. The specific fluorescent index is shown. Patients were categorized according to their diagnosis as having non-inflammatory neurological diseases (NIND, n514; marked by dot) or inflammatory neurological disease (IND, n58) including 5 MS patients (marked by x) and 3 patients with other inflammatory neurological disease (diamond). A Mann-Whitney U test was performed to assess statistical significance. (B) Histogram of a representative staining in a patient with NIND. doi:10.1371/journal.pone.0115488.g004 doi:10.1371/journal.pone.0115488.g004 and in the one chronic inactive lesion compared to control brain and normal appearing white matter of a MS brain (Fig. 6). B7-H1 has been previously reported to be upregulated on microglial/macrophage-like cells within active MS lesions. Due to the limited amount of human material, we here refrained from reproducing the results on B7-H1 obtained by Ortler et al. [17]. Expression of the immunoregulatory molecules ILT3, ILT4, and B7-H3 in MS lesions and control tissue Lesions were characterized and graded for the activity of demyelination by luxol fast blue, hematoxylin/eosin staining and CD68 immunohistochemistry (Fig. 5). Lesions were characterized as active when they showed recent demyelination, macrophage infiltration, and perivascular cuffing. Seven lesions were classified as active lesions. One lesion was defined as chronic inactive. As controls, three samples from white matter of control brains and one sample from normal appearing white matter of a MS brain were included. Quantitative RT-PCR revealed an upregulation of ILT3, ILT4, and B7-H3 transcripts in active lesions PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 6 / 15 doi:10.1371/journal.pone.0115488.g005 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 Discussion The immune system has to preserve immune tolerance to self while maintaining the ability to fight foreign pathogens. In addition to central tolerance mechanisms, Fig. 5. Characterization of MS tissue blocks. (A) Overview of a demyelinating lesion with perivascular cuffing. Myelin is stained blue with LFB. Lesion border is indicated by a dotted line. Perivascular cuff is marked by an arrow. Scale bar indicates 400 mm. (B) CD68 positive macrophages in demyelinated lesion. CD68 positive cells are stained brown and marked by arrows. Scale bar indicates 50 mm. (C) Demyelinated lesion in higher magnification with cells reminiscent of macrophages that phagocytosed LFB positive myelin debris (marked with arrows). Scale bar indicates 50 mm. Fig. 5. Characterization of MS tissue blocks. (A) Overview of a demyelinating lesion with perivascular cuffing. Myelin is stained blue with LFB. Lesion border is indicated by a dotted line. Perivascular cuff is marked by an arrow. Scale bar indicates 400 mm. (B) CD68 positive macrophages in demyelinated lesion. CD68 positive cells are stained brown and marked by arrows. Scale bar indicates 50 mm. (C) Demyelinated lesion in higher magnification with cells reminiscent of macrophages that phagocytosed LFB positive myelin debris (marked with arrows). Scale bar indicates 50 mm. Fig. 5. Characterization of MS tissue blocks. (A) Overview of a demyelinating lesion with perivascular cuffing. Myelin is stained blue with LFB. Lesion border is indicated by a dotted line. Perivascular cuff is marked by an arrow. Scale bar indicates 400 mm. (B) CD68 positive macrophages in demyelinated lesion. CD68 positive cells are stained brown and marked by arrows. Scale bar indicates 50 mm. (C) Demyelinated lesion in higher magnification with cells reminiscent of macrophages that phagocytosed LFB positive myelin debris (marked with arrows). Scale bar indicates 50 mm. doi:10.1371/journal.pone.0115488.g005 7 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Fig. 6. ILT3, ILT4 and B7-H3 mRNA expression in MS tissue samples. Expression of ILT3, ILT4, and B7- H3 in CNS autopsy samples measured by quantitative PCR and normalized to PUM1. For each gene, results from normal white matter from control brains are shown on the left (open circles), and results from 7 active lesions (black squares) and 1 inactive lesion (grey square) and 1 normal appearing white matter (open square) are shown on the right. Discussion Vertical axis shows the expression relative to PUM1 calculated according to the formula in the Methods section (p,0.05, Mann-Whitney U test). doi:10 1371/journal pone 0115488 g006 Fig. 6. ILT3, ILT4 and B7-H3 mRNA expression in MS tissue samples. Expression of ILT3, ILT4, and B7- H3 in CNS autopsy samples measured by quantitative PCR and normalized to PUM1. For each gene, results from normal white matter from control brains are shown on the left (open circles), and results from 7 active lesions (black squares) and 1 inactive lesion (grey square) and 1 normal appearing white matter (open square) are shown on the right. Vertical axis shows the expression relative to PUM1 calculated according to the formula in the Methods section (p,0.05, Mann-Whitney U test). Fig. 6. ILT3, ILT4 and B7-H3 mRNA expression in MS tissue samples. Expression of ILT3, ILT4, and B7- Fig. 6. ILT3, ILT4 and B7-H3 mRNA expression in MS tissue samples. Expression of ILT3, ILT4, and B7- H3 in CNS autopsy samples measured by quantitative PCR and normalized to PUM1. For each gene, results from normal white matter from control brains are shown on the left (open circles), and results from 7 active lesions (black squares) and 1 inactive lesion (grey square) and 1 normal appearing white matter (open square) are shown on the right. Vertical axis shows the expression relative to PUM1 calculated according to the formula in the Methods section (p,0.05, Mann-Whitney U test). doi:10.1371/journal.pone.0115488.g006 the control of T cell activation in the periphery is essential to prevent autoimmunity [18]. Both antigen-presenting cells and T cells are endowed with a large repertoire of activating and inhibitory molecules. It is the integration of these signals that keeps T cells in check and dictates the outcome of an APC-T cell encounter [11, 19]. Here we report that IFN beta, an immunomodulatory drug widely used to treat multiple sclerosis, induces the expression of the immune inhibitory receptors ILT3 and ILT4 on monocytes. Our findings confirm in vitro data by Jensen et al. [20] and further show that upregulation of monocytic ILT3 expression by IFN beta is not an in vitro artefact, but can also be found in patients with relapsing-remitting multiple sclerosis undergoing IFN beta therapy as demonstrated by quantitative PCR. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS act together in modulating disease activity in MS [27]. However, on a cellular level the mechanisms of this interaction remain elusive. The bioactive form of vitamin D, 1a,25(OH)2D3 has previously been reported to be a potent inducer of ILT3 on APC [16]. Here we demonstrate that IFN beta and 1a,25(OH)2D3 work together in inducing ILT3 receptor expression in monocytes. Thereby, vitamin D co- treatment could contribute to the beneficial effects of the disease-modifying drug. Interestingly, IFN beta and 1a,25(OH)2D3 have contrasting effects on ILT4 expression levels on monocytes. By counteracting the effects of 1a,25(OH)2D3 on ILT4 expression IFN beta may further contribute to the tolerogenic properties of these cells. Interestingly, ILT3 has bidirectional signalling properties [28]. In addition to the signal transduced via its intracellular domain, the extracellular Ig-like domain of ILT3 binds to T cells and may directly modulate. [28]. Accordingly, soluble forms of ILT3 are effective inhibitors of T cell proliferation, even in the absence of APC and may be of interest for future therapeutic use as immunomodulators in autoimmune diseases. [6, 12, 28, 29]. So far, ILT3 serum levels have not yet been systematically analyzed in patients with autoimmune diseases. In our study only 2 out of 15 MS patients tested positive for soluble ILT3 in serum, which is in the range of healthy individuals [12]. As a positive control we studied patients with melanoma. In contrast to a previous publication [12], the rate of detectable sILT3 in serum was low in this cohort (,25%). The discrepancy of our findings may be due to the fact that the previous publication was confined to patients with advanced-stage melanoma, while our cohort also included patients that, although at high risk for spreading, still had localized disease. In addition the patients examined by Suciu-Foca et al were enrolled for treatment with high-dose IL-2 which may also account for higher sILT3 serum levels in their cohort [12]. Current concepts of MS pathogenesis presume that breakdown of immune tolerance in the periphery is an initiating step in the cascade that finally leads to CNS infiltration by autoreactive T cells, demyelination and axonal loss [11]. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 Discussion In addition, we demonstrate that the extent of ILT3 and ILT4 upregulation in monocytes is comparable to that of the coinhibitory molecule B7-H1 which has been reported as a target of IFN beta therapy in MS [21]. In contrast, the expression level of the coinhibitory B7 homologue B7-H3 remained unaffected, pointing to a selective effect of IFN beta on these immunomodulatory molecules and monocyte immunobiology. the control of T cell activation in the periphery is essential to prevent autoimmunity [18]. Both antigen-presenting cells and T cells are endowed with a large repertoire of activating and inhibitory molecules. It is the integration of these signals that keeps T cells in check and dictates the outcome of an APC-T cell encounter [11, 19]. Here we report that IFN beta, an immunomodulatory drug widely used to treat multiple sclerosis, induces the expression of the immune inhibitory receptors ILT3 and ILT4 on monocytes. Our findings confirm in vitro data by Jensen et al. [20] and further show that upregulation of monocytic ILT3 expression by IFN beta is not an in vitro artefact, but can also be found in patients with relapsing-remitting multiple sclerosis undergoing IFN beta therapy as demonstrated by quantitative PCR. In addition, we demonstrate that the extent of ILT3 and ILT4 upregulation in monocytes is comparable to that of the coinhibitory molecule B7-H1 which has been reported as a target of IFN beta therapy in MS [21]. In contrast, the expression level of the coinhibitory B7 homologue B7-H3 remained unaffected, pointing to a selective effect of IFN beta on these immunomodulatory molecules and monocyte immunobiology. Vitamin D insufficiency has emerged as a potential risk factor for multiple sclerosis and may be associated with a higher disease activity [22, 23, 24, 25]. Larger studies addressing effects of oral high-dose vitamin D are on their way [26], including a phase II trial of vitamin D and IFN beta cotreatment (SOLAR, NCT01285401). A previous clinical study suggested that vitamin D and IFN beta PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 8 / 15 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS ILT3 in the basic mechanisms of CNS immune surveillance. However, since CSF was not available from healthy donors this remains speculative. Silencing of ILT3 in monocyte-derived dendritic cells potentiates stimulus- induced release of chemokines, that may be involved in T cell trafficking to the CNS [5]. Therefore, ILT3 expressing monocytes within CSF-filled spaces have the potential not only to dampen T cell activity but also to counteract the attraction of pathogenic cells. Further evidence for a possible role of ILT3 at the site of inflammation in MS comes from our expression analysis demonstrating the presence of ILT3 in active MS lesions but not in normal-appearing white matter and white matter from control brains. Since immunohistochemical staining of ILT3 in brain tissue could not be established with the available antibodies we cannot tell the exact cellular source of ILT3 expression. However, since ILT3 expression is confined to professional and non-professional APC [2, 29] it can be assumed that macrophages and microglia account for a large fraction of the ILT3- expressing cells. Similar to ILT3, ILT4 was also found to be upregulated in active MS lesions.However, in contrast to ILT3, ILT4 was not analyzed on CSF monocytes in this study which limits the conclusions drawn from this finding. This work provides in vitro and ex vivo evidence for an induction of the immune inhibitory molecules ILT3 and ILT4 on APCs by IFN beta. The IFN mediated induction of ILT3 can also be potentiated by vitamin D. Increased expression of ILT3 on CSF monocytes and in MS lesions together with other inhibitory receptors indicate a pivotal role of these molecules in the regulation of neuroinflammation. Targeting ILT3 may therefore represent a future therapeutic option in autoimmune diseases. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 Given the potent inhibitory capacity of ILT3 on T cell activation and proliferation, upregulation of this immune inhibitory receptor on monocytes in IFN beta- treated MS patients may indeed represent a therapeutic mechanism that helps to prevent overshooting immune responses. Demonstration of increased ILT3 expression on CSF monocytes and within active MS lesions suggests that the immunomodulatory effects of ILT3 are not restricted to the peripheral immune compartment but may also play a role at the site of inflammation in MS. Compared to the peripheral blood, CSF monocytes display high levels of ILT3 on their surface, with ILT3+ monocytes accounting for approximately two thirds of CD14+ cells in the CSF. This enrichment of ILT3+ monocytes in CSF can be observed not only in patients with inflammatory but also in individuals with non- inflammatory neurological diseases. Accordingly, the high expression of ILT3 on CSF monocytes seems not to be reactive to neuroinflammation but rather a primary attribute of monocytes in the CSF, suggesting a possible involvement of PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 9 / 15 Clinical Samples The study was approved by the local ethics committee (Nr. 4203, University Erlangen-Nuremburg). Written informed consent was obtained from all patients and and all clinical investigations have been conducted according to the principles expressed in the Declaration of Helsinki. Patients with relapsing-remitting multiple sclerosis (RRMS) according to the revised McDonald criteria (2005) [30] and clinically isolated syndrome (CIS) were eligible for participation in this study. All patients were nai¨ve to immunomodulatory (IFN beta, glatiramer acetate, dimethylfumarate, teriflunomide, fingolimod, natalizumab, alemtuzumab) and immunosuppressive drugs (mitoxantrone, cyclophosphamide) except for patients presented in Fig. 3 from whom samples were obtained before and at least 3 months after initiation of IFN beta treatment and patients whose serum was obtained for sILT3 analysis. The individual treatment status with IFN beta in serum donors is stated in S1 Table. Patients treated with intravenous glucocorticoids or plasmapheresis during the preceding six weeks were excluded. In addition age- and sex-matched healthy volunteers were recruited. Control sera were obtained from patients with malignant melanoma treated at the Ludwig- glucocorticoids or plasmapheresis during the preceding six weeks were excluded. In addition age- and sex-matched healthy volunteers were recruited. Control sera were obtained from patients with malignant melanoma treated at the Ludwig- PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 10 / 15 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS Maximilian University of Munich and the University of Regensburg. Again the individual treatment status with IFN is indicated in S1 Table. Maximilian University of Munich and the University of Regensburg. Again the individual treatment status with IFN is indicated in S1 Table. CSF and paired blood samples were collected from patients that were referred to the Dept. of Neurology of the University of Erlangen for diagnostic lumbar puncture. Patients were categorized into two groups: inflammatory neurological diseases (IND, e.g. viral encephalitis, multiple sclerosis) and non-inflammatory neurological diseases (NIND, e.g. normal pressure hydrocephalus). None of the CSF donors were undergoing IFN beta treatment. Nine MS and three control tissue samples were supplied by the UK Multiple Sclerosis Tissue Bank (UK Multicentre Research Ethics Committee, MREC/02/2/ 39), funded by the Multiple Sclerosis Society of Great Britain and Northern Ireland (registered charity 207495). Autoptic brain samples were analysed by luxol fast blue, hematoxylin/eosin staining and CD68 immunohistochemistry as previously described [31]. Information on the pre-mortem treatment of these patients with disease modifying drugs was not available. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 Clinical Samples Isolation and cultivation of peripheral blood mononuclear cells PBMC were isolated by centrifugation on a Lymphoprep (Fresenius Kabi Norge AS, Oslo, Norway) density gradient and cultured in RPMI 1640 (Gibco Invitrogen GmbH, Karlsruhe, Germany) supplemented with 10% of heat inactivated human AB serum (PAA, Pasching, Austria), glutamine and antibiotics. Cells were grown at a densitiy of 26106/ml in 12-well plates (Gibco Invitrogen GmbH, Karlsruhe, Germany) under standard conditions (37uC, 5% CO2). For some experiments monocytes were purified from PBMC using anti-CD14 microbeads (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany) and magnetic cell separation (MACS technology). IFN beta-1a (Rebif, 1000 IE/ml) or 1a,25-Dihydroxyvitamin D3 (100 nM) were added to the medium as indicated. Isolation and cultivation of peripheral blood mononuclear cells PBMC were isolated by centrifugation on a Lymphoprep (Fresenius Kabi Norge AS, Oslo, Norway) density gradient and cultured in RPMI 1640 (Gibco Invitrogen GmbH, Karlsruhe, Germany) supplemented with 10% of heat inactivated human AB serum (PAA, Pasching, Austria), glutamine and antibiotics. Cells were grown at a densitiy of 26106/ml in 12-well plates (Gibco Invitrogen GmbH, Karlsruhe, Germany) under standard conditions (37uC, 5% CO2). For some experiments monocytes were purified from PBMC using anti-CD14 microbeads (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany) and magnetic cell separation (MACS technology). IFN beta-1a (Rebif, 1000 IE/ml) or 1a,25-Dihydroxyvitamin D3 (100 nM) were added to the medium as indicated. Sandwich ELISA detection of soluble ILT3 Soluble ILT3 ELISA was performed as previously described [12]. In brief, Maxisorp 96-well plates (Nalge Nunc International) were coated overnight with 1.0 mg/well anti-ILT3 mAb (clone ZL5.7). Free binding sites were blocked with 5% BSA solution for 1 h at room temperature. After washing with PBS with 0.1% Tween 20, serum samples were added at a 1:3 and 1:9 dilution and incubated for 1 hour. Detection was performed using 100 ml of biotinylated polyclonal anti- ILT3 antibody (R&D Systems) at a concentration of 33 ng/ml followed by HRP- conjugated streptavidin (BD Biosciences). Tetramethylbenzidine was used as a substrate. RNA isolation and quantitative PCR RNA from PBMCs was isolated using the Rneasy Mini Kit (Qiagen, Hilden, Germany). RNA from frozen tissue sections was purified using RNAzol (Sigma- Aldrich, Taufkirchen, Germany) according to the manufacturer’s instructions, but with two subsequent phase separation steps to improve RNA purity. RNA was transcribed using random hexamers and MuLV reverse transcriptase (Promega, Mannheim, Germany) or high capacity cDNA Kit (ABI, Darmstadt, Germany). For expression analysis in PBMCs, PCR and quantification of PCR products was performed using Ssofast EVA Green Supermix (Biorad Laboratories, Hercules, CA, USA) on a MyIQ5 Real Time PCR System. The following primers for ILT3 were used during the amplification: forward, 59-CTGCCGAGTCCTCTT- GTGACC-39; reverse, 59 TGG AGG ACG TTG GAA ATC AGC-39 [12]. Beta- Actin was used as an endogenous control (forward, 59-AGG ATG CAG AAG GAG ATC ACT-39, reverse 59-GGG TGT AAC GCA ACT AAG TCA TAG-39). Quantification of gene expression was performed according to the comparative cycle threshold method (22DDCT). Samples were normalized to beta-actin and compared to a reference sample used in all the experiments. Expression of ILT3, ILT4, B7-H3 and PUM1 in autopsy samples was measured using an Applied Biosystems 7500 with probes supplied by Applied Biosystems (Taqman Gene Expression Assays Hs00429000, Hs01629548, Hs00987207, and Hs00206469, respectively). Relative expression was calculated using PUM1 levels for normal- isation. IFN Beta and Vitamin D Modulate Expression of ILT3 in MS the specific fluorescent indices [geometric mean of the specific antibody fluorescence divided by the geometric mean of the isotype control antibody fluorescence] were calculated. Flow cytometry PBMC were washed twice followed by Fc receptor blocking with human IgG (Sigma-Aldrich, Munich, Germany). Afterwards cells were incubated for 30 minutes at 4uC with specific monoclonal antibodies: anti-CD14 APC (MOP9, BD Bioscience, Heidelberg, Germany), anti-CD86 PE (IT2.2, eBioscience, Frankfurt, Germany), anti-B7-H1 PE (MIH1, eBioscience, Frankfurt, Germany), anti-B7-H3-PE (RnD Systems), anti-ILT3 PE (ZM4.1, eBioscience, Frankfurt, Germany), anti-ILT4 PE (42D1, eBioscience, Frankfurt, Germany) and the respective fluorochrome-conjugated isotypic controls. For CSF analysis, cells were pelleted by centrifugation and resuspended in 100 ml of 4uC PBS 1% BSA within 30 minutes after the lumbar puncture. CSF and paired EDTA blood samples were stained with APC conjugated anti-CD14 antibody and PE-conjugated anti-ILT3 antibody or the respective isotype controls (blood only). Following erythrocyte lysis, cells were analyzed on a FACSCanto II (BD Biosciences, Heidelberg, Germany). Percentages of positive cells, as well as PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 11 / 15 Supporting Information S1 Fig. Effects of 1a,25 Dihydroxyvitamin D3 in combination with IFN beta on ILT3 and ILT4 expression in healthy donors. PBMC derived from RRMS patients were stimulated with 1a,25 Dihydroxyvitamin D3 (100 nM) and/or IFN 12 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS beta (1000IE) over a 48 h period. ILT3 and ILT4 protein expression on CD14+ cells was assessed by flow cytometry. The fold induction is shown (mean + SEM). A repeated measurement ANOVA and Bonferroni multiple comparison test was performed to assess statistical significance ( p,0.05; **p,0.005). doi:10 1371/journal pone 0115488 s001 (TIF) S1 Table. Soluble ILT3 in serum samples derived from MS or malignant melanoma patients. Serum samples were analyzed for soluble ILT3 protein content by ELISA. From some of the RRMS and melanoma patients, samples derived from before (untreated) and after initiation of IFN beta or IFN alpha therapy (IFN therapy) were available. A slash (/) indicates a missing sample, n.d. (not detectable) indicates that soluble ILT3 serum concentrations were below the lower level of detection of the assay. Detectable levels of ILT3 were found in 2 patients with RRMS and 2 patients with melanoma. In contrast to the increased surface ILT3 expression after IFN beta treatment, soluble ILT3 concentrations were not elevated (at least not above the detection threshold) by IFN beta therapy. doi:10.1371/journal.pone.0115488.s002 (TIF) Author Contributions Conceived and designed the experiments: AW TD MM SS. Performed the experiments: AW MK NS TD GV DT. Analyzed the data: AW TD NS MK GV. Contributed reagents/materials/analysis tools: SS MM TD MK. Wrote the paper: AW TD NS MK GV SS MM DT. Conceived and designed the experiments: AW TD MM SS. Performed the experiments: AW MK NS TD GV DT. Analyzed the data: AW TD NS MK GV. Contributed reagents/materials/analysis tools: SS MM TD MK. Wrote the paper: AW TD NS MK GV SS MM DT. Acknowledgments We are grateful to the UK MS tissue bank for providing autopsy samples. We thank K. Bitterer and K. Lehner for excellent technical assistance. We thank C. Hafner (Dept. of Dermatology, University of Regensburg), R. Rupec, M. Volkenandt, T. Maier and T. Vogt (Dept. of Dermatology, University of Munich) for melanoma patient samples. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 5. Chang CC, Liu Z, Vlad G, Qin H, Qiao X, et al. (2009) Ig-like transcript 3 regulates expression of proinflammatory cytokines and migration of activated T cells. J Immunol 182: 5208–5216. References 1. Lu HK, Rentero C, Raftery MJ, Borges L, Bryant K, et al. (2009) Leukocyte Ig-like receptor B4 (LILRB4) is a potent inhibitor of FcgammaRI-mediated monocyte activation via dephosphorylation of multiple kinases. J Biol Chem 284: 34839–34848. 2. Cella M, Dohring C, Samaridis J, Dessing M, Brockhaus M, et al. (1997) A novel inhibitory receptor (ILT3) expressed on monocytes, macrophages, and dendritic cells involved in antigen processing. J Exp Med 185: 1743–1751. 3. Chang CC, Ciubotariu R, Manavalan JS, Yuan J, Colovai AI, et al. (2002) Tolerization of dendritic cells by T(S) cells: the crucial role of inhibitory receptors ILT3 and ILT4. Nat Immunol 3: 237–243. 4. Manavalan JS, Rossi PC, Vlad G, Piazza F, Yarilina A, et al. (2003) High expression of ILT3 and ILT4 is a general feature of tolerogenic dendritic cells. Transpl Immunol 11: 245–258. 5. Chang CC, Liu Z, Vlad G, Qin H, Qiao X, et al. (2009) Ig-like transcript 3 regulates expression of proinflammatory cytokines and migration of activated T cells. J Immunol 182: 5208–5216. 13 / 15 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS 6. Vlad G, Chang CC, Colovai AI, Vasilescu ER, Cortesini R, et al. (2010) Membrane and soluble ILT3 are critical to the generation of T suppressor cells and induction of immunological tolerance. Int Rev Immunol 29: 119–132. 7. Vlad G, Suciu-Foca N (2012) Induction of antigen-specific human T suppressor cells by membrane and soluble ILT3. Exp Mol Pathol 93: 294–301. 8. Korn T (2008) Pathophysiology of multiple sclerosis. J Neurol 255 Suppl 6: 2–6. 9. Salama AD, Chitnis T, Imitola J, Ansari MJ, Akiba H, et al. (2003) Critical role of the programmed death-1 (PD-1) pathway in regulation of experimental autoimmune encephalomyelitis. J Exp Med 198: 71–78. 10. Rottman JB, Smith T, Tonra JR, Ganley K, Bloom T, et al. (2001) The costimulatory molecule ICOS plays an important role in the immunopathogenesis of EAE. Nat Immunol 2: 605–611. 11. Joller N, Peters A, Anderson AC, Kuchroo VK (2012) Immune checkpoints in central nervous system autoimmunity. Immunol Rev 248: 122–139. 12. Suciu-Foca N, Feirt N, Zhang QY, Vlad G, Liu Z, et al. (2007) Soluble Ig-like transcript 3 inhibits tumor allograft rejection in humanized SCID mice and T cell responses in cancer patients. J Immunol 178: 7432–7441. 13. Vlad G, D’Agati VD, Zhang QY, Liu Z, Ho EK, et al. References (2008) Immunoglobulin-like transcript 3-Fc suppresses T-cell responses to allogeneic human islet transplants in hu-NOD/SCID mice. Diabetes 57: 1878–1886. 14. Vlad G, Suciu-Foca N (2013) Tolerogenic dendritic cells and induction of Tsuppressor cells in transplant recipients. Methods Mol Biol 1034: 359–371. 15. Schreiner B, Mitsdoerffer M, Kieseier BC, Chen L, Hartung HP, et al. (2004) Interferon-beta enhances monocyte and dendritic cell expression of B7-H1 (PD-L1), a strong inhibitor of autologous T- cell activation: relevance for the immune modulatory effect in multiple sclerosis. J Neuroimmunol 155: 172–182. 16. Penna G, Roncari A, Amuchastegui S, Daniel KC, Berti E, et al. (2005) Expression of the inhibitory receptor ILT3 on dendritic cells is dispensable for induction of CD4+Foxp3+ regulatory T cells by 1,25- dihydroxyvitamin D3. Blood 106: 3490–3497. 17. Ortler S, Leder C, Mittelbronn M, Zozulya AL, Knolle PA, et al. (2008) B7-H1 restricts neuroantigen- specific T cell responses and confines inflammatory CNS damage: implications for the lesion pathogenesis of multiple sclerosis. Eur J Immunol 38: 1734–1744. 18. Mueller DL (2010) Mechanisms maintaining peripheral tolerance. Nat Immunol 11: 21-27. 19. Smith-Garvin JE, Koretzky GA, Jordan MS (2009) T cell activation. Annu Rev Immunol 27: 591–619. 20. Jensen MA, Yanowitch RN, Reder AT, White DM, Arnason BG (2010) Immunoglobulin-like transcript 3, an inhibitor of T cell activation, is reduced on blood monocytes during multiple sclerosis relapses and is induced by interferon beta-1b. Mult Scler 16: 30–38. 21. Wiendl H, Mitsdoerffer M, Schneider D, Chen L, Lochmuller H, et al. (2003) Human muscle cells express a B7-related molecule, B7-H1, with strong negative immune regulatory potential: a novel mechanism of counterbalancing the immune attack in idiopathic inflammatory myopathies. FASEB J 17: 1892–1894. 22. Ascherio A, Munger KL, Simon KC (2010) Vitamin D and multiple sclerosis. Lancet Neurol 9: 599–612. 23. Runia TF, Hop WC, de Rijke YB, Buljevac D, Hintzen RQ (2012) Lower serum vitamin D levels are associated with a higher relapse risk in multiple sclerosis. Neurology 79: 261–266. 24. Simpson S Jr, Taylor B, Blizzard L, Ponsonby AL, Pittas F, et al. (2010) Higher 25-hydroxyvitamin D is associated with lower relapse risk in multiple sclerosis. Ann Neurol 68: 193–203. 25. Ascherio A, Munger KL, White R, Kochert K, Simon KC, et al. (2014) Vitamin D as an Early Predictor of Multiple Sclerosis Activity and Progression. JAMA Neurol. 26. James E, Dobson R, Kuhle J, Baker D, Giovannoni G, et al. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 28. Kim-Schulze S, Scotto L, Vlad G, Piazza F, Lin H, et al. (2006) Recombinant Ig-like transcript 3-Fc modulates T cell responses via induction of Th anergy and differentiation of CD8+ T suppressor cells. J Immunol 176: 2790–2798. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 References (2013) The effect of vitamin D-related interventions on multiple sclerosis relapses: a meta-analysis. Mult Scler 19: 1571–1579. 27. Stewart N, Simpson S Jr, van der Mei I, Ponsonby AL, Blizzard L, et al. (2012) Interferon-beta and serum 25-hydroxyvitamin D interact to modulate relapse risk in MS. Neurology 79: 254–260. PLOS ONE \| DOI:10.1371/journal.pone.0115488 December 31, 2014 14 / 15 IFN Beta and Vitamin D Modulate Expression of ILT3 in MS 28. Kim-Schulze S, Scotto L, Vlad G, Piazza F, Lin H, et al. (2006) Recombinant Ig-like transcript 3-Fc modulates T cell responses via induction of Th anergy and differentiation of CD8+ T suppressor cells. J Immunol 176: 2790–2798. 28. Kim-Schulze S, Scotto L, Vlad G, Piazza F, Lin H, et al. (2006) Recombinant Ig-like transcript 3-Fc modulates T cell responses via induction of Th anergy and differentiation of CD8+ T suppressor cells. J Immunol 176: 2790–2798. 29. Gleissner CA, Zastrow A, Klingenberg R, Kluger MS, Konstandin M, et al. (2007) IL-10 inhibits endothelium-dependent T cell costimulation by up-regulation of ILT3/4 in human vascular endothelial cells. Eur J Immunol 37: 177–192. 30. Polman CH, Reingold SC, Edan G, Filippi M, Hartung HP, et al. (2005) Diagnostic criteria for multiple sclerosis: 2005 revisions to the "McDonald Criteria". Ann Neurol 58: 840–846. 31. Lassmann H, Raine CS, Antel J, Prineas JW (1998) Immunopathology of multiple sclerosis: report on an international meeting held at the Institute of Neurology of the University of Vienna. J Neuroimmunol 86: 213–217. 15 / 15
https://openalex.org/W3000680906	https://www.mdpi.com/1660-3397/18/1/54/pdf?version=1580701669	English	null	Recent Discovery of Heterocyclic Alkaloids from Marine-Derived Aspergillus Species	Marine drugs	2,020	cc-by	11,335	Received: 27 December 2019; Accepted: 11 January 2020; Published: 14 January 2020 Abstract: Nitrogen heterocycles have drawn considerable attention due to of their signiﬁcant biological activities. The marine fungi residing in extreme environments are among the richest sources of these basic nitrogen-containing secondary metabolites. As one of the most well-known universal groups of ﬁlamentous fungi, marine-derived Aspergillus species produce a large number of structurally unique heterocyclic alkaloids. This review attempts to provide a comprehensive summary of the structural diversity and biological activities of heterocyclic alkaloids that are produced by marine-derived Aspergillus species. Herein, a total of 130 such structures that were reported from the beginning of 2014 through the end of 2018 are included, and 75 references are cited in this review, which will beneﬁt future drug development and innovation. Keywords: Aspergillus; metabolite; marine; alkaloid; biological activity marine drugs marine drugs Recent Discovery of Heterocyclic Alkaloids from Marine-Derived Aspergillus Species Kuo Xu 1 , Xiao-Long Yuan 1, Chen Li 2,3 and Xiao-Dong Li 2,3,* 1 Tobacco Research Institute of Chinese Academy of Agricultural Sciences, Qingdao 266101, China; xukuoworld@126.com (K.X.); yuanxiaolong@caas.cn (X.-L.Y.) 2 Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China; Lychees6601@163.com 3 Key Laboratory of marine biotechnology in Universities of Shandong (Ludong University), School of Life Sciences, Ludong University, Yantai 264025, China * Correspondence: imnli@163.com; Tel.: +86-535-210-9018 1 Tobacco Research Institute of Chinese Academy of Agricultural Sciences, Qingdao 266101, China; xukuoworld@126.com (K.X.); yuanxiaolong@caas.cn (X.-L.Y.) 2 Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China; Lychees6601@163.com 3 Key Laboratory of marine biotechnology in Universities of Shandong (Ludong University), Scho Sciences, Ludong University, Yantai 264025, China * Correspondence: imnli@163.com; Tel.: +86-535-210-9018 Mar. Drugs 2020, 18, 54; doi:10.3390/md18010054 marine drugs marine drugs 1. Introduction Heterocyclic alkaloids are one of the most challenging natural product classes to characterize, not only because of their structurally unique skeletons that arise from distinct amino acids, but also because of their potential bioactivities. These nitrogen heterocycles are among the most active molecules and they are currently in various phases of human clinical trials for treating various diseases [1–3]. The ocean is a rich underexploited source of novel and bioactive molecules, because extreme marine conditions, including low temperature, high pressure, reduced light, and the presence of predators, can cause marine organisms to develop machinery for the construction of a greater diversity of metabolites than terrestrial organisms [4–8]. Fungi living in the extreme environments that are typical of marine ecosystems are very sensitive to culture media and are more liable to produce novel metabolites than fungi living in less extreme environments. As one of the most well-known universal groups of ﬁlamentous fungi, marine-derived Aspergillus species have one of the main sources of new heterocyclic alkaloids in recent years. This mini-review attempts to provide a comprehensive summary of the structural diversity and biological activities of the nitrogen-containing secondary metabolites that are produced by marine-derived Aspergillus species. Mar. Drugs 2020, 18, 54; doi:10.3390/md18010054 www.mdpi.com/journal/marinedrugs 2 of 22 Mar. Drugs 2020, 18, 54 A series of excellent reviews on various aspects of secondary metabolites that are derived from the genus Aspergillus have been published in the past ﬁve years (from 2014 to present) [9–16]. However, there is only one work speciﬁcally aimed at Aspergillus species from the marine environment. In 2018, K.W. Wang and P. Ding summarized the information on 232 new bioactive secondary metabolites from marine-derived Aspergillus species, which had been reported from 2006 to 2016, with classiﬁcation on the basis of biological activity and chemical structure [12]. As part of our ongoing investigations of biological compounds from endophytic Aspergillus species that reside on the marine brown algae Leathesia nana (Chordariaceae) [17], a detailed and comprehensive literature survey disclosed that the previously published structures might not be adequately represented. To the best of our knowledge, a total of approximately 400 new compounds were isolated from marine-derived Aspergillus species from the beginning of 2014 to the end of 2018 (see Supplementary Materials), of which 130 could be classiﬁed as heterocyclic alkaloids. This review aims to provide an update on the recent discoveries of the heterocyclic alkaloids that are produced by marine-derived Aspergillus species. 1. Introduction The selection of original articles was of greatest importance because these papers had a direct impact on the ﬁndings and the ﬁnal results. This review included all original articles registered with the relevant subject in the Web of Science Core Collection database between 2014 and 2018. The literature search was performed while using a previously reported search method [18,19]. The search strategy was as follows: “Title: (from Aspergillus); Reﬁned by: Topic (marine) and Document types (article); Timespan: 2014–2018; Indexes: SCI-EXPANDED, CPCI-S”. Notably, the present work was preliminarily planned in April 2019, and the studies that were published or being submitted in the current year might not be accurately indexed in the Web of Science Core Collection database; thus, the timespan of the literature search was from 2014 to 2018. With this approach, 166 records were ﬁnally identiﬁed and were considered to cover most of the related studies. After retrieving the records that were related to the ﬁeld of natural product chemistry, 123 original articles were indexed in the Web of Science Core Collection database over a period of ﬁve years (from the beginning of 2014 to the end of 2018). During this period, 398 naturally occurring compounds were isolated and characterized from marine-derived Aspergillus species. The 130 nitrogen-heterocyclic compounds included accounted for 32.7% of all newly reported secondary metabolites. Supplementary Materials lists all 123 original articles and the structures of these 398 newly reported secondary metabolites. This critical review focuses on the structural diversity, biological activities, and sources of these newly reported heterocyclic alkaloids. 2.1. Indole Alkaloids Indole alkaloids serve as the active moiety in several clinical drugs, such as reserpine, and several well-known drugs, such as sumatriptan, tadalaﬁl, ﬂuvastatin, and rizatriptan, were designed on the basis of the indole framework [20]. The indole moiety is present in a wide range of marine natural products, especially fungal metabolites [8,21,22]. Figure 1 lists the structures of indole alkaloids produced by marine-derived Aspergillus species. Compound 1 was isolated from a culture broth of a gorgonian-originating fungal strain, A. sp. SCSIO 41501, and then characterized as a new linear peptide with three amino acid residues, d-Tyr, d-Val, and l-Trp [23]. Compound 2 was obtained from the coral-associated fungus A. terreus, whose structure featured an unusual (E)-oxime group, which is rare in natural products [24]. The structure of compound 3 was deﬁned and characterized as a previously unreported bis-indolyl benzenoid and it was isolated from cultures of the marine sponge-derived fungus A. candidus KUFA0062 [25]. The C-3 position of the indole fragment in compounds 1–3 was substituted by methylene and phenyl groups, which were considered the same type of substituent. Chemical investigation of the algal-derived endophytic fungus A. alabamensis EN-547 led to the isolation of two new compounds, 4 and 5, possessing a rare diketomorpholine fragment [26]. The rice-based culture of a marine-associated fungal strain A. sp. MEXU 27854 was extensively chromatographed to produce ﬁve dioxomorpholine derivatives 6–10 [27]. The structure of compound 11, which was obtained from a marine sediment-derived Aspergillus sp. CMB-M081F, was identiﬁed and characterized as a dioxomorpholine derivative [28]. Compounds 12–15 were isolated from the fungal strain Aspergillus sp. from an unidentiﬁed colonial ascidian and then characterized as four new indole-diterpene alkaloids [29]. Compounds 16 and 17 were isolated and identiﬁed from the cultures of the endophytic fungus A. nidulans EN-330 that were collected from the marine red alga Polysiphonia scopulorum [30]. Interestingly, the structures of compounds 12 and 14–16 contain a halogen atom, which is rare in fungal secondary metabolites. Compound 18 represents the ﬁrst example of an N-isopentenyl tryptophan methyl ester with a phenyl propanoic amide arm and it was identiﬁed from the marine sponge-derived fungus A. sp. SCSIO XWS03F03 [31]. Compound 19 was obtained through chromatographic separation of the crude organic extract of a sponge-associated fungal strain A. sp., whose structure was a tryptophan-derived indole alkaloid [32]. 2. Structural Diversity Figures 1–6 present the structures of newly reported heterocyclic alkaloids (1–130) produced by marine-derived Aspergillus species from 2014 to 2018, in which the nitrogen-containing heterocyclic rings are marked in red. These heterocyclic alkaloids could be classiﬁed into six major categories: indole alkaloids (1–31), diketopiperazine alkaloids (32–58), quinazoline alkaloids (59–72), pyrrolidine alkaloids (73–96), cyclopeptide alkaloids (97–108), and other heterocyclic alkaloids (109–130) based on their structural patterns. Mar. Drugs 2020, 18, 54 3 of 22 2.1. Indole Alkaloids Compound 20 was characterized as a new polycyclic alkaloid, which was isolated by further chemical investigation of a coral-associated fungal strain A. versicolor LZD-14-1 [33]. Compounds 21 and 22 were identiﬁed and characterized as two new indole diterpenoids and they were isolated from the fermentation broth of the fungal strain A. ﬂavus OUCMDZ-2205 [34]. Compounds 23–30 were reported as eight new cyclopiazonic acid (CPA)-type alkaloids, which are usually composed of three structural units: an indole, a tetramic acid unit, and a malonic acid unit. More precisely, compounds 23–29 were isolated from the culture of an epiphytic fungal strain of A. oryzae residing in marine sediments that were collected from Langqi Island, China, while compound 30 was obtained from the culture of a marine isopod-associated endophytic fungus, A. sp. Z-4, which was derived from the marine isopod Ligia oceanica [35–37]. Compound 31 was characterized as a new prenylated indole alkaloid, which was isolated from a coculture of the marine-derived fungi A. sulphureus KMM 4640 and Isaria felina KMM 4639 [38]. 4 of 22 4 of 22 Mar. Drugs 2020, 18, 54 Figure 1. Indole alkaloids produced by marine-derived Aspergillus species (1–31). Figure 1. Indole alkaloids produced by marine-derived Aspergillus species (1–31). 2.2. Diketopiperazine Alkaloids Diketopiperazine alkaloids are common metabolites of microorganisms that are widely distributed in ﬁlamentous fungi, especially in the genera Aspergillus and Penicillium of the phylum Ascomycota or sac fungi [39]. Interestingly, an indole fragment is typically present in the structures of these diketopiperazine alkaloids. Figure 2 lists the structures of diketopiperazine alkaloids that are produced by marine-derived Aspergillus species. Compounds 32–35 were isolated from a coculture of the marine sediment-derived fungi A. sulphureus KMM 4640 and Isaria felina KMM 4639 and identiﬁed as four new prenylated indole diketopiperazine alkaloids [38]. Compounds 36–43 were characterized as eight linearly fused prenylated indole diketopiperazines featuring an unusual pyrano[3,2-f]indole unit, which were isolated from the culture of a fungal strain, A. versicolor, residing in mud from the South China Sea [40]. Compounds 44 and 45 were isolated from the Antarctic marine-derived A. sp. SF-5976 obtained from an unidentiﬁed marine organism that was collected in the Ross Sea [41]. Compounds 46–51 were identiﬁed as six new prenylated indole diketopiperazines and they were isolated from a culture of the marine sediment-derived fungus A. versicolor HDN08-60 [42]. These compounds are characterized by a 6/6/5/8/6/5 hexacyclic ring system that possesses a hydrogenated azocine unit. Compounds 52–56 were reported as four new bis-indole diketopiperazine alkaloids characterized by the presence of two indole diketopiperazines in their structures. More speciﬁcally, compounds 52 and 53 were isolated from an organic extract of the sponge-derived fungal strains A. sp. SF-5280 and A. violaceofuscus, respectively, while compounds 54–56 were from a culture of a fungus residing in marine shrimp that were collected along the coast of Dinghai, China [43–45]. Compounds 57 and 58 were 5 of 22 Mar. Drugs 2020, 18, 54 also isolated from a culture of the marine sediment-derived fungus A. versicolor HDN08-60, and their structures featured the presence of only a diketopiperazine fragment, but no indole ring. Structurally, compound 57 possesses an unprecedented skeleton of a 2,5-dihydro-1H-azepino[4,3-b]quinoline system, while 58 contains a novel 6/6/11/6/5 pentacyclic ring system. Moreover, the former compound is considered to be a derivative of the latter [42]. Figure 2. Diketopiperazine alkaloids produced by marine-derived Aspergillus species (32–58). Figure 2. Diketopiperazine alkaloids produced by marine-derived Aspergillus species (32–58). 6 of 22 Figure 3. Quinazoline alkaloids produced by marine-derived Aspergillus species (59–72). Figure 3. Quinazoline alkaloids produced by marine-derived Aspergillus species (59–72). 2.3. Quinazoline Alkaloids Figure 3 lists the structures of quinazoline alkaloids that are produced by marine-derived Aspergillus species. A quinazoline moiety was found in all these alkaloids, which might provide insights into the biogenetic relationships of quinazoline-containing indole alkaloids. Compounds 59–61 were characterized as two new quinazoline alkaloids and they were isolated from a culture of the deep-sea-derived fungus A. fumigatus SCSIO 41012 [46]. Compound 62 originated from a culture of the Australian marine sediment-derived A. sp. CMB-M081F, whose structure was elucidated by detailed spectroscopic analysis and biosynthetic considerations [28]. A solid-substrate culture of strain A. sp. F452 residing in submerged decaying wood was extensively chromatographed to produce ﬁve new quinazoline-containing alkaloids 63–67. Among them, compound 63 represents a new member of the fumiquinazoline class of alkaloids, which has been reported in a number of marine-derived Aspergillus, Acremonium, and Scopulariopsis fungal strains [47]. Compounds 68–72 were characterized as six new polycyclic alkaloids and they were isolated from a culture of a coral-associated fungus, A. versicolor LZD-14-1 [33]. Mar. Drugs 2020, 18, 54 6 of 22 2.4. Pyrrolidine Alkaloids Figure 4 lists the structures of pyrrolidine alkaloids that are produced by marine-derived Aspergillus species. Bioactivity-guided chemical investigation of cultures of the marine sponge-associated fungal strain A. ﬂocculosus 16D-1 facilitated the isolation of nine pyrrolidine alkaloids 73–81. The structures and conﬁgurations of these compounds were elucidated by detailed spectroscopic analysis, the modiﬁed Mosher’s method, and comparisons with literature data [48]. Compound 82 was isolated and characterized as a new hydroxypyrrolidine alkaloid from cultures of the marine sponge-associated fungus A. candidus KUFA 0062 [25]. Compound 83 was produced by a coculture of gorgonian-derived fungal strains of A. sclerotiorum and P. citrinum and characterized as a pyrrole analog [49]. Compounds 84–91 were identiﬁed and characterized as having a spiro-heterocyclic γ-lactam skeleton and they were isolated from a culture broth of the marine ﬁsh-associated endophytic fungi A. fumigatus [50]. Compounds 92–96 were characterized as four aspochalasin analogs and they were obtained from the intestines of the marine isopod Ligia oceanica, collected along the coast of Dinghai in Zhoushan, Zhejiang Province, China [51–53]. Aspochalasins are a small group of cytochalasans structurally featuring a macrocyclic ring system and perhydroisoindol-1-one unit with an isobutyl side chain. Among them, the structure of compound 92 includes a unique 5/6/6 tricyclic ring fused with the skeleton of aspochalasin [51]. Compounds 94 and 95 represent the ﬁrst thiomethyl-substituted aspochalasin analogs [52]. Compound 96 is rare, in that it contains two nitrogen atoms in its molecular structure and an unusual skeleton that includes an azabicyclo moiety [53]. 7 of 22 7 of 22 Mar. Drugs 2020, 18, 54 Figure 4. Pyrrolidine alkaloids produced by marine-derived Aspergillus species (73–96). Figure 4. Pyrrolidine alkaloids produced by marine-derived Aspergillus species (73–96). 8 of 22 Figure 5. Cyclic peptide alkaloids produced by marine-derived Aspergillus species (97–108). Figure 5. Cyclic peptide alkaloids produced by marine-derived Aspergillus species (97–108). 2.5. Cyclopeptide Alkaloids Cyclopeptide alkaloids are mainly constructed from proteinogenic or nonproteinogenic amino acids that are joined together by amide bonds [54]. These alkaloids can be widely synthesized by both terrestrial and marine organisms. A diversity of cyclopeptide alkaloids with intriguing structures and possible pharmaceutical activities has been identiﬁed from marine fungi, a well-known producer [55]. To the best of our knowledge, twelve cyclic peptides (97–105) were published from 2014 to 2018. Figure 5 lists the structures of cyclopeptide alkaloids that are produced by marine-derived Aspergillus species. Compound 97 was isolated and characterized as a novel cyclic dipeptide with a skeleton of cyclo-(anthranilic acid-l-N-Me-Tyr) and it could also be considered a benzodiazepine alkaloid of the cyclopenin group [56]. Compounds 98 and 99 were isolated from the EtOAc extract of an endophytic fungal strain, A. violaceofuscus, residing in the interior of the marine sponge Reniochalina sp. collected from the Xisha Islands in the South China Sea. The structure of compound 98 was established as an aspochracin-type cyclic tripeptide, while that of 99 was elucidated as a cyclic tetrapeptide with the sequence cyclo-[l-Thr-l-O-Me-Tyr-l-N-Me-Ala-l-Ile] [44]. Compounds 100–104 were isolated from the fungal strain A. versicolor ZLN-60, which was obtained from marine sediment in the Yellow Sea in China. More precisely, the structures of compounds 100–103 were established as four cyclic peptides that possess a rare amide linkage between the carboxylic acid in anthranilic acid and the nitrogen in an indole moiety, while that of 104 was an anthranilic acid-containing hexapeptide [57,58]. Compound 105 was elucidated as a new cyclic tetrapeptide with a skeleton of cyclo[anthranilic acid-3(S)-OH-N-Me-Phe-d-Val-l-Ala] [59]. Compound 106 was isolated from a culture broth of the gorgonian-derived fungus A. terreus SCSGAF0162 and was characterized as a cyclic tetrapeptide with a skeleton of cyclo[l-Val-(N-Me-)-d-Tyr-(O-Me-)-l-Tyr-(O-Me-)-l-Tyr-l-Pro] [23]. Compound 107 was established as a new cyclohexapeptide with the sequence [cyclo (anthranilic acid-l-Val-d-Leu-l-Ala-N-methyl-l-Leu-d-pipecolic acid)] and it was isolated from the sponge-derived fungus A. similanensis KUFA 0013. Its amino acid sequence was the same as that of a previously reported compound (PF1171C), but the absolute conﬁguration was diﬀerent [60]. Compound 108 was isolated from the gorgonian-associated fungus A. versicolor TA01-14 that was collected from the South China Sea. Its structure was identiﬁed and characterized as a centrally symmetric cyclohexapeptide with a skeleton of cyclo [l-Phe-(anthranilic acid)-l-Pro-l-Phe-(anthranilic acid)-l-Pro] [61]. Mar. Drugs 2020, 18, 54 8 of 22 2.6. Other Heterocyclic Alkaloids The remaining heterocyclic alkaloids (109–130) that were produced by marine-derived Aspergillus species are summarized in this section, and their structures are listed in Figure 6. Compounds 109–111 were identiﬁed and characterized as three new pyridine derivatives. In contrast, compound 109 was obtained from the EtOAc extract of cultures of the marine sponge-derived fungal strain A. similanensis KUFA 0013, while compounds 110 and 111 were identiﬁed from the cultures of the marine alga-derived fungus A. niger SCSIO Jcsw6F30 [60,62]. Compounds 112 and 113 were characterized as two new prenylated dihydroquinolone derivatives that were isolated from the mycelia of a gorgonian-derived Aspergillus fungus and they represent the ﬁrst examples of prenylated dihydroquinolone derivatives containing an amino acid residue in the side chain [63]. Compounds 114 and 115 were isolated from the cultures of the marine sediment-derived epiphytic fungi A. versicolor and A. ﬂavus KMM 4650, respectively, while compounds 116 and 117 were identiﬁed from the gorgonian-derived endophytic fungus A. versicolor [64–66]. Structurally, compounds 114–117 represent four pyrimidine derivatives, and compounds 115–117 are aromatic nucleosides. Compounds 118–122 were isolated from the fermentation broth of the marine coral-derived halotolerant A. ochraceus LCJ11-102 that was cultivated in nutrient-limited medium containing 10% NaI. Compounds 123 and 124 were isolated from a coculture of the marine gorgonian-derived P. citrinum SCSGAF 0052 and A. sclerotiorum [49,67]. These seven heterocyclic alkaloids were characterized as new pyrazinone alkaloids. The structures of compounds 125–127 were characterized as three open-chain peptides with an unusual skeleton of 1,3-dimethyllumazine-6-carboxylic acid, coupled to glutamine and anthranilic acid methyl ester [68,69]. In contrast, compound 125 was isolated from a culture of the mangrove-derived fungal strain A. sp. (33241) [68], while compounds 126 and 127 were obtained from cultures of the marine sediment-derived fungal strain A. terreus FA009 [69]. Compound 128 was identiﬁed as a novel oxadiazin derivative and it was also isolated from a coculture of the marine gorgonian-derived P. citrinum SCSGAF 0052 and A. sclerotiorum [49]. Compound 129 was isolated and identiﬁed from a static culture of the marine sediment-originated fungus A. sydowii SP-1 and possesses a 1H-imidazo[2,1-b]purin-4(5H)-one skeleton [70]. Compound 130 was elucidated as a novel hybrid polyketide-terpenoid with a unique skeleton of fused polycyclic fragments and it was isolated and identiﬁed from a crab collected from a Kueishantao hydrothermal vent in China [71]. 9 of 22 9 of 22 Mar. Drugs 2020, 18, 54 Figure 6. 2.6. Other Heterocyclic Alkaloids Other alkaloids isolated from marine-derived Aspergillus species (109–130). Figure 6. Other alkaloids isolated from marine-derived Aspergillus species (109–130). 3. Production Environment Moreover, the producing strains of 21, 22, and 54–56 originated from marine shrimp or prawns [34,45], while those of compounds 12–15 were derived from an unidentiﬁed colonial ascidian that was collected at Shikotan Island in the Paciﬁc Ocean [29]. In addition to these producing strains of marine invertebrate origin, only one fungal strain, which produced 84–91, was obtained from the marine ﬁsh Mugil cephalus, representing vertebrates [50]. It appears that the number of the fungal strains of marine plant origin that produce heterocyclic alkaloids is much less than that of marine invertebrate and vertebrate origin. For example, the producing strains of the heterocyclic alkaloids 4, 5, 16, 17, 110, and 111 originated from the fungi residing in marine algal species, including Ceramium japonicum [26], Polysiphonia scopulorum [30], and Sargassum sp. [58,62], while those of 63–67 were derived from submerged decaying wood at Jeju Island, Korea [47]. Moreover, the producing strain of 125 originated from the mangrove Bruguiera sexangula var. rhynchopetala that was collected in the South China Sea [68]. Interestingly, two heterocyclic alkaloids (44 and 45) were isolated from the fungi residing in an unidentiﬁed marine organism that was collected in the Ross Sea without any detailed description for species identiﬁcation [41]. Table 1. The producing strain, biological activities of these heterocyclic alkaloids (1–130). NO. Producing Strain Environmental Source Biological Activities Ref. 1 A. sp. SCSIO 41501 the gorgonian Melitodes squamata collected from the South China Sea, Sanya, China moderate antiviral activity against HSV-1 under non-cytotoxic concentrations against Vero cells [23] 2 A. terreus the coral Sarcophyton subviride collected from the coast of Xisha Island in the South China Sea potent inhibition on LPS-induced NO production; nonsigniﬁcant inhibition on α-Glucosidase [24] 3 A. candidus KUFA0062 the marine sponge Epipolasis sp. collected at Similan Island National Park (15–20 m), Thailand weak cytotoxic activity against eight cell lines; nonsigniﬁcant antibacterial activity [25] 4 A. alabamensis EN-547 the fresh inner tissue of marine alga Ceramium japonicum collected at Qingdao, China moderate antimicrobial activities against E. coli, M. luteus, Ed. ictaluri and V. alginolyticus [26] 5 A. alabamensis EN-547 the fresh inner tissue of marine alga Ceramium japonicum collected at Qingdao, China moderate antimicrobial activities against E. coli, M. luteus, Ed. ictaluri and V. alginolyticus [26] 6 A. sp. MEXU 27854 sandy soil collected in the intertidal zone located in Caleta Bay, Acapulco, Guerrero, Mexico nonsigniﬁcant cytotoxic activities [27] 7 A. sp. 3. Production Environment Endophytic and epiphytic fungi have proven to be proliﬁc sources of bioactive natural products with unique structures and potent pharmaceutical activity; such fungi harmoniously colonize the internal tissues of their hosts usually without causing obvious damage to the hosts [72,73]. Marine-derived fungi could be isolated from every possible marine habitat, such as marine sediments, marine invertebrates (sponges, corals, ascidians, and holothurians), and vertebrates (mainly ﬁsh), as well as marine plants (algae, driftwood, and mangrove plants) [74,75]. As shown in Table 1, a total of 44 heterocyclic alkaloids (11, 23–29, 31–43, 46–51, 57–62, 97, 100–104, 114, 115, 126, 127, and 129) originated from the fungi residing in marine sediments, which accounted for 33.8% of the 130 nitrogen-heterocyclic secondary metabolites. More precisely, the producing strains of 59–61, 97, and 114 originated from deep-sea sediment (deeper than 100 m) [46,56,64], while the other strains were collected from marine sediments at depths above 100 m or even tideland mud [28,35,36,40,42,57,65,69,70]. Unfortunately, the source of the producing strain of 31–35 (A. sulphureus KMM 4640 and I. felina KMM 4639) was not described [39]. To the best of our knowledge, an overwhelming majority of these fungal strains were isolated from marine invertebrates, including corals, sponges, crabs, shrimps, ascidians, and some isopods. For instance, a total of 24 heterocyclic alkaloids (1, 2, 20, 68–72, 83, 105, 106, 108, 112, 113, 116–124, and 128) were identiﬁed from the endophytic fungal strains of marine gorgonian species, including Melitodes squamata [23], Sarcophyton subviride [24], Pseudopterogorgia sp. LZD-14 [33], Muricella ﬂexuosa [49], Echinogorgia aurantiaca [59], Carijoa sp. GX-WZ-2010001 [61], Muricella abnormaliz [63], and Dichotella gemmacea [66,67], and they accounted for 18.5% of the newly reported heterocyclic alkaloids. A total of 19 heterocyclic alkaloids (3, 18, 19, 52, 53, 73–82, 98, 99, 107, and 109) were identiﬁed from the fungi residing in marine sponges, including Epipolasis sp. [25], Tethya aurantium [32], Reniochalina sp. [44], Phakellia fusca [48], Rhabdermia sp. [60], and an unidentiﬁed species [31,43], and 10 of 22 Mar. Drugs 2020, 18, 54 they accounted for 14.6% of the newly reported heterocyclic alkaloids. Six heterocyclic alkaloids (30 and 92–96) were also isolated from the endophytic fungi (A. sp. Z-4) residing in the marine isopod Ligia oceanica [37,51–53], while only one heterocyclic alkaloid (130) was identiﬁed from the endophytic fungi obtained from the marine crab Xenograpsus testudinatus [71]. 3. Production Environment MEXU 27854 sandy soil collected in the intertidal zone located in Caleta Bay, Acapulco, Guerrero, Mexico no biological activity was tested [27] 8 A. sp. MEXU 27854 sandy soil collected in the intertidal zone located in Caleta Bay, Acapulco, Guerrero, Mexico nonsigniﬁcant cytotoxic activities [27] 9 A. sp. MEXU 27854 sandy soil collected in the intertidal zone located in Caleta Bay, Acapulco, Guerrero, Mexico no biological activity was tested [27] 10 A. sp. MEXU 27854 sandy soil collected in the intertidal zone located in Caleta Bay, Acapulco, Guerrero, Mexico no biological activity was tested [27] 11 A. sp. CMB-M081F the marine sediment collected at an intertidal depth of 1 m near Shorncliﬀe, Queensland, Australia nonsigniﬁcant cytotoxic activities; potent inhibition on P-glycoprotein-mediated drug eﬄux [28] Table 1. The producing strain, biological activities of these heterocyclic alkaloids (1–130). 11 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 12 A. sp. KMM 4676 an unidentiﬁed colonial ascidian (Shikotan Island, Paciﬁc Ocean) potent cytotoxicity against 22Rv1, while moderate cytotoxicity against PC-3 and LNCaP [29] 13 A. sp. KMM 4676 an unidentiﬁed colonial ascidian (Shikotan Island, Paciﬁc Ocean) no biological activity was tested [29] 14 A. sp. KMM 4676 an unidentiﬁed colonial ascidian (Shikotan Island, Paciﬁc Ocean) nonsigniﬁcant cytotoxic activities against PC-3, LNCaP and 22Rv1 cell lines [29] 15 A. sp. KMM 4676 an unidentiﬁed colonial ascidian (Shikotan Island, Paciﬁc Ocean) no biological activity was tested [29] 16 A. nidulans EN-330 The marine red alga P. scopulorum var. villum collected from Yantai coastline of north China moderate antimicrobial activities against four human- and aqua-pathogens [30] 17 A. nidulans EN-330 The marine red alga P. scopulorum var. villum collected from Yantai coastline of north China weak antimicrobial activities against four human- and aqua-pathogens [30] 18 A. sp. SCSIO XWS03F03 a sponge collected from the sea area Xuwen County, Guangdong, China potent cytotoxic activity against HL-60 and LNCap cell lines [31] 19 A. sp. the sponge Tethya aurantium (Pallas 1766) collected at the entrance of Limski kanal (a depth of 20 m) moderate selective activity against marine bacteria; nonsigniﬁcant cytotoxicity against L5178Y cells [32] 20 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; potent inhibitory activity against TrxR [33] 21 A. 3. Production Environment ﬂavus OUCMDZ-2205 the prawn Penaeus vannamei collected in Lianyungang sea area, Jiangsu Province of China moderate antibacterial and cytotoxic activities, as well as PKC-beta inhibition [34] 22 A. ﬂavus OUCMDZ-2205 the prawn Penaeus vannamei collected in Lianyungang sea area, Jiangsu Province of China moderate cytotoxicity; nonsigniﬁcant antibacterial activity and PKC-beta inhibition [34] 23 A. oryzae the marine sediments collected from Langqi Island, Fujian, China weak cytotoxic activities against Hela and MGC803 cell lines [35] 24 A. oryzae the marine sediments collected from Langqi Island, Fujian, China nonsigniﬁcant cytotoxic activities against Hela, HL-60, and K562 cell lines [36] 25 A. oryzae the marine sediments collected from Langqi Island, Fujian, China nonsigniﬁcant cytotoxic activities against Hela and MGC803 cell lines [35] 26 A. oryzae the marine sediments collected from Langqi Island, Fujian, China nonsigniﬁcant cytotoxic activities against Hela and MGC803 cell lines [35] 27 A. oryzae the marine sediments collected from Langqi Island, Fujian, China weak cytotoxic activities against Hela and MGC803 cell lines [35] 28 A. oryzae the marine sediments collected from Langqi Island, Fujian, China nonsigniﬁcant cytotoxic activities against Hela, HL-60, and K562 cell lines [36] 29 A. oryzae the marine sediments collected from Langqi Island, Fujian, China nonsigniﬁcant cytotoxic activities against Hela, HL-60, and K562 cell lines [36] 30 A. sp Z-4 the marine isopod Ligia oceanica collected in Zhoushan, Zhejiang, China nonsigniﬁcant cytotoxicity against PC3 and HCT116 [37] 31 A. sulphureus KMM 4640 and I. felina KMM 4639 marine sediments (no detailed description) nonsigniﬁcant cytotoxic activities against MRC-9, HEK 293T, 22Rv1, PC-3, and LNCaP [38] 12 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 32 A. sulphureus KMM 4640 and I. felina KMM 4639 marine sediments (no detailed description) nonsigniﬁcant cytotoxic activities against MRC-9, HEK 293T, 22Rv1, PC-3, and LNCaP [38] 33 A. sulphureus KMM 4640 and I. felina KMM 4639 marine sediments (no detailed description) nonsigniﬁcant cytotoxic activities against MRC-9, HEK 293T, 22Rv1, PC-3, and LNCaP [38] 34 A. sulphureus KMM 4640 and I. felina KMM 4639 marine sediments (no detailed description) nonsigniﬁcant cytotoxic activities against MRC-9, HEK 293T, 22Rv1, PC-3, and LNCaP [38] 35 A. sulphureus KMM 4640 and I. felina KMM 4639 marine sediments (no detailed description) nonsigniﬁcant cytotoxic activities against MRC-9, HEK 293T, 22Rv1, PC-3, and LNCaP [38] 36 A. 3. Production Environment versicolor the mud of the South China Sea moderate inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 37 A. versicolor the mud of the South China Sea moderate inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 38 A. versicolor the mud of the South China Sea nonsigniﬁcant inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 39 A. versicolor the mud of the South China Sea moderate inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 40 A. versicolor the mud of the South China Sea potent inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 41 A. versicolor the mud of the South China Sea nonsigniﬁcant inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 42 A. versicolor the mud of the South China Sea nonsigniﬁcant inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 43 A. versicolor the mud of the South China Sea nonsigniﬁcant inhibitory activities against LPS-induced NO production and iNOS enzyme [40] 44 A. sp. SF-5976 an unidentiﬁed marine organism collected in the Ross Sea weak inhibitory activities against LPS-induced NO production in RAW 264.7 and BV2 cells [41] 45 A. sp. SF-5976 an unidentiﬁed marine organism collected in the Ross Sea moderate inhibitory activities against LPS-induced NO production in RAW 264.7 and BV2 cells [41] 46 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 47 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 48 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 49 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 50 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 51 A. 3. Production Environment versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HCT-116, HL-60 and K562 cell lines [42] 52 A. sp. SF-5280 an unidentiﬁed sponge collected at Cheju Island, Korea moderate inhibitory eﬀects against PTP1B activity [43] Table 1. Cont. Table 1. Cont. 13 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. Table 1. Cont. Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 53 A. violaceofuscus the inner part of marine sponge Reniochalina sp. collected from Xisha Islands in South China Sea potent anti-inﬂammatory activity against IL-10 expression of the LPS-induced THP-1 cells [44] 54 A. sp. DX4H marine shrimp collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China weak cytotoxic activities against PC3 cell line [45] 55 A. sp. DX4H marine shrimp collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China weak cytotoxic activities against PC3 cell line [45] 56 A. sp. DX4H marine shrimp collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China weak cytotoxic activities against PC3 cell line [45] 57 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China moderate cytotoxic activities and selective PTK inhibitory activities [42] 58 A. versicolor HDN08-60 the sediments (at a depth of 35 m) collected in the South China Sea, China nonsigniﬁcant cytotoxic activities against HeLa, HL-60, and K562 cell lines [42] 59 A. fumigatus SCSIO 41012 the deep-sea sediments (3614 m) collected from the Indian Ocean potent antifungal and antibacterial activities [46] 60 A. fumigatus SCSIO 41012 the deep-sea sediments (3614 m) collected from the Indian Ocean potent antibacterial activities [46] 61 A. fumigatus SCSIO 41012 the deep-sea sediments (3614 m) collected from the Indian Ocean potent antibacterial activities [46] 62 A. sp. CMB-M081F the marine sediment collected at an intertidal depth of 1 m near Shorncliﬀe, Queensland, Australia nonsigniﬁcant cytotoxic activities and inhibition on P-glycoprotein-mediated drug eﬄux [28] 63 A. sp. F452 submerged decaying wood oﬀthe shore of Jeju Island, Korea moderate cytotoxicity; nonsigniﬁcant antibacterial activity; weak inhibition against Na+/K+-ATPase [47] 64 A. sp. F452 submerged decaying wood oﬀthe shore of Jeju Island, Korea moderate cytotoxicity; nonsigniﬁcant antibacterial activity; weak inhibition against Na+/K+-ATPase [47] 65 A. sp. 3. Production Environment F452 submerged decaying wood oﬀthe shore of Jeju Island, Korea moderate cytotoxicity; nonsigniﬁcant antibacterial activity; weak inhibition against Na+/K+-ATPase [47] 66 A. sp. F452 submerged decaying wood oﬀthe shore of Jeju Island, Korea moderate cytotoxicity; nonsigniﬁcant antibacterial activity; weak inhibition against Na+/K+-ATPase [47] 67 A. sp. F452 submerged decaying wood oﬀthe shore of Jeju Island, Korea moderate cytotoxicity; nonsigniﬁcant antibacterial activity; weak inhibition against Na+/K+-ATPase [47] 68 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; nonsigniﬁcant inhibitory activity against TrxR [33] 69 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; potent inhibitory activity against TrxR [33] 70 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; nonsigniﬁcant inhibitory activity against TrxR [33] 71 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; nonsigniﬁcant inhibitory activity against TrxR [33] 72 A. versicolor LZD-14-1 the gorgonian Pseudopterogorgia sp. (LZD-14) collected from the South China Sea weak cytotoxic activity against A549; nonsigniﬁcant inhibitory activity against TrxR [33] 73 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 14 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. Table 1. Cont. Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 74 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 75 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 76 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 77 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China potent inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 78 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 79 A. 3. Production Environment ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China potent inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 80 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China potent inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 81 A. ﬂocculosus 16D-1 the inner tissue of marine sponge Phakellia fusca collected from Yongxing Island, China moderate inhibitory activity against IL-6 production in LPS-induced THP-1 cells [48] 82 A. candidus KUFA0062 the marine sponge Epipolasis sp. collected at Similan Island National Park (15–20 m), Thailand weak cytotoxic activity against eight cell lines; nonsigniﬁcant antibacterial activity [25] 83 A. sclerotiorum and P. citrinum the gorgonian Muricella ﬂexuosa collected from the South China Sea, Sanya, Hainan Province, China moderate brine shrimp lethality; nonsigniﬁcant antibacterial and anti-bioﬁlm activities [49] 84 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 85 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 86 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 87 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 88 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 89 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 90 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 91 A. fumigatus the marine ﬁsh Mugil cephalus cytotoxic tests are in progress [50] 92 A. sp. Z-4 the marine isopod Ligia oceanica collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China weak cytotoxic activity against PC3 cell line [51] 93 A. sp. Z-4 the marine isopod Ligia oceanica collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China weak cytotoxic activity against PC3 cell line [51] 94 A. sp. Z-4 the marine isopod Ligia oceanica collected in seaside of Dinghai, Zhejiang Province of China moderate cytotoxic activities against PC3 and HCT116 cell lines [52] 95 A. sp. Z-4 the marine isopod Ligia oceanica collected in seaside of Dinghai, Zhejiang Province of China no biological activity was tested [52] 96 A. sp. Z-4 the intestinal of the marine isopod Ligia oceanica weak cytotoxic activities against PC3 cell line [53] 97 A. sp. 3. Production Environment SCSIOW2 the deep marine sediment (2439 m) collected in the South China Sea weak inhibitory activity on NO production induced by lipopolysaccharide (LPS)/INF-γ [56] 98 A. violaceofuscus the inner part of marine sponge Reniochalina sp. collected from Xisha Islands in South China Sea nonsigniﬁcant anti-inﬂammatory activity against IL-10 expression of the LPS-induced THP-1 cells [44] 15 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 99 A. violaceofuscus the inner part of marine sponge Reniochalina sp. collected from Xisha Islands in South China Sea potent anti-inﬂammatory activity against IL-10 expression of the LPS-induced THP-1 cells [44] 100 A. versicolor ZLN-60 the mud (depth, 20 m) of the Yellow Sea, China nonsigniﬁcant cytotoxic activities and lipid-lowering eﬀect [57] A. sp. BM-05 and BM-05ML a brown algal species belonging to the genus Sargassum collected oﬀ Helgoland, North Sea, Germany moderate cytotoxicities against K562, HCT116, A2780, and A2780CisR cell lines [58] 101 A. versicolor ZLN-60 the mud (depth, 20 m) of the Yellow Sea, China nonsigniﬁcant cytotoxic activities and lipid-lowering eﬀect [57] 102 A. versicolor ZLN-60 the mud (depth, 20 m) of the Yellow Sea, China potent lipid-lowering eﬀect; nonsigniﬁcant cytotoxic activities [57] 103 A. versicolor ZLN-60 the mud (depth, 20 m) of the Yellow Sea, China nonsigniﬁcant cytotoxic activities and lipid-lowering eﬀect [57] 104 A. versicolor ZLN-60 the mud (depth, 20 m) of the Yellow Sea, China nonsigniﬁcant cytotoxic activities and lipid-lowering eﬀect [57] 105 A. terreus SCSGAF0162 the gorgonian coral Echinogorgia aurantiaca in the South China Sea nonsigniﬁcant antifouling activity towards larvae of the barnacle B. amphitrite [59] 106 A. sp. SCSIO 41501 the gorgonian Melitodes squamata collected from the South China Sea, Sanya, China moderate antiviral activity against HSV-1 under non-cytotoxic concentrations against Vero cells [23] 107 A. similanensis KUFA 0013 the marine sponge Rhabdermia sp. collected in coral reef of Similan Islands, Phang Nga, Thailand nonsigniﬁcant cytotoxic and antibacterial activities [60] 108 A. versicolor TA01-14 a gorgonian Carijoa sp. GX-WZ-2010001 collected in Weizhou coral reefs in the South China Sea weak cytotoxic activity; nonsigniﬁcant brine shrimp lethality, antibacterial and antiviral activities, as well as AChE, Top I, and α-glucosacharase inhibition [61] 109 A. similanensis KUFA 0013 the marine sponge Rhabdermia sp. collected in coral reef of Similan Islands, Phang Nga, Thailand weak cytotoxicity; nonsigniﬁcant antibacterial activities against four reference strains [60] 110 A. 3. Production Environment niger SCSIO Jcsw6F30 a marine alga Sargassum sp. collected in Yongxing Island, South China Sea potent cytotoxic activity against TZM-bl cells; moderate anti-HIV-1 activity against HIV-1 SF162 [62] 111 A. niger SCSIO Jcsw6F30 a marine alga Sargassum sp. collected in Yongxing Island, South China Sea no biological activity was tested [62] 112 A. sp. XS20090B15 the Muricella abnormaliz gorgonian collected from the Xisha Islands coral reef in South China Sea nonsigniﬁcant antiviral activity against RSV virus-induced cytopathogenicity in Hep-2 cells [63] 113 A. sp. XS20090B15 the Muricella abnormaliz gorgonian collected from the Xisha Islands coral reef in South China Sea potent antiviral activity against RSV virus-induced cytopathogenicity in Hep-2 cells [63] 114 A. versicolor A-21-2-7 the deep-sea sediment (3002 m) in South China Sea no biological activity was tested [64] 115 A. ﬂavus KMM 4650 Sakhalin Bay marine sediments (32 m, Sea of Okhotsk) nonsigniﬁcant antimicrobial activity [65] 116 A. versicolor the inner part of gorgonian D. gemmacea collected from the Xisha Islands coral reef of the South China Sea moderate antibacterial activities and brine shrimp lethality; nonsigniﬁcant cytotoxicities [66] 117 A. versicolor the inner part of gorgonian D. gemmacea collected from the Xisha Islands coral reef of the South China Sea moderate antibacterial activities and brine shrimp lethality; nonsigniﬁcant cytotoxicities [66] 16 of 22 16 of 22 Mar. Drugs 2020, 18, 54 Table 1. Cont. Table 1. Cont. NO. Producing Strain Environmental Source Biological Activities Ref. 118 A. ochraceus LCJ11-102 the gorgonian Dichotella gemmacea (Valenciennes) collected in Lingao, Hainan province of China moderate antimicrobial activity against E. aerogenes; nonsigniﬁcant cytotoxic activities [67] 119 A. ochraceus LCJ11-102 the gorgonian Dichotella gemmacea (Valenciennes) collected in Lingao, Hainan province of China nonsigniﬁcant antimicrobial and cytotoxic activities [67] 120 A. ochraceus LCJ11-102 the gorgonian Dichotella gemmacea (Valenciennes) collected in Lingao, Hainan province of China nonsigniﬁcant antimicrobial and cytotoxic activities [67] 121 A. ochraceus LCJ11-102 the gorgonian Dichotella gemmacea (Valenciennes) collected in Lingao, Hainan province of China moderate antimicrobial activity against E. aerogenes; nonsigniﬁcant cytotoxic activities [67] 122 A. ochraceus LCJ11-102 the gorgonian Dichotella gemmacea (Valenciennes) collected in Lingao, Hainan province of China nonsigniﬁcant antimicrobial and cytotoxic activities [67] 123 A. sclerotiorum and P. citrinum the gorgonian Muricella ﬂexuosa collected from the South China Sea, Sanya, Hainan Province, China potent brine shrimp lethality and cytotoxic activities; nonsigniﬁcant antibacterial and anti-bioﬁlm activities [49] 124 A. sclerotiorum and P. 3. Production Environment citrinum the gorgonian Muricella ﬂexuosa collected from the South China Sea, Sanya, Hainan Province, China moderate brine shrimp lethality and cytotoxic activities; nonsigniﬁcant antibacterial and anti-bioﬁlm activities [49] 125 A. sp. (33241) the mangrove Bruguiera sexangula var. rhynchopetala collected in the South China Sea nonsigniﬁcant antibacterial and cytotoxic activities [68] 126 A. terreus FA009 the marine sediment collected in Jeju Island, Korea moderate enhancement eﬀect on insulin sensitivity [69] 127 A. terreus FA009 the marine sediment collected in Jeju Island, Korea moderate enhancement eﬀect on insulin sensitivity [69] 128 A. sclerotiorum and P. citrinum the gorgonian Muricella ﬂexuosa collected from the South China Sea, Sanya, Hainan Province, China weak brine shrimp lethality; nonsigniﬁcant cytotoxic, antibacterial and anti-bioﬁlm activities [49] 129 A. sydowii SP-1 the marine sediment sample collected from site in the Antarctic Great Wall Station weak antimicrobial activities against MRSA and MRSE [70] 130 A. sp. WU 243 the crab Xenograpsus testudinatus collected from a Kueishantao hydrothermal vent, Taiwan, China no biological activity was tested [71] 4. Biological Activities The biological activities of these heterocyclic alkaloids are detailed in Table 1. Anticancer and antimicrobial activities, as well as anti-inﬂammatory activity, were the three main indexes that were used to assess the pharmacological activity of these natural heterocyclic alkaloids. In this section, alkaloids with potent biological activities are the focus, and detailed descriptions are provided below. 4.1. Anticancer Activities Figure 7 lists the anticancer heterocyclic alkaloids. Compound 11 at 20 µM was demonstrated to be a noncytotoxic inhibitor of P-glycoprotein-mediated drug eﬄux in multidrug-resistant (MDR) human colon cancer cells and it might be used to improve the prognosis for MDR cancer chemotherapy [28]. Compound 12 showed cytotoxicity against human PC-3, LNCaP, and 22Rv1 cells, with IC50 values of 69.4 µM, 47.8 µM, and 4.86 µM, respectively. The reference substance (Docetaxel) displayed IC50 values of 15.4 nM, 3.8 nM, and 12.7 nM, respectively. This compound was able to induce the apoptosis of 22Rv1 cells at low micromolar concentrations. Cell cycle progression analyses of 22Rv1 cells that were treated with 12 also revealed discrete G2/M-phase arrest [29]. Compound 18 exhibited potent 17 of 22 17 of 22 Mar. Drugs 2020, 18, 54 cytotoxic activity against HL-60 and LNCap cells with IC50 values of 3.1 and 4.9 µM, respectively, but with no signiﬁcant cytotoxicity against the rest of the tested cell lines (HepG2, Hela, A375, A549, HT29, SK-BR-3, and MCF-7) [31]. Compounds 20 and 69 exhibited signiﬁcant inhibitory activities against thioredoxin reductase with IC50 values of 12.2 ± 0.7 and 13.6 ± 0.6 µM (the IC50 of the positive control curcumin was 25 µM), but weak toxicity against A549 cells (IC50 > 10 µM), which suggested that these two compounds might act as microenvironmental regulators of tumor progression and metastasis [33]. Compound 123 possesses selective cytotoxicity against U937 cells with an IC50 value of 4.2 µM and mild cytotoxicity against HeLa and MCF-7 cells, with IC50 values of 29.3 µM and 24.8 µM, respectively. The IC50 values of doxorubicin (positive control) towards U937, HeLa, and MCF-7 cells were 0.06 µM, 0.8 µM, and 23.1 µM, respectively [49]. Figure 7. Anticancer heterocyclic alkaloids produced by marine-derived Aspergillus species. Figure 7. Anticancer heterocyclic alkaloids produced by marine-derived Aspergillus species. 4.2. Antimicrobial Activities Figure 8 lists the antimicrobial heterocyclic alkaloids. Compounds 59 and 61 were tested for their antimicrobial activities. Compound 59 showed comparable or even higher antibacterial activity than the other tested compounds. This compound also showed excellent antifungal activity against F. oxysporum with an MIC of 1.5 µg/mL. Compound 60 exhibited high activity against S. aureus (16339 and 29213), with MIC values of 1.565 µg/mL and 0.78 µg/mL, respectively, while compound 61 exhibited signiﬁcant activity against A. baumanii ATCC 19606 with an MIC of 6.25 µg/mL [46]. Although compounds 83 and 123 showed non-signiﬁcant antimicrobial activity against two common bacterial strains (S. aureus and P. aeruginosa) and three marine-derived bacteria (P. nigrifaciens, B. amyloliquefaciens, and B. stearothermophilus), they increased the growth of S. aureus one-fold at 100 µg/mL, and 123 increased the bioﬁlm formation of S. aureus 1.3-fold at 25 µg/mL [49]. Figure 8. Antimicrobial heterocyclic alkaloids that are produced by marine-derived Aspergillus species. Figure 8. Antimicrobial heterocyclic alkaloids that are produced by marine-derived Aspergillus species. 4.4. Other Biological Activities Figure 10 lists other bioactive alkaloids. Compound 102 was found to possess potent lipid-lowering eﬀects, but non-signiﬁcant cytotoxicity [57]. Compound 110 exhibited signiﬁcant HIV-1 inhibitory activities against SF162 infection in TZM-bl cells, with IC50 and CC50 values of 4.7 ± 0.4 and 35.0 ± 2.1 µM (selectivity index of 7.5), respectively, which might be beneﬁcial for the development of heterocyclic alkaloids as anti-HIV agents [62]. Compound 112 showed strong toxicity towards brine shrimp with an LC50 value of 6.1 µM as compared with the positive control toosendanin (LC50 = 1.73 µM). Compound 113 possessed outstanding anti-RSV activity with an IC50 value of 42 nM, being approximately 500-fold stronger than that of the positive control ribavirin (IC50 = 20 µM), as well as a higher therapeutic ratio (TC50/IC50 = 520) [63]. Compound 123 also showed strong toxicity, with an LC50 value of 6.1 µM as compared with the positive control toosendanin (LC50 = 1.73 µM) [49]. Figure 10. Other bioactive heterocyclic alkaloids produced by marine-derived Aspergillus species. Figure 10. Other bioactive heterocyclic alkaloids produced by marine-derived Aspergillus species. 4.3. Anti-Inﬂammatory Activities Figure 9 lists the anti-inﬂammatory heterocyclic alkaloids. Compound 2 showed potent anti-inﬂammatory activity against NO production with an IC50 of 24.64 µM [24]. Compound 40 exhibited an excellent inhibition of iNOS with an IC50 of 5.39 µM, but weak activity against Raw 264.7 cells. The inhibitory eﬀects might be the result of cell viability independent of concentration. 18 of 22 18 of 22 Mar. Drugs 2020, 18, 54 Molecular docking studies with 40 and iNOS showed that it could adopt an extended conformation and ﬁt well into the ligand binding site of mutant iNOS [40]. Compounds 53 and 99 were evaluated for their inhibitory activities against the production of cytokines in the serum of THP-1 by using the human inﬂammation cytometric bead array assay. The THP-1 cells that were pretreated with 53 and 99 showed a signiﬁcant decrease in the LPS-induced expression of IL-10, with inhibitory rates of 78.1% and 84.3% (p < 0.01), respectively. Moreover, these two compounds did not show cytotoxicity against THP-1 cells after 24 h of treatment [44]. Compounds 77, 79, and 80 showed potent inhibitory activity against IL-6 production, with IC50 values of 0.11 µM, 0.19 µM, and 2.3 µM, respectively [48]. Figure 9. Anti-inﬂammatory heterocyclic alkaloids produced by marine-derived Aspergillus species. Figure 9. Anti-inﬂammatory heterocyclic alkaloids produced by marine-derived Aspergillus species. References 1. Rida, P.C.; LiVecche, D.; Ogden, A.; Zhou, J.; Aneja, R. The noscapine chronicle: A pharmaco-historic biography of the opiate alkaloid family and its clinical applications. Med. Res. Rev. 2015, 35, 1072–1096. [CrossRef] [PubMed] 1. Rida, P.C.; LiVecche, D.; Ogden, A.; Zhou, J.; Aneja, R. The noscapine chronicle: A pharmaco-historic biography of the opiate alkaloid family and its clinical applications. Med. Res. Rev. 2015, 35, 1072–1096. [CrossRef] [PubMed] 2. Pirillo, A.; Catapano, A.L. Berberine, a plant alkaloid with lipid- and glucose-lowering properties: From in vitro evidence to clinical studies. Atherosclerosis 2015, 243, 449–461. [CrossRef] [PubMed] 2. Pirillo, A.; Catapano, A.L. Berberine, a plant alkaloid with lipid- and glucose-lowering properties: From in vitro evidence to clinical studies. Atherosclerosis 2015, 243, 449–461. [CrossRef] [PubMed] 3. Ferraz, C.A.A.; de Oliveira Júnior, R.G.; Picot, L.; da Silva Almeida, J.R.G.; Nunes, X.P. Pre-clinical investigations of β-carboline alkaloids as antidepressant agents: A systematic review. Fitoterapia 2019, 137, 104196. [CrossRef] [PubMed] 3. Ferraz, C.A.A.; de Oliveira Júnior, R.G.; Picot, L.; da Silva Almeida, J.R.G.; Nunes, X.P. Pre-clinical investigations of β-carboline alkaloids as antidepressant agents: A systematic review. Fitoterapia 2019, 137, 104196. [CrossRef] [PubMed] 4. Wang, B.G.; Gloer, J.B.; Ji, N.Y.; Zhao, J.C. Halogenated organic molecules of Rhodomelaceae origin: Chemistry and biology. Chem. Rev. 2013, 113, 3632–3685. [CrossRef] 4. Wang, B.G.; Gloer, J.B.; Ji, N.Y.; Zhao, J.C. Halogenated organic molecules of Rhodomelaceae origin: Chemistry and biology. Chem. Rev. 2013, 113, 3632–3685. [CrossRef] 5. Gogineni, V.; Schinazi, R.F.; Hamann, M.T. Role of marine natural products in the genesis of antivira Chem. Rev. 2015, 115, 9655–9706. [CrossRef] 5. Gogineni, V.; Schinazi, R.F.; Hamann, M.T. Role of marine natural products in the genesis of antivira Chem. Rev. 2015, 115, 9655–9706. [CrossRef] 6. Bideau, F.L.; Kousara, M.; Chen, L.; Wei, L.; Dumas, F. Tricyclic sesquiterpenes from marine origin. Chem. Rev. 2017, 117, 6110–6159. [CrossRef] 6. Bideau, F.L.; Kousara, M.; Chen, L.; Wei, L.; Dumas, F. Tricyclic sesquiterpenes from marine origin. Chem. Rev. 2017, 117, 6110–6159. [CrossRef] 7. Soldatou, S.; Baker, B.J. Cold-water marine natural products, 2006 to 2016. Nat. Prod. Rep. 2017, 34, 585–626. [CrossRef] 7. Soldatou, S.; Baker, B.J. Cold-water marine natural products, 2006 to 2016. Nat. Prod. Rep. 2017, 34, 585–626. [CrossRef] 8. Carroll, A.R.; Copp, B.R.; Davis, R.A.; Keyzers, R.A.; Prinsep, M.R. Marine natural products. Nat. Prod. Rep. 2019, 36, 122–173. [CrossRef] 9. Guo, C.J.; Wang, C.C.C. Recent advances in genome mining of secondary metabolites in Aspergillus terreus. Front. 5. Conclusions This review summarized the ﬁndings, including the biological activities, on a total of 130 nitrogen-containing secondary metabolites that originate from marine-derived Aspergillus species reported from the beginning of 2014 through the end of 2018. All of the original literature in the Web of Science database, which we believe covers most of the newly reported naturally occurring heterocyclic alkaloids from speciﬁc sources, was searched. However, several works may not have 19 of 22 19 of 22 Mar. Drugs 2020, 18, 54 been retrieved by the literature method used in this review. In the process of preparing this review, compound 23 was reported as speradine B and it was shown to possess an identical planar structure to that of speradine G (24), which was not explained in the original articles [35,36]. A careful comparison of the one-dimensional NMR data makes us boldly propose that these two compounds are a pair of diastereoisomers. Further, it is quite interesting that psychrophilin E (100) was reported as a new compound by two completely independent research teams in the same year [57,58]. Therefore, for natural product chemists, the research results should be published in a timely manner. At the end of this review, compounds with potent bioactivities were comprehensively described, which will be beneﬁcial in future drug development and innovation. Supplementary Materials: The following are available online at http://www.mdpi.com/1660-3397/18/1/54/s1, the structures of all 398 new metabolites produced by marine-derived Aspergillus species from 2014 to 2018, as well as references. Author Contributions: Conceptualization, K.X. and X.-D.L.; writing—Original draft preparation, K.X.; writing—Review and editing, X.-L.Y. and C.L.; funding acquisition, K.X. and X.-D.L. All authors have read and approved the ﬁnal manuscript. Funding: This research was supported by the National Natural Science Foundation of China (No. 81803375), the Agricultural Science and Technology Innovation Program (No. ASTIP-TRIC05), the Key Research and Development Program of Shandong Province (No. 2019GSF107091), the Fundamental Research Funds for the Central Non-Proﬁt Scientiﬁc Institution (No. 1610232019006), the National Postdoctoral Program for Innovative Talents (No. BX201700247), and the China Postdoctoral Science Foundation (No. 2018M630804). Acknowledgments: All of the authors would like to thank American Journal Experts (AJE) for their professional language editing. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Nat. Prod. Rep. 2010, 27, 57–78. [CrossRef] [PubMed] 22. Netz, N.; Opatz, T. Marine indole alkaloids. Mar. Drugs 2015, 13, 4814–4914. [CrossRef] [PubMed] 23. Ma, X.; Nong, X.H.; Ren, Z.; Wang, J.; Liang, X.; Wang, L.; Qi, S.H. Antiviral peptides from marine gorgonian-derived fungus Aspergillus sp. SCSIO 41501. Tetrahedron Lett. 2017, 58, 1151–1155. [CrossRef] 24. Liu, M.T.; Sun, W.G.; Wang, J.P.; He, Y.; Zhang, J.W.; Li, F.L.; Qi, C.X.; Zhu, H.C.; Xue, Y.B.; Hu, Z.X.; et al. Bioactive secondary metabolites from the marine-associated fungus Aspergillus terreus. Bioorg. Chem. 2018, 80, 525–530. [CrossRef] 25. Buttachon, S.; Ramos, A.A.; Inacio, A.; Dethoup, T.; Gales, L.; Lee, M.; Costa, P.M.; Silva, A.M.S.; Sekeroglu, N.; Rocha, E.; et al. Bis-indolyl benzenoids, hydroxypyrrolidine derivatives and other constituents from cultures of the marine sponge-associated fungus Aspergillus candidus KUFA0062. Mar. Drugs 2018, 16, 119. [CrossRef] 26. Yang, S.Q.; Li, X.M.; Li, X.; Chi, L.P.; Wang, B.G. Two new diketomorpholine derivatives and a new highly conjugated ergostane-type steroid from the marine algal-derived endophytic fungus Aspergillus alabamensis EN-547. Mar. Drugs 2018, 16, 114. [CrossRef] 27. Aparicio-Cuevas, M.A.; Rivero-Cruz, I.; Sanchez-Castellanos, M.; Menendez, D.; Raja, H.A.; Joseph-Nathan, P.; Gonzalez, M.D.; Figueroa, M. Dioxomorpholines and derivatives from a marine-facultative Aspergillus species. J. Nat. Prod. 2017, 80, 2311–2318. [CrossRef] 28. Khalil, Z.G.; Huang, X.C.; Raju, R.; Piggott, A.M.; Capon, R.J. Shornephine A: Structure, chemical stability, and P-glycoprotein inhibitory properties of a rare diketomorpholine from an Australian marine-derived Aspergillus sp. J. Org. Chem. 2014, 79, 8700–8705. [CrossRef] 29. Ivanets, E.V.; Yurchenko, A.N.; Smetanina, O.F.; Rasin, A.B.; Zhuravleva, O.I.; Pivkin, M.V.; Popov, R.S.; von Amsberg, G.; Aﬁyatullov, S.S.; Dyshlovoy, S.A. Asperindoles A-D and a p-terphenyl derivative from the ascidian-derived fungus Aspergillus sp. KMM 4676. Mar. Drugs 2018, 16, 232. [CrossRef] 30. Zhang, P.; Li, X.M.; Li, X.; Wang, B.G. New indole-diterpenoids from the algal-associated fungus Aspergillus nidulans. Phytochem. Lett. 2015, 12, 182–185. [CrossRef] 31. Zhou, R.; Liao, X.J.; Li, H.B.; Li, J.; Peng, P.J.; Zhao, B.X.; Xu, S.H. Isolation and synthesis of misszrtine A: A novel indole alkaloid from marine sponge-associated Aspergillus sp. SCSIO XWS03F03. Front. Chem. 2018, 6, 212. [CrossRef] [PubMed] 32. Zhou, Y.M.; Debbab, A.; Wray, V.; Lin, W.H.; Schulz, B.; Trepos, R.; Pile, C.; Hellio, C.; Proksch, P.; Aly, A.H. Marine bacterial inhibitors from the sponge-derived fungus Aspergillus sp. Tetrahedron Lett. 2014, 55, 2789–2792. [CrossRef] 33. Cheng, Z.B.; Liu, D.; Cheng, W.; Proksch, P.; Lin, W.H. References Microbiol. 2014, 5, 717. [CrossRef] 10. Yaegashi, J.; Oakley, B.R.; Wang, C.C.C. Recent advances in genome mining of secondary metabolite biosynthetic gene clusters and the development of heterologous expression systems in Aspergillus nidulans. J. Ind. Microbiol. Biotechnol. 2014, 41, 433–442. [CrossRef] 1. Anyaogu, D.C.; Mortensen, U.H. Heterologous production of fungal secondary metabolites in Asper Front. Microbiol. 2015, 6, 77. [CrossRef] [PubMed] 12. Wang, K.W.; Ding, P. New bioactive metabolites from the marine-derived fungi Aspergillus. Mini-Rev. Med. Chem. 2018, 18, 1072–1094. [CrossRef] [PubMed] 20 of 22 Mar. Drugs 2020, 18, 54 20 of 22 13. Frisvad, J.C.; Moller, L.L.H.; Larsen, T.O.; Kumar, R.; Arnau, J. Safety of the fungal workhorses of industrial biotechnology: Update on the mycotoxin and secondary metabolite potential of Aspergillus niger, Aspergillus oryzae, and Trichoderma reese. Appl. Microbiol. Biotechnol. 2018, 102, 9481–9515. [CrossRef] [PubMed] 13. Frisvad, J.C.; Moller, L.L.H.; Larsen, T.O.; Kumar, R.; Arnau, J. Safety of the fungal workhorses of industrial biotechnology: Update on the mycotoxin and secondary metabolite potential of Aspergillus niger, Aspergillus oryzae, and Trichoderma reese. Appl. Microbiol. Biotechnol. 2018, 102, 9481–9515. [CrossRef] [PubMed] 4. He, Y.; Wang, B.; Chen, W.P.; Cox, R.J.; He, J.R.; Chen, F.S. Recent advances in reconstructing micro secondary metabolites biosynthesis in Aspergillus spp. Biotechnol. Adv. 2018, 36, 739–783. [CrossRef] 15. Salvatore, M.M.; Nicoletti, R.; Salvatore, F.; Naviglio, D.; Andolﬁ, A. GC-MS approaches for the screening of metabolites produced by marine-derived Aspergillus. Mar. Chem. 2018, 206, 19–33. [CrossRef] 16. Romsdahl, J.; Wang, C.C.C. Recent advances in the genome mining of Aspergillus secondary metabolites (covering 2012–2018). Medchemcomm 2019, 10, 840–866. [CrossRef] 17. Xu, K.; Guo, C.; Meng, J.; Tian, H.; Guo, S.; Shi, D. Discovery of natural dimeric naphthopyrones as potential cytotoxic agents through ROS-mediated apoptotic pathway. Mar. Drugs 2019, 17, 207. [CrossRef] 18. Soosaraei, M.; Khasseh, A.A.; Fakhar, M.; Hezarjaribi, H.Z. A decade bibliometric analysis of global research on leishmaniasis in Web of Science database. Ann. Med. Surg. (Lond.) 2018, 26, 30–37. [CrossRef] 19. Chang, H.T.; Lin, M.H.; Hwang, I.H.; Chen, T.J.; Lin, H.C.; Hou, M.C.; Hwang, S.J. Scientiﬁc publications in gastroenterology and hepatology in Taiwan: An analysis of Web of Science from 1993 to 2013. J. Chin. Med. Assoc. 2017, 80, 80–85. [CrossRef] Singh, T.P.; Singh, O.M. Recent progress in biological activities of indole and indole alkaloids. Mini-Rev. Med Chem. 2018, 18, 9–25. [CrossRef] 21. Li, S.M. Prenylated indole derivatives from fungi: Structure diversity, biological activities, biosynthesis and chemoenzymatic synthesis. References Versiquinazolines L-Q, new polycyclic alkaloids from the marine-derived fungus Aspergillus versicolor. RSC Adv. 2018, 8, 31427–31439. [CrossRef] 34. Sun, K.L.; Li, Y.; Guo, L.; Wang, Y.; Liu, P.P.; Zhu, W.M. Indole diterpenoids and isocoumarin from the fungus, Aspergillus ﬂavus, isolated from the prawn, Penaeus vannamei. Mar. Drugs 2014, 12, 3970–3981. [CrossRef] 21 of 22 21 of 22 Mar. Drugs 2020, 18, 54 35. Hu, X.; Xia, Q.W.; Zhao, Y.Y.; Zheng, Q.H.; Liu, Q.Y.; Chen, L.; Zhang, Q.Q. Speradines B–E, four novel tetracyclic oxindole alkaloids from the marine-derived fungus Aspergillus oryzae. Heterocycles 2014, 89, 1662–1669. 36. Hu, X.; Xia, Q.W.; Zhao, Y.Y.; Zheng, Q.H.; Liu, Q.Y.; Chen, L.; Zhang, Q.Q. Speradines F–H, three new oxindole alkaloids from the marine-derived fungus Aspergillus oryzae. Chem. Pharm. Bull. 2014, 62, 942–946. [CrossRef] 37. Wang, P.M.; Zhao, S.Z.; Liu, Y.; Ding, W.J.; Qiu, F.; Xu, J.Z. Asperginine, an unprecedented alkaloid from the marine-derived fungus Aspergillus sp. Nat. Prod. Commun. 2015, 10, 1363–1364. [CrossRef] 38. Aﬁyatullov, S.S.; Zhuravleva, O.I.; Antonov, A.S.; Berdyshev, D.V.; Pivkin, M.V.; Denisenko, V.A.; Popov, R.S.; Gerasimenko, A.V.; von Amsberg, G.; Dyshlovoy, S.A.; et al. Prenylated indole alkaloids from co-culture of marine-derived fungi Aspergillus sulphureus and Isaria felina. J. Antibiot. 2018, 71, 846–853. [CrossRef] 39. Ma, Y.M.; Liang, X.A.; Kong, Y.; Jia, B. Structural diversity and biological activities of indole diketopiperazine alkaloids from fungi. J. Agric. Food Chem. 2016, 64, 6659–6671. [CrossRef] 40. Li, H.Q.; Sun, W.G.; Deng, M.Y.; Zhou, Q.; Wang, J.P.; Liu, J.J.; Chen, C.M.; Qi, C.X.; Luo, Z.W.; Xue, Y.B.; et al. Asperversiamides, linearly fused prenylated indole alkaloids from the marine-derived fungus Aspergillus versicolor. J. Org. Chem. 2018, 83, 8483–8492. [CrossRef] 41. Kwon, J.; Lee, H.; Ko, W.; Kim, D.C.; Kim, K.W.; Kwon, H.C.; Guo, Y.Q.; Sohn, J.H.; Yim, J.H.; Kim, Y.C.; et al. Chemical constituents isolated from Antarctic marine-derived Aspergillus sp. SF-5976 and their anti-inﬂammatory eﬀects in LPS-stimulated RAW 264.7 and BV2 cell. Tetrahedron 2017, 73, 3905–3912. [CrossRef] 42. Peng, J.X.; Gao, H.Q.; Li, J.; Ai, J.; Geng, M.Y.; Zhang, G.J.; Zhu, T.J.; Gu, Q.Q.; Li, D.H. Prenylated indole diketopiperazines from the marine-derived fungus Aspergillus versicolor. J. Org. Chem. 2014, 79, 7895–7904. [CrossRef] [PubMed] 43. Cho, K.H.; Sohn, J.H.; Oh, H. Isolation and structure determination of a new diketopiperazine dimer from marine-derived fungus Aspergillus sp. SF-5280. Nat. Prod. Res. 2018, 32, 214–221. [CrossRef] [PubMed] 44. Liu, J.T.; Gu, B.B.; Yang, L.J.; Yang, F.; Lin, H.W. References Res. 2016, 30, 52–57. [CrossRef] 57. Peng, J.X.; Gao, H.Q.; Zhang, X.M.; Wang, S.; Wu, C.M.; Gu, Q.Q.; Guo, P.; Zhu, T.J.; Li, D.H. Psychrophilins E-H and Versicotide C, cyclic peptides from the marine-derived fungus Aspergillus versicolor ZLN-60. J. Nat. Prod. 2014, 77, 2218–2223. [CrossRef] 58. Ebada, S.S.; Fischer, T.; Hamacher, A.; Du, F.Y.; Roth, Y.O.; Kassack, M.U.; Wang, B.G.; Roth, E.H. Psychrophilin E, a new cyclotripeptide, from co-fermentation of two marine alga-derived fungi of the genus Aspergillus. Nat. Prod. Res. 2014, 28, 776–781. [CrossRef] 59. Nong, X.H.; Zhang, X.Y.; Xu, X.Y.; Qi, S.H. Antifouling compounds from the marine-derived fungus Aspergillus terreus SCSGAF0162. Nat. Prod. Commun. 2015, 10, 1033–1034. [CrossRef] 60. Prompanya, C.; Fernandes, C.; Cravo, S.; Pinto, M.M.M.; Dethoup, T.; Silva, A.M.S.; Kijjoa, A. A new cyclic hexapeptide and a new isocoumarin derivative from the marine sponge-associated fungus Aspergillus similanensis KUFA 0013. Mar. Drugs 2015, 13, 1432–1450. [CrossRef] 61. Hou, X.M.; Zhang, Y.H.; Hai, Y.; Zheng, J.Y.; Gu, Y.C.; Wang, C.Y.; Shao, C.L. Aspersymmetide A, a new centrosymmetric cyclohexapeptide from the marine-derived fungus Aspergillus versicolor. Mar. Drugs 2017, 15, 363. [CrossRef] [PubMed] 62. Zhou, X.F.; Fang, W.; Tan, S.Y.; Lin, X.P.; Xun, T.R.; Yang, B.J.; Liu, S.W.; Liu, Y.H. Aspernigrins with anti-HIV-1 activities from the marine-derived fungus Aspergillus niger SCSIO Jcsw6F30. Bioorg. Med. Chem. Lett. 2016, 26, 361–365. [CrossRef] 63. Chen, M.; Shao, C.L.; Meng, H.; She, Z.G.; Wang, C.Y. Anti-respiratory syncytial virus prenylated dihydroquinolone derivatives from the gorgonian-derived fungus Aspergillus sp. XS-20090B15. J. Nat. Prod. 2014, 77, 2720–2724. [CrossRef] [PubMed] 64. Wu, Z.H.; Liu, D.; Xu, Y.; Chen, J.L.; Lin, W.H. Antioxidant xanthones and anthraquinones isolated from a marine-derived fungus Aspergillus versicolor. Chin. J. Nat. Med. 2018, 16, 219–224. [CrossRef] 65. Zhuravleva, O.I.; Kirichuk, N.N.; Denisenko, V.A.; Dmitrenok, P.S.; Pivkin, M.V.; Afiyatullov, S.S. New kipukasin from marine isolate of the fungus Aspergillus flavus. Chem. Nat. Compd. 2016, 52, 266–268. [CrossRef] 6. Chen, M.; Fu, X.M.; Kong, C.J.; Wang, C.Y. Nucleoside derivatives from the marine-derived fungus Asperg versicolor. Nat. Prod. Res. 2014, 28, 895–900. [CrossRef] 67. Peng, X.P.; Wang, Y.; Zhu, T.H.; Zhu, W.M. Pyrazinone derivatives from the coral-derived Aspergillus ochraceus LCJ11-102 under high iodide salt. Arch. Pharm. Res. 2018, 41, 184–191. [CrossRef] 68. Zheng, C.J.; Wu, L.Y.; Li, X.B.; Song, X.M.; Niu, Z.G.; Song, X.P.; Chen, G.Y.; Wang, C.Y. Structure and absolute conﬁguration of Aspergilumamide A, a novel lumazine peptide from the mangrove-derived fungus Aspergillus sp. Helv. Chim. References New anti-inﬂammatory cyclopeptides from a sponge-derived fungus Aspergillus violaceofuscus. Front. Chem. 2018, 6, 226. [CrossRef] 5. Xu, J.Z.; Hu, Q.; Ding, W.J.; Wang, P.M.; Di, Y.N. New asymmetrical bispyrrolidinoindoline diketopiperaz from the marine fungus Aspergillus sp. DX4H. Nat. Prod. Res. 2018, 32, 815–820. [CrossRef] 46. Limbadri, S.; Luo, X.W.; Lin, X.P.; Liao, S.R.; Wang, J.F.; Zhou, X.F.; Yang, B.; Liu, Y.H. Bioactive novel indole alkaloids and steroids from deep sea-derived fungus Aspergillus fumigatus SCSIO 41012. Molecules 2018, 23, 2379. [CrossRef] 47. Liao, L.; You, M.; Chung, B.K.; Oh, D.C.; Oh, K.B.; Shin, J. Alkaloidal metabolites from a marine-derived Aspergillus sp. fungus. J. Nat. Prod. 2015, 78, 349–354. [CrossRef] 48. Gu, B.B.; Jiao, F.R.; Wu, W.; Jiao, W.H.; Li, L.; Sun, F.; Wang, S.P.; Yang, F.; Lin, H.W. Preussins with inhibition of IL-6 expression from Aspergillus ﬂocculosus 16D-1, a fungus isolated from the marine sponge Phakellia fusca. J. Nat. Prod. 2018, 81, 2275–2281. [CrossRef] 49. Bao, J.; Wang, J.; Zhang, X.Y.; Nong, X.H.; Qi, S.H. New furanone derivatives and alkaloids from the co-culture of marine-derived fungi Aspergillus sclerotiorum and Penicillium citrinum. Chem. Biodivers. 2017, 14, e1600327. [CrossRef] 50. Yamada, T.; Kimura, H.; Arimitsu, K.; Kajimoto, T.; Kikuchi, T.; Tanaka, R. Absolute conﬁguration of eight cephalimysins isolated from the marine-derived Aspergillus fumigatus. ChemistrySelect 2017, 2, 10936–10940. [CrossRef] 51. Li, X.Y.; Ding, W.J.; Wang, P.M.; Xu, J.Z. Two novel aspochalasins from the gut fungus Aspergillus sp. Z4. Mar. Drugs 2018, 16, 343. [CrossRef] [PubMed] 52. Liu, Y.; Zhao, S.Z.; Ding, W.J.; Wang, P.M.; Yang, X.W.; Xu, J.Z. Methylthio-aspochalasins from a marine-derived fungus Aspergillus sp. Mar. Drugs 2014, 12, 5124–5131. [CrossRef] [PubMed] 53. Li, X.Y.; Zhao, Z.H.; Ding, W.J.; Ye, B.; Wang, P.M.; Xu, J.Z. Aspochalazine A, a novel polycyclic aspochalasin from the fungus Aspergillus sp. Z4. Tetrahedron Lett. 2017, 58, 2405–2408. [CrossRef] 54. Wang, X.; Lin, M.; Xu, D.; Lai, D.; Zhou, L. Structural diversity and biological activities of fungal cyclic peptides, excluding cyclodipeptides. Molecules 2017, 22, 2069. [CrossRef] 55. Lee, Y.; Phat, C.; Hong, S.C. Structural diversity of marine cyclic peptides and their molecular mechanisms for anticancer, antibacterial, antifungal, and other clinical applications. Peptides 2017, 95, 94–105. [CrossRef] 22 of 22 Mar. Drugs 2020, 18, 54 56. Zhou, X.; Fang, P.Y.; Tang, J.Q.; Wu, Z.Q.; Li, X.F.; Li, S.M.; Wang, Y.; Liu, G.; He, Z.D.; Gou, D.M.; et al. A novel cyclic dipeptide from deep marine-derived fungus Aspergillus sp. SCSIOW2. Nat. Prod. References Acta 2015, 98, 368–373. [CrossRef] 69. You, M.; Liao, L.; Hong, S.H.; Park, W.; Kwon, D.I.; Lee, J.; Noh, M.; Oh, D.C.; Oh, K.B.; Shin, J. Lumazine peptides from the marine-derived fungus Aspergillus terreus. Mar. Drugs 2015, 13, 1290–1303. [CrossRef] 70. Li, W.T.; Luo, D.; Huang, J.N.; Wang, L.L.; Zhang, F.G.; Xi, T.; Liao, J.M.; Lu, Y.Y. Antibacterial constituents from Antarctic fungus, Aspergillus sydowii SP-1. Nat. Prod. Res. 2018, 32, 662–667. [CrossRef] 70. Li, W.T.; Luo, D.; Huang, J.N.; Wang, L.L.; Zhang, F.G.; Xi, T.; Liao, J.M.; Lu, Y.Y. Antibacterial constituents from Antarctic fungus, Aspergillus sydowii SP-1. Nat. Prod. Res. 2018, 32, 662–667. [CrossRef] 71. Ding, C.H.; Wu, X.D.; Auckloo, B.N.; Chen, C.T.; Ye, Y.; Wang, K.W.; Wu, B. An unusual stress metabolite from a hydrothermal vent fungus Aspergillus sp. WU 243 induced by cobalt. Molecules 2016, 21, 105. [CrossRef] [PubMed] 72. Strobel, G.; Daisy, B. Bioprospecting for microbial endophytes and their natural products. Microbiol. Mol. Biol. Rev. 2003, 67, 491–502. [CrossRef] [PubMed] 73. Bugni, T.S.; Ireland, C.M. Marine-derived fungi: A chemically and biologically diverse group of microorganisms. Nat. Prod. Rep. 2004, 21, 143–163. [CrossRef] [PubMed] 74. Rateb, M.E.; Ebel, R. Secondary metabolites of fungi from marine habitats. Nat. Prod. Rep. 2011, 28, 290–344. [CrossRef] 75. Zhang, P.; Li, X.; Wang, B.G. Secondary metabolites from the marine algal-derived endophytic fungi: Chemical diversity and biological activity. Planta Med. 2016, 82, 832–842. [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/).
https://openalex.org/W4313440897	https://acopen.umsida.ac.id/index.php/acopen/article/download/5478/1492	Indonesian	null	The Relationship between Parental Social Support with Achievement Motivation in Senior High School Students	Academia Open	2,022	cc-by	363	Artikel by Pradnya Ayu Paramita Submission date: 26-Sep-2022 06:05PM (UTC+0700) Submission ID: 1909327851 File name: Artikel_Pradnya.docx (63.98K) Word count: 3421 Character count: 21862 16 3 7 14 14 7 23 6 10 22 8 8 10 24 9 2 2 22 1 1 4 11 17 21 5 5 5 1 11 17 21 1 11 21 14 5 5 5 14 20 19 12 12 11 18 18 15 3 3 3 3 13 13 7 1 20% SIMILARITY INDEX 20% INTERNET SOURCES 15% PUBLICATIONS % STUDENT PAPERS 1 2% 2 1% 3 1% 4 1% 5 1% 6 1% 7 1% 8 1% 9 1% Artikel ORIGINALITY REPORT PRIMARY SOURCES ejurnal.undana.ac.id Internet Source journal.uinmataram.ac.id Internet Source e-journal.hamzanwadi.ac.id Internet Source ejournal.uin-suska.ac.id Internet Source jurnal.unublitar.ac.id Internet Source jurnal.primagraha.ac.id Internet Source www.scilit.net Internet Source kemendikbud.my.id Internet Source repository.metrouniv.ac.id Internet Source Artikel ORIGINALITY REPORT repository.stei.ac.id 18 jurnal.primagraha.ac.id Internet Source 8 9 1% % 10 1% 11 1% 12 1% 13 1% 14 1% 15 1% 16 1% 17 1% 18 1% vdocuments.mx Internet Source Joshy Herliani Joshy Herliani, Jumaini Jumaini, Erna Marni Erna Marni. "HUBUNGAN DUKUNGAN SUAMI TERHADAP MOTIVASI IBU DALAM MERAWAT ANAK DENGAN AUTIS", Jurnal Keperawatan Hang Tuah (Hang Tuah Nursing Journal), 2021 Publication tunasbangsa.ac.id Internet Source eprints.umpo.ac.id Internet Source ejournal.kopertais4.or.id Internet Source ejournal.uinib.ac.id Internet Source journal3.um.ac.id Internet Source repository.stei.ac.id Internet Source repository.unwidha.ac.id:880 Internet Source 1% vdocuments.mx Internet Source vdocuments.mx vdocuments.mx Internet Source 10 % 1% Joshy Herliani Joshy Herliani, Jumaini Jumaini, Erna Marni Erna Marni. "HUBUNGAN DUKUNGAN SUAMI TERHADAP MOTIVASI IBU DALAM MERAWAT ANAK DENGAN AUTIS", Jurnal Keperawatan Hang Tuah (Hang Tuah Nursing Journal), 2021 Publication Joshy Herliani Joshy Herliani, Jumaini Jumaini, Erna Marni Erna Marni. "HUBUNGAN DUKUNGAN SUAMI TERHADAP MOTIVASI IBU DALAM MERAWAT ANAK DENGAN AUTIS", Jurnal Keperawatan Hang Tuah (Hang Tuah Nursing Journal), 2021 Publication 11 % 1% 12 % www.journal.uad.ac.id Internet Source 19 % 20 1% 21 1% 22 1% 23 1% 24 1% Citra Wahyuni, Emiel Yusuf Costadinov. "HUBUNGAN ANTARA DUKUNGAN TEMAN SEBAYA DENGAN KEPERCAYAAN DIRI BERBICARA DI DEPAN UMUM PADA MAHASISWA", Jurnal Psikologi Malahayati, 2020 Publication fpsi.mercubuana-yogya.ac.id Internet Source journal.upgris.ac.id Internet Source ojs.unm.ac.id Internet Source sangkudaapi.wordpress.com Internet Source 20 % journal.upgris.ac.id 24 Exclude quotes On Exclude bibliography On
https://openalex.org/W3016666583	https://europepmc.org/articles/pmc7202535?pdf=render	English	null	High miR-31-5p expression promotes colon adenocarcinoma progression by targeting TNS1	Aging	2,020	cc-by	5,693	ABSTRACT Overexpression of the miR-31-5p contributes to tumorigenesis and metastasis in diverse neoplasms. In this study, we evaluated expression of miR-31-5p in patients with colon adenocarcinoma (COAD). We found that miR-31-5p was overexpressed in four cohorts (GSE30454, GSE41655, GSE18392, GSE108153) of COAD patients. Importantly, a LinkedOmics analysis revealed that high miR-31-5p expression was associated with poor overall survival of COAD patients. At total of 133 putative target genes of miR-31-5p were identified from TargetScan, miRDB, and TargetMiner. After integrating the target genes with 1,556 deregulated genes in COAD, 8 were acquired that may be targeted by miR-31-5p and contribute to COAD progression. Among these, tensin 1 (TNS1) showed the greatest prognostic ability in COAD and was strongly correlated with M2 macrophages, regulatory T cells, and other immune cells. These findings indicate that, in COAD, miR-31-5p is a potential prognostic factor that affects immune infiltration by targeting TNS1. High miR-31-5p expression promotes colon adenocarcinoma progression by targeting TNS1 1Department of Orthopedics, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China 2Department of Cardiology, Beijing Chaoyang Integrative Medicine Emergency Medical Center, Beijing 100029, China 3Department of oncology, The Third Affiliated Hospital, Beijing University of Chinese Medicine, Beijing 100029, China 4Surgery Department, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115, USA *Equal contribution rrespondence to: Guohui Liu, Jiayang Sai; email: liuguohui@hust.edu.cn, saijiayang90@126.com ywords: colorectal adenocarcinoma, miR-31-5p, TNS1, immune profiling, tumor microenvironment ceived: December 11, 2019 Accepted: March 30, 2020 Published: April 21, 2020 Correspondence to: Guohui Liu, Jiayang Sai; email: liuguohui@hust.edu.cn, saijiayang90@126.com Keywords: colorectal adenocarcinoma, miR-31-5p, TNS1, immune profiling, tumor microenvironment Received: December 11, 2019 Accepted: March 30, 2020 Published: April 21, 2020 Copyright: Mi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. AGING 2020, Vol. 12, No. 8 www.aging-us.com Research Paper AGING 2020, Vol. 12, No. 8 AGING 2020, Vol. 12, No. 8 INTRODUCTION and participate in various physiologic processes of cells, including posttranscriptional regulation of gene expression [5, 6]. In cancer, miRNAs regulate various biologic processes of tumor cells, including proliferation, migration, and apoptosis [7, 8]. Notably, aberrant expression of miRNAs contributes to development and progression of COAD [9]. The strong evidence that dysregulation of miRNAs contributes to COAD patho- genesis provides a rationale for targeting miRNAs in cancer treatment. Colorectal cancer (CRC) is a common malignancy throughout the world. CRC accounts for 8% of all cancer-related deaths [1], and the incidence of CRC is increasing, especially in Asian countries [2]. Colon adenocarcinoma (COAD), which accounts for 90% of all CRCs, is the most common histologic subtype [3]. Despite advances in technology, patients with COAD still have a poor prognosis, especially when metastasis to the lymph nodes or distant organs is present. Therefore, understanding the underlying mechanisms in COAD is important to provide new concepts for novel therapies for advanced COAD [4]. Using bioinformatic analysis, miRNAs that exhibit oncogenic activity in carcinogenesis have been identified, including miR-21 [10], miR-155 [11], and miR-142 [12]. Literature indicates that miR-31-5p is overexpressed in diverse tumor types [13], and studies have explored its role in lung and breast cancer MicroRNAs (miRNAs), a family of noncoding RNAs, are prevalent in multicellular and complex eukaryotes AGING 7480 www.aging-us.com metastasis [14, 15]. However, the underlying mechanisms of the role of miR-31-5p in COAD remain unknown. Although it has been postulated that miR-31- 5p regulates the WNT and Hippo signaling pathways to promote epithelial regeneration following injury [16, 17], we wanted to investigate its function in COAD. Thus, we aimed to evaluate the prognostic value of miR-31-5p and its putative oncogenic functions in COAD and demonstrate the underlying potential mechanism through data mining. hepatocellular carcinoma, and head and neck cancer (Figure 1A). To further validate high expression of miR-31-5p in COAD, microarray data from GEO databases were collected and comparisons between COAD and normal colon tissues were conducted by GEO2R. COAD tissue had significantly higher miR-31- 5p expression than the control group tissue in the GSE30454, GSE41655, GSE18392, and GSE108153 data sets (Figure 1B). miR-31-5p expression in COAD In COAD patients with microsatellite stability (MSS) phenotype, miR-31-5p expression was significantly different depending on histologic type (P = 5.97e–4), number of involved lymph nodes (P = 2.5e–3), and pathologic stage (P = 5.4e–3). In silico exploration of miR-31-5p targets and their prognostic value To identify the mechanisms underlying miR-31-5p involvement in different biologic processes of COAD development, the potential targets of miR-31-5p were identified with miRWalk 3.0 target prediction tools. We retrieved target sets from three website servers: TargetScan, miRDB, and TargetMiner. The results of the three predicted target gene sets were integrated by drawing a Venn diagram. One hundred thirty-three overlapping genes were identified as promising targets of miR-31-5p (Figure 2C). miR-31-5p expression in COAD To assess whether miR-31-5p could be used as a prognostic predictor of COAD, a Kaplan-Meier curve was generated using the LinkedOmics Database. As depicted in Figure 2A, 2B, no statistical difference in survival outcome was observed between COAD patients exhibiting lower and higher expression of miR-31-5p (P = 0.091). Because the miR-31-5p expression in COAD patients was significantly different depending on pathologic N stage (P = 0.038), we To evaluate the expression level of miR-31-5p in various cancers, we performed a systematic pancancer analysis based on The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) miRNA databases. The results showed that miR-31-5p is over- expressed in many tumor types, including CRC, breast cancer, lung squamous cell carcinoma, liver gure 1. Expression validation of miR-31-5p. (A) Pancancer expression of miR-31-5p in GEDS. (B) Upregulation of miR-31-5p in the icroarrays tissues, based on Gene Expression Omnibus (GEO) data sets. Notes: a, GSE18392; b, GSE108153; c, GSE30454; d, GSE41655. Figure 1. Expression validation of miR-31-5p. (A) Pancancer expression of miR-31-5p in GEDS. (B) Upregulation of miR-31-5p in the microarrays tissues, based on Gene Expression Omnibus (GEO) data sets. Notes: a, GSE18392; b, GSE108153; c, GSE30454; d, GSE41655. AGING 7481 www.aging-us.com DEGs were set as follows: \| Log2 fold change (FC) \| ≥ 2 and P value < 0.01. After integrating the 133 miR-31-5p target genes with the 1,556 DEGs in COAD, we identified 8 specific genes (SLC6A6, SATB2, TNS1, KRT80, FGF7, CACNB2, MAP1B, and DMD) that might be targeted by miR-31-5p and play a role in COAD progression (Figure 2E). Expression of the 8 genes was evaluated using the UALCAN database. As shown in Figure 3A, 6 genes (SATB2, TNS1, FGF7, CACNB2, MAP1B, and DMD) were underexpressed in tissue samples of COAD patients, which might be related to the overexpression of miR-31-5p in COAD. CBioPortal for Cancer Genomics (cBioPortal, https://www.cbioportal.org/) was applied to explore genetic alterations of these 6 genes. Analysis showed that DMD, TNS1, and MAP1B are the most frequently altered genes with a high ratio of missense mutations based on 619 samples from DFCI COAD data sets (Figure 3B). further evaluated the prognostic value of miR-31-5p in different pathologic stages and found that high miR-31- 5p expression was significantly correlated with a poor prognosis in patients with pathologic stage IV COAD (P = 0.016). Validation of TNS1 expression and its prognostic value To further determine the prognostic value of the 6 genes in COAD, we performed an overall survival (OS) analysis using GEPIA tools, which showed that high From GEPIA, 1,556 deregulated genes (DEGs) in COAD were identified (Figure 2D). The thresholds of Figure 2. Prognostic value of miR-31-5p and target genes related with COAD. (A, B) Kaplan-Meier curve for miR-31-5p in clinical COAD samples. The P values of the Kaplan-Meier curve for COAD patients and pathologic stage IV COAD patients were 0.091 and 0.017, respectively. (C) Integration of miR-31-5p predictive genes from TargetScan, miRDB, and TargetMiner. (D) The differentially expressed genes in COAD retrieved from GEPIA. The thresholds were set as follows: \| Log2 fold change (FC) \| ≥ 2 and P value<0.01. (E) Venn diagram for overlap analysis of miR-31-5p target genes related to COAD. Figure 2. Prognostic value of miR-31-5p and target genes related with COAD. (A, B) Kaplan-Meier curve for miR-31-5p in clinical COAD samples. The P values of the Kaplan-Meier curve for COAD patients and pathologic stage IV COAD patients were 0.091 and 0.017, respectively. (C) Integration of miR-31-5p predictive genes from TargetScan, miRDB, and TargetMiner. (D) The differentially expressed genes in COAD retrieved from GEPIA. The thresholds were set as follows: \| Log2 fold change (FC) \| ≥ 2 and P value<0.01. (E) Venn diagram for overlap analysis of miR-31-5p target genes related to COAD. AGING AGING 7482 www.aging-us.com expression of TNS1 is correlated with a poor prognosis (P = 0.03). Furthermore, the prognostic value of TNS1 was validated using LinkedOmics (P = 0.055), the UALCAN database (TCGA samples; P = 0.022), and the PrognoScan database (P = 0.019; expression histogram: 221747_at) (Figure 4A–4D). Data from LinkedOmics indicated that expression of TNS1 is significantly different depending on number of involved lymph nodes, pathologic N stage, and pathologic stage (P = 0.005, 0.011, and 0.048, respectively). TNS1 protein levels in COAD and normal colon tissues were acquired from The Human Protein Atlas (THPA) and are presented in Figure 4E. Distinctly positive TNS1 protein was observed in the epithelium of normal colon tissues, whereas the majority of malignant cells displayed weak cytoplasmic immunoreactivity. Validation of TNS1 expression and its prognostic value GEPIA, which indicated TNS1 is closely associated with naïve T cells (r = 0.43, P = 4.3e–14), effector T cells (r = 0.48, P = 0), central memory T cells (r = 0.43, P = 4.4e–14), resident memory T cells (r = 0.41, P = 1.3e–12), exhausted T cells (r = 0.42, P = 3.9e–13), resting regulatory T (Treg) cells (r = 0.43, P = 1.3e–13), and effector Treg cells (r = 0.41, P = 1.1e–12). In addition, correlation of TNS1 with markers of the M1 (PTGS2, IRF5) and M2 (CD163, MS4A4A) phenotypes was analyzed, resulting in correlation indices of 0.1, 0.21, 0.44, and 0.48, with P values of 0.087, 0.00, 1.4e−14, and 0.00, respectively (Figure 6). The results suggest that TNS1 is more closely related to M2 than M1, indicating TNS1 might regulate macrophage polarization in COAD. Correlation of TNS1 expression with immune infiltration level in COAD MiR-31 is a significant prognostic predictor in various neoplasms [18] and plays a major role in regulating tumorigenesis in ovarian, breast, lung, and renal cell carcinoma [19–21]. RNA immunoprecipitation results from a microarray study showed that LINC01116 competed with VEGFA to bind with miR-31-5p in tumorigenesis of glioma [22]. Studies have shown that miR-31 expression correlates inversely with metastasis in breast cancer patients, which is achieved via coordinated repression of RhoA [14]. In addition, miR-31 We assessed the correlation of TNS1 expression with immune infiltration level in COAD using TIMER. As shown in Figure 5, TNS1 expression was significantly correlated with tumor purity, macrophages (r = 0.602, P = 3.15e–41), CD4+ T cells (r = 0.527, P = 3.92e–30), dendritic cells (r = 0.468, P = 3.19e–23), and neutrophils (r = 0.403, P = 4.38e–17). The correlation of TNS1 expression with T cells was further evaluated using Figure 3. Expression and mutation analysis of miR-31-5p target genes related to COAD. (A) Expression of integrated genes in COAD and normal tissues based on TCGA samples analyzed by the UALCAN database. (B) OncoPrint of integrated gene alterations in COAD. Genomic alterations of the 8 genes are mutually exclusive. Figure 3. Expression and mutation analysis of miR-31-5p target genes related to COAD. (A) Expression of integrated genes in COAD and normal tissues based on TCGA samples analyzed by the UALCAN database. (B) OncoPrint of integrated gene alterations in COAD. Genomic alterations of the 8 genes are mutually exclusive. Figure 3. Expression and mutation analysis of miR-31-5p target genes related to COAD. (A) Expression of integrated genes in COAD and normal tissues based on TCGA samples analyzed by the UALCAN database. (B) OncoPrint of integrated gene alterations in COAD. Genomic alterations of the 8 genes are mutually exclusive. AGING 7483 www.aging-us.com is also involved in immune and inflammation responses, such as regulating T-cell exhaustion during chronic viral infection [23] and acting as a negative regulator of the noncanonical NF-κB pathway in adult T-cell leukemia [24]. of miR-31 was shown to promote COAD cell growth, invasion, and migration in vitro by repressing its target gene SATB2 [25]. In addition, elevated expression of the long noncoding RNA MIR31HG, the host transcript of miR-31-5p, has been associated with poor prognosis in COAD patients independent of consensus molecular subtypes and cytotoxic T lymphocyte and fibroblast infiltration [26]. Correlation of TNS1 expression with immune infiltration level in COAD Recent studies also revealed that miR- The role of miR-31 in COAD has been explored in several studies. In one study, exogenetic overexpression Kaplan-Meier curve and histochemistry for TNS1 in clinical COAD samples. (A) Kaplan-Meier curve for TNS1 of COAD alyzed in the LinkedOmics database. (B) Kaplan-Meier curve for TNS1 of COAD patients analyzed in the UALCAN database. (C) er curve for TNS1 of COAD patients analyzed in GEPIA. (D) Kaplan-Meier curve for TNS1 of COAD patients analyzed in the database. (E) Histochemistry of TNS1 in normal colon and colorectal adenocarcinoma tissues. The expression distribution of TNS1 olon tissue and CRC patient samples was evaluated in THPA. Normal HPA036089 (ID1857) and tumor HPA036089 (ID 3550) were Figure 4. Kaplan-Meier curve and histochemistry for TNS1 in clinical COAD samples. (A) Kaplan-Meier curve for TNS1 of COAD patients analyzed in the LinkedOmics database. (B) Kaplan-Meier curve for TNS1 of COAD patients analyzed in the UALCAN database. (C) Kaplan-Meier curve for TNS1 of COAD patients analyzed in GEPIA. (D) Kaplan-Meier curve for TNS1 of COAD patients analyzed in the PrognoScan database. (E) Histochemistry of TNS1 in normal colon and colorectal adenocarcinoma tissues. The expression distribution of TNS1 in normal colon tissue and CRC patient samples was evaluated in THPA. Normal HPA036089 (ID1857) and tumor HPA036089 (ID 3550) were presented. AGING 7484 www.aging-us.com was performed, which identified a pivotal miR-31-5p target gene, TNS1, which is downregulated in COAD patients and closely correlated with COAD progression. This is consistent with a recent study that found that TNS1 level was negatively correlated with miR-31 in COAD tumor tissues [31]. TNS1 is a key component of specialized cellular adhesions that bind to extra- cellular fibronectin fibrils [32]. One study demonstrated that TNS1 was expressed in normal tissues but had greatly reduced expression in tumor tissues [33] and was associated with tumorigenesis. However, it is controversial whether TNS1 plays a negative or positive role in carcinogenesis. Zhou et al. reported that TNS1 was highly expressed in human CRC cell lines SW620 and RKO and promoted CRC cell proliferation and invasion [34]. Zhang et al. reported that miR-548j promoted human breast cancer invasiveness by downregulating TNS1 expression [33]. An in vitro study indicated that higher expression of TNS1 promoted metastasis [35]. Interestingly, we observed that TNS1 was markedly decreased in COAD samples; however, OS analysis showed that high expression of TNS1 was correlated with poor survival outcome. Correlation of TNS1 expression with immune infiltration level in COAD A recent study demonstrated that TNS1 is required for fibronectin fibrillogenesis on extracellular vesicle fractions by promoting clustering of extracellular matrix–bound integrins and that its depletion significantly inhibits pulmonary metastasis [36]. 31-5p may be a useful prognostic biomarker for anti- EGFR therapy in CRC because high miR-31-5p expression was associated with shorter progression-free survival [27]. Furthermore, a comprehensive miRNA expression profiling study identified elevated miR-31- 5p expression in BRAF-mutant COAD, which highlights its possible functional role in the serrated pathway [28]. In addition, miR-31-5p is also associated with resistance to chemotherapy, such as oxaliplatin [29] and sorafenib [20]. The present study evaluated miR-31-5p expression level in a pancancer analysis and concluded that miR-31-5p is broadly overexpressed in most neoplasms, indicating that miR-31-5p might be a clinically useful biomarker for human malignancies. Our Kaplan-Meier plot revealed that miR-31-5p contributed to tumor progression based on pathologic stage, suggesting miR- 31-5p might contribute to tumorigenesis in COAD. In addition, a previous study reported that miR-31-5p is upregulated in all four murine COAD stages and one of the most upregulated miRNAs in the earliest stages, suggesting it may be involved in COAD initiation [30]. Based on out study, miR-31-5p could be involved in both the initiation and metastasis of COAD. To elucidate the potential mechanism of miR-31-5p involvement in COAD, a computational target prediction Figure 5. Correlation of TNS1 expression with immune infiltration level in COAD. TNS1 expression is significantly correlated with tumor purity and has strong correlations with macrophages, CD4+ T cells, dendritic cells, and neutrophils. Figure 5. Correlation of TNS1 expression with immune infiltration level in COAD. TNS1 expression is significantly correlated with tumor purity and has strong correlations with macrophages, CD4+ T cells, dendritic cells, and neutrophils. AGING 7485 www.aging-us.com enhancing M2 polarization. Because a high Treg cell ratio in tumors is associated with poor survival [39], the relationship between TNS1 and Treg cells observed in the present study further validates the involvement of TNS1 in COAD tumorigenesis. In fact, previous reports have indicated that miR-31 can mediate immune reaction. However, whether miR-31 interacts with tumor-infiltrating immune cells by targeting TNS1 and whether miR-31 plays different roles in CRC according to microsatellite instability status warrant further investigation. In addition, although miR-31 has been validated as a pivotal marker involved in COAD, a combination of different markers may provide better prognostic prediction [40, 41]. REFERENCES 1. Ferlay J, Colombet M, Soerjomataram I, Mathers C, Parkin DM, Piñeros M, Znaor A, Bray F. Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer. 2019; 144:1941–53. Expression and survival analysis of overlapping genes that can serve to predict prognosis in patients with COAD, especially in those with pathologic stage IV disease. Bioinformatics analyses identified TNS1 as a potential gene targeted by miR-31-5p that may regulate immune cell infiltration and play a vital role in TME and tumorigenesis of COAD. The expression and methylation of overlapping genes were evaluated using the UALCAN database (http:// ualcan.path.uab.edu/index.html) [44]. To evaluate the prognostic value of overlapping genes in COAD, we analyzed the overall survival (OS) rate of those genes in COAD using the GEPIA tool. The thresholds were set as follows: median group cutoff, 95% confidence interval, and P = 0.05 significance level. Then, we further validated the OS of those genes in COAD using UALCAN database and PrognoScan (http://dna00. bio.kyutech.ac.jp/PrognoScan/) [45]. In addition, immunohistochemical results of pivotal targeted genes in normal colon tissue and COAD tissue were obtained from The Human Protein Atlas (THPA, https://www. proteinatlas.org/). Prediction of miR-31-5p targets To identify the targets of miR-31-5p, we searched the miRNAwalk3.0 website (http://mirwalk.umm.uni- heidelberg.de/), and three databases (TargetScan, miRDB, and TargetMiner) were mined. Only those genes predicted by all three databases were recognized as target genes. Then, we collected differentially expressed genes (DEGs) associated with COAD from the Gene Expression Profiling Interactive Analysis (GEPIA) online database (http://gepia. cancer-pku.cn/). Genes that were both miRNA target genes and COAD-related genes were determined using a Venn diagram. Gene correlation analysis with immune infiltration The online TIMER database was used to assess the correlation of gene expression with the level of immune infiltrates [46]. The gene module in TIMER allows users to analyze the gene expression with CD4+ T cells, CD8+ T cells, macrophages, dendritic cells, and neutrophils. Furthermore, the gene expression with the T-cell infiltrates in COAD were confirmed using GEPIA. ACKNOWLEDGMENTS The authors would like to express their gratitude to the Chinse Scholar Counsel for their support. Expression level of miR-31-5p in COAD Gene Expression Display Server (GEDS, http://bioinfo. life.hust.edu.cn/web/GEDS/) was used to explore the expression pattern of miR-31-5p in COAD [42]. GEDS is a platform that collects 40 tissues and 1,594 cells lines from The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), Cancer Cell Line Encyclopedia (CCLE), and MD Anderson Cell Lines Project (MCLP), providing information on human gene, miRNA, and protein expression in tissues and cell lines. In addition, we collected microarray data from the GEO database (GSE30454, GSE41655, GSE18392, and GSE108153) to compare the expression of miR-31-5p in COAD and normal tissue. Analysis of survival and miR-31-5p in COAD The correlation between miR-31-5p and survival in COAD was analyzed using the LinkedOmics database (http://www.linkedomics.org/) [43]. The LinkedOmics database collected multiomics data and clinical data of 32 cancer types from The Cancer Genome Atlas (TCGA) project. The thresholds were set according to the following steps: Step 1: TCGA_Colorectal adenocarcinoma (COADREAD); Step 2: miRNASeq, HS miR platform; Step 2b: histological_type colona- denocarcinoma [N:391]; Step 3: miR-31-5p; Step 4: TCGA_COADREAD, Clinical data type, clinical platform; and Step 5: Non-Parametric Test (Attribute Dependent). FUNDING This study was supported by the National Science Foundation of China (grant no. 81772345). CONFLICTS OF INTEREST The authors declare no conflicts of interest. Correlation of TNS1 expression with immune infiltration level in COAD Phospho-TNS1 was highly elevated in EMT cells after TGFβ treatment and was specifically observed in tissue samples of patients with poor-prognosis lung adenocarcinoma [37]. Based on this evidence, TNS1 negatively impacts COAD tumorigenesis and may accelerate metastasis by regulating epithelial to mesenchymal transition (EMT). Phospho-TNS1 was highly elevated in EMT cells after TGFβ treatment and was specifically observed in tissue samples of patients with poor-prognosis lung adenocarcinoma [37]. Based on this evidence, TNS1 negatively impacts COAD tumorigenesis and may accelerate metastasis by regulating epithelial to mesenchymal transition (EMT). Immune infiltrate correlation analysis indicates that TNS1 is strongly correlated with macrophages, which are the most abundant hematopoietic cells in the COAD tumor microenvironment (TME) [38]. We speculate that TNS1 may play an important role in COAD TME. It is believed that M1 and M2 macrophages are active in tumor prevention and tumor promotion, respectively. Our results indicate that TNS1 is more closely associated with M2 macrophages, suggesting TNS1 might negatively affect tumorigenesis of COAD by In summary, the current study confirmed the overexpression of miR-31-5p in COAD. More importantly, miR-31-5p may be a latent tumor biomarker Figure 6. Correlation of TNS1 expression with immune infiltration level in COAD. TNS1 is closely related with abundance of T cells and has a stronger relationship with macrophage M2 cells than M1 cells. Figure 6. Correlation of TNS1 expression with immune infiltration level in COAD. TNS1 is closely related with abundance of T cells and has a stronger relationship with macrophage M2 cells than M1 cells. AGING 7486 www.aging-us.com Expression and survival analysis of overlapping genes https://doi.org/10.1038/s41419-019-1811-y 13. Creighton CJ, Fountain MD, Yu Z, Nagaraja AK, Zhu H, Khan M, Olokpa E, Zariff A, Gunaratne PH, Matzuk MM, Anderson ML. Molecular profiling uncovers a p53-associated role for microRNA-31 in inhibiting the proliferation of serous ovarian carcinomas and other cancers. Cancer Res. 2010; 70:1906–15. https://doi.org/10.1158/0008-5472.CAN-09-3875 PMID:20179198 5. Hobert O. Gene regulation by transcription factors and microRNAs. Science. 2008; 319:1785–86. https://doi.org/10.1126/science.1151651 PMID:18369135 https://doi.org/10.1053/j.gastro.2019.02.023 PMID:30779922 17. Lv C, Li F, Li X, Tian Y, Zhang Y, Sheng X, Song Y, Meng Q, Yuan S, Luan L, Andl T, Feng X, Jiao B, et al. MiR-31 promotes mammary stem cell expansion and breast tumorigenesis by suppressing Wnt signaling antagonists. Nat Commun. 2017; 8:1036. https://doi.org/10.1038/s41467-017-01059-5 PMID:29051494 PMID:20179198 6. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005; 120:15–20. https://doi.org/10.1016/j.cell.2004.12.035 PMID:15652477 14. Valastyan S, Reinhardt F, Benaich N, Calogrias D, Szász AM, Wang ZC, Brock JE, Richardson AL, Weinberg RA. A pleiotropically acting microRNA, miR-31, inhibits breast cancer metastasis. Cell. 2009; 137:1032–46. https://doi.org/10.1016/j.cell.2009.03.047 PMID:19524507 14. Valastyan S, Reinhardt F, Benaich N, Calogrias D, Szász AM, Wang ZC, Brock JE, Richardson AL, Weinberg RA. A pleiotropically acting microRNA, miR-31, inhibits breast cancer metastasis. Cell. 2009; 137:1032–46. https://doi.org/10.1016/j.cell.2009.03.047 PMID:19524507 7. Ling H, Fabbri M, Calin GA. MicroRNAs and other non- coding RNAs as targets for anticancer drug development. Nat Rev Drug Discov. 2013; 12:847–65. https://doi.org/10.1038/nrd4140 PMID:24172333 15. Song F, Xuan Z, Yang X, Ye X, Pan Z, Fang Q. Identification of key microRNAs and hub genes in non- small-cell lung cancer using integrative bioinformatics and functional analyses. J Cell Biochem. 2020; 121:2690–703. incidence of colorectal cancer in Asia: implications for screening. Lancet Oncol. 2005; 6:871–76. https://doi.org/10.1016/S1470-2045(05)70422-8 PMID:16257795 3. Barresi V, Reggiani Bonetti L, Ieni A, Caruso RA, Tuccari G. Histological grading in colorectal cancer: new insights and perspectives. Histol Histopathol. 2015; 30:1059–67. 12. Jia Y, Duan Y, Liu T, Wang X, Lv W, Wang M, Wang J, Liu L. LncRNA TTN-AS1 promotes migration, invasion, and epithelial mesenchymal transition of lung adenocarcinoma via sponging miR-142-5p to regulate CDK5. Cell Death Dis. 2019; 10:573. https://doi.org/10.1002/ijc.31937 PMID:30350310 2. Sung JJ, Lau JY, Goh KL, Leung WK, and Asia Pacific Working Group on Colorectal Cancer. Increasing AGING 7487 7487 www.aging-us.com Neurotensin signaling activates microRNAs-21 and - 155 and Akt, promotes tumor growth in mice, and is increased in human colon tumors. Gastroenterology. 2011; 141:1749–61.e1. https://doi.org/10.1053/j.gastro.2011.07.038 PMID:21806946 incidence of colorectal cancer in Asia: implications for screening. Lancet Oncol. 2005; 6:871–76. https://doi.org/10.1016/S1470-2045(05)70422-8 PMID:16257795 8. Di Leva G, Briskin D, Croce CM. MicroRNA in canc new hopes for antineoplastic chemotherapy. Up Med Sci. 2012; 117:202–16. https://doi.org/10.3109/03009734.2012.660551 PMID:22348396 8. Di Leva G, Briskin D, Croce CM. MicroRNA in cancer: new hopes for antineoplastic chemotherapy. Ups J Med Sci. 2012; 117:202–16. https://doi.org/10.3109/03009734.2012.660551 PMID:22348396 https://doi.org/10.1002/jcb.29489 PMID:31692035 https://doi.org/10.1002/jcb.29489 PMID:31692035 16. Tian Y, Xu J, Li Y, Zhao R, Du S, Lv C, Wu W, Liu R, Sheng X, Song Y, Bi X, Li G, Li M, et al. MicroRNA-31 reduces inflammatory signaling and promotes regeneration in colon epithelium, and delivery of mimics in microspheres reduces colitis in mice. Gastroenterology. 2019; 156:2281–2296.e6. https://doi.org/10.1053/j.gastro.2019.02.023 PMID:30779922 9. Emami SS, Akbari A, Zare AA, Agah S, Masoodi M, Talebi A, Minaeian S, Fattahi A, Moghadamnia F. MicroRNA expression levels and histopathological features of colorectal cancer. J Gastrointest Cancer. 2019; 50:276–84. https://doi.org/10.14670/HH-11-633 PMID:26004398 4. Agarwal E, Robb CM, Smith LM, Brattain MG, Wang J, Black JD, Chowdhury S. Role of Akt2 in regulation of metastasis suppressor 1 expression and colorectal cancer metastasis. Oncogene. 2017; 36:3104–18. https://doi.org/10.1038/onc.2016.460 PMID:28068324 https://doi.org/10.1007/s12029-018-0055-x 10. Sheedy FJ, Palsson-McDermott E, Hennessy EJ, Martin C, O’Leary JJ, Ruan Q, Johnson DS, Chen Y, O’Neill LA. Negative regulation of TLR4 via targeting of the proinflammatory tumor suppressor PDCD4 by the microRNA miR-21. Nat Immunol. 2010; 11:141–47. https://doi.org/10.1038/ni.1828 PMID:19946272 18. Wang S, Hu J, Zhang D, Li J, Fei Q, Sun Y. Prognostic role of microRNA-31 in various cancers: a meta-analysis. Tumour Biol. 2014; 35:11639–45. https://doi.org/10.1007/s13277-014-2492-x PMID:25139099 18. Wang S, Hu J, Zhang D, Li J, Fei Q, Sun Y. Prognostic role of microRNA-31 in various cancers: a meta-analysis. Tumour Biol. 2014; 35:11639–45. https://doi.org/10.1007/s13277-014-2492-x PMID:25139099 11. Bakirtzi K, Hatziapostolou M, Karagiannides I, Polytarchou C, Jaeger S, Iliopoulos D, Pothoulakis C. AGING 7488 www.aging-us.com 27. Igarashi H, Kurihara H, Mitsuhashi K, Ito M, Okuda H, Kanno S, Naito T, Yoshii S, Takahashi H, Kusumi T, Hasegawa T, Sukawa Y, Adachi Y, et al. Association of MicroRNA-31-5p with clinical efficacy of anti-EGFR therapy in patients with metastatic colorectal cancer. Ann Surg Oncol. 2015; 22:2640–48. https://doi.org/10.1245/s10434-014-4264-7 PMID:25472647 19. Chou J, Werb Z. MicroRNAs play a big role in regulating ovarian cancer-associated fibroblasts and the tumor microenvironment. Cancer Discov. 2012; 2:1078–80. https://doi.org/10.1158/2159-8290.CD-12-0465 PMID:23230184 19. Chou J, Werb Z. MicroRNAs play a big role in regulating ovarian cancer-associated fibroblasts and the tumor microenvironment. Cancer Discov. 2012; 2:1078–80. 20. He J, He J, Min L, He Y, Guan H, Wang J, Peng X. Extracellular vesicles transmitted miR-31-5p promotes sorafenib resistance by targeting MLH1 in renal cell carcinoma. Int J Cancer. 2020; 146:1052–63. https://doi.org/10.1002/ijc.32543 PMID:31259424 20. He J, He J, Min L, He Y, Guan H, Wang J, Peng X. Extracellular vesicles transmitted miR-31-5p promotes sorafenib resistance by targeting MLH1 in renal cell carcinoma. Int J Cancer. 2020; 146:1052–63. 28. Ito M, Mitsuhashi K, Igarashi H, Nosho K, Naito T, Yoshii S, Takahashi H, Fujita M, Sukawa Y, Yamamoto E, Takahashi T, Adachi Y, Nojima M, et al. MicroRNA-31 expression in relation to BRAF mutation, CpG island methylation and colorectal continuum in serrated lesions. Int J Cancer. 2014; 135:2507–15. https://doi.org/10.1002/ijc.28920 PMID:24752710 21. Mitamura T, Watari H, Wang L, Kanno H, Kitagawa M, Hassan MK, Kimura T, Tanino M, Nishihara H, Tanaka S, Sakuragi N. microRNA 31 functions as an endometrial cancer oncogene by suppressing Hippo tumor suppressor pathway. Mol Cancer. 2014; 13:97. 29. Hsu HH, Kuo WW, Shih HN, Cheng SF, Yang CK, Chen MC, Tu CC, Viswanadha VP, Liao PH, Huang CY. https://doi.org/10.1007/s12029-018-0055-x FOXC1 regulation of miR-31-5p confers oxaliplatin resistance by targeting LATS2 in colorectal cancer. Cancers (Basel). 2019; 11:E1576. https://doi.org/10.3390/cancers11101576 PMID:31623173 https://doi.org/10.1002/ijc.32483 PMID:31144303 23. Moffett HF, Cartwright AN, Kim HJ, Godec J, Pyrdol J, Äijö T, Martinez GJ, Rao A, Lu J, Golub TR, Cantor H, Sharpe AH, Novina CD, Wucherpfennig KW. The microRNA miR-31 inhibits CD8+ T cell function in chronic viral infection. Nat Immunol. 2017; 18:791–99. https://doi.org/10.1038/ni.3755 PMID:28530712 31. Zhang WW, Ming XL, Rong Y, Huang CQ, Weng H, Chen H, Bian JM, Wang FB. Diagnostic value investigation and bioinformatics analysis of miR-31 in patients with lymph node metastasis of colorectal cancer. Anal Cell Pathol (Amst). 2019; 2019:9740475. https://doi.org/10.1155/2019/9740475 PMID:31934534 24. Yamagishi M, Nakano K, Miyake A, Yamochi T, Kagami Y, Tsutsumi A, Matsuda Y, Sato-Otsubo A, Muto S, Utsunomiya A, Yamaguchi K, Uchimaru K, Ogawa S, Watanabe T. Polycomb-mediated loss of miR-31 activates NIK-dependent NF-κB pathway in adult T cell leukemia and other cancers. Cancer Cell. 2012; 21:121–35. 32. Bernau K, Torr EE, Evans MD, Aoki JK, Ngam CR, Sandbo N. Tensin 1 is essential for myofibroblast differentiation and extracellular matrix formation. Am J Respir Cell Mol Biol. 2017; 56:465–76. https://doi.org/10.1165/rcmb.2016-0104OC PMID:28005397 https://doi.org/10.1186/1476-4598-13-97 PMID:24779718 22. Ye J, Zhu J, Chen H, Qian J, Zhang L, Wan Z, Chen F, Sun S, Li W, Luo C. A novel lncRNA-LINC01116 regulates tumorigenesis of glioma by targeting VEGFA. Int J Cancer. 2020; 146:248–61. https://doi.org/10.1002/ijc.32483 PMID:31144303 22. Ye J, Zhu J, Chen H, Qian J, Zhang L, Wan Z, Chen F, Sun S, Li W, Luo C. A novel lncRNA-LINC01116 regulates tumorigenesis of glioma by targeting VEGFA. Int J Cancer. 2020; 146:248–61. 30. De Robertis M, Mazza T, Fusilli C, Loiacono L, Poeta ML, Sanchez M, Massi E, Lamorte G, Diodoro MG, Pescarmona E, Signori E, Pesole G, Vescovi AL, et al. EphB2 stem-related and EphA2 progression-related miRNA-based networks in progressive stages of CRC evolution: clinical significance and potential miRNA drivers. Mol Cancer. 2018; 17:169. https://doi.org/10.1186/s12943-018-0912-z PMID:30501625 https://doi.org/10.1016/j.ccr.2011.12.015 PMID:22264793 25. Yang MH, Yu J, Chen N, Wang XY, Liu XY, Wang S, Ding YQ. Elevated microRNA-31 expression regulates colorectal cancer progression by repressing its target gene SATB2. PLoS One. 2013; 8:e85353. https://doi.org/10.1371/journal.pone.0085353 PMID:24386467 33. Zhan Y, Liang X, Li L, Wang B, Ding F, Li Y, Wang X, Zhan Q, Liu Z. MicroRNA-548j functions as a metastasis promoter in human breast cancer by targeting Tensin1. Mol Oncol. 2016; 10:838–49. https://doi.org/10.1016/j.molonc.2016.02.002 PMID:26949125 26. Eide PW, Eilertsen IA, Sveen A, Lothe RA. Long noncoding RNA MIR31HG is a bona fide prognostic marker with colorectal cancer cell-intrinsic properties. Int J Cancer. 2019; 144:2843–53. https://doi.org/10.1002/ijc.31998 PMID:30447009 34. Zhou H, Zhang Y, Wu L, Xie W, Li L, Yuan Y, Chen Y, Lin Y, He X. Elevated transgelin/TNS1 expression is a AGING 7489 www.aging-us.com potential biomarker in human colorectal cancer. Oncotarget. 2017; 9:1107–13. https://doi.org/10.18632/oncotarget.23275 PMID:29416680 40. Talesa VN, Antognelli C, Del Buono C, Stracci F, Serva MR, Cottini E, Mearini E. Diagnostic potential in prostate cancer of a panel of urinary molecular tumor markers. Cancer Biomark. 2009; 5:241–51. https://doi.org/10.3233/CBM-2009-0109 PMID:20037200 35. Zhou HM, Fang YY, Weinberger PM, Ding LL, Cowell JK, Hudson FZ, Ren M, Lee JR, Chen QK, Su H, Dynan WS, Lin Y. Transgelin increases metastatic potential of colorectal cancer cells in vivo and alters expression of genes involved in cell motility. BMC Cancer. 2016; 16:55. 41. Antognelli C, Mezzasoma L, Mearini E, Talesa VN. Glyoxalase 1-419C>A variant is associated with oxidative stress: implications in prostate cancer progression. PLoS One. 2013; 8:e74014. https://doi.org/10.1371/journal.pone.0074014 PMID:24040147 41. Antognelli C, Mezzasoma L, Mearini E, Talesa VN. Glyoxalase 1-419C>A variant is associated with oxidative stress: implications in prostate cancer progression. PLoS One. 2013; 8:e74014. h //d / / l https://doi.org/10.1186/s12885-016-2105-8 PMID:26847345 36. Shinde A, Paez JS, Libring S, Hopkins K, Solorio L, Wendt MK. Transglutaminase-2 facilitates extracellular vesicle-mediated establishment of the metastatic niche. Oncogenesis. 2020; 9:16. https://doi.org/10.1038/s41389-020-0204-5 PMID:32054828 42. Xia M, Liu CJ, Zhang Q, Guo AY. GEDS: a gene expression display server for mRNAs, miRNAs and proteins. Cells. 2019; 8:E675. https://doi.org/10.3390/cells8070675 PMID:31277321 36. Shinde A, Paez JS, Libring S, Hopkins K, Solorio L, Wendt MK. Transglutaminase-2 facilitates extracellular vesicle-mediated establishment of the metastatic niche. Oncogenesis. 2020; 9:16. 43. Vasaikar SV, Straub P, Wang J, Zhang B. LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res. 2018; 46:D956–63. https://doi.org/10.1093/nar/gkx1090 PMID:29136207 37. Okayama A, Miyagi Y, Oshita F, Ito H, Nakayama H, Nishi M, Kurata Y, Kimura Y, Ryo A, Hirano H. Identification of tyrosine-phosphorylated proteins upregulated during epithelial-mesenchymal transition induced with TGF-beta. J Proteome Res. 2015; 14:4127–36. 44. Chandrashekar DS, Bashel B, Balasubramanya SA, Creighton CJ, Ponce-Rodriguez I, Chakravarthi BV, Varambally S. UALCAN: a portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia. 2017; 19:649–58. https://doi.org/10.1016/j.neo.2017.05.002 PMID:28732212 https://doi.org/10.1021/acs.jproteome.5b00082 PMID:26216473 38. Bindea G, Mlecnik B, Tosolini M, Kirilovsky A, Waldner M, Obenauf AC, Angell H, Fredriksen T, Lafontaine L, Berger A, Bruneval P, Fridman WH, Becker C, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity. 2013; 39:782–95. https://doi.org/10.1016/j.immuni.2013.10.003 PMID:24138885 45. Mizuno H, Kitada K, Nakai K, Sarai A. PrognoScan: a new database for meta-analysis of the prognostic value of genes. BMC Med Genomics. 2009; 2:18. https://doi.org/10.1186/1755-8794-2-18 PMID:19393097 39. Timperi E, Pacella I, Schinzari V, Focaccetti C, Sacco L, Farelli F, Caronna R, Del Bene G, Longo F, Ciardi A, Morelli S, Vestri AR, Chirletti P, et al. Regulatory T cells with multiple suppressive and potentially pro-tumor activities accumulate in human colorectal cancer. OncoImmunology. 2016; 5:e1175800. https://doi.org/10.1080/2162402X.2016.1175800 PMID:27622025 46. Li B, Severson E, Pignon JC, Zhao H, Li T, Novak J, Jiang P, Shen H, Aster JC, Rodig S, Signoretti S, Liu JS, Liu XS. Comprehensive analyses of tumor immunity: implications for cancer immunotherapy. Genome Biol. 2016; 17:174. https://doi.org/10.1186/s13059-016-1028-7 PMID:27549193 AGING 7490 www.aging-us.com
W2794287851.txt	https://bmcbioinformatics.biomedcentral.com/track/pdf/10.1186/s12859-019-2664-1	en		Predicting clinically promising therapeutic hypotheses using tensor factorization	bioRxiv (Cold Spring Harbor Laboratory)	2,018	cc-by	8,321	Yao et al. BMC Bioinformatics (2019) 20:69 https://doi.org/10.1186/s12859-019-2664-1 RESEARCH ARTICLE Open Access Predicting clinically promising therapeutic hypotheses using tensor factorization Jin Yao1* , Mark R. Hurle1, Matthew R. Nelson2 and Pankaj Agarwal1 Abstract Background: Determining which target to pursue is a challenging and error-prone first step in developing a therapeutic treatment for a disease, where missteps are potentially very costly given the long-time frames and high expenses of drug development. With current informatics technology and machine learning algorithms, it is now possible to computationally discover therapeutic hypotheses by predicting clinically promising drug targets based on the evidence associating drug targets with disease indications. We have collected this evidence from Open Targets and additional databases that covers 17 sources of evidence for target-indication association and represented the data as a tensor of 21,437 × 2211 × 17. Results: As a proof-of-concept, we identified examples of successes and failures of target-indication pairs in clinical trials across 875 targets and 574 disease indications to build a gold-standard data set of 6140 known clinical outcomes. We designed and executed three benchmarking strategies to examine the performance of multiple machine learning models: Logistic Regression, LASSO, Random Forest, Tensor Factorization and Gradient Boosting Machine. With 10-fold cross-validation, tensor factorization achieved AUROC = 0.82 ± 0.02 and AUPRC = 0.71 ± 0.03. Across multiple validation schemes, this was comparable or better than other methods. Conclusion: In this work, we benchmarked a machine learning technique called tensor factorization for the problem of predicting clinical outcomes of therapeutic hypotheses. Results have shown that this method can achieve equal or better prediction performance compared with a variety of baseline models. We demonstrate one application of the method to predict outcomes of trials on novel indications of approved drug targets. This work can be expanded to targets and indications that have never been clinically tested and proposing novel target-indication hypotheses. Our proposed biologically-motivated cross-validation schemes provide insight into the robustness of the prediction performance. This has significant implications for all future methods that try to address this seminal problem in drug discovery. Keywords: Drug target discovery, Clinical trial outcomes, Tensor factorization Background Drug discovery and development often begin with a drug target, through which the drug exerts its therapeutic effect in patients with a certain disease or clinical condition (termed as an indication). A target is a broad term which includes many biological entities such as proteins, genes, and RNA, whose modulation (increase or decrease in activity) can provide a therapeutic benefit to a patient. Although selecting an efficacious drug target is the first and most important step in drug development, * Correspondence: jin.8.yao@gsk.com 1 Computational Biology, GSK R&D, 1250 S. Collegeville Road, UP12-200, Collegeville, PA, USA Full list of author information is available at the end of the article more than half of clinical trials still fail due to lack of efficacy, i.e., modulating the target’s activity did not provide a statistically significant benefit to patients [1, 2]. Target selection is critical in drug discovery given the long-time frame and high expense of drug development. Often drug targets come from research publication where evidence is generated to support a hypothesis that inhibition or activation of a target may result in a therapeutic effect for a specific disease indication. For example, amyloid precursor protein is a target suggested for Alzheimer’s Disease (AD). A piece of important evidence to support this hypothesis is that familial AD patients commonly have genetic mutations in the corresponding gene which lead to the production and © 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. Yao et al. BMC Bioinformatics (2019) 20:69 deposition in the brain of increased amounts of amyloid beta peptide, a major characteristic of AD [3]. With current informatics technology, it is now possible to construct online repositories that aggregate existing knowledge about the association evidence linking potential targets with disease indications. Open Targets [4] is such a platform that provides drug discovery researchers with multiple evidence types including genetic association, pathways, animal models and drugs, that connect targets with indications for validating potential therapeutic hypotheses. At the same time, these online knowledge repositories are also amenable to computational analysis to discover drug target hypotheses using machine learning. One major challenge in framing this problem from a machine learning perspective is that there are very few positive examples (0.005% of target-indication hypotheses included in Open Targets have approved drugs). However, any insights gleaned from the limited number of pursued targets may be useful in delivering new medicines with lower attrition rates. In this paper, we collated historical outcomes of clinical trials and determined if these clinical outcomes can be predicted retrospectively using multiple machine learning models built on existing evidence of the targets’ biological association with indications. A successful prediction model provides an understanding of how informative the evidence is for clinical success, and is also capable of generating new target-indication hypotheses with a higher potential of being developed into successful medicines. Another difficulty in building such a model is that not all biological evidence is available for every pair of target and indication due to reasons such as technological limitation and limited disease coverage. For example, as of June 2017, Open Targets contained 26,122 targets, 9150 diseases with 2,857,732 positive associations from 15 a Page 2 of 12 evidence sources. Though Open Targets contains over 2.8 million associations, that is still only 0.08% of the possible combinations covered by this data, suggesting that a great deal of association evidence (99.92%) is still to be determined by biomedical researchers and clinicians. Traditional paradigms of machine learning algorithms, learning a mapping from input features (biological evidence) to output prediction (clinical outcomes), may be inadequate in this context. We explored if tensor factorization is useful in the analysis of this sparse biological dataset. Tensor extends the concept of a matrix to a multidimensional array where each dimension corresponds to one “axis”, called mode, of a tensor [5]. Data in many applications can be naturally organized into a tensor format. Figure 1a shows a three-mode tensor representing different types of evidence associating targets with disease indications and one extra “slice” represents clinical outcomes. Tensor factorization decomposes a tensor into factor matrices that compactly store information encoded in a tensor and integrate interaction across different modes even when a large portion of entries of a tensor is missing [5]. This technique has a wide range of applications such as in recommendation systems [6], knowledge graph systems [7] and multiple biomedical domains [8]. There are several lines of previous work that are related to this paper. One line of work is focused on answering the question of what makes a good drug target by investigating features of targets that are correlated with clinical successes in the context of genetics [9], tissue mRNA expression [10], human protein interactome [11] and publication trends [12]. Another line of work is focused on disease gene prediction, where the goal is to predict genes mechanistically involved in a given disease [13–17]. Our work is different from these efforts in that b Fig. 1 Data representation and model benchmark schematic. a Tensor representation of the dataset. The last “slice” matrix represents the clinical outcomes of target-indication pairs. b Illustration of three schemes of benchmarking models on predicting clinical outcomes. Each matrix represents the clinical outcomes of targets (rows) and indications (columns). Grey and green cells are target-indications pairs used for training and testing, respectively. Blank cells represent unknown clinical outcomes of target-indication pairs Yao et al. BMC Bioinformatics (2019) 20:69 we leverage a novel computational method to integrate multiple evidence types and directly assess the models’ performance of predicting clinical outcomes of drug target hypotheses. In the following sections, we first describe the dataset we have collected and then introduce the basic formulation of matrix factorization, a special form of tensor factorization. Then we explain our selection of a specific algorithm of tensor factorization based on characteristics of our data. We then discuss how we design experiments to benchmark the method against a series of baseline models under three scenarios of drug discovery. We demonstrate that the model can capture known biological mechanisms of human diseases and can identify opportunities of approved drug targets to novel indications. Methods Data collection and processing We created a dataset which combined clinical outcomes from the commercial database Pharmaprojects [18] with evidence from Open Targets [4] and other sources (Tables 1, 2). In total, we collected 17 association evidence sources that connect potential targets with disease indications. These 17 association evidence sources can be further grouped into seven evidence types: Genetics (is germline mutation in the target associated with the disease?), somatic mutations (is somatic mutation in the target associated with the disease, typically cancer?), pathways (is the target part of a pathway involved in the disease?), mRNA disease expression (does the target’s expression significantly change in the disease?), mRNA tissue overexpression (is the target’s expression overexpressed in disease-related tissues?), literature (is there association between the target and the indication identified through text mining of scientific literature?) and animal models (does the knockout of the target in Table 1 17 sources of target-indication evidence Evidence Type Sources Animal models Phenodigm Genetics European Variant Archive, Uniprot, Uniprot literature, GWAS catalog, STOPGAP [9] Somatic mutation Cancer Gene Census, European Variant Archive somatic Literature Europe PMC, TERMITE mRNA disease expression Expression Atlas, Internal expression data mRNA tissue overexpression GeneLogic, GTEx [49], Human Protein Atlas [50] Pathways Reactome, Metabase Evidence data were obtained from Open Targets [1] except for TERMITE: www.scibite.com/products/termite; GeneLogic: GeneLogic Division, Ocimum Biosolutions, Inc., Internal expression data, and those explicitly referenced Page 3 of 12 Table 2 Six sources of target-only categorical attributes Attribute Type Sources (# of categories) Mutation Tolerance ExAc_LoF(3), ExAc_Missense (3), RVIS (3), Mouse Protein identity (2) Other Target Characteristics Target Location (7), Target Topology (5) Genes were broken into non-overlapping categories based on available data. Genes were classified as tolerant, intolerant and unclassified based on data from the Exome Aggregation Consortium [51] and the percentile rank of Residual Variation Intolerance Score [52]. Genes were based on the identification of > = 75% protein homology between human and mouse, data downloaded from BioMart [53]. Target Location and Topology were derived from a review of information from Gene Ontology, InterPro, PFAM, and UniProt animal models manifest phenotypes that are concordant with the human disease?). Besides these 17 association evidence sources, we also collected information about properties of targets from six sources (Table 2) as previous studies have found that successful FDA-approved drugs are enriched with targets with properties that are independent of disease indications [19, 20]. The collected evidence covers the data space of 21,437 targets, 2211 indications, and 17 evidence sources. For clinical outcome data, if at least one drug asset for a given target-indication pair was identified as successful, then the target-indication pair was classified as Succeeded. Of the remaining target-indication pairs, if at least one asset had a clinical failure then it was classified as a Clinical Failure. Open Targets presents evidence from each individual source as a numerical value for a target-indication pair, with a positive value representing the strength of evidence. To simplify the further collation of target-indication evidence with target-only attributes (Table 2), we converted numerical evidence value into binary values: 1 indicates a positive association, 0 means that there is no association and unknown evidence is represented as null. We encoded categorical data, typically present in target-only attributes, as multiple binary values with each category converted into a binary value, i.e., having the property or not having the property. Here, we analyzed data mapped to the 574 non-cancer indications (a subset of 2211 indications) with at least one clinical outcome and the corresponding 875 targets (a subset of 21,437 targets). Oncology indications were excluded, as studies have observed that features of successful targets for cancer differ from features of successful targets for other indications [21, 22], moreover, cancer trials fail more frequently than trials for other indications. [23] Matrix factorization The clinical outcomes of existing target-indication pairs can be represented in a matrix format as R ∈ ℝM × N, where the M rows represent targets and the N columns Yao et al. BMC Bioinformatics (2019) 20:69 represent indications. Rij = 1 if there is at least one drug that modulates target i and is marketed for indication j. Rij = 0 if all the drugs modulating target i are reported failed for indication j in the clinic (from Phase I to Phase III). For target-indication pairs that have no outcomes in the clinic, the corresponding Rij is empty. The goal is to predict clinical outcomes for all possible pairs of targets and indications i.e. fill out the empty Rij′s. Thus, we can treat the problem as completing the target-indication matrix of clinical outcomes. Matrix completion problem has been widely studied in the machine learning community in the context of recommendation systems [6, 24]. A famous application is Netflix’s movie recommendation system, where each user has ratings on a small number of movies and the task is to recommend movies for each user based on existing ratings of other users with similar patterns of movie ratings. Matrix factorization is recognized as one of the more successful methods for this task [6, 25, 26]. The method assumes that the true completed matrix is of low rank and can be approximated by a product of two low-dimensional latent factor matrices that represent rows and columns of a matrix in a joint D-dimensional latent space, i.e. R ≈ DM UT V, where U ¼ fui gM , V ¼ fv j gNj¼1 ∈RDN and i¼1 ∈R D D ui ∈ ℝ , vj ∈ ℝ are column vectors of U and V, respectively. The predicted entries in Rij is achieved by the inner product of ui and vj. Learning of U and V can be formulated as an optimization problem by minimizing the mean squared error between observed and predicted entries. To avoid overfitting, regularization on the latent factor matrices is added to the minimization problem that can be solved by methods such as stochastic gradient descent and alternating least square [6]. Bayesian tensor factorization Many matrix-factorization based methods have been proposed for recommendation systems. To choose an appropriate method to predict clinical outcomes, we considered three aspects of our problem. First, some of the evidence is target-indication specific such as human genetic evidence for each disease, and this has been suggested as related to clinical outcome [9]. Second, in our data, there are several target-only attributes independent of indications, such as target protein location, tolerance of mutation. Thus, the chosen method should also take target-only information into consideration. Third, in drug discovery, it is not uncommon that targets or indications that have never been tested in clinical trials. In the case of movie recommendation systems, this corresponds to recommending movies to users who have not rated any movies in the system or recommending new movies that do not have any ratings in the system. The chosen method should be able to handle this situation. Page 4 of 12 Given these three aspects, we investigated a method based on tensor factorization, called Macau, that is capable of naturally handling all the three aspects in a unified Bayesian framework and was originally used to predict drug-protein interaction [27]. Tensor extends the matrix concept to a multidimensional array, where each dimension corresponds to one mode of a tensor. Our data can be organized into a three-mode tensor: target × indication × evidence T ∈RMNK ;where one entry tijk indicates the association score in kth evidence between target i and indication j and one “slice” of the tensor corresponding to one evidence source organized as a matrix. M, N, K are the number of targets, indications and evidence sources, respectively. To predict clinical outcomes, we appended the clinical outcome matrix R as one extra “slice” to the evidence tensor (Fig. 1a) and factorized the resulting tensor X ∈RMNðK þ1Þ . Similar to matrix factorization, tensor factorization decomposes a tensor into a series of low-dimensional latent factor matrices where each matrix represents one mode of the tensor. One direct way to decompose a three-mode tensor is to assume that each entry xijk can be expressed as the sum of the elementwise product of three low-dimensional vectors: ui, vj.and ek , representing ith target, jth indication and kth evidence (including the clinical outcomes), respectively in a joint latent factor space, P i.e. xijk ≈ D d¼1 udi vdj edk , where D is the dimensionality of the latent factors. The latent factors can be further orgaDM nized in three factor matrices: U ¼ fui gM , V i¼1 ∈ℝ D þ1 ¼ fv j gNj¼1 ∈ℝ DN , E ¼ fek gKk¼1 ∈ℝ DðK þ1Þ and ui ∈ ℝ , D D vj ∈ ℝ ,ek ∈ ℝ are column vectors of U, V and E, respectively. Here the eK + 1 column of E corresponds with the latent factor of the clinical outcome. The prediction of any entry xijk of the tensor can be achieved by the sum of the elementwise product of the low-dimensional vectors of target ui, indication vjand evidence ek. Since the factorized tensor included the clinical outcome matrix, we can use the low-dimensional vector corresponding to the clinical outcome to perform prediction, i.e. the predicted outcome of Tmodulating target i for the treatment of indication j is 1 (ui ∘ vj ∘ eK + 1), where 1 is an all one vector and ∘ is the elementwise product. The specified tensor factorization method we chose is based on Bayesian probabilistic modeling, which assumes each observed cell of the tensor X is a random variable following a normal distribution, xijk N ð1T ðui ∘v j ∘ek Þ; α−1 Þ, where α is the precison of the normal distribution. In this model, the mean of the normal distribution is determined by the three low-dimensional latent factors: ui, vj.and ek. Each latent factor is assumed to have a Gaussian prior with a Gaussian-Wishart hyper prior placed on its hyperparameters: ui N ðμtarget þ ΒT Yao et al. BMC Bioinformatics (2019) 20:69 −1 g i ; Λ−1 target Þ, v j N ðμindication ; Λtarget Þ and ek N ðμevidence G×D ; Λ−1 where Β ∈ ℝ linearly projects the target Þ , G i target-only attributes g ∈ ℝ into the latent space, providing prediction ability to targets that do not have any observed clinical outcomes. The inference of model parameters is carried out by sampling from the posterior of the model parameters by Markov Chain Monte Carlo (MCMC) technique, except for α, which is set to 1 by default and the number of latent factors D, which is determined by a heuristic approach (see Additional file 1). Specifically, we used the Julia implementation of the method [28] and followed a common practice of MCMC inference where we “burn-in” samples generated in the beginning and collect samples after that to approximate posterior distribution over model parameters [29]. In our case, the first 500 samples were discarded and the posterior distribution over parameters were estimated using 300 samples after the “burn-in” process. The predictive distribution is approximated from the 300 samples of the model parameters and the average over samples is used to make predictions. Generally, we did not observe further improvement on prediction performance if we let the chain run longer. Model benchmark experiments We performed a nested cross-validation experiment to evaluate the method in three different scenarios (Fig. 1b). In each experiment, we divided the target-indication pairs with clinical outcomes (6140) into K folds and tested the prediction results on a held-out (one of the K) fold using a model trained with the rest (K-1) of folds. This is the outer loop for the cross-validation. In the inner loop, we determine the model parameters using five-fold cross-validation. In the first experiment, we did a standard ten-fold cross-validation in the outer loop, where each fold is randomly determined but retains the same fraction of successes. In drug discovery, we know that certain sub-classes of targets and indications have different properties. In order to, assess if the model can be generalized to sub-classes of targets and indications different from those used in the training stage, we devised two other cross-validation experiments where each time the clinical outcomes of one pre-defined group of targets (indications) are left out as the test set. Specifically, for the second experiment, i.e. leave-one-target-group-out, we used the grouping defined by the Target Class (See Table S1 in Additional file 1). A given target is assigned to one of ten target classes (thus K = 10) based on the target’s protein family retrieved from ChEMBL hierarchical target classification system [30]. For the third experiment, Page 5 of 12 i.e. leave-one-indication-group-out, we defined eight indication groups (thus K = 8) by de novo clustering indications based on the similarity of the indications in terms of their relative positions in MeSH (Medical Subject Heading) hierarchical tree and co-occurrence frequency in the literature (see Table S2 in Additional file 1). Baseline models For comparison purposes, we also ran the cross-validation experiments using four additional machine learning models. For these models, each target-indication pair is treated as a data point and the corresponding 17 association evidence and six target-only attributes are treated as its features. As the target-only attributes are not directly linked with specific indications, for each target, we duplicated its feature values across all indications. The task is being cast as a binary classification problem. To allow these four models to handle missing values, we treated the association scores as categorical variables with three categories: no association (0), positive association (1) and unknown (missing) association. Each categorical variable is then encoded as two binary variables (also called one-hot encoding). The four models that we tested are: 1. Logistic Regression (LR), a simple linear model. 2. LASSO [31], which is a generalized linear model with L1 regularization implemented in the glmnet R package where the regularization parameters were determined using cross-validation. 3. Random Forest (RF), an ensemble model of decision trees where the parameters are determined by the Out-of-Bag error estimate using the tuneRF function in randomForest R package. 4. Gradient Boosting Machine (GBM) [32], a boosting method which is implemented in xgboost [33] where we tuned the following parameters using cross-validation: the feature shrinkage rate, maximum depth of a tree, subsample ratio of features and number of iterations. 5. Matrix Factorization (MF): We also included another baseline model where we only used the clinical outcome matrix and applied matrix factorization to complete the matrix for prediction. Specifically, we used a nuclear norm regularized matrix factorization method that is implemented in the softimpute [34] R package and the regularization parameter is determined through cross-validation. Performance metrics We used two metrics to measure the prediction performance of the evaluated methods: area under receiver operator curve (AUROC) and area under precision-recall curve. (AUPRC). The AUROC measures the probability of Yao et al. BMC Bioinformatics (2019) 20:69 a model ranking a randomly chosen positive example higher than a randomly chosen negative example and is commonly used in assessing the performance of models for binary classification tasks. AUROC treats positive and negative examples equally, this metric is of limited value when the number of positive examples is relatively low. Given the low success rate in drug development, we chose AUPRC as the primary evaluation metric as it focuses on the performance of positive examples. Here the precision is the proportion of correctly predicted positives out of all predicted positives and recall is the proportion of correctly predicted positives out of all positives. Results Model benchmark results We performed a standard cross-validation experiment to benchmark various types of machine learning models (Fig. 1b, panel 1). The best model is the matrix factorization (MF) model (AUROC = 0.83 ± 0.02, AUPRC = 0.77 ± 0.02) (Fig. 2), which only factorizes the clinical outcomes matrix without considering any other evidence in the dataset. Due to the highly-correlated structure within the clinical outcomes of target-indication pairs, the standard way of randomly splitting them into training and test sets may overestimate the predictability of clinical outcomes. This may explain the high performance of MF; knowing which targets have succeeded against which indications in the training data may provide enough information to predict the outcome status of new indications for these targets. Many drug targets are from the same gene family, and in a random training-test split, it is likely that targets within the same gene Page 6 of 12 family are split across training and test set, though the same drug may bind to multiple members in each set. This leads to an overestimate of prediction accuracy for truly independent and novel targets. A similar effect may relate indications with different subtypes, as drug targets are often tried against many closely related diseases. To mitigate this problem and obtain an unbiased estimation of predictive capacity, we designed two benchmark experiments, where a group of similar targets (Fig. 1b, panels 2 and 3) or indications is held out as a test set and models trained on the other target or indication groups, respectively, are evaluated against the held-out set. We categorized targets into ten target classes largely derived from the ChEMBL hierarchical target classification system [30], and grouped indications into eight clusters that are based on MeSH hierarchy and co-occurrence frequency in the literature (see Additional file 1). In the leave-one-target-class-out cross-validation experiment (Fig. 2), the performance of MF decreases dramatically as there is no information in the training set to predict the clinical outcomes of the held-out target class. All the other methods perform similarly and the overall performance is not as good as in the standard cross-validation setting. This implies that it is difficult to predict candidate indications for targets that have not been assessed in clinical trials. In the leave one disease cluster out validation experiment, the performance of MF again dropped below that of the other methods as there is no information about clinical outcomes of the held-out disease clusters in the training step. Fig. 2 Benchmark performance of models. Prediction performance comparison in three benchmark schemes in terms of Area Under Receiving Operation Curve (AUROC, Top) and Area Under Precision Recall Curve (AUPRC, Bottom). Error bars are calculated from cross-validation (LR: Logistic Regression; GBM: Gradient Boosting Machine; RF: Random Forest; MF: Matrix Factorization; BTF: Bayesian Tensor Factorization) Yao et al. BMC Bioinformatics (2019) 20:69 However, the Bayesian tensor factorization (BTF) model scored as the best model in the disease group benchmark (AUROC = 0.73 ± 0.05, AUPRC = 0.58 ± 0.09) and the second to best model in standard cross-validation (AUROC = 0.82 ± 0.02, AUPRC = 0.71 ± 0.03). It is counter-intuitive that BTF does not out-perform the MF method in the standard cross-validation case, as it incorporated more data. MF approach may be taking maximum advantage of the highly-related nature of the outcomes, given the poor performance of MF in the target class and disease group benchmarks. MF also only needs to learn latent factors to explain the clinical outcomes, while BTF needs to learn latent factors to explain the clinical outcomes and all the evidence as well, which is inherently a more difficult task. In general, the performance of models that explicitly use evidence as predictors did not vary too much across three validation settings. Among these models, ensemble-based methods (RF and GBM) worked slightly better than linear model-based methods (LR and Page 7 of 12 LASSO). Although MF performed relatively well in the standard validation case, its performance was inconsistent among validation settings. BTF combined both evidence and inter-relationship among targets and indications and performed consistently well in all three validation scenarios. In addition to AUROC and AUPRC, we also evaluated performance using F-score, precision@30, and recall@30 (see Additional file 4), but the comparison across methods was not affected. Leave one out experiments One advantage of this leave one target/disease group out validation scheme is that we can assess how trained models can be generalized to groups of targets/diseases that the models have never trained on before. Figure 3 shows the prediction performance of the six models on the held-out target classes (Fig. 3a) and disease clusters (Fig. 3b). In the leave-one-target-class-out case, the prediction performance averaged over the six models varies between target classes (AUPRC ranges from 0.24 to 0.58; AUROC ranges from 0.53 to 0.68). Specifically, we a b Fig. 3 Benchmark performance of leave one out experiments. Model performance on predicting clinical outcomes of target classes (a) and disease clusters (b) in the leave-one-out experiments in terms of Area Under Receiving Operation Curve (AUROC, x-axis) and Area Under Precision Recall Curve (AUPRC, y-axis). 95% confidence interval is calculated using 1000 bootstraps. Dotted lines mark the AUROC (vertical) and AUPRC (horizontal) of a random guess, which is 0.5 and the fraction of positives in the testing set, respectively. The percentage of target-indication pairs in each held-out set is listed after the pipe symbol (\|) in the titles. (LR: Logistic Regression; GBM: Gradient Boosting Machine; RF: Random Forest; MF: Matrix Factorization; BTF: Bayesian Tensor Factorization) Yao et al. BMC Bioinformatics (2019) 20:69 notice that the models perform consistently poorly for transcriptional factor targets and miscellaneous enzymes, which implies that these target classes are quite different from the other target classes. On the other hand, most models perform relatively well in protease targets. We note the performance is consistent among models within each target class, but this low variability is not repeated in the leave-one-disease-cluster-out case, where the prediction performance shows higher variability among disease clusters. For example, the BTF model performs better than the other models in the metabolic, GI and urologic and oral disease clusters, and performs as well as any other model in the other disease clusters. Latent factors capture disease relationship within three disease areas After benchmarking the performance of the BTF model in the cross-validation experiments, we fitted the model to the whole dataset. We chose 11 latent factors (see Additional file 1). Before using the fitted BTF model to make any predictions, we explored whether the latent factors learned from the model are biologically meaningful so that we can increase our trust in the prediction made by the model. To do so, we reduced the 11 latent factors to two dimensions using t-SNE [35] to visualize how indications are distributed and examined whether the learned latent a Page 8 of 12 factors can capture inter-relationship among indications. t-SNE is a dimension reduction technique used to visualize high-dimensional dataset where similar points in high dimensional space are transformed to neighboring points in a low dimensional space and dissimilar points are transformed to distant points in the low dimensional embedding. Figure 4a shows the two-dimensional t-SNE embedding of the 574 indications with at least one clinical outcome, where three distinct clusters are present on the map. We further checked the MeSH annotations of the diseases in each cluster and found that the three clusters are enriched with three distinct disease categories including Central nervous system diseases, Digestive system diseases, and Hemic & lymphatic diseases, respectively. Interestingly, auto-immune diseases, such as rheumatoid arthritis, asthma, psoriasis and Crohn’s disease that manifest in different organs are localized in the same neighboring area on the map. This implies that latent factors are able to capture the intrinsic relationships of diseases within these disease areas. For the rest of the diseases, we did not observe distinct clustering patterns using t-SNE. This could either be because the latent factors are not rich enough to capture the relationship among these diseases or these diseases are inherently interconnected by sharing similar pathological mechanisms [36]. b Fig. 4 Validation of BTF model prediction. a t-SNE visualization of indications based on the latent factors learned in BTF model. Each dot represents one indication and the size of the dot is proportional to the number of targets that have been clinically failed. The inserted pie charts show diseases composition of representative clusters of indications in the 2D visualization. 2D embedding was obtained by using perplexity = 30 in t-SNE and the visualization is consistent using different perplexity values in the range from 10 to 50. b BTF prediction scores of targetindication pairs in Phase I-III clinical trials. The numbers are P-values (Wilcoxon rank sum tests) from comparing prediction scores of targetindication pairs between any two phases Yao et al. BMC Bioinformatics (2019) 20:69 Prediction scores of target-indication pairs under clinical trials To validate the prediction made by the BTF model, we chose 1246 novel target-indication pairs that were in clinical trials (Phase I-III) at the time when we collected the data (May 2016), and thus did not have clinical outcome readouts. We compared the prediction scores generated by the BTF model on these target-indication pairs and noticed that the prediction scores of later phase pairs are significantly higher than those of earlier phase pairs (Fig. 4b), which recapitulates the observation that drugs in later phases on average have a higher likelihood of approval [37]. Since we did not include phase information of these target-indication pairs when training the model, these pairs serve as an independent test set and the results increase our confidence in the predictions of the model. Next, we conducted a literature search on the top 63 hypotheses of the 1246 pairs based on a prediction score threshold, which corresponds with 0.8 precision and 0.27 recall in the standard cross-validation experiment. We list 15 of these 63 hypotheses along with a relevant literature reference in Table 3; the complete list of 63 can be found in Additional file 1. As an example, interleukin 6 (IL6) is an approved drug target for giant lymph node hyperplasia (Table 3). Our results suggest that the current trials for psoriatic arthritis, which includes a Phase IIb trial of a monoclonal antibody against this protein [38], have a greater than random chance of success. Psoriatic arthritis is chronic inflammatory arthritis that is associated with psoriasis and thus somewhat related to the successful indication Page 9 of 12 for IL6, a cytokine with a wide variety of biological functions. It induces the acute phase response and is involved in the final differentiation of B-cells into Ig-secreting cells in lymphocyte and monocyte differentiation. It acts on B-cells, T-cells, hepatocytes, hematopoietic progenitor cells and is required for the generation of T(H)17 cells. It also acts as a myokine and is discharged into the bloodstream after muscle contraction and acts to increase the breakdown of fats and to improve insulin resistance [39]. Genetic polymorphism of IL6 has been shown to be significantly associated with a form of psoriatic arthritis [40], and serum IL6 is considered as a biomarker for assessing disease activity in patients with psoriasis, as well as for predicting responsiveness of joint symptoms to biologic treatment [41]. Another target of interest is angiotensin II receptor type 1 (AGTR1), an important effector controlling blood pressure and volume in the cardiovascular system. It has been approved for many cardiovascular indications such as heart failure, myocardial infarction, and hypertension. The predicted indication for AGTR1 is hypercholesterolemia, also known as high cholesterol. AGTR1 antagonism improves hypercholesterolemia-associated endothelial dysfunction [42] and attenuates the inflammatory and thrombogenic responses to hypercholesterolemia in venules [43]. Significant association of AGTR1 polymorphism with hypercholesterolemia was also observed in hypertension patients [44]. Discussion In this paper, we focused on the problem of predicting clinically promising therapeutic hypotheses using Table 3 High Scoring Pairs of Interest from TF Model Target High Scoring Indication in Clinical Pipeline (Phase) PubmedID Related Approved Indication (Total Approved Indications) ABCC8 Glucose Intolerance (III) 23903354 Diabetes Mellitus (1) ADRB1 Cachexia (II) 20426789 Ischemia (13) ADRB2 Hypoglycemia (I) 22013013 Glaucoma (15) ADRB2 Myocardial Infarction (III) 26692153 Heart Failure (15) AGTR1 Hypercholesterolemia (III) 12117739 Hyperlipidemias (7) CYP3A4 Hepatitis C (II) 20938912 HIV Infections (1) IL2 Behcet Syndrome (II) 26654556 Graft Rejection (1) IL6 Waldenstrom Macroglobulinemia (I) 26238488 Giant Lymph Node Hyperplasia (1) IL6 Arthritis, Psoriatic (II) 27789987 Giant Lymph Node Hyperplasia (1) OPRM1 Schizophrenia (III) 27397309 Migraine Disorders (22) RYR1 Muscular Dystrophy, Duchenne (I) 26793121 Malignant Hyperthermia (1) SERPINC1 Hemophilia (II) 27099538 Blood Coagulation Disorders (16) TNFSF11 Hypercalcemia (II) 27904108 Osteoporosis (1) VDR Alopecia (I) 27932380 Keratosis (9) VDR Cachexia (I) 22497530 Chronic Kidney Failure (9) New indications of approved targets in clinical trials (Phase as of May 27, 2016) that have the highest probability of eventual clinical success as measured by the tensor factorization model. The full list is available in the supplement. For illustrative purposes, we list a related indication approved for assets for each target Yao et al. BMC Bioinformatics (2019) 20:69 associative knowledge of targets and indications. We compared tensor factorization with other traditional machine learning methods in a variety of benchmarking experiments and identified two interesting findings from the evaluation of this method: 1) the latent factors learned from the model align with known biological relationships among three human disease areas, and 2) the method can be applied to different scenarios of drug discovery and achieves competitive prediction performance. However, there are some limitations worth discussing before deploying tensor factorization to propose novel target-indication hypotheses. First, the model relies on the available compilation of evidence sources. Open Targets provided us with a good foundation, but clearly, more sources could be gathered. Second, we treated every clinical failure equally. Our preliminary analysis has shown that some target-indications pairs have been tried multiple times and are still being pursued clinically, while some failed only once and were never tested again. Although the probabilistic framework of the model can potentially mitigate this problem, the model does not explicitly differentiate definitive failures from those that have not been thoroughly explored and may become successful drugs in the future. Lastly, we only applied the technique to a dataset of targets and indications with at least one clinical outcome; thus, the application as benchmarked here is constrained to applying approved drug targets to new indications. The methodology, however, can be expanded to any target and any indication so long as their evidence is encoded in the data. Such an application may result in the identification of novel target-indication hypotheses with a high predicted probability of being successfully translated into medicines. Computational prediction of drug targets has been widely studied in the context of predicting disease-associated genes [14–16, 45–47]. These disease-associated genes can facilitate the discovery of drug targets by narrowing down the search space of potential targets. The prediction performance (precision) of the models varies from 0.5 to 0.9 depending on the methods and data used in the studies. Many related studies design models to infer novel associations by leveraging similarity information between biological entities and biomolecular network information encoded in a protein-protein interaction database [47, 48]. One example is FASCINATE [17], which is able to infer cross-layer dependencies on multi-layered biological networks. This method can be used for this problem by collapsing all evidence and augmenting the data with disease similarity information. Conclusion In this work, we evaluated a machine learning technique called tensor factorization on the problem of predicting clinical outcomes of therapeutic hypotheses using Page 10 of 12 existing association evidence between drug targets and disease indications. We illustrate that the method can achieve equal or better prediction performance compared with a variety of baseline models across three scenarios of drug discovery, and the learned model can capture the known biological mechanism of human diseases. Furthermore, we demonstrated an application of the method to predict outcomes of trials on novel indications of approved drug targets. Future work includes expanding this method to targets and indications that previously have never been clinically tested and proposing novel target-indication hypotheses that can be developed into medicines with predicted high probabilities of success. Additional files Additional file 1: Supplementary material. (PDF 490 kb) Additional file 2: Target-indication association evidence data file. (ZIP 227 kb) Additional file 3: Target-only attribute data file. (ZIP 5 kb) Additional file 4: Results of benchmark experiments. (XLSX 11 kb) Abbreviations AUPRC: Area Under Precision Recall Curve; AUROC: Area Under Receiver Operator Curve; BTF: Bayesian Tensor Factorization; GBM: Gradient Boosting Machine; LR: Logistic Regression; MeSH: Medical Subject Heading; MF: Matrix Factorization; RF: Random Forest Acknowledgments We would like to thank Andrew Rouillard for his helpful discussion and valuable comments on the manuscript. We thank Johannes Freudenberg for supplying the data for disease grouping. We thank Gautier Koscielny for facilitating access to Open Target data. We also thank Mandy Bergquist, Enrico Ferrero, Victor Neduva and Naruemon Pratanwanich for helpful discussions. Funding The authors received no specific funding for this work. Availability of data and materials The target-indication association evidence was downloaded from Open Target https://www.targetvalidation.org/downloads/data. Clinical outcomes, tissue overexpression from GeneLogic and pathway information from Metabase are derived from commercial databases, thus needs to be licensed. The majority of the data are supplied in Additional file 2 (target-indication association) and Additional file 3 (target only evidence). Authors’ contributions JY, MH, and MN conceived the project. MH performed ontology mapping and curated target annotation; MN calculated MeSH similarity of indications. JY and PA designed the experiments. JY performed the experiments and analyzed the results. JY, MH, and PA interpreted the results and wrote the manuscript. All authors read and approved the final manuscript. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests JY, MH, MN, and PA are full-time employees of GSK. Yao et al. BMC Bioinformatics (2019) 20:69 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 Computational Biology, GSK R&D, 1250 S. Collegeville Road, UP12-200, Collegeville, PA, USA. 2Genetics, GSK R&D, 1250 S. Collegeville Road, UP12-200, Collegeville, PA, USA. Received: 26 September 2018 Accepted: 30 January 2019 References 1. Arrowsmith J, Miller P. Trial watch: phase II and phase III attrition rates 20112012. Nat Rev Drug Discov. 2013;12(8):569. 2. Cook D, Brown D, Alexander R, March R, Morgan P, Satterthwaite G, Pangalos MN. Lessons learned from the fate of AstraZeneca's drug pipeline: a five-dimensional framework. Nat Rev Drug Discov. 2014;13(6):419–31. 3. Bertram L, Tanzi RE. Thirty years of Alzheimer's disease genetics: the implications of systematic meta-analyses. Nat Rev Neurosci. 2008;9:768. 4. Koscielny G, An P, Carvalho-Silva D, Cham JA, Fumis L, Gasparyan R, Hasan S, Karamanis N, Maguire M, Papa E. Open targets: a platform for therapeutic target identification and validation. Nucleic Acids Res. 2016;45(D1):D985–94. 5. Kolda TG, Bader BW. Tensor decompositions and applications. SIAM Rev. 2009;51(3):455–500. 6. Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems. Computer. 2009;42(8). 7. Nickel M, Murphy K, Tresp V, Gabrilovich E. A review of relational machine learning for knowledge graphs. Proc IEEE. 2016;104(1):11–33. 8. Luo Y, Wang F, Szolovits P. Tensor factorization toward precision medicine. Brief Bioinform. 2017;18(3):511–4. 9. Nelson MR, Tipney H, Painter JL, Shen J, Nicoletti P, Shen Y, Floratos A, Sham PC, Li MJ, Wang J. The support of human genetic evidence for approved drug indications. Nat Genet. 2015;47(8):856. 10. Rouillard AD, Hurle MR, Agarwal P. Systematic interrogation of diverse Omic data reveals interpretable, robust, and generalizable transcriptomic features of clinically successful therapeutic targets. PLoS Comput Biol. 2018;14(5): e1006142. 11. Sun J, Zhu K, Zheng WJ, Xu H. A comparative study of disease genes and drug targets in the human protein interactome. BMC Bioinformatics. 2015; 16(5):S1. 12. Heinemann F, Huber T, Meisel C, Bundschus M, Leser U. Reflection of successful anticancer drug development processes in the literature. Drug Discov Today. 2016;21(11):1740–4. 13. Moreau Y, Tranchevent L-C. Computational tools for prioritizing candidate genes: boosting disease gene discovery. Nat Rev Genet. 2012;13:523. 14. Ghiassian SD, Menche J, Barabási A-L. A DIseAse MOdule detection (DIAMOnD) algorithm derived from a systematic analysis of connectivity patterns of disease proteins in the human Interactome. PLoS Comput Biol. 2015;11(4):e1004120. 15. Carson MB, Lu H. Network-based prediction and knowledge mining of disease genes. BMC Med Genet. 2015;8(2):S9. 16. Yang P, Li X, Chua H-N, Kwoh C-K, Ng S-K. Ensemble positive unlabeled learning for disease gene identification. PLoS One. 2014;9(5):e97079. 17. Chen C, Tong H, Xie L, Ying L, He Q: FASCINATE: fast cross-layer dependency inference on multi-layered networks. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; San Francisco, California, USA. 2939784: ACM 2016: 765–774. 18. Pharmaprojects Database [https://citeline.com/products/pharmaprojects]. 19. Yao L, Rzhetsky A. Quantitative systems-level determinants of human genes targeted by successful drugs. Genome Res. 2008;18(2):206–13. 20. Bull SC, Doig AJ. Properties of protein drug target classes. PLoS One. 2015; 10(3):e0117955. 21. Mitsopoulos C, Schierz AC, Workman P, Al-Lazikani B. Distinctive behaviors of Druggable proteins in cellular networks. PLoS Comput Biol. 2015;11(12): e1004597. 22. Mora A, Donaldson IM. Effects of protein interaction data integration, representation and reliability on the use of network properties for drug target prediction. BMC Bioinformatics. 2012;13(1):294. 23. Harrison RK. Phase II and phase III failures: 2013–2015. Nat Rev Drug Discov. 2016;15:817. Page 11 of 12 24. Ma H, Yang H, Lyu MR, King I: Sorec: social recommendation using probabilistic matrix factorization. In: Proceedings of the 17th ACM conference on information and knowledge management: 2008. ACM: 931– 940. 25. Mnih A, Salakhutdinov RR. Probabilistic matrix factorization. In: Advances in neural information processing systems; 2008. p. 1257–64. 26. Salakhutdinov R, Mnih A: Bayesian probabilistic matrix factorization using Markov chain Monte Carlo. In: Proceedings of the 25th international conference on machine learning: 2008. ACM: 880–887. 27. Simm J, Arany A, Zakeri P, Haber T, Wegner JK, Chupakhin V, Ceulemans H, Moreau Y: Macau: scalable bayesian multi-relational factorization with side information using MCMC. arXiv preprint arXiv:150904610 2015. 28. Julia implementation of Bayesian tensor factorization algorithm [https:// github.com/jaak-s/BayesianDataFusion.jl]. 29. Gelman A, Carlin JB, Stern HS, Dunson DB, Vehtari A, Rubin DB. Bayesian Data Analysis, vol. 2. Boca Raton, FL: CRC press; 2014. 30. Bento AP, Gaulton A, Hersey A, Bellis LJ, Chambers J, Davies M, Krüger FA, Light Y, Mak L, McGlinchey S. The ChEMBL bioactivity database: an update. Nucleic Acids Res. 2014;42(D1):D1083–90. 31. Friedman J, Hastie T, Tibshirani R. Regularization paths for generalized linear models via coordinate descent. J Stat Softw. 2010;33(1):1. 32. Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat. 2001:1189–232. 33. Chen T, Guestrin C: Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining: 2016. ACM: 785–794. 34. Mazumder R, Hastie T, Tibshirani R. Spectral regularization algorithms for learning large incomplete matrices. J Mach Learn Res. 2010;(11):2287–322. 35. Maaten Lvd HG. Visualizing data using t-SNE. J Mach Learn Res. 2008;9(Nov): 2579–605. 36. Menche J, Sharma A, Kitsak M, Ghiassian SD, Vidal M, Loscalzo J, Barabási AL. Uncovering disease-disease relationships through the incomplete interactome. Science. 2015;347(6224):1257601. 37. Hay M, Thomas DW, Craighead JL, Economides C, Rosenthal J. Clinical development success rates for investigational drugs. Nat Biotechnol. 2014; 32:40. 38. Mease PJ, Gottlieb AB, Berman A, Drescher E, Xing J, Wong R, Banerjee S. The efficacy and safety of clazakizumab, an anti–interleukin-6 monoclonal antibody, in a phase IIb study of adults with active psoriatic arthritis. Arthritis Rheumatol. 2016;68(9):2163–73. 39. Uniprot entry of IL6 [http://www.uniprot.org/uniprot/P05231]. 40. Cubino N, Montilla C, Usategui-Martín R, Cieza-Borrela C, Carranco T, CaleroPaniagua I, Quesada A, Cañete J, Queiro R, Sánchez M. Association of IL1Β (−511 a/C) and IL6 (−174 G> C) polymorphisms with higher disease activity and clinical pattern of psoriatic arthritis. Clin Rheumatol. 2016;35(7):1789–94. 41. Muramatsu S, Kubo R, Nishida E, Morita A. Serum interleukin-6 levels in response to biologic treatment in patients with psoriasis. Mod Rheumatol. 2017;27(1):137–41. 42. Wassmann S, Hilgers S, Laufs U, Böhm M, Nickenig G. Angiotensin II type 1 receptor antagonism improves hypercholesterolemia-associated endothelial dysfunction. Arterioscler Thromb Vasc Biol. 2002;22(7):1208–12. 43. Petnehazy T, Stokes KY, Russell JM, Granger DN. Angiotensin II type-1 receptor antagonism attenuates the inflammatory and thrombogenic responses to hypercholesterolemia in venules. Hypertension. 2005;45(2): 209–15. 44. Morisawa T, Kishimoto Y, Kitano M, Kawasaki H, Hasegawa J. Influence of angiotensin II type 1 receptor polymorphism on hypertension in patients with hypercholesterolemia. Clin Chim Acta. 2001;304(1):91–7. 45. Costa PR, Acencio ML, Lemke N: A machine learning approach for genomewide prediction of morbid and druggable human genes based on systemslevel data. In: BMC Genomics: 2010. BioMed Central: S9. 46. Yang P, Li X-L, Mei J-P, Kwoh C-K, Ng S-K. Positive-unlabeled learning for disease gene identification. Bioinformatics. 2012;28(20):2640–7. 47. Emig D, Ivliev A, Pustovalova O, Lancashire L, Bureeva S, Nikolsky Y, Bessarabova M. Drug target prediction and repositioning using an integrated network-based approach. PLoS One. 2013;8(4):e60618. 48. Barabási A-L, Gulbahce N, Loscalzo J. Network medicine: a network-based approach to human disease. Nat Rev Genet. 2010, 12:56. 49. Lonsdale J, Thomas J, Salvatore M, Phillips R, Lo E, Shad S, Hasz R, Walters G, Garcia F, Young N, et al. The genotype-tissue expression (GTEx) project. Nat Genet. 2013;45:580. Yao et al. BMC Bioinformatics (2019) 20:69 50. Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson Å, Kampf C, Sjöstedt E, Asplund A, et al. Tissue-based map of the human proteome. Science. 2015;347(6220). 51. Lek M, Karczewski KJ, Minikel EV, Samocha KE, Banks E, Fennell T, O’DonnellLuria AH, Ware JS, Hill AJ, Cummings BB, et al. Analysis of protein-coding genetic variation in 60,706 humans. Nature. 2016;536:285. 52. Petrovski S, Wang Q, Heinzen EL, Allen AS, Goldstein DB. Genic intolerance to functional variation and the interpretation of personal genomes. PLoS Genet. 2013;9(8):e1003709. 53. Smedley D, Haider S, Durinck S, Pandini L, Provero P, Allen J, Arnaiz O, Awedh MH, Baldock R, Barbiera G, et al. The BioMart community portal: an innovative alternative to large, centralized data repositories. Nucleic Acids Res. 2015;43(W1):W589–98. Page 12 of 12
https://openalex.org/W3125588101	https://europepmc.org/articles/pmc5981319?pdf=render	English	null	Application of a Novel S3 Nanowire Gas Sensor Device in Parallel with GC-MS for the Identification of Rind Percentage of Grated Parmigiano Reggiano	null	2,018	cc-by	9,794	Received: 11 April 2018; Accepted: 15 May 2018; Published: 18 May 2018 Abstract: Parmigiano Reggiano cheese is one of the most appreciated and consumed foods worldwide, especially in Italy, for its high content of nutrients and taste. However, these characteristics make this product subject to counterfeiting in different forms. In this study, a novel method based on an electronic nose has been developed to investigate the potentiality of this tool to distinguish rind percentages in grated Parmigiano Reggiano packages that should be lower than 18%. Different samples, in terms of percentage, seasoning and rind working process, were considered to tackle the problem at 360◦. In parallel, GC-MS technique was used to give a name to the compounds that characterize Parmigiano and to relate them to sensors responses. Data analysis consisted of two stages: Multivariate analysis (PLS) and classiﬁcation made in a hierarchical way with PLS-DA ad ANNs. Results were promising, in terms of correct classiﬁcation of the samples. The correct classiﬁcation rate (%) was higher for ANNs than PLS-DA, with correct identiﬁcation approaching 100 percent. Keywords: electronic nose; nanowire gas sensors; food quality control; Parmigiano Reggiano; multivariate data analysis; artiﬁcial neural network Article Application of a Novel S3 Nanowire Gas Sensor Device in Parallel with GC-MS for the Identiﬁcation of Rind Percentage of Grated Parmigiano Reggiano Marco Abbatangelo 1,* ID , Estefanía Núñez-Carmona 1 ID , Veronica Sberveglieri 2,3 ID , Dario Zappa 1, Elisabetta Comini 1,3 and Giorgio Sberveglieri 1,3 1 Department of Information Engineering, University of Brescia, Via Branze 38, 25123 Brescia, Italy; e.nunezcarmona@unibs.it (E.N.-C); dario.zappa@unibs.it (D.Z.); elisabetta.comini@unibs.it (E.C.); giorgio.sberveglieri@unibs.it (G.S.) 1 Department of Information Engineering, University of Brescia, Via Branze 38, 25123 Brescia, Italy; e.nunezcarmona@unibs.it (E.N.-C); dario.zappa@unibs.it (D.Z.); elisabetta.comini@unibs.it (E.C.); giorgio.sberveglieri@unibs.it (G.S.) g g g 2 CNR-IBBR, Institute of Biosciences and Bioresources, Via Madonna del Piano 10, 50019 Sesto Fiorentino (FI) Italy; veronica.sberveglieri@ibbr.cnr.it y g 3 NANO SENSOR SYSTEMS S.r.l., Via Branze 38, 25123 Brescia, Italy * Correspondence: m.abbatangelo@unibs.it; Tel.: +39-348-842-3503 * Correspondence: m.abbatangelo@unibs.it; Tel.: +39-348-842-3503 sensors sensors sensors Sensors 2018, 18, 1617; doi:10.3390/s18051617 www.mdpi.com/journal/sensors 1. Introduction Parmigiano Reggiano (PR) cheese is among the most typical Italian foods and one of the oldest traditional cheeses produced in Europe. It is also the most important Protected Designation of Origin (PDO) Italian cheese in terms of commercial importance [1]. Its production is regulated by the Parmigiano Reggiano Cheese Consortium (CFPR). According to European Regulation 510/2006, this designation can be exclusively assigned to the cheese only when it is made with a traditional established production technology in a restricted area of Italy (provinces of Parma, Reggio Emilia, Modena, Mantova and Bologna) from milk produced in the same area [2]. PR can be found on the market in different forms. It can be portioned or grated and cannot be subjected to any treatment like lyophilization, drying and freezing [3]. All the procedures, which must be followed to obtain the original PR, make this cheese a high-value product. This leads to a ﬁnal product that has various nutritional properties: its dry weight is mostly composed of proteins and lipids, it is lactose- and galactose-free and it is rich in organic acids, such as lactic acid, acetic acid, propionic and butyric acids [4]. The semi-fat composition, due to natural creaming of skimmed Sensors 2018, 18, 1617; doi:10.3390/s18051617 www.mdpi.com/journal/sensors 2 of 15 Sensors 2018, 18, 1617 unpasteurized milk [5], is produced by cattle that consume only locally grown forage because supply of silage and fermented feeds is not permitted [6]. For these reasons, PR has a high cost when compared to similar hard cheeses. This encourages the appearance on the market of counterfeited products that bear the PR brand at a lower price. The rate of fraud is estimated to be between 20% and 40%, the latter predominantly in the grated form [7]. As established in the procedural guideline, grated PR cheese must follow some technical and technological parameters: moisture no less than 25% and no more than 35%, at least 12 months of ripening, rind percentage compared to pulp not over 18% (by weight), typical amino-acid composition of the cheese, absence of additives, not crumbly in aspect and with homogeneous particles that have a diameter inferior to 0.5 mm and do not exceed 25% [8]. In order to determine if a PR cheese package conforms to the rules, the aromatic proﬁle of grated PR can be analyzed thanks to the volatile organic compounds (VOCs). 1. Introduction VOCs of various dairy products have received a great deal of attention in recent years. Until now, about 600 volatile compounds have been identiﬁed for cheese [9]. However, only a small part of these compounds is responsible for cheese ﬂavor [10]. Cheese aroma is considered the result of the equilibrium between various VOCs that, separately, do not reﬂect the overall odor [11]. Hydrocarbons, alcohols, aldehydes, ketones, esters and lactones were the major classes of compounds found in the neutral fraction of cheese [12]. In this work, an electronic nose has been used in order to analyze rind percentage in grated PR cheese through emitted VOCs. In recent years, this kind of device has received considerable attention for its potentialities; it has been applied in various ﬁelds, such as environment [13–17], health [18–22] and food, with excellent results [23–25]. Regarding food applications, some examples of electronic nose applications are the detection of microorganisms in tomato sauce [26] and of different molds in coffee [27], the determination of the shelf life of milk [28], the detection of additives in fruit juices [29], the identiﬁcation of fruit [30] and the discrimination of cheese varieties [31,32]. These few examples show how e-noses have the potential to be used in different ways to assess food quality and identity. Placed side by side with e-nose analysis, Gas Chromatography coupled with Solid Phase Micro Extraction (SPME) was used. SPME has received a great deal of attention in the literature to ﬁnd VOCs that characterize food matrices. Many foods have been studied, including dairy products, such as milk [33], butter [34] and cheese [35,36]. The aim of this preliminary research is to determine the rind percentage of the sample under analysis with an innovative and rapid methodology, in order to identify frauds and therefore have available an affordable and reliable instrument to reduce them, thanks to the different VOCs, in terms of presence and amount, between products. 2.2. GC-MS Analysis The Gas Chromatograph (GC) used during the analyses was a Shimadzu GC2010 PLUS (Kyoto, KYT, Japan), equipped with a Shimadzu single quadrupole Mass Spectometer (MS) MS-QP2010 Ultra (Kyoto, KYT, Japan) and an autosampler HT280T (HTA s.r.l., Brescia, Italy). The GC-MS analysis was coupled with the Solid-Phase Micro Extraction (SPME) method in order to ﬁnd the most signiﬁcant VOCs, which allows for recognition of the different kinds of cheeses. The ﬁber used for the adsorption of volatiles was a DVB/CAR/PDMS-50/30 µm (Supelco Co. Bellefonte, PA, USA). The ﬁber was exposed to the headspace of the vials after heating the samples in the HT280T oven, thermostatically regulated at 50 ◦C for 15 min, with the aim of creating the headspace equilibrium. The length of the ﬁber in the headspace was kept constant. Desorption of volatiles took place in the injector of the GC-MS for 6 min at 250 ◦C. The gas chromatograph was operated in the direct mode throughout the run, with the mass spectrometer in electron ionization (EI) mode (70 eV). GC separation was performed on a MEGA-WAX capillary column (30 m × 0.25 mm × 0.25 µm, Agilent Technologies, Santa Clara, CA, USA). Ultrapure helium (99.99%) was used as the carrier gas, at the constant ﬂow rate of 1.3 mL/min. The following GC oven temperature programming was applied. At the beginning, the column was held at 40 ◦C for 8 min, and then raised from 40 to 190 ◦C at 4 ◦C/min; then, the temperature was maintained at 190 ◦C for 5 min. Next, the temperature was raised from 190 ◦C to 210 ◦C, with a rate of 5 ◦C/min; ﬁnally, 210 ◦C was maintained for 5 min. The GC-MS interface was kept at 200 ◦C. The mass spectra were collected over the range of 45 to 500 m/z in the Total Ion Current (TIC) mode, with scan intervals at 0.3 s. The identiﬁcation of the volatile compounds was carried out using the NIST11 and the FFNSC2 libraries of mass spectra. The described parameters have been optimized for this speciﬁc application. Each sample was analyzed one time. 2.1. Samples Preparation and Experimental Design Analyzed samples were packaged under vacuum at the headquarters of CFPR. They came from two different ripening stages: 12 and 24 months. For each of these, ﬁve different combinations of pulp-rind were prepared (expressed in rind percentage): 0%, 18%, 26%, 45% and 100%. In addition, two kinds of rind working processes were considered: washed-rind (WR) and scraped-rind (SR). The only exceptions are represented by 0% samples, for which only the 24-month ripening was taken into account, and 100% samples, for which there is one for WR, and one for SR, that corresponds to 24-month and 12-month seasoning, respectively. For each sample, 14 replicas were arranged for a total of 210 (14 replicas × 15 samples). Samples were stored at 4 ◦C until the moment when they were prepared for the analysis. The amount of 2 g of grated cheese was positioned in 20 mL glass headspace vials and sealed by a metal cap with a PTFE-silicon membrane, crimped with an aluminum crimp. 3 of 15 Sensors 2018, 18, 1617 2.3. S3 Analysis The innovative Small Sensors System S3 device used in the present work has been completely designed and constructed at SENSOR Laboratory (University of Brescia, Italy) in collaboration with NASYS S.r.l., a spin-off of the University of Brescia. The tool comprises a metal oxide (MOX) gas sensors array, ﬂow sensors, temperature and humidity sensors, ﬂuidodinamic system, electronic control system. In particular, the sensors used in this study are 8 MOX gas sensors. Three of them are nanowires of MOX, as presented in References [37,38]. Two of them are tin oxides nanowires sensors, both grown by means of the Vapor Liquid Solid technique [39], using a gold catalyst on the alumina substrate and functionalizing one of them with gold clusters; the third sensor has an active layer of copper oxide nanowires. The working temperature is 350 ◦C, 350 ◦C and 400 ◦C, respectively. The other three sensors are prepared with Rheotaxial Growth and Thermal Oxidation (RGTO) thin ﬁlm technology; one is tin oxide functionalized with gold clusters (working at 400 ◦C), while the other two are pure tin oxide (working at 300 ◦C and at 400 ◦C, respectively). The last two are commercial MOX sensors produced by Figaro Engineering Inc. (Osaka, Japan). In particular, they are the TGS2611 and TGS2602, which are sensitive to natural gases and odorous gases like ammonia, respectively, according to the datasheet of the company. Commercial sensors have been mounted on our e-nose in order to evaluate the performances of nanowire sensors. Details of S3 sensors made at SENSOR Laboratory are summarized in Table 1. Response to 5 ppm of ethanol, selectivity (response ethanol/response carbon monoxide) and limit of detection (LOD) of ethanol are also included. The MOX nanowires are gas sensors with a high sensitivity to a broad range of chemicals; they exhibit physical properties that are signiﬁcantly different from their polycrystalline counterpart. The nanowires have a high degree of crystallinity, atomically-sharp terminations and an extraordinary length-to-width ratio, resulting in enhanced sensing capability as well as long-term material stability for prolonged operation. In addition, the three-dimensional network formed by the nanowires 4 of 15 Sensors 2018, 18, 1617 increases the adsorption surface and the catalytic activity, improving the response and decreasing the instrument threshold [40]. SnO2 and CuO nanowires structures obtained from SEM are shown in Figure 1A,B, respectively. 2.3. S3 Analysis S3 analyzes the head space (HS), i.e., the volatile fraction of the samples formed when the equilibrium of the solid–liquid phase and the vapor phase of all volatile compounds is reached. The creation of the HS depends on the test substance (vapor pressure) and the conditioning temperature of the sample. The compounds are extracted at the equilibrium point between the solid phase and the vapor in a dynamic head space. This characteristic allows for a non-destructive samples analysis. In this case, the sensor base line is obtained from the air of the surrounding environment; no gas cylinder of chromatographic air is required (an essential feature that makes it a portable instrument). The environmental air was ﬁltered using a small metal cylinder (21.5 cm in length, 5 cm in diameter) ﬁlled with activated carbons. Table 1. Type, composition, morphology, operating temperature, response (∆R/R), selectivity (response ethanol/response carbon monoxide) and limit of detection (LOD) of ethanol for S3 sensors made at the SENSOR Laboratory. Table 1. Type, composition, morphology, operating temperature, response (∆R/R), selectivity (response ethanol/response carbon monoxide) and limit of detection (LOD) of ethanol for S3 sensors made at the SENSOR Laboratory. SENSOR Laboratory. Materials (Type) Composition Morphology Operating Temperature (◦C) Response to 5 ppm of Ethanol Selectivity Limit of Detection (LOD) of Ethanol (ppm) SnO2Au (n) SnO2 functionalized with Au clusters RGTO 400 ◦C 6.5 3 0.5 SnO2 (n) SnO2 RGTO 300 ◦C 3.5 2.5 1 SnO2 (n) SnO2 RGTO 400 ◦C 4 2 0.8 SnO2Au+Au (n) SnO2 grown with Au and functionalized with gold clusters Nanowire 350 ◦C 7 2.5 0.5 SnO2Au (n) SnO2 grown with Au Nanowire 350 ◦C 5 2.1 1 CuO (p) CuO Nanowire 400 ◦C 1.5 1.5 1 The volatile fraction is then aspirated and transported to the sensor chamber to be analyzed. In order to avoid any inﬂuence of the surrounding environment to the sensor response, the chamber has been thermostated and isolated. To prevent the absorption of volatile substances that could be released during subsequent analysis, the chamber and the connection between the elements’ tires are made using steel. The air is ﬂown into the sensor chamber using a pump through a needle valve. This is used to adjust the total airﬂow, which is measured by a ﬂowmeter downstream from the pump. 2.3. S3 Analysis The instrument was also provided with the auto-sampler head space system HT2010H, supporting a 42-loading-sites carousel and a shaking oven to equilibrate the sample head space. The vials were placed in a randomized mode into the carousel. Each vial was incubated at 50 ◦C for 5 min in the auto-sampler oven and shaken for 1 min during the incubation. The sample head space was then extracted from the vial in the dynamic head space path and released into the carried ﬂow (80 sccm). Figure 1C shows the experimental setup (S3 and auto-sampler with its carousel ﬁlled with vials). The analysis timeline can be divided into three different steps for a duration of 420 s (7 min) per sample, which are preceded by a warm-up step that allows for the achievement of the base line for the entire system: • Injection: the sample HS is ﬂowed in the sensor chamber for 60 s (actual analysis time); then, Injection: the sample HS is ﬂowed in the sensor chamber for 60 s (actual analysis time); then, for • Injection: the sample HS is ﬂowed in the sensor chamber for 60 s (actual analysis time); then, for 30 s, environmental air ﬂows through the same tube to clean it from any residual VOCs; • Injection: the sample HS is ﬂowed in the sensor chamber for 60 s (actual analysis time); then, for 30 s, environmental air ﬂows through the same tube to clean it from any residual VOCs; j p ( y ); , 30 s, environmental air ﬂows through the same tube to clean it from any residual VOCs; • Restore: when the injection period is ﬁnished, the ﬁltered air is ﬂowed into the sensors camber. During this time (330 s), the sensors restore the original condition of the base line. • Restore: when the injection period is ﬁnished, the ﬁltered air is ﬂowed into the sensors camber. During this time (330 s), the sensors restore the original condition of the base line. Thanks to the processor integrated in the S3 instrument, the frequency at which the equipment works is equal to 1 Hz. 5 of 15 Sensors 2018, 18, 1617 Figure 1. (A) SEM image of SnO2 nanowires. (B) SEM image of CuO nanowires. (C) Experimental setup formed by S3 and autosampler. Figure 1. (A) SEM image of SnO2 nanowires. (B) SEM image of CuO nanowires. 2.4. Data Analysis D t l i 2.4. Data Analysis Data analysis was performed using MATLAB® R2015a software (MathWorks, USA). First of all, sensors responses in terms of resistance (Ω) were normalized when compared to the first value of the acquisition (R0). For all the sensors, the difference between the first value and the minimum value during the analysis time was calculated. Hence, the dataset was composed of ΔR/R0 parameters. In the second step the normal distribution of the variables was checked using the Jarque Bera Data analysis was performed using MATLAB® R2015a software (MathWorks, USA). First of all, sensors responses in terms of resistance (Ω) were normalized when compared to the ﬁrst value of the acquisition (R0). For all the sensors, the difference between the ﬁrst value and the minimum value during the analysis time was calculated. Hence, the dataset was composed of ∆R/R0 parameters. In the second step, the normal distribution of the variables was checked using the Jarque-Bera (JB) test, with a significance level equal to 0.05 chosen. This test is a goodness-of-fit test of whether sample data have the skewness and kurtosis matching a normal distribution. The null hypothesis is a joint hypothesis from both the skewness and the excess kurtosis being zero. Based on the test result Partial Least Squares (PLS) method was used both to view how the In the second step, the normal distribution of the variables was checked using the Jarque-Bera (JB) test, with a signiﬁcance level equal to 0.05 chosen. This test is a goodness-of-ﬁt test of whether sample data have the skewness and kurtosis matching a normal distribution. The null hypothesis is a joint hypothesis from both the skewness and the excess kurtosis being zero. Based on the test result, Partial Least Squares (PLS) method was used both to view how the groups of samples were represented thanks to sensors responses and to build the model that was used to classify the samples themselves. Finally classification was performed comparing two different classifiers: Partial Least Squares Based on the test result, Partial Least Squares (PLS) method was used both to view how the groups of samples were represented thanks to sensors responses and to build the model that was used to classify the samples themselves. Finally, classification was performed, comparing two different classifiers: Partial Least Squares Discriminant Analysis (PLS-DA) and Artificial Neural Networks (ANNs). 2.3. S3 Analysis (C) Experimental setup formed by S3 and autosampler. Figure 1. (A) SEM image of SnO2 nanowires. (B) SEM image of CuO nanowires. (C) Experimental setup formed by S3 and autosampler. Figure 1. (A) SEM image of SnO2 nanowires. (B) SEM image of CuO nanowires. (C) Experimental setup formed by S3 and autosampler. 2.4. Data Analysis D t l i 2.4. Data Analysis PLS-DA was successfully applied in different fields where products had to be recognized according to their place of origin or the presence of contamination, such as in milk [41], honey [42], wine [43] and cheese [44]. ANNs are complex structures that try to mimic what the human brain does. They are formed by elemental units called neurons that work like real neurons: once information arrives, they elaborate it and give an output. Each neuron is characterized by an activation function and coefficients of connectivity called weights. The overall structure is mainly composed of an input layer, hidden layers and an output layer [45]. ANNs can be used to resolve regression and classification problems, and function approximation. They found a lot of space in food applications for the analysis of data collected with electronic noses [46–49]. In this work, a feed-forward ANN trained with the Levenberg-Marquardt algorithm is used. For PLS dataset was split in two parts a training set and test set using Venetian Blinds (VB) Finally, classiﬁcation was performed, comparing two different classiﬁers: Partial Least Squares Discriminant Analysis (PLS-DA) and Artiﬁcial Neural Networks (ANNs). PLS-DA was successfully applied in different ﬁelds where products had to be recognized according to their place of origin or the presence of contamination, such as in milk [41], honey [42], wine [43] and cheese [44]. ANNs are complex structures that try to mimic what the human brain does. They are formed by elemental units called neurons that work like real neurons: once information arrives, they elaborate it and give an output. Each neuron is characterized by an activation function and coefﬁcients of connectivity called weights. The overall structure is mainly composed of an input layer, hidden layers and an output layer [45]. ANNs can be used to resolve regression and classiﬁcation problems, and function approximation. They found a lot of space in food applications for the analysis of data collected with electronic noses [46–49]. In this work, a feed-forward ANN trained with the Levenberg-Marquardt algorithm is used. For PLS, dataset was split in two parts—a training set and test set—using Venetian Blinds (VB) as the cross-validation procedure. This method divides the whole dataset in j cross-validation groups; in each one, one sample is put in the test set and the others in the training set on the first step. 2.4. Data Analysis D t l i 2.4. Data Analysis For PLS, dataset was split in two parts—a training set and test set—using Venetian Blinds (VB) as the cross-validation procedure. This method divides the whole dataset in j cross-validation groups; 6 of 15 Sensors 2018, 18, 1617 in each one, one sample is put in the test set and the others in the training set on the ﬁrst step. Subsequently, in every group the sample after the previous one is taken into the test set, and the others into the training set, and so on. In this work, the number of cross-validation groups chosen was equal to 10. For classiﬁcation with PLS-DA, a toolbox made for MATLAB® and released by Milano Chemometrics was used [50]. Instead, ANNs were created using the function nntool of the same software. This tool allows for the random splitting of the dataset in test and training sets by default. 3.1. GC-MS Analysis Results From the comparison between samples chromatograms, substantial differences were found. The main difference between 12-months and 24-months ripened grated PR lies in the amount of fatty acids that characterize this product. They are acetic acid, butanoic acid, hexanoic acid, octanoic acid, n-decanoic acid, and their presence is much greater in 24-months PR. In Figure 2, histograms for each of the aforementioned compounds are shown: results are presented in terms of mean ± standard deviation of the mean with an arbitrary unit. This result is widely conﬁrmed in the literature. Indeed, it is well known that these fatty acids are the results of fermentation processes, especially in butter and seasoned cheese. Some studies revealed that the amount of acetic acid and butanoic acid doubles in the period between 12 and 24 months [51,52]. The same trend was observed in the other compounds, since biochemical processes that lead to their formation are very similar. Figure 2. Cont. Figure 2. Cont. 7 of 15 Sensors 2018, 18, 1617 Figure 2. Comparison of acetic acid, butanoic acid, hexanoic acid, octanoic acid and n-decanoic acid amount between 12-months and 24-months samples. Results are presented in terms of mean ± standard deviation of the mean. Figure 2. Comparison of acetic acid, butanoic acid, hexanoic acid, octanoic acid and n-decanoic acid amount between 12-months and 24-months samples. Results are presented in terms of mean ± standard deviation of the mean. Figure 2. Comparison of acetic acid, butanoic acid, hexanoic acid, octanoic acid and n-decanoic acid amount between 12-months and 24-months samples. Results are presented in terms of mean ± standard deviation of the mean. Figure 2. Comparison of acetic acid, butanoic acid, hexanoic acid, octanoic acid and n-decanoic acid amount between 12-months and 24-months samples. Results are presented in terms of mean ± standard deviation of the mean. Differences were also found in comparing samples with different percentages of rind and the same rind working process; the same trend is valid both for 12-months and 24-months ripened PR. It turned out that in increasing the quantity of rind, the presence of three compounds increases. Besides butanoic and hexanoic acid, 2-nonanone has the same behavior. It is a member of the class of methyl ketones and it can be found in several foods, such as milk and cheese [53]. It is produced by the oxidative degradation of fatty acids [54]. 3.1. GC-MS Analysis Results These results suggest that both the fermentation and the degradation happen closer to the rind than in the central part of the cheese. 3.2. S3 Analysis Results 2 Once data were acquired, sensors responses were checked ﬁrst. Since the ﬁrst measures of each session were very different from the others, they were discarded. Consequently, there is a different number of replicas for each sample. Most likely, experimental conditions of ﬁrst acquisitions were not the same as the following measures in terms of the temperature of the auto-sampler oven that the vials 8 of 15 s were hat the Sensors 2018, 18, 1617 number of replica t th th were put in, as explained in Section 2.3. S3 Analysis. In Table 2, a detailed description of the number of samples that were considered for the following analysis is shown. vials were put in, as explained in Section 2.3. S3 Analysis. In Table 2, a detailed description of the number of samples that were considered for the following analysis is shown. were put in, as explained in Section 2.3. S3 Analysis. In Table 2, a detailed description of the number of samples that were considered for the following analysis is shown. vials were put in, as explained in Section 2.3. S3 Analysis. In Table 2, a detailed description of the number of samples that were considered for the following analysis is shown. bl d d l d d d f d d d k Table 2. Considered samples divided for ripening stage, rind percentage and rind working processes (WR = washed-rind, SR = scraped-rind). SeasoningPercentage 0% 18% 26% 45% 100% WR SR WR SR WR SR WR SR 12 months - 11 12 13 11 12 14 - 14 24 months 12 14 14 13 13 11 13 13 - Table 2. Considered samples divided for ripening stage, rind percentage and rind working processes (WR = washed-rind, SR = scraped-rind). Seasoning\Percentage 0% 18% 26% 45% 100% WR SR WR SR WR SR WR SR 12 months - 11 12 13 11 12 14 - 14 24 months 12 14 14 13 13 11 13 13 - Table 2. Considered samples divided for ripening stage, rind percentage and rind working processes (WR = washed-rind, SR = scraped-rind). Table 2. Considered samples divided for ripening stage, rind percentage and rind working processes (WR = washed-rind, SR = scraped-rind). Seasoning\Percentage 0% 18% 26% 45% 100% The choice to extract ∆R/R0 as a feature was made after viewing the sensors responses. 3.2. S3 Analysis Results In Figure 3, the resistance value, as a function of time during the injection phase, is presented for four sensors that represent the four types of MOX in the S3. They are CuO, SnO2Au-RGTO, SnO2Au+Au-Nanowire and TGS2602. Since the starting point is equal for all the measures, the variation of normalized resistance exhibit that all the sensors are capable of distinguishing samples with different concentrations of rind (samples colors: red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45%). In addition, they show also the ability to recognize the two ripening degrees, characterized with a solid line for 24-months samples and with a dotted line for the others. Finally, responses to the working processes are highlighted using a thicker line for WR samples as compared to SR ones. / g p Figure 3, the resistance value, as a function of time during the injection phase, is presented for four sensors that represent the four types of MOX in the S3. They are CuO, SnO2Au-RGTO, SnO2Au+Au- Nanowire and TGS2602. Since the starting point is equal for all the measures, the variation of normalized resistance exhibit that all the sensors are capable of distinguishing samples with different concentrations of rind (samples colors: red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45%). In addition, they show also the ability to recognize the two ripening degrees, characterized with a solid line for 24-months samples and with a dotted line for the others. Finally, responses to the working processes are highlighted using a thicker line for WR samples as compared to SR ones. TSG2602 and SnO2Au+Au nanowires seem to be the best MOX to identify ripening and rind working process at ﬁxed concentration, since minimum resistance varies mostly for samples with the same rind percentage. Conversely, CuO and SnO2Au (RGTO) responses are more useful to recognize “pure” samples (0% and 100% of rind) from mixtures, since in the ﬁrst case ∆R is bigger. TSG2602 and SnO2Au+Au nanowires seem to be the best MOX to identify ripening and rind working process at fixed concentration, since minimum resistance varies mostly for samples with the same rind percentage. Conversely, CuO and SnO2Au (RGTO) responses are more useful to recognize “pure” samples (0% and 100% of rind) from mixtures, since in the first case ΔR is bigger. Figure 3. Cont. Figure 3. Cont. 3.2. S3 Analysis Results 9 of 15 Sensors 2018, 18, 1617 2018, 18, x igure 3. CuO, SnO2Au-RGTO, SnO2Au+Au-Nanowire and TGS2602 responses as functions of tim amples colors: Red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45%. So ne for 24-months samples and dotted line for 12-months. Finally, thicker lines represent WR and t thers SR. gure 3. CuO, SnO2Au-RGTO, SnO2Au+Au-Nanowire and TGS2602 responses as functions of tim amples colors: Red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45 olid line for 24-months samples and dotted line for 12-months. Finally, thicker lines represent WR a e others SR. Sensors 2018, 18, x Figure 3. CuO, SnO2Au-RGTO, SnO2Au+Au-Nanowire and TGS2602 responses as functions of time. Samples colors: Red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45%. Solid line for 24-months samples and dotted line for 12-months. Finally, thicker lines represent WR and the others SR. Figure 3. CuO, SnO2Au-RGTO, SnO2Au+Au-Nanowire and TGS2602 responses as functions of time. Samples colors: Red for 100% rind, green for 0%, blue for 18%, cyan for 26% and black for 45%. Solid line for 24-months samples and dotted line for 12-months. Finally, thicker lines represent WR and the others SR. 10 of 15 Sensors 2018, 18, 1617 10 of 15 In Figure 4, boxplots of TGS2602 response that include ∆R/R0 for each sample are shown. This sensor represents the general trend that can be observed in all the sensors. Obviously, since different sensing materials are used, there are differences in the highlighted groups that overlap. On the left part of the ﬁgure, there are 24-months seasoned samples, in the upper part, grated cheese with SR, while in the lower, PR with WR. In the right part, there are 12-months ripened samples, and they follow the same trend. The ﬁrst boxplot is relative to samples of 0% rind; its ∆R/R0 is different with respect to all the other groups, but it is more similar to WR grated PR, both at seasoning stage. This result reﬂects the fact that they are characterized by a greater amount of humidity. In Figure 4, boxplots of TGS2602 response that include ΔR/R0 for each sample are shown. This sensor represents the general trend that can be observed in all the sensors. 3.2. S3 Analysis Results Obviously, since different sensing materials are used, there are differences in the highlighted groups that overlap. On the left part of the figure, there are 24-months seasoned samples, in the upper part, grated cheese with SR, while in the lower, PR with WR. In the right part, there are 12-months ripened samples, and they follow the same trend. The first boxplot is relative to samples of 0% rind; its ΔR/R0 is different with respect to all the other groups, but it is more similar to WR grated PR, both at seasoning stage. This result reflects the fact that they are characterized by a greater amount of humidity. Figure 4. Boxplots of TGS2602 feature ΔR/R0. Four groups are highlighted: In blue, 24-months SR; in green, 24-months WR; in black, 12-months SR; and in orange, 12-months WR. Af h ki h l f JB li d h d O l 4 f Figure 4. Boxplots of TGS2602 feature ∆R/R0. Four groups are highlighted: In blue, 24-months SR; in green, 24-months WR; in black, 12-months SR; and in orange, 12-months WR. Figure 4. Boxplots of TGS2602 feature ΔR/R0. Four groups are highlighted: In blue, 24-months SR; in green, 24-months WR; in black, 12-months SR; and in orange, 12-months WR. Figure 4. Boxplots of TGS2602 feature ∆R/R0. Four groups are highlighted: In blue, 24-months SR; in green, 24-months WR; in black, 12-months SR; and in orange, 12-months WR. After checking the general sensors performances, JB test was applied to the dataset. Only 4 of the eight parameters followed a normal distribution (p < 0.05); they correspond to the features extracted by the two tin oxide nanowires and RGTO sensors. This was the main reason for choosing PLS. In Figure 5, PLS score plot was made, considering the first two latent variables (LV) for a total explained variance equal to 99.95% (99.87% for LV1 and 0.08% for LV2). The plot measures are divided by seasoning degree. It can be observed that the 24-months class is in the central part of the graph, while the other one is divided in the left and right part. For this reason, classification techniques were used in a hierarchical way. In addition, another i f hi h i i lif l ifi i d l i hi i 15 l bl H After checking the general sensors performances, JB-test was applied to the dataset. 3.2. S3 Analysis Results Only 4 of the eight parameters followed a normal distribution (p < 0.05); they correspond to the features extracted by the two tin oxide nanowires and RGTO sensors. This was the main reason for choosing PLS. In Figure 5, PLS score plot was made, considering the ﬁrst two latent variables (LV) for a total explained variance equal to 99.95% (99.87% for LV1 and 0.08% for LV2). The plot measures are divided by seasoning degree. It can be observed that the 24-months class is in the central part of the graph, while the other one is divided in the left and right part. motive for this choice was to simplify classification models, since this is a 15-class problem. Hence, in the first step, classifiers were used to distinguish the seasoning degree; in a second step, for each ripening state the different working processes were discriminated; finally, ring percentage was taken into account. In Figure 6, a scheme of the steps is shown. Regarding ANNs structures, three different ones were considered, one for each step. In the first case a two-layers architecture with 3 neurons in the input layer and 1 in the output layer was For this reason, classiﬁcation techniques were used in a hierarchical way. In addition, another motive for this choice was to simplify classiﬁcation models, since this is a 15-class problem. Hence, in the ﬁrst step, classiﬁers were used to distinguish the seasoning degree; in a second step, for each ripening state the different working processes were discriminated; ﬁnally, ring percentage was taken into account. In Figure 6, a scheme of the steps is shown. case, a two layers architecture with 3 neurons in the input layer and 1 in the output layer was considered. For the second stage, the same number of layers was used, but two neurons were put in the first one. Finally, the third ANN had the same structure as the previous ones, but with 6 neurons in the input layer. For all the neurons, hyperbolic tangent sigmoid transfer function was chosen. Regarding ANNs structures, three different ones were considered, one for each step. In the ﬁrst case, a two-layers architecture with 3 neurons in the input layer and 1 in the output layer was considered. For the second stage, the same number of layers was used, but two neurons were put in the ﬁrst one. 3.2. S3 Analysis Results In Table 3, overall classification rate of the two classifiers is put side by side. In general, AN Figure 6. Step-by-step scheme for classification analysis. Figure 6. Step-by-step scheme for classiﬁcation analysis. Figure 6. Step-by-step scheme for classification analysis. In Table 3, overall classification rate of the two classifiers is put side by side. In general, ANN Figure 6. Step-by-step scheme for classification analysis. Figure 6. Step-by-step scheme for classiﬁcation analysis. g p y p y classification rate of the two classifiers is put side by Figure 6. Step-by-step scheme for classification analysis. Figure 6. Step-by-step scheme for classiﬁcation analysis. classification rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the samples based on seasoning and rind working processes. PLS-DA performances are lower, although it can reach good classification rates. The distinction between rind percentage shows that both classifiers can classify samples with SR better than those with WR. A possible explanation for this result could be the different amount of humidity: WR samples have a higher content of humidity because of water treatment and this could cause the occupation of the adsorption sites by water molecules instead of the ones that characterize the volatile fingerprint of the samples. In Table 3, overall classification rate of the two classifiers is put side by side. In general, ANN classification rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the samples based on seasoning and rind working processes. PLS-DA performances are lower, although it can reach good classification rates. The distinction between rind percentage shows that both classifiers can classify samples with SR better than those with WR. A possible explanation for this result could be the different amount of humidity: WR samples have a higher content of humidity because of water treatment and this could cause the occupation of the adsorption sites by water molecules instead of the ones that characterize the volatile fingerprint of the samples. In Table 3, overall classiﬁcation rate of the two classiﬁers is put side by side. In general, ANN classiﬁcation rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the samples based on seasoning and rind working processes. PLS-DA performances are lower, although it can reach good classiﬁcation rates. 3.2. S3 Analysis Results Finally, the third ANN had the same structure as the previous ones, but with 6 neurons in the input layer. For all the neurons, hyperbolic tangent sigmoid transfer function was chosen. 11 of 15 11 of 15 Sensors 2018, 18, 1617 sors 2018, 18, x 11 of 15 Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Figure 6. Step-by-step scheme for classification analysis. In Table 3, overall classification rate of the two classifiers is put side by side. In general, ANN ssification rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the mples based on seasoning and rind working processes. PLS-DA performances are lower, although can reach good classification rates. The distinction between rind percentage shows that both ssifiers can classify samples with SR better than those with WR. A possible explanation for this ult could be the different amount of humidity: WR samples have a higher content of humidity cause of water treatment and this could cause the occupation of the adsorption sites by water olecules instead of the ones that characterize the volatile fingerprint of the samples. Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in ﬁrst two LV. sors 2018, 18, x 11 of 1 Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Figure 6. Step-by-step scheme for classification analysis. In Table 3, overall classification rate of the two classifiers is put side by side. In general, ANN ssification rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the mples based on seasoning and rind working processes. PLS-DA performances are lower, although can reach good classification rates. The distinction between rind percentage shows that both ssifiers can classify samples with SR better than those with WR. 3.2. S3 Analysis Results A possible explanation for thi ult could be the different amount of humidity: WR samples have a higher content of humidity cause of water treatment and this could cause the occupation of the adsorption sites by wate olecules instead of the ones that characterize the volatile fingerprint of the samples. Figure 6. Step-by-step scheme for classiﬁcation analysis. In Table 3, overall classiﬁcation rate of the two classiﬁers is put side by side. In general, ANN ssiﬁcation rates are better than those of PLS-DA. Indeed, ANN is able to recognize correctly all the mples based on seasoning and rind working processes. PLS-DA performances are lower, although i n reach good classiﬁcation rates. The distinction between rind percentage shows that both classiﬁer n classify samples with SR better than those with WR. A possible explanation for this result could be e different amount of humidity: WR samples have a higher content of humidity because of wate atment and this could cause the occupation of the adsorption sites by water molecules instead o h h i h l il ﬁ i f h l so s 0 8, 8, o Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in ﬁrst two LV. nsors 2018, 18, x 11 of 1 Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Sensors 2018, 18, x 11 of Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in ﬁrst two LV. Figure 5. PLS score plot for all the measures divided by seasoning degree: in red circle, 12-months; in green square, 24-months. Total explained variance equal to 99.95% in first two LV. Figure 6. Step-by-step scheme for classification analysis. Author Contributions: Conceptualization, M.A. and V.S.; Methodology, M.A. and E.N.-C.; Software, M.A.; Validation, M.A., E.N.-C. and V.S.; Formal Analysis, M.A.; Investigation, M.A.; Resources, E.N.-C., D.Z., V.S., E.C. and G.S.; Data Curation, M.A.; Writing-Original Draft Preparation, M.A.; Writing-Review & Editing, V.S., E.C. and G.S.; Visualization, M.A.; Supervision, V.S. and E.C.; Project Administration, V.S.; 3.2. S3 Analysis Results The distinction between rind percentage shows that both classiﬁers can classify samples with SR better than those with WR. A possible explanation for this result could be the different amount of humidity: WR samples have a higher content of humidity because of water treatment and this could cause the occupation of the adsorption sites by water molecules instead of the ones that characterize the volatile ﬁngerprint of the samples. 12 of 15 Sensors 2018, 18, 1617 Table 3. Classiﬁcation rates of Partial Least Squares Discriminant Analysis (PLS-DA) and Artiﬁcial Neural Networks (ANNs) divided per steps. First Step Ripening Stage Second Step Working Processes Third Step Rind Percentage PLD-DA 94.7% 12 months: 100% WR: 61.1% SR: 90.2% 24 months: 79% WR:90.2% SR: 95% ANN 100% 12 months: 100% WR: 63.8% SR: 96.1% 24 months: 100% WR: 58.8% SR: 100% Table 3. Classiﬁcation rates of Partial Least Squares Discriminant Analysis (PLS-DA) and Artiﬁcial Neural Networks (ANNs) divided per steps. Table 3. Classiﬁcation rates of Partial Least Squares Discriminant Analysis (PLS-DA) and Artiﬁcial Neural Networks (ANNs) divided per steps. To the author’s knowledge, only few researches have been carried out regarding rind composition of grated cheeses. The preparation of the samples in some works was carried out through the grating process, although the aim was to classify the different varieties of cheeses, like Swiss [55] or Emmental cheese [56]. For the latter, it has been tried unsuccessfully to ﬁnd the “rind-taste” off-ﬂavor [57]; in this case, the lack of positive results could be due to non-volatile compounds that change only the taste but not the aroma. As regards PR, cheese aroma authenticity and rind percentage recognition have been achieved with an electronic nose equipped with SnO2 and ZnO sensors made at SENSOR Laboratory of University of Brescia [58]. In this case, the tool was also able to distinguish samples with little differences in terms of rind percentage, such as 18% and 19%. However, unlike this study, only one type of ripening was considered (12-months) and the different working processes were not taken into consideration. Finally, the comparison with this study allows for the assessment of the utility of S3 for fraud detection, since the results point in the same direction. Acknowledgments: Authors are grateful to Parmigiano Reggiano Consortium (CFPR) for giving us sample Conﬂicts of Interest: The authors declare no conﬂict of interest. Acknowledgments: Authors are grateful to Parmigiano Reggiano Consortium (CFPR) for giving us samples. Conﬂicts of Interest: The authors declare no conﬂict of interest. Acknowledgments: Authors are grateful to Parmigiano Reggiano Consortium (CFPR) for giving us samples. ﬂ f ﬂicts of Interest: The authors declare no conﬂict of interest. 4. Conclusions This study aimed at verifying the possible distinction between grated PR with different rind percentages with an electronic nose, taking into account two other variables: seasoning degree and working processes of rind. In parallel, a consolidated technique, i.e., SPME GC-MS, has been used to understand which VOCs characterized analyzed samples. This combined analysis has produced promising results that pave the way to assess cheese quality and avoid frauds. First of all, with GC-MS, the VOCs that characterize grated cheeses have been individuated. The results concerning PR are compliant with those found in the literature. Indeed, fatty acids that describe the aroma and taste proﬁle of PR have been found in greater quantity for 24-months seasoned samples as compared to 12-months ones. In addition, VOCs, whose amount is bigger in rind compared to pulp, were found and they are acquiescent with chemical reactions that take place in this product. The multivariate statistical analysis made with PLS indicated how to proceed during the classification stage. A hierarchical approach was used, both for PLS-DA and ANNs. ANNs classification rates are the highest, suggesting that in future they could be improved to increase their performances. These first results are encouraging, and further research is in progress to add more samples and to acquire greater statistical significance for the achieved results. Author Contributions: Conceptualization, M.A. and V.S.; Methodology, M.A. and E.N.-C.; Software, M.A.; Validation, M.A., E.N.-C. and V.S.; Formal Analysis, M.A.; Investigation, M.A.; Resources, E.N.-C., D.Z., V.S., E.C. and G.S.; Data Curation, M.A.; Writing-Original Draft Preparation, M.A.; Writing-Review & Editing, V.S., E.C. and G.S.; Visualization, M.A.; Supervision, V.S. and E.C.; Project Administration, V.S.; Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 13 of 15 13 of 15 Sensors 2018, 18, 1617 References 1. Zannoni, M. Evolution of the sensory characteristics of Parmigiano-Reggiano cheese to the present day. Food Qual. Preference 2010, 21, 901–905. [CrossRef] 2. Caligiani, A.; Nocetti, M.; Lolli, V.; Marseglia, A.; Palla, G. Development of a Quantitative GC-MS Method for the Detection of Cyclopropane Fatty Acids in Cheese as New Molecular Markers for Parmigiano Reggiano Authentication. J. Agric. Food Chem. 2016, 64, 4158–4164. [CrossRef] [PubMed] 3. Gazzetta Ufﬁiale. Available online: www.gazzettaufﬁciale.it/atto/serie_generale/caricaArticolo? art.progressivo=0&art.idArticolo=1&art.versione=1&art.codiceRedazionale=16A06126&art. dataPubblicazioneGazzetta=2016-08-22&art.idGruppo=1&art.idSottoArticolo1=10&art.idSottoArticolo= 1&art.ﬂagTipoArticolo=1 (accessed on 27 March 2018). 4. Bottari, B.; Quartieri, A.; Prandi, B.; Raimondi, S.; Leonardi, A.; Rossi, M.; Ulrici, A.; Gatti, M.; Sforza, S.; Nocetti, M.; et al. Characterization of the peptide fraction from digested Parmigiano Reggiano cheese and its effect on growth of lactobacilli and biﬁdobacteria. Int. J. Food Microbiol. 2017, 255, 32–41. [CrossRef] [PubMed] 5. Malacarne, M.; Summer, A.; Formaggioni, P.; Franceschi, P.; Sandri, S.; Pecorari, M.; Vecchia, P.; Mariani, P. Dairy maturation of milk used in the manufacture of Parmigiano-Reggiano cheese: Effects on physico-chemical characteristics, rennet-coagulation aptitude and rheological properties. J. Dairy Res. 2008, 75, 218–224. [CrossRef] [PubMed] 6. Sberveglieri, V.; Bhandari, M.P.; Carmona, E.N.; Betto, G.; Sberveglieri, G. A novel MOS nanowire gas sensor device (S3) and GC-MS-based approach for the characterization of grated parmigiano reggiano cheese. Biosensors 2016, 6, 60. [CrossRef] [PubMed] 7. Popping, B.; De Dominicis, E.; Dante, M.; Nocetti, M. Identiﬁcation of the Geographic Origin of Parmigiano Reggiano (P.D.O.) Cheeses Deploying Non-Targeted Mass Spectrometry and Chemometrics. Foods 2017, 6, 13. [CrossRef] [PubMed] 8. Tipologia Grattugiato. Available online: https://www.parmigiano-reggiano.it/consorzio/disciplinare_ produzione/tipologia_grattugiato.aspx (accessed on 17 May 2018). 9. Maarse, H.; Visscher, C.A. Volatile Compounds in Foods: Qualitative and Quantitative Data; TNO-CIVO Food Analysis Institute: Zeist, The Netherlands, 1989. 10. Curioni, P.M.G.; Bosset, J.O. Key odorants in various cheese types as determined by gas chromatography- olfactometry. Int. Dairy J. 2002, 12, 959–984. [CrossRef] y y 11. Moio, L.; Addeo, F. Grana Padano cheese aroma. J. Dairy Res. 2000, 65, 317–333. [CrossRef] 12. Qian, M.; Reineccius, G. Identiﬁcation of Aroma Compounds in Parmigiano-Reggiano Cheese by Gas Chromatography/Olfactometry. J. Dairy Sci. 2002, 85, 1362–1369. [CrossRef] 3. Carmona, E.N.; Sberveglieri, V.; Comini, E.; Zappa, D.; Pulvirenti, A. Nanowire technology for the detec of microorganisms in potable water. Procedia Eng. 2014, 87, 1453–1456. [CrossRef] 4. Wilson, A. D Review of Electronic-nose Technologies and Algorithms to Detect Hazardous Chemicals in Environment. Procedia Technol. 2012, 1, 453–463. [CrossRef] 15. Wilson, A.D.; Baietto, M. Applications and Advances in Electronic-Nose Technologies. References Sensors 2009, 9, 5099–5148. [CrossRef] [PubMed] 16. Muñoz, R.; Sivret, E.C.; Parcsi, G.; Lebrero, R.; Wang, X.; Suffet, I.H.; Stuetz, R.M. Monitoring techniques for odour abatement assessment. Water Res. 2010, 44, 5129–5149. [CrossRef] [PubMed] 17. G˛ebicki, J.; Dymerskib, T.; Namie´snik, J. Monitoring of Odour Nuisance from Landﬁll Using Electronic Nose. Chem. Eng. Trans. 2014, 40, 85–90. [CrossRef] 18. D’Amico, A.; Pennazza, G.; Santonico, M.; Martinelli, E.; Roscioni, C.; Galluccio, G.; Paolesse, R.; Di Natale, C. An investigation on electronic nose diagnosis of lung cancer. Lung Cancer 2010, 68, 170–176. [CrossRef] [PubMed] 19. Dutta, R.; Morgan, D.; Baker, N.; Gardner, J.W.; Hines, E.L. Identiﬁcation of Staphylococcus aureus infections in hospital environment: Electronic nose based approach. Sens. Actuators B Chem. 2005, 109, 355–362. [CrossRef] 20. Pavlou, A.K.; Magan, N.; McNulty, C.; Jones, J.M.; Sharp, D.; Brown, J.; Turner, A.P.F. Use of an electronic nose system for diagnoses of urinary tract infections. Biosens. Bioelectron. 2002, 17, 893–899. [CrossRef] 14 of 15 14 of 15 Sensors 2018, 18, 1617 21. Kodogiannis, V.S.; Lygouras, J.N.; Tarczynski, A.; Chowdrey, H.S. Artiﬁcial Odor Discrimination System Using Electronic Nose and Neural Networks for the Identiﬁcation of Urinary Tract Infection. IEEE Trans. Inf. Technol. Biomed. 2008, 12, 707–713. [CrossRef] [PubMed] 22. Covington, J.A.; Westenbrink, E.W.; Ouaret, N.; Harbord, R.; Bailey, C.; O’Connel, N.; Cullis, J.; Williams, N.; Nwokolo, C.; Bardhan, K.D.; et al. Application of a Novel Tool for Diagnosing BIle Acid Diarrhoea. Sensors 2013, 13, 11899–11912. [CrossRef] [PubMed] 23. Loutﬁ, A.; Coradeschi, S.; Mani, G.K.; Shankar, P.; Rayappan, J.B.B. Electronic Noses for Food Quality: A Review. J. Food Eng. 2014, 144, 103–111. [CrossRef] 24. Macias, M.M.; Agudo, J.E.; Manso, A.G.; Orellana, C.J.G.; Velasco, H.M.G.; Caballero, R.G. A Compact and Low Cost Electronic Nose for Aroma Detection. Sensors 2013, 13, 5528–5541. [CrossRef] [PubMed] 25. Wojnowski, W.; Majchrzak, T.; Dymerski, T.; G˛ebicki, J.; Namie´snik, J. Portable Electronic Nose Based on Electrochemical Sensors for Food Quality Assessment. Sensors 2017, 17, 2715. [CrossRef] [PubMed] 26. Sberveglieri, V.; Falasconi, M.; Gobbi, E.; Carmona, E.N.; Zambotti, G.; Pulvirenti, A. Candida milleri detected by Electronic noce in tomato sauce. Procedia Eng. 2014, 87, 584–587. [CrossRef] 27. Sberveglieri, V.; Comini, E.; Zappa, D.; Pulvirenti, A.; Carmona, E.N. Electronic nose for the early detection of different types of indigenous mold contamination in green coffee. In Proceedings of the 2013 Seventh International Conference on Sensing Technology (ICST), Wellington, New Zealand, 3–5 December 2013; pp. 461–465. [CrossRef] 28. References [CrossRef] [PubMed] 15 of 15 15 of 15 Sensors 2018, 18, 1617 42. Lenhardt, L.; Bro, R.; Zekovi´c, I.; Drami´canin, T.; Drami´canin, M.D. Fluorescence spectroscopy coupled with PARAFAC and PLS DA for characterization and classiﬁcation of honey. Food Chem. 2015, 175, 284–291. [CrossRef] [PubMed] 43. Cozzolino, D.; Smyth, H.E.; Cynkar, W.; Dambergs, R.G.; Gishen, M. Usefulness of chemometrics and mass spectrometry-based electronic nose to classify Australian white wines by their varietal origin. Talanta 2005, 68, 382–387. [CrossRef] [PubMed] 44. Consonni, R.; Cagliani, L.R. Ripening and geographical characterization of Parmigiano Reggiano cheese by 1H NMR spectroscopy. Talanta 2008, 76, 200–205. [CrossRef] [PubMed] 45. Wang, S.C. Artiﬁcial Neural Network. In Interdisciplinary Computing in Java Programming; The Springer International Series in Engineering and Computer Science; Springer: Boston, MA, USA, 2003; Volume 743. 46. Yu, H.; Wang, J.; Yao, C.; Zhang, H.; Yu, Y. Quality grade identiﬁcation of green tea using E-nose by CA and ANN. LWT Food Sci. Technol. 2008, 41, 1268–1273. [CrossRef] 47. Berrueta, L.A.; Alonso-Salces, R.M.; Heberger, K. Supervised pattern recognition in food analysis. J. Chromatogr. A 2007, 1158, 196–214. [CrossRef] [PubMed] 48. Tan, J.; Kerr, W. L Determining Degree of Roasting in Cocoa Beans by Artiﬁcial Neural Network (ANN) based Electronic Nose System and Gas Chromatography/Mass Spectrometry (GC/MS). J. Sci. Food Agric. 2018. [CrossRef] [PubMed] 49. Adak, M.F.; Yumusak, N. Classiﬁcation of E-Nose Aroma Data of Four Fruit Types by ABC-Based Neural Network. Sensors 2016, 16, 304. [CrossRef] [PubMed] 50. Ballabio, D.; Consonni, V. Classiﬁcation tools in chemistry. Part 1: Linear models. PLS-DA. Anal. Methods 2013, 5, 3790–3798. [CrossRef] 51. Tosi, F.; Sandri, S.; Tedeschi, G.; Malacarne, M.; Fossa, E. Variazioni di composizione e prprietà ﬁsico-chimiche del Parmigiano Reggiano durante la maturazione e in differenti zone della forma. Sci. Tec. Latt. Casearia 2008, 59, 507–528. 52. Nanni, M.; Coppola, R.; Iorizzo, M.; Sorrentino, A.; Sorrentino, E.; Grazia, L. La microﬂora lattica nella maturazione del formaggio Parmigiano Reggiano. Sci. Tec. Latt. Casearia 1997, 48, 211–216. 53. 2-Nonanone. Available online: https://pubchem.ncbi.nlm.nih.gov/compound/2-nonanone (accessed on 29 March 2018). 54. Jollivet, N.; Chataud, J.; Vayssier, Y.; Bensoussan, M. Production of volatile compounds in model milk and cheese media by eight strains of Geotrichum candidum. J. Dairy Res. 1994, 61, 241–248. [CrossRef] 55. Jou, K.D.; Harper, W.J. Pattern recognition of Swiss cheese aroma compounds by SPME/GC and an electronic nose. Milchwissenschaft 1998, 53, 259–263. 56. Pillonel, L.; Ampuero, S.; Tabacchi, R.; Bosset, J.O. References Labreche, S.; Bazzo, S.; Cade, S.; Chanie, E. Shelf life determination by electronic nose: Application to milk. Sens. Actuators B 2005, 106, 19–206. [CrossRef] 29. Qiu, S.; Wang, J. The prediction of food additives in the fruit juice based on electronic nose with chemometrics. Food Chem. 2017, 230, 208–214. [CrossRef] [PubMed] 30. Baietto, M.; Wilson, A.D. Electronic-Nose Applications for Fruit Identiﬁcation, Ripeness and Quality Grading. Sensors 2015, 15, 899–931. [CrossRef] [PubMed] 31. Gursoy, O.; Somervuo, P.; Alatossava, T. Preliminary study of ion mobility based electronic nose GD-1 for discrimination of hard cheeses. J. Food Eng. 2009, 92, 202–207. [CrossRef] 32. Pais, V.F.; Oliveira, J.A.B.P.; Gomes, M.T.S.R. An Electronic Nose Based on Coated Piezoelectric Quartz Crystals to Certify Ewes’ Cheese and to Discriminate between Cheese Varieties. Sensors 2012, 12, 1422–1436. [CrossRef] [PubMed] 33. Marsili, R.T. Comparison of solid-phase microextraction and dynamic headspace methods for the gas chromatographic-mass spectrometric analysis of light-induced lipid oxidation products in milk. J. Chromatogr. Sci. 1999, 37, 17–23. [CrossRef] [PubMed] 34. Shooter, D.; Jayatissa, N.; Renner, N. Volatile reduced sulphur compounds in butter by solid phase microextraction. J. Dairy Res. 1999, 66, 115–123. [CrossRef] 35. Frank, D.C.; Owen, C.M.; Patterson, J. Solid phase microextraction (SPME) combined with gas-chromatography and olfactometry-mass spectrometry for characterization of cheese aroma compounds. LWT Food Sci. Technol. 2004, 37, 139–154. [CrossRef] 36. Peres, C.; Viallon, C.; Berdague, J. Solid-Phase Microextraction-Mass Spectrometry: A New Approach to the Rapid Characterization of Cheeses. Anal. Chem. 2001, 73, 1030–1036. [CrossRef] [PubMed] 37. Ponzoni, A.; Zappa, D.; Comini, E.; Sberveglieri, V.; Faglia, G.; Sberveglieri, G. Metal oxide nanowire gas sensors: Application of conductometric and surface ionization architectures. Chem. Eng. Trans. 2012, 30, 31–36. [CrossRef] 38. Sberveglieri, G.; Concina, I.; Comini, E.; Falasconi, M.; Ferroni, M.; Sberveglieri, V. Synthesis and integration of tin oxide nanowires into an electronic nose. Vacuum 2012, 86, 532–535. [CrossRef] 39. Wagner, R.S.; Ellis, W.C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth. Appl. Phys. Lett. 1964, 4, 89–90. [CrossRef] 40. Carmona, E.N.; Sberveglieri, V.; Ponzoni, A.; Zappa, D.; Pulvirenti, A. Detection of microbial contamination in potable water by Nanowire technology. In Proceedings of the 8th International Conference on Sensing Technology, Liverpool, UK, 2–4 September 2014. 41. Botelho, B.G.; Reis, N.; Oliveira, L.S.; Sena, M.M. Development and analytical validation of a screening method for simultaneous detection of ﬁve adulterants in raw milk using mid-infrared spectroscopy and PLS-DA. Food Chem. 2015, 181, 31–37. References Analytical methods for the determination of the geographic origin of Emmental cheese, Volatile compounds by GC-MS-FID and electronic nose. Eur. J. Food Res. Technol. 2003, 216, 179–183. [CrossRef] 57. Schaller, E.; Bosset, J.O.; Escher, F. Feasibility study: Detection of rind taste off-ﬂavour in Swiss Emmental cheese using an electroinc nose and a GC-MS. Mitt. Lebensm. Hyg. 2000, 91, 610–615. 58. Sberveglieri, V. Validation of Parmigiano Reggiano Cheese Aroma Authenticity, Categorized through the Use of an Array of Semiconductors Nanowire Device (S3). Materials 2016, 9, 81. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4386135238	https://www.researchsquare.com/article/rs-3279400/latest.pdf	English	null	Comparison of temperature and renal tissue thermal damage by Holmium laser with different energy parameters during lithotripsy: in vitro porcine kidney model	Research Square (Research Square)	2,023	cc-by	4,777	Page 1/13 Comparison of temperature and renal tissue thermal damage by Holmium laser with different energy parameters during lithotripsy: in vitro porcine kidney model Wei Wei Guangzhou University of Chinese Medicine Ming Chen Guangzhou University of Chinese Medicine Le Xie Guangzhou University of Chinese Medicine Yuan Mai Guangzhou University of Chinese Medicine Huacai Zhu Guangzhou University of Chinese Medicine Zhanping Xu Guangzhou University of Chinese Medicine Research Article Keywords: holmium laser, lithotripsy, thermal damage, intrarenal temperature Posted Date: August 24th, 2023 DOI: https://doi.org/10.21203/rs.3.rs-3279400/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Additional Declarations: No competing interests reported. Version of Record: A version of this preprint was published at International Urology and Nephrology on March 14th, 2024. See the published version at https://doi.org/10.1007/s11255-024-03943-8. Comparison of temperat thermal damage by Holm energy parameters durin kidney model Wei Wei Guangzhou University of Chinese Medicine Ming Chen Guangzhou University of Chinese Medicine Le Xie Guangzhou University of Chinese Medicine Yuan Mai Guangzhou University of Chinese Medicine Huacai Zhu Guangzhou University of Chinese Medicine Zhanping Xu Guangzhou University of Chinese Medicine Research Article Keywords: holmium laser, lithotripsy, thermal da Posted Date: August 24th, 2023 DOI: https://doi.org/10.21203/rs.3.rs-3279400/v License:   This work is licensed under a Crea Read Full License Additional Declarations: No competing interests Version of Record: A version of this preprint was March 14th, 2024. See the published version at h Research Article Additional Declarations: No competing interests reported. Version of Record: A version of this preprint was published at International Urology and Nephrology on March 14th, 2024. See the published version at https://doi.org/10.1007/s11255-024-03943-8. Page 1/13 Page 1/13 Conclusion The higher power and longer time of holmium laser excitation will raise the intrarenal temperature significantly and cause a certain degree of thermal damage to the kidney tissue, but overall it is safe and reliable, and urologists can avoid more side damage through surgical experience skills. The holmium laser percutaneous nephrolithotripsy was simulated by porcine kidney calculus model in vitro to investigate the thermal damage of renal tissue by holmium laser with different energy parameters The holmium laser percutaneous nephrolithotripsy was simulated by porcine kidney calculus model in vitro to investigate the thermal damage of renal tissue by holmium laser with different energy parameters. RESULTS The four independent models were lithotripsy with 30W and 60W laser for 5 and 10 minutes, respectively; the mean temperature of 30W vs. 60W within 5 minutes was 36.06°C vs. 39.21°C (t = 5.36, P = 0.00) and the highest temperature was 43.60°C vs. 46.60°C; the mean temperature of 30W vs. 60W within 10 minutes was 37.91°C vs. 40.13 ℃ (t = 5.28, P = 0.00), maximum temperature 46.80℃ vs. 49.20℃. Pathologically, each kidney was observed to have different degrees of thermal injury lesions, and the higher the power and longer the time the more severe the injury, but the injury was mainly limited to the uroepithelial and subepithelial tissues, with rare damage to the renal tubules, etc. Page 1/13 Page 2/13 METHODS We placed human kidney calculus specimen in a fresh vitro porcine kidney, then insert thermocouple temperature probes into the submucosa of the renal pelvis and rewarmed in a 37°C water bath. The renal parenchyma was penetrated with a percutaneous nephrological sheath through moderate irrigation rate of 30 ml/s in 18℃, and the Holmium laser was used to fragment the stone under the nephroscope and record the temperature. 2.1 Experimental model Kidney model: fresh porcine kidney was obtained directly from the meat plant, stored in frozen transit at -4 ~ 0℃, and used for the study within 1 hour after isolation. Porcine kidney morphology and size were basically the same, no other lesions, perirenal fat, envelope were trimmed and cleaned before the experiment operating. Calculus: human urinary calculus after harmless treatment were used for the study. The main components are calcium oxalate dihydrate and calcium oxalate pentahydrate, the most common ones. The calculus were removed from the kidney or ureter after incisional stone-extraction procedures. Patients signed an informed consent form agreeing to the use of calculus specimens for this study. The calculus were broken down and polished to approximately 2-2.5 cm, inserted into the incised porcine renal pelvis. Then the assembly system were stitched with silk thread.(Fig. 1a) 1. Background The incidence of urinary calculus is gradually increase in the world, and some complex calculus often require surgical intervention. With the continuous development of minimally invasive surgery, the treatment of upper urinary tract stones has changed from open surgery to endoluminal surgery. Percutaneous nephrolithotomy (PCNL) and flexible ureteroscopy (FURS) have been widely performed in clinical practice because of their less invasive, precise efficacy and maturity of application [1]. Also driving the progress of minimally invasive surgery are the lumpectomy energy output modalities that Page 3/13 have kept pace with the times. Lasers have been formally used in medicine since the 1980s, and the unique physical effect of the holmium laser on human calculus makes it highly efficient for lithotripsy. Thought it has been shown that thulium fiber laser is not inferior for this purpose, holmium laser is still a state-of-the-art device for endoscopic lithotripsy.[2] Holmium laser is the most commonly used laser in urology surgery, with its outstanding advantages about 95% of its energy absorbed by the surrounding 5mm water medium and good directivity[3]. While the gradual popularity of holmium laser, we also found that after holmium laser lithotripsy, patients have higher frequency of complications such as ureteral and pelvic stenosis adhesions or even impaired renal function [4], which increases the medical and economic burden on individuals and society. The causes of these are not only the urinal infection and operation skill, but the thermal effect of holmium laser. Most of the previous thermal studies of holmium laser did not consider the absorption of laser energy by calculus, nor did they discuss in depth from the pathological changes of tissues, so we need more detailed experiment to simulate the reality.We performed the study with reference to various sources and summarized the experience of existing studies, and used vitro porcine kidney model to investigate the thermal impact caused by holmium laser at different energies for a certain period of time. 2.2 Equipment The equipment included holmium laser lithotripsy system (Pudong Optoelectronics, Shenzhen, China), endoscopic monitoring system (Wolf, Germany), nephroscope (F9.5, Wolf, Germany), renal puncture sheath (F20, Veili Medical, Guangzhou, China), thermostatic water bath system (China), thermocouple thermometer and probe (Suzhou TASI Electronics Co.,Ltd, TA612 Series,Jiangsu,China), lavage pump. (Fig. 1b) 2.3 Procedure (3) Experiment groups: The study was divided into 4 groups with different laser parameters: Group A setting 30W (2J, 15Hz), for 5 minutes; Group B setting 60W (3J, 20Hz), 5 minutes; Group C setting 30W (1J, 30Hz), 10 minutes; Group D setting 60W (2J, 30Hz), 10 minutes. (3) Experiment groups: The study was divided into 4 groups with different laser parameters: Group A setting 30W (2J, 15Hz), for 5 minutes; Group B setting 60W (3J, 20Hz), 5 minutes; Group C setting 30W (1J, 30Hz), 10 minutes; Group D setting 60W (2J, 30Hz), 10 minutes. (4) After reaching the proposed time, lithotripsy stopped, the model kidney was dissected and histopathological changes in the lithotripsy area were observed visually. The tissue at the site of possible thermal injury was taken and fixed in formalin. Microscopic pathology was observed after preparation of the films. 2.4 Data Statistics Temperature recording software was provided by TASI, Inc. with a recording interval of 5 seconds. Two probe temperature data within the same kidney model were recorded by taking the average value after data export. All data were imported into Microsoft Excel and SPSS 22.0 to produce graphs. The mean temperature of each group was compared, and the difference was examined by t test, the calibration level α = 0.05. 2.3 Procedure Page 4/13 (1) A fresh vitro porcine kidneys combined with urinary calculus model were prepared, and two thermocouple temperature probes were inserted into the kidney tissue in the proposed lithotripsy area, with The probe phase distance is 1cm, the probes were placed as close as possible to the urothelial layer of the collecting system, about 5 mm away from the epithelial tissue, avoiding penetration into the renal pelvis cavity. (1) A fresh vitro porcine kidneys combined with urinary calculus model were prepared, and two thermocouple temperature probes were inserted into the kidney tissue in the proposed lithotripsy area, with The probe phase distance is 1cm, the probes were placed as close as possible to the urothelial layer of the collecting system, about 5 mm away from the epithelial tissue, avoiding penetration into the renal pelvis cavity. (1) A fresh vitro porcine kidneys combined with urinary calculus model were prepared, and two thermocouple temperature probes were inserted into the kidney tissue in the proposed lithotripsy area, with The probe phase distance is 1cm, the probes were placed as close as possible to the urothelial layer of the collecting system, about 5 mm away from the epithelial tissue, avoiding penetration into the renal pelvis cavity. (2) The porcine kidneys were rewarmed in a 37℃ thermostatic saline bath, experiment was started by monitoring the kindey temperature about to 37℃. The renal pelvis was punctured with F20 nephroscopic sheath, and the nephroscopic combined with holmium laser was used to smash the stones in kidney. Machine controlled continuous saline irrigation was used with a uniform moderate rate of 30 ml/s and an irrigation fluid temperature of 18°C (room temperature). In combination with urologists’ surgical habits, the holmium laser was excitated in a trial by the same surgeon for 20-25s per continuous relay with a 5s pause. (2) The porcine kidneys were rewarmed in a 37℃ thermostatic saline bath, experiment was started by monitoring the kindey temperature about to 37℃. The renal pelvis was punctured with F20 nephroscopic sheath, and the nephroscopic combined with holmium laser was used to smash the stones in kidney. Machine controlled continuous saline irrigation was used with a uniform moderate rate of 30 ml/s and an irrigation fluid temperature of 18°C (room temperature). In combination with urologists’ surgical habits, the holmium laser was excitated in a trial by the same surgeon for 20-25s per continuous relay with a 5s pause. 3. Results Two kidney models were used in each group for two independent trail. The maximum temperature of Group A for two trials was 43.60℃, the mean temperature was 36.01℃ and 36.10℃, respectively. While the maximum temperature of Group B was 46.60℃, the mean temperature was 39.39℃ and 39.02℃. The temperature profiles of Group A/B are shown in Fig. 2a,2b. After data consolidation, the mean temperature in Group B was higher(39.21°C vs. 36.06°C, t = 5.36, P = 0.00), and the heat generated by the 60W laser during the 5-minute laser lithotripsy had a greater effect on the elevated kidney temperature. Holmium laser excitation for 10 minutes in Group C was compared with Group D. The maximum temperature of Group C was 46.8°C with mean temperatures of 37.56°C and 38.25°C in both trials. While the maximum temperature of Group D was 49.2°C with mean temperatures of 40.08°C and Page 5/13 Page 5/13 40.18°C,respectively (Fig. 2c,2d). Group D with 60W laser for 10 minutes, compared to Group C 30W, had more significant elevation of renal parenchymal temperature and may have caused thermal damage to the tissue (40.13°C vs. 37.91°C, t = 5.28, P = 0.00). We dissected each kidney after laser excitation. For each kidney we took 3–5 pieces of tissue for pathological observation, near the thermocouple probe. Mild to moderate thermal injury was more common in Group A and C. Microscopically, the mucosal epithelium of the renal pelvis was detached to varying degrees, with some of the superficial umbrella cells only missing and some of the whole layer missing, exposing the bare basement membrane and edema of the mucosal lamina propria (Fig. 3a, 3b). More severe thermal injury was seen in specimens from groups B and D. The mucosal epithelium was degenerated and detached, with edema of the lamina propria and varying degrees of degeneration and necrosis of the mucosal muscle, myocytes and collagen fibers of the outer membrane (Fig. 3c). 3. Results We classified the damage from the 53 pathological sections produced, defining 19 sections with severe thermal damage, 2 in Group A, 4 in Group B, 5 in Group C and 8 in Group D, respectively; 24 sections with moderate thermal damage, 8 in Group A, 6 in Group B, 6 in Group C and 4 in Droup D, respectively; and 10 sections with mild damage, 3 in Group A, 3 in Group B, 2 in Group C and 2 in Group D. We ranked the proportion of severe thermal injuries from highest to lowest as Group D (57.1%), Group C (38.5%), Group B (30.8%), and Group A (15.4%). 4. Discussion Holmium laser lithotripsy has become the best method for treatment of urolithiasis. Although previous researchers have conducted some similar experiments in vitro and in vivo. However, most of the previous experiments on isolated animal organs or equipment were only for the purpose of observing temperature changes, and less attention was paid to actual histopathological damage[5]. In vivo animal experiments, the influence of blood flow on temperature changes is fully considered, and the monitored temperature is closer to the reality[6], but it is very difficult to make a urinary calculus model in vivo animals, and the accuracy of temperature monitoring is not as good as that in vitro. Moreover, the model consistency and sample size are both major challenges to the control of bias. Our study innovative integration of the advantages of in vitro and in vivo experiment, especially in the production of calculus models and pathological observation, which better simulate the actual situation compared with the previous ones. Holmium laser relies mainly on the photothermal mechanism of lithotripsy, where the aqueous medium at the tip of the transmitting fiber absorbs energy and vaporizes rapidly to form microbubbles[7]. However, not all of the energy is absorbed by the water and transmitted to the target stone. This laser- emitted energy that does not reach the stone can spill over, elevating the intrarenal temperature, damaging the urinary epithelium, and possibly creating complications[8]. Direct action of laser energy on the flushing fluid is less likely to occur in real procedures. Therefore, human calculus samples were included in our study. Page 6/13 Page 6/13 Page 6/13 The water flow rate is one of the most important factors affecting the temperature of lithotripsy [9]. Higher water flow rate can greatly reduce intrarenal temperature, but it brings higher intrarenal pressure, which increases the risk of infection. Some studies have pointed out that the water flow rate, the risk is higher, and the temperature after 10 and 60s of laser activation did not exceed 38.5°C at a rinse fluid flow rate of 40 mL/min, regardless of the laser configuration [10]. There are complicated factors such as bleeding and infection in the actual operation. In order to ensure the safety of the operation, urologists usually control the flow rate to the minimum value to ensure the visual field through the water valve of the nephroscope, so it is difficult to quantify the flow velocity. Therefore, in our study, we unified the ideal perfusion flow velocity of 30ml/ min. And we apply constant pressure pump irrigation without considering factors such as lithotripsy discharge and complete clarity of field of view, and therefore there is no need to manually adjust the water valve in the nephroscope. According to our surgical habit, the laser was excited for about 25 seconds with a pause of 5 seconds to allow the irrigation fluid to properly cool the kidney, hence the existence of peaks and valleys in our temperature profile. The proportion of laser energy reaching the stone can be used to estimate the efficiency of laser lithotripsy. An in vitro study reported that 52% of the emitted pulses reached the stones when using a frequency of 20 Hz; when using a frequency of 50 Hz, only 23% reached the target, and at 80 Hz only 4% [11]. It seems that the frequency of laser and lithotripsy efficiency are inversely proportional for the same power. A ratio that cannot be accurately measured in human surgery, but does the remaining energy generate higher heat and affect the intrarenal temperature? At this point, we used different frequency settings in groups A and C with a laser power of 30W, 15Hz in Group A versus 30Hz in Group C. During the first 5 minutes of intermittent excitation, no significant difference in mean temperature was observed between the two groups(36.06℃ vs. 36.36℃, t = 0.52, P > 0.05). Page 6/13 Thermal damage to tissues depends on the temperature reached and its ability to be maintained over time (thermal dose), which is why it is believed that tissue and cellular damage occurs at 43°C and is maintained for 240 minutes [15]. Body temperature > 42°C in humans causes substantial cellular damage, when the local temperature is > 45°C, there will be damage to the tissue, and when the local temperature is > 60°C, the tissue proteins are thermally coagulated and denatured and irreversible damage occurs [16]. A study used an vitro rabbit kidney placed in a closed container for holmium laser excitation, and there was an increase in water temperature and rabbit kidney tissue damage for a certain period of time, but this did not have the conditions of flowing water flushing and energy absorption by stones also[17]. Our results that the mean submucosal temperature did not exceed 40°C during the trail in all groups; the highest temperature was found in Group D, which used 60W excitation, at 49.2°C. In each group, there were periods of time when the temperature was above 42°C, but thanks to the same intermittent laser excitation and constant-rate irrigation as in human surgery, the high-temperature periods were relatively brief, lasting up to about 20s, and the temperature dropped back quickly after the pause in laser excitation. After the individual model temperature data were recorded completely, the porcine kidney was immediately handed over to an assistant for dissection. Different degrees of thermal injury lesions were observed visually within each renal parenchyma, as the holmium laser excitation time was longer and the number of lesions was higher in Groups C and D. We mainly cut the tissues near the temperature probe for pathological analysis. It was found in the specimen sections that Group D had the most serious lesions, with 60W holmium laser lithotomy for 10 minutes. The number of lesions with serious urothelium thermal damage accounted for 57.1%, while Group A with the least serious damage was only 15.4%, but there was no significant difference between Group B and group C. From this, we can judge that the higher power and longer Holmium laser excitation caused more thermal damage to the local kidney tissues, mainly the urothelium tissues. Page 6/13 Similarly, in groups B and D with laser power of 60W set at frequencies of 20 Hz and 30 Hz, the mean temperatures of the two groups were almost the same(39.21℃ vs. 39.27℃, t = 0.09, P > 0.05). We conclude that under sufficient lavage and good laser excitation habit, the residual energy of different frequency settings, which does not act on the stone, has little difference on the renal temperature. A comparative study of high versus low power laser lithotripsy showed significantly shorter procedure times but higher joules/mm3 values for high power lasers [12]. Another laboratory study simulating the renal calyx, authors investigated a study that included configuring the laser with different energies and frequencies to measure the temperature of the medium at different irrigation flows and found that the higher the energy delivered, the greater the increase in fluid temperature, reaching 50°C and 70°C in 10 and 60s without irrigation, respectively [10]. Another study by Aldoukhi, applied a flexible ureteroscope to excite a 40W holmium laser in a living pig renal pelvis, reaching peak temperatures of up to 84.8, 63.9 and 43.6°C in no, medium and high water flow rates, it showed that thermal damage at lower water flow with higher energy lasers [13]. However, most of the experiments still didn’t include the calculus factor. In our study, we observed that the higher the power, the greater the energy spillover and the higher and faster the intrarenal temperature increase is obvious, but the exponential increase in power does not translate into a significant increase in temperature; rather, the higher power may result in an increase in Page 7/13 lithotripsy efficiency and a decrease in operative time, which in the real world means a lower risk of surgical bleeding and infection. Although a recent meta-analysis showed no evidence of an advantage of higher power lasers over lower power lasers in terms of shorter operative times when evaluating the same calculus volume [14], this is strongly related to the fact that calculus composition varies in different populations. Therefore, the surgeon must fully weigh the mutual advantages and disadvantages of holmium laser thermal damage and procedure time in the context of the actual situation, such as the patient's calculus location, load, and infection status. Author's Contribution Wei Wei: Project development, Data management, Pathological observation, Manuscript writing Zhanping Xu: Project development, Manuscript editing Ming Chen: Data management, Manuscript editing Le Xie: Data collection, Pathological observation Yuan Mai: Data collection Huacai Zhu: Data collection Yuan Mai: Data collection Huacai Zhu: Data collection Page 6/13 In contrast, the thermal damage was similar between the high power, short time (Group B) and low power, long time (Group C), although this difference may gradually increase after further extension of the laser excitation time, pending our further study to prove it. At the same time, we also observed very limited degenerative necrosis of the submucosal tissue glomeruli and tubules at the injury site (< 25%, mild) from all sections. There are still some limitations in our study. Because of the isolated organ experiment, we could not simulate the effect of blood flow on the renal temperature. In the vivo experiment or real world, the actual temperature may be lower and the thermal damage may be smaller than the experiment due to the Page 8/13 regulating effect of blood flow. More samples in the study will make the results more accurate, which needs to be further investigated. Declarations 1. The study and manuscript had no disclosure of potential conflicts of interest. 1. The study and manuscript had no disclosure of potential conflicts of int 2. The study did not involve Human Participants and/or Animals. 2. The study did not involve Human Participants and/or Animals. 3. The experimental calculus was surgically removed from human. Patients signed informed consent for the calculus to be used in this study. 3. The experimental calculus was surgically removed from human. Patients signed informed consent for the calculus to be used in this study. Conclusion Different energy parameter configurations of holmium laser for PCNL, the energy will cause different degrees of thermal damage to the kidney, higher laser power and longer time for a certain irrigation speed lead to higher intrarenal temperature and more obvious damage, but the effect of rapid and high power excitation on intrarenal temperature is not a concern. Moderate irrigation and intermittent laser excitation habits are very good at avoiding excessive elevation of intrarenal temperature. Little damage to the renal tubules from heat was observed in vitro experiment, with effects mostly limited to the renal mucosa and subepithelium. In conclusion, holmium laser applications for lithotripsy are safe and reliable overall under different parameter. Urologists can rely on surgical experience and skill to avoid additional side effects. References 1. Tok A, Akbulut F, Buldu I, et al. Comparison of microperc and minipercutaneous nephrolithotomy for medium-sized lower calyx stones. Urolithiasis, 2016, 44(2): 155–9. 1. Tok A, Akbulut F, Buldu I, et al. Comparison of microperc and minipercutaneous nephrolithotomy fo medium-sized lower calyx stones. Urolithiasis, 2016, 44(2): 155–9. 2. Dmitry Enikeev, Thomas R W Herrmann, et al. Thulium fiber laser in endourology: current clinical evidence. Curr Opin Urol. 2023;33(2):95–107. doi: 10.1097/MOU.0000000000001057. Page 9/13 Page 9/13 Analysis of Laser Fiber to Stone Distance During Different Modes of Laser Lithotripsy. J Urol. 2020, 203, e626 12. Mekayten M, Lorber A, Katafigiotis I, et al. Will stone density stop being a key factor in endourology? The impact of stone density on laser time using Lumenis laser p120w and standard 20w laser-A comparative study. J Endourol. 2019, 33, 585–589. 13. Aldoukhi A.H, Hall T.L, Ghani K.R, Maxwell A.D, et al. Caliceal Fluid Temperature during High-Power Holmium Laser Lithotripsy in an in Vivo Porcine Model. J Endourol. 2018, 32, 724–729. 14. Ventimiglia E, Pauchard F, Quadrini F, et al. High- and low-power laser lithotripsy achieve similar results: A systematic review and meta-analysis of available clinical series. J Endourol. 2021, 35, 1146–1152. 15. Sapareto S.A, Dewey W.C. Thermal dose determination in cancer therapy. Int J Radiat. Oncol. Biol. Phys. 1984, 10, 787–800. 16. Izzard AS, Rizzoni D, et al. Small artery structure and hypertension: adaptive changes and target organ damage. J Hypertens. 2005;23(2):247–50. 17. Molina WR, Silva IN, Donalisio da Silva R, et al. Influence of saline on temperature profile of laser lithotripsy activation. J Endo urol.2015,29(2):235-9. Page 9/13 3. Skolarikos, A, Jung, H.U, Somani, B, et al. Guidelines Urolithiasis; EAU Guidelines Office: Arnhem, The Netherlands, 2022; ISBN 978-94-92671-16-5. 4. Dong H, Peng Y, Li L, Gao X, et al. Prevention strategies for ureteral stricture following ureteroscopic lithotripsy. Asian J Urol 5(2):94–100. 4. Dong H, Peng Y, Li L, Gao X, et al. Prevention strategies for ureteral stricture following ureteroscopic lithotripsy. Asian J Urol 5(2):94–100. 5. Héctor Gallegos, Juan Cristóbal Bravo, et al. Intrarenal temperature measurement associated with holmium laser intracorporeal lithotripsy in an ex vivo model. Cent European J Urol. 2021;74(4):588– 594. doi: 10.5173/ceju.2021.0092. 5. Héctor Gallegos, Juan Cristóbal Bravo, et al. Intrarenal temperature measurement associated with holmium laser intracorporeal lithotripsy in an ex vivo model. Cent European J Urol. 2021;74(4):588– 594. doi: 10.5173/ceju.2021.0092. 6. Hein Simon, Petzold Ralf, et al. Thermal effects of Ho:YAG laser lithotripsy during retrograde intrarenal surgery and percutaneous nephrolithotomy in an ex vivo porcine kidney model World J 6. Hein Simon, Petzold Ralf, et al. Thermal effects of Ho:YAG laser lithotripsy during retrograde 6. Hein Simon, Petzold Ralf, et al. Thermal effects of Ho:YAG laser lithotripsy during retrograde intrarenal surgery and percutaneous nephrolithotomy in an ex vivo porcine kidney model. World J Urol. 2020;38(3):753–760. doi: 10.1007/s00345-019-02808-5. Epub 2019 May 16. , , p y g g intrarenal surgery and percutaneous nephrolithotomy in an ex vivo porcine kidney model. World J Urol. 2020;38(3):753–760. doi: 10.1007/s00345-019-02808-5. Epub 2019 May 16. 7. Cecchetti W, Zattoni F, Nigro F, et al. Plasma bubble formation induced by holmium laser: an in vitro study [J]. Urology, 2004, 63(3): 586–90. 7. Cecchetti W, Zattoni F, Nigro F, et al. Plasma bubble formation induced by holmium laser: an in vitro study [J]. Urology, 2004, 63(3): 586–90. 8. Pauchard Felipe, Ventimiglia Eugenio,et al.A Practical Guide for Intra-Renal Temperature and Pressure Management during Rirs: What Is the Evidence Telling Us. J Clin Med. 2022;11(12):3429. doi: 10.3390/jcm11123429. 8. Pauchard Felipe, Ventimiglia Eugenio,et al.A Practical Guide for Intra-Renal Temperature and Pressure Management during Rirs: What Is the Evidence Telling Us. J Clin Med. 2022;11(12):3429. doi: 10.3390/jcm11123429. 9. Pauchard Felipe, Ventimiglia Eugenio, et al.A Practical Guide for Intra-Renal Temperature and Pressure Management during Rirs: What Is the Evidence Telling Us. J Clin Med. 2022;11(12):3429. doi: 10.3390/jcm11123429. 10. Aldoukhi A.H, Ghani K.R, Hall T.L, et al. Thermal Response to High-Power Holmium Laser Lithotripsy. J Endourol. 2017, 31, 1308–1312. 11. Aldoukhi A, Hall T, Ghani, K, et al. Figures Page 10/13 Figure 1 Experiment model and procedure (1a. Porcine kidney with calculus model, 1b. Equipment and procedure ) Figure 1 Experiment model and procedure (1a. Porcine kidney with calculus model, 1b. Equipment and procedure ) (1a. Porcine kidney with calculus model, 1b. Equipment and procedure ) (1a. Porcine kidney with calculus model, 1b. Equipment and procedure ) Page 11/13 gure 2 Figure 2 Temperature profiles of different laser energy parameters Temperature profiles of different laser energy parameters Temperature profiles of different laser energy parameters (2a.Group A/B 1st trail , 2b. Group A/B 2nd trail, 2c. Group C/D 1st trail, 2d. Group C/D 2nd trail) (2a.Group A/B 1st trail , 2b. Group A/B 2nd trail, 2c. Group C/D 1st trail, 2d. Group C/D 2nd trail) Page 12/13 Figure 3 Pathological pictures of thermal damage renal tissue (3a. mild, 3b. moderate, 3c. severe) Figure 3 Figure 3 Figure 3 Pathological pictures of thermal damage renal tissue Pathological pictures of thermal damage renal tissue (3a. mild, 3b. moderate, 3c. severe) Pathological pictures of thermal damage renal tissue (3a. mild, 3b. moderate, 3c. severe) Page 13/13
https://openalex.org/W4362425454	https://aacr.figshare.com/articles/journal_contribution/Supplementary_Data_from_Amplicon_Mapping_and_Expression_Profiling_Identify_the_Fas-Associated_Death_Domain_Gene_as_a_New_Driver_in_the_11q13_3_Amplicon_in_Laryngeal_Pharyngeal_Cancer/22439743/1/files/39890578.pdf	English	null	Supplementary Data from Amplicon Mapping and Expression Profiling Identify the Fas-Associated Death Domain Gene as a New Driver in the 11q13.3 Amplicon in Laryngeal/Pharyngeal Cancer	null	2,023	cc-by	372	Supplementary Methods RNA was extracted from snap frozen tumor tissue or cell lines using Trizol® (Invitrogen, Breda, Netherlands). From 10 slides of 10 µm RNA was extracted using Trizol® according to manufacturer’s instructions. Extracted RNA was DNAse-treated by incubating RNA for 1 hr at 37°C with 3µl (5 U) of RNAse free DNAse I (Zymo Research, Orange, CA, USA) in a total volume of 30 µl. Ten μl of RNA extracted from tumor tissue or 1 μg of cell line RNA was transcribed into cDNA using 200 units of Superscript II-reverse transcriptase (Invitrogen) and 300 ng of random hexamers (Invitrogen) following the manufacturer’s instructions in a final volume of 20 μl. Quantitative measurements of mRNA content were performed using a TaqMan-assay on an ABI PRISM® 7900 HT and the SDS software 2.1 (Applied Biosystems, Nieuwerkerk a/d lJssel, The Netherlands). Taqman primers and probes are listed in supplementary table 1. Primers were designed on different exons separated by intron sequences. Probes were labeled with a quencher (FAM) and a reporter (TAMRA). In order to be function appropriately, probes were designed to span exon-exon boundaries, to avoid specific splice variants preferably and to prevent detection of contaminating genomic DNA. PCR was performed with 2x Eurogentec Master Mix (Eurogentec, Seraing, Belgium) using 5 μl of diluted cDNA from an RT initiated with 5ng of RNA, 200 nM probe, and 900 nM primers. The relative copy number of each unknown sample (ΔCt) was obtained by normalization to TBP (TATA-box binding protein). Normalization was verified with the expression of a second housekeeping gene (HPRT). The ΔΔCT value was calculated by using the average ΔCt value of the second group as a calibrator. Therefore, all significant deviations from 1 are considered to be caused by amplification. dered as an Assay-on-demand (Applied Biosytems) and Taqman primers used for cyclin D1, HPR de probes for CPT1A, MRLP21, IGHMBP2, FLJ42258, TMEM16A, PPFIA1, CTTN and SHANK2 ttps://products.appliedbiosystems.com). Reverse primer Probe TGGAATCGTGGATCCCAAA CCGGGAGGAAATCAAACCAATTCGTC CCAAGCAGCGTGAAGTTGTC ACTGGAGAAGGTCCTGCTGGTTGGG CAATGACCACCACGTCGAAGT CAAACACAGGTGCGTCTGCCGATG 1) GGCACCCTTGTCTCCCTTGT TTTGCTGCTGGAGCTGGTGACCG CCTCCTCCTCGCACTTCTGT CTCGCAGACCTCCAGCATCCAGGT TCGCTTCTGTGTATCTCTTTCACAT CGCAGCCAGGATGGTTGAGCAAAC TCATCTTTCTGCTGTCCTTCTCAGT TGCTTTTGCATGGAAATGTCTACTGCA 1) 1) 1) TTC CCACAAAGTCGCTCATGAACAT CGGCTGGCGTTTGTCATCGTCTT 1) TCGTGGTTCATGTCCCCATA CCCCGCCCACATCTAGAACGACACT GAT TCTCGATGTTGGTGATGATGTCA AGGCTGCGGGCGATGATGAGATCT CAA CCTGCCGGATCATGTTCAC TGTCAAAGTCGGCCACAGGCAGGT GGAAACCATCCGATGTG CGCGGTGACAGGGTCAAAGTTCTG GCTGTCGCTTCCGTTGGAT CGGGACACAAGACAGGCAGTGCAAC CTCTTAGGAACAGGGATTG TGGACTAATGGTCAGGTGTGACAGAC CCTGTGCCCGAAGTGCAC CTTTTGCATTCATCCCTGCTGAGATCC GATAGATCAAGGGTCTCAGT AGATGGTGGACAAGAAGGTCTCGTGC GCCTCCAGCCAAGTATTCCA CCCACAGCTCATCTCCCACGAGGTA GTC GG G T C GCA CT T T G
https://openalex.org/W3011140494	https://periodicos.ufpi.br/index.php/lingedusoc/article/download/1083/934	Portuguese	null	A AUTORIA DISCENTE NA CULTURA DIGITAL	RELPE	2,019	cc-by	11,132	RESUMO O artigo aborda um problema referente à resistência dos alunos do ensino superior, que se envolvem em atividades que promovam a sua autoria. Apesar da adesão dos usuários em produzir conteúdo por meio de aplicativos (recursos digitais) que viabiliza a autoria, quando se trata do contexto educacional, nem sempre as interfaces presentes nos recursos digitais conseguem angariar a participação de discentes, principalmente se ainda não estão inseridos na Cultura Digital. Essa deficiência no envolvimento do discente pode estar relacionada ao desenvolvimento de uma produção textual, em razão da falta de aperfeiçoamento da escrita acadêmica dissertativa. Dessa forma, a exigência do uso da linguagem formal associada ao frágil domínio da modalidade culta da língua, pode provocar a inibição e a insegurança na participação, não viabilizando assim sua autoria digital. Nesse sentido, a questão da pesquisa é: como os recursos digitais, que vão além da linguagem textual, podem contribuir para a autoria discente? Para essa questão, objetiva-se identificar as possibilidades de autoria discente na produção de mapa conceitual e vídeo curto. Dessa forma, realizou-se um estudo teórico, no qual refletiu-se se os elementos da autoria (identidade e autonomia) podem ser identificados nos recursos digitais sugeridos. O artigo conclui que utilizar recursos que dispõem de interfaces no qual é viabilizada a construção de mapa conceitual e vídeo curto, como alternativas à produção de conteúdo textual, podem contemplar diferentes estilos de aprendizagem. Palavras-chave: Autoria digital. Mapa conceitual. Vídeo curto. p g CLEIDE JANE DE SÁ ARAÚJO COSTA Doutora em Educação pela Université de Provence Aix-Marseille I e em Linguística pela UFAL. Professora do Programa de Pós-graduação em Educação e do Programa de Modelagem Computacional do Conhecimento da UFAL. E-mail: cleidejanesa@gmail.com ORCID: https://orcid.org/0000-0002-2152-0465 Doutora em Educação pela Université de Provence Aix-Marseille I e em Linguí p g p Programa de Pós-graduação em Educação e do Programa de Modelagem Computacional do Conhecimento da UFAL. E-mail: cleidejanesa@gmail.com ORCID: https://orcid.org/0000-0002-2152-0465 Doutor em Educação (UFAL). Professor do Programa de Pós-graduação em Educação da UFAL E-mail: prof.fernandoscp@gmail.com ORCID: https://orcid.org/0000-0002-9180-8691 491 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 491 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 491 ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) FERNANDA JOSIRENE DE MELO FERREIRA Doutoranda em Educação pela Universidade Federal de Alagoas (UFAL). Mestre em em Modelagem Computacional de Conhecimento na linha de pesquisa em Educação pela UFAL. E-mail: fynanda@gmail.com ORCID: https://orcid.org/0000-0002-1391-5790 Á Ú ABSTRACT ABSTRACT The paper is about a problem regarding the resistance of students of higher education to being involved in activities that promotes their authorship. Despite the users' adhesion to produce content through applications (digital resources) that makes authorship possible, when it comes to the educational context, the interfaces present in the digital resources aren’t always be able to attract the participation of students, especially if they aren’t yet part of Digital Culture. This deficiency in the student's involvement may be related to the development of a textual production, due to the lack of improvement of the academic writing. Thus, the requirement of the use of formal language associated with the fragile domain of the cultured language modality can provoke inhibition and insecurity in the participation, not allowing its digital authorship. In this sense, the research question is: How did digital resources, which go beyond textual language, contribute to student authorship? For this question, the goal is to identify the possibilities of student authorship in the production of conceptual map and interactive short video. By this, a theoretical study was carried out, in which it was reflected if the 1 Agradecimento ao apoio financeiro FAPEAL/CAPES. Dedicamos esse artigo à memoria da Professora Anamelea de Campos Pinto. 1 Agradecimento ao apoio financeiro FAPEAL/CAPES. Dedicamos esse artigo à memoria da Professora Anamelea de Campos Pinto. 492 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 492 elements of the authorship (identity and autonomy) can be identified in the digital resources suggested. The article concludes that using resources that have interfaces in which it is feasible to construct a conceptual map and interactive short video, as alternatives to the production of textual content, can contemplate different styles of learning. Keywords: Digital authorship. Conceptual map. Short video. RESUMEN El artículo aborda un problema con respecto a la resistencia de los estudiantes de educación superior, que participan en actividades que promueven su autoría. A pesar de la adhesión de los usuarios para producir contenido a través de aplicaciones (recursos digitales) que hacen posible la autoría, cuando se trata del contexto educativo, las interfaces presentes en los recursos digitales no siempre pueden atraer la participación de los estudiantes, especialmente si no están insertos en la Cultura Digital. Esta deficiencia en la participación de los estudiantes puede estar relacionada con el desarrollo de una producción textual, debido a la falta de mejora de la escritura académica disertativa. De esa forma, la exigencia del uso del lenguaje formal asociado con el frágil dominio de la modalidad lengua culta, puede causar inhibición e inseguridad en la participación, lo que no hace posible su autoría digital. En este sentido, la pregunta de investigación es: ¿Cómo los recursos digitales, que van más allá del lenguaje textual, pueden contribuir a la autoría de los estudiantes? Para esta pregunta, el objetivo es identificar las posibilidades de autoría de los estudiantes en la producción de mapas conceptuales y videos cortos. Así, se realizó un estudio teórico, que reflexionó si los elementos de autoría (identidad y autonomía) pueden identificarse en los recursos digitales sugeridos. El artículo concluye que usar recursos que tienen interfaces en el cual permite la construcción de un mapa conceptual y un video corto, como alternativas la producción de contenido textual, pueden contemplar diferentes estilos de aprendizaje. Palabras clave: Autoría digital. Mapa conceptual. Video corto. Introdução A produção de conteúdo na Web 1.0 limitava-se a uma minoria detentora de conhecimentos técnicos de informática (ACEDO, 2012). Com isso, estabelecia-se uma comunicação unidirecional, de forma que o conteúdo era transmitido de um emissor para muitos receptores (BRAVO; COSLADO, 2012). Dessa forma, os usuários em massa restringiam-se a consumir o conteúdo de forma passiva. Contudo, com a chegada da web 2.0, as Tecnologias da Informação e Comunicação (TIC) têm disponibilizado uma infinidade de recursos digitais, que dispõem de interfaces nas quais priorizam a linguagem natural (escrita) como principal meio de exercer a autoria, a exemplo do fórum de discussão, blogs, wikis, bate-papo. Assim, estabelece a comunicação bidirecional (BRAVO; COSLADO, 2012), na qual o usuário passa a ser não só receptor ou consumidor de conteúdo, mas também emissor, produtor de conteúdo, logo autor. 493 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 493 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Embora exista um número crescente de usuários ativos na web 2.0 (KEMP, 2018), que integra a Cultura Digital, quando se trata do contexto educacional, no que se concerne à autoria, nem sempre as interfaces presentes nos recursos digitais conseguem angariar a participação deles, principalmente se não estão inseridos nessa cultura. Essa resistência tem sido observada em maior parte pelos professores, que normalmente são representados pelo grupo baby boomers aos da Geração X (BRAVO; COSLADO, 2012), que engessam a produção de conteúdos digitais autorais. Além de professores, também existe alunos que não estão inseridos na Cultura Digital, razão pela qual pode tornar deficiente o envolvimento na participação destes em interfaces que promovam a autoria no contexto educacional. Os autores Braga e Senem (2017) já relatavam preocupação dos alunos de graduação e pós-graduação na educação presencial quanto a escrita para (rea)aprender a escrever texto dissertativo, aperfeiçoar e melhorar a escrita acadêmica. A mesma dificuldade está presente ao utilizar interfaces dos recursos digitais, no contexto da educação online, pois Cunha (2006) sugere que o fato de o aluno não possuir domínio da modalidade culta da língua, pode deixá- lo inibido e inseguro em participar, e consequentemente desenvolver sua autoria digital. Nesse sentido, Bezerra (2011) defende que uma possível razão seja a exigência do uso da linguagem formal. Introdução Contudo, como maior parte dos alunos já exercem sua autoria em assuntos cotidianos nas interfaces disponíveis nos recursos digitais, parece oportuno trabalhar/incentivar a autoria discente, especialmente se representados por usuários inseridos na Cultura Digital. Diante do contexto abordado, busca-se responder ao seguinte questionamento: como os recursos digitais, que vão além da linguagem textual, podem contribuir para a autoria discente? Para responder ao questionamento, este artigo tem como objetivo identificar as possibilidades de autoria discente na produção de mapa conceitual e vídeo. Espera-se que ao diversificar os recursos digitais, o aluno tenha possibilidades de exercer a autoria, assim como já exerce nos recursos que utilizam a linguagem textual e nos demais recursos, como Facebook, Instagram e WhatsApp, que utilizam diferentes linguagens para fins de entretenimento. Para isso, realizou-se um estudo teórico, no qual foi observado se os elementos da autoria (identidade e autonomia) podem ser identificados nos recursos digitais sugeridos nesse artigo. O artigo encontra-se organizado da seguinte forma: na próxima seção é realizada uma revisão de literatura a respeito da autoria na cultura digital; na segunda seção são explanadas as potencialidades de autoria do mapa conceitual e vídeo como recursos digitais; e por fim as considerações finais. 494 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 494 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Autoria na cultura digital Autoria na cultura digital Antes de iniciar a revisão da literatura de autoria no contexto da Cultura Digital é explorado de forma breve o conceito que permeia esse contexto. Segundo Barato e Crespo (2013), o termo cibercultura é conceituado por meio da cultura, por sua derivação, até encontrar a expressão Cultura Digital e entendê-la como sinônimo. Sabendo da afirmação de Lévi-Strauss (1963), quando diz que a linguagem é o principal meio pelo qual aprendemos sobre nossa cultura, e sendo assim, ponte essencial para haver comunicação nas relações sociais, podemos compreender a Cultura Digital por meio de suas linguagens, de suas formas de comunicar, do que elas comunicam, do espaço/lugar que elas ocorrem, nas quais permeiam as relações sociais. As relações sociais da Cultura Digital encontram-se em um espaço virtual, no qual é denominada por Santaella (2007) como “não lugar”. Conforme Levy (2009, p. 47) “é virtual toda entidade ‘desterritorializada’, capaz de gerar diversas manifestações concretas em diferentes momentos e locais determinados, sem, contudo, estar ela mesma presa a um lugar ou tempo em particular”. Em outras palavras, o espaço virtual permite que o usuário se comunique no tempo e lugar que ele desejar, desde que tenha uma rede de conexão disponível e dispositivo para conectá-lo. Na Cultura Digital essa comunicação pode trazer uma informação nova, sendo ele autor, ou uma informação pré-existente proveniente de outros usuários, por meio de um conteúdo digital, no qual é possível compartilhá-lo com outros usuários, que Santaella (2007, p. 180) denomina de ‘disponibilizar’, como a “primeira palavra de ordem do ciberespaço”. As informações são transmitidas por meio da interatividade proporcionada pelas interfaces com o usuário, de forma que a linguagem utilizada varia conforme o recurso digital que se esteja utilizando. Contudo, na Cultura Digital é comum que uma mesma interface seja compatível para integrar ao mesmo tempo texto, imagem, áudio e vídeo. Portanto, diante do exposto, a Cultura Digital é fundada em um espaço e tempo livre, no qual seu desenvolvimento depende dos processos cognitivos socializados por meio de uma linguagem suportada pelo recurso digital definido. Revisão de Literatura Revisão de Literatura Revisão de Literatura 495 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 495 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) A revisão de literatura, do tema proposto nessa seção, buscou resultados obtidos no Periódicos Capes2 a partir de 2012, independentemente do idioma. 2 Maiores informações <www.periodicos.capes.gov.br/>. Autoria na cultura digital 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 496 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) educador promove e orienta a aprendizagem coletiva. Como os autores fazem um estudo teórico, o estudo não identifica o tipo de linguagem investigada na autoria, ficando aberta em abranger dessa forma, todas. Santarosa, Conforto e Basso (2012) apresentam o Eduquito, uma ferramenta de autoria e de colaboração acessível na perspectiva da Web 2.0, que tem como objetivo apoiar processos de inclusão sociodigital, conforme os princípios de acessibilidade e de desenho universal, normatizados pela WAI/W3C. Na ferramenta Eduquito é explorada a Oficina Multimídia, na qual o processo de criação dos documentos contempla a utilização de diferentes tipos de mídias: textos, imagens, vídeos ou áudios. A Oficina Multimídia passou por uma fase de validação com usuários de diferentes limitações sensoriais, cognitivas e motoras, no Brasil e na Espanha. Comprovou-se que a agregação de recursos em diversos formatos, permitida pela Oficina Multimídia, impulsionaria a construção de uma biblioteca com as produções dos diferentes interagentes, promovendo e configurando um espaço para construções coletivas e individuais. O trabalho de Azevedo, Nascimento e Oliveira (2016) traz como objetivo evidenciar como uma proposta didática e pedagógica que utiliza TIC pode contribuir com a formação de autores de textos “verbovocovisuais” (denominação para a integração verbal, visual e sonora), produzidos na esfera literária e digital, em duas escolas públicas de educação básica. Ao analisar um miniconto multimodal o trabalho indica critérios que favorecem a compreensão das atitudes responsivas de estudantes em aulas de língua portuguesa. Os autores concluem que as práticas escolares associadas às TIC estimulam o protagonismo do aluno. Embora o objetivo das autoras Pimenta, Momesso e Assolini (2016) foi observar e refletir sobre as práticas de leitura de um curta-metragem da cena de Hamlet presente no YouTube, elas indicam a possibilidade de coautoria, por parte do sujeito adolescente. Para as autoras, a leitura por meio de vídeo produz efeitos de sentido na constituição do sujeito discursivo. Para elas, os efeitos de sentido são construídos na relação com o outro e em seu ambiente, e, para isso, o sujeito precisa se apropriar do campo da linguagem e fazer questionamentos sobre si mesmo. Autoria na cultura digital Portanto, serão considerados apenas artigos que abrangem os últimos seis anos (2012-2018), pois se trata de um tema recente, no qual envolve a autoria na Cultura Digital a partir dos recursos digitais da web 2.0, que oferece interfaces de comunicação bidirecional e viabiliza o usuário ativo na produção de conteúdo. A pesquisa realizou-se por assunto, por meio de uma busca avançada, no qual definiu-se o termo “autoria” associado ao termo “digital”. Com isso, foram obtidos 880 resultados, filtrados, por resumo das pesquisas ou quando não disponível pelo texto completo, considerando apenas a autoria digital discente em produções que vão além de texto. Considerou-se ainda apenas os 100 primeiros artigos, ordenados por relevância e distribuídos nas dez páginas iniciais. Dessa forma, foram ignorados os resultados que não estivessem simultaneamente dentro do contexto educacional e digital e os que contemplassem somente a linguagem textual. Logo, o resultado abrangeu trabalhos que discutissem ou viabilizassem a autoria digital discente no contexto educacional, por meio de recursos, denominados nos artigos de ferramenta ou software, que não se limitassem a linguagem textual. Os autores Trentin, Teixeira e Feldkircher (2012) descrevem o processo do desenvolvimento de um software de autoria, tanto para professores como para alunos, nos aspectos técnicos e educacionais. O software é denominado de “Guri”, no qual permite desenvolver conteúdo hipermídia para a TV digital. Contudo, a pesquisa em questão vislumbrava que em 2016 todos os brasileiros teriam acesso ao sinal televisivo digital, e que tal fato seria um grande potencial para processos de inclusão social/digital via acesso a aplicações interativas na TV digital. Mas, estamos em 2019, e esse processo de migração para TV digital, apenas iniciou-se. Portanto, associar a televisão digital como um grande potencial para a educação parece ainda um estágio distante e prematuro, tal como foi a promessa das TIC resolverem os problemas da educação. O estudo de Foresti e Teixeira (2012) teve como objetivo associar os conceitos de aprendizagem de Paulo Freire, Seymour Papert e George Siemens para a proposta de um conceito de aprendizagem no contexto da Cultura Digital. Como resultado deste estudo, fica evidente que a aprendizagem na era digital deve ter como principal elemento a criação de estratégias eficazes, a exemplo da autoria colaborativa, dentro de um contexto de interação, comunicação e feedback. Nesse sentido, o educando assume um papel de nó ativo na rede e o 2 Maiores informações <www.periodicos.capes.gov.br/>. 496 Teresina, Ano 24, n. Autoria na cultura digital Na pesquisa desenvolvida por Maddalena e Santos (2017) buscou-se compreender as potencialidades da narrativa digital e da linguagem hipermidiática que discentes e docentes produziram sobre suas histórias pessoais de formação na UERJ, desencadeadas pela crise do estado. O movimento de repúdio e luta mostra como as redes sociais na Internet atuam como espaços de autoria nos quais a indignação e a esperança dão lugar a múltiplas manifestações 497 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 497 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) de protestos. As narrativas digitais foram construídas baseadas no método “Digital Storyelling” (contação de história), no qual é utilizado recurso digital. Um dos recursos utilizados priorizou a linguagem audiovisual. Nesse sentido, o vídeo foi construído por meio da associação de imagens narradas com duração de 3 a 5 minutos. Embora não foi uma preocupação dos autores, como é do presente artigo, no método utilizado para construir o vídeo, as dificuldades quanto a sua exposição e memorização de longos discursos são atenuadas. Primeiro porque o aluno não será filmado, mas usado em imagens fotográficas que ilustrem o contexto narrado e segundo porque o áudio estará separado de sua imagem, permitindo mais liberdade de consulta ao texto, embora precise haver sincronia da narração com as imagens. Apesar de o artigo visar à autoria apenas na perspectiva discente, o trabalho de Maddalena e Santos (2017) também perpassa a perspectiva do professor. Os seis trabalhos foram apresentados em ordem cronológica, de forma que um deles segue uma linha teórica, dois deles referem-se à ferramenta de autoria (Guri e Eduquito) e os demais (a outra metade – três deles) exploram o termo autoria fazendo análises. Nenhum dos artigos abordados considera o mapa conceitual de forma explícita, como possibilidade de autoria, enquanto que grande parte deles, considera o vídeo. Isso porque o trabalho de Azevedo, Nascimento e Oliveira menciona o termo “verbovocovisuais”, sem deixar claro se o visual se trata de um mapa conceitual. Contudo, os que consideram vídeo, não fazem referência às possibilidades efêmeras impostas pelos aplicativos que produzem história por meio da junção de vídeos curtos, como explanará o presente artigo. De uma forma geral, os trabalhos contribuem para a visão de que o usuário comum, como o aluno, pode agora ser autor, contribuindo para a geração de conhecimento a partir de suas reflexões agora compartilhadas no espaço digital. Autoria na cultura digital Mas nem sempre isso foi possível, é o que veremos na próxima seção. O tratamento da Autoria precedente a Cultura Digital O tratamento da Autoria precedente a Cultura Digital Para adentrar no conceito de autoria no contexto da Cultura Digital, inicialmente é abordada uma síntese do tratamento do conceito de autoria precedente a essa cultura. Para Foucault (1992), a autoria tem a função de organizar a circulação do discurso3 de determinada época da sociedade. Esse discurso está sujeito a receber influências do indivíduo, denominado como posições de discursividade. Apesar disso, a função autor não está ligada a 3 “Chamaremos de discurso um conjunto de enunciados” (FOUCAULT, 2002, p. 135). 3 “Chamaremos de discurso um conjunto de enunciados” (FOUCAULT, 2002, p. 135). 498 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 498 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) um indivíduo, mas pode “dar lugar a vários ‘eus’ em simultâneo, a várias posições-sujeitos que classes diferentes de indivíduos podem ocupar” (FOUCAULT, 1992, p. 56-57) ou a várias identidades. Bennett (2005) concorda ao colocar a autoria como produção individual em segundo plano, colocando o autor como anônimo, pois o interesse não é nele e, sim, no discurso. Isso pode justificar a visão de que o autor é apenas um meio para expressão da inspiração divina, pois conforme Burke (1995) Deus é o verdadeiro autor. Foucault (2002) se preocupa em diferenciar o autor do sujeito do discurso/enunciado, no qual afirma que o autor é o indivíduo real, que se responsabiliza pelo discurso que escreveu perante a sociedade, enquanto o sujeito do discurso é uma representação de ser, tratando-se, portanto, de uma suposição. Para Fortunado (2003, p. 20-21) o sujeito do enunciado é aquele que “se projeta através do discurso [...] que dissimula a identidade do autor real”, ou seja, para ser aquele que não é. A notícia “Homem se passa por policial federal e aplica golpes em Brumadinho” (ZAMBONI, 2019) traz um exemplo claro de quando o sujeito se coloca na posição de algo que não é. Embora posteriormente sua real identidade tenha sido descoberta, o sujeito conseguiu representar um suposto policial. Em outros contextos, o sujeito convidado para se pronunciar na justiça, numa entrevista, ou atuar na defesa de um criminoso pode incorporar uma identidade que ele julga aceitável à sociedade ou aos seus princípios ideológicos ou ainda que cumpra com a sua função de advogado ou que respeite as cláusulas contratuais, mas que pode ter uma opinião como autor diferente da divulgada ou publicada. O tratamento da Autoria precedente a Cultura Digital Em outras palavras, o sujeito do discurso pode esconder sua real identidade como autor. Fazendo “um rápido retrospecto histórico sobre a questão da autoria”, Mantovani, Dias e Liesenberg (2006, p. 260) afirmam que desde a invenção da imprensa foi atribuída ao autor a categoria de proprietário de seus textos. Nesses textos “nada se podia acrescentar, suprimir ou alterar [...] esta situação criou entre autores e leitores um distanciamento significativo, assumido pelos últimos como quase intransponível” (MANTOVANI; DIAS; LIESENBERG, 2006, p. 260). Fazendo uma analogia, isso nos remete a comunicação unidirecional, de um autor para muitos leitores impedidos de interagir com o texto, na qual a produção de conteúdo limitava-se aos detentores de conhecimentos de informática. Apesar da analogia acima, Liang (2011) revela que no final da idade média, no século XIV, os textos podiam sofrer mudanças, seja adicionando, recebendo adaptação. Nos olhares atuais, isso caracteriza uma autoria coletiva, pois os leitores podiam interferir com anotações que permeavam o texto original, formando assim, outra obra. Esse relato pode ser associado à 499 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 499 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) comunicação bidirecional que se faz presente na Cultura Digital, ou seja, quando o leitor também pode ser autor. Isso justifica a afirmação de Ricardo e Vilarinho (2006), ao defender que há alteração da relação entre leitor e autor com o texto. Conforme Chartier (1999) ainda nos tempos da disseminação da imprensa, os termos escritor e autor não denotavam o mesmo significado. Isso porque o escritor tratava-se daquele que escreveu um texto manuscrito sem circulação e o autor tratava-se daquele que escreveu um texto impresso publicado, ganhando assim notoriedade, pois seu nome próprio adquiria identidade e autoridade ao texto. Por outro lado, os recursos da cultura digital proporcionam que aquele que escreve já seja o autor, às vezes injustamente, diante da prática de plágio sem controle, pois conforme Fortunato (2003, p. 117) “toda informação digital é suscetível de reprodução”. Neste sentido, “é fundamental que o aluno perceba que ao plagiar está abrindo mão de sua autoria, de sua singularidade e individualidade, e da chance de criar e produzir conhecimento” (RICARDO; VILARINHO, 2006, p. 71). O tratamento da Autoria precedente a Cultura Digital Contudo, o ato de querer ocultar-se, demanda esforço do autor em desenvolver uma identidade, ainda que fictícia, pois precisará apagar ou manipular os rastros de formação do discurso, influenciando na interpretação que o interlocutor fará de seu discurso (PÊCHEUX, 1995). Isso significa, que se por um lado o aluno desperdiça a autoria quando plagia, por outro o plágio demanda esforço para criar uma identidade fantasiosa no desenvolvimento do discurso. Ou seja, se há esforço cognitivo para qualquer uma das situações descritas, não há porque o aluno investir tal esforço no lado ilícito da construção de sua identidade como autor. Embora essa seção se prestou a tratar a autoria nos tempos antecessores da cultura digital, sempre que possível realizou-se comparações a essa cultura. Portanto, essa seção abordou que a autoria já foi colocada em segundo plano, em razão da valorização do discurso; abordou a diferença entre autor e sujeito do discurso, bem como a diferença entre autor e escritor com o surgimento da imprensa; abordou o autor como proprietário de seu texto, mas impossibilitado de receber contribuição de seus leitores na época da chegada da imprensa, enquanto a autoria coletiva foi possível em plena idade média e nos dias de hoje. Neste sentido, Foucault (2000) afirma que ao longo do tempo a função autor não parou de enfraquecer. Portanto, enquanto a autoria estava associada a uma inspiração externa, vinda de Deus, com a chegada da cultura digital, ela passa a ser associada a uma interiorização individual, Portanto, enquanto a autoria estava associada a uma inspiração externa, vinda de Deus, com a chegada da cultura digital, ela passa a ser associada a uma interiorização individual, 500 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 500 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) autônoma, valorizando a expressão de sua interpretação pessoal, sua identidade. Desta forma, na próxima subseção é abordado três aspectos da autoria dentro do contexto da cultura digital. Identidade, Autonomia e o Desaparecimento do Autor na Cultura Digital ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 501 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) fantasias, dificultando o reconhecimento e caracterização do sujeito (FORTUNATO, 2003). Neste sentido, a cultura digital parece ser a de autores e de conteúdos descartáveis ou efêmeros, diante do mundo globalizado, veloz e ávido por novidades da sociedade da informação. Mas o sujeito para ser autor precisa não só desenvolver sua própria identidade, mas também ter uma postura autônoma, como requer a educação (PINTO; BASTOS FILHO, 2012). A palavra autônomo “vem do grego: autos (eu mesmo, si mesmo) e nomos (lei, norma, regra). Aquele que tem o poder para dar a si mesmo a regra, a norma, a lei, é autônomo e goza de autonomia ou liberdade. Autonomia significa autodeterminação” (CHAUI, 2000, p. 566). No caso do termo autonomia, na quinta definição o dicionário refere-se à “propriedade pela qual o homem pretende poder escolher as leis que regem sua conduta” (FERREIRA, 1986, p. 203). Desta forma, a definição dos termos autônomo e autonomia convergem em regras nas quais o sujeito possui o controle de escolhê-las. Segundo Petroni e Souza (2010, p. 357) “a autonomia só poderá ser exercida por sujeitos que se constituam como autorregulados, capazes de tomar para si a regulação de sua própria conduta”, mesmo diante da regulação externa. A regulação externa refere-se às regras ou costumes que a sociedade apresenta, enquanto que a autorregulação refere-se ao sujeito que internaliza essas regras mediadas nas e pelas interações sociais. Desta forma, o sujeito se torna autorregulado, ou seja, capaz de dominar sua própria conduta, quando atribui ou formula um sentido/significado próprio a essas regras (PETRONI; SOUZA, 2009). Isso significa que o sujeito autonômo controla suas escolhas conforme o que lhe traz significado, pois pode decidir quais das regras apresentadas pela sociedade fazem sentido para ele. Fazendo uma analogia para o contexto educacional, especialmente no que se refere à Cultura Digital, o sujeito aluno é autônomo quando o conjunto de regras que determinam seu comportamento em um ambiente online é acordado com o professor, de forma que as regras deixam de ser impostas e passam a ser negociadas. Esse conjunto de regras, pré-estabelecidas, é chamado de contrato didático (BROUSSEAU, 1986), no qual são estabelecidos direitos e deveres não só do aluno, como também do professor. Identidade, Autonomia e o Desaparecimento do Autor na Cultura Digital Para Fichte (1791), todo indivíduo tem seu próprio processo de pensamento, seu próprio modo de expressar ideias e de formar conceitos e conectá-los entre si. Podemos associar esse próprio processo de pensamento ao que Orlandi (2001, p. 79) denomina de interiorização, quando explica que o autor constrói sua identidade a partir do momento que ele “estabelece uma relação com a exterioridade, ao mesmo tempo em que se remete à sua própria interioridade”. Nesse sentido, Foucault (1992) afirma que a autoria é relevante quando se reconhece a identidade do autor, pois cada autor tem seu estilo próprio, com palavras e formas de expressão, caracterizando sua escrita, ou seja, sua marca autoral. Essa marca autoral remete ao que Possenti (2002, p. 114) se referiu de “o jeito, o como” do autor. Ainda fazendo associação com o próprio processo de pensamento, citado acima por Fichte, a possibilidade de compartilhá-lo nas interfaces presentes na Cultura Digital é um grande feito das TIC. Isso porque de forma instantânea a produção de conteúdo se torna pública se comparada ao processo autoral que ocorria de forma isolada. Dessa forma, segundo Martins (2012), ao compartilhar o conteúdo produzido, o processo de autoria digital fica sujeito a receber constantes colaborações, caracterizando assim, um discurso aberto e inacabado. Ou seja, enquanto recebe colaboração o conhecimento é ressignificado mediante reflexão e análise crítica para acatá-la ou não. Apesar das possíveis colaborações a serem recebidas, Weissberg (2003) afirma que a identidade de cada indivíduo é valorizada. Isso porque as interfaces dos recursos digitais geralmente dispõem de funções que facilitam que a contribuição registrada seja pelo nome do usuário ou creditada por menções pré-dispostas, a exemplo de como ocorrem na Wikipédia, Facebook, WhatsApp e Twitter. No Twitter essa menção é chamada de ReTweet ou simplesmente RT. Por outro lado, Fortunato (2003, p. 130) não vê a identidade do indivíduo valorizada na cultura digital, mas obscurecida, sem relevância, “pois não há tempo para o culto ao autor. Em tempo de internet, o alvo da atenção é a informação e não a sua autoria”. Isso se deve em razão da internet ser um espaço sem pátria nem dono; em razão da velocidade de informações que circulam, envelhecem e são substituídas por outras, transformando discursos em eventos; mas também em razão de autorias fictícias, que criam personagens anônimos para dar vazão a 501 Teresina, Ano 24, n. 43, set./dez. 2019. Identidade, Autonomia e o Desaparecimento do Autor na Cultura Digital Dessa forma, as opiniões de ambos os lados são consideradas e respeitadas, e assim o sujeito aluno tem autonomia para tomar decisões por si próprio. De acordo Pinto e Bastos Filho (2012), para se ter autonomia o aluno precisa buscar ser independente para que encontre a sua afirmação, ou seja, a sua identidade, enquanto indivíduo. Para estes autores, o ponto de partida para essa independência pode estar nas De acordo Pinto e Bastos Filho (2012), para se ter autonomia o aluno precisa buscar ser independente para que encontre a sua afirmação, ou seja, a sua identidade, enquanto indivíduo. Para estes autores, o ponto de partida para essa independência pode estar nas 502 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 502 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) contribuições, ou seja, quando o aluno ao refletir e interpretar suas leituras de estudo, expressa seu modo de pensar, suas ideias e conceitos de forma própria, autêntica. Dessa forma, o aluno assume sua potencialidade cognitiva e desenvolve sua autonomia. Petroni e Souza (2010, p. 358) classificam o sujeito autônomo como aquele que ao contribuir: (...) se percebe no mundo, que se torna ator e autor de sua história, consciente de que não está sozinho, vendo-se como diferente e aprendendo com as diferenças; aquele que dispõe de recursos para expressar-se livremente e ser compreendido pelo outro, em um exercício permanente do diálogo e da reflexão, em que exerce sua liberdade (PETRONI; SOUZA, 2010, p. 358). Nesse sentido, é a partir das interações sociais que o sujeito tem oportunidade de emitir e receber contribuições em um espaço plural de conhecimentos, propiciando assim, reflexões que podem influenciar novos significados aos envolvidos. Conforme Pinto e Bastos Filho (2012), mesmo que nas primeiras contribuições sejam expressos pensamentos dos outros e sejam mínimas, com pouca criatividade e de pouca relevância, gradativamente o aluno desenvolverá seu estilo, que revelará sua autenticidade e identidade, singular e original, dentro do ritmo e limitações de cada um. Portanto, é imprescindível que na Cultura Digital o aluno se interesse em desenvolver sua autonomia o quanto antes, para que suas contribuições gradativamente possam ser fortalecidas com identidade própria, autoral. Recursos digitais como possibilidades de autoria discente Segundo Romera (2014), a partir da década de 1990 começaram a surgir os primeiros programas informáticos destinados à criação de conteúdo digital para a educação, ou seja, o que chamamos hoje de ferramentas de autoria. Para o autor, neste momento, o conteúdo digital passou a ser visto como substituto do livro, numa perspectiva docente. No entanto, Johnson (2001) orienta que o termo ferramenta é inadequado, de forma que o correto é utilizar interfaces. Isso porque “a ferramenta está para a sociedade industrial como instrumento individual”, enquanto “a interface, no contexto sociotécnico do computador online, é espaço coletivo de comunicação entre duas ou mais faces humanas [...] geograficamente dispersas” (JOHNSON, 2001, p. 19). Portanto, nas próximas subseções são abordados dois recursos digitais, que dispõem de interfaces que abrangem a linguagem diagramática, por meio de mapas conceituais e da linguagem audiovisual, por meio de vídeo curto. Identidade, Autonomia e o Desaparecimento do Autor na Cultura Digital Para Menezes (2013), “o autor pode ser considerado como o sujeito que expressa sua percepção intelectual por meio de suas obras publicadas e, desse modo, contribui com a expansão do conhecimento social”. Na cultura digital, dentro do contexto acadêmico, as publicações se tratam de artigos científicos que entram em circulação na sociedade por meio de revistas digitais online. Mas dentro do contexto da Cultura Digital, é necessário matar o autor como Foucault (1992) sugere? Nesse contexto, a morte do autor já foi consumada pela necessidade de acabar com a monopolização da produção do conteúdo. É uma forma figurada que indica o intuito de valorizar a posição do leitor, que agora também é autor. Então, a morte do autor representa apenas uma retração de sua posição hierárquica, pois na Cultura Digital estão presentes leitores que são capazes de opinar, comentar, concordar ou refutar o autor, surgindo assim novos autores, em um processo cíclico de morte e surgimento de novos autores ou coautores. Essas contribuições recebidas pelo autor podem implicar em ressignificação de conhecimentos, contribuindo assim para um aprendizado constante. 503 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 503 Teresina, Ano 24, n. 43, set./dez. 2019. SN 2526-8449 (Eletrônico) 1518-0743 (Impress , , , ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Recursos digitais como possibilidades de autoria discente Mapas conceituais e autoria discente O mapa conceitual surge como uma das possibilidades de autoria discente. Para Moreira e Masini (1982), mapa conceitual é a técnica de Joseph Novak, que viabiliza a aprendizagem significativa de Ausubel. Sua linguagem gráfica organiza e representa o conhecimento por meio do relacionamento de conceitos entre nova informação e conceitos prévios na estrutura cognitiva do aluno, formando uma proposição que representa um significado (unidade semântica) (MACHADO; OSTERMANNÉ, 2006). Um exemplo de mapa conceitual é encontrado na figura 1. FIGURA 1 – Exemplo de Mapa Conceitual construído no CmapTools: Web 1.0 versus 2.0 GURA 1 – Exemplo de Mapa Conceitual construído no CmapTools: Web 1.0 versus 2.0 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Fonte: Autores, 2019 504 Teresina, Ano 24, n. 43, set./dez. 2019. N 2526-8449 (Eletrônico) 1518-0743 (Impre Teresina, Ano 24, n. 43, set./dez. 2019. Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Fonte: Autores, 2019 Organizar o relacionamento de conceitos, na construção do mapa conceitual, pode ser o ponto de partida para o que aluno exerça a autonomia na formação do sujeito autor. Para estabelecer esse relacionamento, Marriott e Torres (2014, p. 177) afirmam, que é preciso “refletir, pensar, ponderar, buscar as informações no texto novamente e analisar, por exemplo, se o conceito A é ‘decorrente’ ou ‘gerador’ do conceito B e se o conceito C deve ser ligado ao conceito A ou B, etc”. Dessa forma, o aluno estará no caminho para ser independente, e assim, vir a se tornar autônomo, como condição necessária para desenvolver sua autoria. No mapa conceitual o aluno tem oportunidade de expressar seu processo de pensamento, seu próprio modo de formar e organizar conceitos e conectá-los, imprimindo dessa forma, seu estilo com palavras e formas de expressão, caracterizando assim sua identidade autoral. Nesse sentido, quando é solicitado que cada aluno construa seu mapa conceitual sobre o mesmo assunto, as produções recebidas são distintas, devido segundo Machado e Ostermanné (2006) a idiossincrasia do mapa. Isso significa que em cada mapa é expressa a identidade do autor. Sendo assim, os autores concluem que não há o mapa conceitual correto, pois depende do vocabulário e de conceitos prévios utilizados por cada aluno autor. Segundo Marriott e Torres (2014, p. Mapas conceituais e autoria discente 183) “quando os mapas são gerados a partir de um texto, a leitura desse texto também passa a ser um exercício novo”, pois sua construção “promove uma mudança na maneira de estudar”. Com isso, o aluno deixa de ler de forma passiva e passa a identificar os conceitos relevantes e suas relações para gerar grupos Segundo Marriott e Torres (2014, p. 183) “quando os mapas são gerados a partir de um texto, a leitura desse texto também passa a ser um exercício novo”, pois sua construção “promove uma mudança na maneira de estudar”. Com isso, o aluno deixa de ler de forma passiva e passa a identificar os conceitos relevantes e suas relações para gerar grupos Segundo Marriott e Torres (2014, p. 183) “quando os mapas são gerados a partir de um texto, a leitura desse texto também passa a ser um exercício novo”, pois sua construção “promove uma mudança na maneira de estudar”. Com isso, o aluno deixa de ler de forma passiva e passa a identificar os conceitos relevantes e suas relações para gerar grupos 505 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 505 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) semânticos (MARRIOTT; TORRES, 2014). Essa maneira de se comportar diante da leitura do texto pode despertar no aluno processos de pensamentos, que por meio da interligação de conceitos novos com conceitos prévios, o faça ter as primeiras reflexões que possam contribuir para um pensamento crítico que desenvolva sua autoria. 505 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) semânticos (MARRIOTT; TORRES, 2014). Essa maneira de se comportar diante da leitura do texto pode despertar no aluno processos de pensamentos, que por meio da interligação de conceitos novos com conceitos prévios, o faça ter as primeiras reflexões que possam contribuir para um pensamento crítico que desenvolva sua autoria. Mesmo que o aluno se sinta inseguro e desconfortável no início na construção de mapas conceituais, de forma que inicie construindo mapas pequenos, sem tantas ramificações, e tenha dúvida em relação à relevância dos conceitos selecionados “essa inquietação faz parte do processo de aprendizagem e crescimento pessoal” (MARRIOTT; TORRES, 2014, p. 183). Logo, o aluno só irá aprender a fazer um mapa conceitual se der os primeiros passos para construí-lo, mesmo diante das incertezas e dificuldades, até que se desenvolva sua identidade autoral nos mapas seguintes. 4 Informações: <http://compendiuminstitute.net/download/download.htm> e < http://compendiumld.open.ac.uk/>. 5 Outras informações: <http://www.conzilla.org/>. 6 Licença Lesser General Public License. Detalhes: <https://opensource.org/licenses/lgpl-license>. 7 Maiores informações: <http://www.inspiration.com>. 8 Demais informações: <http://copyright.com.br/Direito-Autoral-Direito-Legal.html>. 9 Visual Understanding Environment. Informações: <http://vue.tufts.edu/>. 10 Detalhes da licença: <https://opensource.org/licenses/ecl2.php>. 11 Outras informações: <https://cmap.ihmc.us/cmaptools/> 12 Mais detalhes: <https://opensource.org/faq#free-software>. Mapas conceituais e autoria discente No contexto da Cultura Digital, a construção do mapa conceitual pode ser facilitada pela grande variedade de interfaces disponíveis online ou para ser instaladas no computador ou no celular, tais como: Compendium4 (desenvolvido em parte pela Open University Learning Design e financiada pela Open University) e Conzilla5 (desenvolvido pelo grupo KMR), nos quais possuem licença LGPL6; Inspiration7 (recurso comercial protegido pela Copyright8, possui versão para criança); Vue9 (Education Community License v2) 10 cujo desenvolvimento recebeu o suporte da Tufts University e o mais conhecido CmapTools11 (licença free12), no qual foi desenvolvido pelo Institute for Human and Machine Cognition (IHMC) em parceria com a West Florida University. É interessante observar que a maior parte dos recursos citados foi desenvolvida em parcerias com os cuidados de uma universidade, com exceção do Conzilla e do Inspiration, apesar disso esses recursos também têm seus objetivos voltados para o contexto educacional, especialmente com a aprendizagem. Entre os benefícios que os recursos digitais citados proporcionam, encontramos: destacar as elipses dos conceitos e as setas das relações por meio de diferentes espessuras, definir seus tipos, colori-los, arrastá-los para melhor harmonia do espaço e organização 4 Informações: <http://compendiuminstitute.net/download/download.htm> e < http://compendiumld.open.ac.uk/>. 5 Outras informações: <http://www.conzilla.org/>. 6 Licença Lesser General Public License. Detalhes: <https://opensource.org/licenses/lgpl-license>. 7 Maiores informações: <http://www.inspiration.com>. 8 Demais informações: <http://copyright.com.br/Direito-Autoral-Direito-Legal.html>. 9 Visual Understanding Environment. Informações: <http://vue.tufts.edu/>. 10 Detalhes da licença: <https://opensource.org/licenses/ecl2.php>. 11 Outras informações: <https://cmap.ihmc.us/cmaptools/> 12 Mais detalhes: <https://opensource.org/faq#free-software>. 506 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 506 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) cognitiva, apagá-los ou retornar ou avançar uma ação realizada. Desta forma, desenvolver mapa conceitual digital evita corriqueiros borrões quando construído/desenhado em papel e editado com lápis e borracha. Com o mapa conceitual construído, o aluno pode exportar em formato de imagem ou em formato de documento portátil (.pdf) e compartilhá-lo em um fórum de discussão sobre o tema, por exemplo. Assim, o aluno recebe feedback de seu mapa e após refleti-lo pode ressignificar seus conceitos e relações cognitivas e assim externar tais mudanças no mesmo arquivo salvo no qual o mapa foi construído anteriormente. Dessa forma, o mapa ganhará uma nova versão, atualizada. Produção de vídeos e autoria discente Além do mapa conceitual, o vídeo também pode ser trabalhado como outra possibilidade de autoria discente. Essa possibilidade, na presente pesquisa, refere-se apenas a vídeos curtos, cuja produção é viabilizada por meio de aplicativos que limitam o tempo de gravação e geram uma história, a partir da união sequencial desses vídeos. Dessa forma, conforme a sequência de envio, os pequenos vídeos são visualizados de forma contínua e sequencial. O tempo limitado do vídeo pode variar em 10, 15, 20, 30 ou 60 segundos, a depender do aplicativo. Outra característica presente nos aplicativos que produzem história é a duração de permanência do vídeo publicado, que pode variar entre 24h a 7 dias, em outras palavras, o vídeo desaparece após o tempo imposto. Logo, o conteúdo torna-se efêmero. Para Borges e Pacheco (2016) o conteúdo efêmero propicia que o aprendiz se apresse para assistir e se sinta mais à vontade na hora de postar, ao saber que o conteúdo desaparece. Dessa forma o aluno desenvolve a autonomia na organização de seu tempo, sendo responsável por cumprir atividades dentro dos prazos e evitando que se prolongue e atrase as atividades. Quanto ao tempo limitado do conteúdo produzido, faz-se lembrar do Twitter, um microblog, no qual a publicação é limitada em 280 caracteres (iniciou-se com 140). Essa condição elimina a preocupação de memorizar longos discursos e propicia que o usuário desenvolva uma informação curta e simples, exigindo dessa forma objetividade do texto. Tratando-se de vídeo curto, essa condição também pode beneficiar discursos mais objetivos e significativos, evitando rodeios de informações conceituais. Nesse sentido, Glance, Forsey e Riley (2013) afirmam que vídeos curtos podem ajudar a focar nos conceitos. Além do mais, como o conteúdo final produzido será uma história produzida a partir da união de partes pequenas, pequenos trechos do vídeo, isso contribuirá para que aluno se enxergue capaz de 507 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 507 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) desenvolver um conteúdo de vídeo, e por isso se torne autônomo dessa produção. Essa produção é individual, pois cada perfil de usuário produzirá sua própria história. Os aplicativos existentes ainda não permitem a formação de uma história produzida de forma coletiva, tratando-se apenas de uma autoria individual. Produção de vídeos e autoria discente 2017) e Story no próprio Facebook (mar. 2017). Um estudo comparativo desses aplicativos é realizado no trabalho de Ferreira e Costa (2018). Embora seja fácil encontrar na literatura bastante trabalho de autoria discente, a Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) FIGURA 2 – Lentes (máscaras): filtros animados aplicados em diferentes vídeos Fonte: Autores, 2019 FIGURA 2 – Lentes (máscaras): filtros animados aplicados em diferentes vídeos FIGURA 2 – Lentes (máscaras): filtros animados aplicados em diferentes vídeos Fonte: Autores, 2019 Fonte: Autores, 2019 Fonte: Autores, 2019 No contexto da Cultura Digital, a construção de vídeos curtos pode ser viabilizada por interfaces disponíveis para dispositivos móveis, especialmente smartphones. O SnapChat foi o primeiro aplicativo criado exclusivamente para formação de histórias por meio de vídeo curto associado ao uso de máscaras. No entanto, já existem outros aplicativos de diversas finalidades, que adicionaram a mesma função do SnapChat, temendo seu forte crescimento mundial de usuários ativos. Como exemplo, temos um lançamento sequencial de todos os aplicativos que têm o Facebook como proprietário: Stories do Instagram (ago. 2016), o Status13 do WhatsApp (fev. 2017) e Story no próprio Facebook (mar. 2017). Um estudo comparativo desses aplicativos é realizado no trabalho de Ferreira e Costa (2018). Produção de vídeos e autoria discente Contudo, nem todo aluno se sente confortável em publicar seu próprio vídeo e ser assistido, sobretudo se ele não está familiarizado com as TIC. Sendo assim, as máscaras podem atenuar esse desconforto, contribuindo para que os mais inibidos sejam encorajados a participar. Esse encorajamento pode impulsionar a autonomia do aluno. Além do mais, a escolha de uma máscara auxiliará o aluno a dar o primeiro passo para o desenvolvimento de sua identidade. Apenas o primeiro passo, porque a identidade não só é formada pela imagem que o aluno autor quer expor, mas também por outros estilos identitários, como a entonação da voz, articulação corporal, de palavras, a escolha de seu vocabulário e o conteúdo exposto. É quase unânime a disponibilidade de máscaras nos aplicativos que viabilizam a produção de conteúdo de vídeo curto em formato de história. Como é possível visualizar na figura 2, existem diversas máscaras a ser usadas conforme a identidade que o usuário quer expor no momento. Algumas das finalidades das máscaras podem ser visualizadas na figura 2. De forma geral, podemos citar: cobrir/esconder totalmente o rosto, por meio de um rosto de urso de pelúcia, de um personagem, pintura no rosto caracterizando animais ou de uma máscara com pele enrugada aparentando ser uma pessoa de idade avançada; disfarces como peruca caracterizando personagens como mangá, enfeites (chapéu e bigode fazendo alusão ao personagem Charlie Chaplin, flores no cabelo e orelhas de coelho, gato) acompanhados de melhoramento da imagem com efeitos de maquiagem conforme proposta da máscara utilizada. 508 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 508 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 508 508 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) FIGURA 2 – Lentes (máscaras): filtros animados aplicados em diferentes vídeos Fonte: Autores, 2019 No contexto da Cultura Digital, a construção de vídeos curtos pode ser viabilizada por interfaces disponíveis para dispositivos móveis, especialmente smartphones. O SnapChat foi o primeiro aplicativo criado exclusivamente para formação de histórias por meio de vídeo curto associado ao uso de máscaras. No entanto, já existem outros aplicativos de diversas finalidades, que adicionaram a mesma função do SnapChat, temendo seu forte crescimento mundial de usuários ativos. Como exemplo, temos um lançamento sequencial de todos os aplicativos que têm o Facebook como proprietário: Stories do Instagram (ago. 2016), o Status13 do WhatsApp (fev. 13 O Status do WhatsApp é o único aplicativo de vídeos curtos em formato de história que ainda não possui máscaras, em compensação torna-se um aplicativo que exige menos recursos de memória, tornado-a mais leve. Apesar disso, o Status aceita vídeos produzidos no MSQRD ou Snow, por exemplo, que são aplicativos de vídeo curto com máscaras, embora não possuam histórias e nem contatos para receber interação no próprio aplicativo. Embora seja fácil encontrar na literatura bastante trabalho de autoria discente, a maioria está associada à autoria textual. Sendo assim, esse artigo traz o diferencial de apresentar possibilidades de autoria que envolve duas linguagens diferentes (audiovisual e 13 O Status do WhatsApp é o único aplicativo de vídeos curtos em formato de história que ainda não possui máscaras, em compensação torna-se um aplicativo que exige menos recursos de memória, tornado-a mais leve. Apesar disso, o Status aceita vídeos produzidos no MSQRD ou Snow, por exemplo, que são aplicativos de vídeo curto com máscaras, embora não possuam histórias e nem contatos para receber interação no próprio aplicativo. 509 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 509 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) diagramática) da textual. Entre as linguagens indicadas na literatura (apresentada da seção anterior), a audiovisual é citada em quase todos os trabalhos, enquanto que a diagramática não foi cogitada. Apesar disso, existem trabalhos, não recentes, que citam ferramentas de autoria, como recursos digitais nos quais auxiliam a construção de mapa conceitual, mas a autoria em si, como por exemplo no artigo de Anacleto et al. (2008), não é investigado. Quanto aos artigos que trabalham com a linguagem audiovisual, não é considerada a dificuldade que os discentes podem encontrar ao produzir um vídeo de sua autoria, a expor sua identidade e a preocupação em gravar longos discursos sem gaguejar. Para isso, foi considerado nesse artigo utilizar aplicativos, como recursos digitais, que viabilizassem a produção de vídeos com máscaras e construídos em partes pequenas para ser compilados em formato de história. Dessa forma, os empecilhos para a produção de vídeo autoral por parte dos discentes podem ser atenuados. Apesar de existirem recursos digitais que promovam a autoria discente, por si só eles nunca serão eficientes se o aluno não se esforçar/motivar em busca de sua própria autonomia para que sua identidade autoral seja aos poucos construída. Para isso, o professor precisa auxiliar o seu uso e criar estratégias didáticas para que o recurso escolhido seja contextualizado conforme os objetivos de aprendizagem. Caso contrário, serão “utilizadas para continuar reproduzindo as velhas concepções pedagógicas da reprodução e do isolamento” (APARICI, 2012, p. 10). QUADRO 1 – Relação dos elementos da autoria versus recursos digitais Elementos da Autoria Quando ocorre? Identidade Autonomia Recursos Digitais Mapa Conceitual Ao expressar seu processo de pensamento, ou seja, seu próprio modo de formar e organizar conceitos, de forma a imprimir seu estilo com palavras e formas de expressão. Ao ler um texto de forma ativa: identificando os conceitos relevantes e refletindo na organização das relações desses conceitos. Vídeo Curto Ao escolher máscaras para sua exposição, na entonação da voz, articulação facial e de palavras e escolha do vocabulário. Ao organizar seu tempo para ser responsável por cumprir atividades efêmeras. Fonte: Autores, 2019 UADRO 1 – Relação dos elementos da autoria versus recursos digitais QUADRO 1 – Relação dos elementos da autoria versus recursos digitais Ao expressar seu processo de pensamento, ou seja, seu próprio modo de formar e organizar conceitos, de forma a imprimir seu estilo com palavras e formas de expressão. Ao escolher máscaras para sua exposição, na entonação da voz, articulação facial e de palavras e escolha do vocabulário. Fonte: Autores, 2019 Em um cenário em que a educação online dispõe, em sua maioria, de recursos de autoria que priorizem a linguagem textual, por meio de fórum de discussão, chat, portfólio, blog, os recursos digitais que utilizam linguagem diagramática e audiovisual mostraram-se também promissores quanto a possibilidade de autoria. Dessa forma, foi possível perceber a Em um cenário em que a educação online dispõe, em sua maioria, de recursos de autoria que priorizem a linguagem textual, por meio de fórum de discussão, chat, portfólio, blog, os recursos digitais que utilizam linguagem diagramática e audiovisual mostraram-se também promissores quanto a possibilidade de autoria. Dessa forma, foi possível perceber a Em um cenário em que a educação online dispõe, em sua maioria, de recursos de autoria que priorizem a linguagem textual, por meio de fórum de discussão, chat, portfólio, blog, os recursos digitais que utilizam linguagem diagramática e audiovisual mostraram-se também promissores quanto a possibilidade de autoria. Dessa forma, foi possível perceber a 510 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 510 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) relação existente dos elementos da autoria (identidade e autonomia) com os recursos digitais que utilizem mapa conceitual e vídeo curto, como expõe a síntese no Quadro 1. Nesse sentido, Palloff e Pratt (2004) reconhecem a importância de diversificar a linguagem textual, pois sugerem que quando se utiliza variadas linguagens são contempladas diferentes abordagens de aprendizagem. O artigo trabalhou os conceitos que envolvem a autoria no contexto da cultura digital, concluindo que a autoria digital requer autonomia e identidade nas contribuições e a morte do autor é necessária em um parâmetro figurado. Foi verificado que existe relação entre os elementos identidade e autonomia com os recursos digitais que utilizam mapa conceitual e vídeo curto, e por isso há possibilidade de autoria ao utilizá-los. Como maior parte dos recursos digitais utilizados na educação promovem a autoria por meio da linguagem textual, esse artigo apontou alternativas mostrando possibilidades de autoria utilizando a linguagem diagramática e audiovisual. Portanto, essas alternativas podem contemplar diferentes estilos de aprendizagem como a visual e verbal, respectivamente. Como trabalho futuro, pretende-se desenvolver experiências a fim de colher dados da autoria discente utilizando recursos digitais que vão além da linguagem textual. Referências ACEDO, S. O. Interatuantes e interatuados na web 2.0. In: APARICI, R. (Org.). Conectados no ciberespaço. São Paulo: Paulinas, 2012. p. 147-163. ACEDO, S. O. Interatuantes e interatuados na web 2.0. In: APARICI, R. (Org.). Conectados no ciberespaço. São Paulo: Paulinas, 2012. p. 147-163. ANACLETO, J. C. et al. Mapas Conceituais como Árvore Navegacional de Conteúdos Web: Um estudo de Caso sobre o Cognitor. In: Simpósio Brasileiro de Informática na Educação, 19, 2008, Fortaleza. Anais... Porto Alegre: Sociedade Brasileira de Computação, 2008. p. 420-429. Disponível em: <http://www.br- ie.org/pub/index.php/sbie/article/download/724/710>. Acesso em: 01 jul. 2018. ANACLETO, J. C. et al. Mapas Conceituais como Árvore Navegacional de Conteúdos Web: Um estudo de Caso sobre o Cognitor. In: Simpósio Brasileiro de Informática na Educação, 19, 2008, Fortaleza. Anais... Porto Alegre: Sociedade Brasileira de Computação, 2008. p. 420-429. Disponível em: <http://www.br- ie.org/pub/index.php/sbie/article/download/724/710>. Acesso em: 01 jul. 2018. APARICI, R. (org). Conectados no ciberespaço. São Paulo: Paulinas, 2012. AZEVEDO, I. C. M.; NASCIMENTO, D. S.; OLIVEIRA, V. M. S. A autoria na composição verbovocovisual de minicontos multimodais a partir de selfies. Fórum Linguístico, v. 13, n. 3, p. 1492-1505, 2016. Disponível em: < https://periodicos.ufsc.br/index.php/forum/article/download/1984- 8412.2016v13n3p1492/32714>. Acesso em: 17 jun. 2018. p p https://periodicos.ufsc.br/index.php/forum/article/download/1984- 8412.2016v13n3p1492/32714>. Acesso em: 17 jun. 2018. 511 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 511 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) BARATTO, S. S.; CRESPO, L. F. Cultura digital ou cibercultura: definições e elementos constituintes da cultura digital, a relação com aspectos históricos e educacionais. Revista Científica Eletrônica UNISEB, v. 1, n. 2, p. 16-25, 2013. Disponível em: <http://estacioribeirao.com.br/revistacientifica/arquivos/jul-2.pdf>. Acesso em: 29 jun. 2018. BENNETT, A. The author. New York: Routledge, 2005. BEZERRA, B. G. Usos da Linguagem em Fóruns de Ead. Revista Investigações, v. 24, n. 2, p. 11-33, 2011. Disponível em: < https://periodicos.ufpe.br/revistas/INV/article/view/1314/1000>. Acesso em: 20 jul. 2018. BORGES, T.; PACHECO, C. Modernos! Professores inovam e aderem ao Snapchat para incentivar alunos. Correio 24 horas, Salvador, 10 ago. 2016. Disponível em: <http://www.correio24horas.com.br/detalhe/enem/noticia/modernos-professores-inovam-e- aderem-ao-snapchat-para-incentivar-alunos/?cHash=07cf151deca62b5c96800d7db1c28d45>. Acesso em: 10 nov. 2016. BRAGA, S.; SENEM, J. O aluno na posição de autor: uma experiência com a resenha na universidade. Fórum Linguístico, v. 14, n. 4, p. 2685-2702, 2017. Disponível em: < https://periodicos.ufsc.br/index.php/forum/article/download/1984- 8412.2017v14n4p2685/35783>. Acesso em: 3 jun. 2018. BRAVO, C. B.; COSLADO, A. B. Uma geração de usuários da mídia digital. In: APARICI, R. (Org.). Conectados no ciberespaço São Paulo: Paulinas, 2012. p. 117-144. BROUSSEAU, G. Fondements et méthodes de la didactique dês mathématiques. Recherches Em Didactique dês Mathématiques, v. 7, n. 2, p. 33-115, 1986. BURKE, S. Reconstructing the author. In: Séan Burke. Authorship: from Plato to the postmodern. Edinburgh: Edinburgh University Press, 1995. CHARTIER, R. A aventura do livro: do leitor ao navegador. São Paulo: UNESP, 1999. CHAUÍ, M. Convite à filosofia. São Paulo: Ática, 2000. Disponível em: <https://edisciplinas.usp.br/pluginfile.php/1936981/mod_resource/content/3/aula%201_CHA U%C3%8D%2C%20Marilena.%20Convite%20%C3%A0%20Filosofia.pdf>. Acesso em: 31 mar. 2019. CUNHA, A. L. A interação na educação a distância: cuidados com o uso da linguagem em cursos online. In: Seminário Nacional ABED de Educação a Distância (SENAED), 4., 2006, v. 14, n. 2, Brasília. Anais...São Paulo: ABED, 2006. p.1-9. Disponível em: <www.abed.org.br/seminario2006/pdf/tc011.pdf>. Acesso em: 6 abr. 2018. FERREIRA, F. J. M.; COSTA, C. J. S. A. Um parâmetro atual da viabilidade de vídeo curto na Educação. In: Encontro Internacional Virtual Educa, 19, 2018, Bahia. Anais...Salvador: Virtual Educa, 2018,p. 1-20. Disponível em: <https://encuentros.virtualeduca.red/storage/ponencias/bahia2018/uDpS9dQGTzTl3gFMB3O Xr5RsTRjjkxg3KnJJZ8HO.pdf>. Acesso em: 1 jul. 2018. 512 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 512 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) FERREIRA, A. B. H. Dicionario da lingua portuguesa. Rio de janeira: Nova Fronteira, 1986. FERREIRA, A. B. H. Dicionario da lingua portuguesa. Rio de janeira: Nova Fronteira, 1986. FICHTE, J. G. Beweis der Unrechtmäßigkeit des Büchernachdrucks: Ein Räsonnement und eine Parabel. Berlinische Monatsschrift, 1793. Disponível em: <http://archiviomarini.sp.unipi.it/17/1/fichde.pdf>. Acesso em: 22 jun. 2018. FORESTI, A.; TEIXEIRA, A. C. Proposta de um conceito de aprendizagem para a era digital. Revista Latinoamericana de Tecnología Educativa - RELATEC, v. 11, n. 2, p. 55-68, 2012. Disponível em: <https://relatec.unex.es/article/view/919/667>. Acesso em: 24 jun. 2018. FORTUNATO, M.V. Autoria sob a materialidade do discurso. 2003. 148f. Dissertação (Mestrado em Educação). Universidade de São Paulo, São Paulo, 2003. Disponível em: < https://www.teses.usp.br/teses/disponiveis/48/48134/tde-13022004- 190509/publico/mfortunato.pdf>. Acesso em: 6 abr. 2019. FOUCAULT, M. O que é um autor? 3. ed. (Trad. António Fernando Cascais e Eduardo Cordeiro). Lisboa: Vega/Passagens, 1992. FOUCAULT, M. A Ordem do Discurso. 5. ed. Tradução: Laura Fraga de Almeida Sampaio. São Paulo: Loyola, 2000. FOUCAULT, M. A Arqueologia do Saber. 6. ed. Tradução: Luiz Felipe Baeta Neves. Rio de Janeiro: Forense Universitária, 2002. GLANCE, D. G.; FORSEY, M.; RILEY, M. The pedagogical foundations of massive open online courses. First Monday, v.18, n. 5, 2013. Disponível em: <http://firstmonday.org/article/view/4350/3673>. Acesso em: 4 jan. 2017. JOHNSON, S. A cultura da interface: como o computador transforma nossa maneira de criar e de comunicar. Rio de Janeiro: Zahar, 2001. KEMP, S. Digital in 2018: World’s internet users pass the 4 billion mark. We Are Social, Nova York, jan. 2018. Disponível em: <https://wearesocial.com/blog/2018/01/global-digital- report-2018>. Acesso em: 17 jun. 2018. LÉVI-STRAUSS, C. Linguistic and Anthropology. In: Claude Lévi-Strauss. Structural Anthropology, (p. 67-80). New York: Basic Books, 1963. Disponível em: < https://monoskop.org/images/e/e8/Levi-Strauss_Claude_Structural_Anthropology_1963.pdf>. Acesso em: 25 jun. 2018. LÉVY, P. Cibercultura. (Trad. Carlos Irineu da Costa). São Paulo: Editora 34, 2009. LIANG, L. A brief history of the Internet from the 15th to the 18th century. In: LOVINK, G.; TKACZ, N. (Orgs.). Critical point of view: A Wikipedia Reader. Amsterdam: Institute of Network Cultures, 2011. p. 50-62. Disponível em: <https://www.networkcultures.org/_uploads/%237reader_Wikipedia.pdf>. Acesso em: 18 jun. 2018. 513 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 513 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) MACHADO, M. A.; OSTERMANN, F. Unidades didáticas para a formação de docentes das séries iniciais do ensino fundamental. Desenvolvimento de material didático ou instrucional: Textos de Apoio ao Professor de Física, v. 17, n. 06, 2006, 74p. Porto Alegre: UFRGS. Disponível em: <http://www.if.ufrgs.br/ta/v17n6_Araujo_Ostermann.pdf>. Acesso em: 26 jun. 2018. MADDALENA, T. L.; SANTOS, E. #UERJRESISTE: relatos digitales de formación en tiempos de crisis. FERREIRA, A. B. H. Dicionario da lingua portuguesa. Rio de janeira: Nova Fronteira, 1986. Revista Observatório, v. 3, n. 4, p. 92-115, 2017. Disponível em: <https://sistemas.uft.edu.br/periodicos/index.php/observatorio/article/view/3488/9778>. Acesso em: 17 jun. 2018. MANTOVANI, O.; DIAS, M. H. P.; LIESENBERG, H. Conteúdos abertos e compartilhados: novas perspectivas para a educação. Sociedade & Educação, v. 27, n. 94, p. 257-276, 2006. Disponível em: <http://www.scielo.br/pdf/es/v27n94/a13v27n94.pdf>. Acesso em: 17 jun. 2018. MARRIOTT, R. C. V.; TORRES, P. L. Mapas Conceituais uma Ferramenta para a MARRIOTT, R. C. V.; TORRES, P. L. Mapas Conceituais uma Ferramenta para a Construção de uma Cartografia do Conhecimento. In: TORRES, P. L. (Org.). Complexidade - Redes e Conexões na Produção do Conhecimento. Curitiba: KAIRÓS Edições, 2014. p. 173- 211. Disponível em: <https://www.researchgate.net/profile/Patricia_Torres7/publication/271136094_Mapas_Conc eituais_uma_ferramenta_para_a_construcao_de_uma_cartografia_do_conhecimento/links/54b e973b0cf28ad7e718415c/Mapas-Conceituais-uma-ferramenta-para-a-construcao-de-uma- cartografia-do-conhecimento.pdf>. Acesso em: 26 jun. 2018. MARTINS, B. C. Autoria em rede: um estudo dos processos autorais interativos de escrita nas redes de comunicação. 2012. 155f. Tese (Doutorado em Ciências da Comunicação) – Universidade de São Paulo, São Paulo, 2012. Disponível em: < http://www.teses.usp.br/teses/disponiveis/27/27154/tde-31082012- 103436/publico/Beatriz_Martins_corrigida.pdf>. Acesso em: 12 jul. 2018. MENEZES, M. E. L. Tecnologias e mídias digitais no processo educativo e a autoria de alunos: limites, contribuições e possibilidades. 2013. 170f. Dissertação (Mestrado em Educação: currículo) - Pontifícia Universidade Católica de São Paulo, São Paulo, 2013. Disponível em: < https://tede2.pucsp.br/bitstream/.../Maria%20Eduarda%20de%20Lima%20Menezes.pdf>. Acesso em: 31 mar. 2019. MOREIRA, M. A.; MASINI, E. F. S. Aprendizagem Significativa: A Teoria de David Ausubel. São Paulo: Editora Moraes, 1982. ORLANDI, E. P. Discurso e leitura. Campinas: UNICAMP, 2001. PALLOFF, R.; PRATT, K. O aluno virtual. Porto Alegre: Artmed, 2004. PÊCHEUX, M. Semântica e Discurso: uma crítica à afirmação do óbvio. Tradução: Eni Pulcinelli Orlandi et al. 2 ed. Campinas: Unicamp, 1995. Disponível em: PÊCHEUX, M. Semântica e Discurso: uma crítica à afirmação do óbvio. Tradução: Eni Pulcinelli Orlandi et al. 2 ed. Campinas: Unicamp, 1995. Disponível em: 51 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 514 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) <https://www.scribd.com/doc/300138178/PECHEUX-Michel-Semantica-e-Discurso-uma- critica-a-afirmacao-do-obvio>. Acesso em 14 abr. 2019. <https://www.scribd.com/doc/300138178/PECHEUX-Michel-Semantica-e-Discurso-uma- critica-a-afirmacao-do-obvio>. Acesso em 14 abr. 2019. PETRONI, A. P.; SOUZA, V. L. T. Vigotski e Paulo Freire: contribuições para a autonomia do professor. Revista Diálogo Educacional, v. 9, n. 27, p. 351-361, 2009. Disponível em: <https://periodicos.pucpr.br/index.php/dialogoeducacional/article/download/3601/3517>. Acesso em: 31 mar. 2019. PETRONI, A. P.; SOUZA, V. L. T. As Relações na Escola e a Construção da Autonomia: um estudo da perspectiva da psicologia. Psicologia & Sociedade, v. 22, n. 2, p. 355-364, 2010. FERREIRA, A. B. H. Dicionario da lingua portuguesa. Rio de janeira: Nova Fronteira, 1986. Disponível em: < http://www.scielo.br/pdf/psoc/v22n2/16.pdf>. Acesso em: 10 jul. 2019. PIMENTA, L. A.; MOMESSO, M. R.; ASSOLINI, F. E. P. To be or not to be nas malhas híbridas do suporte digital YouTube: práticas de leitura, linguagens e o sujeito. Acta Scientiarum. Language and Culture (UEM), v. 38, n. 4, p. 385-395, 2016. Disponível em: <www.periodicos.uem.br/ojs/index.php/ActaSciLangCult/article/download/26426/pdf>. Acesso em: 17 jun. 2018. PINTO, A. C; BASTOS FILHO, J. B. Autoria, autonomia e ética na educação a distância. Perspectiva, v. 30, n. 1, p. 155-172, 2012. Disponível em: < https://periodicos.ufsc.br/index.php/perspectiva/article/download/2175- 795X.2012v30n1p155/21917>. Acesso em: 22 mar. 2018. POSSENTI, S. Indícios de Autoria. Perspectiva, v. 20, n. 1, p.105-124, 2002. Disponível em: <https://periodicos.ufsc.br/index.php/perspectiva/article/viewFile/10411/9677>. Acesso em: 2 abr. 2019. RICARDO, E. J.; VILARINHO, L. R. G. A construção da autoria na aprendizagem online: um desafio da pós-graduação. Revista Brasileira de Pós-Graduação, v. 3, n. 5, p. 59-78, 2006. Disponível em: < http://www.capes.gov.br/images/novo_portal/documentos/RBPG/Edicoes_revistas/v._3_n._5 _2006.zip>. Acesso em: 28 ago. 2019. ROMERA, D. L. Influencia de la Web 2.0 en el uso de las herramientas de autoría multimedia: un estúdio preliminar. Revista de Comunicación, n. 7, p. 92-109, 2014. Disponível em: <http://revistesdigitals.uvic.cat/index.php/obradigital/article/view/48/49>. Acesso em: 25 jun. 2018. SANTAELLA, L. Linguagens liquidas na era da mobilidade. São Paulo: Paulus, 2007. SANTAROSA, L. M. C.; CONFORTO, D.; BASSO, L. O. Eduquito: ferramentas de autoria e de colaboração acessíveis na perspectiva da web 2.0 Revista Brasileira de Educação Especial, v. 18, n. 3, p. 449-468, 2012. Disponível em: <http://www.scielo.br/pdf/rbee/v18n3/a07.pdf>. Acesso em: 17 jun. 2018. TRENTIN, M. A. S.; TEIXEIRA, A. C.; FELDKIRCHER, C. Software de autoria colaborativa de materiais educacionais para a TV digital. Revista Brasileira de Computação Aplicada, v. 4, n. 1, p. 92-101, 2012. Disponível em: < http://seer.upf.br/index.php/rbca/article/download/2115/1599>. Acesso em: 24 jun. 2018. 515 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 515 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) 515 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) WEISSBERG, J.-L. Entre produção e recepção: hipermediação, uma mutação dos saberes simbólicos. In: COCCO, G.; GALVÃO, A. P.; SILVA, G. (Orgs.). Capitalismo cognitivo: trabalho, rede e inovação. Rio de Janeiro: DP&A, 2003. p. 109-131. ZAMBONI, C. Homem se passa por policial federal e aplica golpes em Brumadinho. Tropical FM, Treze Tílias - SC, 4 fev. 2019. Brasil, notícia. Disponível em: <https://www.tropicalfm99.com.br/noticia/18392/homem-se-passa-por-policial-federal-e- aplica-golpes-em-brumadinho>. Acesso em: 2 abr. 2019. Recebido em: 30.08.2019 Aceito em: 06.12.2019 515 Teresina, Ano 24, n. FERREIRA, A. B. H. Dicionario da lingua portuguesa. Rio de janeira: Nova Fronteira, 1986. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) WEISSBERG, J.-L. Entre produção e recepção: hipermediação, uma mutação dos saberes simbólicos. In: COCCO, G.; GALVÃO, A. P.; SILVA, G. (Orgs.). Capitalismo cognitivo: trabalho, rede e inovação. Rio de Janeiro: DP&A, 2003. p. 109-131. 515 Teresina, Ano 24, n. 43, set./dez. 2019. ISSN 2526-8449 (Eletrônico) 1518-0743 (Impresso) WEISSBERG, J.-L. Entre produção e recepção: hipermediação, uma mutação dos saberes simbólicos. In: COCCO, G.; GALVÃO, A. P.; SILVA, G. (Orgs.). Capitalismo cognitivo: trabalho, rede e inovação. Rio de Janeiro: DP&A, 2003. p. 109-131. WEISSBERG, J.-L. Entre produção e recepção: hipermediação, uma mutação dos saberes simbólicos. In: COCCO, G.; GALVÃO, A. P.; SILVA, G. (Orgs.). Capitalismo cognitivo: trabalho, rede e inovação. Rio de Janeiro: DP&A, 2003. p. 109-131. ZAMBONI, C. Homem se passa por policial federal e aplica golpes em Brumadinho. Tropical FM, Treze Tílias - SC, 4 fev. 2019. Brasil, notícia. Disponível em: <https://www.tropicalfm99.com.br/noticia/18392/homem-se-passa-por-policial-federal-e- aplica-golpes-em-brumadinho>. Acesso em: 2 abr. 2019. R bid 30 08 2019 ZAMBONI, C. Homem se passa por policial federal e aplica golpes em Brumadinho. Tropical FM, Treze Tílias - SC, 4 fev. 2019. Brasil, notícia. Disponível em: <https://www.tropicalfm99.com.br/noticia/18392/homem-se-passa-por-policial-federal-e- aplica-golpes-em-brumadinho>. Acesso em: 2 abr. 2019. Recebido em: 30.08.2019 Aceito em: 06.12.2019 aplica-golpes-em-brumadinho>. Acesso em: 2 abr. 2019. Recebido em: 30.08.2019 Aceito em: 06.12.2019 Recebido em: 30.08.2019 Aceito em: 06.12.2019
https://openalex.org/W3022388663	https://research-information.bris.ac.uk/ws/files/238113364/s41598_020_64761_3.pdf	English	null	Fluconazole resistance in Candida albicans is induced by Pseudomonas aeruginosa quorum sensing	Scientific reports	2,020	cc-by	13,863	Bandara, H. M. H. N., Wood, D. L. A., Vanwonterghem, I., Hugenholtz, P., Cheung, B. P. K., & Samaranayake, L. P. (2020). Fluconazole resistance in Candida albicans is induced by Pseudomonas aeruginosa quorum sensing. Scientific Reports, 10(1), Article 7769 (2020). https://doi.org/10.1038/s41598-020-64761-3 Bandara, H. M. H. N., Wood, D. L. A., Vanwonterghem, I., Hugenholtz, P., Cheung, B. P. K., & Samaranayake, L. P. (2020). Fluconazole resistance in Candida albicans is induced by Pseudomonas aeruginosa quorum sensing. Scientific Reports, 10(1), Article 7769 (2020). https://doi.org/10.1038/s41598-020-64761-3 Publisher's PDF, also known as Version of record License (if available): CC BY Link to published version (if available): 10.1038/s41598-020-64761-3 This is the final published version of the article (version of record). It first appeared online via Springer Nature at https://www.nature.com/articles/s41598-020-64761-3. Please refer to any applicable terms of use of the publisher. Bandara, H. M. H. N., Wood, D. L. A., Vanwonterghem, I., Hugenholtz, P., Cheung, B. P. K., & Samaranayake, L. P. (2020). Fluconazole resistance in Candida albicans is induced by Pseudomonas aeruginosa quorum sensing. Scientific Reports, 10(1), Article 7769 (2020). https://doi.org/10.1038/s41598-020-64761-3 Bandara, H. M. H. N., Wood, D. L. A., Vanwonterghem, I., Hugenholtz, P., Cheung, B. P. K., & Samaranayake, L. P. (2020). Fluconazole resistance in Candida albicans is induced by Pseudomonas aeruginosa quorum sensing. Scientific Reports, 10(1), Article 7769 (2020). https://doi.org/10.1038/s41598-020-64761-3 OPEN g q g H. M. H. N. Bandara1 ✉, D. L. A. Wood2, I. Vanwonterghem2, P. Hugenholtz2, B. P. K. Cheun & L. P. Samaranayake4 Microorganisms employ quorum sensing (QS) mechanisms to communicate with each other within microbial ecosystems. Emerging evidence suggests that intraspecies and interspecies QS plays an important role in antimicrobial resistance in microbial communities. However, the relationship between interkingdom QS and antimicrobial resistance is largely unknown. Here, we demonstrate that interkingdom QS interactions between a bacterium, Pseudomonas aeruginosa and a yeast, Candida albicans, induce the resistance of the latter to a widely used antifungal fluconazole. Phenotypic, transcriptomic, and proteomic analyses reveal that P. aeruginosa’s main QS molecule, N-(3- Oxododecanoyl)-L-homoserine lactone, induces candidal resistance to fluconazole by reversing the antifungal’s effect on the ergosterol biosynthesis pathway. Accessory resistance mechanisms including upregulation of C. albicans drug-efflux, regulation of oxidative stress response, and maintenance of cell membrane integrity, further confirm this phenomenon. These findings demonstrate that P. aeruginosa QS molecules may confer protection to neighboring yeasts against azoles, in turn strengthening their co-existence in hostile polymicrobial infection sites. Microbial communities residing within the human body, either transiently or permanently, play a pivotal role in human health and disease1. In particular, interkingdom polymicrobial infections due to pathogenic fungi and bacteria are relatively common and are seen in the oral cavity, respiratory tract, gastrointestinal system, skin, and urinary tract1,2. For instance, the focal fungal pathogen of our study, Candida albicans, contributes to >50% of the total microbial burden in mixed fungal-bacterial chronic wound infections, and has been frequently co-isolated with bacterial pathogens including Pseudomonas aeruginosa and Staphylococcus aureus3–5. C. albicans is consid- ered an independent risk factor for ventilator associated pneumonia and co-exists with P. aeruginosa in 26% of these infections6. When superinfected with Candida, the prognosis of P. aeruginosa infections in cystic fibrosis lungs are significantly poorer compared to the bacterial infection alone7,8. Alarmingly, 27–56% of nosocomial C. albicans blood stream infections are associated with Staphylococcus epidermidis, S. aureus and Enterococcus spe- cies9. Moreover, Candida spp. are co-isolated with vaginal streptococci in 20–34% of recurrent vulvovaginal can- didiasis, and with oral streptococci in 50–75% of denture stomatitis cases10–12. Candidal-bacterial polymicrobial infections are responsible for a high incidence of mortality and morbidity due to their increased dissemination, antimicrobial resistance, and the lack of sensitive diagnostics9,13. University of Bristol – Bristol Research Portal General rights This document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/red/research-policy/pure/user-guides/brp-terms/ www.nature.com/scientificreports Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ and virulence, bacterial sporulation, formation of fungal fruiting bodies, conjugal plasmid transfer and anti- microbial resistance, and antibiotic synthesis15–17. QS interactions can occur among microbes from the same species (intraspecies QS), different species (interspecies QS) or even between members of different kingdoms (interkingdom QS)14. However, most studies have focused on intra- and interspecies QS, and our understanding of interkingdom QS is limited. g Q Candida QS interactions with the respiratory pathogen P. aeruginosa have been extensively studied due to their frequent co-isolation in cystic fibrosis lungs, wound infections, indwelling devices and nosocomial infections18–20. Farnesol, a major QSM secreted by C. albicans, is known to supress P. aeruginosa by inhibiting its homoserine lac- tone synthesis that leads to subsequent reduction in bacterial swarming, and pyocyanin and quinolone signaling (PQS, 2-alkyl-4-quinolones)15,20–22. Farnesol also acts on C. albicans itself by inhibiting hyphal development (fil- amentation) through repression of adenylyl cyclase (Cyr1p) in the Ras1–cyclic AMP–protein kinase A pathway, which positively regulates hyphal growth23. In addition, farnesol triggers cellular oxidative stress and apoptosis in C. albicans. Exposure to azole antifungal agents significantly increases farnesol synthesis in C. albicans24,25, and recent studies have shown that nonlethal concentrations of farnesol enhance the efficacy of azole antifun- gals by suppressing ABC multidrug efflux transporters and accumulating reactive oxygen species (ROS)26,27. Interestingly, among the wide array of QSMs secreted by P. aeruginosa, N-(3-Oxododecanoyl)-L-homoserine lactone (C16H27NO4, C12AHL) has a significant structural resemblance to farnesol28,29. Therefore, C12AHL also inhibits C. albicans hyphal development using the same mechanism as farnesol28,29. However, despite being struc- turally similar to farnesol, the effects of C12AHL on C. albicans’ cellular mechanisms upon exposure to antifun- gal agents, including multidrug efflux activity, cellular fitness, and ergosterol synthesis (the molecular target of azoles), are largely unknown.l ) g y We recently demonstrated that the co-delivery of C12AHL with fluconazole in a liposomal drug carrier increases the efficacy of the antifungal agent in elimination of C. albicans biofilms. However, free forms of drug + C12AHL failed to demonstrate similar antifungal efficacy30 suggesting the effects of C12AHL on C. albicans upon exposure to antifungal agents are drug and C12AHL formulation dependent. Owing to the recognized clinical importance of Pseudomonas-Candida interactions in various pathological states, lack of synergistic effects of free C12AHL + fluconazole on C. albicans biofilms observed in our recent study30, and the sparsity of data on the role of Pseudomonas QSMs on C. Results C lbi C. albicans sensitivity to fluconazole decreases in the presence of C12AHL. We hypothesised that C12AHL would make C. albicans more sensitive to fluconazole due to its known inhibitory properties on the yeast, therefore the minimum inhibitory concentrations (MIC50 and MIC80) for fluconazole in the presence and absence of C12AHL was determined. Unexpectedly, the MIC50 of fluconazole exhibited a 16-fold increase in the presence of 100 µg mL−1 C12AHL (0.156 µg mL−1 vs 2.5 µg mL−1, Supplementary Fig. S1 and Supplementary Table S2) and 8-fold increase in the presence of 12.5–50 µg mL−1 C12AHL (0.156 µg mL−1 vs 1.25 µg mL−1, Supplementary Fig. S1). No MIC80 of fluconazole was observed when C. albicans was exposed to the antifungal agent alongside C12AHL within the concentration ranges assessed in this study. Therefore, MIC80 of fluconazole appears to increase more than 8-fold when treated with C12AHL with a concentration range of 12.5–100 µg mL−1 (1.25 vs >10 µg mL−1, Supplementary Fig S1 and Supplementary Table S2). C12AHL demonstrated a 20% maxi- mum inhibition of C. albicans growth when treated with 100 µg mL−1. C12AHL stimulates the multidrug efflux activity of C. albicans. Efflux of antifungal drugs via trans- port proteins is one of the main mechanisms employed by C. albicans when developing antifungal resistance31. Therefore, the activity of transport proteins was assessed in the presence of various treatment groups. When exposed to C12AHL or C12AHL + fluconazole, C. albicans pumped out significantly higher quantities of the indicator dyeR6G compared to fluconazole treated or the solvent (DMSO) controls for up to an exposure period of 24 h (Fig. 1A,B, P < 0.05). C12AHL + fluconazole exposure showed significantly higher R6G efflux compared to C12AHL treated C. albicans in the early stages of the exposure (up to 1 h of observation, Fig. 1A, P < 0.05). However, the latter difference appeared to wane during prolonged exposure to the QSM ± antifungal (up to 24 h of exposure, Fig. 1B, P > 0.05). In contrast, C. albicans exposed to fluconazole alone did not show any notable changes of rhodamine efflux compared to the control (Fig. 1A,B, P > 0.05).fl gfl p ( g ) Drug efflux pumps in C. albicans are mainly encoded by CDR1, CDR2 and MDR132, therefore, respective mutant strains were used to verify the efflux activity observed with the indicator dye. Efflux pump mutant strains of C. www.nature.com/scientificreports/ albicans antifungal sensitivity/resistance, we evaluated the cellu- lar and molecular responses of C. albicans on in vitro exposure to a widely-used anti-fungal fluconazole in the presence of the QSM C12AHL. We assessed the minimum inhibitory concentration (MIC) of the active agents (Fluconazole, C12AHL, C12AHL + fluconazole) using broth dilution assay with a checkerboard approach. C. albicans’ multidrug efflux pump activity when exposed to the active agents was quantified by measuring the efflux of an indicator dye, rhodamine 6 g (Rhodamine 6 g Assay) and further verified based on the expression of genes coding for efflux pumps by qPCR. Changes in the C. albicans transcriptome in response to the active agents were assessed using next generation sequencing (RNA-Seq) and their effect on yeast protein synthesis was evaluated via two-dimensional gel electrophoresis and mass spectrometry. We demonstrate that P. aeruginosa C12AHL induces C. albicans’ fluconazole resistance through multiple mechanisms, predominantly by facilitating fungal ergosterol synthesis and restoring its cell wall integrity. OPEN Hence, fungal-bacterial interkingdom infections represent an, as yet, understudied health issue warranting further investigation.h The severity and outcome of polymicrobial infections are dictated not only by the nature and the composi- tion of the constituent microbiota, but also by the chemical communications between co-habitants. Quorum sensing (QS) is a universal chemical messenger system used by microorganisms to interact with each other. QS is defined as a cell-density dependent, coordinated gene expression in microbial communities in response to threshold concentrations of specific chemical signalling molecules (quorum sensing molecules; QSMs) leading to a synchronized population response14. QS is essential for microbes to optimize their survival in dynamic, con- stantly challenging niches, as the chemical messengers help correlate individual cellular functions to microbial community-based requirements14. These include regulation of biofilm development and maturation, motility 1Oral Microbiology, Bristol Dental School, University of Bristol, Lower Maudlin Street, Bristol, BS1 2LY, UK. 2Australian Centre for Ecogenomics, School of Chemistry and Molecular Biosciences, The University of Queensland, St Lucia, QLD 4072, Australia. 3Faculty of Dentistry, The University of Hong Kong, 34 Hospital Rd, Sai Ying Pun, Hong Kong SAR, China. 4College of Dental Medicine, The University of Sharjah, P.O. Box, 27272, Sharjah, UAE. ✉e-mail: nihal.bandara@bristol.ac.uk www.nature.com/scientificreports/ Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 Results C lbi albicans, cdr1Δ (DSY448), mdr1Δ (DSY465), cdr1Δ/cdr2Δ (DSY653), and cdr1Δ/cdr2Δ/mdr1Δ (DSY1050) demonstrated a 2.0 to 2.8-fold increase in rhodamine 6 g efflux when exposed to C12AHL or C12AHL + fluconazole for 1 h compared to the solvent control (Fig. 1C, P < 0.05). Exposure to fluconazole alone failed to increase R6G efflux significantly (Fig. 1C P > 0.05). Functions of the efflux pumps are known to be affected by the composition of cell membrane sterols (ergosterol in particular)33,34. Therefore, the mutant strains Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Figure 1. Mean drug efflux activity of C. albicans when exposed to C12AHL, fluconazole or their combination. (A) 1 h exposure, (B) 24 h exposure. Note the extracellular accumulation of significantly higher quantities of rhodamine 6 g when exposed to C12AHL or C12AHL + fluconazole compared to solvent control or fluconazole exposed samples. Standard deviations are presented as error bars (n = 18) (C) drug efflux activity of C. albicans mutants for a period of 1 h. There was a 2.0 to 2.8-fold increase in rhodamine 6 g efflux when CDR and MDR mutants were exposed to C12AHL or C12AHL + fluconazole for 1 h compared to solvent control, and insignificant changes in the ERG mutants (p-value < 0.05). C. albicans CAF2–1, the parental strain of the mutants tested, is included for comparison purposes. Figure 1. Mean drug efflux activity of C. albicans when exposed to C12AHL, fluconazole or their combination. (A) 1 h exposure, (B) 24 h exposure. Note the extracellular accumulation of significantly higher quantities of rhodamine 6 g when exposed to C12AHL or C12AHL + fluconazole compared to solvent control or fluconazole exposed samples. Standard deviations are presented as error bars (n = 18) (C) drug efflux activity of C. albicans mutants for a period of 1 h. There was a 2.0 to 2.8-fold increase in rhodamine 6 g efflux when CDR and MDR mutants were exposed to C12AHL or C12AHL + fluconazole for 1 h compared to solvent control, and insignificant changes in the ERG mutants (p-value < 0.05). C. albicans CAF2–1, the parental strain of the mutants tested, is included for comparison purposes. of ERG11 and ERG3, the genes encoding rate limiting enzymes in ergosterol synthesis and sterol intermediates synthesis pathways31 were used to verify the role of sterols and the efflux activity observed with the indicator dye. Ergosterol mutant strains of C. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 Results C lbi albicans, erg3Δ (DSY1751), erg11Δ (DSY1769), erg3Δ erg11Δ (DSY1764) did not exhibit significant changes in efflux activity when exposed to any of the treatments for 1 h compared to the solvent control (Fig. 1C, P > 0.05). C12AHL modulates the transcriptomic response of C. albicans when exposed to flucona- zole. Transcriptomic sequencing was performed to determine which molecular mechanisms of C. albicans are modified in the presence of the QSM C12AHL, the antifungal fluconazole, or the combination of these molecules. First, an overall comparison of gene expression profiles was performed to assess whether there was an effect of the type of treatment. Then, differentially expressed genes were assessed for each treatment relative to the sol- vent control [(Fluconazole vs Control), (C12AHL vs Control), and (C12AHL + fluconazole vs Control)], as well as between treatments [(Fluconazole vs C12AHL), (Fluconazole vs C12AHL + fluconazole), and (C12AHL vs C12AHL + fluconazole)] to determine significant up- and/or downregulation of genes across the various treat- ments groups (adjusted p-value < 1e−5) (Supplementary Tables S3-S7). There was a significant influence of the type of treatment on the gene expression, with expression pro- files for the fluconazole alone treated C. albicans being significantly different from the control, C12AHL and C12AHL + fluconazole samples (PERMANOVA, p-value < 0.05). This can also be observed graphically in a principle component analysis (Supplementary Fig. S8), where the fluconazole treated samples are statistically distinguishable from the control and other treatments in the second principle component (PC2), which accounts for almost a quarter of the total variation in the data. These results suggest that the effects of fluconazole were being ameliorated in the presence of C12AHL, and this is further confirmed when looking at the differentially expressed genes. Results C lbi Multiple genes were up- or downregulated in each treatment relative to the control samples, Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Gene Name Annotation Fluconazole treated compared to control: log2FoldChange Adjusted P value C12AHL treated compared to control: log2FoldChange Adjusted P value CSH1 Aldo-keto reductase 3.36 5.11E-23 1.87 5.61E-07 C3_03460C_A Protein of unknown function 3.35 5.16E-23 3.19 4.58E-20 C1_01510W_A Protein of unknown function 2.78 1.81E-16 2.46 2.65E-12 C1_04010C_A IFD1/IFD6 alias Protein with a NADP-dependent oxidoreductase domain 2.67 1.06E-19 1.68 1.47E-07 LPG20 Aldo-keto reductase family protein 2.44 4.3E-13 1.65 8.38E-06 CRH11 GPI-anchored cell wall transglycosylase 2.30 7.62E-08 −2.36 1.36E-07 WOR4 Predicted C2H2 zinc finger protein 2.09 7.17E-12 2.40 1.18E-14 NRG1 Transcription factor/repressor 1.86 1.93E-12 2.18 4.02E-16 IFD6 Aldo-keto reductase 1.75 6.62E-08 3.02 1.11E-21 SRR1 Two-component system response regulator 1.62 1.27E-06 2.24 1.98E-11 C2_01750C_A Protein with a life-span regulatory factor domain 1.33 8.99E-07 1.51 4.57E-08 RGS2 Protein of RGS superfamily 1.29 1.1E-07 1.39 3.28E-08 BCR1 Transcription factor 1.26 3.25E-07 1.21 3.28E-06 CR_10230W_A Histone acetyltransferase activity 0.85 3.82E-06 1.36 2.94E-14 CDR4 Putative ABC transporter superfamily −1.43 2.18E-08 1.70 6.09E-11 PHHB Putative 4a-hydroxytetrahydrobiopterin dehydratase −1.62 5.69E-07 −1.72 2.87E-07 IFR2 Zinc-binding dehydrogenase −1.79 2.02E-08 −1.78 8.07E-08 RME1 Zinc finger protein −1.95 4.98E-07 2.17 3.28E-08 ATX1 Putative cytosolic copper metallochaperone −2.22 8.01E-10 2.75 2.42E-14 Table 1. Gene expression data; The C. albicans genes that were significantly affected by both fluconazole only and C12AHL only exposure compared to untreated controls [(Fluconazole vs Control) vs (C12AHL vs Control)]. Genes highlighted in blue are known to be associated with antifungal sensitivity/resistance. (http:// www.candidagenome.org/). Table 1. Gene expression data; The C. albicans genes that were significantly affected by both fluconazole only and C12AHL only exposure compared to untreated controls [(Fluconazole vs Control) vs (C12AHL vs Control)]. Genes highlighted in blue are known to be associated with antifungal sensitivity/resistance. (http:// www.candidagenome.org/). including the genes involved in ergosterol synthesis and antifungal resistance (Fig. 2, Supplementary Fig. S9 and S10, Supplementary Tables S3-S7). Genes were also differentially expressed between treatments with the excep- tion of C12AHL vs C12AHL + fluconazole (Tables 1–3, Fig. 3, Supplementary Figs. S9 and S10). To determine the mechanisms of how C12AHL may be amending the effects of the antifungal molecule, we examined the differen- tially expressed pathways in more detail. C12AHL modulates gene expression in the C. albicans ergosterol synthesis pathway upon exposure to fluconazole. The C. Table 1. Gene expression data; The C. albicans genes that were significantly affected by both fluconazole onl and C12AHL onl e pos re compared to ntreated controls [(Fl conazole s Control) s (C12AHL Results C lbi albicans genes were collated into 156 different molecular pathways using published data from the Candida Genome Database. Once grouped, two molecular pathways in particular showed significant differences when comparing treatments, i.e. the ergosterol biosynthesis pathway and the pentose phos- phate pathway. Comparison of genes involved in the ergosterol biosynthesis pathway (as shown in Fig. 4) revealed significant upregulation of ERG1, ERG2, ERG4, ERG5, ERG6, ERG10, ERG11, ERG24, ERG26, and ERG27 in flu- conazole treated samples compared to solvent controls, C12AHL treated, and C12AHL + fluconazole treated C. albicans. ERG3 was upregulated in fluconazole treated C. albicans compared to solvent controls, while ERG7 and ERG9 were upregulated in fluconazole treated C. albicans compared to C12AHL + fluconazole treated samples (Fig. 4). Genes involved in the oxidative branch of the pentose phosphate pathway (SOL3, GND1, ZWF1), were significantly upregulated in fluconazole treated C. albicans compared to C12AHL treated and C12AHL + flucona- zole treated C. albicans (Fig. 5). C12AHL alters protein expression in C. albicans exposed to fluconazole. In order to further assess the findings of the transcriptomic analyses, protein expression of C. albicans exposed to fluconazole with/without C12AHL was investigated. According to 2-dimensional electrophoresis and mass spectrometric data, addition of C12AHL led to differential expression of a variety of C. albicans proteins in the presence of fluconazole. A total of 17 under-expressed (Supplementary Table S11) and seven over-expressed (Supplementary Table S12) pro- teins were identified in the C12AHL + fluconazole treated C. albicans compared to the fluconazole only control (p-value < 0.05).l p Several proteins that are known to be induced by fluconazole and/or other antifungals exposure (Gcy1p, Lsc1p, Pda1p, Atp1p, Mxr1p and Ach1p) were identified in fluconazole only treated C. albicans whereas they were absent in C12AHL + fluconazole treated samples (Supplementary Tables S11 and S12). C12AHL upregulates multidrug efflux pumps-coding genes in C. albicans exposed to fluconazole. Surprisingly, there was no indication of significantly increased expression of multidrug efflux pump genes at Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ w ntificreports/ Figure 2. Comparison of gene expression profiles between each tre showing RNA-Seq data from each treatment [(A) C12AHL, (B) flu relative to the control. The dashed lines represent the cut-off values (=2) to identify significantly different gene expression. The plots ar differentially expressed genes are represented in grey, those with lo with p-value < 10−6 are coloured in blue, and those with both log2 f in red. Results C lbi Genes that represent proteins involved in the ergosterol bios plots. Figure 2. Comparison of gene expression profiles between each treatment and the control. Volcano plots showing RNA-Seq data from each treatment [(A) C12AHL, (B) fluconazole and (C) C12AHL + fluconazole] relative to the control. The dashed lines represent the cut-off values for p-value (=10−6) and log2 fold change (=2) to identify significantly different gene expression. The plots are coloured so that non-significant differentially expressed genes are represented in grey, those with log2 fold change >2 are shown in green, genes with p-value < 10−6 are coloured in blue, and those with both log2 fold change >2 and p-value < 10−6 are shown in red. Genes that represent proteins involved in the ergosterol biosynthesis pathway have been labelled in the plots. Figure 2. Comparison of gene expression profiles between each treatment and the control. Volcano plots showing RNA-Seq data from each treatment [(A) C12AHL, (B) fluconazole and (C) C12AHL + fluconazole] relative to the control. The dashed lines represent the cut-off values for p-value (=10−6) and log2 fold change (=2) to identify significantly different gene expression. The plots are coloured so that non-significant differentially expressed genes are represented in grey, those with log2 fold change >2 are shown in green, genes with p-value < 10−6 are coloured in blue, and those with both log2 fold change >2 and p-value < 10−6 are shown in red. Genes that represent proteins involved in the ergosterol biosynthesis pathway have been labelled in the plots. Figure 2. Comparison of gene expression profiles between each treatment and the control. Volcano plots showing RNA-Seq data from each treatment [(A) C12AHL, (B) fluconazole and (C) C12AHL + fluconazole] relative to the control. The dashed lines represent the cut-off values for p-value (=10−6) and log2 fold change (=2) to identify significantly different gene expression. The plots are coloured so that non-significant differentially expressed genes are represented in grey, those with log2 fold change >2 are shown in green, genes with p-value < 10−6 are coloured in blue, and those with both log2 fold change >2 and p-value < 10−6 are shown in red. Genes that represent proteins involved in the ergosterol biosynthesis pathway have been labelled in the plots. Results C lbi Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Gene Name Annotation C12AHL-Fluconazole treated compared to control: log2FoldChange Adjusted P value Fluconazole treated compared to control: log2FoldChange Adjusted P value C3_03460C_A Protein of unknown function 3.45 3.54E-23 3.35 5.16E-23 IFD6 Aldo-keto reductase 3.00 2.14E-21 1.75 6.62E-08 ATX1 Putative cytosolic copper metallochaperone 2.74 3.92E-14 −2.22 8.01E-10 C1_01510W_A Protein of unknown function 2.41 1.41E-11 2.78 1.81E-16 SRR1 Two-component system response regulator 2.15 1.81E-10 1.62 1.27E-06 WOR4 Predicted C2H2 zinc finger protein 2.08 6.53E-11 2.09 7.17E-12 RME1 Zinc finger protein 1.98 9E-07 −1.95 4.98E-07 NRG1 Transcription factor/repressor 1.95 9.01E-13 1.86 1.93E-12 CSH1 Aldo-keto reductase 1.91 3.78E-07 3.36 5.11E-23 CDR4 Putative ABC transporter superfamily 1.76 1.68E-11 −1.43 2.18E-08 LPG20 Aldo-keto reductase family protein 1.71 3.87E-06 2.44 4.3E-13 C1_04010C_A Protein with a NADP-dependent oxidoreductase domain 1.57 1.54E-06 2.67 1.06E-19 RGS2 Protein of RGS superfamily 1.17 7.97E-06 1.29 1.1E-07 CR_10230W_A Histone acetyltransferase activity 1.07 8.15E-09 0.85 3.82E-06 ENA21 Predicted P-type ATPase sodium pump −1.59 9.11E-06 −2.00 1.19E-09 PHHB Putative 4a-hydroxytetrahydrobiopterin dehydratase −1.78 1.04E-07 −1.62 5.69E-07 IFR2 Zinc-binding dehydrogenase −1.82 4.01E-08 −1.79 2.02E-08 Table 2. Gene expression data; The C. albicans genes that were significantly affected by both fluconazole only and C12AHL + fluconazole exposure compared to untreated controls [(Fluconazole vs Control) vs (C12AHL + fluconazole vs Control)]. Genes highlighted in blue are known to be associated with antifungal resistance/sensitivity. (http://www.candidagenome.org/). Table 2. Gene expression data; The C. albicans genes that were significantly affected by both fluconazole only and C12AHL + fluconazole exposure compared to untreated controls [(Fluconazole vs Control) vs (C12AHL + fluconazole vs Control)]. Genes highlighted in blue are known to be associated with antifunga resistance/sensitivity. (http://www.candidagenome.org/). either the RNA or protein level after 24 h exposure to fluconazole in the presence of C12AHL. This is despite the observed increase in indicator dye efflux activity of the C. albicans strain in the presence of C12AHL + flucona- zole. To understand whether there was any involvement of multidrug efflux pump genes on indicator dye efflux activity at the early stages of C12AHL + fluconazole exposure (1 h), the expression of these genes was assessed using qPCR32. Exposure to C12AHL + fluconazole for 1 h significantly upregulated C. albicans CDR1 (2.4-fold) and CDR2 (5-fold) genes compared to solvent controls (Supplementary Fig. S13, P < 0.05). C12AHL treated C. albicans also showed upregulation of CDR2 (1.9-fold), however, CDR1 and MDR1 were downregulated com- pared to solvent controls. Results C lbi In contrast, both CDR genes were downregulated when exposed to fluconazole alone (CDR1:2.5 fold and CDR2:1.4-fold). MDR1 expression was not affected by C12AHL + fluconazole treatment. Discussion In host-associated environments, fungi and bacteria interact physically (e.g. co-aggregation) and chemically (e.g. quorum sensing), thereby impacting their immediate surroundings as well as the host immune response35. Such mutualistic/synergistic interactions have evolved to facilitate epithelial cohabitation and efficient use of met- abolic by-products, while competitive antagonistic approaches have also developed during co-colonization36. Although the fundamental role of QS in optimizing microbial survival in polymicrobial environments has been well-studied, very little is known of the interactions and effects of interkingdom QS systems during antimicrobial therapy. Here we studied the effect of the P. aeruginosa quorum sensing molecule, C12AHL, on the cellular and molecular responses of C. albicans when exposed to the anti-fungal molecule fluconazole.h p p gl The role of C12AHL in interactions between C. albicans and P. aeruginosa has been increasingly studied in recent years. For example, it is known that P. aeruginosa cells preferentially adhere to C. albicans hyphae using their surface adherence proteins37,38. Using C12AHL defective mutants, Ovchinnikova et al. (2012) demon- strated that C12AHL is essential for the expression of P. aeruginosa adhesion proteins, and therefore is critical for adherence to C. albicans hyphae in a mixed fungal-bacterial environment37. Previous studies based on gas chromatography-mass spectrometric analyses, have reported that P. aeruginosa biofilms may contain over 600 µM (~178.5 µg mL−1) of peak C12AHL28,39. It is also noteworthy that QSM concentrations within the microbial popu- lations vary depending on the different stages of biofilm growth, with lower concentrations during early stages of biofilm formation and biofilm dispersal, and higher concentrations during maturation. Subsequent studies have further suggested that >200 μM of C12AHL (59.5 µg mL−1) suppresses C. albicans hyphal growth whilst >500 μM (148.5 µg mL−1) of C12AHL inhibits fungal growth completely28. Hence, from the range of concentrations tested, we used sub-growth and sub-hyphal inhibitory concentration (50 µg mL−1, 168 µM) of C12AHL throughout our study. The inability to form hyphae with concentrations of C12AHL used (12.5–100 µg mL−1) was confirmed via a germ tube assay (Data not shown). Discussion Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Gene Name Annotation C12AHL- Fluconazole treated compared to control: log2FoldChange Adjusted P value C12AHL treated compared to control: log2FoldChange Adjusted P value C3_03460C_A Protein of unknown function 3.45 3.54E-23 3.19 4.58E-20 IFD6 Aldo-keto reductase 3.00 2.14E-21 3.02 1.11E-21 UGT51C1 UDP-glucose:sterol glucosyltransferase 2.20 1.95E-19 2.38 1.45E-22 CRZ2 C2H2 transcription factor 2.30 2.41E-14 2.51 2.86E-17 ATX1 Putative cytosolic copper metallochaperone 2.74 3.92E-14 2.75 2.42E-14 ALK2 N-Alkane inducible cytochrome P450 1.93 2.75E-13 1.63 1.09E-09 NRG1 Transcription factor/repressor 1.95 9.01E-13 2.18 4.02E-16 C1_01510W_A Protein of unknown function 2.41 1.41E-11 2.46 2.65E-12 CDR4 Putative ABC transporter superfamily 1.76 1.68E-11 1.70 6.09E-11 YOR1 ABC-type plasma membrane transporter 1.35 3.45E-11 1.41 2.33E-12 WOR4 Predicted C2H2 zinc finger protein 2.08 6.53E-11 2.40 1.18E-14 ADA2 Zinc finger and homeodomain transcriptional coactivator 1.51 8.98E-11 1.67 2.43E-13 C6_02100W_A Secreted protein −3.73 1.15E-10 −3.54 9.31E-10 OPT3 Oligopeptide transporter 2.16 1.39E-10 2.15 1.60E-10 SRR1 Two-component system response regulator 2.15 1.81E-10 2.24 1.98E-11 C7_04090C_A Predicted mitochondrial cardiolipin-specific phospholipase 2.33 2.33E-10 2.35 1.59E-10 C1_09210C_A Putative transporter 1.72 2.42E-10 1.47 8.96E-08 C3_05450C_A Protein of unknown function 2.38 4.54E-10 2.53 2.17E-11 ZCF1 Zn(II)2Cys6 transcription factor 1.43 6.29E-10 1.59 2.33E-12 RFG1 HMG domain transcriptional repressor 1.81 7.32E-10 1.94 3.12E-11 MOH1 Ortholog of S. cerevisiae Moh1 1.61 2.14E-09 1.68 2.64E-10 CR_09100C_A Aldo-keto reductase 2.30 4.70E-09 2.35 1.36E-09 FBA1 Fructose-bisphosphate aldolase −1.79 5.31E-09 −1.64 8.92E-08 EFG1 bHLH transcription factor 1.88 7.64E-09 1.97 9.31E-10 CR_10230W_A Ortholog(s) have histone acetyltransferase activity 1.07 8.15E-09 1.36 2.94E-14 HSP78 Heat-shock protein 2.23 1.33E-08 2.22 1.27E-08 RPN4 C2H2 transcription factor 1.34 1.56E-08 1.40 3.13E-09 SNQ2 Protein similar to S. cerevisiae Snq2p transporter 1.74 1.56E-08 1.55 6.73E-07 CR_06140W_A Protein of unknown function 1.96 1.56E-08 1.98 1.08E-08 C7_00770W_A Protein of unknown function 2.19 1.73E-08 1.81 6.81E-06 HSP104 Heat-shock protein 2.54 1.83E-08 2.47 3.99E-08 GOR1 Ortholog(s) have glyoxylate reductase activity 2.16 1.86E-08 2.00 2.37E-07 IFR2 Zinc-binding dehydrogenase −1.82 4.01E-08 −1.78 8.07E-08 AAF1 Possible regulatory protein 2.13 5.63E-08 2.18 2.21E-08 YIM1 Protein similar to protease of mitochondrial inner membrane 1.95 6.00E-08 2.00 2.04E-08 PHHB Putative 4a-hydroxytetrahydrobiopterin dehydratase −1.78 1.04E-07 −1.72 2.87E-07 ERO1 Ortholog of S. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 Discussion Genes highlighted in blue are known to be associated with antifungal resistance/sensitivity. (http://www.candidagenome.org/). The 16-fold measured increase in the MIC of fluconazole indicates that C. albicans sensitivity to flucona- zole decreases when it is simultaneously exposed to the antifungal and C12AHL (100 µg mL−1). Indeed, even at lower concentrations (12.5–50 µg mL−1) C12AHL induced fluconazole resistance in C. albicans compared to fluconazole-treated controls (Fig. 1A, p-value < 0.05). In our latest published work, we witnessed this lack of inhibitory properties of C12AHL and fluconazole when delivered to C. albicans biofilms as free forms without encapsulating into a drug carrier molecule (i.e. liposomes)30. Although the exact mechanism is yet to decipher, the synergy we observed with fluconazole and C12AHL in the form of liposomes is likely to be attributed to the characteristics of the liposomal formulation (i.e. enhanced penetration), slow and sustained drug release, the encapsulated dosage of the active agents and the intrinsic properties of Candida biofilm phenotype. This evidence further supported aforementioned hypothesis of fluconazole and C12AHL antagonism. In line with this finding, recent studies have also shown an increase in C. albicans’ antifungal resistance in C. albicans-P. aeruginosa mixed species biofilms via upregulated C. albicans proteins associated with drug resistance and virulence40. However, it is known that C. albicans exhibits distinctive responses to antifungal agents and the environmental stresses (such as to bacterial QSMs) based on its physiological forms (i.e. biofilm vs planktonic phases). Therefore, in part, the lack of C. albicans inhibition observed with C12AHL ± fluconazole in our study compared to the existing Candida biofilm literature is likely to be associated the variations in the experimental design as this study focuses on planktonic phenotype, the usage of sub-growth and sub-hyphal inhibitory concentrations of C12AHL and other unknown fungal mechanisms. Accordingly, cautions must be taken when generalised inferences are made. g g y g Intracellular R6G indicator dye accumulation is commonly used to identify azole-resistant C. albicans strains as it has been shown that the retention of R6G within fungal cells is inversely correlated with the expression of the drug efflux pump protein, Cdr1p, in C. albicans41,42. Our R6G assay reveals for the first time that exposure to the QSM C12AHL can upregulate the efflux activity of C. albicans. The presence of fluconazole in combination with C12AHL further enhances this increased efflux activity, indicating a potential mechanism of C12AHL-mediated fluconazole resistance in the yeast. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 Discussion cerevisiae Ero1 1.91 1.20E-07 2.00 1.82E-08 MNN42 Protein of unknown function 1.25 1.48E-07 1.43 6.88E-10 HAL9 Putative Zn(II)2Cys6 transcription factor 1.35 2.48E-07 1.40 5.22E-08 ADH3 Putative NAD-dependent (R,R)-butanediol dehydrogenase 3.05 2.81E-07 3.25 2.36E-08 INO2 Transcriptional activator 1.76 3.13E-07 1.92 1.24E-08 CSH1 Aldo-keto reductase 1.91 3.78E-07 1.87 5.61E-07 C6_00290W_A Protein of unknown function 1.54 4.63E-07 1.67 2.04E-08 GRP2 Methylglyoxal reductase 2.50 4.86E-07 2.62 8.07E-08 ISA1 Putative mitochondrial iron-sulfur protein 1.39 7.93E-07 1.55 1.62E-08 RME1 Zinc finger protein 1.98 9.00E-07 2.17 3.28E-08 C1_04010C_A Protein with a NADP-dependent oxidoreductase domain 1.57 1.54E-06 1.68 1.47E-07 GAL102 UDP-glucose 4,6-dehydratase 1.85 1.54E-06 2.05 3.84E-08 C5_04030W_A Protein of unknown function 1.85 2.15E-06 2.08 3.99E-08 C4_03020W_A Putative mitochondrial GTPase 0.98 2.64E-06 1.18 4.87E-09 Continued Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Gene Name Annotation C12AHL- Fluconazole treated compared to control: log2FoldChange Adjusted P value C12AHL treated compared to control: log2FoldChange Adjusted P value ALS7 ALS family protein 0.95 2.83E-06 1.12 1.07E-08 MNN12 Predicted alpha-1,3-mannosyltransferase −1.90 2.89E-06 −1.93 1.55E-06 C3_02630C_A Protein of unknown function 1.20 3.87E-06 1.42 1.29E-08 SIS1 Putative Type II HSP40 co-chaperone 1.98 3.87E-06 1.94 5.25E-06 LPG20 Aldo-keto reductase family protein 1.71 3.87E-06 1.65 8.38E-06 HSP90 Essential chaperone 1.71 3.87E-06 1.85 2.86E-07 MHP1 Protein similar to S. cerevisiae Mhp1p 1.10 3.93E-06 1.09 4.69E-06 C3_00360W_A Protein of unknown function 1.43 4.78E-06 1.88 2.21E-10 C1_03990W_A Ortholog(s) have proteasome binding activity 1.64 6.48E-06 1.95 2.07E-08 CR_06960W_A Ortholog(s) have ATP binding, DNA replication origin binding activity 1.29 6.48E-06 1.35 1.73E-06 C1_01130W_A Putative ubiquitin ligase complex component 1.06 7.23E-06 1.09 2.80E-06 RGS2 Protein of RGS superfamily 1.17 7.97E-06 1.39 3.28E-08 CR_07480W_A Predicted auxin family transmembrane transporter 1.02 9.33E-06 1.22 2.41E-08 C4_02740W_A Protein of unknown function 1.48 9.33E-06 1.54 2.80E-06 Table 3. Gene expression data; The C. albicans genes that were significantly affected by both C12AHL only and C12AHL + fluconazole exposure compared to untreated controls [(C12AHL vs Control) vs (C12AHL + fluconazole vs Control)]. Genes highlighted in blue are known to be associated with antifungal resistance/sensitivity. (http://www.candidagenome.org/). Table 3. Gene expression data; The C. albicans genes that were significantly affected by both C12AHL only and C12AHL + fluconazole exposure compared to untreated controls [(C12AHL vs Control) vs Table 3. Gene expression data; The C. albicans genes that were significantly affected by both C12AHL only and C12AHL + fluconazole exposure compared to untreated controls [(C12AHL vs Control) vs (C12AHL + fluconazole vs Control)]. Table 3. Gene expression data; The C. albicans genes that were significantly affected by both C12AHL l d C12AHL+fl l d t t t d t l [(C12AHL C t l) Discussion Unlike C12AHL, its structural analogue farnesol produced by C. albicans is known to inhibit C. albicans drug efflux during fluconazole exposure, thereby potentiating the activity of the anti- fungal drug26,27. This result demonstrates the functional diversity of C12AHL and farnesol, despite their structural resemblance43.fl C. albicans possesses three major plasma membrane drug efflux pump proteins: Cdr1p, Cdr2p [ATP-binding cassette (ABC) pumps] and Mdr1p [the major facilitator superfamily (MFS) transporters] that are known to regulate azole efflux32. Using a set of isogenic C. albicans strains lacking the genes for one or more of these drug efflux pumps (CDR1, CDR2 and MDR1), Mukherjee et al. concluded that drug efflux pumps play a significant role in candidal resistance to fluconazole in early planktonic and biofilm phases (0–6 h)44. Our qPCR data indicated that short-term exposure (1 h) of C. albicans (azole-sensitive strain) to fluconazole in the presence of C12AHL upregulates the expression of CDR1 and CDR2. This observation supports a previous study that showed upreg- ulation of CDR1 in C. albicans biofilms when exposed to P. aeruginosa secretory factors45. Taken together, these findings suggest that C12AHL ± fluconazole triggers phenotypic and transcriptional changes in multidrug efflux mechanisms in C. albicans within a short period of exposure (~1 h), suggesting a potential mechanism of early azole resistance. Further mechanistic investigations on C. albicans early antifungal resistance are necessary to confirm this hypothesis.fli i yp In contrast to previous reports on the temporal nature of the efflux activity and the absence of signifi expression of CDR1 and CDR2 in the latter stages of planktonic/biofilm phases (e.g. 24 h), we noted that www.nature.com/scientificreports/ w ntificreports/ Figure 3. Comparison of gene expression profiles between the dif comparison of RNA-Seq data between the different treatments [(A fluconazole vs C12AHL, and (C) fluconazole vs C12AHL + flucon off values for p-value (=10−6) and log2 fold change (=2) to identify plots are coloured so that non-significant differentially expressed g fold change >2 are shown in green, genes with p-value < 10−6 are c fold change >2 and p-value < 10−6 are shown in red. Genes that re biosynthesis pathway have been labelled in the plots. Figure 3. Comparison of gene expression profiles between the different treatments. Volcano plots showing the comparison of RNA-Seq data between the different treatments [(A) C12AHL vs C12AHL + fluconazole, (B) fluconazole vs C12AHL, and (C) fluconazole vs C12AHL + fluconazole]. Discussion ERG1: Squalene epoxidase, ERG2: C-8 sterol isomerase, ERG3: C-5 sterol desaturase, ERG4: sterol C-24 reductase, ERG5: C-22 sterol desaturase, ERG6: Delta (24)-sterol C-methyltransferase, ERG7: 2,3-epoxysqualene-lanosterol cyclase (lanosterol synthase), ERG8: phosphomevalonate kinase, ERG9: farnesyl-diphosphate farnesyl transferase (squalene synthase), ERG10: Acetyl-CoA acetyltransferase, ERG11: Lanosterol 14-alpha-demethylase, ERG12: mevalonate kinase, ERG13: 3-hydroxy-3-methylglutaryl coenzyme A synthase, ERG20: farnesyl pyrophosphate synthetase, ERG24: C-14 sterol reductase, ERG25: C-4 methyl sterol oxidase, ERG26: C-3 sterol dehydrogenase, ERG27: 3-Keto sterol reductase, HMG1: HMG-CoA reductase, IDI1: isopentenyl-diphosphate delta-isomerase, and MVD: Mevalonate diphosphate decarboxylase. efflux activity in C. albicans remained significantly higher for 24 h upon exposure to C12AHL ± fluconazole. Interestingly, transcriptomic data did not show significant changes in CDR1, CDR2, MDR1, FLU1 or TAC1 gene expression after 24 h treatment. In addition, we also noted that C. albicans cdr1Δ, mdr1Δ, cdr1Δ/cdr2Δ, and cdr1Δ/cdr2Δ/mdr1Δ strains continued to transport R6G outside of the cell, albeit to a lesser degree compared to the wild type, despite the respective mutations. Hence, the persistent efflux activity observed here is likely regu- lated via another mechanism(s).fl efflux activity in C. albicans remained significantly higher for 24 h upon exposure to C12AHL ± fluconazole. Interestingly, transcriptomic data did not show significant changes in CDR1, CDR2, MDR1, FLU1 or TAC1 gene expression after 24 h treatment. In addition, we also noted that C. albicans cdr1Δ, mdr1Δ, cdr1Δ/cdr2Δ, and cdr1Δ/cdr2Δ/mdr1Δ strains continued to transport R6G outside of the cell, albeit to a lesser degree compared to the wild type, despite the respective mutations. Hence, the persistent efflux activity observed here is likely regu- lated via another mechanism(s).fl Fungal ABC superfamily efflux pump proteins (Cdr1p in particular) are sensitive to imbalanced lipid compo- sition in the yeast plasma membrane, and their mutual interactions with membrane sterols (mainly ergosterol) are critical for the sorting and functioning of efflux pump proteins in C. albicans33,34. Therefore, we hypothesized that increased efflux pump activity of C. albicans when exposed to C12AHL ± fluconazole is likely to be asso- ciated with changes in ergosterol biosynthesis. In order to test this, we assessed the R6G efflux activity of two mutant C. albicans strains with knockout of key genes involved in the ergosterol biosynthesis pathway (ERG11 and ERG3). Interestingly, we noted that erg11Δ, erg3Δ and erg3Δ/erg11Δ strains were incapable of altering their efflux activity in response to 1 h exposure to C12AHL ± fluconazole, in contrast to the wild type. Discussion The dashed lines represent the cut- off values for p-value (=10−6) and log2 fold change (=2) to identify significantly different gene expression. The plots are coloured so that non-significant differentially expressed genes are represented in grey, those with log2 fold change >2 are shown in green, genes with p-value < 10−6 are coloured in blue, and those with both log2 fold change >2 and p-value < 10−6 are shown in red. Genes that represent proteins involved in the ergosterol biosynthesis pathway have been labelled in the plots. Figure 3. Comparison of gene expression profiles between the different treatments. Volcano plots showing the comparison of RNA-Seq data between the different treatments [(A) C12AHL vs C12AHL + fluconazole, (B) fluconazole vs C12AHL, and (C) fluconazole vs C12AHL + fluconazole]. The dashed lines represent the cut- off values for p-value (=10−6) and log2 fold change (=2) to identify significantly different gene expression. The plots are coloured so that non-significant differentially expressed genes are represented in grey, those with log2 fold change >2 are shown in green, genes with p-value < 10−6 are coloured in blue, and those with both log2 fold change >2 and p-value < 10−6 are shown in red. Genes that represent proteins involved in the ergosterol biosynthesis pathway have been labelled in the plots. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Figure 4. C. albicans molecular pathways analyses; The genes in the ergosterol biosynthesis pathway that are affected by fluconazole, C12AHL or C12AHL + fluconazole exposure. Comparisons denoted with * are significant (adjusted p-value < 1e−5). ERG1: Squalene epoxidase, ERG2: C-8 sterol isomerase, ERG3: C-5 sterol desaturase, ERG4: sterol C-24 reductase, ERG5: C-22 sterol desaturase, ERG6: Delta (24)-sterol C-methyltransferase, ERG7: 2,3-epoxysqualene-lanosterol cyclase (lanosterol synthase), ERG8: phosphomevalonate kinase, ERG9: farnesyl-diphosphate farnesyl transferase (squalene synthase), ERG10: Acetyl-CoA acetyltransferase, ERG11: Lanosterol 14-alpha-demethylase, ERG12: mevalonate kinase, ERG13: 3-hydroxy-3-methylglutaryl coenzyme A synthase, ERG20: farnesyl pyrophosphate synthetase, ERG24: C-14 sterol reductase, ERG25: C-4 methyl sterol oxidase, ERG26: C-3 sterol dehydrogenase, ERG27: 3-Keto sterol reductase, HMG1: HMG-CoA reductase, IDI1: isopentenyl-diphosphate delta-isomerase, and MVD: Mevalonate diphosphate decarboxylase. Figure 4. C. albicans molecular pathways analyses; The genes in the ergosterol biosynthesis pathway that are affected by fluconazole, C12AHL or C12AHL + fluconazole exposure. Comparisons denoted with * are significant (adjusted p-value < 1e−5). Discussion This suggests that the effect of C12AHL ± fluconazole on efflux pump activity may be associated with changes in the ergosterol synthesis pathway. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 10 www.nature.com/scientificreports/ Figure 5. C. albicans molecular pathways analyses; Genes in the oxidative branch of the pentose phosphate pathway that are affected by fluconazole, C12AHL or C12AHL + fluconazole exposure. Comparisons denoted with * are significant (adjusted p-value < 1e−5). ZWF1: Glucose-6-phosphate dehydrogenase, GND1: 6-phosphogluconate dehydrogenase, and SOL3: 6-phosphogluconolactonase. Figure 5. C. albicans molecular pathways analyses; Genes in the oxidative branch of the pentose phosphate pathway that are affected by fluconazole, C12AHL or C12AHL + fluconazole exposure. Comparisons denoted with * are significant (adjusted p-value < 1e−5). ZWF1: Glucose-6-phosphate dehydrogenase, GND1: 6-phosphogluconate dehydrogenase, and SOL3: 6-phosphogluconolactonase. Azoles and many other antifungal drugs primarily target ergosterol biosynthesis in C. albicans. The sterol bio- synthesis pathway possesses three distinct sub-pathways; mevalonate, late and alternate pathways (Supplementary Fig. S14). The mevalonate pathway, the first step in the sterol synthesis process, entails the production of farnesyl pyrophosphate (FPP) from acetyl-coenzyme A (acetyl-CoA)46. The resulting FPP is fed into many different cel- lular pathways as it is an essential intermediate in the biosynthesis of sterols (i.e. ergosterol), heme, ubiquinone, dolichol, and prenylated proteins46,47. The pathway responsible for the catalysation of FPP to synthesise ergosterol is identified as the late pathway. When antifungal agents such as azoles interfere with the late pathway, it branches out to the alternate pathway that produces sterol intermediates instead of ergosterol. Some of these sterol inter- mediates are known to be toxic and their intracellular accumulation arrests cell growth48,49. Fluconazole suppresses C14α-demythylase encoded by ERG11 in the late pathway, which normally catalyzes lanosterol to C14-demethyl-lanosterol and would ultimately lead to the synthesis of ergosterol. Suppression of ERG11 reroutes the late pathway to the alternate pathway by expressing C24 methyl transferase (ERG6) with the synthesis of various sterol intermediates as a result. One particular intermediate is the toxic compound 14α-methyl-3,6-diol, which is catalysed by C5 desaturase (ERG3) in the final step and ultimately arrests fungal growth46. Ergosterol is a major sterol component of the yeast cell wall and mitochondrial membrane, and is vital in maintaining membrane fluidity and permeability, enzyme activity, cell cycle progression and cell morphol- ogy50. In addition, sterols and sphingolipids together form lipid rafts, i.e. Discussion a type of microdomain located in the fungal cell membrane, that is enriched with numerous molecules such as efflux pumps, sodium and potassium pumps, receptors, and nutrient transporters51,52. p p p p In this study, gene and protein expression data provided strong evidence to suggest that C12AHL mediated induction of fluconazole resistance in C. albicans is associated with ergosterol biosynthesis. Previous studies have established that prolonged exposure to azoles (fluconazole, itraconazole, ketoconazole, clotrimazole, and micona- zole) can trigger over expression of ERG11 and other genes associated with the alternate pathway of sterol syn- thesis (ERG9, ERG1, ERG7, ERG3), particularly during the logarithmic growth phase of the yeast50,53,54. Our gene expression data also demonstrated similar findings, for example, all genes of both the late and alternate pathways (except ERG25) of sterol biosynthesis were significantly upregulated when C. albicans was exposed to fluconazole but remained unaffected with either C12AHL + fluconazole or C12AHL exposure. Therefore, these results sug- gest that the effect of fluconazole on C. albicans’ late and alternate pathways of sterol synthesis is suppressed in the presence of C12AHL. Functional investigations using relevant key mutant strains of the ergosterol synthesis pathway could provide valuable mechanistic insights to further support the observed changes in gene expression. Th h l h l bi h i h l d i b h i i fi p y p g pp g g p The enzymes that catalyse the sterol biosynthesis pathway are regulated in part by the zinc-cysteine finger transcription factor paralogs Upc2p in C. albicans55,56. Upc2p senses sterol levels within the yeast and when these levels are reduced, for example due to fluconazole interference, it activates genes for sterol biosynthesis and uptake55,56. Our gene expression data confirmed significant upregulation of UPC2 (codes for Upc2p) in flucona- zole treated C. albicans as a result of fluconazole-mediated inhibition of ergosterol synthesis. This finding explains why not only genes in the alternate pathway were upregulated in the fluconazole treated samples, but there was also indirect upregulation of the genes in the late pathway. Experiments using UPC2 mutant C. albicans could be performed to further confirm this observation.l pi Notably, neither C12AHL nor C12AHL + fluconazole treated samples elicited the changes observed in the presence of fluconazole alone, indicating that the regulation of sterol biosynthesis/uptake in the yeast in the presence of the QSM was unaffected. www.nature.com/scientificreports/ treated C. albicans. UGT51C1 codes for UDP-glucose:sterol glucosyltransferase that catalyses the biosynthesis of sterol glycosides from ergosterol59. Upregulation of this gene indicates that there is likely to be a continual supply of ergosterol as substrate to the enzyme, thus further supporting the hypothesis of unaffected ergosterol synthesis in C. albicans by fluconazole when C12AHL is present. yl p C. albicans can use three antioxidant systems (i.e. catalase, thioredoxin and glutathione) and two major oxidative stress signalling pathways (i.e. Cap1 and Hog 1) to respond to oxidative stress induced by antifun- gals. Oxidative stress induced by antifungals stimulates NADPH production in C. albicans via the oxidative branch of the pentose phosphate pathway (PPP)60–62. NADPH is an essential cofactor for glutathione- and thioredoxin-dependent enzymes in antioxidant systems (thioredoxin and glutathione, respectively) that neutral- ize reactive oxygen species (ROS)63–65. Therefore, the oxidative branch of the PPP is critical for fungal survival against oxidative stress66,67. Glucose-6-phosphate-1-dehydrogenase coded by ZWF1 regulates the rate limiting first step of the oxidative branch of PPP68 and the gene expression profiles from this experiment showed signifi- cant upregulation of genes in the oxidative arm of the PPP, in particular ZWF1, in the fluconazole treated samples. This effect was however not observed when C12AHL was present. In addition, as observed in the protein expres- sion data, several key proteins that play a role in protecting the fungus from oxidative stress, Sod1p (superoxide dismutase69), Pst1p (Flavodoxin-like protein70), Mxr1p (methionine sulfoxide reductase62, Adh4p (alcohol dehy- drogenase71), and Cyp5p (Peptidyl-prolyl cis-trans isomerase72), were downregulated in C. albicans treated with C12AHL + fluconazole compared to fluconazole alone treated samples. These results suggest that the presence of C12AHL prevents the oxidative stress otherwise imposed by fluconazole on the yeast cells.ii p p yl y Another interesting finding was that the presence of C12AHL appears to increase the overall fitness of C. albicans when challenged with fluconazole. For instance, we observed significant upregulation of genes GAL102, C2_00770W_A, and DAG7 that lower the sensitivity of the yeast to toxic sterol analogues accumulated via the alternate pathway73,74. Similarly, GAL102 plays an important role in yeast cell wall synthesis and resistance to antifungal drugs by stabilizing the cell wall73. Upregulation of GAL102 together with other genes that are known to regulate yeast cell wall synthesis and repair (i.e. www.nature.com/scientificreports/ INO2, ADA2, PHR1 and MNN12) may further indicate that the presence of C12AHL prevents the impact of fluconazole on yeast cell wall integrity and improves the overall cellular fitness75,76.f i In summary, our data suggest that the presence of C12AHL favorably affects C. albicans challenged with fluconazole by preventing changes in sterol biosynthesis, increasing drug efflux pump activity, reducing the oxi- dative stress response, and maintaining yeast cell membrane integrity. These conclusions are largely based on our transcriptomic data; therefore, appropriate functional assessments as indicated are necessary to verify these claims. Further investigations on sterol analyses (including total cellular sterol and sterol intermediates) as well as changes in plasma membrane composition of the yeast are necessary to confirm this hypothesis. In addition, recent studies have highlighted some of the complex interactions between C. albicans and P. aeruginosa in pol- ymicrobial infections. For example, certain compounds produced by C. albicans, that remain to be characterized, have been shown to stimulate the synthesis of virulence factors (e.g. phenazine production) by Pseudomonas spp., as well as to reduce swarming motility which leads to enhanced biofilm development20,77. Therefore, further inves- tigations on C. albicans and P. aeruginosa cocultured in a polymicrobial biofilm environment would be beneficial to understand the specific interactions between these two microorganisms when exposed to fluconazole. Selective physical interactions between P. aeruginosa and C. albicans filaments, together with mutual inhibitory and benefi- cial effects of the QSMs C12AHL and farnesol, speak to the importance of co-existence and the interdependence of P. aeruginosa and C. albicans for their survival in mixed microbial communities. Hence, the core finding of our study, that C12AHL induces antifungal resistance in C. albicans, thereby protecting the fungal population, is likely to be another control mechanism employed by P. aeruginosa in optimizing its survival in challenging polymicrobial environments. Discussion This is further evidenced by no significant change in the expression of ECM33 relative to the control, which codes for protein molecules within lipid rafts that are sensitive to changes in the cell membrane composition. The maintenance of lipid rafts is critically important for proper functioning of a variety of cellular processes, cell signalling, protein sorting, virulence, stress responses, and environmental adaptations33,34,51,57. ECM33 is known to be significantly upregulated during exposure to fluconazole as observed in our data58. We also noted an upregulation of UGT51C1 in C12AHL + fluconazole treated but not in fluconazole Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Gene expression analyses Gene expression analyses Next generation sequencing (RNA-Seq). Changes in the C. albicans transcriptome were assessed with next generation sequencing (RNA-Seq). C. albicans SC5314 standard suspension was prepared as described above, treated with either fluconazole, C12AHL, C12AHL + fluconazole, or DMSO and incubated at 37 °C stat- ically for 24 h. Cells were washed 3 times in PBS and total RNA was extracted using the SV total RNA isolation system (Catalog No. Z3100, Promega, Madison, WI). Three biological replicates were processed for each treat- ment group. RNA-Seq libraries were prepared using Illumina ScriptSeq Complete Gold (Yeast) Kit (Illumina, Inc., San Diego, CA) according to manufacturer’s instructions. One µg of total RNA from each sample was used for library preparation. All libraries were sequenced 2 ×150 bp high output v2 kit (100 Gb) on the Illumina NextSeq 500 platform. p Reads were mapped to the Candida genome using the RNA-seq processing pipeline STAR v2.5.2a80 in AlignReads mode with a maximum intron size of 30Kb. Gene expression was quantified by counting reads using htseq v0.6.180 for all genes in the Candida Genome Database (http://www.candidagenome.org, Stanford Genome Technology Centre) gene model C. albicans SC5314 version A22-s07-m01-r30. RNA-Seq data was analyzed in R v3.6.1 (https://www.R-project.org/)81 by broadly exploring differences between samples using principle compo- nent analysis after normalizing read counts into centered log ratio values. Statistical comparison of gene expres- sion between treatment types was performed using PERMANOVA. Differentially expressed genes were then identified using DESeq. 2 v1.14.182, with the sample group set as the design formula and contrasts between groups used to identify differentially abundant genes, and visualized using EnhancedVolcano v1.2.0 (https://github.com/ kevinblighe/EnhancedVolcano)83. The Candida genome database was used to determine pathways affected by the different treatments, using an adjusted p-value < 1e−5 as the cut-off for statistically significant gene expression comparison. Real‐time PCR assay. Changes in the expression of CDR1, CDR2 and MDR1 (the genes coding C. albi- cans drug efflux pumps) when exposed to fluconazole, C12AHL or C12AHL + fluconazole for a shorter duration (1 h) were quantitatively assessed by real‐time polymerase chain reaction (qPCR). C. albicans SC5314 suspen- sions (1 × 103 cells mL−1) were prepared as mentioned above and treated with fluconazole, C12AHL or both (C12AHL + fluconazole) for 1 h at 37 °C in static conditions. Cells were harvested, washed 3 times with PBS, and RNA was extracted using the SV total RNA isolation system (Catalog No. Material and methods The MIC50 and MIC80 were defined as the lowest concentration of the tested agent that inhibited 50% and 80%, respectively, of fungal growth compared to solvent controls. The assay was performed as quadruplicates three separate times (n = 12). Determination of minimum inhibitory concentration (MIC). The MIC was determined by a broth microdilution assay in accordance with the CLSI guidelines78. Briefly, C. albicans suspensions (1 × 103 cells mL−1) were treated with fluconazole, C12AHL or both using a checker-board approach (C12AHL: 12.5 µg mL−1– 100 µg mL−1 and Fluconazole: 0.078 µg mL−1–80 µg mL−1) and incubated in a 96-well microtiter plate for 24 h at 37 °C. At the end of this incubation, the optical density of the fungal growth was measured spectrometrically at 595 nm and MICs were determined. The MIC50 and MIC80 were defined as the lowest concentration of the tested agent that inhibited 50% and 80%, respectively, of fungal growth compared to solvent controls. The assay was performed as quadruplicates three separate times (n = 12). Treatment groups and doses. Three test groups (fluconazole, C12AHL, fluconazole+C12AHL) and one solvent control group (DMSO; the solvent for C12AHL and fluconazole) were used. Following concentrations were used throughout the study unless otherwise specified; 1.25 µg mL−1 fluconazole, 50 µg mL−1 C12AHL, 1.25 µg mL−1 fluconazole + 50 µg mL−1 C12AHL, or DMSO (Control, 2% V/V). The chosen concentration of flu- conazole is the minimum concentration required to inhibit 80% of C. albicans cells (MIC80). A sub-growth and sub-hyphal inhibitory concentration of C12AHL (50 µg mL−1, 168 µM)28 was chosen to prevent growth or hyphae development associated effects on C. albicans. Drug efflux activity assay. The activity of C. albicans drug efflux pumps when treated with C12AHL and/or fluconazole was assessed using an indicator dye, rhodamine 6 g (R6G), as described by Holmes, A. R. et al. 201879. Briefly, standard suspensions of C. albicans SC5314 and mutant strains DSY448, DSY465, DSY654, DSY1050, DSY1751, DSY1764 and DSY1769 were prepared in PBS and starved for 2 h at room temperature (25 °C). R6G was added (10 µM final concentration) and incubated in dark conditions for further 1 h at 37 °C and 200 rpm. At the end of the incubation, cells were washed three times with PBS, resuspended and 100 µl was added to wells in a 96 well plate. Rhodamine 6 g loaded C. Material and methods Microorganisms and quorum sensing molecules. C. albicans SC5314 (a fluconazole sensitive strain) was used throughout this study. Microbial identity was reconfirmed with commercially available API 32 C for Candida strains (Biomérieux, Mercy I’Etoile, France). Mutant C. albicans strains DSY448, DSY465, DSY654, DSY1050, DSY1751, DSY1764, DSY1769 and parental strain C. albicans CAF2–1 (Supplementary Table S15) were kindly gifted by Associate Professor Dominique Sanglard from the Institute of Microbiology, University Hospital Lausanne, Switzerland. All isolates were stored in multiple aliquots at –70 °C, after confirming their purity. C12AHL from P. aeruginosa (Catalogue No. O9139) and fluconazole (Catalogue No. F8929) were purchased from Sigma Aldrich (St. Louis, MO), dissolved in dimethyl sulfoxide (DMSO) and stored at −20°C until further icroorganisms and quorum sensing molecules ti C12AHL from P. aeruginosa (Catalogue No. O9139) and fluconazole (Catalogue No. F8929) were purchased from Sigma Aldrich (St. Louis, MO), dissolved in dimethyl sulfoxide (DMSO) and stored at −20 °C until further use. Growth media. Sabouraud dextrose agar and yeast nitrogen base with amino acids (YNB; Catalogue No. Y1250; Sigma Aldrich, St. Louis, MO) solution supplemented with 100 mm glucose were used for culturing C. albicans. RPMI 1640 media supplemented with MOPS (morpholinepropanesulfonic acid) was used for broth microdilution assays. Yeast inocula. Before each experiment, both C. albicans wild type and mutant strains were subcultured on Sabouraud Dextrose Agar for 18 h at 37 °C. A single colony from overnight C. albicans growth was inoculated into YNB medium and incubated for 18 h in an orbital shaker (150 rpm) at 37 °C. The resultant culture was har- vested, washed twice in phosphate-buffered saline (PBS, pH 7.2) and resuspended in YNB. Cell suspensions were adjusted to 1 × 107 cells mL−1 (standard unless otherwise specified) by spectrophotometry and confirmed by hemocytometric counting. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Determination of minimum inhibitory concentration (MIC). The MIC was determined by a broth microdilution assay in accordance with the CLSI guidelines78. Briefly, C. albicans suspensions (1 × 103 cells mL−1) were treated with fluconazole, C12AHL or both using a checker-board approach (C12AHL: 12.5 µg mL−1– 100 µg mL−1 and Fluconazole: 0.078 µg mL−1–80 µg mL−1) and incubated in a 96-well microtiter plate for 24 h at 37 °C. At the end of this incubation, the optical density of the fungal growth was measured spectrometrically at 595 nm and MICs were determined. Material and methods albicans were treated with either fluconazole, C12AHL, C12AHL + flu- conazole, or DMSO and the plate was incubated at 37 °C in the dark. After 5 min of post-treatment, 1 mM glucose was added to each well and further incubated. The cell suspensions were removed at given time points (every 10 min up to 1 h, hourly up to 5 h, 18 h, and 24 h), centrifuged (10 min, 13000 rpm, 25 °C), and the amount of R6G released into the supernatant was read using a spectrophotometer at 485 nm excitation/535 nm emission. Each assay was conducted in sextuplicate at 3 different occasions (n = 18). References 1. Nobile, C. J. & Johnson, A. D. Candida albicans Biofilms and Human Disease. Annu. Rev. Microbiol. 69, 71–92, https://doi. org/10.1146/annurev-micro-091014-104330 (2015). g 2. Dhamgaye, S., Qu, Y. & Peleg, A. Y. Polymicrobial infections involving clinically relevant Gram-negative bacteria and fungi. Cel Microbiol. 18, 1716–1722, https://doi.org/10.1111/cmi.12674 (2016). p g 3. Kalan, L. & Grice, E. A. Fungi in the Wound Microbiome. Adv. Wound Care 7, 247–255, https://doi.org/10.1089/wound.2017.0756 (2018). 4. Chellan, G. et al. Spectrum and prevalence of fungi infecting deep tissues of lower-limb wounds in patients with type 2 diabetes. J. Clin. Microbiol. 48, 2097–2102, https://doi.org/10.1128/JCM.02035-09 (2010). p g 5. Dowd, S. E. et al. Survey of fungi and yeast in polymicrobial infections in chronic wounds. J. Wound Care 20, 40–47, https://doi org/10.12968/jowc.2011.20.1.40 (2011). 6. Azoulay, E. et al. Candida colonization of the respiratory tract and subsequent Pseudomonas ventilator-associated pneumonia. Ches 129, 110–117, https://doi.org/10.1378/chest.129.1.110 (2006). 7. Hamet, M. et al. Candida spp. airway colonization could promote antibiotic-resistant bacteria selection in patients with suspected ventilator-associated pneumonia. Intensive Care Med. 38, 1272–1279, https://doi.org/10.1007/s00134-012-2584-2 (2012). p p g 8. Navarro, J. et al. Factors associated with poor pulmonary function: cross-sectional analysis of data from the ERCF. Euro Epidemiologic Registry of Cystic Fibrosis Eur Respir J 18, 298–305, https://doi org/10 1183/09031936 01 00068901 (2001) avarro, J. et al. Factors associated with poor pulmonary function: cross-sectional analysis of data from the ERCF. European d l f b h d ( ) 8. Navarro, J. et al. Factors associated with poor pulmonary function: cross sectional analysis of data from the ERCF. Eur Epidemiologic Registry of Cystic Fibrosis. Eur. Respir. J. 18, 298–305, https://doi.org/10.1183/09031936.01.00068901 (2001). 9. Klotz, S. A., Chasin, B. S., Powell, B., Gaur, N. K. & Lipke, P. N. Polymicrobial bloodstream infections involving Candida species: analysis of patients and review of the literature. Diagn. Microbiol. Infect. Dis. 59, 401–406, https://doi.org/10.1016/j.diagmicrobio.2007.07.001 (2007). 0. Esim Buyukbayrak, E. et al. Diagnosis of vulvovaginitis: comparison of clinical and microbiological diagnosis. Arch. Gynecol. Obstet 282, 515–519, https://doi.org/10.1007/s00404-010-1498-x (2010).i g 1. Buchta, V. & Spacek, J. Microbiological findings in patients with recurrent vulvovaginal candidiasis in the Hradec Kralove Faculty Hospital 1995-2002. Ceska Gynekol. 69, 7–14 (2004). p y 2. Cumming, C. G., Wight, C., Blackwell, C. L. & Wray, D. Denture stomatitis in the elderly. Oral. Microbiol. Immunol. 5, 82–85, https:/ doi.org/10.1111/j.1399-302X.1990.tb00232.x (1990). 13. Pammi, M., Zhong, D., Johnson, Y., Revell, P. & Versalovic, J. www.nature.com/scientificreports/ www.nature.com/scientificreports/ electrophoreses were performed as described previously84. Protein gels derived from three biological replicates (with three technical replicates) from each group were analysed using a two‐dimensional analysis software (PD Quest; Bio‐Rad Laboratories). Protein spots were identified using default settings and verified manually to elim- inate background noise and artefacts. Only consistent and reproducible protein spots (in all replicates) were pro- gressed for further analysis, and spots that were missing in either fluconazole or fluconazole + C12AHL samples (in all replicates) were considered as differentially expressed in C. albicans in response to respective exposure84.f ( p )f y p p p p Proteins that were differentially expressed were in-gel digested, peptides were extracted and subjected to tandem mass spectrometry as described previously84. Briefly, matrix‐assisted laser desorption/ionization time‐ of‐flight mass spectroscopy/mass spectroscopy (MALDI TOF MS/MS) was performed using a Bruker Autoflex III MALDI TOF/TOF Mass Spectrometer (Bruker Daltonics, Bremen, Germany) and Dionex UltiMate 3000 nano‐LC system using a 50‐Hz frequency laser beam. Candidate proteins were identified in the NCBI nr database using Mascot software (http://www.matrixscience.com/) (parameters used: Type of search: MS/MS Ion Search, Enzyme: Trypsin/P, Fixed modifications: Carbamidomethyl (C), Variable modifications: Oxidation (M), Mass values: Monoisotopic, Protein Mass: Unrestricted, Peptide Mass Tolerance: ±50 ppm, Fragment Mass Tolerance: ±0.5 Da, Max Missed Cleavages: 1, Instrument type: MALDI‐TOF‐TOF). Protein scores were derived from ion scores as a non‐probabilistic basis for ranking the protein hits at a significance level of p-value < 0.0584. Identified proteins were functionally characterized, and encoding genes were determined using the Candida genome data- base, NCBI database (http://www.ncbi.nlm.nih.gov/), SWISSPROT and TrEMBL non‐redundant protein data- bases (http://www.expasy.ch/sport/)84. Statistical analyses. All other assays not mentioned in the sections above were performed using nonpar- ametric Mann—Whitney U-tests with SPSS software (version 16.0) for comparison of test conditions to corre- sponding control groups. A p-value < 0.05 was considered statistically significant. Data availability y Sequencing data that support the findings of this study have been deposited in NCBI Sequence Read Archive (SRNA) under Bio Project Accession No. PRJNA599446. (https://www.ncbi.nlm.nih.gov/bioproject/599446). Received: 2 March 2020; Accepted: 22 April 2020; Published: xx xx xxxx Received: 2 March 2020; Accepted: 22 April 2020; Published: xx xx xxxx Received: 2 March 2020; Accepted: 22 April 2020; Published: xx xx xxxx Gene expression analyses Z3100, Promega, Madison, WI) using 2 μg template for reverse transcription with Superscript II (Invitrogen, Carlsbad, CA). qPCR was performed as described previously84 using primers shown in Supplementary Table S16. Relative gene expression was quantified using EFB1 as the housekeeping (reference) gene85. All experiments were carried out in duplicate on three differ- ent occasions (n = 6). Protein expression analysis. The changes in C. albicans protein expression when treated with C12AHL + fluconazole compared to fluconazole were assessed with 2-dimensional gel electrophoresis and mass spectrometry. C. albicans SC5314 standard suspension was prepared as described above, treated with either fluconazole or C12AHL + fluconazole and incubated at 37 °C statically for 24 h. At the end of the incubation, cells were washed 3 times with PBS and total protein were extracted, after which first‐ and second-dimension Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 13 References 53, 2392–2401, https://doi org/10.1128/AAC.01551-08 (2009). 26. Sharma, M. & Prasad, R. The quorum-sensing molecule farnesol is a modulator of drug efflux mediated by ABC multidrug transporters and synergizes with drugs in Candida albicans. Antimicrob. Agents Chemother. 55, 4834–4843, https://doi.org/10.1128/ AAC.00344-11 (2011).fl ( ) 27. Cernakova, L., Dizova, S., Gaskova, D., Jancikova, I. & Bujdakova, H. Impact of Farnesol as a Modulator of Efflux Pumps Fluconazole-Resistant Strain of Candida albicans. Microb. Drug. Resist. https://doi.org/10.1089/mdr.2017.0332 (2019).l uco a o e Res sta t St a o Candida albicans. icrob. rug. Resist. ttps://do .o g/ 0. 089/ d . 0 7.033 ( 0 9). 28. Hogan, D. A., Vik, A. & Kolter, R. A Pseudomonas aeruginosa quorum-sensing molecule influences Candida albicans morphology. Mol. Microbiol. 54, 1212–1223, https://doi.org/10.1111/j.1365-2958.2004.04349.x (2004). g p g 28. Hogan, D. A., Vik, A. & Kolter, R. A Pseudomonas aeruginosa quorum-sensing molecule influences Ca Mol. Microbiol. 54, 1212–1223, https://doi.org/10.1111/j.1365-2958.2004.04349.x (2004).h 8. Hogan, D. A., Vik, A. & Kolter, R. A Pseudomonas aeruginosa quorum-sensing molecule influences Candida albicans morphology Mol. Microbiol. 54, 1212–1223, https://doi.org/10.1111/j.1365-2958.2004.04349.x (2004).h p g j ( ) 29. Hall, R. A. et al. The quorum-sensing molecules farnesol/homoserine lactone and dodecanol operate via distinct modes of action in Candida albicans. Eukaryot. Cell 10, 1034–1042, https://doi.org/10.1128/EC.05060-11 (2011). h ans. Eukaryot. Cell 10, 1034–1042, https://doi.org/10.1128/EC.0506 30. Bandara, H. M. H. N., Hewavitharana, A. K., Shaw, P. N., Smyth, H. D. C. & Samaranayake, L. P. A novel, quorum sensor-infused liposomal drug delivery system suppresses Candida albicans biofilms. Int. J. Pharm. 578, 119096, https://doi.org/10.1016/j. ijpharm.2020.119096 (2020). jp 1. Berkow, E. L. & Lockhart, S. R. Fluconazole resistance in Candida species: a current perspective. Infect. Drug. Resist. 10, 237–245 https://doi.org/10.2147/IDR.S118892 (2017).fl p g 32. Cannon, R. D. et al. Efflux-mediated antifungal drug resistance. Clin Microbiol Rev 22, 291-321, Table of Contents, https://doi. org/10.1128/CMR.00051-08 (2009). g 33. Mukhopadhyay, K. et al. Membrane sphingolipid-ergosterol interactions are important determinants of multidrug resistance in Candida albicans. Antimicrob. Agents Chemother. 48, 1778–1787, https://doi.org/10.1128/aac.48.5.1778-1787.2004 (2004).f 34. Pasrija, R., Panwar, S. L. & Prasad, R. Multidrug transporters CaCdr1p and CaMdr1p of Candida albicans display different lipid specificities: both ergosterol and sphingolipids are essential for targeting of CaCdr1p to membrane rafts. Antimicrob. Agents Chemother. 52, 694–704, https://doi.org/10.1128/AAC.00861-07 (2008). p g ( ) 5. Peleg, A. Y., Hogan, D. A. & Mylonakis, E. Medically important bacterial-fungal interactions. Nat. Rev. Microbiol. 8, 340–349, https:/ doi.org/10.1038/nrmicro2313 (2010).f g 6. Peters, B. M., Jabra-Rizk, M. References Polymicrobial bloodstream infections in the neonatal intensive care unit are associated with increased mortality: a case-control study. BMC Infect. Dis. 14, 390, https://doi.org/10.1186/1471-2334-14-390 (2014). 4. Bandara, H. M., Lam, O. L., Jin, L. J. & Samaranayake, L. Microbial chemical signaling: a current perspective. Crit. Rev. Microbiol 38, 217–249, https://doi.org/10.3109/1040841X.2011.652065 (2012). 15. Diggle, S. P., Crusz, S. A. & Camara, M. Quorum sensing. Curr. Biol. 17, R907–910, https://doi.org/10.1016/j.cub.2007.08.045 (2007). 16. Fuqua, W. C., Winans, S. C. & Greenberg, E. P. Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. J. Bacteriol. 176, 269–275, https://doi.org/10.1128/jb.176.2.269-275.1994 (1994). 6. Fuqua, W. C., Winans, S. C. & Greenberg, E. P. Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. J. Bacteriol. 176, 269–275, https://doi.org/10.1128/jb.176.2.269-275.1994 (1994). , , , g, Q g R y y p riptional regulators. J. Bacteriol. 176, 269–275, https://doi.org/10.1128/jb.176.2.269-275.1994 (1994). 7. Whitehead, N. A., Barnard, A. M., Slater, H., Simpson, N. J. & Salmond, G. P. Quorum-sensing in Gram-negative bacteria. FEMS Microbiol. Rev. 25, 365–404, https://doi.org/10.1111/j.1574-6976.2001.tb00583.x (2001). 8. Fourie, R. et al. Candida albicans and Pseudomonas aeruginosa Interaction, with Focus on the Role of Eicosanoids. Front. Physiol. 7 64, https://doi.org/10.3389/fphys.2016.00064 (2016).i p g p y 9. El-Azizi, M. A., Starks, S. E. & Khardori, N. Interactions of Candida albicans with other Candida spp. and bacteria in the biofilms. J Appl. Microbiol. 96, 1067–1073, https://doi.org/10.1111/j.1365-2672.2004.02213.x (2004). pp p g j 0. McAlester, G., O’Gara, F. & Morrissey, J. P. Signal-mediated interactions between Pseudomonas aeruginosa and Candida albicans. J Med. Microbiol. 57, 563–569, https://doi.org/10.1099/jmm.0.47705-0 (2008). p g j 21. Nickerson, K. W., Atkin, A. L. & Hornby, J. M. Quorum sensing in dimorphic fungi: farnesol and beyond. Appl. Env. Microbiol. 72, 3805–3813, https://doi.org/10.1128/AEM.02765-05 (2006). Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ 22. Cugini, C. et al. Farnesol, a common sesquiterpene, inhibits PQS production in Pseudomonas aeruginosa. Mol. Microbio 896–906, https://doi.org/10.1111/j.1365-2958.2007.05840.x (2007).f p g j 3. Davis-Hanna, A., Piispanen, A. E., Stateva, L. I. & Hogan, D. A. Farnesol and dodecanol effects on the Candida albicans Ras1-cAMP signalling pathway and the regulation of morphogenesis. Mol. Microbiol. 67, 47–62, https://doi.org/10.1111/j.1365-2958.2007.06013.x (2008). ( ) 24. Hornby, J. M. & Nickerson, K. W. Enhanced production of farnesol by Candida albicans treated with four azoles. Antimicrob. Agents Chemother. 48, 2305–2307, https://doi.org/10.1128/AAC.48.6.2305-2307.2004 (2004).f p g 5. Shirtliff, M. E. et al. Farnesol-induced apoptosis in Candida albicans. Antimicrob. Agents Chemother. References A., O’May, G. A., Costerton, J. W. & Shirtliff, M. E. Polymicrobial interactions: impact on pathogenesi and human disease. Clin. Microbiol. Rev. 25, 193–213, https://doi.org/10.1128/CMR.00013-11 (2012). 37. Ovchinnikova, E. S., Krom, B. P., van der Mei, H. C. & Busscher, H. J. Force microscopic and thermodynamic analysis of the adhesion between Pseudomonas aeruginosa and Candida albicans. Soft Matter 8, 6454–6461, https://doi.org/10.1039/C2SM25100K (2012). 38. Hogan, D. A. & Kolter, R. Pseudomonas-Candida interactions: an ecological role for virulence factors. Science 296, 2229–2232, https://doi.org/10.1126/science.1070784 (2002). p g 39. Charlton, T. S. et al. A novel and sensitive method for the quantification of N-3-oxoacyl homoserine lactones using gas chromatography-mass spectrometry: application to a model bacterial biofilm. Env. Microbiol. 2, 530–541, https://doi.org/10.1046/ j.1462-2920.2000.00136.x (2000). j 0. Trejo-Hernandez, A., Andrade-Dominguez, A., Hernandez, M. & Encarnacion, S. Interspecies competition triggers virulence and mutability in Candida albicans-Pseudomonas aeruginosa mixed biofilms. ISME J. 8, 1974–1988, https://doi.org/10.1038 ismej.2014.53 (2014).fl ismej.2014.53 (2014). 41. Gbelska, Y., Toth Hervay, N., Dzugasova, V. & Konecna, A. Measurement of Energy-dependent Rhodamine 6G Efflux in Yeast Species. Bio-protocol 7, e2428, https://doi.org/10.21769/BioProtoc.2428 (2017). j 1. Gbelska, Y., Toth Hervay, N., Dzugasova, V. & Konecna, A. Measurement of Energy-dependent Rhodamine 6G Efflux in Yeas Species. Bio-protocol 7, e2428, https://doi.org/10.21769/BioProtoc.2428 (2017). p p p g ( ) 42. Maesaki, S., Marichal, P., Vanden Bossche, H., Sanglard, D. & Kohno, S. Rhodamine 6G efflux for the detection of CDR1- overexpressing azole-resistant Candida albicans strains. J. Antimicrob. Chemother. 44, 27–31, https://doi.org/10.1093/jac/44.1.27 (1999). 43. Boon, C. et al. A novel DSF-like signal from Burkholderia cenocepacia interferes with Candida albicans morphological transi ISME J. 2, 27–36, https://doi.org/10.1038/ismej.2007.76 (2008).li 44. Mukherjee, P. K., Chandra, J., Kuhn, D. M. & Ghannoum, M. A. Mechanism of fluconazole resistance in Candida albicans biofilms: phase-specific role of efflux pumps and membrane sterols. Infect. Immun. 71, 4333–4340, https://doi.org/10.1128/IAI.71.8.4333- 4340.2003 (2003).i 45. Holcombe, L. J. et al. Pseudomonas aeruginosa secreted factors impair biofilm development in Candida albicans. Microbiology 1476–1486, https://doi.org/10.1099/mic.0.037549-0 (2010). 46. Bhattacharya, S., Esquivel, B. D. & White, T. C. Overexpression or Deletion of Ergosterol Biosynthesis Genes Alters Doubling Time, Response to Stress Agents, and Drug Susceptibility in Saccharomyces cerevisiae. MBio 9, https://doi.org/10.1128/mBio.01291-18 (2018).l 47. Albertsen, L. et al. Diversion of flux toward sesquiterpene production in Saccharomyces cerevisiae by fusion of host and heterologous enzymes. Appl. Env. Microbiol. 77, 1033–1040, https://doi.org/10.1128/AEM.01361-10 (2011). ( ) ( ) y pp p g 8. Vale-Silva, L. A. et al. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 References Microbiol. 29, 605–615, https://doi.org/10.1046/j.1365-2958.1998.00957.x (1998). References et al. Genome-wide expression profiling of the response to azole, polyene, echinocandin, and pyrimidine antifungal agent i C did lbi A ti i b A t Ch th 49 2226 2236 htt //d i /10 1128/AAC 49 6 2226 2236 2005 (2005) 59. Warnecke, D. et al. Cloning and functional expression of UGT genes encoding sterol glucosyltransferases from Saccharomyces cerevisiae, Candida albicans, Pichia pastoris, and Dictyostelium discoideum. J. Biol. Chem. 274, 13048–13059, https://doi.org/10.1074/ jbc.274.19.13048 (1999). j ( ) 60. Enjalbert, B. et al. Role of the Hog1 stress-activated protein kinase in the global transcriptional response to stress in the fu pathogen Candida albicans. Mol. Biol. Cell 17, 1018–1032, https://doi.org/10.1091/mbc.e05-06-0501 (2006).i g g 61. Ralser, M. et al. Metabolic reconfiguration precedes transcriptional regulation in the antioxidant response. Nat. Biotechno 604–605, https://doi.org/10.1038/nbt0709-604 (2009).i p g ( ) 62. Petti, A. A., Crutchfield, C. A., Rabinowitz, J. D. & Botstein, D. Survival of starving yeast is correlated with oxidative stress response and nonrespiratory mitochondrial function. Proc. Natl Acad. Sci. USA 108, E1089–1098, https://doi.org/10.1073/pnas.1101494108 (2011). 63. Wang, Y. et al. Cap1p is involved in multiple pathways of oxidative stress response in Candida albicans. Free. Radic. Biol. Med 1201–1209, https://doi.org/10.1016/j.freeradbiomed.2005.11.019 (2006). p g j 4. Chechik, G. et al. Activity motifs reveal principles of timing in transcriptional control of the yeast metabolic network. Nat Biotechnol. 26, 1251–1259, https://doi.org/10.1038/nbt.1499 (2008). p g 5. Lee, J. et al. Yap1 and Skn7 control two specialized oxidative stress response regulons in yeast. J. Biol. Chem. 274, 16040–16046 https://doi.org/10.1074/jbc.274.23.16040 (1999). p g j 6. Minard, K. I. & McAlister-Henn, L. Sources of NADPH in yeast vary with carbon source. J. Biol. Chem. 280, 39890–39896, https:/ doi.org/10.1074/jbc.M509461200 (2005). g j ( ) 7. Minard, K. I. & McAlister-Henn, L. Antioxidant function of cytosolic sources of NADPH in yeast. Free. Radic. Biol. Med. 31 832–843, https://doi.org/10.1016/S0891-5849(01)00666-9 (2001).l p g ( ) ( ) 8. Christensen, B., Gombert, A. K. & Nielsen, J. Analysis of flux estimates based on (13)C-labelling experiments. Eur. J. Biochem. 269 2795–2800, https://doi.org/10.1046/j.1432-1033.2002.02959.x (2002). 69. Hwang, C. S. et al. Copper- and zinc-containing superoxide dismutase (Cu/ZnSOD) is required for the protection of Candida albicans against oxidative stresses and the expression of its full virulence. Microbiology 148, 3705–3713, https://doi. org/10.1099/00221287-148-11-3705 (2002). g ( ) 70. Li, L., Naseem, S., Sharma, S. & Konopka, J. B. Flavodoxin-Like Proteins Protect Candida albicans from Oxidative Stress and Promote Virulence. PLoS Pathog. Acknowledgements g We thank Dr. Nicola Angel (Australian Centre for Ecogenomics, the University of Queensland, Australia) for her assistance in RNA-Seq and Associate Professor Dominique Sanglard (the University Hospital Lausanne, Switzerland) for providing C. albicans mutant strains. Author contributions H.M.H.N.B. and L.P.S. conceptualized the study; H.M.H.N.B., B.P.K.C. and D.L.A.W. designed the experiments; H.M.H.N.B. and B.P.K.C. performed experiments; H.M.H.N.B., I.V., B.P.K.C., D.L.A.W., P.H. and L.P.S. analyzed and interpreted data; H.M.H.N.B., D.L.A.W. and I.V. wrote the manuscript; I.V., P.H. and L.P.S. revised the manuscript. References Azole resistance by loss of function of the sterol Δ(5),(6)-desaturase gene (ERG3) in Candida albicans does no necessarily decrease virulence. Antimicrob. Agents Chemother. 56, 1960–1968, https://doi.org/10.1128/AAC.05720-11 (2012). y pp g 48. Vale-Silva, L. A. et al. Azole resistance by loss of function of the sterol Δ(5),(6)-desaturase gene (ERG3) in Candida alb necessarily decrease virulence Antimicrob Agents Chemother 56 1960 1968 https://doi org/10 1128/AAC 05720 11 y g , , p g ( ) 9. Lupetti, A., Danesi, R., Campa, M., Del Tacca, M. & Kelly, S. Molecular basis of resistance to azole antifungals. Trends Mol. Med. 8 76–81, https://doi.org/10.1016/S1471-4914(02)02280-3 (2002). p g 50. Henry, K. W., Nickels, J. T. & Edlind, T. D. Upregulation of ERG genes in Candida species by azoles and other sterol biosynthesis inhibitors. Antimicrob. Agents Chemother. 44, 2693–2700, https://doi.org/10.1128/AAC.44.10.2693-2700.2000 (2000).t 51. Mollinedo, F. Lipid raft involvement in yeast cell growth and death. Front. Oncol. 2, 140, https://doi.org/10.3389/fonc.2012.00140 (2012).f 52. Kodedova, M. & Sychrova, H. Changes in the Sterol Composition of the Plasma Membrane Affect Membrane Potential, Salt Tolerance and the Activity of Multidrug Resistance Pumps in Saccharomyces cerevisiae. PLoS One 10, e0139306, https://doi. org/10.1371/journal.pone.0139306 (2015).h g j p 53. Song, J. L., Harry, J. B., Eastman, R. T., Oliver, B. G. & White, T. C. The Candida albicans lanosterol 14-alpha-demethylase (ERG11) gene promoter is maximally induced after prolonged growth with antifungal drugs. Antimicrob. Agents Chemother. 48, 1136–1144, https://doi.org/10.1128/AAC.48.4.1136-1144.2004 (2004).i 54. Rogers, P. D. & Barker, K. S. Genome-wide expression profile analysis reveals coordinately regulated genes associated with stepwise acquisition of azole resistance in Candida albicans clinical isolates. Antimicrob. Agents Chemother. 47, 1220–1227, https://doi. org/10.1128/AAC.47.4.1220-1227.2003 (2003). g ( ) 5. Yang, H. et al. Structural mechanism of ergosterol regulation by fungal sterol transcription factor Upc2. Nat. Commun. 6, 6129 https://doi.org/10.1038/ncomms7129 (2015). Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 15 www.nature.com/scientificreports/ 56. Silver, P. M., Oliver, B. G. & White, T. C. Role of Candida albicans transcription factor Upc2p in drug resistance and s metabolism. Eukaryot. Cell 3, 1391–1397, https://doi.org/10.1128/EC.3.6.1391-1397.2004 (2004).t 7. Martin, S. W. & Konopka, J. B. Lipid raft polarization contributes to hyphal growth in Candida albicans. Eukaryot. Cell 3, 675–684 https://doi.org/10.1128/EC.3.3.675-684.2004 (2004).i p g ( ) 58. Liu, T. T. et al. Genome-wide expression profiling of the response to azole, polyene, echinocandin, and pyrimidine antifungal agents in Candida albicans. Antimicrob. Agents Chemother. 49, 2226–2236, https://doi.org/10.1128/AAC.49.6.2226-2236.2005 (2005). p g ( ) 8. Liu, T. T. References (Clinical Laboratory Standards Institute, 2017). 79. Holmes, A. R. et al. ABC transporter Cdr1p contributes more than Cdr2p does to fluconazole efflux in fluconazole-resistant Candida albicans clinical isolates Antimicrob Agents Chemother 52 3851 3862 https://doi org/10 1128/AAC 00463 08 (2008) 78. CLSI. Reference method for broth dilution antifungal susceptibility testing of yeasts. (Clinical Laboratory Standards Institute, 2017). 79. Holmes, A. R. et al. ABC transporter Cdr1p contributes more than Cdr2p does to fluconazole efflux in fluconazole-resistant Candida albicans clinical isolates. Antimicrob. Agents Chemother. 52, 3851–3862, https://doi.org/10.1128/AAC.00463-08 (2008). albicans clinical isolates. Antimicrob. Agents Chemother. 52, 3851 3862, https://doi.org/10.1128/AAC.00463 08 (2008). 80. Anders, S., Pyl, P. T. & Huber, W. HTSeq–a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169, https://doi.org/10.1093/bioinformatics/btu638 (2015). 81. R-Core-Team. R: A language and environment for statistical co 8 . R Co e ea . R: a guage a d e o e t o stat st ca co put g, ttps:// .R p oject.o g/ ( 0 9). 82. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq. 2. Genome Biol. 15, 550, https://doi.org/10.1186/s13059-014-0550-8 (2014). g g g j g 82. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq. 2. Genome Biol. 15, 550, https://doi.org/10.1186/s13059-014-0550-8 (2014). g 83. Blighe, K., Rana, S. & Lewis, M. EnhancedVolcano: Publication-ready volcano plots with enhanced colouring and labeling, ht github.com/kevinblighe/EnhancedVolcano (2019). 84. Bandara, H. M. H. N., Cheung, B. P. K., Watt, R. M., Jin, L. J. & Samaranayake, L. P. Pseudomonas aeruginosa lipopolysaccharide inhibits Candida albicans hyphae formation and alters gene expression during biofilm development. Mol. Oral. Microbiol. 28, 54–69, https://doi.org/10.1111/omi.12006 (2013).f 84. Bandara, H. M. H. N., Cheung, B. P. K., Watt, R. M., Jin, L. J. & Samaranayake, L. P. Pseudomonas aeruginosa lipopolysaccharide inhibits Candida albicans hyphae formation and alters gene expression during biofilm development. Mol. Oral. Microbiol. 28, 54–69, https://doi.org/10.1111/omi.12006 (2013).f 85. Schaller, M., Schafer, W., Korting, H. C. & Hube, B. Differential expression of secreted aspartyl proteinases in a model of human oral candidosis and in patient samples from the oral cavity. Mol. Microbiol. 29, 605–615, https://doi.org/10.1046/j.1365-2958.1998.00957.x (1998). 85. Schaller, M., Schafer, W., Korting, H. C. & Hube, B. Differential expression of secreted aspartyl proteinases in a model of human oral candidosis and in patient samples from the oral cavity. Mol. References 11, e1005147, https://doi.org/10.1371/journal.ppat.1005147 (2015). 1. DeRisi, J. L., Iyer, V. R. & Brown, P. O. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278, 680–686, https://doi.org/10.1126/science.278.5338.680 (1997). 278, 680–686, https://doi.org/10.1126/science.278.5338.680 (1997). 2. Cowen, L. E. et al. Population genomics of drug resistance in Candida albicans. Proc. Natl Acad. Sci. USA 99, 9284–9289, https://doi g 2. Cowen, L. E. et al. Population genomics of drug resistance in Candida albicans. Proc. Natl Acad. Sci. USA 99, 9284–9289, https://doi org/10.1073/pnas.102291099 (2002). g p 73. Sen, M. et al. UDP-glucose 4, 6-dehydratase activity plays an important role in maintaining cell wall integrity and virulence of Candida albicans. PLoS Pathog. 7, e1002384, https://doi.org/10.1371/journal.ppat.1002384 (2011).hi g p g j pp 74. Skrzypek, M. S. et al. The Candida Genome Database (CGD): incorporation of Assembly 22, systematic identifiers and of high throughput sequencing data Nucleic Acids Res 45 D592–D596 https://doi org/10 1093/nar/gkw924 (2017) 4. Skrzypek, M. S. et al. The Candida Genome Database (CGD): incorporation of Assembly 22, systematic identifiers and visualization of high throughput sequencing data. Nucleic Acids Res. 45, D592–D596, https://doi.org/10.1093/nar/gkw924 (2017).fl g g p q g p g g 5. Sorgo, A. G. et al. Effects of fluconazole on the secretome, the wall proteome, and wall integrity of the clinical fungus Candida albicans. Eukaryot. Cell 10, 1071–1081, https://doi.org/10.1128/EC.05011-11 (2011). 75. Sorgo, A. G. et al. Effects of fluconazole on the secretome, the wall proteome, and wall integ albicans. Eukaryot. Cell 10, 1071–1081, https://doi.org/10.1128/EC.05011-11 (2011). 6. Arnaud, M. B. et al. Candida Genome Database http://www.candidagenome.org/, http://pathway.candidagenome.org/CALBI/NEW IMAGE?object=ARG-PRO-PWY. 76. Arnaud, M. B. et al. Candida Genome Database http://www.candidagenome.org/, http://pathway.candidagenome.org/CALBI/NEW- IMAGE?object=ARG-PRO-PWY. 77. Gibson, J., Sood, A. & Hogan, D. A. Pseudomonas aeruginosa-Candida albicans interactions: localization and fungal toxicity of a p g g p p y g g IMAGE?object=ARG-PRO-PWY. 77. Gibson, J., Sood, A. & Hogan, D. A. Pseudomonas aeruginosa-Candida albicans interactions: localization and fungal toxicity of a h i d i ti Appl E Mi bi l 75 504 513 htt //d i /10 1128/AEM 01037 08 (2009) j 7. Gibson, J., Sood, A. & Hogan, D. A. Pseudomonas aeruginosa-Candida albicans interactions: localization and fungal toxicity of a phenazine derivative. Appl. Env. Microbiol. 75, 504–513, https://doi.org/10.1128/AEM.01037-08 (2009). p pp p g 78. CLSI. Reference method for broth dilution antifungal susceptibility testing of yeasts. (Clinical Laboratory Standards Institute, 2 78. CLSI. Reference method for broth dilution antifungal susceptibility testing of yeasts. Competing interestsh p g The authors declare no competing interests. Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 www.nature.com/scientificreports/ Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-64761-3. Correspondence and requests for materials should be addressed to H.M.H.N.B. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2020 Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-64761-3. Correspondence and requests for materials should be addressed to H.M.H.N.B. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2020 Scientific Reports \| (2020) 10:7769 \| https://doi.org/10.1038/s41598-020-64761-3 17
https://openalex.org/W2801185906	https://discovery.ucl.ac.uk/id/eprint/10049496/1/e018620.full.pdf	English	null	‘Treading water but drowning slowly’: what are GPs’ experiences of living and working with mental illness and distress in England? A qualitative study	BMJ open	2,018	cc-by	7,105	To cite: Riley R, Spiers J, Chew- Graham CA, et al. ‘Treading water but drowning slowly’: what are GPs’ experiences of living and working with mental illness and distress in England? A qualitative study. BMJ Open 2018;8:e018620. doi:10.1136/ bmjopen-2017-018620 Strengths and limitations of this study Objectives This paper provides an in-depth account of general practitioners’ (GPs) experiences of living and working with mental illness and distress, as part of a wider study reporting the barriers and facilitators to help-seeking for mental illness and burn-out, and sources of stress/ distress for GP participants. ► ►Few studies employing qualitative methods, using in-depth interviews, have been used to examine this topic and depict the experience of distress among general practitioners (GPs). on 29 May 2018 by guest. Protect http://bmjopen.bmj.com/ 0 on 3 May 2018. Downloaded from ► ►This study carried out 47 interviews with GPs from across England and contributes to a growing body of research illuminating and examining the experience of living with chronic stress and distress among this population of doctors. Design Qualitative study using in-depth interviews with 47 GP participants. The interviews were audio recorded, transcribed, anonymised and imported into NVivo V.11 to facilitate data management. Data were analysed using a thematic analysis employing the constant comparative method. ► ►Prepublication history for this paper is available online. To view these files, please visit the journal online (http://dx.doi. org/10.1136/bmjopen-2017- 018620). ► ►Participants were self-selecting, which may be per- ceived as a limitation; however, the recruitment of participants ensured that the sample was varied in terms of age, gender, number of years as practis- ing GPs, level of seniority/employment status in the practice and geographical location. 1Institute of Applied Health Research, University of Birmingham, Birmingham, UK 2Centre for Academic Primary Care, University of Bristol, Bristol, UK 3Research Institute, Primary Care and Health Sciences, Keele University, Keele, UK 4Faculty of Health Sciences, University of Bristol, Bristol, UK 5Research Department of Primary Care and Population Health, University College London, London, UK Correspondence to Dr Ruth Riley; r.riley@bham.ac.uk ► ►In recruiting individuals with experience of mental illness and burn-out, we also included the perspec- tives of GPs who had no disclosed history of mental illness which enables the data to be more widely transferable to the UK GP population. Conclusions These findings paint a concerning picture of the situation affecting primary care doctors, with participants’ accounts suggesting there is a considerable degree of mental ill health and reduced well-being among GPs. The solutions are complex and lie in prevention and provision. There needs to be greater recognition of the components and cumulative effect of occupational stressors for doctors, such as the increasing workload and the clinical and emotional demands of the job, as well as the need for a culture shift within medicine to more supportive and compassionate work environments. 1Institute of Applied Health Research, University of Birmingham, Birmingham, UK 2Centre for Academic Primary Care, University of Bristol, Bristol, UK ► ►The multidisciplinary research team independently analysed a subset of transcripts in order to contrib- ute to the generation and refinement of codes to maximise rigour. guest. Protected by copyright. ► ►Emergent themes were subsequently discussed by the whole team to ensure credibility and confirmability. high levels of psychological distress linked to chronic stress, anxiety, depression and burn-out.1–4 These doctors also have higher rates of mental illness and suicide.5–8 ‘Treading water but drowning slowly’: what are GPs’ experiences of living and working with mental illness and distress in England? A qualitative study on 29 May 2018 by guest. Protected by copyr http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018. Downloaded from BMJ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018 Ruth Riley,1 Johanna Spiers,2 Carolyn A Chew-Graham,3 Anna K Taylor,4 Gail A Thornton,2 Marta Buszewicz5 Ruth Riley,1 Johanna Spiers,2 Carolyn A Chew-Graham,3 Anna K Taylor,4 Gail A Thornton,2 Marta Buszewicz5 Open Access Open Access Research ‘Treading water but drowning slowly’: what are GPs’ experiences of living and working with mental illness and distress in England? A qualitative study 1Institute of Applied Health Research, University of Birmingham, Birmingham, UK 2Centre for Academic Primary Care, University of Bristol, Bristol, UK 3Research Institute, Primary Care and Health Sciences, Keele University, Keele, UK 4Faculty of Health Sciences, University of Bristol, Bristol, UK 5Research Department of Primary Care and Population Health, University College London, London, UK Correspondence to Dr Ruth Riley; r.riley@bham.ac.uk Setting England. Participants A purposive sample of GP participants who self-identified as: (1) currently living with mental distress, (2) returning to work following treatment, (3) off sick or retired early as a result of mental distress or (4) without experience of mental distress. Interviews were conducted face to face or over the telephone. Received 10 July 2017 Revised 29 November 2017 Accepted 2 January 2018 Received 10 July 2017 Revised 29 November 2017 Accepted 2 January 2018 ► ►This variation provided a range of perspectives; however, uniform differences between the groups were not observed. Results The findings report GP participants’ in-depth experiences of distress and mental illness with many recollecting their distressing experiences and significant psychological and physical symptoms relating to chronic stress, anxiety, depression and/or burn-out, and a quarter articulating thoughts of suicide. Many talked about their shame, humiliation and embarrassment at their perceived inability to cope with the stresses of their job and/or their symptoms of mental illness. ► ►The researchers did not employ respondent valida- tion; however, the second coders included academic GPs and team members with lived experience of mental illness which afforded some checks and bal- ances to the validity of the analytical process, inter- pretation of data and generalisability of the research findings. Ruth Riley,1 Johanna Spiers,2 Carolyn A Chew-Graham,3 Anna K Taylor,4 Gail A Thornton,2 Marta Buszewicz5 Methods on 29 May 2018 by guest http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018. Downloaded from This multicentre, qualitative study employed in-depth interviews with 47 GPs in England. Information about the study was advertised through professional publica- tions such as Pulse, study newsletters, social media and national and local GP networks (including local medical committees) in Bristol, Manchester and London. A subsa- mple (n=12) was recruited through a specialist treatment service. GPs interested in taking part were asked to self-se- lect into the following groups: (1) Living with anxiety, depression, stress and/or, burn-out; (2) Returning to work following treatment; (3) Off sick or retired early due to psychological ill health (4) No mental ill health. GPs were offered reimbursement of £80 to recognise the time for their participation. p p GPs who expressed an interest in participating were purposively sampled to represent as even a spread as possible across these four groups, although the largest number of participants were in group 1. We intended to recruit approximately 10 participants per group. However, the majority of GPs who contacted us self-se- lected into group 1 and, due to the emergent rich data, continued recruitment to this group and further explo- ration of emerging themes was justified in meeting the study’s aims and objectives. We endeavoured to recruit more participants into groups 2 and 4 using targeted publicity information, but because of time constraints those groups remained marginally under-recruited. In the event, many GPs who identified as living with no stress reported as having had experiences of stress and distress at some juncture in their career. More female GPs contacted the study team expressing an interest in partic- ipating, and therefore the disparity in numbers inter- viewed reflects this. The iterative process of recruitment, sampling and analysis ensured that emerging concepts and themes could be tested out among participants with different characteristics (eg, partners vs locum GPs). The increasing prevalence of mental ill health among the medical profession, including GPs, has been attributed to a number of factors, most notably the esca- lating occupational stresses related to workload, long hours and the emotional demands of the job.4 18–23 In addition, sharp increases in patient demand19 24 and transfer of care from specialist to primary care,25 as well as reductions in resources,25 have resulted in the pres- sure on primary care in England being at its highest ever. Open Access Evidence from industrialised countries (eg, UK, USA, Australia) indicates that suicide rates are higher among GPs, psychiatrists and anaesthetists compared with other medical specialties.6 8 Doctors undergoing fitness for practice investigations are at particularly high risk of suicide.9 Compared with the general population, doctors also have higher rates of drug and alcohol misuse and dependency, with a higher rate of misuse of prescription medicines.10 Psychological difficulties or distress expe- rienced by doctors can be further compounded by self and societal stigmatising associated with mental illness, with evidence that doctors avoid or delay disclosure and help-seeking due to concerns about loss of confidenti- ality, denial of illness and logistical barriers to accessing support.11–15 The prevailing hegemonic culture among medical practitioners and non-disclosure of vulnera- bility, with shame attached to mental health problems or psychological difficulties and the consequent perceived inability to cope within the job, also hinders help- seeking.4 11 16 17 It has been suggested that personality characteristics such as perfectionism and obsessionality, while being positively associated with hard-working and conscientious professionals, may play a role in predis- posing doctors towards chronic stress and/or mental ill health.4 Introduction Compared with the general population, doctors, including general practitioners (GPs) working in primary care, experience 1 Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 1 Open Access Introduction and background y g 4. Suicidal ideation—thoughts of suicide. ► ►Describe average working day in practice (hours, surgery, home vis- its) and any additional responsibilities. 5. Shame, as well as feelings of failure and humiliation, resulting from a perceived lack of resilience and in- ability to cope. Current well-being ► ►Explore current well-being, feelings about work, levels of stress, work– life balance. It should be noted that some participants described or disclosed symptoms that could be clearly attributed to chronic stress, anxiety, depression and burn-out. Other participants described or framed their symptoms in terms of diagnostic concepts associated with these categories, while others talked about their feelings and emotional state without necessarily labelling them. Notably, the symptomatology was identified across all groups including group 4, that is, individuals who self-identified as having no mental health concerns. The themes also applied to ► ►Explore causes of stress/distress (workload, hours, admin, clinical caseloads, organisational issues, lack of support, personal issues, pre-existing mental health symptoms). p g y p ) ► ►Explore reasons for early retirement/sickness (if relevant). Methods This pressure is predicted to increase in the future.25 Findings such as these highlight the potentially stressful context in which GPs are currently working, leaving many vulnerable to distress and/or mental health chal- lenges.4 13 14 23 Face-to-face or telephone interviews lasting between 27 and 126 min (mean=69 min) were conducted between April and September 2016. The majority of interviews were conducted in participants’ homes. The in-depth interviews were conducted by two authors (JS, RR) using a flexible topic guide (see box 1). This was informed by the existing literature, input from GPs on the study team and PPI consultation exercises conducted with GP networks prior to obtaining funding. The interviews were audio recorded, transcribed, anonymised and imported into NVivo V.11 to facilitate data management. Analysis and data collection were conducted iteratively. A thematic analysis was performed involving a process of constant comparison between cases.33 Analysis commenced with JS generating an initial coding framework grounded in the data, which was added to and refined, with material regrouped and recoded as new data were gathered. Codes were gradually built into broader categories through comparison across transcripts, and higher-level recurring themes were developed until data saturation was reached when no new themes were arising from the data.34 Evidence of the extent of mental ill health and burn-out among GPs is largely derived from survey data, with some small-scale qualitative studies exploring the topic.13 26–30 To date, few studies have explored the experiences and accounts of GPs living and working with enduring levels of stress, mental ill health and psychological distress using in-depth qualitative methods. The employment of in-depth interviews in this study allowed GP participants to describe the full extent and complexity of their distress. In addition, by providing the ‘patient’s view’,31 qualitative findings can fully elucidate details and context about phenomena in a way that quantitative work cannot.32 This paper reports GPs’ detailed experiences of living and working with mental illness and emotional distress, as part of a wider study reporting the sources of stress and distress23 for GPs and the barriers and facilitators to help-seeking for mental illness and burn-out14 among this medical population. May 2018 by guest. Protected by copyright. uest. Protected by copyright. Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 2 Open Access 3. Depression—distressed at work, insomnia and ex- haustion, negative thoughts, low self-esteem, irritabil- ity and anger. Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 Managing stress ► ►Explore how general practitioners (GPs) manage their workload/ stress in their day-to-day work life, what they do to relax, to look af- ter themselves (self-care strategies: supportive relationships, sport, exercise relaxation techniques). ) ► ►Explore relationship with colleagues and whether/how/if concerns are raised, how they are responded. Table 1 Participant and practice characteristics GP characteristics n=47 N Sex (female) 33 Age 20–29 1 30–39 12 40–49 20 50–59 14 Group 1 19 2 9 3 11 4 8 Years since qualified <10 19 No of sessions contracted per week <5 sessions (mean actual hours worked) 12 (15) >5 sessions (mean actual hours worked) 32 (38) Fully retired 3 Mean size of practice 12 624 Range 3600–37 000 Employment status Partner 17 Salaried 11 Locum 5 Registrar 4 Retired 3 Sick leave 5 More than one role 2 GP, general practitioner. ► ►Explore if GPs have received informal/formal supervision or mentor (1:1 or group) and experience/value of group. ► ►Explore thoughts/feelings about seeking help, barriers to seek- ing help (stigma/shame, fears about confidentiality, uncertainty of where to go). Reflexivity was employed throughout the research. Both interviewers were experienced qualitative researchers who both reflected on and discussed the impact of the data on their cognitive and emotional sensing throughout the study. Both researchers also discussed and made explicit how their epistemological (JS with a background in psychology and RR in medical sociology) and experi- ential backgrounds may have oriented the data collection and analytical process. Members of the multidisciplinary research team inde- pendently analysed a subset of transcripts in order to contribute to the generation and refinement of codes to maximise rigour. Emergent themes were discussed by the whole team to ensure credibility and confirmability. Symptoms of depression participants irrespective of seniority or role in the prac- tice, age or gender. There was no clear distinction in terms of presentation or severity of symptoms between the groups and so no typology could be developed along these lines. Symptoms of depression Many GP participants described symptoms relating to depression, which included: feeling distressed at work, low mood, hopelessness, sleep problems, low self-esteem, guilt, restlessness, fatigue and suicidal feelings. Some participants employed a range of poignant and powerful metaphors which illustrate the extent of those difficult emotions: Symptoms of burn-out Some participants reported feeling a loss of empathy or compassion fatigue, feeling empty, exhaustion, difficulty making decisions, low job satisfaction and loss of enjoy- ment for their work, using language which highlights the depth of the feelings: And, to be honest with you, that kind of er—the best way that I could describe that is that it’s kind of like a dark shadow that’s in the corner of the room […] I basically felt like I was treading water but drowning slowly. (P45, Male Salaried) I’d lost my empathy with my patients. You know, and er I’m just—I was just—you’re just like a ketchup bottle in a production line […] But from a personal point of view I’m still feeling a bit washed out by it, you know, I feel a little bit empty. You know, I feel like I’m burnt out really a bit, to be honest. (P25, Male Salaried) Results Willing participants were purposively sampled from GPs who expressed an interest in the study, based on demo- graphic and practice characteristics as well as self-selected group type. The characteristics of the participants are found in table 1. For this paper, we will present the following five main themes and corresponding subthemes relating to: 1. Burn-out—loss of empathy or compassion fatigue, feeling empty and/or exhausted, difficulty making decisions, low job satisfaction and loss of enjoyment in seeing patients. g p 2. Anxiety—psychological symptoms relating to anxiety, including feeling anxious or on edge, worrying, irrita- bility, inability to relax and physical symptoms related to anxiety such as panic attacks, hyperventilating, pal- pitations, sweating and nausea. Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 3 Distressed at work Many participants recalled their experiences of feeling distressed, anxious and tearful, with some describing breaking down at home, after or during work or in front of patients: I was saying all the right things but […] I felt quite detached from it. (P27, Female Salaried) I mean there were times certainly when I arrived at work and just sat outside and cried before I even went in the building. And I often cried on the way home. I’d often cry when I came in from work, um just col- lapse in a heap, and get to bed actually crying. (P31, Female Locum) on 29 May 2018 by guest. Protec http://bmjopen.bmj.com/ 0 on 3 May 2018. Downloaded from The following participant describes how feeling over- worked can lead to a reduced ability to offer compassion and support to patients: If you find yourself working too many sessions, I can, you know, I feel it, and you almost feel like you lose the milk of human kindness. And you can’t do that job if you’re not (pause) at your best. (P21, Female, Salaried and Locum) I remember once actually sitting under the desk, hug- ging my knees to my chest, I was so anxious. (P25, Male Salaried) I would just end up in tears all the time at the end of each session. (P53, Female Salaried) Negative thoughts/low self-esteem Negative thoughts/low self-esteem Um, and then I think the hard thing, the worst thing I think I found was when people were feeling suicidal, and they were saying they were feeling suicidal and why they were feeling suicidal. And normally I’d be like, ‘Look, you know, it’s temporary, it’ll get better. These are the things that we can do,’ but I just sat there thinking, ‘Well, you’re probably right, it is rub- bish,’ you know. (P27, Female Partner) Um, and then I think the hard thing, the worst thing I think I found was when people were feeling suicidal, and they were saying they were feeling suicidal and why they were feeling suicidal. And normally I’d be like, ‘Look, you know, it’s temporary, it’ll get better. These are the things that we can do,’ but I just sat there thinking, ‘Well, you’re probably right, it is rub- bish,’ you know. (P27, Female Partner) Some participants talked about grappling with cycles of negative thoughts, symptomatic of low self-esteem and a common symptom of depression, as this participant illustrates: And so then I’m really struggling, because I’m strug- gling with my internal dialogue saying, you know, ‘Not good enough, could be doing a better job. Not really, not really er making people better.’ […] per- sistent negative cycles of er (pause), yeah, of negative thoughts about yourself (P20, Male Partner) Irritability and anger Some study participants described feeling irritable and sometimes angry; a possible symptom of depression in this context: Some study participants described feeling irritable and sometimes angry; a possible symptom of depression in this context: But I think, towards the end when I stopped lo- cum-ing again, I was beginning to feel very tired as a result and that it was taking more out of me than it ought to really, and feeling um, yeah, feeling more irritable. (P47, Female on sick leave) And you do feel a huge sense of shame, huge sense of shame. Because even though I treat patients day in, day out with anxiety and depression, it felt like this was my not coping. And I was thinking, ‘Why are the partners all able to cope with this but not me?’ and that sort of thing. So it was all rolled into one really. (P4, Female Partner) on 29 May 2018 by guest. Protec http://bmjopen.bmj.com/ 0 on 3 May 2018. Downloaded from This account demonstrates how feeling stressed and irritable can sometimes manifest itself within a work context: Er (sigh) yeah, I think for me it was very embarrass- ing. Didn’t want to admit it. Couldn’t believe that like I’d got to this stage. Cos I’d never been someone like that […] yeah, I did find it really hard, I think, to actu- ally accept that this had happened. Um (pause) yeah, bit of a failure? I suppose, in a way. Kind of feel like you’re not strong enough? (P44, Female Registrar) …it was fairly um traumatic really, I have to say… I think I’d become quite stressed at work actually and I’d lost my temper in meetings at times and um nobody quite knew what to do about it. (P39, Male Partner) The participant below highlights how the denial and intolerance of mental ill health among some colleagues can enter into an individual’s internal dialogue, thus compounding existing feelings of shame: Shame and feelings of failure Perhaps symptomatic of the stigma attached to mental illness and the culture of invulnerability within medicine, some GP participants described how they felt ashamed, embarrassed, humiliated and a sense of having failed, due to their perceived lack of resilience and inability to cope. These quotations illustrate the depth of that emotion for two of the participants: Suicidal ideation One in four GP participants described having had suicidal thoughts, with the following participants reflecting on occasions when they had contemplated and ruminated on ways of taking their own lives: ‘Yeah, well you don’t get ill, do you? You know, you, you refuse to accept any illness in yourself, don’t you? It’s just a question of that’s why doctors present late with illnesses because they don’t like to make a fuss. We see it as a sign of weakness, which I think proba- bly is why we are intolerant of illness in our partners, because we’re doctors. And I think it’s just part of the sort of macho thing which is first drummed into you at medical school and the end product of that gener- ation, which—if you can’t take the heat get out of the kitchen, type thing’. (P49, Female, Retired) And it just came on me like a wave. And then I started thinking, ‘I’m going to jump onto those train tracks.’ (P45, Male Salaried) And it just came on me like a wave. And then I started thinking, ‘I’m going to jump onto those train tracks.’ (P45, Male Salaried) And um constant ruminations on suicide and doing myself in. And the trick with suicide is to persuade yourself that those around you would be better off without you, um and I got to that point and I decided on ways that I might (pause) do myself in and rumi- nated about that all the time. Just, it just filled my waking and sleeping nights. (P20, Male Partner) ay 2018 by guest. Protected by copyright. uest. Protected by copyright. This latter data extract demonstrates how suicidal thoughts can be both constant and normalised for some GPs. Symptoms of anxiety Among the myriad of symptoms, insomnia and exhaustion were a commonplace experience for study participants: Among the myriad of symptoms, insomnia and exhaustion were a commonplace experience for study participants: The majority of participants, including those who self-identified as having no mental health concerns (group 4), described psychological symptoms relating to anxiety, including feeling on edge or hyperalert, persistent worrying, irritability and an inability to relax. The following examples demonstrate how difficult and distressing these symptoms and feelings were for some participants to negotiate: Um - so I wasn’t sleeping and I wasn’t—wasn’t eating very well, and—and it just all sort of took its toll. (P23, Female Salaried) It’s very hard to function when you’re not ever get- ting refreshing sleep […] And I just wouldn’t stop. And absolute, incapacitating terror. Um—So, but one’s having um—night, night terrors maybe two, three, four times a week. Sometimes more than one a night. (P24, Female Partner) Everything was running around in my mind at 200 miles an hour. I was in sort of hyper drive and, yeah, you wake up to super stimulated, super awake, super, you know, on edge and anxious. (P36, Male Partner) Many GP participants detailed their exhaustion, which was often linked to insomnia and could be symptomatic of their anxiety, depression and/or feeling burnt-out, demonstrating the ways in which different elements of GP experience combine to add to distress: Many participants also described physical symptoms related to their anxiety, such as panic attacks, hyperventi- lating, palpitations, sweating and nausea: guest. Protected by copyright. Erm (pause) and I, and I was getting quite bad chest pain, palpitations, I was feeling anxious in many cir- cumstances, like the school run, just meeting people, going to meetings and it was becoming incapacitat- ing where it felt like somebody had got a knife in my chest all the time. And I knew it wasn’t anything to do with my heart. (P31, Female Locum) Probably like sort of swimming through treacle, so everything seemed—um—much harder. (P38, Male Partner) So like you can’t, you can’t avoid that fatigue. You might be able to put it off a bit (laughs) but you’re going to pay interest on it. (P7, Female Locum) Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 4 Open Access Discussion This findings highlight that internalised and perceived stigma still prevent doctors from seeking support.11 There has been a paucity of in-depth qualitative research exploring this topic area with doctors, including GPs.14 This study highlights that many GPs are living and working with extreme levels of emotional distress. A notable proportion of participants vividly recalled and articulated both fleeting and recurring thoughts of suicide, which is reflective of the extent of individuals’ distress and suffering. These candid, evocative and some- times disturbing accounts, some hitherto unvoiced, reveal a population of doctors with generally very poor mental health. Access to personal, relational and environmental resources are protective factors for good mental health and well-being and impact on individuals’ resilience.39 Crucially, collegial support is a protective factor for good mental health—support from mentors, supervisors and colleagues is associated with resilience and reduced sickness.22 The findings from this study and the related papers14 23 were based on in-depth interviews with 47 GPs from across England and contribute to a growing corpus of research illuminating and examining the causes, experiences and impact of chronic stress and distress among GPs.20 35 The participants were self-selecting, which may be seen as a limitation; however, the recruitment of participants ensured that the sample was varied in terms of age, gender, number of years as practising GPs, level of seniority/ employment status in the practice and geographical loca- tion. In addition to recruiting individuals with self-re- ported mental ill health and burn-out, we also included the perspectives of GPs who had no disclosed history of mental health problems, which enables the data to be more transferable to the wider GP population. However, despite this heterogeneity within the sample, we did not observe any marked differences between the groups, suggesting a certain uniformity of distress for the group despite demographic differences. Balint groups or similarly structured group work or supervision continue to be employed in general practice and are valued by GPs,40 yet are not used across all prac- tices. Individual or group supervision aims to provide a safe and supportive space where staff can openly discuss the pressures and emotional challenges of their work and may, as previous evidence suggests, provide GPs with the support they need while offering protection against compassion fatigue and burn-out.41 Its wider use may need to be considered. Tackling the culture of invulnerability early on in medical training is also key. Discussion The accounts of the 47 participants in this study suggest there may be an undetermined population of GPs who are living and working with poor mental health. Partici- pants vividly described and recollected their distressing experiences and troubling symptoms, with one in four GPs articulating thoughts of suicide. Such findings are The following GP reflects on the advice she gives to her patients who express suicidal thoughts and how her feel- ings mirrored her patients’ expressed hopelessness and despair, again showing the ways in which GP experiences can combine to add to distress: 5 Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 Open Access currently under-reported in the literature with few studies providing evidence of suicidal thoughts and extreme levels of distress among this GP population. The GP participants described numerous troubling psychological and physical symptoms relating to chronic stress, anxiety, depression and/or burn-out with several breaking down at work—crying, feeling despair and some contemplating suicide. Many also talked about their shame, humiliation and embarrassment at their perceived inability to cope with the stresses of their job and/or their symptoms of mental ill health. The experiences and feelings of many GP participants depict a dispiriting emotional landscape, with notable signs of poor mental health and striking levels of distress. and coping on a day-to-day basis with symptoms of chronic stress, anxiety and depression, ranging from sleeplessness to persistent anxiety. Evidence indicates that chronic stress negatively impacts on physical health and is a key risk factor for mental illness, including anxiety and depression, and can lead to the adoption of maladaptive coping mechanisms such as nicotine, alcohol, and illicit drug use.37 on 29 May 2018 by guest. Protected by copyright. http://bmjopen.bmj.com/ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018. Downloaded from g Normalising and acknowledging chronic stress as an occupational hazard associated with being a doctor is helpful, but only if solutions are sought which focus on prevention and the provision of support to maintain the health and well-being of frontline clinicians. Since January 2017, GPs and GP trainees in England suffering from mental ill health and addiction have been able to access free support through the National Health Service’s GP Health Service.38 Further research and evaluation must be undertaken to assess the effectiveness and accessibility of this service. The emphasis now needs to focus on preventing mental ill health and improving the support available to doctors. Conclusion 4. Brooks SK, Gerada C, Chalder T. Review of literature on the mental health of doctors: are specialist services needed? J Ment Health 2011;20:146–56. These findings, together with those from the wider study,14 23 paint a concerning picture of the situa- tion affecting primary care doctors, with participants’ accounts suggesting there is a population of GPs expe- riencing a considerable degree of mental ill health, and overall reduced well-being among GPs. The solutions are complex and prevention lies in addressing the systemic causes of stress (related to diminishing resources, increased workload, etc). In addition to recognising the components and cumulative effect of occupational stressors for doctors, such as the increasing workload and the clinical and emotional demands of the job, there is need for a culture shift within medicine to more supportive and compassionate work environments where it is safe and acceptable for doctors to talk or seek help when they need it. Medical schools, medical institutions and individuals need to address and tackle the stigma of mental illness and engender a greater tolerance and acceptance of vulnerability. 5. Dobbin A. Burnt out or fired up? Helping recovery in distressed patients can increase our own resilience. Br J Gen Pract 2014;64:497–8. 6. Hawton K, Clements A, Sakarovitch C, et al. Suicide in doctors: a study of risk according to gender, seniority and specialty in medical practitioners in England and Wales, 1979-1995. J Epidemiol Community Health 2001;55:296–300. y ; 7. Richings JC, Khara GS, McDowell M. Suicide in young doctors. Br J Psychiatry 1986;149:475–8. y ; 7. Richings JC, Khara GS, McDowell M. Suicide in young doctors. Br J Psychiatry 1986;149:475–8. 8. Schernhammer ES, Colditz GA. Suicide rates among physicians: a quantitative and gender assessment (meta-analysis). Am J Psychiatry 2004;161:2295–302. 8. Schernhammer ES, Colditz GA. Suicide rates among physicians: a quantitative and gender assessment (meta-analysis). Am J Psychiatry 2004;161:2295–302. 9. GMC. Doctors who commit suicide while under GMC fitness to practise investigation. 2015. p g 10. Brooks SK, Chalder T, Gerada C. Doctors vulnerable to psychological distress and addictions: treatment from the Practitioner Health Programme. J Ment Health 2011;20:157–64. g 11. Henderson M, Brooks SK, Del Busso L, et al. Shame! self- stigmatisation as an obstacle to sick doctors returning to work: a qualitative study. BMJ Open 2012;2:e001776. 12. Cohen D, Winstanley SJ, Greene G. Understanding doctors’ attitudes towards self-disclosure of mental ill health. Occup Med 2016;66:383–9. 13. Conclusion Thompson WT, Cupples ME, Sibbett CH, et al. Challenge of culture, conscience, and contract to general practitioners' care of their own health: qualitative study. BMJ 2001;323:728–31. q y 14. Spiers J, Buszewicz M, Chew-Graham CA, et al. Barriers, facilitators, and survival strategies for GPs seeking treatment for distress: a qualitative study. Br J Gen Pract 2017;67:e700–e708. Acknowledgements The authors would like to thank all volunteer participants who kindly gave their time to share their experiences of working in English general practice; and collaborators Clare Gerada and Chris Manning for endorsing this study, providing valuable comments on the funding application and assisting with recruitment. q y 15. Cheshire A, Hughes J, Lewith G, et al. GPs' perceptions of resilience training: a qualitative study. Br J Gen Pract 2017;67:e709–e715. g y 16. Department of Health. Mental health and ill health in do London: Department of Health, 2008. Collaborators Clare Gerada; Chris Manning. 17. Gold KJ, Andrew LB, Goldman EB, et al. "I would never want to have a mental health diagnosis on my record": a survey of female physicians on mental health diagnosis, treatment, and reporting. Gen Hosp Psychiatry 2016;43:51–7. Contributors RR, JS, CAC-G and MB: substantial contributions to conception and design, acquisition of data or analysis and interpretation of data; drafting the article or revising it critically for important intellectual content and final approval of the version to be published. AKT and GAT: analysis and interpretation of data; drafting the article or revising it critically for important intellectual content; and final approval of the version to be published. p y y 18. Baird B, Charles A, Honeyman M, et al. Understanding pressures in general practice. London: King’s Fund, 2016. 19. Hobbs FDR, Bankhead C, Mukhtar T, et al. Clinical workload in UK primary care: a retrospective analysis of 100 million consultations in England, 2007–14. The Lancet 2016;387:2323–30. Funding The study was funded by NIHR School for Primary Care Research. g 20. Doran N, Fox F, Rodham K, et al. Lost to the NHS: a mixed methods study of why GPs leave practice early in England. Br J Gen Pract 2016;66:e128–e135. Disclaimer The views and opinions expressed there in are those of the authors and do not necessarily reflect those of the NIHR School for Primary Care Research, NIHR, NHS or the Department of Health. 21. Firth-Cozens J. A perspective on stress and depression. on 29 May 2018 by guest. Protected b http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018. Downloaded from © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted. ► ►Greater recognition of the components and cumula- tive effect of the occupational stressors for doctors, such as increasing workload and the clinical and emotional demands of the job. j ► ►A culture change within medicine, with a shift to more supportive and compassionate work cultures. References 1. Chopra SS. Physician burnout. JAMA 2004;291:633–33. S S C S 1. Chopra SS. Physician burnout. JAMA 2004;291:633–33. M l LJ Si h k S C i SM l R f ► ►The destigmatising of mental ill health and the recog- nition of GPs’ experiences and emotional responses as symptomatic of occupational stress as opposed to any perceived failure to cope in the individual. 2. Merlo LJ, Singhakant S, Cummings SM, et al. Reasons for misuse of prescription medication among physicians undergoing monitoring by a physician health program. J Addict Med 2013;7:349–53. 3. Orton P, Orton C, Pereira Gray D. Depersonalised doctors: a cross- sectional study of 564 doctors, 760 consultations and 1876 patient reports in UK general practice. BMJ Open 2012;2:e000274. Open Access provided the original work is properly cited. See: http://creativecommons.org/ licenses/by/4.0/ In the light of the findings of this study, we recommend the following: provided the original work is properly cited. See: http://creativecommons.org/ licenses/by/4.0/ Discussion Schwartz Centre Rounds, for instance, are currently being piloted in medical schools with early evidence supporting their value.42 First,43 buddy systems, and access to regular supervision or mentorship can also offer support and a reflective space. The British Medical Association has produced guidance on recog- nising signs of mental distress among colleagues with suggestions about what action to take.44 Perhaps caution needs to be applied to the over-reliance on personal resil- ience as a solution as this may shift the responsibility for maintaining good mental health entirely on to the indi- vidual. While self-care is paramount in maintaining good mental and physical health, it is not solely the respon- sibility of individual doctors to look after their mental health; employers and colleagues need to engender a more supportive and compassionate culture. Doctors are not invulnerable to suffering, most especially their own. p g p These findings present a concerning picture of psycho- logical difficulties and reduced well-being affecting primary care doctors. The experience of chronic stress and anxiety in GPs (eg, feeling tense, restless, anxious, worried or having disrupted sleep, panic attacks, hyper- ventilating, palpitations, sweating and nausea) appears commonplace, including in those who did not identify themselves as having mental ill health. Our findings reflect the evidence that doctors frequently normalise, minimise, deflect or downplay how stressed they feel, which may preclude help-seeking and the availability of appropriate support.36 Strikingly, one GP participant normalised the way they routinely ruminated on their suicidal thoughts, with other participants indicating that they were living uest. Protected by copyright. Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 6 Conclusion In: Cox JK J, Hutchinson A, McAvoy P, eds. Understanding doctors’ performance. Oxford: Radcliffe Publishing, 2006:22–5. Competing interests None declared. 22. Firth-Cozens JH J. A little too much, in how to survive in medicine personally and professionally. Oxford: Wiley-Blackwell, 2010. Patient consent We sought consent from clinicians and doctors as patients. Patient consent We sought consent from clinicians and doctors as patients. 23. Riley R, Spiers J, Buszewicz M, et al. What are the sources of stress and distress for general practitioners working in England? A qualitative study. BMJ Open 2018;8:e017361. Ethics approval South West—Frenchay Research Ethics Committee (reference number: 15/SW/0350). Ethics approval South West—Frenchay Research Ethics Committee (reference number: 15/SW/0350). q y p 24. Hippisley-Cox J, Vinogradova Y. Trends in consultation rates in general practice 1995 to 2008: analysis of the QResearch database. London: NHS Information Centre, 2009. Provenance and peer review Not commissioned; externally peer reviewed. Provenance and peer review Not commissioned; externally peer reviewed. Data sharing statement This study has not received ethical approval to share confidential data with any third party other than the study research team. Data sharing statement This study has not received ethical approval to share confidential data with any third party other than the study research team. 25. King’s Fund. Improving the quality in General Practice. London: King’s Fund, 2011. Open Access This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, g 26. Ahola K, Hakanen J. Job strain, burnout, and depressive symptoms: a prospective study among dentists. J Affect Disord 2007;104:103–10. 7 Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620 Open Access Open Access 27. Clarke J, O'Sullivan Y, Maguire N. A study of self-care among Irish doctors. Ir Med J 1998;91:175–6. in a working age population seeking primary care--an observational study. BMC Fam Pract 2015;16. on 29 May 2018 by guest. Protected by copyrig http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2017-018620 on 3 May 2018. Downloaded from y 38. NHS GP Health Service. Secondary NHS GP Health Service 2017. https://www.england.nhs.uk/gp/gpfv/workforce/health-service/ 28. Sibbald B, Bojke C, Gravelle H. National survey of job satisfaction and retirement intentions among general practitioners in England. BMJ 2003;326:22. 39. Cusack L, Smith M, Hegney D, et al. Exploring Environmental Factors in Nursing Workplaces That Promote Psychological Resilience: Constructing a Unified Theoretical Model. Front Psychol 2016;7:1–8. 29. Dyrbye LN, Varkey P, Boone SL, et al. Physician satisfaction and burnout at different career stages. Mayo Clin Proc 2013;88:1358–67 30. Firth-Cozens J. Predicting stress in general practitioners: 10 year follow up postal survey. BMJ 1997;315:34–5. 40. Van Roy K, Vanheule S, Inslegers R. Research on Balint groups: A literature review. Patient Educ Couns 2015;98:685–94. y 31. Armstrong D. The patient's view. Soc Sci Med 1984;18:737–44. 41. Benson J, Magraith K. Compassion fatigue and burnout: the role of Balint groups. Aust Fam Physician 2005;34:497–8. 32. Smith JA. Qualitative psychology: A practical guide to research methods. Washington, D.C: Sage, 2007. 33. Fram SM. The constant comparative analysis method outside of grounded theory. The Qualitative Report 2013;18:1–25. 42. Gishen F, Gill D. Introducing schwartz rounds at UCL Medical School. Secondary Introducing Schwartz rounds at UCL Medical School. https://uclmsnews.wordpress.com/2014/09/29/ introducing-schwartz-rounds-at-ucl-medical-school/ (accessed 15 Sep 2017). 34. Sandelowski M. Sample size in qualitative research. Res Nurs Health 1995;18:179–83. 35. Fisher RF, Croxson CH, Ashdown HF, et al. GP views on strategies to cope with increasing workload: a qualitative interview study. Br J Gen Pract 2017;67:e148–e156. 43. RCGP. http://www.rcgp.org.uk/membership/about-rcgp- membership/join-rcgp-newly-qualified-gps.aspx. accessed 24 Nov 2017 36. Thompson NJ, Corbett SS, Welfare M. A qualitative study of how doctors use impression management when they talk about stress in the UK. Int J Med Educ 2013;4:236–46. 44. BMA. Are you concerned about a colleague? Secondary are you concerned about a colleague? https://www.bma.org.uk/advice/work- life-support/your-wellbeing/concerned-about-a-colleague (accessed 24 Nov 2017). 37. W L, Hange D, Björkelund C, et al. Prevalence of perceived stress and associations to symptoms of exhaustion, depression and anxiety 8 8 Riley R, et al. BMJ Open 2018;8:e018620. doi:10.1136/bmjopen-2017-018620
https://openalex.org/W3001479080	https://hal.sorbonne-universite.fr/hal-02557129/document	English	null	Early fate of exogenous promoters in E. coli	Nucleic acids research	2,020	cc-by	9,149	To cite this version: Malikmohamed Yousuf, Ilaria Iuliani, Reshma T Veetil, Aswin Sai Narain Seshasayee, Bianca Sclavi, et al.. Early fate of exogenous promoters in E. coli. Nucleic Acids Research, 2020, 48 (5), pp.2348-2356. 10.1093/nar/gkz1196. hal-02557129 Early fate of exogenous promoters in E. coli Malikmohamed Yousuf, Ilaria Iuliani, Reshma T Veetil, Aswin Sai Narain Seshasayee, Bianca Sclavi, Marco Cosentino Lagomarsino ABSTRACT Gene gain by horizontal gene transfer is a major pathway of genome innovation in bacteria. The cur- rent view posits that acquired genes initially need to be silenced and that a bacterial chromatin pro- tein, H-NS, plays a role in this silencing. However, we lack direct observation of the early fate of a horizontally transferred gene to prove this theory. We combine sequencing, ﬂow cytometry and sort- ing, followed by microscopy to monitor gene expres- sion and its variability after large-scale random in- sertions of a reporter gene in a population of Es- cherichia coli bacteria. We ﬁnd that inserted promot- ers have a wide range of gene-expression variability related to their location. We ﬁnd that high-expression clones carry insertions that are not correlated with H- NS binding. Conversely, binding of H-NS correlates with silencing. Finally, while most promoters show a common level of extrinsic noise, some insertions show higher noise levels. Analysis of these high- noise clones supports a scenario of switching due to transcriptional interference from divergent ribo- somal promoters. Altogether, our ﬁndings point to evolutionary pathways where newly-acquired genes are not necessarily silenced, but may immediately explore a wide range of expression levels to probe the optimal ones. The high fraction of mobile genes in bacterial genomes is a source of a great diversity of phenotypes. This large diversity challenges the very concept of species, and has enormous importance for understanding pathogenicity and antibiotic resistance (1). At the genetic level, Escherichia coli genomes vary dramatically in their sizes ranging from 4.5 to 6 Mb. Comparative genomic surveys of E. coli have shown that there is a core set of genes which is highly conserved across the species and coexists with a large pangenome, the set of genes that can be gained by horizon- tal gene acquisition from other species (2). Indeed, bacteria acquire exogenous DNA by transformation (of naked DNA from the environment), transduction (of DNA from bac- teriophages) or conjugation (from fellow bacteria through molecular pipes such as pili) (3). In order to be functional, exogenous acquired genes often need for the metabolic and the regulatory circuitry of the cell to be rewired (4,5). Fur- thermore, expression of a foreign gene can interfere with the resources allocated for endogenous gene expression. There- fore, horizontally acquired genes must be regulated (6). C⃝The Author(s) 2020. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Malikmohamed Yousuf1,2,†, Ilaria Iuliani1,3,†, Reshma T. Vee Aswin Sai Narain Seshasayee4, Bianca Sclavi 1,3 and Marco Cosentino Lagomarsino 6,7,8,9,* 1LBPA, UMR 8113, CNRS, ENS Paris-Saclay, 61 Avenue du President Wilson, 94235 Cachan, France, 2Current Afﬁliation: Centre for Clinical Brain Sciences, The University of Edinburgh, Edinburgh EH16 4SB, UK, 3Current Afﬁliation: LCQB, UMR 7238, Sorbonne Universit´e, 4 Place Jussieu, 75005 Paris, France, 4National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bangalore 560065, Karnataka, India, 5School of Life science, The University of Trans-Disciplinary Health Sciences and Technology (TDU), Bengaluru 560064, Karnataka, India, 6Sorbonne Universit´e, Campus Pierre and Marie Curie, 4 Place Jussieu, 75005 Paris, France, 7CNRS, UMR7238, 4 Place Jussieu, 75005 Paris, France, 8Current Afﬁliation: IFOM, FIRC Institute of Molecular Oncology, Via Adamello 16, 20143 Milan, Italy and 9Current Afﬁliation: Physics Department, University of Milan, and I.N.F.N., Via Celoria 16, 20133 Milan, Italy Downloaded from https://academic.oup.com/nar/article-abstract/48/5/2348/5710780 b Received October 16, 2019; Revised December 05, 2019; Editorial Decision December 08, 2019; Accepted December 20, 2019 eceived October 16, 2019; Revised December 05, 2019; Editorial Decision December 08, 2019; Accepted Decem *To whom correspondence should be addressed. Tel: +39 02 5743 03200; Email: marco.cosentino-lagomarsino@ifom.eu †The authors wish it to be known that, in their opinion, the first two authors should be regarded as Joint First Authors. HAL Id: hal-02557129 https://hal.sorbonne-universite.fr/hal-02557129v1 Submitted on 28 Apr 2020 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. 2348–2356 Nucleic Acids Research, 2020, Vol. 48, No. 5 doi: 10.1093/nar/gkz1196 Published online 21 January 2020 Efficient protocol for production and characterization of sys- tematic exogenous reporter insertions The promoter chosen to control GFP expression in the ran- domly inserted cassette is the rrnBP1 promoter of the rrnB ribosomal operon. We chose this well-characterized highly expressed promoter because it is regulated by changes in DNA supercoiling and by the abundant nucleoid proteins Fis and H-NS (30). We also considered, as a control, a shortened version of the promoter lacking regulation by the nucleoid proteins Fis and H-NS but still regulated by DNA topology. Additionally, at least through the action of H-NS, which is a notorious nucleoid-shaping protein (20), the dynam- ics of horizontal transfers is related to the physical orga- nization of the chromosome. An important question to be addressed is whether and how the organizational features of the E. coli chromosome -such as the ‘macrodomain’ architecture (21–24)- are correlated with gene acquisition and control of gene expression, particularly of acquired genes (25). Figure 1A describes our pipeline (see Methods and Sup- plementary Figure S1). On the order of 105 transposed colonies (this estimate was based on manual counting as we knew the number of transposed colonies in each plate) were mixed and grown overnight in minimal medium. This pop- ulation was regrown to a fixed OD in the same medium, and initially assayed by population sequencing (Figure 1B and flow cytometry Figure 1C–E). We then used a cell sorter to select sub-populations based on gene expression levels (Figure 1C–E). Each of these subpopulations was grown overnight, regrown to exponential phase and then sorted again as a function of GFP content, for a total of four rounds (Supplementary Figure S1). Finally, 658 randomly hand-picked clonal populations were individually charac- terized by flow cytometry. A subset of 96 from the 658 clonal populations representing different sorted populations were randomly selected and sequenced, and 90 of these were used to measure gene expression and growth rate in a plate- reader assay (see Supplementary File SF1). A smaller se- lected subset of clones was used to measure the dynamics of gene expression in single-cell microcolony growth assays by epifluorescence microscopy. Horizontally transferred genes are often clustered along the genome (3,23,26–28). In part, this reflects joint transfer of functionally co-dependent genes that would provide no benefit if transferred independently. In part, however, this reflects the existence of ‘permissive’ zones along the chro- mosome, which experience recurrent integration and high turnover. Efficient protocol for production and characterization of sys- tematic exogenous reporter insertions Permissive zones can originate or be reinforced through physical integration biases, where the presence of integrases and/or recombinogenic sites facilitates acquisi- tion of genetic material (1). Additionally, in many species including E. coli, horizontally acquired genes preferentially accumulate near the (AT-rich) terminus region (1,2,23), possibly to avoid deleteriously high expression near the ori- gin due to gene copy number effects. The genome sequence organization of a given species is a result of selection pressure and architectural constraints. Some of these have been clearly identified (3,6,17,28,29). For example, highly expressed, newly acquired genes must be kept from interfering with the expression of essential genes. However, comparatively little is known about the early dynamics of acquired genes. Do clear physical inser- tion biases emerge? What are the phenotypic impacts of in- serted genes and how are they linked with expression levels? How, in turn, is gene expression a consequence of the locus of insertion? Are they immediately silenced and do H-NS and nucleoid organization play a role? RESULTS Efficient protocol for production and characterization of sys- tematic exogenous reporter insertions ABSTRACT A i d b hi h h i f h i ll A primary mode by which the expression of horizontally- acquired genes is regulated is believed to be transcriptional repression, which is achieved by proteins such as H-NS in enterobacteria, including E. coli ((7–10), reviewed in (11)). Many previous studies support both the need of initial re- pression of acquired genes, and the view that H-NS repres- sion is relevant for the successful establishment of these genes (5,6,12,13). H-NS is among several ‘global’ transcrip- tional regulators that affect the expression of hundreds of genes in E. coli. It binds to AT-rich or intrinsically bent DNA sequences and forms structures such as stiff rods or Nucleic Acids Research, 2020, Vol. 48, No. 5 2349 inserted and initially maintained in specific chromosomal contexts, as well as to characterize its fate in terms of both gene expression activity and noise of the transcription re- porter constructs at different insertion sites on the genome. DNA–protein–DNA bridges, which might act as geometri- cal motifs for transcriptional silencing (14,15). Many H-NS binding regions are up to a few kilobases long. The length of these binding regions correlates with the degree of tran- scriptional repression imposed on the target gene (9,16) and genes regulated by H-NS are very highly expressed in the absence of this repressive control (17,18). Since the levels and activity of H-NS depend on environmental conditions and growth rate, this level of regulation allows for a coor- dinated gene-expression change needed for cellular adap- tation. Studies of gene expression of inserted reporter cas- settes at different genomic locations (16,19) have demon- strated that gene expression of an identical regulatory sys- tem can vary greatly, beyond the effects of gene dosage, for three main reasons, supercoiling, activity of neighbour pro- moters and H-NS regulatory activity. A gene’s local envi- ronment can thus provide a fitness advantage, associated to the selection of the gene’s position over evolution. MATERIALS AND METHODS Detailed Methods are available as Supplementary Materi- als. Bimodal distribution of gene expression in parental popula- tions and low-expression sub-populations Insets in panels (D) and (E) show insertions found by population sequencing (y-axis are counts in logarithmic scale), with overall similarity but local differences. B D E D D E Downloaded from https://academic.oup.com/nar/article-abstract/48/5/2348/5710780 by C E Figure 1. Insertion localization and sorting by gene expression. (A) Experimental pipeline. Massive transposon insertion of a GFP reporter gene cassette in ∼100 000 founder strains was tested by plating on kanamycin-selective agar and PCR. Surviving colonies were mixed, grown overnight in LB, resuspended and grown to a fixed OD. (B) Sequencing of resulting parental populations yields the locations of the insertions, shown in the top-left panel (y axis are counts in logarithmic scale). The bottom panel compares a 3 kb sliding average of the coverage (black line, y-axis rescaled for comparison) with the prediction from gene dosage, and the experimental dosage (red dashed line) measured by whole-genome sequencing (blue line) the right panels are controls that the trend of insertions copy number is not due to ribosomal genes (orange line) and to the insertions with top 10% coverage (>3000 reads/bin, purple line). (C–E) Forward scatter versus GFP expression measured by flow-cytometry. FACS Sorting by the level of fluorescence was performed on a total of four rounds (see Supplementary Figure S1). Selecting for high expression (RH) from the parental population (C) yielded a population with a similar distribution of gene expression (D), while selecting for low expression yielded a population with a bimodal distribution of gene expression (E). Insets in panels (D) and (E) show insertions found by population sequencing (y-axis are counts in logarithmic scale), with overall similarity but local differences. populations derived from the low GFP expression popula- tions, particularly those filtered for very low expression lev- els, showed high-frequency insertions in ribosomal regions (Supplementary Figure S2). compared to the parental one. This variability is the combi- nation of the variability of promoter expression across sin- gle cells that are clonal (i.e. where the insertion is in the same exact position) and the variability of mean expression be- tween clones with different insertion locations. Thus, the clonal variability should be considerably high in order to account for the fact that the overall pattern of variability is robust in the sorted sub-populations (which contain less clonal variants). Bimodal distribution of gene expression in parental popula- tions and low-expression sub-populations Comparison of the fluorescence distribution in the sorted populations obtained from the high- (RH) and low- expressing (RL) fractions of the parental population (Fig 1C–E) shows that the low-expressing (RL) population have a sub-population of clones with very low expression. In the parental population (Fig 1C), some of these low- expression clones are already visible (green box). Sorting them from the RL population gave rise to the population RLR1V1L. To access some of the above questions, we devised an ex- perimental assay (Figure 1) where a cassette including an antibiotic resistance gene and a GFP reporter under the control of a highly expressed ribosomal promoter is inserted systematically in the genome, and the resulting mixed and clonal populations are analysed by sequencing and single- cell biology methods. This methodology allows us to de- scribe statistical tendencies for a reference promoter to be Outside of this low-expression peak, the distribution of gene expression has little variation in the sorted populations 2350 Nucleic Acids Research, 2020, Vol. 48, No. 5 A B C D E Figure 1. Insertion localization and sorting by gene expression. (A) Experimental pipeline. Massive transposon insertion of a GFP reporter gene cassette in ∼100 000 founder strains was tested by plating on kanamycin-selective agar and PCR. Surviving colonies were mixed, grown overnight in LB, resuspended and grown to a fixed OD. (B) Sequencing of resulting parental populations yields the locations of the insertions, shown in the top-left panel (y axis are counts in logarithmic scale). The bottom panel compares a 3 kb sliding average of the coverage (black line, y-axis rescaled for comparison) with the prediction from gene dosage, and the experimental dosage (red dashed line) measured by whole-genome sequencing (blue line) the right panels are controls that the trend of insertions copy number is not due to ribosomal genes (orange line) and to the insertions with top 10% coverage (>3000 reads/bin, purple line). (C–E) Forward scatter versus GFP expression measured by flow-cytometry. FACS Sorting by the level of fluorescence was performed on a total of four rounds (see Supplementary Figure S1). Selecting for high expression (RH) from the parental population (C) yielded a population with a similar distribution of gene expression (D), while selecting for low expression yielded a population with a bimodal distribution of gene expression (E). Bimodal distribution of gene expression in parental popula- tions and low-expression sub-populations There are, however, some important dif- ferences in the distributions, mirrored by differences in the location and frequency of the insertions in the sorted pop- ulations, which turn out to be significant (see below). We tested a possible role of gene dosage in the origin- to-terminus bias of insertion frequency. The samples are in early log phase in LB medium at 37◦C when they are ex- posed to the transposon. There is therefore a higher num- ber of copies of the chromosome close to the origin than to the terminus. Estimating the dosage from the Cooper– Helmstetter model (34), and assuming an insertion rate pro- portional to the dosage, we computed the expected insertion bias, keeping into account the population age-structure (see SI text). Insertions are non-uniform and sparse and are more biased towards the replication origin than justified by gene dosage ) Figure 1B shows that the dosage estimated theoretically agrees very well with whole-genome sequencing of genome copy number, but is not sufficient to explain the stronger origin-terminus bias of the insertions. We also verified that this bias was not due to the insertions with top 10% cov- erage and to the insertions on ribosomal genes. The addi- tional bias may be due to additional factors such as DNA supercoiling or biased binding of nucleoid proteins and dif- ferences in nucleoid compaction (35–38). Additionally, the density of insertions shows a slight left-right asymmetry with respect to the origin, which is visible when the sliding average of insertions is compared with the prediction from dosage (Figure 1B). We verified that a model with time- dependent insertion rate, i.e. where the insertion rate r in- creases with time t, r(x, t), can fit the data, using insertion rate growing as a power law in time (see SI text). However, TraDIS Sequencing of FACS-sorted populations based on GFP expression shows the presence of transposon inser- tions at different chromosomal positions. Coverage of the insertions is uneven and sparse (Figure 1B, D, E). In ad- dition, there is a bias with respect to genome coordinate, with a higher insertion frequency close to the replication origin and a lower insertion frequency close to the terminus. The distributions of insertion frequencies in the parental populations of the P1-short and P1-long promoter inser- tions showed the same qualitative features as a function of genome coordinate (Supplementary Figure S2). Popu- lations derived from the high or medium GFP expression populations show a bias for insertions closer to the origin of replication (Supplementary Figure S2). We also noted that Nucleic Acids Research, 2020, Vol. 48, No. 5 2351 of the background sequences of the genome. A one-tailed Kolmogorov–Smirnov test (P-value <10−16) suggests that there is a significant difference between the two distribu- tions, with the sequences surrounding the insertions being richer in AT than the background ones (Supplementary Figure S2). The role of H-NS has been proposed to inhibit- ing insertions, in addition to repressing events of spurious transcription (12,18,41,42). Our results lead us to conclude that, in the tested conditions, at these fast growth rates, H- NS does not appear to inhibit physical events of transpo- son insertion efficiently. H-NS binding sites are enriched at insertions positions In order to better characterize the genomic positions of the insertions, we investigated the statistical tendencies for lo- calization of insertions using the gene lists from the NuST database (33). This database contains a large panel of pub- lished gene sets measuring several genomic properties such as binding of nucleoid-associated proteins, including several H-NS data sets (see Supplementary Table S1 for a detailed description of each data set). To score for significance, we compared the co-occurrences of insertions and genes with 5000 realizations of a shuffling null model (see Supplemen- tary Methods for details). Note that the null model sub- tracts the empirical sliding average of insertions, and not the dosage, thus the results are net of the overall enrich- ment around the origin. The analysis was applied to the population-sequencing data for the insertion sites in both the parental populations as well as in the ones that were sorted for gene-expression levels. g p This analysis, summarized in Figure 2A, shows that H- NS binding is the main property associated with any in- sertions (even before any sorting by gene expression is per- formed). The light blue circles in this figure refer to different genome-wide H-NS occupancy (ChIP-ChIP and ChIP-seq) data sets, obtained in different conditions (see Supplemen- tary Table S1). The dark blue circle refers to genes that are sensitive to H-NS knockout under perturbations that make supercoiling more positive (39). The crossed square refers to generic transcriptionally silenced extended protein oc- cupancy domains, which are largely made of H-NS bound regions. All these sets are correlated but not identical, and essentially contain different categories of H-NS bound re- gions. Insertions are non-uniform and sparse and are more biased towards the replication origin than justified by gene dosage The correlation analysis performed here does not allow us to conclude that AT-richness causes the enrichment of both insertions and H-NS binding. An- other causal chain is possible, but appears less likely, where H-NS binding facilitates insertions, thereby driving them towards AT-rich regions. We also note that the Tn5 trans- poson has been reported to be biased towards GC-rich re- gions (43), by a similar analysis than that performed in Sup- plementary Figure S2, but in different conditions, and in a different organism. This previous result makes the positive association that we find between insertions and AT-rich re- gions more intriguing. Additionally, Figure 2B shows that the significant enrichments of the different gene sets for in- sertions are consistent across the two promoters used here (P1-short and P1-long) used here (see also Supplementary Table S5), as expected from a lack of a role of the donor sequence on insertion bias. there is no empirical motivation to assume such cooperative behaviour in insertions that occur in different cells. Alterna- tively, since the exponent linking expected dosage and mea- sured insertions is close to three, one can also hypothesize a cooperative effect of technical or biological origin, but we could not produce a technical or biological explanation for such a simple cooperativity. H-NS binding is the sole over-represented signal in low- expression clonal populations From these, 90 clones where chosen to measure both the flu- orescence and the growth rate in a plate reader in different growth media. Finally, we compare the parental population with the ones sorted for gene expression levels (Figure 2C and Supple- mentary Table S6). The comparison of the insertion sites of the high and low expressing populations from the P1- long promoter strains shows that the low expressing popu- lation is found preferentially within H-NS binding regions, while the high expressing population is not. Indeed, the low- expression (RL) population maintains a similar association as the parental population with H-NS, and no other binding protein. Conversely, the RH (high expression) populations show no association with H-NS, and only maintain some enrichment with FNR (Figure 2C and Supplementary Table S6). Hence, despite the lack of an effect in inhibiting trans- poson insertion, H-NS does appear to regulate the level of gene expression of the inserted sequences. This analysis yields the following main results (see Sup- plementary Figures S3–S5) -There is agreement between a clone’s level of fluorescence and the average level of fluorescence of its original pop- ulation (Supplementary Figure S3), as measured by both flow cytometry and the fluorimeter. Specifically, the clones from the low-expressing populations have a significantly lower average level of fluorescence. -The magnitude of the difference between high- and low- expressing clones depends on the growth medium. The difference is greater in the faster growth medium (Supple- mentary Figure S4). -The very low-expression strains (Figure 1E) show low ex- pression regardless of the growth media, except for a few outliers showing out-of-trend expression (higher than ex- pected) in the poorer medium (Supplementary Figure S4). The last two observations are consistent with the known growth-rate dependence of H-NS activity and ribosomal operons transcription rate (16,45,46). This result shows that different regions of the genome can have different growth-rate dependent properties. Overall, these results point to a more complex role than is expected for H-NS in modulating genome accessibility and gene expression of recently acquired genes. Other global regulators are enriched at insertions positions 2352 Nucleic Acids Research, 2020, Vol. 48, No. 5 A B C Figure 2. Enrichment of insertions for H-NS and other global regulators. (A) Z-score of enrichment tests for different gene lists (see Supplementary Table S1 for a full legend). H-NS binding sites (from ChIP-seq and ChIp-ChIP data) and H-NS perturbations experiments (from (39)) are highly enriched (circles), indicating a strong positive association of insertions to H-NS binding regions starting from the parental colony. Other global nucleoid regulators (FNR, Fis, IHF, CRP, see legend), and a list of horizontal transfer genes (HT, see legend) also show positive association, lists of essential genes (filled squares) show strong negative enrichment. (B) Comparison of the two different promoter tested (with and without Fis and H-NS binding sites) shows a similar behaviour. (C) Comparison of parental and sorted populations (see Figure 1C–E) shows that H-NS association maintains a strong significance in the low-expression population, and loses significance in the high-expression population, where FNR sites remain highly enriched. C A A B Figure 2. Enrichment of insertions for H-NS and other global regulators. (A) Z-score of enrichment tests for different gene lists (see Supplementary Table S1 for a full legend). H-NS binding sites (from ChIP-seq and ChIp-ChIP data) and H-NS perturbations experiments (from (39)) are highly enriched (circles), indicating a strong positive association of insertions to H-NS binding regions starting from the parental colony. Other global nucleoid regulators (FNR, Fis, IHF, CRP, see legend), and a list of horizontal transfer genes (HT, see legend) also show positive association, lists of essential genes (filled squares) show strong negative enrichment. (B) Comparison of the two different promoter tested (with and without Fis and H-NS binding sites) shows a similar behaviour. (C) Comparison of parental and sorted populations (see Figure 1C–E) shows that H-NS association maintains a strong significance in the low-expression population, and loses significance in the high-expression population, where FNR sites remain highly enriched. Other global regulators are enriched at insertions positions We now proceed to discuss other gene sets that share en- richment for any insertions (visible in Fig 2A). Of notable significance are targets of global regulators Fis (which al- ters the nucleoid state to aid transcription in exponential growth) and FNR (which alters the distribution of RNA polymerase in response to oxygen starvation). This could be related to AT-richness bias of the binding site of these pro- teins or to high transcriptional activity (and thus accessibil- ity for insertions) of these genes (41,42). It is reasonable to expect that Fis targets are more active in LB medium. How- ever, we found that transcriptionally active RNAP binding regions (measured by ChIP-chip during rapid growth (44)) are under-represented for insertions, which suggests a nega- tive interaction between RNAP binding or transcriptional activity and insertion frequency (Supplementary Table S3). Finally, a milder but significant over-representation for in- sertions was found for CRP and IHF targets, genes that are sensitive to supercoiling perturbations in an H-NS knock- out background, and genes with trans-membrane domains (Supplementary Table S4).i Importantly, a strong enrichment is shared with puta- tive horizontal transfers detected from sequence properties (among which AT-richness (40), red diamond in Figure 2A). Indeed, the full list of insertion sites is enriched in H-NS tar- get genes regardless of the expression level (Supplementary Table S2). We also found an enrichment on H-NS binding sites in the surroundings (10 kb regions) of the insertions compared to random sites (Supplementary Figure S2). The positive local association of H-NS binding sites with insertion sites is in agreement with a common pref- erence for AT-rich regions. As previously mentioned, ge- nomic insertions generally have a reported bias for AT- rich regions (1,3,25), and AT-rich regions are also the pre- ferred binding targets for H-NS (9,10,20,23). We looked for a correlation between all insertions (regardless of their association with H-NS) and AT-rich regions. In order to do this, we compared the distribution of AT-bias in the sequences surrounding insertions with a random sample Conversely, essential genes (filled green squares in Fig- ure 2A) are the most under-represented set for insertions, as expected (Supplementary Table S3). The other under- represented gene sets for insertions comprise RNAP targets in rapid growth (crossed dark-green circle in Figure 2A), genes whose promoters are sensitive to supercoiling changes and highly transcribed occupancy domains, suggesting a negative correlation between insertion probability and tran- scriptional activity. Flow-cytometry analysis of clonal populations shows variable noise and gene-expression properties Each of the sorted populations was plated separately to yield individual isogenic (clonal) colonies. To gain further insight into these differences in gene expression, 658 indi- vidual clones were hand-picked from the different popula- tions and grown in 96-well plates to measure the average fluorescence and its standard deviation by flow cytometry. g p p p -Each clonal population (with a single insertion site, as ver- ified by sequencing) shows a distribution of fluorescence whose spread does not depend solely on the average ex- pression level (Figure 3A and Supplementary Figure S5). g p p p -Each clonal population (with a single insertion site, as ver- ified by sequencing) shows a distribution of fluorescence whose spread does not depend solely on the average ex- pression level (Figure 3A and Supplementary Figure S5). Nucleic Acids Research, 2020, Vol. 48, No. 5 2353 A B C ations than expected. All of these high-noise clones were very-low expression bacteria obtained from the RL (low- expression filtered) population. Hence, these clones likely explain the very-low expression sub-population (in green in Figure 1E) in the distribution of gene expression of the RL (low-expression filtered) population. When tested for their distribution in single-cell gene expression by flow cytom- etry, these low-expression high-noise clones did not show bimodal distributions. Rather, they showed disperse and skewed distributions, whose range overlaps with the expres- sion level of other low-expression clones (Supplementary Figure S5A, B). Supplementary Figure S5C–G recapitu- lates the noise properties of all the picked clones from differ- ent rounds of selection. It is clear that high-noise clones be- come more frequent in successive sorting rounds where low or very-low expression cells are selected. The following two sections deal with a more detailed characterization of the properties of these low-expression, high-noise insertions. A A A B C D Figure 3. Noisy promoters emerge from transcriptional interference within the rrlE ribosomal operon. (A) Characterization of noise of 658 clonal populations from flow cytometry. Left: Standard deviation vs mean GFP expression. Right: Noise (CV2) versus mean GFP expression. (B) All the tested ‘noisy’ clones by sequencing typically show association with the rrlE ribosomal RNA operon (see Supplementary File SF1). (C) Characteriza- tion of promoter noise by microcolony time-lapse growth assay. Micro- colonies were grown, imaged, segmented and tracked for three generations. (D) An example of the resulting lineage-specific gene expression data is shown on the left. Flow-cytometry analysis of clonal populations shows variable noise and gene-expression properties The plot on the right quantifies the lineage divergence of gene expression (time average of the absolute gene expression difference) of the different clones, comparing it with the CV measured from flow cytome- try (see text). High-noise clones show a large lineage divergence - indicating a possible switching behavior. B B C Noisy sites are associated with the insertions within ribosomal operons Noisy promoters emerge from transcriptional interference within the rrlE ribosomal operon. (A) Characterization of noise of 658 clonal populations from flow cytometry. Left: Standard deviation vs mean GFP expression. Right: Noise (CV2) versus mean GFP expression. (B) All the tested ‘noisy’ clones by sequencing typically show association with the rrlE ribosomal RNA operon (see Supplementary File SF1). (C) Characteriza- tion of promoter noise by microcolony time-lapse growth assay. Micro- colonies were grown, imaged, segmented and tracked for three generations. (D) An example of the resulting lineage-specific gene expression data is shown on the left. The plot on the right quantifies the lineage divergence of gene expression (time average of the absolute gene expression difference) of the different clones, comparing it with the CV measured from flow cytome- try (see text). High-noise clones show a large lineage divergence - indicating a possible switching behavior. Insertion within ribosomal operons is tolerated because E. coli has seven copies of ribosomal operons. To discover the orientation of the cloned sequences with respect to the rrlE promoter position, we used the blast results of de novo assembled contigs with the flanks of cloned insert sequence and we compared them with the reference genome. Most of the insertions showed an opposite orientation of the in- serted promoter with respect to the rrlE promoter sequence. Separately, we checked the orientation of the inserted GFP cassette in few selected Illumina samples for which the blast results showed reasonable overlap with the genome locus. This confirmed the opposite orientation of GFP with re- Noisy sites are associated with the insertions within ribosomal operons To characterize the set of clonal colonies from different pop- ulations, we performed whole-genome sequencing on 90 se- lected clones. The locations of all the insertions of these clones are listed in Supplementary File SF1. Thirty two clones out of the 90 selected samples show the presence of insertion within a rRNA operon from whole-genome sequencing data. These clones were very-low expressing clones derived from RL filtered populations. For exam- ple, in the clones from the RLR1V1L sorted populations we tested a total of 16 insertions, of which nine were in rRNA regions (Supplementary File SF1). In order to verify these short-read assignments of insertions sites, long-read nanopore sequencing was used on five of these 32 clonal samples. This analysis confirmed the presence of the trans- poson insertion within the 23S ribosomal RNA (rrlE) of the rrnE operon. To recapitulate these results, Supplementary Figure S6A shows the association of clones with high-noise promoters with rrlE insertions. Some of the non-rrlE high- noise clones revealed the presence of multiple (up to three) insertion sites, which can explain the large variance in gene expression of these clones (see column D of Supplemen- tary File SF1 for a list of the multiple insertions and their coordinates). Indeed, we found that normalization of gene expression by the copy number of the promoter removed these outliers (Supplementary Figure S6B, C). Hence, we believe that the true high-noise phenotype should almost exclusively be associated with rrlE insertions. D D Figure 3. Noisy promoters emerge from transcriptional interference within the rrlE ribosomal operon. (A) Characterization of noise of 658 clonal populations from flow cytometry. Left: Standard deviation vs mean GFP expression. Right: Noise (CV2) versus mean GFP expression. (B) All the tested ‘noisy’ clones by sequencing typically show association with the rrlE ribosomal RNA operon (see Supplementary File SF1). (C) Characteriza- tion of promoter noise by microcolony time-lapse growth assay. Micro- colonies were grown, imaged, segmented and tracked for three generations. (D) An example of the resulting lineage-specific gene expression data is shown on the left. The plot on the right quantifies the lineage divergence of gene expression (time average of the absolute gene expression difference) of the different clones, comparing it with the CV measured from flow cytome- try (see text). High-noise clones show a large lineage divergence - indicating a possible switching behavior. Figure 3. Noisy promoters may perform switching An important technical point to address is the role of kanamycin selection in these experiments. If the donor se- quence including the kanR cassette is inserted in a locus where it is completely silenced by H-NS, one might not be able to see the insertion. However, our data show that se- lection itself does not preclude the identification of an in- sertion site, since insertions are not excluded from H-NS occupancy-rich regions, as it would be expected if complete silencing of KanR expression had taken place. We do ob- serve that promoters inserted in H-NS rich regions are on average expressed less than the others. We also note that we carried out the transposon reaction in mid exponential phase cells growing in a rich medium, LB. In these condi- tions the concentration of H-NS is lower due to a high di- lution rate (16). However, at the faster growth rates, H-NS is known to still play a role in the repression of ribosomal promoters by binding to higher affinity sites (37,53). Our results suggest that there is probably not enough protein to also cover the lower affinity (nonspecific) binding to AT- rich regions, in order to inhibit transposon insertion (11). This is in contrast to a previous study in Vibrio cholerae that has shown that only in the absence of H-NS there was a higher probability of insertion in AT-rich regions of the genome (12). These results lead us to further suggest that the role of H-NS in regulating the probability of genomic insertion of horizontally acquired genes may depend on the growth conditions and on the specific strain. The previous analyses strongly indicate an association of the high-noise insertions with an interference of the inser- tions with ribosomal operons. To gain more insight on the temporal dynamics of these high-noise inserted promot- ers, we measured gene expression noise in time-lapse mi- croscopy data on growing microcolonies (Figure 3C, D). We compared the change in GFP gene expression over time of single cells from clones carrying noisy and non-noisy pro- moters for three to four generations and quantified the dif- ferences between gene expression in different lineages. The divergence between lineages was quantified as the time av- erage of the absolute value of the gene expression difference between sister cells. Bacteria were grown on an agar pad to form a micro- colony. A set of ‘noisy’ insertion sites We found that in all the clones the standard deviation of gene expression is proportional to the mean, meaning that most of the noise shown by the promoter is extrinsic, regard- less of expression level and insertion location (Figure 3A). This is expected from the high level of expression of the rrnBP1 promoter (47). However, a subset of clones show a higher level of noise where the scaling of the standard deviation with the mean follows a steeper slope. In other words, these clones show much larger gene expression vari- 54 Nucleic Acids Research, 2020, Vol. 48, No. 5 2354 second, insertion probability is higher in AT-rich regions of the genome. This may seem surprising, because these are regions that are preferentially bound by the H-NS protein, which has been proposed to act as a barrier to horizon- tal gene transfer (3,6,8,10,28,48). However, physical com- ponents such as differences in DNA supercoiling (35,49,50) and the biophysical properties of AT-rich DNA (lower melt- ing barriers, different stacking energy, etc.) may play a role in establishing these biases. In particular, the measured origin-to-terminus gradients of supercoling (51) and gy- rase binding (52) are compatible with the hypothesis that the probability of insertion is increased in regions with higher negative supercoiling, which may explain the origin- to-terminus positive bias for insertions (not justified by origin-to-terminus differences in gene dosage). Note that this insertion bias (only transposon type) should not be con- fused to the transcriptional consequences of supercoling on the donor promoter. A different promoter sequence is not expected to affect the insertion probability. On the other hand, changes in gene expression levels depend on promoter sensitivity to supercoiling and local context. spect to the rrlE promoter sequence in most samples. Strong promoter competition may explain the very low levels of GFP transcript production by which these clones were iso- lated, as well as the high variability. A set of ‘noisy’ insertion sites , g y We found that the trends of gene expression with growth rate were consistent with the hypothesis of competition with a strong promoter: while most of the other clones increase their expression with growth rate (in agreement with the known regulation of the rrnBP1 ribosomal promoter) the noisy clones decrease in expression with increasing growth rate, in agreement with the idea that their expression is re- pressed by an interference with transcription of the increas- ingly transcribed ribosomal operon (Supplementary Fig- ures S3 and S4). Additionally, the mild reduction in growth rate for insertions giving different mean GFP expression suggests that the cost associated to GFP expression and the possible interference with the ribosomal operon are not dominant for these insertions (Supplementary Figures S3 and S4). Downloaded from https://academic.oup.com/nar/article-abstract/48/5/2348/5710780 by guest on 28 April 2020 Noisy promoters may perform switching The time-lapse data in the formation of the micro- colony was segmented to obtain the change in the average cell fluorescence as a function of time Figure 3C. An ex- ample of gene expression of two lineages, one from a noisy clone, as measured by flow cytometry, and one from a con- trol clone (where the cassette was inserted specifically be- tween two converging genes, AidB and yjfN) is shown in the left panel of Figure 3D. The right panel of Figure 3D quantifies the divergence of gene expression along lineages for different clonal microcolonies, corresponding to clones where the promoter is inserted in different positions. Fig- ure 3D shows that in microcolonies from high-noise clones different lineages emanating from the same single cell tend to diverge more in gene expression as time progresses than in the control or low-noise clones. This result points to the possible presence of switching behaviour in the high-noise clones. Our results also indicate that the dilution of H-NS in rapidly growing cells does not prevent the establishment of a low-expressing population biased for the insertions asso- ciated with H-NS binding. Hence, they are consistent with the role of H-NS as a silencer of newly acquired genes. In- deed, we observe that once the full length rrnBP1 promoter cassette is inserted in the genome, its level of expression is lower if it is found near H-NS rich regions. This cassette includes a higher affinity H-NS binding site within the full length rrnBP1 promoter and an AT-rich gfpmut2 gene se- quence stabilizing the formation of H-NS dependent re- pressing complex. The shorter version of the promoter (P1- short), lacking the high affinity H-NS binding site, does not REFERENCES 1. Touchon,M., Bobay,L.M. and Rocha,E.P.C. (2014) The chromosomal accommodation and domestication of mobile genetic elements. Curr. Opin. Microbiol., 22, 22–29. 1. Touchon,M., Bobay,L.M. and Rocha,E.P.C. (2014) The chromosomal accommodation and domestication of mobile genetic elements. Curr. Opin. Microbiol., 22, 22–29. 2. Lawrence,J.G. and Ochman,H. (1998) Molecular archaeology of the Escherichia coli genome. Proc. Natl. Acad. Sci. U.S.A., 95, 9413–9417. 2. Lawrence,J.G. and Ochman,H. (1998) Molecular archaeology of the Escherichia coli genome. Proc. Natl. Acad. Sci. U.S.A., 95, 9413–9417. Altogether, the main novelty of our study is the support for a high (initial) tolerance for insertions with a wide range of expression levels, which challenges the standard view that H-NS not only silences horizontally acquired genes (be- cause they are more AT-rich) but also inhibits insertion in AT-rich regions of the genome. We find that in the tested growth conditions, H-NS does not inhibit insertions in AT- rich regions. However, it can still decrease expression after the insertion has taken place. These findings support the fol- lowing evolutionary scenario. When a novel gene enters the genome, it is more likely found in a region that is controlled by H-NS, for reasons that most likely have nothing to do with fitness, but have to do with the physico-chemical prop- erties of the DNA in AT-rich regions. However, the wide range of expression levels that we find show that the gene is not necessarily immediately silenced. Rather, the different insertion positions allow it to sample a wide range of ex- pression levels (including silencing), at (initially) equal pro- 3. Oliveira,P.H., Touchon,M., Cury,J. and Rocha,E.P.C. (2017) The chromosomal organization of horizontal gene transfer in bacteria. Nat. Commun., 8, 841. 4. Ochman,H., Lawrence,J.G. and Groisman,E.A. (2000) Lateral gene transfer and the nature of bacterial innovation. Nature, 405, 299–304. 5. Lercher,M.J. and P´al,C. (2008) Integration of horizontally transferred genes into regulatory interaction networks takes many million years. Mol. Biol. Evol., 25, 559–567. 6. Park,C. and Zhang,J. (2012) High expression hampers horizontal gene transfer. Genome Biol. Evol., 4, 523–532. 7. Lucchini,S., Rowley,G., Goldberg,M.D., Hurd,D., Harrison,M. and Hinton,J.C. (2006) H-NS mediates the silencing of laterally acquired genes in bacteria. PLoS Pathog., 2, e81. 8. Navarre,W.W., McClelland,M., Libby,S.J. and Fang,F.C. (2007) Silencing of xenogeneic DNA by H-NS facilitation of lateral gene transfer in bacteria by a defense system that recognizes foreign DNA. Gene. Dev., 21, 1456–1471. 9. Kahramanoglou,C., Seshasayee,A.S.N., Prieto,A.I., Ibberson,D., Schmidt,S., Zimmermann,J., Benes,V., Fraser,G.M. and Luscombe,N.M. DISCUSSION Our results directly show that the probability of DNA in- sertion in the E. coli genome by a transposon is biased, be- fore any long-term selection may act, other than that re- lated to overnight growth. First, there is a stronger origin- to-terminus bias than explained by gene dosage imbalance, Nucleic Acids Research, 2020, Vol. 48, No. 5 2355 appear show a stable sub-population of very-low fluores- cence clones (data not shown), showing that the high affin- ity site of the promoter is important in nucleating the re- pressing oligomeric structure, a question that we are still exploring. In summary, the level of expression of the clones can thus be very heterogeneous, depending on local proper- ties of the site of insertion and the sequence of the fragment. moter strength, while interacting from the start with the cell’s housekeeping physiology. We believe that this inherent bet-hedging and exploratory stage may be a key ingredient of genome plasticity, and is underestimated in our current narrative of the process of horizontal transfer, which is cen- tered on the average outcome, and establishes a strict time hierarchy between stages where an exogenous gene is first silenced and then reactivated. The cell-to-cell variability of gene expression within a given isogenic clone can vary significantly, but it typically scales with the mean level of expression as expected from ex- trinsic noise (45,47,54). This is expected from the promoter used here, rrnBP1, which is a strong promoter, resulting in a high level of expression. The change in the CV as a func- tion of mean expression therefore remains for the most part relatively flat, corresponding to the extrinsic noise regime. DATA AVAILABILITY The sequencing data are available on the Sequence Read Archive (SRA) (https://www.ncbi.nlm.nih.gov/sra) under BIoproject accession IDs numbers PRJNA575574 (WGS Data) and PRJNA575567 (TRADIS data). The (processed) cloned insert FASTA file is available as Supplementary File SF2. Downloaded from https://academic.oup.com/nar/article-abstract/48/5/2348/5710780 b yl , p g g However, in some of the very low expression clones the noise varies in a way that is not expected from the known pattern of gene expression noise correlations that have been described previously (45,47). We therefore characterized those clones that have a higher level of gene expression noise and found that in these cases the insertion has taken place within a ribosomal operon. Escherichia coli has seven copies of ribosomal operons, therefore insertion inside one of them does not have a high cost and is not selected against, at least in the short term of this experiment. This results in inter- ference between two transcription processes driven by very similar promoters, of similar strength. Furthermore, the ini- tiation frequency of ribosomal promoters is high enough at fast growth rates that most of the time the operon sequence can be assumed to be covered by transcribing RNA poly- merase ‘trains’ (55). This creates a block for RNA poly- merase to bind to the promoter that is found within the operon, creating a stable ‘off state’. However, from time to time RNA polymerase manages to bind to the newly in- serted promoter, perhaps after the DNA replication forks have erased the memory from the competing process, start- ing its own ‘train’ of GFP production. Such transcriptional interference is a well-known phenomenon (56). Our result on promoter noise also suggests that in the tightly packed bacterial genomes, transcription interference with newly in- serted genes might be a natural source of innovations in terms of gene expression noise on evolutionary time scales, as previously speculated for eukaryotes (57). SUPPLEMENTARY DATA Supplementary Data are available at NAR Online. Supplementary Data are available at NAR Online. FUNDING IFCPAR/CEFIPRA (Indo-French Centre for the Promo- tion of Advanced Research) [5103-3]; DBT/Wellcome Trust India Alliance Intermediate Fellowship [IA/I/16/2/502711 to A.S.N.S.]. Funding for open access charge: IFOM insti- tutional funds.l Conflict of interest statement. None declared. Conflict of interest statement. None declared. ACKNOWLEDGEMENTS We are grateful to Vittore Scolari, Qing Zhang, Awadesh Pandit and Terence Christie for help and useful discussions. Illumina and Nanopore sequencing were performed at the NGS facility at NCBS. up.com/nar/article-abstract/48/5/2348/5710780 by guest on 28 April 2020 REFERENCES (2015) The genome-scale interplay amongst xenogene silencing, stress response and chromosome architecture in Escherichia coli. Nucleic Acids Res., 43, 295–308. 43. Green,B., Bouchier,C., Fairhead,C., Craig,N.L. and Cormack,B.P. (2012) Insertion site preference of Mu, Tn5, and Tn7 transposons. Mobile DNA, 3, 3. 44. Grainger,D.C., Hurd,D., Harrison,M., Holdstock,J. and Busby,S.J.W. (2005) Studies of the distribution of Escherichia coli cAMP-receptor protein and RNA polymerase along the E. coli chromosome. Proc. Natl. Acad. Sci. U.S.A., 102, 17693–17698. 18. Singh,S.S., Singh,N., Bonocora,R.P., Fitzgerald,D.M., Wade,J.T. and Grainger,D.C. (2014) Widespread suppression of intragenic transcription initiation by H-NS. Gene. Dev., 28, 214–219. 19. Bryant,J.A., Sellars,L.E., Busby,S.J.W. and Lee,D.J. (2014) Chromosome position effects on gene expression in Escherichia coli K-12. Nucleic Acids Res., 42, 11383–11392. 45. Zaslaver,A., Kaplan,S., Bren,A., Jinich,A., Mayo,A., Dekel,E., Alon,U. and Itzkovitz,S. (2009) Invariant distribution of promoter activities in Escherichia coli. PLoS Comput. Biol., 5, e1000545. 20. Dorman,C.J. (2009) Nucleoid-associated proteins and bacterial physiology. Adv. Appl. Microbiol., 67, 47–64. 46. Zhang,Q., Brambilla,E., Li,R., Shi,H., Cosentino Lagomarsino,M. and Sclavi,B. (2019) A decrease in transcription capacity limits growth rate upon translation inhibition. bioRxiv doi: https://doi.org/10.1101/599183, 5 April 2019, preprint: not peer reviewed. 21. Esp´eli,O. and Boccard,F. (2006) Organization of the Escherichia coli chromosome into macrodomains and its possible functional implications. J Struct. Biol., 156, 304–310. p 22. Esnault,E., Valens,M., Esp´eli,O. and Boccard,F. (2007) Chromosome structuring limits genome plasticity in Escherichia coli. PLoS Genet., 3, e226. 47. Taniguchi,Y., Choi,P.J., Li,G.W., Chen,H., Babu,M., Hearn,J., Emili,A. and Xie,X.S. (2010) Quantifying E. coli proteome and transcriptome with single-molecule sensitivity in single cells. Science 329, 533–538. 23. Zarei,M., Sclavi,B. and Cosentino Lagomarsino,M. (2013) Gene silencing and large-scale domain structure of the E. coli genome. Mol. Biosyst., 9, 758–767. 48. Touchon,M. and Rocha,E.P.C. (2007) Causes of insertion sequences abundance in prokaryotic genomes. Mol. Biol. Evol., 24, 969–981. 24. Dame,R.T., Kalmykowa,O.J. and Grainger,D.C. (2011) Chromosomal macrodomains and associated proteins: implications for DNA organization and replication in gram negative bacteria. PLoS Genet., 7, e1002123. 49. Dorman,C.J. (2013) Co-operative roles for DNA supercoiling and nucleoid-associated proteins in the regulation of bacterial transcription. Biochem. Soc. Trans., 41, 542–547. 25. Dorman,C.J. (2013) Genome architecture and global gene regulation in bacteria: making progress towards a unified model? Nat. Rev. Microbiol., 11, 349–355. 50. Ouafa,Z.A., Reverchon,S., Lautier,T., Muskhelishvili,G. and Nasser,W. (2012) The nucleoid-associated proteins H-NS and FIS modulate the DNA supercoiling response of the pel genes, the major virulence factors in the plant pathogen bacterium Dickeya dadantii. REFERENCES (2011) Direct and indirect effects of H-NS and Fis 2356 Nucleic Acids Research, 2020, Vol. 48, No. 5 on global gene expression control in Escherichia coli. Nucleic Acids Res., 39, 2073–2091. on global gene expression control in Escherichia coli. Nucleic Acids Res., 39, 2073–2091. et al. (2007) High-affinity DNA binding sites for H-NS provide a molecular basis for selective silencing within proteobacterial genomes. Nucleic Acids Res., 35, 6330–6337. et al. (2007) High-affinity DNA binding sites for H-NS provide a molecular basis for selective silencing within proteobacterial genomes. Nucleic Acids Res., 35, 6330–6337. , , 10. Dorman,C.J. (2007) H-NS, the genome sentinel. Nat. Rev. Microbiol., 5, 157–161. genomes. Nucleic Acids Res., 35, 6330–6337. 37. Berger,M., Farcas,A., Geertz,M., Zhelyazkova,P., Brix,K., Travers,A. and Muskhelishvili,G. (2010) Coordination of genomic structure and transcription by the main bacterial nucleoid-associated protein HU. EMBO Rep., 11, 59–64. 11. Singh,K., Milstein,J.N. and Navarre,W.W. (2016) Xenogeneic silencing and its impact on bacterial genomes. Annu. Rev. Microbiol., 70, 199–213. 12. Kimura,S., Hubbard,T.P., Davis,B.M. and Waldor,M.K. (2016) The nucleoid binding protein H-NS biases genome-wide transposon insertion landscapes. mBio, 7, 1–6. 38. Lioy,V.S., Cournac,A., Marbouty,M., Duigou,S., Mozziconacci,J., Espli,O., Boccard,F. and Koszul,R. (2018) Multiscale structuring of the E. coli chromosome by nucleoid-associated and condensin proteins. Cell, 172, 771–783. 13. Sorek,R., Zhu,Y., Creevey,C.J., Francino,M.P., Bork,P. and Rubin,E.M. (2007) Genome-wide experimental determination of barriers to horizontal gene transfer. Science, 318, 1449–1452. 39. Blot,N., Mavathur,R., Geertz,M., Travers,A. and Muskhelishvili,G. (2006) Homeostatic regulation of supercoiling sensitivity coordinates transcription of the bacterial genome. EMBO Rep., 7, 710–715. 14. Dame,R.T., Noom,M.C. and Wuite,G.J.L. (2006) Bacterial chromatin organization by H-NS protein unravelled using dual DNA manipulation. Nature, 444, 387–390. 40. Garcia-Vallve,S., Guzman,E., Montero,M.A. and Romeu,A. (2003) HGT-DB: a database of putative horizontally transferred genes in prokaryotic complete genomes. Nucleic Acids Res., 31, 187–189. 15. Liu,Y., Chen,H., Kenney,L.J. and Yan,J. (2010) A divalent switch drives H-NS/DNA-binding conformations between stiffening and bridging modes. Gene. Dev., 24, 339–344. 41. Browning,D.F., Grainger,D.C. and Busby,S.J. (2010) Effects of nucleoid-associated proteins on bacterial chromosome structure and gene expression. Curr. Opin. Microbiol., 13, 773–780. 16. Brambilla,E. and Sclavi,B. (2015) Gene regulation by H-NS as a function of growth conditions depends on chromosomal position in Escherichia coli. G3, 5, 605–614. 42. Grainger,D.C., Goldberg,M.D., Lee,D.J. and Busby,S.J. (2008) Selective repression by Fis and H-NS at the Escherichia coli dps promoter. Mol. Microbiol., 68, 1366–1377. 17. Srinivasan,R., Scolari,V.F., Cosentino Lagomarsino,M. and Seshasayee,A.S.N. REFERENCES Nucleic Acids Res., 40, 4306–4319. 26. Touchon,M., Hoede,C., Tenaillon,O., Barbe,V., Baeriswyl,S., Bidet,P., Bingen,E., Bonacorsi,S., Bouchier,C., Bouvet,O. et al. (2009) Organised genome dynamics in the Escherichia coli species results in highly diverse adaptive paths. PLoS Genet., 5, e1000344. 51. Lal,A., Dhar,A., Trostel,A., Kouzine,F., Seshasayee,A.S.N. and Adhya,S. (2016) Genome scale patterns of supercoiling in a bacterial chromosome. Nat. Commun., 7, 11055. 27. Junier,I., Martin,O. and K´ep`es,F. (2010) Spatial and Topological Organization of DNA Chains Induced by Gene Co-localization. PLoS Comput. Biol., 6, e1000678. 52. Jeong,K.S., Ahn,J. and Khodursky,A.B. (2004) Spatial patterns of transcriptional activity in the chromosome of Escherichia coli. Genome Biol., 5, R86. p 28. Dilthey,A. and Lercher,M.J. (2015) Horizontally transferred genes cluster spatially and metabolically. Biol. Direct., 10, 72. 53. Sobetzko,P., Travers,A. and Muskhelishvili,G. (2012) Gene order and chromosome dynamics coordinate spatiotemporal gene expression during the bacterial growth cycle. Proc. Natl. Acad. Sci. U.S.A., 109, E42–E50. p y y 29. Bobay,L.M. and Ochman,H. (2017) The evolution of bacterial genome architecture. Front. Genet., 8, 72. 30. Dennis,P.P., Ehrenberg,M. and Bremer,H. (2004) Control of rRNA synthesis in Escherichia coli: a systems biology approach. Microbiol. Mol. Biol. R.: MMBR, 68, 639–668. 54. Swain,P.S., Elowitz,M.B. and Siggia,E.D. (2002) Intrinsic and extrinsic contributions to stochasticity in gene expression. Proc. Natl. Acad. Sci. U.S.A., 99, 12795–12800.i 33. Scolari,V.F., Zarei,M., Osella,M. and Cosentino Lagomarsino,M. (2012) NuST: analysis of the interplay between nucleoid organization and gene expression. Bioinformatics, 28, 1643–1644. 55. Klumpp,S. and Hwa,T. (2008) Stochasticity and traffic jams in the transcription of ribosomal RNA: Intriguing role of termination and antitermination. Proc. Natl. Acad. Sci. U.S.A., 105, 18159–18164. 34. Cooper,S. and Helmstetter,C.E. (1968) Chromosome replication and the division cycle of Escherichia coli B/r. J. Mol. Biol., 31, 519–540. 56. Shearwin,K.E., Callen,B.P. and Egan,J.B. (2005) Transcriptional interference–a crash course. Trends Genet.: TIG, 21, 339–345. 35. Travers,A. and Muskhelishvili,G. (2005) DNA supercoiling - a global transcriptional regulator for enterobacterial growth? Nat. Rev. Microbiol., 3, 157–169. 57. Wang,G.Z., Lercher,M.J. and Hurst,L.D. (2011) Transcriptional coupling of neighboring genes and gene expression noise: evidence that gene orientation and noncoding transcripts are modulators of noise. Genome Biol. Evol., 3, 320–331. 36. Lang,B., Blot,N., Bouffartigues,E., Buckle,M., Geertz,M., Gualerzi,C.O., Mavathur,R., Muskhelishvili,G., Pon,C.L., Rimsky,S.
https://openalex.org/W3087161209	https://link.springer.com/content/pdf/10.1007/s11406-020-00268-5.pdf	English	null	Modal Dispositionalism and the (T) Axiom	Philosophia	2,020	cc-by	7,761	* Matthew James Collier Matthew.collier@oriel.ox.ac.uk https://doi.org/10.1007/s11406-020-00268-5 Philosophia (2021) 49:977–988 https://doi.org/10.1007/s11406-020-00268-5 Philosophia (2021) 49:977–988 Abstract Yates has recently argued that modal dispositionalism invalidates the (T) axiom. Both Yates and Allen have advanced responses to the objection: Yates’s response proposes installing truth into the possibility biconditional, and Allen’s response requires that all properties be construed as being essentially dispositional. I argue that supporters of Borghini and Williams’s modal dispositionalist theory cannot accept these responses, given critical tenets of their theory. But, since these responses to the objection are the most plausible in the literature, I conclude that the threat that Borghini and Williams’s modal dispositionalist theory invalidates the (T) axiom still looms large. Keywords Modal Dispositionalism . Modal logic . Dispositions . Categorical and dispositional properties Modal Dispositionalism and the (T) Axiom Matthew James Collier1 Received: 5 May 2020 /Revised: 28 August 2020 /Accepted: 7 September 2020 # The Author(s) 2020 / Published online: 21 2020 September 1 Oriel College Oriel Square, Oxford OX1 4EW, UK or the most dominant possible worlds accounts, see Lewis (1986), Plantinga (1974) and Adams (1974). 2 There are two reasons for my choosing Borghini and Williams’s MD: The first reason is that I regard it as one of the most easily understood dispositionalist accounts, and the second reason is that, by focussing on the resources wielded by MD, I can rebut various counter-objections to the chief objection raised in this paper: MD’s invalidation of (T). For other MD-like accounts, see: Vetter (2015), Pruss (2011), Contessa (2010), Jacobs (2010), Bird (2007), Mumford (2004), Molnar (2003), and Ellis (2001). 3 It is worth noting at the outset that neither Yates (2015) nor Allen (2017) are interested exclusively in MD, but rather in a more general modal dispositionalism. Indeed, Yates accepts that his solution to his objection entails a weak version of dispositionalism, and Allen is concerned with the extent to which modal dispositionalism requires a development of an implausible or an overly-complex ontology to maintain formal and material adequacy. However, I speak as if their responses specifically target MD, not only for simplicity’s sake, but also to highlight that their responses to Yates’s objection fail when applied to MD; thus, by treating their responses as specifically targeted at MD, I cast light upon the tenability of MD. 4 The stimulus conditions are generally considered to consist solely of other dispositions. 5 Here I follow Borghini and Williams’s (2008) usage of states of affairs: that is, objects’ having of properties are states of affairs, even if the relevant states of affairs fail to obtain. Thus, both manifested and unmanifested dispositions are construed in terms of states of affairs. 6 Now, the term ‘disposition’ has many alternatives: for instance, ‘power’, ‘potency’, ‘potentiality’, ‘capabil- ity’, etc.… I just use ‘disposition’ however. 1 Introduction Modal dispositionalism (MD) semantically reduces, and ontologically explicates, pos- sibility and necessity in terms of dispositionality. For MD, the dispositions of actual objects are taken as the truth-makers of modal claims.1 So, (roughly) some object’s having the disposition to bring it about that p makes-true the modal claim <possibly: p >. Borghini and Williams’s (2008) version of MD (‘MD’ shall refer exclusively to the 1For the most dominant possible worlds accounts, see Lewis (1986), Plantinga (1974) and Adams (1974). * Matthew James Collier Matthew.collier@oriel.ox.ac.uk * Matthew James Collier Matthew.collier@oriel.ox.ac.uk 1 Oriel College Oriel Square, Oxford OX1 4EW, UK Philosophia (2021) 49:977–988 978 theory) is the chief dispositionalist theory of concern here – however some of the points I raise generalize to other modal dispositionalist theories.2 The roadmap for this paper is the following: Section 2 sets up MD’s ontology by explaining what dispositions and dispositional properties are, and provides an exposition of MD; and, section 3 examines an instance of MD’s formal deficiency – specifically, I examine Yates’s (2015) objection that, in the case of certain actualized necessities, modal dispositionalism’s biconditionals for possibility and necessity invalidate the (T)-axiom, and I apply this objection to MD. To Yates’s objection, I assess the two most plausible responses – one of which is owed to Yates (which is also provided in his (2015)), and the other to Allen (2017). I argue that both responses fail to save MD: crucially, they ignore critical aspects of MD.3 Indeed, Yates’s response requires one to ignore MD’s universal dispositionalist aim insofar as his response threatens to render dispositions redundant in MD’s explication of modality, and Allen’s response requires one either to ignore MD’s naturalism or to accept that MD is – at best – ontologically bizarre or – at worst – absurd. I conclude that neither of the two most plausible responses to Yates’s objection are successful. Thus, MD should be rejected – whilst some modal logical axioms may be negotiable, some are not; certainly (T) is not. 2 There are two reasons for my choosing Borghini and Williams’s MD: The first reason is that I regard it as one of the most easily understood dispositionalist accounts, and the second reason is that, by focussing on the resources wielded by MD, I can rebut various counter-objections to the chief objection raised in this paper: MD’s invalidation of (T). For other MD-like accounts, see: Vetter (2015), Pruss (2011), Contessa (2010), Jacobs (2010), Bird (2007), Mumford (2004), Molnar (2003), and Ellis (2001). 2 Dispositions, Dispositional Properties and MD Let us Get a Grip on MD’s Ontology. Dispositions, when under suitable stimulus conditions,4 manifest, and thereby produce or bring about certain states of affairs.5 That is to say, dispositions are considered abilities of objects to produce certain states of affairs, and so, standardly, dispositionality is understood in modal terms.6 Furthermore, (at least some) dispositions can manifest under multiple different stimulus conditions: for instance, a teacup will shatter if it is thrown against a hard surface – but, a teacup will also shatter if it is knocked with a mallet. Moreover, it is conventional for dispositions to be characterized by their manifestation-types – so, objects have dispo- sitions for certain manifestations: for instance, a teacup’s fragile disposition is characterised by its manifestation being such that, under the appropriate stimulus Philosophia (2021) 49:977–988 979 conditions, the teacup shatters. And so, a teacup’s being fragile constitutes a disposition for shattering. Herein, I follow the above conventions. conditions, the teacup shatters. And so, a teacup’s being fragile constitutes a disposition for shattering. Herein, I follow the above conventions. The view that objects have dispositions through having certain properties is called dispositional realism.7 Central to dispositional realism is the notion that objects may possess dispositions that may never manifest. Thus, objects do not have certain dispositions because of the properties they had or will have, but instead because of the properties that they currently have. Importantly, there are two factions amongst the realists: the reductivists and the non-reductivists (/essentialists). Reductivists contend: the properties that ground dispositions are essentially non-dispositional (or categorical) – viz., non-dispositional properties ground dispositions and their manifestations.8 Non- reductivists, however, contend: essentially, irreducibly dispositional properties ground dispositions and their manifestations.9 So, essentially dispositional properties essential- ly dispose their object-bearers to behave in particular ways. Now, Borghini and Williams (2008) are in the non-reductivist camp, but are agnostic over whether all properties are essentially dispositional; certainly, they do not explicitly endorse dispositional monism.10 For MD, it is sufficient for an object to have dispositions if it instantiates certain dispositional properties.11 But, Borghini and Williams make a distinction between dispositions and the dispositional properties that those dispositions are grounded by. 7 Here I follow Borghini and Williams (2008) in following Prior’s (1985) characterisation of realism. 8 g p p p g ( ) 9 For defences of reductivism see: Lewis (1997), Armstrong (1997), Jackson (1995), Prior (1985), Jackso et al. (1982) and Mackie (1977). For defences of non-reductivism see: Bird (2007), Mumford (2004), Moln (2003), Heil (2003), Ellis (2002) and Mellor (2000). 10 8 I return to categorical properties when exploring MD’s invalidation of (T). 9 11 It is worth noting that MD is neutral on the ontological status of dispositional properties: for instance, o MD, properties can be immanent universals or tropes. However, properties cannot be (as per Lewis (1986 sets of individuals that reside in possible worlds. 7 Here I follow Borghini and Williams (2008) in following Prior’s (1985) characterisation of realism. 8 I return to categorical properties when exploring MD’s invalidation of (T). 9 For defences of reductivism see: Lewis (1997), Armstrong (1997), Jackson (1995), Prior (1985), Jackson et al. (1982) and Mackie (1977). For defences of non-reductivism see: Bird (2007), Mumford (2004), Molnar (2003), Heil (2003), Ellis (2002) and Mellor (2000). 10 Later I return to dispositional monism. 11 It is worth noting that MD is neutral on the ontological status of dispositional properties: for instance, on MD, properties can be immanent universals or tropes. However, properties cannot be (as per Lewis (1986)) sets of individuals that reside in possible worlds. 12 The former case is entitled by Molnar (2003: 194) ‘pleitropy’ and the latter ‘polygeny’. 13 Borghini and Williams (2008: 31) provide biconditionals besides the biconditional for atomic possibility – e.g. they provide biconditionals for conjunctive, disjunctive and existentially quantified possibilities. However, these other biconditionals are not relevant here. 10 Later I return to dispositional monism. 11 12 The former case is entitled by Molnar (2003: 194) ‘pleitropy’ and the latter ‘polygeny’. 13 13 Borghini and Williams (2008: 31) provide biconditionals besides the biconditional for atomic possibility – e.g. they provide biconditionals for conjunctive, disjunctive and existentially quantified possibilities. However, these other biconditionals are not relevant here. 14 Strictly, AP should include the term ‘iterated’ (or ‘n-iterated’ where ‘n’ = ≥1) such that AP reads ‘...some (iterated) dispositional properties…’ and ‘…some (iterated) disposition…’. The notion of n-iterated-disposi- tions, however, is not relevant here. See Borghini and Williams (2008: 30–31) for an elucidation. 15 An anonymous reviewer suggests that the point made within this paragraph might provide a way for MDists to avert problems concerning truthmakers for necessary truths that I discuss later. This might be true. But either way, I will be interested in mathematical truths and not just some set of necessary truths. I discuss the options for truthmakers for mathematical truths below. 16 See Vance (2014) and Austin (2015) for alternative types of objection to MD-like theories. 17 For objections of this type, to theories of MD’s type, see Allen (2017), Vetter (2015), Wang (2015), Contessa (2010) and Cameron (2008). 2 Dispositions, Dispositional Properties and MD They cite two reasons (and both are owed to Molnar (2003)): Firstly, (for some properties) from one dispositional property, many different manifestations can arise; and secondly, when combined with different com- binations of dispositional properties, the same dispositional property can give rise to different manifestations.12 Moreover, Borghini and Williams (2008) maintain: dispo- sitional properties are not extrinsic, but are intrinsic, and this is why dispositions need never manifest. Now, as mentioned, dispositions are modal; and so, MD is not fully modally reduc- tive. Borghini and Williams (2008: 33) maintain, however: “dispositions are something we need in our ontology anyway” whatever modal theory one endorses; so, they ask: why would one wish to reduce possibility to something else “if the dispositions can deal with [possibility] themselves?”. MD, then, is the view that metaphysical modality is reducible to one modal notion: dispositionality. And so, to analyse atomic possibility in terms of dispositionality,13 MD provides the following biconditional: 980 Philosophia (2021) 49:977–988 (AP) For any atomic state of affairs, S, S is possible ↔some actual objects have some dispositional properties d1,…, dn, which ground some disposition D, which has as its manifestation, S.14 (AP) For any atomic state of affairs, S, S is possible ↔some actual objects have some dispositional properties d1,…, dn, which ground some disposition D, which has as its manifestation, S.14 One can observe that AP coheres with dispositional realism’s contention that disposi- tions need not ever actually manifest. Likewise, it need only be the case that some actual object bears D, for S to be possible. Further, S will, in addition to being possible, be actual, if D manifests at some actual time, t; so, any actual state of affairs will also be possible given that the actual manifestations of dispositions ground actual states of affairs; and so, the (T)-axiom, ‘◻A→A’, is affirmed by MD. Additionally, MD admits that if anything is a constituent of D’s manifestation, that constituent is rendered possible by D’s existence. Thus, for any relevant, say, “general” possibility, D’s existence will ground the truth of that general possibility if it is a constituent of D’s manifestation (Borghini and Williams 2008: 27). 2 Dispositions, Dispositional Properties and MD Thus, in addition to grounding the possibility of shattering, a teacup’s being fragile grounds the following possibility: <possibly: at least one teacup exists>.15 And so, MD is an actualist and naturalist theory that semantically reduces and ontologically explicates possibility in terms of dispositionality: MD is actualist, because the dispositions of actual objects are the truth-makers of all possibility claims; and MD is naturalist, because objects’ dispositions are spatiotemporal and causal. 18 It is important to remember that Yates (2015) and Allen (2017) concern themselves with modal dispositionalist type theories, and not exclusively with Borghini and Williams’s (2008) MD. Here I speak as if Yates and Allen are referring to MD, for simplicity, but also to show that some of their responses to Yates’s objection miscarry given MD’s specifics. j y g p 19 This might not be quite right in perfect generality, if we admit impossible worlds into our modal theory 20 Given the truth of the axiom: ‘□A ↔¬◊¬A’, NEC is a substitution instance of POSS. 18 It is important to remember that Yates (2015) and Allen (2017) concern themselves with modal dispositionalist type theories, and not exclusively with Borghini and Williams’s (2008) MD. Here I speak as if Yates and Allen are referring to MD, for simplicity, but also to show that some of their responses to Yates’s objection miscarry given MD’s specifics. 19 This might not be quite right in perfect generality, if we admit impossible worlds into our modal theory. 20 Given the truth of the axiom: ‘□A ↔¬◊¬A’, NEC is a substitution instance of POSS. 3 MD’s Formal Deficiency: The Invalidation of (T) Now, there are (at least) two types of objection to MD.16 The first objection-type concerns MD’s material deficiency, and asks: does MD ascribe the correct truth values to certain intuitively true modal claims? Those who answer negatively point to certain intuitively true unactualised possibilities, and demonstrate MD’s inability to ascribe the correct truth values to claims pertaining to those possibilities by highlighting the fact that, on MD, such claims lack appropriate truth-makers.17 The second objection-type concerns MD’s formal deficiency, and asks: does MD validate all desirable modal logical axioms? Those who answer negatively point to certain intuitively true actualised possibilities (or necessities), and demonstrate MD’s inability to validate certain modal logical axioms, through the lack of truth-makers for claims pertaining to those possibilities. Philosophia (2021) 49:977–988 981 Here I argue for MD’s formal deficiency. I argue: neither Yates’s (2015) nor Allen’s (2017) respective responses to Yates’s (2015) objection that (T) is invalid on MD is successful.18 Recall: for MD, since actual manifestations of dispositions ground actual states of affairs, any actual state of affairs will also be possible. Accordingly, this affirms (T): ‘A →◊A’ (or, contrapositively: ‘□A →A’). The truth of (T) seems indisputable: were (T) not included in our modal logic, ‘□A’ would not express the claim ‘necessarily: A’, since the notion of it being necessary that A is the notion that it could not possibly be the case that not-A, and so A cannot be anything but true if □A is true. Thus, it goes without saying that any modal theory should validate (T).19 However, as noted by Yates (2015), MD invalidates (T). To provide the objection, I use Yates’s (2015: 413) biconditionals for the claims of the forms ‘◊A’ and ‘□A’, which use second-order quantification over dispositions: (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) Thus, since, for MD, all modal claims are made-true by dispositions, which are grounded by dispositional properties, any claim that has categorical properties as a truth-maker can be used to demonstrate that MD invalidates (T). Now, there are two plausible responses in the literature: one is owed to Yates (2015) and the other to Allen (2017). I argue that both responses fail. The first response proposes installing truth into MD’s biconditionals, so by stipula- tion (T) holds – here are the biconditionals (Yates 2015: 419): (POSS) ◊p ↔{p ∨∃D〉[p](D)}. (NEC) □p ↔{p ∧¬∃D〉[¬p](D)}. o, since ‘1 + 2 = 3’ is true, ‘1 + 2 = 3’ is possible. Thence, (T) holds. p Now, Yates (2015: 412) claims that “[he] see[s] no reason why” MDists should not “embrace [his] amendments” to the biconditionals. I do see why, however: installing truth into the biconditionals is tantamount to rejecting MD’s core claim: i.e., that all modal facts are made-true by dispositions alone. Thus, MDists welcome this solution on pain of rejecting MD’s principal tenet and thereby limiting dispositions’ capacity to ground modality. Now, Yates does accept that his solution renders MD-like theories weak rather than strong. But, I am not as sanguine about adopting the solution as he is – MDists, I argue, should not adopt the solution: installing truth into the biconditionals is not mere tinkering – Yates’s solution requires a fundamental change to the nature of MD.21 Indeed, Yates’s response here should seem odd to MDists: MDists wish to know why it is that claims such as ‘1 + 2 = 3′ are true and necessarily true, and MD should informatively explain this – indeed, MDists do not merely wish to know whether such claims simply are true, they wish to know why they are true. Certainly, on possible worlds accounts, truth is installed into the biconditionals: ‘possibly: p ↔at some world: p’ and ‘necessarily: p ↔at all worlds: p’. However, I take it that part of MDists’ aversion to possible worlds accounts of modality is precisely that truth is installed into the biconditionals (very often) with no further informative explanation for why, say, a certain proposition just so happens to be true at some world. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) (POSS) ◊p ↔∃D〉[p](D) (NEC) □p ↔¬∃D〉[¬p](D).20 Now, consider the claim: ‘1 + 2 = 3’. By NEC, if ‘¬∃D〉[‘¬(1 + 2 = 3)’](D)’ is true, then ‘1 + 2 = 3’ is necessarily true. Since this condition seems satisfied, we therefore get: ‘□(1 + 2 = 3)’. This is the correct result: true mathematical propositions should be necessarily true. Moreover, since whatever is necessary is also possible, we also get: ‘◊(1 + 2 = 3)’. Now, by POSS, if ‘∃D〉[‘1 + 2 = 3′](D)’ is true, ‘1 + 2 = 3′ is possible. This condition, however, seems unsatisfied: specifically, there fails to exist an object with a disposition which has ‘1 + 2 = 3′ as its manifestation. Certainly, there exist dispositions that, if manifested, ‘1 + 2 = 3′ would be true – but this is distinct to there existing a particular disposition that grounds the truth of ‘1 + 2 = 3′. So, by POSS, we therefore get: ‘¬◊(1 + 2 = 3)’ (which, given ‘¬◊A ↔□¬A’, we also get: ‘□¬(1 + 2 = 3)’). g ( ) ( g g ( ) ) So, on MD, when considering ‘1 + 2 = 3’, we get two sets of claims: {‘□(1 + 2 = 3)’, ‘¬◊(1 + 2 = 3)’} and {‘□¬(1 + 2 = 3)’, ‘◊(1 + 2 = 3)’}. Clearly then, (T) fails: ‘□(1+2=3) →1+2=3’ is invalid, since ‘□¬(1 + 2 = 3)’ is true, and ‘□¬(1 + 2 = 3)’ entails that necessarily ‘1 + 2 = 3’ is false; and ‘1+2=3 →◊(1+2=3)’ is invalid, since ‘¬◊(1 + 2 = 3)’ is true. The problem, I believe, stems from the fact that the truth-makers for ‘1 + 2 = 3’ seem to be categorical properties. So, if this is the case, it is perhaps not surprising that claims like ‘1 + 2 = 3’ deliver odd results on MD’s biconditionals – after all, MD’s Given the truth of the axiom: ‘□A ↔¬◊¬A’, NEC is a substitution instance of POSS. 982 Philosophia (2021) 49:977–988 biconditionals analyse modal claims by appealing to dispositions, which, on MD, are exclusively ontologically supported by dispositional properties, not categorical prop- erties. 21 Again, it is worth remembering that Yates (2015) is not specifically interested in MD; however, here I am treating his response as specifically targeted at MD for the reasons I give above. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) It seems that a virtue of MD is that, through its actualism, naturalism, and dispositionalism, MDists can provide true modal claims with informative explanations for why they are true: there is some causal and spatiotemporal explanation for why <possibly: p > is true – that possibility claim is true, not simply by virtue of its being true at the world, but by virtue of causal and spatiotemporal features of the world, namely, dispositions. Indeed, it seems (at least part of) the reason for propounding MD is to provide an informative explanation of why certain claims are necessarily true, but if we can now simply say that (at least some) necessity claims are true if their corresponding non-modal claims are true with no positive dispositionalist grounding to account for their being true, dispositions seem 983 Philosophia (2021) 49:977–988 redundant. And, if dispositions are now not required to do the work that MDists thought they did, why should MDists not provide a simpler account, and analyse all modal claims purely in terms of truth? That is, why should MDists not provide the following biconditionals? redundant. And, if dispositions are now not required to do the work that MDists thought they did, why should MDists not provide a simpler account, and analyse all modal claims purely in terms of truth? That is, why should MDists not provide the following biconditionals? (POSS) ◊p ↔{p}. (NEC) □p ↔{p ∧¬(¬p)}. (NEC*) □p ↔{p ∧¬(¬p)}. The answer from MDists, I take it, would be: Part of our motivation for positing MD, and part of our reason for rejecting possible worlds, is that we wish to provide a causal and spatiotemporal (that is: a naturalist) explanation of why it is that certain true claims are necessarily true: analysing modality in terms of dispositions allows us to provide such an explanation. Necessary truths should not float free from, or be prior to, or be more basic than, dispositions; dispositions should be more basic than, or be prior to, or not float free from, necessary truths. 22 Now, Borghini and Williams (2008: 21–22, fn2) note that they are content with not providing an account of logical necessity – however, mathematical necessity falls within the scope metaphysical necessity, and Borghini and Williams (2008: 21, fn2) claim that they wish to account for metaphysical possibility. Thus, if they cannot account for mathematical necessity, their theory fails to do what they wish it to do. 23 I thank an anonymous reviewer for pressing me on this point. 24 I thank an anonymous reviewer for this suggestion. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) But, at any rate, it seems like there is a huge amount that needs to be done here – all of it fascinating, but I think an in-depth discussion warrants a separate paper. p p p But, even if all the above were worked out there might still be a problem with this response. Even if it is true that our best natural science requires the existence of real mathematical objects which determine mathematical truths, or that our best natural science determines mathematical truths without the need for realist mathematical ontology, and the dispositions posited by MD act as truthmakers for our best natural science, it seems there will still be mathematical truths that are not required by our best natural science and thus the dispositions posited by MD are not ultimately or indirectly truthmakers for those truths. For instance, take the claim that the foundation of mathematics is set theory. Assume this is true. Let us suppose that the axioms of ZFC are true. Does our best natural science require the truths of the axioms of ZFC? Well, if the foundation of mathematics can be taken to be non-set theoretic, but category theoretic instead, and so all mathematics can be done with category theory instead of with set theory (or at least all the mathematics required by our best natural science can be done with either theory), then science will not require the truths of certain mathematical claims: for example, claims concerning the foundation of mathematics. That is, if our best natural science is neutral over whether set theory or category theory is the foundation of mathematics (since either theory can be used for all the mathematics required by our best natural science), then our best natural science does not require, say, set theory to be true and category theory to be false; and, the antecedent in this conditional seems right (indeed, it would be odd if our best natural science were not neutral on claims concerning, say, the nature of the cardinality of sets). However, it might be that the mathematics required by our best natural science entails that set theory is the foundation of mathematics, if necessarily set theory is the foundation of mathematics. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) 25 An anonymous reviewer suggests that an indispensability argument (for examples, see Quine (1976, 1980a, 1980b, 1981a, 1981b) and Putnam (1979a, 1979b)) might be called upon such that MD’s posited dispositions act as (indirect) truthmakers for mathematical claims via our best natural science – that is, MD’s dispositions make true our best natural science, and our best natural science requires the truths of mathematical claims, and so MD’s dispositions are indirect truthmakers for mathematical claims. Here is the reviewer: “Since the dispositions postulated by MD act as truth makers for our best scientific theory and that theory requires the existence of mathematical ontology which determines necessary mathematical truths (or else, that our scientific theory somehow determines that mathematical truths hold without the extra mathematical ontology), then the physical dispositions indirectly determine the mathematical necessities. … Of course, it is not generally true that the claim that ‘X determines the truth of Y which requires Z’ entails that ‘Z is true in virtue of X’. But if we take on board the MD perspective that there isn’t anything else to make Z true, then the MD theorists have given a way in which the class of Z truths (maths) are dependent on X (physical or natural dispositions), which was what they needed. So this response is consistent with MD, although it could not be used in an argument for it without begging the question”. This response is certainly fascinating, and certainly warrants in-depth discussion. Indeed, a full defence of the claim that our best natural science requires the real existence of mathematical objects which determine mathematical truths would be required, which would depend in part upon the soundness of an indispensability argument about mathematical objects – and this is no easy task. (For an excellent introduction to indispens- ability arguments, see Colyvan (2019)). Moreover, the perhaps weaker claim that our best natural science determines mathematical truths, but does not require the real existence of mathematical objects, would need to admit of a full defence, if the stronger claim were not held – indeed, some argument might have to be given as to why, say, natural science rules out an error theory of mathematical claims; that is, natural science might have to exclude the possibility that mathematics is simply a useful, but strictly speaking false, theory (assuming mathematics is not simply a useful, but false, theory). (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) Thus, if MDists were happy to accept Yates’s conclusion that MD cannot account for, say, mathematical necessities in terms of dispositions, then rather than accept Yates’s amendment of installing truth into the biconditionals to account for mathematical truths, it seems MDists should instead concern themselves exclusively with providing truthmakers for, say, nomic possibilities and necessities.22 Now, one might contend that Yates’s modal distinction which entails that necessary mathematical truths are not grounded by dispositions is justifiable; indeed, Yates only revises the biconditional to deal with such necessary truths.23 I am unconvinced here: certainly, it is sometimes claimed that necessary facts are brute; but, as I note above, I take it that the MDist would reject this claim on explanatory grounds. But, even if such a distinction can be justified, such that MDists can accept that some necessities are brute, and so can accept that their naturalist explanatory project does not extend to such necessities, it is unclear that such a distinction would render the theory preferable to the possible worlds framework, all things considered: sure, MD might be able to provide a more satisfactory explanation of why, say, it is possible that I be a concert pianist, but a possible worlds theory can provide theoretical unity. It is at best unclear which virtue is preferable. Perhaps, however, it could be contended that the truth claims within the respective right-halves of the biconditionals POSS and NEC* (and POSS and NEC) are true by virtue of the actual manifestations of dispositions.24 Certainly, one may attempt to provide a dispositionalist account of, say, mathematical truths – and I consider this strategy below – however, it is then unclear why truth is installed into the bicondi- tionals’ respective right-halves; if, for instance, the claim ‘p’ within ‘{p ∨∃D〉[p](D)}’ 984 Philosophia (2021) 49:977–988 is true by virtue of an actual manifestation of a disposition, then why should the ‘p’ be included in the biconditional at all? The ‘p’ would be redundant.25 is true by virtue of an actual manifestation of a disposition, then why should the ‘p’ be included in the biconditional at all? The ‘p’ would be redundant.25 For these reasons, then, I believe Yates’s response is not available to the MDist. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) But, then, it seems that we are back at the original problem: sure, there is perhaps no disposition which has as its manifestation the falsity of the claim that set theory is the foundation of mathematics, but what is the truthmaker for that claim so that, say, ZFC possible? The truthmaker(s) cannot indirectly make true that claim via our best natural science, since our best natural science is neutral on the truth of that claim; thus, we have to look elsewhere. Below, I explore some ontological options for the truthmakers of necessary mathematical truths. Now, I have not done full justice to the reviewer’s suggestion, but I hope to have shown that, despite its being a fascinating response, it will be no easy task in the first instance, and in the second, it may not even give the MDist what he needs after all: there may well be mathematical truths that are still not rendered possible. Philosophia (2021) 49:977–988 985 So, let us examine the second response, suggested by Allen (2017). Her re- sponse, in essence, is as follows: MDists should ontologically construe all proper- ties as essentially dispositional properties; that is, to espouse, what I referred to earlier as, dispositional monism. Now, Wang (2015: 456) claims: “defender[s] of essentially dispositional properties need not ban categorical properties”. But, to avoid invalidating (T), MDists may have to.26 So, paradigmatic categorical prop- erties are now taken to be dispositional properties which support dispositions that, following Hütteman’s (2013) terminology, continuously manifest. Thus, the truth- maker for ‘1 + 2 = 3’ is a continuously manifesting disposition, which is grounded by certain dispositional properties. But, one must ask: which objects have these dispositional properties? The first option is to say: all objects. This seems absurd, however: indeed, (without calling upon merely ad hoc reasons) why do I, my cat and my laptop each have dispositional properties that support a continuously- manifesting disposition for ‘1 + 2 = 3’? Thus, we must move to the second option, which is to say: some objects.27 At this point Allen provides two ontological options, which both – she believes – help MD validate (T). Now, strictly speaking, Allen is correct here, both options do validate (T) – I do not deny that, like I did not deny that Yates’s response validated (T). 29 My objection is not entirely distinct from Aristotelianism, since Aristotelianism allows MD to be naturali 30 S (201 ) h ( ) l f b y 28 For an elaboration of Aristotelianism in mathematics, see Mendell (2004). 29 26 Now, strictly speaking, Wang may be correct: MDists could espouse a dual ontology: i.e. every categorical property has a dispositional counterpart. This, however, delivers an unnecessarily bloated ontology. 27 Allen (2015: 39) notes that Vetter hinted at this response when replying to a paper on 26th November 2014, University of Oxford My objection is not entirely distinct from Aristotelianism, since Aristotelianism allows MD to be naturalist. See Rosen (2017) on the (non-)causal powers of abstracta. 30 See Rosen (2017) on the (non-)causal powers of abstracta. 26 Now, strictly speaking, Wang may be correct: MDists could espouse a dual ontology: i.e. every categorical property has a dispositional counterpart. This, however, delivers an unnecessarily bloated ontology. 27 Allen (2015: 39) notes that Vetter hinted at this response when replying to a paper on 26th November 2014, University of Oxford. 28 For an elaboration of Aristotelianism in mathematics, see Mendell (2004). 29 My objection is not entirely distinct from Aristotelianism, since Aristotelianism allows MD to be naturalist. 30 See Rosen (2017) on the (non-)causal powers of abstracta. 30 See Rosen (2017) on the (non-)causal powers of abstracta 29 My objection is not entirely distinct from Aristotelianism, since Aristotelianism allows MD to be naturalist. 30 See Rosen (2017) on the (non-)causal powers of abstracta. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) But, I do argue, in the way that I argued against Yates’s response, that Allen’s prescribed solutions validate (T) only at a high cost to MD. Let us examine Allen’s ontological options. The first option Allen sketches is to claim that numbers, as abstracta, instantiate certain dispositional properties that ground certain dispositions. So, the disposi- tional properties of the abstract mathematical objects, 1, 2 and 3, when combined, produce the continuously-manifesting-disposition that make-true the claim: ‘1 + 2 = 3’. Now, Allen briefly mentions that, whilst a Platonist about dispositions might welcome this suggestion, an Aristotelian might not – and arguably, MDists are Aristotelians.28 But, there is, however, a sufficiently strong objection to Allen’s first suggestion, which, I believe, demonstrates why MDists should not adopt such an ontological construal of mathematical objects (or at any rate should be rather cautious about adopting such an ontological picture). Specifically, this option, I argue, jars with MD’s naturalism.29 Naturalism is the view that objects’ disposi- tions are spatiotemporal and causal. Now, since abstracta are not spatiotemporal, it is reasonable to suppose that neither are their dispositions, if they have any. Moreover, it is conventional to regard abstracta as causally inert, that is, abstract entitles are non-causal; and, since the entities are non-causal, it is reasonable to suppose that neither are their dispositions, (again) if they have any.30 So, I argue: MDists are permitted to use the dispositions of numbers, which are taken as abstract entities, to ground mathematical necessities, only if they are willing to violate MD’s naturalism – a core feature of MD. Of course, MDists are at liberty, for a price, to 986 Philosophia (2021) 49:977–988 violate their naturalism – indeed, often the point of raising objections to theories and proposals is to better understand the costs involved in adopting those theories and proposals. However, it seems, to me at least, that since naturalism is such a core feature of MD’s account of modality, permitting certain violations of naturalism, at best, presents a high price for MDists to pay, and, at worst, distorts MD to the extent that it no longer sufficiently resembles Borghini and Williams’s original theory. So, we have Allen’s second ontological option: adopt physicalism about numbers – that is, numbers existentially depend upon actual, concrete objects. So, the constantly- manifesting dispositions of (at least some) concrete objects make-true mathematical necessities. See Mill (1843) for a defence of a similar ontological account of numbers. 32 Now, some may think that one, albeit not physical but concrete, object grounds mathematical necessities: God. I am very sympathetic to this suggestion (see Leftow (2012) for an exploration of this topic and related issues). However, it should be obvious that unless God is taken to be a spatiotemporal object – or at any rate, a natural object – MDists should reject such a suggestion. 33 An anonymous reviewer suggests that the proposal in footnote 25 may be applied here. Again, the same problems I mentioned therein arise. The reviewer also suggests that maybe necessary mathematical truths are “determined by dispositions belonging to worlds”. Now, I am unsure what to make of this suggestion: if the point is that the universe’s dispositions ground necessary mathematical truths, taking the universe as an object, we can ask: which parts of that object have the dispositional properties that support the universe’s disposi- tion(s) to determine necessary mathematical truths? That is, we are back at the original question: do all parts of the universe have dispositional properties that support a disposition that determines necessary mathematical truths or do only some? 34 For instance, Borghini and Williams (2008: 32, fn.28) explicitly reject axiom-(4). 31 See Mill (1843) for a defence of a similar ontological account of numbers. 32 (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) This view – although congruous with MD’s naturalism – is certainly, I believe, unusual and antiquated (and plausibly antiquated for a good reason (although, I do not think that something’s merely being antiquated is reason in itself to reject that something!)).31 It additionally seems that we have returned to the question: which objects have the relevant dispositions to ground mathematical necessities? Indeed, we originally thought it absurd that all objects have dispositions that ground mathematical necessities, so if MDists agree with that being absurd, they must now claim that only some concrete objects have dispositions that ground mathematical necessities. But this latter claim, I believe, seems more absurd than the former: why is it that members of one concrete object-type, and not members of another, have dispositions that ground mathematical necessities? Surely it would seem extremely odd that certain concrete objects have certain dispositions that ground mathematical necessities and certain other concrete objects do not.32 So, I argue: if MDists adopt physicalism about numbers, their theory is at best extremely bizarre and at worst absurd. Now, accepting either conse- quence is perhaps a price worth paying; but, since I am not an MDist, the price is not mine to pay, and, at any rate, MDists will have a better understanding of how large their purses are.33 And so: contra Allen, both ontological options, I believe, fail, or at least present substantial costs; thus, this second response, I believe, is also unavailable to MDists – or at least is available only at a high price. Therefore, since both the most plausible responses to MD’s invalidation of (T) are unsuccessful (or at least present rather high costs), MD should be rejected (or the cost better be demonstrated to be worth it). Unlike perhaps some logical axioms, rejecting (T) is non-negotiable.34 34 For instance, Borghini and Williams (2008: 32, fn.28) explicitly reject axiom-(4). 987 Philosophia (2021) 49:977–988 Acknowledgements I would like to thank Farbod Akhlaghi, Sam Speight and Martin Pickup for very kind providing me with comments on earlier drafts of this paper. 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. (Where ‘∃D〉[p](D)’ = ‘there exists a disposition D such that D has p as its manifestation’) The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Adams, R. (1974). Theories of actuality. Noûs, 8, 211–231. Allen, S. R. (2017). From possibility to properties? Or from properties to possibility? Philosophy, 92(1), 21–49 Austin, C. J. (2015). The Truthmaking argument against Dispositionalism. Ratio, 28(3), 271–285. Bird, A. (2007). Nature’s metaphysics. Oxford: Oxford University Press. Borghini, A., & Williams, N. E. (2008). A dispositional theory of possibility. Dialectica, 62(1), 21– Cameron, R. (2008). Truthmakers and modality. Synthese, 164, 261–280. Colyvan, M., 2019. Indispensability arguments in the philosophy of mathematics. [online]. Stanford Encylopedia of Philosophy. E. N. Zalta ed. available at: <https://plato.stanford.edu/entries/mathphil- indis/> [Accessed 30 July 2020]. Contessa, G. (2010). Modal Truthmakers and two varieties of Actualism. Synthese, 174(3), 341–353. Ellis, B. (2002). The philosophy of nature – A guide to the new essentialism. Montreal: McGill-Queen’s University Press. Ellis, B. (2001). Scientific Essentialism. New York: Cambridge University Press. Heil, J. (2003). From an ontological point of view. New York: Oxford University Press. Hütteman, A. (2013). A disposition-based process-theory of causation. In S. Mumford & M. Tugby (E Metaphysics of science (pp. 101–122). Oxford: Oxford University Press. ckson, F. (1995). From metaphysics to ethics: A Defence of conceptual analysis. New York: Oxford University Press. Jackson, F., Prior, E., & Pargetter, R. (1982). Three theses about dispositions. American Philosophical Quarterly, 19, 251–256. Jacobs, J. D. (2010). A powers theory of modality: Or, how I learned to stop worrying and reject possible worlds. Philosophical Studies, 151(2), 227–248. Leftow, B. (2012). God and necessity. Oxford: Oxford University Press. Leftow, B. (2012). God and necessity. Oxford: Oxford University Press. Lewis, D. K. (1997). Finkish dispositions. Reprinted. In Papers in Metaphysics and Epistemology (Vol. 199 pp. 133–151). New York: Cambridge University Press. Lewis, D. K. (1997). Finkish dispositions. Reprinted. In Papers in Metap pp. 133–151). New York: Cambridge University Press. Lewis, D. K. (1986). On the Plurality of Worlds. Oxford: Clarendon Press. Mackie, J. L., 1977. Dispositions, grounds, and causes. Reprinted in: R. Tuomela, ed., Dispositions. Dordrecht: D. Reidel Publishing. 1978. Pp. 99–107. Mellor, D. H. (2000). The semantics and ontology of dispositions. Mind, 109, 757–780. Mendell, H., 2004. Aristotle and mathematics. [online] Stanford Encylopedia of Philosophy. E. N. Zalta ed. available at: <https://plato.stanford.edu/entries/aristotle-mathematics/#7.1> [Accessed 14 April 2020]. Mendell, H., 2004. Aristotle and mathematics. [online] Stanford Encylopedia of Philosophy. E. N. Zalta ed. References il bl t htt // l t t f d d / t i / i t tl th ti /#7 1 [A d 14 A il 2020] Mill, J. S. (1843). A system of logic, ratiocinative and inductive. London: Longmans, Green and Company Molnar, G. (2003). Powers. New York: Oxford University Press. Mumford, S. (2004). Laws in nature. London: Routledge. Plantinga, A. (1974). The nature of necessity. Oxford: Oxford University Press. lantinga, A. (1974). The nature of necessity. Oxford: O Prior, E. (1985). Dispositions. Aberdeen: Aberdeen University Press. Pruss, A. R. (2011). Actuality, possibility, and worlds. New York: Continuum. Putnam, H., 1979a. What is mathematical truth. In: Mathematics Matter and Method: Philosophical Papers, Volume 1. 2nd ed, Cambridge: Cambridge University press, pp. 60–78. Philosophia (2021) 49:977–988 988 Putnam H.,–1979b. Philosophy of Logic. Reprinted in: Mathematics Matter and Method: Philosophic Papers, Volume 1. 2nd ed, Cambridge: Cambridge University Press, pp. 323–357. Quine, W. V. O., 1976. Carnap and logical truth. Reprinted in: The Ways of Paradox and Other Essays. Quine, W. V. O., 1976. Carnap and logical truth. Reprinted in: The Ways of Paradox and Other Essays. Revised ed, Cambridge, MA: Harvard University press, pp. 107–132. Quine, W. V. O. (1980a). On What There Is. Reprinted. In From a Logical Point of View (2nd ed., pp. 1–19). C b id MA H d U i it P Quine, W. V. O. (1980a). On What There Is. Reprinted. In From a Logical Point of View (2nd ed., pp. 1–19). Cambridge, MA: Harvard University Press. Quine, W. V. O. (1980a). On What There Is. Reprinte Cambridge, MA: Harvard University Press. Quine, W. V. O. (1980b). Two Dogmas of Empiricism. Reprinted. In From a Logical Point of View (2nd ed Quine, W. V. O. (1980b). Two Dogmas of Empiricism. Reprinted. In From a Logical Point of View (2nd ed., pp. 20–46). Cambridge, MA: Harvard University Press. pp. 20–46). Cambridge, MA: Harvard University Press. pp. 20–46). Cambridge, MA: Harvard University Press. Quine, W. V. O. (1981a). Things and Their Place in Theories. In Theories and Things (pp. 1–23). Cambridge, MA: Harvard University Press. Quine, W. V. O. (1981b). Success and Limits of Mathematization. In Theories and Things (pp. 148–155). Cambridge, MA: Harvard University Press. Cambridge, MA: Harvard University Press. Rosen G., 2017. Abstract Objects. [Online] Stanford Encylopedia of Philosophy. E. N. Zalta ed. Available a <https://plato.stanford.edu/entries/abstract-objects/> [Accessed: 5 September 2017]. ublisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps d institutional affiliations. References p p j [ p ] Vance, C., 2014. Dispositional Modal Truthmakers and the Necessary Origin Philosophia 41, pp. 1111–1127. Vetter, B. (2015). Potentiality. Oxford: Oxford University Press. Vance, C., 2014. Dispositional Modal Truthmakers and the Necessary Origin Philosophia 41, pp. 1111–112 Vetter, B. (2015). Potentiality. Oxford: Oxford University Press. Wang, J. (2015). The modal limits of Dispositionalism. Noûs, 49(3), 454–469. Yates, D. (2015). Dispositionalism and the modal operators. Philosophy and Phenomenological Research, 91(2), 411–424. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2149557432	https://link.springer.com/content/pdf/10.1007/s11671-008-9230-5.pdf	English	null	Regulation of Transgene Expression in Tumor Cells by Exploiting Endogenous Intracellular Signals	Nanoscale research letters	2,008	cc-by	3,629	Regulation of Transgene Expression in Tumor Cells by Exploiting Endogenous Intracellular Signals Daisuke Asai Æ Jeong-Hun Kang Æ Riki Toita Æ Akira Tsuchiya Æ Takuro Niidome Æ Hideki Nakashima Æ Yoshiki Katayama Received: 11 September 2008 / Accepted: 2 December 2008 / Published online: 17 December 2008 to the authors 2008 DNA complex was observed. Moreover, after microinjec- tion of polymer/GFP-encoding DNA complexes into HepG2 cells at cation/anion (C/A) ratios of 0.5 to 2.0, signiﬁcant expression of GFP was observed in all cases using PPC(S)/DNA complexes, but no GFP expression was observed in the negative control PPC(A)/DNA complex- microinjected cells at C/A ratios of 1.0 and 2.0. On the other hand, GFP expression from PPC(S)/DNA complexes was completely suppressed in cells pretreated with PKCa inhibitor (Ro31-7549). These results suggest that our gene regulation system can be used for tumor cell-speciﬁc expression of a transgene in response to PKCa activity. Abstract Recently, we have proposed a novel strategy for a cell-speciﬁc gene therapy system based on responses to intracellular signals. In this system, an intracellular signal that is speciﬁcally and abnormally activated in the diseased cells is used for the activation of transgene expression. In this study, we used protein kinase C (PKC)a as a trigger to activate transgene expression. We prepared a PKCa-responsive polymer conjugate [PPC(S)] and a neg- ative control conjugate [PPC(A)], in which the phosphorylation site serine (Ser) was replaced with alanine (Ala). The phosphorylation for polymer/DNA complexes was determined with a radiolabel assay using [c-32P]ATP. PPC(S)/DNA complexes were phosphorylated by the addition of PKCa, but no phosphorylation of the PPC(A)/ Keywords Intracellular signal Protein kinase C Gene delivery Nanoparticle Tumor Keywords Intracellular signal Protein kinase C Gene delivery Nanoparticle Tumor Daisuke Asai and Jeong-Hun Kang contributed equally to this work. Nanoscale Res Lett (2009) 4:229–233 DOI 10.1007/s11671-008-9230-5 Nanoscale Res Lett (2009) 4:229–233 DOI 10.1007/s11671-008-9230-5 NANO EXPRESS Daisuke Asai and Jeong-Hun Kang contributed equally to this work. Synthesis of the Peptide Substrate Synthesis of the Peptide Substrate g Living cells contain numerous signal transduction pathways that respond to extracellular signals and regulate or modulate gene expression. Extracellular signals either penetrate the cellular membrane or bind to the extracellular domain of cell surface receptors. The activated receptors are subsequently able to change the amount or intracellular distribution of second messengers through effectors. The second messengers then activate protein targets that ﬁnally control gene expression by acting on further downstream targets. In these intracellular signal transduction pathways, phosphorylation by protein kinases plays an important role and functions through the activation of target proteins [7, 8]. Two peptide substrates, FKKQGSFAKKK and FKKQGA FAKKK, each with a methacryloyl group at the amino- terminus, were synthesized using an automatic peptide synthesizer according to standard Fmoc-chemistry proce- dures. After treatment with triﬂuoroacetic acid (TFA), peptides were puriﬁed on an Inertsil ODS-3 column (250 9 20 mm, 3.5 lm; GL Sciences Inc., Tokyo, Japan) using a BioCAD Perfusion Chromatography system (Ikemoto Scientiﬁc Technology Co., Tokyo, Japan) and a linear A-B gradient at a ﬂow-rate of 8 mL/min, where eluent A was 0.1% TFA in water and eluent B was 0.1% TFA in acetonitrile. Protein kinase C (PKC) is a calcium- and phospholipid- dependent serine/threonine kinase. The PKC isozymes are classiﬁed into three subfamilies based on structural and activational characteristics: conventional or classic PKCs (cPKCs: a, bI, bII, and c), novel or non-classic PKCs (nPKCs: d, e, g, and h), and atypical PKCs (f, i, and k). The activation of cPKCs requires diacylglycerol (DAG) as an activator and phosphatidylserine (PS) and Ca2? as activa- tion cofactors. The nPKCs are regulated by DAG and PS, but do not require Ca2? for activation. In the case of atypical PKCs, their activity is stimulated only by PS, and not by DAG and Ca2? [7–10]. Among these PKC families, PKCa is widely expressed in many tissues and plays key roles in the differentiation and proliferation of tumors such as melanoma, hepatoma, and breast cancer. Upon stimu- lation, PKCa is translocated from the cytosol to the cellular membrane where its subsequent activation by DAG and PS occurs. Increases in the Ca2? concentration increase PKCa translocation to the cellular membrane, leading to activa- tion of proteins that trigger cellular responses, such as proliferation and differentiation. Extraordinary activation of PKCa has been identiﬁed in transformed cell lines and in several cancers [7–10]. Introduction D. Asai H. Nakashima Department of Microbiology, St. Marianna University School of Medicine, 2-16-1 Sugao, Miyamae-ku, Kawasaki 216-8511, Japan D. Asai H. Nakashima Department of Microbiology, St. Marianna University School of Medicine, 2-16-1 Sugao, Miyamae-ku, Kawasaki 216-8511, Japan Gene therapy is recognized as a medical approach for the treatment of diseases that are difﬁcult to cure, such as tumors. Several viral and non-viral carriers for gene transfer have been developed for either in vivo or ex vivo/ in vitro use [1–4]. The most important viral carriers, characterized in laboratory studies and clinical trials, have been inactivated retroviruses, adenoviruses, adeno-associ- ated viruses, and herpes viruses. These viruses show relatively high transfection efﬁciency, but have some clinical safety problems. On the other hand, non-viral methods of gene delivery, including cationic lipofection, calcium phosphate precipitation, gene guns, and injection of naked DNA, generally show low transfection efﬁciency but are not pathogenic [1–4]. However, there is another serious issue regarding the present gene delivery method- ologies. Almost all the gene delivery methods currently D. Asai J.-H. Kang H. Nakashima Y. Katayama CREST, Japan Science and Technology Corporation, 4-1-8 Honcho, Kawaguchi-shi, Saitama 332-0012, Japan J.-H. Kang T. Niidome Y. Katayama (&) Department of Applied Chemistry, Faculty of Engineering, Kyushu University, 744 Motooka, Nishi-Ku, Fukuoka 819-0395, Japan e-mail: ykatatcm@mbox.nc.kyushu-u.ac.jp R. Toita A. Tsuchiya T. Niidome Y. Katayama Graduate School of Systems Life Sciences, Kyushu University, Fukuoka 819-0395, Japan T. Niidome Y. Katayama Center for Future Chemistry, Kyushu University, 744 Motooka, Nishi-Ku, Fukuoka 819-0395, Japan T. Niidome Y. Katayama Center for Future Chemistry, Kyushu University, 744 Motooka, Nishi-Ku, Fukuoka 819-0395, Japan 12 3 3 230 Nanoscale Res Lett (2009) 4:229–233 complexes was detected using a radiolabel assay. More- over, polymer/DNA complexes, microinjected into HepG2 cells, were shown to regulate gene expression. developed have insufﬁcient ability to speciﬁcally recognize and target diseased cells and to distinguish them from normal cells, especially in the same tissue or organ. Thus a number of strategies that address the issue of target- selective gene delivery have been reported. These systems generally involve the use of various ligands that selectively bind to a cell-surface marker on the target cell [5, 6]. Synthesis of Peptide-Pendant Polymer Polymers were synthesized as described previously [11, 12]. Brieﬂy, acrylamide (8.9 mg, 125 lmol) and N-methacry- loylpeptide (7.5 mg, 5.5 lmol), in which the methacryloyl group was attached at the amino terminus of the peptide, were dissolved in water, degassed with nitrogen for 5 min and then polymerized using ammonium persulfate (0.86 mg, 3.74 mmol) and N,N,N0,N0-tetramethylethylenediamine (1.1 lL, 7.38 mmol) as a redox couple at room temperature for 90 min. The synthesized sample was dialyzed against water overnight in a semi-permeable membrane bag with a molecular weight cutoff of 50,000. The dialyzed sample was lyophilized and the ﬁnal sample was obtained as a white powder. The concentration of the peptide was estimated by elemental analysis. Synthesis of the Peptide Substrate If such hyperactivated PKCa can be used for the activation of transgene expression, cancer cell- speciﬁc gene regulation becomes possible. 123 Phosphorylation Assay for Polymer/DNA Complexes The phosphorylation reaction of the polymer/DNA com- plex was quantiﬁed by measuring 32P transfer from [c-32P]ATP into the substrate peptide of the polymer, according to the manufacture’s recommendations (Sigma, Louis, MO). The PPC/DNA complexes were prepared by incubation of the PPC with GFP-encoding DNA (pEGFP- C1, Clontech Laboratories, Inc., Mountain View, CA) in 20 mM Tris–HCl buffer solution (pH 7.5) for 15 min at room temperature. Phosphorylation reactions were carried out in 70 lL of buffer {20 mM Tris–HCl, pH 7.5, 10 mM MgCl2, 0.1 mM CaCl2, a mixture of 100 lM ATP and [c-32P]ATP (1.0 lCi; 6,000 Ci/mmol, GE Healthcare, Recently, we have proposed a novel strategy for cell- speciﬁc gene therapy. In this strategy, an intracellular signal that is speciﬁcally and abnormally activated in the target diseased cells is used for the activation of transgene expression [11, 12]. In this study, we prepared two polymers, a PKCa- responsive polymer [PPC(S)] and a negative control polymer [PPC(A)]. Phosphorylation of polymer/DNA 123 231 Nanoscale Res Lett (2009) 4:229–233 Buckinghamshire, UK), 100 lg/mL PS, and 20 lg/mL DAG} containing polymer/DNA complexes at cation/anion (C/A) ratios of 1.0 and/or 2.0, and 2 ng/mL recombinant human PKCa (Sigma) for 8 h at 37 C. The reaction mixture (20 lL) was spotted onto polyvinylidene diﬂuoride transfer membranes (2 9 2 cm2 Hybond-P, Amersham Biosciences, Piscataway, NJ). The membranes were washed three times with 5% TCA and the radioactivity on each membrane was determined by liquid scintillation counting. Japan) with 0.5 ± 0.2 lm injection pipettes (Femtotips, Eppendorf, NSW). Cytoplasmic microinjections were per- formed under conditions of Pi = 30–50 hPa, Pc = 8 hPa, and an injection time of 180–250 ms. Plasmids were diluted in TE buffer to ﬁnal concentrations of 100–250 ng/lL. PPCs were mixed with plasmid DNA at several C/A ratios (0.5, 1.0, and 2.0) for 10 min at room temperature. After adding Dextran-Texas Red in PBS (1 mg/mL) to the reaction mixture, the solution was transferred to Femtotips and injected into 300 cells. Twenty-four hours after injection, GFP expression was monitored by ﬂuorescence micros- copy, and the ratio of cells expressing GFP to Texas Red positive cells was calculated. Western Blot Analysis HepG2 cells (5 9 105 cells) (JCRB1054) were plated in a 100 mm dish and were cultured in Dulbecco’s modiﬁed Eagle’s medium (Gibco Invitrogen Co., Grand Island, NY) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 lg/mL streptomycin, and 0.25 lg/mL amphotericin B. The cells were incubated in a humidiﬁed atmosphere containing 5% CO2 and 95% air at 37 C. Three days after incubation, the cells were lysed in lysis buffer containing 50 mM Tris–HCl (pH 7.5), 0.15 M NaCl, 1% Nonidet P-40, 2 mM EDTA, 50 mM NaF, and 0.2 mM Na3VO4, supplemented with CompleteTM protease inhibi- tor cocktail (Roche Diagnostics GmbH, Mannheim, Germany) on ice for 15 min. The cell lysate was centri- fuged to remove insoluble materials and the lysate was then immunoblotted with anti-PKCa serum (Cell Signaling, Danvers, MA) and anti-phosphoPKCa (Ser657) serum (UPSTATE, Lake Placid, NY). Bound antibodies were visualized by chemiluminescence. A recombinant active human PKCa (1 ng/lane) was used as a positive control. Fig. 1 Synthetic scheme and chemical structure of the polymer. The polymer was synthesized by polymerization of acrylamide and N- methacryloylpeptide using ammonium persulfate and N,N,N0,N0- tetramethylethylenediamine Cytotoxicity of the Polymer Toward Cells Cytotoxicity of the Polymer Toward Cells To identify cytotoxicity of the polymer toward cells, HepG2 cells were incubated in the presence or absence of the polymer (0–30 lg/mL) for 48 h in a 24-well plate. Staining with Trypan Blue (0.4%; Gibco Invitrogen Co.) was used to calculate the percentage of viable cells. A 0.2 mL aliquot of cell suspension in Hanks’ balanced salt solution (Gibco) was added to Trypan Blue Mix (0.5 mL of 0.4% Trypan Blue and 0.3 mL of Hanks’ balanced salt solution) and incubated for 5 min at room temperature. The number of unstained cells and the total number of cells were counted in a hemocytometer. The percent cell via- bility was calculated by normalizing the cell viability of the treated cells to that of untreated cells. 0 40,000 80,000 120,000 160,000 CPM PKCα + + + - 0. 2 0. 1 C/A Polymer - + + + + + 2.0 + - - + PPC(S) PPC(A) + + Fig. 2 Phosphorylation of PPCs by PKCa in the presence or absence of DNA. The phosphorylation of the polymers or polymer/DNA complexes was determined by the radiolabel assay using [c-32P]ATP. Data show mean ± SD (n = 3). CPM count per minute; C/A cation/ anion charge ratio 1 3 0 40,000 80,000 120,000 160,000 CPM PKCα + + + - 0. 2 0. 1 C/A Polymer - + + + + + 2.0 + - - + PPC(S) PPC(A) + + Fig. 2 Phosphorylation of PPCs by PKCa in the presence or absence of DNA. The phosphorylation of the polymers or polymer/DNA complexes was determined by the radiolabel assay using [c-32P]ATP. Data show mean ± SD (n = 3). CPM count per minute; C/A cation/ anion charge ratio Results and Discussion We prepared a PKCa-responsive conjugate [PPC(S)] and a negative control conjugate [PPC(A)], in which the phosphorylation site serine (Ser) was changed to alanine (Ala), using radical copolymerization (Fig. 1). The content of the peptide, as the side chain of the polymer, was esti- mated to be 5.5 mol% for PPC(S) and 6.5 mol% for PPC(A). Since the substrate peptide has ﬁve cationic amino acids (lysine), the PPCs are able to condense with anionic DNA. The peptide (FKKQGSFAKKK) used in this study is a PKCa-speciﬁc substrate. The Km and Vmax of the peptide are 17.4 lM and 138 nmol/min/mg, respectively [13]. Fig. 3 The level of endogenous PKCa and the activated form of PKCa, phospho-PKCa, in HepG2 cells. a HepG2 cell lysates and b recombinant PKCa (1 ng/lane) were immunoblotted with anti-PKCa serum or anti-phosphoPKCa (Ser657) serum. Bound antibodies were visualized using a chemiluminescence assay The phosphorylation reaction of the polymers or the polymer/DNA complexes was determined with a radiolabel assay using [c-32P]ATP. PPC(S) only and PPC(S)/DNA complex at C/A ratios of 1.0 and 2.0 were phosphorylated by the addition of PKCa, but no phosphorylation of PPC(A) only or of PPC(A)/DNA complex at a C/A ratio of 2.0 was observed (Fig. 2). These results suggest that PKCa can recognize the substrate peptide in the PPC(S)/DNA complex and that PPC(A) is a suitable control polymer. Fig. 3 The level of endogenous PKCa and the activated form of PKCa, phospho-PKCa, in HepG2 cells. a HepG2 cell lysates and b recombinant PKCa (1 ng/lane) were immunoblotted with anti-PKCa serum or anti-phosphoPKCa (Ser657) serum. Bound antibodies were visualized using a chemiluminescence assay Several studies have reported that PKCa was signiﬁ- cantly over-expressed in HepG2 cells, whereas other PKC isozymes (bI, bII, d, e, h, f, k, and i) were expressed at low levels or were not detected [14, 15]. We detected the level of endogenous PKCa and the activated form of PKCa, phospho-PKCa, in HepG2 cells using Western blot analy- sis. As shown in Fig. 3, endogenous PKCa and phospho- PKCa were identiﬁed in the lysate of HepG2 cells. 0 20 40 60 80 100 120 Control (0) 10 20 30 Polymer concentration (µg/ml) Relative viability (%) 24h 48h 0 20 40 60 80 100 120 Control (0) 10 20 30 Polymer concentration (µg/ml) Relative viability (%) 24h 48h Fig. 4 Toxicity of the polymer toward HepG2 cells. Microinjection Study Microinjection Study Cells were microinjected at day 1 post-plating using a Cellinjector CI-2000 system (Fujitsu limited, Kanagawa, 12 3 232 Nanoscale Res Lett (2009) 4:229–233 Results and Discussion Cells were incubated in the presence or absence of the polymer (0–30 lg/mL) for 48 h in a 24-well plate. The cell viability (%) was calculated as the ratio of the number of unstained cells to the total number of cells Relative viability (%) We examined the cytotoxicity of the polymer toward HepG2 cells. The polymer [PPC(S)] concentrations used for the cytotoxicity test at C/A ratios of 0.5 to 2.0 were below 10 lg/mL. The assay results revealed that the polymer hardly affected cell viability ([90%) in the con- centration range of 10 to 30 lg/mL over 48 h (Fig. 4); similar results were obtained from the PPC(A) polymer (data not shown). These results indicate no or very low toxicity of the polymer toward cells. Fig. 4 Toxicity of the polymer toward HepG2 cells. Cells were incubated in the presence or absence of the polymer (0–30 lg/mL) for 48 h in a 24-well plate. The cell viability (%) was calculated as the ratio of the number of unstained cells to the total number of cells For a gene regulation system that responds to intracel- lular PKCa activation, the PPC/DNA complex has to be C P P A +1.0 +2.0 +0.5 +2.0 +2.0 PPC S + Ro31-7549 PPC(S) contained a Ser phosphorylation site for PKCa, but the phosphorylation site was changed to Ala in PPC(A), leading to no phosphorylation by PKCa. Ro31-7549, a PKCa inhibitor C P P A +1.0 +2.0 +0.5 +2.0 +2.0 PPC S + Ro31-7549 PPC(S) contained a Ser phosphorylation site for PKCa, but the phosphorylation site was changed to Ala in PPC(A), leading to no phosphorylation by PKCa. Ro31-7549, a PKCa inhibitor GFP Texas Red Phase-contrast (C/A) C P P yl n o r e m yl o P A N D d e k a N A +1.0 +2.0 +0.5 PPC S +1.0 +2.0 +0.5 +2.0 PPC S + Ro31-7549 Fig. 5 GFP expression after polymer/DNA complex was microin- jected into HepG2 cells. Naked DNA, polymer only, and polymer/ DNA complexes (C/A = 0.5, 1.0, and 2.0) were microinjected into HepG2 cells and GFP expression was detected 24 h after injection. PPC(S) contained a Ser phosphorylation site for PKCa, but the phosphorylation site was changed to Ala in PPC(A), leading to no phosphorylation by PKCa. References 1. D. Ferber, Science 294, 1638 (2001). doi:10.1126/science. 294.5547.1638 2. J.H. Kang, R. Toita, T. Niidome, Y. Katayama, Cancer Lett. 265, 281 (2008). doi:10.1016/j.canlet.2008.02.045 Our polymer consists of a neutral polymer, polyacryl- amide in this case, and a cationic peptide, which is a substrate of PKCa. This polymer forms a complex with DNA through an electrostatic interaction. DNA transcrip- tion from the complex is totally suppressed, probably due to the steric effect of the neutral polymer main chain. However, in target HepG2 cells, in which PKCa is hyperactivated, the pendant substrate peptides become phosphorylated. This event leads to the introduction of phosphate anionic charges into the polymer, thereby attenuating the polymer/DNA interaction. Transcription factors and polymerases are then able to access the DNA, resulting in GFP expression. j 3. M.J. Va¨ha¨-Koskela, J.E. Heikkila¨, A.E. Hinkkanen, Cancer Lett. 3. M.J. Va¨ha¨-Koskela, J.E. Heikkila¨, A.E. Hinkkanen, Cancer Lett. 254, 178 (2007). doi:10.1016/j.canlet.2007.02.002 3. M.J. Va¨ha¨-Koskela, J.E. Heikkila¨, A.E. Hinkkanen 254, 178 (2007). doi:10.1016/j.canlet.2007.02.002 4. Y. Kaneda, Y. Tabata, Cancer Sci. 97, 348 (2006). doi:10.1111/ j.1349-7006.2006.00189.x 5. N. Kasahara, A.M. Dozy, Y.W. Kan, Science 266, 1373 (1994). doi:10.1126/science.7973726 6. C.M. Varga, T.J. Wichham, D.A. Lauffenbruger, Biotechnol. Bioeng. 70, 593 (2000). doi:10.1002/1097-0290(20001220)70: 6\593::AID-BIT1[3.0.CO;2-N 7. H. Hug, T.F. Sarre, Biochem. J. 1291, 329 (1993) 8. A.C. Newton, Curr. Opin. Cell Biol. 9, 161 (1997). doi:10.1016/ S0955-0674(97)80058-0 9. J. Hofmann, Curr. Cancer Drug Targets 4, 125 (2004). doi: 10.2174/1568009043481579 10. C.A. O’Brian, F. Chu, W.G. Bornmann, D.S. Maxwell, Expert Rev Anticancer Ther 6, 175 (2006). doi:10.1586/14737140.6. 2.175 GFP expression was suppressed completely in cells pretreated with PKCa inhibitor (Ro31-7549). Ro-31-7549 is an inhibitor with greatest speciﬁcity toward PKCa ([90% at 5 lM) [16, 17]. These results show that PPC(S) can mediate PKCa-responsive transgene activation. 11. J. Oishi, K. Kawamura, J.H. Kang, K. Kodama, T. Sonoda, M. Murata, T. Niidome, Y. Katayama, J. Control. Release 110, 431 (2006). doi:10.1016/j.jconrel.2005.10.007 12. J.H. Kang, D. Asai, J.H. Kim, T. Mori, R. Toita, T. Tomiyama, Y. Asami, J. Oishi, Y.T. Sato, T. Niidome, B. Jun, H. Nakashima, Y. Katayama, J. Am. Chem. Soc. 130, 14906 (2008). doi: 10.1021/ja805364s Thus we suggest that our polymer/DNA complex can be used for the cell-speciﬁc expression of a transgene responding to PKCa activity, and can be developed into a functional gene regulation system in response to PKCa activation. 13. J.H. Kang, D. Asai, S. Yamada, R. Toita, J. Oishi, T. Mori, T. Niidome, Y. Results and Discussion Ro31-7549, a PKCa inhibitor yl n o r e m yl o P A N D d e k a N Texas Red Phase-contrast PPC(S) contained a Ser phosphorylation site for PKCa, but the phosphorylation site was changed to Ala in PPC(A), leading to no phosphorylation by PKCa. Ro31-7549, a PKCa inhibitor Fig. 5 GFP expression after polymer/DNA complex was microin- jected into HepG2 cells. Naked DNA, polymer only, and polymer/ DNA complexes (C/A = 0.5, 1.0, and 2.0) were microinjected into HepG2 cells and GFP expression was detected 24 h after injection. 123 Nanoscale Res Lett (2009) 4:229–233 233 Grants-in-Aid for Scientiﬁc Research from the Ministry of Education, Science, Sports and Culture of Japan (to D.A., T.N., and Y.K.). phosphorylated by PKCa in cells and gene expression has to be identiﬁed. To prove our concept, the PPC/DNA complexes were microinjected into HepG2 cells. When the PPC(S)/GFP-gene complex was microinjected with C/A ratios of 0.5 to 2.0, signiﬁcant expression of GFP was observed in all cases. However, no GFP expression was found from the negative control PPC(A)/DNA complex- microinjected cells at C/A ratios of 1.0 and 2.0 (Fig. 5). References Katayama, Proteomics 8, 2006 (2008). doi: 10.1002/pmic.200701045 14. L. Ducher, F. Croquet, S. Gil, J. Davy, J. Feger, A. Brehier, Biochem. Biophys. Res. Commun. 217, 546 (1995). doi:10.1006/ bbrc.1995.2810 Acknowledgments We are grateful to Mr Kouki Miyamoto (Fujitsu limited) for his suggestion and technical assistances in the microin- jection study, and to Ms Sigemi Terakubo, Ms Satomi Yamazaki, and Ms Nami Masumoto for their technical assistance. This research was performed with funding support from CREST, Japan Science and Technology Corporation. This work was also supported in part by 15. F. Osaki, Y. Ikeda, T. Suehiro, K. Ota, S. Tsuzura, K. Arii, Y. Kumon, K. Hashimoto, Atherosclerosis 176, 279 (2004) 16. C.T. Murphy, J. Westwick, Biochem. J. 283, 159 (1992) 17. S.E. Wilkinson, P.L. Parker, J.S. Nixon, Biochem. J. 294, 335 (1993) 123 123 12
https://openalex.org/W2462322444	https://europepmc.org/articles/pmc4941518?pdf=render	English	null	Experimental realization of an entanglement access network and secure multi-party computation	Scientific reports	2,016	cc-by	7,509	Experimental realization of an entanglement access network and secure multi-party computation X.-Y. Chang1, D.-L. Deng1,2, X.-X. Yuan1, P.-Y. Hou1, Y.-Y. Huang1 & L.-M. Duan1,2 received: 15 March 2016 accepted: 17 June 2016 Published: 11 July 2016 To construct a quantum network with many end users, it is critical to have a cost-efficient way to distribute entanglement over different network ends. We demonstrate an entanglement access network, where the expensive resource, the entangled photon source at the telecom wavelength and the core communication channel, is shared by many end users. Using this cost-efficient entanglement access network, we report experimental demonstration of a secure multiparty computation protocol, the privacy-preserving secure sum problem, based on the network quantum cryptography. The peculiar quantum correlation, entanglement, provides the crucial resource for quantum information pro- cessing1–3. Generation of remote entanglement is a key step for quantum communication and distributed compu- tation4,5. Distribution of entanglement in a multiple-end quantum network is costly through the point-to-point protocol as one needs to establish an entanglement source and a quantum communication channel between each end4–10. We demonstrate a cost-efficient entanglement access network where multiple end users share the same entanglement source and the core communication channel. We verify entanglement and quantum nonlocality between different end users. Using this entanglement access network, we report the experimental demonstration of a secure multi-party computation protocol where several end users compute cooperatively to solve a joint problem without revealing the information of their actual inputs11–13. This demonstration opens up the prospect of applying the cost-efficient entanglement access network for achieving secure multi-party computation that protects the privacy of end users. With popularity of the internet, distributed computation becomes increasingly more important, where a number of end users need to work cooperatively to solve a common problem. The answer to the problem typ- ically depends on the inputs of all the end users, which need to be communicated in the network. At the same time, the users need to protect their privacy and do not want to reveal their information to the others. Secure multi-party computation is a branch of cryptography and computer science that studies this kind of problems11–13. A well-known primitive example of this field is the millionaire problem, first introduced by Yao11, in which two millionaires want to know which of them is richer without revealing their actual wealth. Secure multi-party com- putation has many applications in e-commerce and data mining where people need to compare numbers which are confidential11–13. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports 1Center for Quantum Information, IIIS, Tsinghua University, Beijing 100084, PR China. 2Department of Physics, University of Michigan, Ann Arbor, Michigan 48109, USA. Correspondence and requests for materials should be addressed to L.-M.D. (email: lmduan@umich.edu) Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 Results The coincidence count rate of the photon pairs is 710 per second for this source. This corresponds to an average pair number of 0.93 × 10−5 per pump pulse registered by the coincidence circuit. The small pair number here includes contribution of the low detection efficiency (~10% for each detector used in our experiment) at the telecom wavelength.hi The entangled photons at the telecom frequency are coupled into coiled optical fibers representing the core communication channel with the total distance about 20 km. The output photons from the fibers are fed into a 1 × 8 optical switch (Dicon MS2-1X8-I2C-15-9/TB-FC/APC-1, with the insertion loss of 0.8 dB max) at each side (called Alice’s and Bob’s side) which is electrically controlled by the computer to deliver the photons to one of the eight output ports according to the electric control signal. This forms an entanglement access network with 8 × 8 possible choices of pairs of end users who can share entangled photons with the entanglement distance more than 20 km, as shown in Fig. 1(b). All the end users share the same entanglement source and the core communi- cation channel. The polarization states of photons are subject to tension and temperature dependent rotations in the optical fibers which are carefully compensated afterwards through the fiber polarization controllers. In our experiment, the polarization entanglement is stable for more than 24 hours under a stable room-temperature environment. In the field experiment with longer communication distance, one may use the polarization locking technique reported in recent experiments to improve the stability of polarization entangled states26.f To demonstrate entanglement between different end users, we randomly choose two users A1 and A2 at Alice’s side and two users B1 and B2 at Bob’s side. For the four pairs of end users AiBj (i, j = 1, 2), we measure their entangle- ment by detecting the photon coincidences in different polarization bases. To rotate the polarization basis, we use a fast switchable electric optical modulator (EOM, Thorlabs EO-AM-NR-C3), which induces a polarization trans- formation \|H〉→cos θ \|H〉 − i sin θ \|V〉 and \|V〉→−i sin θ \|H〉 + cos θ \|V〉 with the angle θ determined by the com- puter controlled electric signal (see the Methods for control of the EOM). The output H and V polarized photons are split by a polarization beam splitter (PBS) and then detected through single photon counters. Results In our experiment, we first report a proof-of-principle demonstration of an entanglement-based network QKD scheme and then use this network QKD setup to realize a multi-party secure-sum computation protocol. In demonstration of the entanglement-based network QKD protocol, we share the same entangled photon source and divide each side into 8 parties with a 1 × 8 optical switch. This makes a 8 × 8 entanglement access network where each party of one side shares entanglement with any party on the other side. By choosing two parties from each side and detecting the photon polarization along three complementary directions, we explicitly demonstrate Bell inequality violation for different pairs of parties and a 4 × 4 entanglement-based QKD network. We then use this 4 × 4 QKD network to realize a four-party secure-sum protocol, where the four agents calculate their total wealth while keeping privacy of their individual wealth. Compared with the QKD network implemented through the BB84 protocol and the wavelength-division multiplexing device23,24, the implementation here has a much lower key exchange rate as the generation rate of entangled photons is much slower than the emission rate of weak coherent pulses used in the BB84 protocol. However, the entanglement-based QKD scheme could offer enhanced security2,17,25,26. In particular, it is a step towards eventual realization of the device-independent quantum cryptog- raphy17–19, the most secure way of communication, although implementation of the latter requires the very chal- lenging condition to close all the loopholes in the Bell inequality test, which is not achieved in this experiment.h The entangled photon source used in this experiment is shown in Fig. 1(a), which is generated by a type-II BBO crystal pumped by ultrafast laser pulses (with the pulse duration less than 150 fs and a pulse repetition rate of 76 MHz) at the wavelength of 775 nm from a Ti:Sapphire laser. The spontaneous parametric down conversion in the BBO crystal produces entangled photon pairs at the telecom wavelength of 1550 nm, which, in the ideal case, are in the polarization entangled state Ψ = + ϕ HV e VH ( )/ 2 i , where \|H〉 and \|V〉 denote respectively the horizontal and the vertical polarization state and ϕ is a controllable phase27. We have verified that the experi- mentally generated state ρ has an entanglement fidelity Fs of (95.12 ± 0.25)% with respect to this ideal state \|Ψ〉 through the quantum state tomography28. Experimental realization of an entanglement access network and secure multi-party computation X.-Y. Chang1, D.-L. Deng1,2, X.-X. Yuan1, P.-Y. Hou1, Y.-Y. Huang1 & L.-M. Duan1,2 Classically, the secure multi-party computation protocols typically rely on computationally hard mathematical problems, which require specific assumptions and are subject to security loopholes in particu- lar under attack by a quantum computer13.f In this paper, we demonstrate an entanglement access network which offers an alternative route for secure multi-party computation based on the network quantum cryptography. Quantum access network is a concept introduced in a recent paper where the expensive quantum resource is shared by many end users8,9. For quantum key distribution (QKD) based on the BB84 protocol14, the photon detector is the relatively expensive part of its implementation and thus shared in the quantum access network demonstrated in ref. 8. Here, we realize an entan- glement access network to efficiently distribute entanglement between network end users. In this network, the entanglement source is the most expensive part of the implementation and thus shared between many end users. Entanglement provides the crucial quantum resource for achieving device-independent quantum cryptography, the most secure way for cryptographic communication15–20. Entanglement is also the critical resource for certified generation of shared randomness21, and for quantum communication and multi-party computation through the entanglement-based schemes3,22. We demonstrate an entanglement access network which efficiently distributes entanglement between a number of end users connected through coiled fibers of more than 20 km length by 1Center for Quantum Information, IIIS, Tsinghua University, Beijing 100084, PR China. 2Department of Physics, University of Michigan, Ann Arbor, Michigan 48109, USA. Correspondence and requests for materials should be addressed to L.-M.D. (email: lmduan@umich.edu) Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 1 www.nature.com/scientificreports/ sharing a single entangled photon source at the telecom frequency. By use of this entanglement access network, we report experimental demonstration of a secure multi-party computation protocol, the secure sum problem, in which several millionaires want to know how much money they have in total but none of them is willing to reveal his wealth to others11,12. This demonstration shows that the entanglement access network provides a cost-efficient way to realize a quantum network with shared resource, opening up the prospect for its applications in secure multi-party cryptography and distributed quantum computation. Results In the Z (or X) basis detection, we register the photon coincidence counts by fixing the angle θA of Ai at 0° (or 45°) and rotating the angle θB of Bj. The resulting coincidence counts as functions of the angle θB are shown in Fig. 2 for different pairs AiBj. The big contrast of these oscillations is a clear demonstration of quantum entanglement. Quantitatively, we calculate the visibilities of these oscillation curves Vz and Vx in the Z and the X basis, respectively, and the entanglement fidelity Fe is then bounded by Fe ≥ (Vz + Vx)/2 (see the Methods for definition of the visibilities and derivation of this criterion29). The visibilities and the corresponding fidelity bounds are listed in Table 1. All the pairs have the entanglement fidelity higher than 90%. The small decrease of the entanglement fidelity compared with the fidelity Fs of the entangled photon source is due to the imperfection in compensation of the polarization rotation in the optical fibers.h i The shared entanglement allows demonstration of quantum key distribution between any pair of the end users in this network using the Ekert protocol2. For this purpose, we need to randomly choose the angle θA from the set {0°, 22.5°, 45°} and θB from the set {22.5°, 45°, 67.5°}, and record the individual measurement outcomes for each coincidence count25. From the counts for the angles θA = {0°, 45°} and θB = {22.5°, 67.5°}, we calculate the expectation value of the Clauser-Horne-Shimony-Holt (CHSH) observable 〈S〉 = 〈0°, 22.5°〉 + 〈45°, 22.5°〉 + 〈45°, 67.5°〉 − 〈0°, 67.5°〉30, where 〈θA, θB〉 denotes the photon coincidence counts with θA, θB at the specified val- ues. The CHSH values are listed in Table 1 for different pairs Ai, Bj. The CHSH inequality 〈S〉 ≤ 2 for any classical Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 2 www.nature.com/scientificreports/ Figure 1. The experimental setup for implementation of an entanglement access network. (a) The setup to generate the polarized entangled photons at the telecom wavelength. The ultrafast pulse at the 775 nm wavelength from the Ti:Sapphire laser is H-polarized and focused by lens onto the BBO crystal cut in the Type-II phase-matching condition, generating polarized entangled photons at the 1550 nm wavelength. Results We use the low density parity check (LDPC) code31,32 to do the error correction for the sifted keys, which is more efficient compared with the conventional CASCADE proto- col25. We obtain error-free shared keys of about 8 × 103 bits, which are then purified by the privacy amplification protocol via a universal-2-class hash function33, yielding shared secret final keys of 1800 bits. The final key rate is about 1.5 bits per second. The residual information available to any potential eavesdroppers is reduced by a factor of 2−300 during the privacy amplification and thus much less than one bit. correlation is clearly violated, demonstrating quantum nonlocality. The significant violation of the CHSH ine- quality for each pair of the end users is a guarantee of the security of the entangled-based QKD protocol15–20. To generate quantum key, the measurement outcomes at θA = θB = {22.5°, 45°} are kept, which yield the sifted keys25. For each pair of the parties, we obtain raw keys of 1.2 × 104 bits with an integration time of 1240 seconds. We ran- domly choose 20% of the keys to estimate the quantum bit error rate (QBER). For our data, the QBER γ is about 5%, smaller than the security threshold of 11%20. We use the low density parity check (LDPC) code31,32 to do the error correction for the sifted keys, which is more efficient compared with the conventional CASCADE proto- col25. We obtain error-free shared keys of about 8 × 103 bits, which are then purified by the privacy amplification protocol via a universal-2-class hash function33, yielding shared secret final keys of 1800 bits. The final key rate is about 1.5 bits per second. The residual information available to any potential eavesdroppers is reduced by a factor of 2−300 during the privacy amplification and thus much less than one bit. g p y pi We now use the entanglement access network to demonstrate a secure multi-party computation protocol: the privacy-preserving secure sum problem12. The problem can be illustrated with the following example: assume several millionaires want to know how much money they have in total, but none of them want to reveal his actual wealth to others. To be concrete, we consider the case of four parties A1, A2, B1, B2. Denote by a1, a2, b1 and b2 the input from A1, A2, B1, B2, respectively. Results To ensure entanglement, we use a compensator made by another BBO crystal at each output, which compensates the temporal walk-off between the H and V polarized photons in the nonlinear crystal. The output photons are filtered by 3 nm interference filters and then coupled into long-distance coiled optical fibers with a fiber coupler. The half-wave plates (HWP) are used for alignment of the polarization axes before and after the fibers. (b) The setup to construct an entanglement access network, where the entanglement source and the core communication channel are shared by many end users. Computer controlled optical switches are used to distribute the entangled photons to different end users. Our experiment demonstrates an entanglement access network with up to 8 × 8 end users, where the two sides are separated by a fiber about 20 km. Among the end users, we have verified entanglement and quantum nonlocality between four pairs of them and used the shared entanglement to demonstrate the four-party secure sum protocol as an example to illustrate its application for implementation of secure multi-party computation. Figure 1. The experimental setup for implementation of an entanglement access network. (a) The setup Figure 1. The experimental setup for implementation of an entanglement access network. (a) The setup to generate the polarized entangled photons at the telecom wavelength. The ultrafast pulse at the 775 nm wavelength from the Ti:Sapphire laser is H-polarized and focused by lens onto the BBO crystal cut in the Type-II phase-matching condition, generating polarized entangled photons at the 1550 nm wavelength. To ensure entanglement, we use a compensator made by another BBO crystal at each output, which compensates the temporal walk-off between the H and V polarized photons in the nonlinear crystal. The output photons are filtered by 3 nm interference filters and then coupled into long-distance coiled optical fibers with a fiber coupler. The half-wave plates (HWP) are used for alignment of the polarization axes before and after the fibers. (b) The setup to construct an entanglement access network, where the entanglement source and the core communication channel are shared by many end users. Computer controlled optical switches are used to distribute the entangled photons to different end users. Our experiment demonstrates an entanglement access network with up to 8 × 8 end users, where the two sides are separated by a fiber about 20 km. Results Among the end users, we have verified entanglement and quantum nonlocality between four pairs of them and used the shared entanglement to demonstrate the four-party secure sum protocol as an example to illustrate its application for implementation of secure multi-party computation. Figure 1. The experimental setup for implementation of an entanglement access network. (a) The setup to generate the polarized entangled photons at the telecom wavelength. The ultrafast pulse at the 775 nm wavelength from the Ti:Sapphire laser is H-polarized and focused by lens onto the BBO crystal cut in the Type-II phase-matching condition, generating polarized entangled photons at the 1550 nm wavelength. To ensure entanglement, we use a compensator made by another BBO crystal at each output, which compensates the temporal walk-off between the H and V polarized photons in the nonlinear crystal. The output photons are filtered by 3 nm interference filters and then coupled into long-distance coiled optical fibers with a fiber coupler. The half-wave plates (HWP) are used for alignment of the polarization axes before and after the fibers. (b) The setup to construct an entanglement access network, where the entanglement source and the core communication channel are shared by many end users. Computer controlled optical switches are used to distribute the entangled photons to different end users. Our experiment demonstrates an entanglement access network with up to 8 × 8 end users, where the two sides are separated by a fiber about 20 km. Among the end users, we have verified entanglement and quantum nonlocality between four pairs of them and used the shared entanglement to demonstrate the four-party secure sum protocol as an example to illustrate its application for implementation of secure multi-party computation. correlation is clearly violated, demonstrating quantum nonlocality. The significant violation of the CHSH ine- quality for each pair of the end users is a guarantee of the security of the entangled-based QKD protocol15–20. To generate quantum key, the measurement outcomes at θA = θB = {22.5°, 45°} are kept, which yield the sifted keys25. For each pair of the parties, we obtain raw keys of 1.2 × 104 bits with an integration time of 1240 seconds. We ran- domly choose 20% of the keys to estimate the quantum bit error rate (QBER). For our data, the QBER γ is about 5%, smaller than the security threshold of 11%20. Step 1.—Using the entanglement-based network QKD, A1 and B1, B1 and A2, A2 and B2, and B2 and A1 share random keys of n bits denoted by a b 1 1  , a b 2 1  , a b 2 2  , and a b 1 2, respectively. The number of bits n is taken to be at least as large as the estimated number of bits for the sum T. Results We want to calculate the sum T = a1 + a2 + b1 + b2 without revealing the inputs a1, a2, b1, b2. The quantum protocol to accomplish this task using the entanglement access network goes as follows: Step 1.—Using the entanglement-based network QKD, A1 and B1, B1 and A2, A2 and B2, and B2 and A1 share random keys of n bits denoted by a b 1 1  , a b 2 1  , a b 2 2  , and a b 1 2, respectively. The number of bits n is taken to be at least as large as the estimated number of bits for the sum T. Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 3 www.nature.com/scientificreports/ Figure 2. The entanglement demonstrated between different pairs of the end users. The figures show the coincidence counts when we fix the rotation angle of the electric optical modulator (EOM) of one end at 0° (for the Z-basis) or 45° (for the X-basis) while rotating the angle of the EOM at the other end. Different sub- figures correspond to different pairs of parties (A1 and B1, A1 and B2, A2 and B1, A2 and B2). The visibilities of the oscillations in the X and the Z bases together bound the entanglement fidelity of the photonic states shared between the corresponding parties. Figure 2. The entanglement demonstrated between different pairs of the end users. The figures show the Figure 2. The entanglement demonstrated between different pairs of the end users. The figures show the coincidence counts when we fix the rotation angle of the electric optical modulator (EOM) of one end at 0° (for the Z-basis) or 45° (for the X-basis) while rotating the angle of the EOM at the other end. Different sub- figures correspond to different pairs of parties (A1 and B1, A1 and B2, A2 and B1, A2 and B2). The visibilities of the oscillations in the X and the Z bases together bound the entanglement fidelity of the photonic states shared between the corresponding parties. Figure 2. The entanglement demonstrated between different pairs of the end users. The figures show the coincidence counts when we fix the rotation angle of the electric optical modulator (EOM) of one end at 0° (for the Z-basis) or 45° (for the X-basis) while rotating the angle of the EOM at the other end. Results At the same time, the inputs a1, a2, b1, b2 remain confidential to the other parties as the public information Xi (i = 1, 2, 3, 4) has no correlation with the inputs due to the one-time pad theorem. For security of this secure sum protocol, we have assumed that dif- ferent parties do not collaborate to steal the input information of the other parties. With the sum T, when three parties collaborate, it is always possible to reveal the input information of the fourth party. When two parties collaborate, whether they can reveal the information of the other two parties depends on whether these two par- ties share a random key in this protocol. If they share a random key (such as A1 and B1), they cannot conspire to reveal the wealth of the other party. For instance, even if A1 and B1 cooperate, they cannot reveal the wealth of A2 or B2 because they do not have the information of a b 2 2  , which is required to read out a2 or b2. On the other hand, if the two parties (such as B1 and B2) do not share a random key, they are able to conspire to reveal the wealth of licly announces X3 to others; B2 calculates = + − X b a b a b 4 2 1 2 2 2   and publicly announces X4 to others. Step 3.—All of them know the sum T by calculating T = X1 + X2 + X3 + X4. At the same time, the inputs a1, a2, b1, b2 remain confidential to the other parties as the public information Xi (i = 1, 2, 3, 4) has no correlation with the inputs due to the one-time pad theorem. For security of this secure sum protocol, we have assumed that dif- ferent parties do not collaborate to steal the input information of the other parties. With the sum T, when three parties collaborate, it is always possible to reveal the input information of the fourth party. When two parties collaborate, whether they can reveal the information of the other two parties depends on whether these two par- ties share a random key in this protocol. If they share a random key (such as A1 and B1), they cannot conspire to reveal the wealth of the other party. Results The experimental data to show entanglement and quantum nonlocality shared between different parties. The table lists the oscillation visibilities of the coincidence counts in the Z and the X bases, whose average gives a lower bound to the entanglement fidelity (the fourth column). The error bars are obtained by assuming a Poissionian distribution for the photon counts and propagated from the measured coincidence counts to the quantities listed in the table through exact Monte Carlo simulation. The last column of the table shows the measured value for the CHSH observable, and a larger-than-2 value indicates violation of the Bell inequality, demonstrating quantum nonlocality shared between the corresponding parties. Step 2.—A1 calculates   = + − X a a b a b 1 1 1 1 1 2 and publicly announces X1 to others; B1 calculates = + − X b a b a b 2 1 2 1 1 1   and publicly announces X2 to others; A2 calculates = + − X a a b a b 3 2 2 2 2 1   and pub- licly announces X3 to others; B2 calculates = + − X b a b a b 4 2 1 2 2 2   and publicly announces X4 to others. Step 3.—All of them know the sum T by calculating T = X1 + X2 + X3 + X4. At the same time, the inputs a1, a2, b b fid l h h h bl f X (i ) h l h Step 2.—A1 calculates   = + − X a a b a b 1 1 1 1 1 2 and publicly announces X1 to others; B1 calculates = + − X b a b a b 2 1 2 1 1 1   and publicly announces X2 to others; A2 calculates = + − X a a b a b 3 2 2 2 2 1   and pub- licly announces X3 to others; B2 calculates = + − X b a b a b 4 2 1 2 2 2   and publicly announces X4 to others. Step 3.—All of them know the sum T by calculating T = X1 + X2 + X3 + X4. Results Different sub- figures correspond to different pairs of parties (A1 and B1, A1 and B2, A2 and B1, A2 and B2). The visibilities of the oscillations in the X and the Z bases together bound the entanglement fidelity of the photonic states shared between the corresponding parties. Z basis visibility X basis visibility Fidelity bound CHSH Value A1B1 (91.68 ± 0.21)% (88.76 ± 0.32)% (90.22 ± 0.26)% 2.428 ± 0.020 A1B2 (92.59 ± 0.19)% (88.11 ± 0.30)% (90.35 ± 0.24)% 2.449 ± 0.019 A2B1 (94.10 ± 0.18)% (87.91 ± 0.36)% (91.01 ± 0.27)% 2.453 ± 0.020 A2B2 (93.54 ± 0.16)% (86.73 ± 0.32)% (90.14 ± 0.24)% 2.458 ± 0.021 Table 1. The experimental data to show entanglement and quantum nonlocality shared between different parties. The table lists the oscillation visibilities of the coincidence counts in the Z and the X bases, whose average gives a lower bound to the entanglement fidelity (the fourth column). The error bars are obtained by assuming a Poissionian distribution for the photon counts and propagated from the measured coincidence counts to the quantities listed in the table through exact Monte Carlo simulation. The last column of the table shows the measured value for the CHSH observable, and a larger-than-2 value indicates violation of the Bell inequality, demonstrating quantum nonlocality shared between the corresponding parties. Z basis visibility X basis visibility Fidelity bound CHSH Value A1B1 (91.68 ± 0.21)% (88.76 ± 0.32)% (90.22 ± 0.26)% 2.428 ± 0.020 A1B2 (92.59 ± 0.19)% (88.11 ± 0.30)% (90.35 ± 0.24)% 2.449 ± 0.019 A2B1 (94.10 ± 0.18)% (87.91 ± 0.36)% (91.01 ± 0.27)% 2.453 ± 0.020 A2B2 (93.54 ± 0.16)% (86.73 ± 0.32)% (90.14 ± 0.24)% 2.458 ± 0.021 Table 1. The experimental data to show entanglement and quantum nonlocality shared between different parties. The table lists the oscillation visibilities of the coincidence counts in the Z and the X bases, whose average gives a lower bound to the entanglement fidelity (the fourth column). The error bars are obtained by assuming a Poissionian distribution for the photon counts and propagated from the measured coincidence counts to the quantities listed in the table through exact Monte Carlo simulation. The last column of the table shows the measured value for the CHSH observable, and a larger-than-2 value indicates violation of the Bell inequality, demonstrating quantum nonlocality shared between the corresponding parties. Table 1. Results We run the above implementation of the secure sum protocol 30 times, and for each time we use different keys a b 1 1  , a b 2 1  , a b 2 2, a b 1 2  with the number of bits n = 25 bits to encode the public information X1, X2, X3, X4. For these 30 experimental runs (each experimental run is an imple- mentation of the full QKD protocol that generates at least 25 bits of final keys a b 1 1, a b 2 1  , a b 2 2  , and a b 1 2  between the corresponding parties), the publicly announced numbers X1, X2, X3, X4 are shown in Fig. 3, which look completely random from trial to trial and reveal no information of the inputs a1, a2, b1, b2. However, for each run, their sum is always fixed to be 4 × 106, which gives the correct calculation result for T = a1 + a2 + b1 + b2. Discussion Similar to the quantum access network realized recently8, we expect that the entanglement access network demonstrated in this paper provides a source-efficient way for network cryptography and secure multi-party computation. In particular, the shared entanglement between each ends of the network opens up the possibility to realize device independent quantum cryptography15–19, which allows most secure communication by closing the security loopholes in conventional quantum cryptography20. Apart from the example demonstrated in this paper, the entanglement access network may allow realization of a number of secure multi-party computation prob- lems13. The entanglement shared in the network could also find applications for certified generation of random numbers21 and for implementation of distributed quantum computation and multiparty quantum cryptography protocols1,3. Results For instance, even if A1 and B1 cooperate, they cannot reveal the wealth of A2 or B2 because they do not have the information of a b 2 2  , which is required to read out a2 or b2. On the other hand, if the two parties (such as B1 and B2) do not share a random key, they are able to conspire to reveal the wealth of Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 4 www.nature.com/scientificreports/ Figure 3. The experimental demonstration of the four-party secure-sum protocol through the entanglement access network. The top four sub-figures show the publically announced data from each of the four parties for 30 rounds of experiments for demonstration of the same secure-sum problem with the same input from each party. The randomly distributed data over different experimental runs is an indication that the announced data reveal no information of the input from each party. The last sub-figure shows the secure sum calculated by each party from the publically announced data, which is identical for different experimental runs and always equal to the true value of the sum for the underlying problem. re 3. The experimental demonstration of the four-party secure-sum protocol through the hi Figure 3. The experimental demonstration of the four-party secure-sum protocol through the entanglement access network. The top four sub-figures show the publically announced data from each of the four parties for 30 rounds of experiments for demonstration of the same secure-sum problem with the same input from each party. The randomly distributed data over different experimental runs is an indication that the announced data reveal no information of the input from each party. The last sub-figure shows the secure sum calculated by each party from the publically announced data, which is identical for different experimental runs and always equal to the true value of the sum for the underlying problem. he other two parties. The above protocol can be extended to more than four parties, and in general we need the ssumption that different parties do not cooperate to simplify the security analysis. f In the experimental demonstration, as an example, we take randomly generated numbers a1 = 55406, b1 = 116559, a2 = 988150 and b2 = 2839885. Methods l f Control of the electric optical modulator. The electro-optic modulator (EOM) used in our experiment is a Pockels cell type modulator consisting of two lithium niobate crystals packaged in a compact housing with an RF input connector. Voltage applied across the crystal structure induces change in the indices of refraction (both ordinary and extraordinary), leading to an electric field dependent birefringence. An optical wave (with Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 5 www.nature.com/scientificreports/ polarization components on both ordinary and extraordinary axes) will experience a change in polarization state after traversing the crystal, from the relative phase delay between these two orthogonal polarization compo- nents. The electro-optic crystal (EOM) thus acts as a wave-plate of variable rotation angle with the angle linearly dependent on the applied voltage. p pp g In our experiment, the applied voltage is controlled by the random numbers generated from a computer pro- gram. The random numbers are fed into a FPGA (filed programmable gate array) board to generate the electric voltage signal. The voltage range from the FPGA board is only from 0–2 V, which is not large enough to induce a polarization rotation of the EOM corresponding to a quarter wave plate operation. The voltage for the latter is required to be 580 V, so we need to apply a voltage amplifier to the output signal of the FPGA board. The switch- ing speed of the whole setup is at 500 Hz, which is limited by the response time of the voltage amplifier. As the recorded coincidence count rate for polarization detection is only about 70 per second at the receiver side, which is significantly less than the switching speed of the EOM device, we have an independent random setting for each pair of the photons that are recorded for quantum keys. Entanglement criterion based on the visibilities. In ref. 29, an entanglement criterion has been derived based on detection of correlations in two complementary bases. Here, we write this criterion into a more compact form in terms of the visibilities. For two correlated photons (or any qubits) detected in two complementary polar- ization bases Z = {\|H〉, \|V〉} and X = {\|+〉, \|−〉}, where ± = ± H V ( )/ 2, ref. References Sy p. o ou dat o s o Co pute Sc e ce ( OCS) p 60 ( 9 12. Sheikh, R., Kumar, B. & Mishra, D. K. Privacy Preserving k-secure sum protocols, International Journal of Computer Science Information Security, ISSN 1947-5500, 6, 2 (2009). 12. Sheikh, R., Kumar, B. & Mishra, D. K. Privacy Preserving k-secure sum protocols, International Information Security, ISSN 1947-5500, 6, 2 (2009). 2. Sheikh, R., Kumar, B. & Mishra, D. K. Privacy Preserving k-secure sum protocols, International Journal of Compu Information Security, ISSN 1947-5500, 6, 2 (2009). y & SahaiSecure, A. Secure Multi-Party Computation. IOS Press (201 . Prabhakaran, M. M. & SahaiSecure, A. Secure Multi-Party Compu 13. Prabhakaran, M. M. & SahaiSecure, A. Secure Multi-Party Computation. IOS Press (2013). y p 14. Bennett, C. H. & Brassard, G. Quantum cryptography: public key distribution and coin tossing. Proceedings of the International Conference on Computers, Systems and Signal Processing, Bangalore, India 175–179 (1984). d f h h l d f l l 15. Mayers, D. & Yao, A. In Proceedings of the 39th Annual Symposium on Foundations of Computer Science, Palo Alto, 1998, p. 503 (IEEE, Washington, DC, 1998). g 16. Barrett, J., Hardy, L. & Kent, A. No signaling and quantum key distribution. Phys. Rev. Lett. 95, 010503 (2005). 17. Acin, A., Brunner, N., Gisin, N., Massar, S., Pironio, S. & Scarani, V. Device-Independent Security of Quantum Cryptography. Rev. Lett. 98, 230501 (2007). 18. Masanes, L., Pironio, S. & Acin, A. Secure device-independent quantum key distribution with causally independent measuremen devices. Nat. Commun. 2, 238 (2011). 19. Vazirani, U. & Vidick, T. Fully Device-Independent Quantum Key Distribution. Phys. Rev. Lett. 113, 140501 (2014)h h y p q y y 21. Pironio, S. et al. Random numbers certified by Bell’s theorem. Nature 464, 1021–1024 (2010). h 21. Pironio, S. et al. Random numbers certified by Bell’s theorem. Nature 464, 1021–1024 (2010). l l k d l l l d d l k h l h 21. Pironio, S. et al. Random numbers certified by Bell’s theorem. Nature 464, 1021–1024 (2010). 22 B tt C H t l T l ti k t t t i d l l i l d Ei t i P d l k R h l Ph R i 22. Bennett, C. H. et al. Teleporting an unknown quantum state via dual classical and Einstein-Podolsky-Rosen channels. Phy Lett. 70, 1895–1899 (1993). Methods l f 29 derived that the entangle- ment fidelity Fe (the overlap with a maximally entangled state) of a mixed state ρ is bounded by ρ ρ ρ ρ ρ ρ ρ ρ ≥ + − + + − − + + −− + − −+ F ( 2 )/2 ( )/2, (1) e H H V V H V V H , , , , , , , , (1) where ρα,β denotes the diagonal matrix elements (correlations) with the first (second) photon in α (β) polariza- tion state. When we detect the visibility in the Z basis, the maximum and the minimum of the correlations are given by ρH,H (ρV,V) and ρH,V (ρV,H), respectively, when the first photon is fixed at H (V) polarization. So we define the average visibility in the Z basis as Vz = (ρH,H + ρV,V − ρH,V − ρV,H)/(ρH,H + ρV,V + ρH,V + ρV,H). The denominator of Vz is simply 1 because of normalization. Similarly, we have Vx = ρ+,+ + ρ−,− − ρ+,− − ρ−,+. Using the inequality that ρ ρ ρ ρ ≤ + 2 H V V H H V V H , , , , , we write the bound for Fe as ≥ + . F V V ( )/2 e z x The entanglement fidelity Fe > 1/2 is a criterion for genuine entanglement29. References 1. Nielsen, M. A. & Chuang, I. L. Quantum Computation and Quantum Information. Cambridge University Press (2000). 2 Ekert A Quantum cryptography based on Bell’s theorem Phys Rev Lett 67 661 663 (1991) yp g p y y 3. Horodecki, R., Horodecki, M. & Horodecki, K. Quantum entanglement. Rev. Mod. Phys. 81, 865–942 (2009). bl Th ( ) g mble, H. J. The Quantum Internet. Nature 453, 1023–1030 (2008). 4. Kimble, H. J. The Quantum Internet. Nature 453, 1023–1030 (2008). h ( ) 5. Duan, L. M. & Monroe, C. Quantum networks with trapped ions. Rev. Mod. Phys. 82, 1209–1224 (2010).i 6. Townsend, P. D. Quantum cryptography on multiuser optical fibre networks. Nature 385, 47–49 (1997). 7. Ursin, R. et al. Entanglement-based quantum communication o 7. Ursin, R. et al. Entanglement-based quantum communication over 144 km. Nat. Phys. 3, 481–486 (2007). g q y 8. Frolich, B., Dynes, J. F., Lucamarini, M., Sharpe, A. W., Yuan, Z.-L. & Shields, A. J. A quantum access network. Nature 501, 69–72 (2013). 9 U i R & H h R Q i f i Sh i N 501 37 38 (2013) 9. Ursin, R. & Hughes, R. Quantum information: Sharing quantum secrets. Nature 501, 37–38 (2013). . Ursin, R. & Hughes, R. Quantum information: Sharing quantum g g 0. Hughes, R. J. et al. Network-centric quantum communications with application to critical infrastructure protection. Preprint a http://arxiv.org/abs/1305.0305 (2013). 11. Yao, A. C. Protocols for secure computations. Proc. 23rd Annu. Symp. on Foundations of Computer Science (FOCS) p 160 (1982). 12 Sh ikh R K B & Mi h D K P i P i k l I i l J l f C S i d 11. Yao, A. C. Protocols for secure computations. Proc. 23rd Annu 11. Yao, A. C. Protocols for secure computations. Proc. 23rd Annu. Symp. on Foundations of Computer Science (FOCS) p 160 (1982). 12. Sheikh, R., Kumar, B. & Mishra, D. K. Privacy Preserving k-secure sum protocols, International Journal of Computer Science and 1. Yao, A. C. Protocols for secure computations. Proc. 23rd Annu. Symp. on Foundations of Computer Science (FOCS) p 160 (1982). 2. Sheikh, R., Kumar, B. & Mishra, D. K. Privacy Preserving k-secure sum protocols, International Journal of Computer Science and Information Security ISSN 1947-5500 6 2 (2009) . ao, . C. otoco s o secu e co putat o s. oc. 3 d u. References 23. Sasaki, M. et al. Field test of quantum key distribution in the Tokyo QKD Network. Optics Express 19,10387–10409 (2011). l l k d b f b d l h d l l k l d h 24. Patel, K. A. et al. Quantum key distribution for 10 Gb/s dense wavelength division multiplexing networks. Applied Physics L 104, 051123 (2014). 5. Jennewein, T., Simon, C., Weihs, G., Weinfurter, H. & Zeilinger, A. Quantum Cryptography with Entangled Photons. Phys. Rev. Lett 84, 4729–4732 (2000).i 6. Honjo, T. et al. Long-distance entanglement-based quantum key distribution over optical fiber. Optics Express 16,19118–19126 (2008). ( ) 27. Kwiat, P. G., Mattle, K., Weinfurter, H., Zeilinger, A., Sergienko, A. V. & Shih, Y. H. New high-intensity source of polarization- entangled photon pairs. Phys. Rev. Lett. 75, 4337–4341 (1995). Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 6 www.nature.com/scientificreports/ 8. James, D. F. V., Kwiat, P. G., Munro, W. J. & White, A. G. Measurement of qubits. Phys. Rev. A 64, 052312–052326 (2001). q y 9. Blinov, B. B., Moehring, D. L., Duan, L.-M. & Monroe, C. Observation of entanglement between a single trapped atom and a single ion. Nature 428, 153–157 (2004). 0. Clauser, J. F., Horne, M. A., Shimony, A. & Holt, R. A. Proposed experiment to test local hidden-variable theories. Phys. Rev. Lett. 23 880–884 (1969).h ( ) 1. Gallager, R. G. Low-density parity-check codes. IEEE Trans. Inf. Theory IT-8, 21–28 (1962). ( ) 31. Gallager, R. G. Low-density parity-check codes. IEEE Trans. Inf. Theory IT-8, 21–28 (1962). h d d d bl f b h 2. Dixon, A. R. & Sato, H. High speed and adaptable error correction for megabit/s rate quantum key distribution. Sci. Rep. 4, 7275 (2014). (2014). 33. Carter, J. L. & Wegman, M. N. Universal classes of hash functions. J. Comput. Sys. Sci. 18, 143–154 (1979). 33. Carter, J. L. & Wegman, M. N. Universal classes of hash functions. J. Comput. Sys. Sci. 18, 143–154 (1979). Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 Acknowledgementsh g This work was supported by the National Basic Research Program of China 2011CBA00302. LMD and DLD acknowledge in addition support from the IARPA MUSIQC program, the AFOSR and the ARO MURI program. Author Contributions L.-M.D. and D.-L.D. proposed the experiment. X.-Y.C., X.-X.Y., P.-Y.H. and Y.-Y.H. carried out the experiment under L.-M.D.’s supervision. D.-L.D. and X.-Y.C. analyzed the data. L.-M.D., X.-Y.C. and D.-L.D. wrote the manuscript. L.-M.D. and D.-L.D. proposed the experiment. X.-Y.C., X.-X.Y., P.-Y.H. and Y.-Y.H. carried out the experiment under L.-M.D.’s supervision. D.-L.D. and X.-Y.C. analyzed the data. L.-M.D., X.-Y.C. and D.-L.D. wrote the manuscript. Additional Informationi Competing financial interests: The authors declare no competing financial interests. How to cite this article: Chang, X.-Y. et al. Experimental realization of an entanglement access network and secure multi-party computation. Sci. Rep. 6, 29453; doi: 10.1038/srep29453 (2016). How to cite this article: Chang, X.-Y. et al. Experimental realization of an entanglement access network and secure multi-party computation. Sci. Rep. 6, 29453; doi: 10.1038/srep29453 (2016). 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/ 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/ Scientific Reports \| 6:29453 \| DOI: 10.1038/srep29453 7
https://openalex.org/W2285025972	https://inldigitallibrary.inl.gov/sites/sti/sti/5806466.pdf	English	null	Final Report - Assessment of Testing Options for the NTR at the INL	null	2,013	public-domain	12,111	INL/EXT-13-28526 INL/EXT-13-28526 INL/EXT-13-28526 Final Report – Assessment of Testing Options for the NTR at the INL Steven D. Howe Travis McLing Michael McCurry Mitchell Plummer Final Report – Assessment of Testing Options for the NTR at the INL Steven D. Howe Travis McLing Michael McCurry Mitchell Plummer February 2013 Steven D. Howe Travis McLing Michael McCurry Mitchell Plummer February 2013 The INL is a U.S. Department of Energy National Laboratory operated by Battelle Energy Alliance INL/EXT-13-28526 INL/EXT-13-28526 Final Report - Assessment of Testing Options for the NTR at the INL Steven D. Howe, Travis McLing, Michael McCurry, Mitchell Plummer Executive Summary One of the main technologies that can be developed to dramatically enhance the human exploration of space is the nuclear thermal rocket (NTR). Several studies over the past thirty years have shown that the NTR can reduce the cost of a lunar outpost, reduce the risk of a human mission to Mars, enable fast transits for most missions throughout the solar system, and reduce the cost and time for robotic probes to deep space. Three separate committees of the National Research Council of the National Academy of Sciences have recommended that NASA develop the NTR. One of the primary issues in development of the NTR is the ability to verify a flight ready unit. Three main methods can be used to validate safe operation of a NTR, after which a fully complete engine could be launched into orbit for the first full power test. The NTR testing options include Three main methods can be used to validate safe operation of a NTR, after which a fully complete engine could be launched into orbit for the first full power test. The NTR testing options include 1) Full power, full duration test in an above ground facility that scrubs the rocket exhaust clean of any fission products; 2) Full power , full duration test using the Subsurface Active Filtering of Exhaust (SAFE) technique to capture the exhaust in subsurface strata; 3) Test of the reactor fuel at temperature and power density in a driver reactor with subsequent first test of the fully integrated NTR in space. The first method, the above ground facility, has been studied in the past [Rocketdyne, INL, MSFC]. The second method, SAFE, has been examined for application at the Nevada Test Site. The third method relies on the fact that the Nuclear Furnace series of tests in 1971 showed that the radioactive exhaust coming from graphite based fuel for the NTR could be completely scrubbed of fission products and the clean hydrogen flared into the atmosphere. Under funding from the MSFC, the Center for Space Nuclear Research (CSNR) at the Idaho National laboratory (INL) has completed a reexamination of Methods 2 and 3 for implementation at the INL site. Introduction One of the main technologies that can be developed to dramatically enhance the human exploration of space is the nuclear thermal rocket (NTR). Several studies over the past thirty years [1-4] have shown that the NTR can reduce the cost of a lunar outpost, reduce the risk of a human mission to Mars, enable fast transits for most missions throughout the solar system, and reduce the cost and time for robotic probes to deep space. Three separate committees of the National Research Council [5-7] of the National Academy of Sciences have recommended that NASA develop the NTR. One of the primary issues in development of the NTR is the ability to verify a flight ready unit. Three main methods can be used to validate safe operation of a NTR: Three main methods can be used to validate safe operation of a NTR: 1) Full power, full duration test in an above ground facility that scrubs the rocket exhaust clean of any fission products; 2) Full power , full duration test using the Subsurface Active Filtering of Exhaust (SAFE) technique to capture the exhaust in subsurface strata; 3) Test of the reactor fuel at temperature and power density in a driver reactor with subsequent first test of the fully integrated NTR in space. The first method, the above ground facility, has been studied in the past [Rocketdyne, INL, MSFC]. The primary barriers are the need for 1) large operations staff, 2) long term operation to maintain capability, and 3) generation of large masses of potentially radioactive waste from the filtration system. The previous studies estimated the cost of the test facility to be between $200- 500 M depending upon the specific size of the NTR. This cost was seen as an up-front expense before any testing could be achieved. Consequently, the method has been seen as a major obstacle in time of decreasing government budgets. The second method, SAFE, has been examined for application at the Nevada Test Site [Howe 99, DRI 2007, Howe 2009-11]. The results of these studies showed that an eight foot diameter hole around 1200 feet deep at the NTS would produce a back pressure on the rocket of around 35 psig. The results indicate that any size engine could be operated for any length of time unlike Method 1 which had to operate at only one power level. Executive Summary In short, the effort performed the following: Under funding from the MSFC, the Center for Space Nuclear Research (CSNR) at the Idaho National laboratory (INL) has completed a reexamination of Methods 2 and 3 for implementation at the INL site. In short, the effort performed the following: x Assess the geology of the INL site and determine a location suitable SAFE testing; x Perform calculations of gas transport throughout the geology; x Produce a cost estimate of a non-nuclear , sub-scale test using gas injection to validate the computational models; x Produce a preliminary cost estimate to build a nuclear furnace equivalent facility to test NTR fuel on a green field location on the INL site. Reexamination of Method 3, involving test of the reactor fuel in a driver reactor indicated that a new Category I facility would be required, which would cost in excess of $250M. Reexamination of Method 2, showed that the INL geology is substantially better suited to the SAFE testing method than the NTS site. The existence of impermeable interbeds just above the sub-surface aquifer ensure that no material from the test, radioactive or not, can enter the water table. Similar beds located just below the surface will prevent any gaseous products from reaching the surface for dispersion. The extremely high permeability of the basalt strata between the interbeds should allow rapid dispersion of the rocket exhaust. In addition, the high permeability suggests that a lower back-pressure may develop in the hole against the rocket thrust, which increases safety of operations. Finally, the cost of performing a sub-scale, non-nuclear verification experiment was determined to be $2,100 K a significant savings over the NTS. Based on the results of this study, a cost estimate for testing a nuclear rocket at the INL site appears to be warranted. Given the fact that a new nuclear fuel may be possible that does not release any fission products, the SAFE testing option appears to be the most affordable. Introduction The estimate of the cost for testing a NTR at NTS was $45 M. The third method relies on the fact that the Nuclear Furnace series of tests in 1971 showed that the radioactive exhaust coming from graphite based fuel for the NTR could be completely scrubbed of fission products and the clean hydrogen flared into the atmosphere. Conceptually, a similar system could be built today to qualify the NTR fuel performance. Then a fully complete engine would be launched into orbit for the first full power test. Under funding from the MSFC, the Center for Space Nuclear Research (CSNR) at the Idaho National laboratory (INL) has completed a reexamination of Methods 2 and 3 for implementation at the INL site. In short, the effort performed the following: x Assess the geology of the INL site and determine a location suitable SAFE testing; x Perform calculations of gas transport for the INL stratigraphy; x Produce a cost estimate of a non-nuclear , sub-scale test using gas injection to validate the computational models; x Produce a preliminary cost estimate to build a nuclear furnace equivalent facility to test NTR fuel on a green field location on the INL site. SAFE Concept The basis of the SAFE concept relies on the porosity of sub-surface strata to transport the exhaust gases and to act as a filter. In essence, the concept proposes to put the nuclear rocket at the top of a hole that has been sealed as depicted in Figure 1. As the rocket fires the effluent into the hole, pressure will build. Eventually the pressure will reach a level where the amount of gas and water vapor driven into the porous rock equals the mass flow of the rocket. Consequently, for a porous matrix of great horizontal extent, the rocket can be operated for long periods over a relatively wide range of power levels. Thus, the requirements of the engine may be determined at a later stage in the program – as the constraints imposed by the capacity of the testing facility are not the primary limitation. A set of calculations using the WAFE code to model the SAFE concept were made in 1999 [2003] for the NTS geology. WAFE is a 2-D model of water, water vapor and non-condensible gas flow and energy transport in permeable soil and rock materials. It was developed initially for the underground nuclear weapons testing program to estimate transient pressure, temperature, and water saturation changes in stemming columns and geologic units surrounding a hot pressurized cavity produced by a nuclear test. Those simulations modeled a vertical borehole with a diameter of 2.4 m, extending to a depth of 360 m, typical of emplacement holes at the NTS. The upper 30 m of the hole is lined with a steel casing. The earth surrounding the hole is alluvium, uniform in properties. Typical values for relevant properties of alluvium at the NTS are a porosity of 35%, a permeability of 8 darcys, and initial pore water saturation of 30%, at a temperature of 20°C. Simulations start with injection of the exhaust gases (H2O and H2) into the borehole at the bottom of the steel liner. Two cases were considered: 100% thrust, and 30% thrust. For the 100% thrust case, a total of 73.4 kg/s of H2O (17.4 kg/s from the engine exhaust plus 56 kg/s of cooling spray) and 0.64 kg/s of excess H2 were injected. SAFE Concept For the 30% thrust case, a total of 20.5 kg/s of H2O (4.9 kg/s from the engine exhaust plus 15.6 kg/s of cooling spray) and 0.33 kg/s of excess H2 were injected. In both cases, injection temperature was assumed to be 600 C. FIGURE 1. Artist’s Concept of the SAFE Configuration. FIGURE 1. Artist’s Concept of the SAFE Configuration. The cooling spray added at the top of the borehole is a necessary feature; otherwise, borehole temperatures would be over 3000 C and would damage or melt the steel casing and cause major chemical changes in the alluvium. Further, other simulations indicated that borehole pressure rise would be considerably higher without the water spray. Results of the simulations showed the pressure rise in the borehole at the mid-depth level. In both cases, the pressure rise exhibits an initial spike of a few psi, which subsides, followed by a more gradual rise. In the 100% thrust case, after 2 hours, the pressure has risen to about 36 psia, and to about 21 psia in the 30% thrust case. The rate of pressure increase is diminishing in both cases with time, as the rate of flow into the surrounding soil increases. Results of the simulations showed the pressure rise in the borehole at the mid-depth level. In both cases, the pressure rise exhibits an initial spike of a few psi, which subsides, followed by a more gradual rise. In the 100% thrust case, after 2 hours, the pressure has risen to about 36 psia, and to about 21 psia in the 30% thrust case. The rate of pressure increase is diminishing in both cases with time, as the rate of flow into the surrounding soil increases. Cooper and Decker (2011) also conducted numerical simulations of the SAFE concept applied at the Nevada Test Site. Based on the permeability range of the stratigraphy there, they determined that a thickness of 100 m would be required to maintain a backpressure of less than their design criterion of 0.24 MPa (35 psi) for the 30% thrust test case. In their simulation, the injected gas consisted of 14.5 kg s-1 hydrogen gas and 15 kg s-1 water vapor, with the hydrogen gas modeled as air. The results of those previous studies are used as comparison for the calculations in this study. Gas Transport Analysis To evaluate the potential for a successful SAFE test at the INL, we conducted gas transport calculations to assess how gas pressures in the vadose zone might respond to a hypothesized rocket exhaust injection test. The conceptual model considered in these calculations is based on a the typical stratigraphy of the vadose zone at the INL (Figure 2), that includes extensive horizontal layers of low-permeability sedimentary deposits separating extensive thicknesses of fractured basalt. In an NTR test at the INL, exhaust would be injected into the fractured basalt between sedimentary layers to limit exhaust flow to the surface or to the underlying aquifer. Detailed discussion of the geology of the site, constraints on possible locations for such a test, and maps of areas that meet the necessary criteria are included in a subsequent geology discussion. Figure 2. Schematic illustrating typical geologic stratigraphy at the INL, that includes horizontally extensive layers of fine sediments separating extensive thicknesses of fractured basalt. Figure 2. Schematic illustrating typical geologic stratigraphy at the INL, that includes horizontally extensive layers of fine sediments separating extensive thicknesses of fractured basalt. Primary criteria for the proposed exhaust injection test, relative to gas transport are x backpressure at the injection point be limited to approximately 0.24 MPa (35 ps x gas velocities should be restricted to a Mach number of less than 1.0, in order to minimize backpressure and allow treatment of the flow problem using weakly compressible flow equations, x induced pressure increase within the borehole should dissipate into the subsurface within a reasonable time after cessation of the injection, and x induced pressure increase within the borehole should dissipate into the subsurface within a reasonable time after cessation of the injection, and x hydrogen injected will not be released to the atmosphere in such a way as to allow inadvertent combustion of the escaped gas. x hydrogen injected will not be released to the atmosphere in such a way as to allow inadvertent combustion of the escaped gas. In our examination of gas transport issues for this assessment, we consider primarily the first two of these criteria because the latter criteria are generally readily met if the forced flow can be accommodated and because the latter criteria depend on continuity of overlying low- permeability layers discussed subsequently in the geology section of this report. Gas Transport Analysis Backpressure develops in the subsurface exhaust injection scheme because of resistance in the well as well as resistance in the permeable porous medium targeted for injection. In the borehole, the resistance to flow depends on the radius of the hole and flow regime that develops under the specified injection rate, as the resistance to flow increases with increasing turbulence. In addition, at very high velocities, the backpressures that develop at locations of contrasting resistance result in changes in fluid density that also depends on pressure. Under the conditions specified for the Cooper and Decker test (2011), the combined volumetric gas flow of water and hydrogen is approximately 580 m3 s-1. The Mach number criterion thus requires that the radius of the borehole be approximately 0.61 m or greater. For any reasonable borehole dimensions, however, the flow regime will be turbulent, and significant backpressures can develop because of the resulting resistance. Assuming a 1.2-m radius borehole, the value used for similar calculations at the NTS, with an intrinsic friction length of 1 mm, the pressure gradient that develops in the borehole is approximately 3 Pa m-1, indicating that if the radius is set to maintain flow in a regime where the gas may be treated as a nearly incompressible fluid, the backpressure associated with the borehole will be insignificant relative to that developed via flow into surrounding rock. In the permeable porous medium, the resistance to flow is a function of the intrinsic permeability of the medium, the presence of other fluids within it, and – again – the flow regime that develops. In the Cooper and Decker (2011) simulations, the intrinsic permeability of the target formations was estimated as 10-11 m2. The transmissivity for the 490-m thickness required to meet their design requirements was thus 4.9 x 10-9 m3. At the INL, which is underlain by extensive basalt flows, permeability of the subsurface is primarily a result of the secondary porosity created by the fracture networks. Fractures are excellent conduits for fluid flow, and the permeability of a fracture varies as the square of the aperture. Thus, one small-aperture fracture can provide the same transmissivity as a much greater thickness of permeable porous medium. Gas Transport Analysis Figure 3, for example, illustrates the intrinsic transmissivity associated with a fracture of varying aperture, from which we can see that 10 horizontal continuous fractures of approximately 2 mm aperture could provide equivalent transmissivity to the entire 490-m thickness of permeable porous medium considered in the Cooper and Decker analysis for the NTS. Figure 3. Transmissivity of a unit of rock in which fluid is transported through a single fracture of aperture specified o the abcissa. Red lines illustrate aperture for a set of 10 horizontal fractures with transmissivity matching that for whic Cooper and Decker demonstrated that injection back pressure in a 1-m radius well would not exceed the desig requirement of 0 24 MPa 0.01 0.1 1 10 1 10 17 u 1 10 16 u 1 10 15 u 1 10 14 u 1 10 13 u 1 10 12 u 1 10 11 u 1 10 10 u 1 10 9 u 1 10 8 u Fracture aperture [mm] Transmissivity [m^3] 4.9 10 10 b 10 mm Figure 3. Transmissivity of a unit of rock in which fluid is transported through a single fracture of aperture specified on the abcissa. Red lines illustrate aperture for a set of 10 horizontal fractures with transmissivity matching that for which Cooper and Decker demonstrated that injection back pressure in a 1-m radius well would not exceed the design requirement of 0.24 MPa. 0.01 0.1 1 10 1 10 17 u 1 10 16 u 1 10 15 u 1 10 14 u 1 10 13 u 1 10 12 u 1 10 11 u 1 10 10 u 1 10 9 u 1 10 8 u Fracture aperture [mm] Transmissivity [m^3] 4.9 10 10 b 10 mm Fracture aperture [mm] Figure 3. Transmissivity of a unit of rock in which fluid is transported through a single fracture of aperture specified on the abcissa. Red lines illustrate aperture for a set of 10 horizontal fractures with transmissivity matching that for which Cooper and Decker demonstrated that injection back pressure in a 1-m radius well would not exceed the design requirement of 0.24 MPa. Gas Transport Analysis Permeability of a volume of fractured rock can thus be considerably higher than for an equivalent thickness of permeable porous medium and a variety of sources indicate that much of the basalts underlying the INL demonstrate this characteristic. Permeability of a volume of fractured rock can thus be considerably higher than for an equivalent thickness of permeable porous medium and a variety of sources indicate that much of the basalts underlying the INL demonstrate this characteristic. Much of the data describing permeability in the basalts underlying the INL stems from the numerous analyses aimed at understanding flow and transport in the prolific Snake River Plain aquifer, the major source of water for eastern Idaho. Ackerman (1991) indicates that permeability measurements from single-well aquifer tests at the INL Site range from less than 1x10-12 m2 to approximately 1.7x10-9 m2. Inverse model derived estimates of intrinsic permeability in the saturated zone at the INL range from 1x10-12 m2 to 4x10-9 m2, and while such measurements are, in part, constrained to match field data such as that described by Ackerman, the calculated value are effectively based on other field data, including abundant water level measurements throughout the Snake River Plain. Anderson, Kuntz, and Davis (1999), for example, reported that hydraulic conductivity in the vicinity of the INL Site ranged from less than 0.01 to more than 7,000 m/d (1x10-14 m2 to 7x10-9 m2), with most estimates exceeding 30 m/d (3.5x10-11 m2). Permeabilities estimated from the transmissivity distribution used in groundwater models of the Snake River Plain aquifer of Robertson (1974) range from 5x10-10 m2 to 2.1x10-9 m2. Similarly, Garabedian’s (1992) calibrated distribution of hydraulic conductivity within the area of the OU 10-08 model domain ranged from 1 to 335 m/d (1x10-12 m2 to 3.35x10- 10 m2). The above described permeability estimates strongly indicate that the permeability of the fractured basalt underlying the INL is frequently high relative to the 10-11 m2 value used by Cooper and Decker for the NTS. If sufficient area with permeabilities toward the higher end of the reported range for the aquifer (1x10-10 m2 to 1x10-9 m2) can be found in the unsaturated zone, the required thickness could theoretically be one to two orders of magnitude less than required for the NTS. Several sources of data provide evidence that the vadose zone at the INL is also highly permeable. Gas Transport Analysis Well logs from numerous water wells provide direct evidence that numerous zones of highly fractured basalt extend across large areas of the site (reference?). Anecdotal evidence from air rotary drilling through the basalt also indicates that that the rock is highly permeable. Magnuson and Sondrup (2006), for example, indicate that before 1994, more than 40 wells were drilled without reverse-air circulation and in most of those wells “circulation (i.e., air recovery) was partially or totally lost below about 18 m. These wells were typically drilled using air pressures of 125 to 250 psi and injection rates of 0.35 to 0.52 m3 s-1. Finally, Mudra and Schmalz (1965) demonstrated high vadose zone permeability during gas injection testing completed at the INL as an appraisal of gaseous waste disposal potential. Where it is desired to constrain the exhaust injection between two of the find-grained, low- permeability, interbeds that extend across large areas of the site, the thickness of fractured basalt available is likely to be at least 150 feet. Assuming that thickness, and a permeability of 2x10-9 m2 (considerably less than the highest reported permeabilities at the site), we can make a preliminary estimate of the backpressure that would develop via the Theis solution to the fluid flow problem to a well. The Theis solution gives the fluid pressure as a function of time and radial distance for a fully penetrating well in a transmissive medium of infinite radial extent. The solution assumes constant transmissivity and relatively small, and constant, compressibility, and the latter is a reasonable approximation for the gas flow if condensation and temperature effects are neglected, because the compressibility, and its pressure dependence, over the pressure difference prescribed by the design criterion, is minimal. The likely effects of heat transfer on the calculated backpressure will be discussed subsequently. For the specified flow rate and well radius of 1.2 m, a transmissivity of 9.1 x 10-8 m3, an effective fluid viscosity of 1.9 x 10-5 Pa s, and a compressibility of 4.9 x 10-6 Pa-1, the pressure vs radial distance profile that develops after 2 hours is shown in Figure 4. The absolute pressure at the exhaust injection point is 0.25 MPa, (gage pressure = 0.150 MPa), indicating that if the thermal effects associated with the injection (including condensation) do not have a negative feedback on the pressure gradient, the assumed conditions could meet the desired design criterion. Gas Transport Analysis Cooling of the gas during transport through the subsurface will have two competing effects on the gas pressure in the injection well. First, cooling will directly reduce the pressure along the flow path according to the relationship described by the ideal gas law and the resultant water vapor condensation will reduce pressure because it effectively removes the steam portion of the gas flow. The former effect could effectively reduce the pressure by a factor of approximately 3, while condensation would effectively remove approximately 10% of the volumetric flux. Second, condensation of water in the pores will reduce the cross-sectional area of the subsurface available for gas flow. Because this reduces fluid permeability, the condensation effect should also cause some increase in the backpressure that develops. As an indicator of the relative importance of these competing effects, we note that Cooper and Decker’s (2011) simulations for such a test at the NTS suggested that the simulated conditions could meet the backpressure design criterion with relative saturations in the water condensation zone of up to 80%, which would have likely reduced permeability to on the order of 1% of their intrinsic permeabilities. Given that permeabilities in the fractured rock at INL are generally much larger than in the NTS subsurface, the NTS simulations strongly suggest the desired maximum injection backpressure could also accommodate subsurface condensation effects at the INL. In addition, because the porosity of fractured basalt is considerably lower than the medium considered in the NTS calculations, the condensation zone is likely to penetrate much farther radially but cause similar changes in relative permeability. The effect of those changes should thus be smaller in a fractured system, because the pressure gradient decreases with radial distance from the injection point and the change in backpressure is proportional to the product of the pressure gradient and relative change in permeability. Figure 4. Pressure vs radial distance for the gas flow and well and flow conditions described in the text, calculated using the Theis solution for radial flow to a well in an infinite aquifer. 20 40 60 80 100 0.16 0.18 0.2 0.22 0.24 0.26 Radius (m) Pressure (MPa) Figure 4. Pressure vs radial distance for the gas flow and well and flow conditions described in the text, calculated using the Theis solution for radial flow to a well in an infinite aquifer. Gas Transport Analysis While permeability data from the Snake River Plain indicates that a zone of sufficiently high permeability for the exhaust injection test should be available at the INL, the porosity of the rock also plays an important role in defining site suitability. While a single fracture with relatively large aperture could theoretically provide the required transmissivity, the gas velocities in such a system would be extremely high, particularly at the borehole face, and the radial penetration would likely extend beyond distances over which confining units could be assumed to continuous. Higher velocities can also be disadvantageous by inducing turbulence, which increases the pressure gradient needed to drive flow through the system of fractures. If we consider a fracture network comprised of, for illustration, equally spaced, uniform, horizontal fractures of infinite extent that could accommodate the required gas flux under nearly incompressible laminar flow, the number of fractures required, and thus the secondary porosity, can be determined by constraining the Mach number to less than 0.3. The required number of fractures is then ~700, and the implied aperture and spacing are, respectively, 1.0 mm and 6 cm and the implied porosity is 0.015. For comparison, a recent large-scale modeling study of the Snake River Plain aquifer underlying the INL used a porosity of 0.062, a value derived via calibration and that is consistent with values used in previous studies. Ackerman et al. (2006), for example, reported a range of porosity estimates from 0.05 to 0.27 derived from previous studies. Considering the volumetric gas flux of 580 m3 s-1 injected into such a system, and the >40% volumetric reduction expected to occur during transport, the radial penetration distance at the end of a 2-hour test would be less than 600 meters in a uniform homogeneous system. In summary, gas flow calculations suggest that the primary design criterion for the SAFE exhaust injection test could be met at the INL, in a zone of fractured rock approximately 150 ft thick with laterally continuous permeability near the upper end of permeabilities measured at multiple locations in the Snake River Plain. Site Selection Criteria Based upon discussions within the project research group, the following specific site selection criteria have been established as geologically relevant NRT site requirements at INL: x 1Vadose zone is required to have at least 450 feet depth below land surface (DBLS); i.e. aquifer is required to be at >450 feet below land surface x Located > ~2 km from site boundary or major roads/highways x Located > 1 km from known contamination sites (e.g., RWMC, INTEC) and developed INL roads x Avoid regions of active seismicity or geothermal activity x Avoid regions of potential flooding x Avoid regions that would provide structural pathways for vertical flow (e.g., faults, dikes, vents or volcanic rifts) x Avoid regions that would provide structural pathways for vertical flow (e.g., faults, dikes, vents or volcanic rifts) x Located > 1 km from known or potential sources of focused surface infiltration (e.g., Big Lost River; diversion ponds; playas; other regions of known rapid infiltration); anthropogenic wastewater recharge areas; and radiologically contaminated vadose zone or groundwater systems x Permeable basalt interval thickness of ≥ ~300 feet ('test interval') x Aquitards2 (sedimentary interbeds) located both below and above fractured basalt test interval. INL Site Geological Assessment The goal of this portion of the report is to provide a high level assessment of the geologic conditions present at the INL that would support the SAFE concept for nuclear propulsion test facility. It provides a summary and analysis of the geologic architecture of the Idaho National laboratory (INL) relative to the needs of the test parameters provided by USRA and intended to provide a rational basis for assessment of possible site locations for field tests of SAFE Nuclear Rocket technology at the Idaho National Laboratory. The identified regions would then be subject to more detailed examination as potential field test sites. This assessment is based on a review of publicly accessible information regarding the geologic framework and hydrogeology of the shallow subsurface for the Idaho National Laboratory, emphasizing physical and chemical characteristics of the earth materials. Information sources include professional scientific publications, and relevant public-domain reports compiled from the DOE/INL (e.g., htttp:www.inl.gov/publications/), US Geological Survey, and State of Idaho. The assessment begins by listing geologically relevant site selection criteria that were established by project team members. It then summarizes key features of the geology and hydrogeology of INL. These features are then evaluated in the context of the site selection criteria. Parts of INL that satisfy those criteria are identified. 1 Vadose zone is defined at the unsaturated zone that exists above a aquifer. In the case of the INL, the vadose zone is very thick - 200’ to 700’ - and is composed of layered basalt and sedimentary interbeds. 2 Aquitards are impermeable layers of sedimentary deposits that restrict the flow of water and gas. Location The Idaho National Laboratory (INL) is located on the northern margin of the Eastern Snake River Plain (ESRP) in southeast Idaho (Figure 5). It is bordered to the north by mountains and The Idaho National Laboratory (INL) is located on the northern margin of the Eastern Snake i l i ( ) i h d h ( i ) i b d d h h b i d The Idaho National Laboratory (INL) is located on the northern margin of the Eastern Snake River Plain (ESRP) in southeast Idaho (Figure 5). It is bordered to the north by mountains and valleys of the northern Basin and Range province (e.g., Anders et al. 1989). Drainage is internal and derived mainly from the north via the Big Lost River, Little Lost River and Birch Creek, all of which converge into a broad sedimentary basin referred to as the Big Lost Trough (Geslin et al. 2002). Modern playa systems developed within this long-lived trough include the Big Lost River and Birch Creek Sinks. Underflow and infiltration have produced a robust open groundwater system referred to as the Eastern Snake River Plain aquifer (e.g., Garabedian 1992). Figure 5. Location and digital elevation map of INL area illustrating geographic features, boundary of INL, and s within INL. Figure 5. Location and digital elevation map of INL area illustrating geographic features, boundary of INL, and sites within INL. A key site selection criterion for this report is that depth to aquifer must exceed 450 feet depth below land surface (DBLS), to ensure that there is an adequate zone for the test gas to diffuse. A key site selection criterion for this report is that depth to aquifer must exceed 450 feet depth below land surface (DBLS), to ensure that there is an adequate zone for the test gas to diffuse. An additional benefit of selecting a location with at thick vadose zone is to provide sufficient protection to groundwater. The United States Geological Survey (USGS) maintains an active and robust monitoring program for the Eastern Snake River Plain and its aquifer (NWIS; Davis, 2010; Bartholomay et al. 2012). Figure 6 illustrates sites at which water levels are monitored periodically as part of their regulatory stewardship. The data have been hand contoured to illustrate spatial variations in depth to the water table below land surface (DBLS). Location DBLS increases generally from north at ~250 feet DBLS to > 700 feet DBLS downgradient of the south margin of the INL. A bold line is drawn to illustrate parts of the INL having water tables depths less than (to the north) and more than (to the south) the 450 feet DBLS level. This line therefore represents a critical datum and constraint that will be referred to in subsequent illustrations and discussion. Figure 6. Illustration of key vadose zone geological site selection criteria. Regional Geology The INL is located on the northern margin of the Yellowstone-Snake River Plain volcanic track (YSRP) (Pierce and Morgan 1992), in the northern Basin and Range tectonic province (e.g., Anders et al. 1989). The volcanic track is the surface manifestation of an active continental hot Figure 6. Illustration of key vadose zone geological site selection criteria. Figure 6. Illustration of key vadose zone geological site selection criteria. Regional Geology The INL is located on the northern margin of the Yellowstone-Snake River Plain volcanic track (YSRP) (Pierce and Morgan 1992), in the northern Basin and Range tectonic province (e.g., Anders et al. 1989). The volcanic track is the surface manifestation of an active continental hot spot and deep mantle plume that is currently located ~100 mi (160 km) to the northeast of INL beneath Yellowstone National Park (Smith et al. 2009; Schmandt et al. 2012). In the INL area, rhyolite volcanism and genetically related tectonic subsidence occurred between ~10-4 Ma (Morgan and McIntosh; Rodgers et al.). Over the last 4 m.y. the post-hot spot track volcanism has been dominated by low intensity effusive basalt volcanism from 100's of widely scattered, overlapping, monogenetic shield volcanoes and northwest trending volcanic rift zones (Kuntz et al. 1992; Smith et al. 1996), and emplacement of scattered rhyolite cryptodomes and lava domes (e.g., McCurry et al. 2008). Basalt lavas incrementally accumulated to a thickness of up to 2 km thick in the center of the ESRP. A strong concentration of vents in the central ESRP produced a broad constructional topographic high that is referred to as the axial volcanic zone (AVZ) (Smith 2004; Hughes et al., 1999, 2002). Concurrent with the basalt volcanism, southward-directed drainage into the ESRP from sources to the north produced layers of clastic sediment that are interlayered with basalt lavas (Bestland et al. 2002; Geslin et al. 2002). Construction of the Axial Volcanic Zone diverted the southward drainage into a broad basin referred to as the Big Lost Trough, that overlaps most of the INL (Figure 6). Smith (2004) summarizes many of the salient geologic features of the shallow subsurface at INL. Local Geologic Features The surficial geology north of the INL and vicinity is comprised of Paleozoic to Mesozoic age carbonate and clastic sedimentary rocks. Those rocks are tilted and faulted by range-bounding northwest trending, Miocene to Quaternary age normal faults (Rodgers et al. 2002). The faults die out to the south into the ESRP (Jackson et al. 2006). The Paleozoic and Mesozoic rocks are also tilted into the Snake River Plain due to crustal subsidence of ESRP over the last 10 m.y. (Rodgers et al. 2002). At INL, the surface and near-surface geology is dominated by olivine tholeiite basalt lava flows (Figure 6). Surficial lavas are dominantly 200 to 600 ka. Ages decrease from northwest to southeast as a result of progressive subsidence of the ESRP (Rodgers et al. 2002). The basalts were erupted effusively from numerous overlapping shield volcanoes (Figure 7). Many of the vents are northwest trending, and also have cogenetic northwest trending fracture systems that formed in response to shallow dike intrusions (e.g., Kuntz et al. 2002). Many of the vents also cluster into northwest trending 'rift zones' that may root at depths ≥1-3 km into deep crustal dike swarms (e.g., Holmes et al. 2008). Figure 6. Surficial geology of INL and surroundings. Geology is simplified from maps by Kuntz et al. (1994; 2003). Locations of buried basalt vents are from Hughes et al. (2002) and Anderson and Liszewski (1998). Figure 6. Surficial geology of INL and surroundings. Geology is simplified from maps by Kuntz et al. (1994; 2003). Locations of buried basalt vents are from Hughes et al. (2002) and Anderson and Liszewski (1998). Figure 7. (A) Diagrammatic representation of interlayered volcanic and sedimentary deposits underlying the Easte Snake River Plain (from Hughes et al. 2002). (B) Volcanic rocks were erupted within NW-trending volcanic rifts that turn were produced by vertical dike swarms. S th t f th INL d l i b ll hi h t ti f b lti t Figure 7. (A) Diagrammatic representation of interlayered volcanic and sedimentary deposits underlying the Eastern Snake River Plain (from Hughes et al. 2002). (B) Volcanic rocks were erupted within NW-trending volcanic rifts that in turn were produced by vertical dike swarms. Figure 7. (A) Diagrammatic representation of interlayered volcanic and sedimentary deposits underlying the Eastern Snake River Plain (from Hughes et al. 2002). Local Geologic Features (B) Volcanic rocks were erupted within NW-trending volcanic rifts that in turn were produced by vertical dike swarms. Southern parts of the INL are underlain by an unusually high concentration of basaltic vents, and also intermediate to rhyolitic lava domes and cryptodomes (McCurry et al. 2008). Basalt lavas as young 12 ka (Cerro Grande lava) to 5.1 ka (Hell's Half acre) were erupted from vents along the AFZ just south of the INL. Basalt lavas at INL are widely overlain by a thin (< few meters), discontinuous veneer of loess (REFs), most of which is <14 ka. They are also overlain and interlayered with clastic fluvial and lacustrine sediments that collected over the last few million years into the Big Lost Trough. Those sediments are thickest along the channel of the Big Lost River and in central and northern parts of INL (Figure 6). Geologic Hazards For more than six decades the INL has been home to a large number of nuclear research projects including the construction of more than 40 reactors. As a consequence, a significant amount of effort has been directed towards characterizing the geologic hazards at the INL, which are primarily flooding, seismicity and volcanism. The characterization of these hazards is discussed in detail in the following references Ostenaa et al. 2002; Northwind (2011), p. 66, Anders et al. (1989); Smith et al. (2009) Jackson et al. (1993). The summary of these assessments is that in spite of the INL’s proximity to active geologic faults, the ESRP has been remarkably free of seismic activity for thousands of years. Borehole studies Over the last six decades numerous boreholes have been constructed to monitor aquifer conditions and also to define the three-dimensional architecture of the vadose and active groundwater systems at INL. These data have recently been integrated into comprehensive models for the southern half of INL (Champion et al. 2011; Hodges et al. 2012; Twining et al. 2008). Additional borehole characterization is available for specific sites in the northern half of INL (e.g., at 2-2A, and near TAN, e.g., Bestland et al. 2002; Anderson and Bowers 1995). Cross-section models depicting the large-scale subsurface architecture at INL, based on surficial geology and borehole data (Figure 7) are shown in Figure 8. Subsurface correlations of basaltic lava flows and flow groups are robust from land surface to between 200 and 400 meters DBLS. They are based upon multiple correlation criteria including lava flow radiometric ages, paleomagnetic inclinations and polarity, bulk rock chemistry and mineralogy, electric log data (e.g., natural gamma), and lava flow facies analysis. Additional criteria exist for specific boreholes, including bulk rock isotopic compositions and phenocryst compositions (e.g., Hughes et al. 2002). Champion et al. (2011) work is augmented by work along the southern margin of INL by numerous other studies (e.g., Hodges et al. 2012; Twining et al. 2008). Collectively these studies yield a clear understanding of the large-scale architecture for much of the southern half of INL, in particular those areas that satisfy minimum vadose thickness requirements (i.e. ≥450 feet). Champion et al. (2011) focused on correlations of lava flow groups, i.e. the collected sequence of cogenetic lavas produced during the life cycle of single monogenetic volcanoes. Sediment interbeds make up <10% of the subsurface architecture and form impermeable barriers to vertical fluid flow. Flow groups vary up to 100 meters thick near centers of the paleo-shield volcanoes forming thick sections of layered basalt that make ideal targets for the SAFE test. Champion et al. (2011) focused on correlations of lava flow groups, i.e. the collected sequence of cogenetic lavas produced during the life cycle of single monogenetic volcanoes. Sediment interbeds make up <10% of the subsurface architecture and form impermeable barriers to vertical fluid flow. Flow groups vary up to 100 meters thick near centers of the paleo-shield volcanoes forming thick sections of layered basalt that make ideal targets for the SAFE test. Figure 7. Location map illustrating locations of surface- and borehole-constrained geologic cross-sections of INL. Borehole studies Figure 7. Location map illustrating locations of surface- and borehole-constrained geologic cross-sections of INL. Figure 8. Cross-sections of vadose zone and upper parts of the Eastern Snake River Plain across southern parts of the INL (from Champion et al. 2011). ! Figure 8. Cross-sections of vadose zone and upper parts of the Eastern Snake River Plain across southern parts of the INL (from Champion et al. 2011). ! Figure 8. Cross-sections of vadose zone and upper parts of the Eastern Snake River Plain across southern parts of the INL (from Champion et al. 2011). Basalt lavas The shallow subsurface architecture for the southern half of INL is dominated by inflationary pahoehoe lava flows (Welhan et al. 2002). A model for the hydrogeologically salient physical features of the lavas is illustrated in Figure 9. Permeability is highly variable in vertical section of flow lobes, and is dominated by fractures and interflow rubbles zones, some of which have extremely high values of hydraulic conductivity (e.g., Ackerman 1991, p. 30; Bartholomay et al. 2000, p. 15). Strong variation is also observed in longitudinal (proximal to distal) and cross- sectional (lobate) views of the lavas. Welhan et al. (2002a, b) summarize key types and hydrogeological scaling properties of the basalt lavas. The net result of the basalt depositional architecture, as shown in Figure 9, is a highly heterogeneous hydrologic system where horizontal permiability is extremely high relative to vertical permeability. Figure 9. A: Vent to toe cross--section illustrating key hydrogeological features of basalts lava flows (from Welhan et al. 2002). Figure 9. A: Vent to toe cross--section illustrating key hydrogeological features of basalts lava flows (from Welhan et al. 2002). to toe cross--section illustrating key hydrogeological features of basalts lava flows (from Welhan et al Summary of site selection criteria 1) Vadose zone thickness ≥450 feet 2) Located > ~2 km from site boundary or major roads/highways 3) Located > 1 km from know contamination sites (e.g., RWMC, INTEC) and maj 3) Located > 1 km from know contamination sites (e.g., RWMC, INTEC) and major roads; 4) Avoid regions of active seismicity or geothermal activity 5) Avoid regions of potential flooding 6) Avoid regions that would provide structural pathways for vertical flow (e.g., faults, dikes, vents or volcanic rifts) 6) Avoid regions that would provide structural pathways for vertical flow (e.g., faults, dikes, vents or volcanic rifts) 7) Located > 1 km from known or potential sources of surface infiltration (e.g., BLR; diversion ponds; playas; other regions of known rapid infiltration); anthropogenic wastewater recharge areas; and radiologically contaminated vadose zone or groundwater systems 8) Permeable basalt interval thickness of ≥ ~300 feet ('test interval') 8) Permeable basalt interval thickness of ≥ ~300 feet ('test interval') 9) Aquitards located both below and above basalt test interval 9) Aquitards located both below and above basalt test interval Application of site selection criteria: The following discussion is a step-wise application of site selection criteria to the INL. The following discussion is a step-wise application of site selection criteria to the INL. Criteria 1-3: Figure 10 illustrates parts of INL, which satisfy criteria 1-3. Parts of the INL satisfying these criteria are shaded green. Depth to water table is contoured specifically for 2010 to present. Criterion 4: Avoid regions having active seismicity or geothermal activity. There is no known geothermal activity at INL. Seismicity is infrequent and of low intensity, and is therefore not considered to be a site exclusion factor at INL. Criterion 5: Avoid regions of potential flooding. Criterion 5: Avoid regions of potential flooding. Criterion 5: Avoid regions of potential flooding. Sedimentary interbreeds The distribution of permeability in the SRPA is spatially and vertically variable and is strongly impacted by the distribution of sedimentary interbeds deposited on the basalt by wind and water. The distribution of sedimentary interbeds within the basalt flows is controlled periods of volcanic quiescence accompanied by fluvial or lucustrine sedimentation. Sedimentary interbeds generally make up less than 10% of the volume of the vadose zone in the southern half of INL. Although volumetrically small the sediment interbeds are composed of fine-grained sediment layers (clays and silt) that act as key aquitards (e.g., Winfield 2005). Fine-grained sediments also infiltrate into underlying basalt lavas reducing their porosity and the vertical permeability of the system. Sediment interbeds vary from a thin mantling of the basalt to over 100 feet in thickness. Lateral (horizontal) extent of the interbeds is often difficult to constrain from borehole data, but appears to vary from 10's meters up to a few kilometers (Stroup et al. 2008; Welhan et al. 2006). Application of site selection criteria: Criteria 1-3 summary: Parts of INL passing site selection criteria 1-7 are illustrated in Figure 13. Areas labeled A1 and A2 are relatively well characterized by borehole data, and are preferred for further site assessment over less well-characterized parts of INL labeled B1 and B2. Criterion 5: Avoid regions of potential flooding. Flood hazards are restricted to regions along the Big Lost River, the Big Lost River diversion area, and in the sinks areas of northern INL. These areas have already been excluded from site consideration because of failure to pass previous site selection criteria. Criterion 6: Avoid regions that would provide preferred pathways for vertical fluid flow (e.g., faults, dikes, vents or volcanic rifts). No significant tectonic faulting occurs on INL. Tectonic faults are therefore not considered to be a site exclusion factor at INL. Other geologic features, such as volcanic vents, dikes or rifts, are concentrated in AVA and labeled rift zones (Figure 11). The H-EB (Howe-East Butte) rift is distinguished from the other rift zones (ARZ, LR-HHA and CB-KB) in lacking evidence for rift- related activity south of the Lost River sinks and north of AVZ. That part of the rift is therefore retained for possible site selection. Site selection criterion six excludes easternmost parts of INL from consideration, as shown in Figure 11. Criterion 7: Located > 1 km from known or potential sources of surface infiltration (e.g., BLR; diversion ponds; playas; other regions of known rapid infiltration); anthropogenic wastewater recharge areas; and radiologically contaminated vadose zone or groundwater systems. Following from Busenberg et al. (2001), Figure 12 illustrates regions at INL that are distinguished by rapid recharge - mainly the sinks, AVZ and Big Lost River diversion and spreading area (yellow), regions affected by anthropogenic recharge to the aquifer (purple). Regions of INL that have tritium groundwater contamination >500 picocuries/liter (an assumed upper limit for site selection) are from Davis et al. (2010), and are shown in pink. Criteria 8 and 9: Parts of potential site selection areas A1 (at LSIT) and A2 (borehole NPR-Test) have been well characterized for their subsurface stratigraphy. Figure 14 illustrates the subsurface stratigraphy in test area A1 near LSIT (Champion et al. 2011, borehole USGS 132). Water table depth is 180 meters (590 feet). Basalt lavas dominate borehole 132. Sedimentary interbeds occur at approximately 10 m (33 feet), 45 (150 feet) and 150 meters (490 feet) DBLS. Interbeds therefore bracket the minimum site selection criteria of ≥350 feet. Lateral extent of the interbed sediments is not well constrained. Figure 15 illustrates a conceptual model of interbeds at SDA (RWMC), ~2-3 km north of LSIT. It is possible that SDA interbeds number BC and FG may correlate with interbeds at similar depth at LSIT (borehole USGS132), suggesting that lateral continuity of those beds occurs at least at the scale of kilometers. Figure 16 illustrates stratigraphy for borehole NPR-Test, located in test area A2. Water table depth at that site is 474 feet, exceeding minimum depth requirement of 450 feet. Basalt lavas dominate borehole stratigraphy. Fine Sediment interbeds occur at depths of 30 m (100 feet), 76 m (250 feet) and 125 m (410 feet) DBLS. Borehole NPR-Test therefore also meets site selection criteria. At NPR-Test, as in the case of LSIT, lateral continuity of the sediment interbeds is poorly constrained. Correlations of basalt lavas between NPR-Test and borehole USGS 123, located near INTEC, and about 5 km west of NPR-Test, suggest that NPR-Test interbeds at 100 and 250 feet DBLS may also be correlated between the two boreholes. However the 410-foot interbed at NPR-Test does not appear to extend as far as the USGS 123 borehole. Summary of application of site selection criteria: The geologic architecture of the Eastern Snake River Plain is well suited for the SAFE concept test, due to the regions layer cake geologic structure, and thick unsaturated zone. The geologic conditions found within the ESRP due to a very high horizontal to vertical permeability. This would allow for rapid dispersal of propulsion gas while limiting the vertical transport to the surface or to the deeper groundwater system. Geologically based site selection criteria have been used to identify areas of the INL for consideration for testing of the SAFE concept. Application of these criteria indicates that three regions within the INL should be considered as potential test sites. Parts of two areas (labeled B1 and B2 in Figure 14) are too poorly characterized by existing subsurface data for evaluation of all site selection criteria. These two areas are not considered, as they would require a significant amount of effort to characterize. Parts of two other areas (A1 and A2) satisfy all the site selection criteria and should be considered for a more detailed assessment. This report recommends that subsequent consideration of test sites focus on regions located at or near LSIT and NPR-Test. Figure 10. Map illustrating parts of INL that pass site selection criteria 1 through 3. Figure 10. Map illustrating parts of INL that pass site selection criteria 1 through 3. Figure 11. Map illustrating parts of INL that pass site selection criteria 1 through 6. Figure 11. Map illustrating parts of INL that pass site selection criteria 1 through 6. Figure 12. Following Busenberg et al. (2001) this figure illustrates regions at INL that are distinguished by rapid rech - mainly the sinks, AVZ and Big Lost River diversion and spreading areas (yellow); and regions affecte anthropogenic recharge to the aquifer (purple). The figure also illustrates flow paths for ground water in the upper of the ESRP aquifer (after Fisher et al. 2012). Figure 12. Following Busenberg et al. (2001) this figure illustrates regions at INL that are distinguished by rapid recharge - mainly the sinks, AVZ and Big Lost River diversion and spreading areas (yellow); and regions affected by anthropogenic recharge to the aquifer (purple). The figure also illustrates flow paths for ground water in the upper parts of the ESRP aquifer (after Fisher et al. 2012). Figure 13. Map illustrating parts of INL that pass site selection criteria 1 through 7. Figure 13. Nuclear Furnace Option In 1955, the Los Alamos Scientific Laboratory began the Rover program to develop a solid core nuclear rocket engine. The basic concept was to allow a graphite-fuel based nuclear reactor to reach high temperatures, to cool the reactor with clean hydrogen, and to exhaust the high-speed hydrogen for thrust. The advantages were seen to be shorter trip times, lower mass in orbit, and no possibility of accidental explosion. In 1963, the Nuclear Engine for Rocket Vehicle Applications (NERVA) began with Aerojet as the prime contractor and Los Alamos as a supporting contributor. The goal of the NERVA program was to transform the nuclear reactor technology developed by Los Alamos and produce a space qualified nuclear engine. Both programs were terminated in 1972. Before termination, however, the Rover/NERVA programs built and tested 23 reactors/engines, achieved fuel temperatures in excess of 5500 F, ran a reactor with a peak power of greater than 4000 megawatts, operated a system for over an hour, demonstrated start-up and shut-down operations, and proved that the graphite based reactor core could withstand the extreme conditions of operation. The exhaust of the engine in the final days of the program was calculated to have a specific impulse of near 850 seconds, almost three times the performance of the kerosene engines of the Saturn V and twice that of the soon-to-be-developed LOX/hydrogen engines of the Space Shuttle. One of the issues that manifested later in the NERVA program was the high level of radioactivity present in the exhaust of the NTR. Because the fuel used a graphite matrix containing uranium carbide particles, the hot hydrogen in the coolant could chemically react through cracks in the cladding of the flow channels. Consequently, both uranium and fission products could escape nto the hydrogen flow and eject out the nozzle. Around 1971, the Rover/NERVA programs demonstrated that the exhaust from a nuclear engine could be “scrubbed” clean of all fission products. As the result of increased restrictions on emission of radioactivity into the atmosphere, the Nuclear Furnace was built in order to continue testing new fuel-element materials. The Furnace consisted of a 45 MW reactor in which many of the fuel elements could be replaced with experimental elements to assess behavior such as corrosion. The Nuclear Furnace reactor was followed by a sequence of filters to clean the effluent. Summary of application of site selection criteria: Map illustrating parts of INL that pass site selection criteria 1 through 7. Figure 14. Summary and correlation fence diagram for subsurface stratigraphy near LSIT (from Champion et al., 2011). Figure 15. Complex interbedding of basalt lavas and sediments at the Subsurface Disposal Area (RWMC area) (from Magnuson, 2004). Figure 14. Summary and correlation fence diagram for subsurface stratigraphy near LSIT (from Champion et al., 2011). Figure 15. Complex interbedding of basalt lavas and sediments at the Subsurface Disposal Area (RWMC area) (from Magnuson, 2004). Figure 14. Summary and correlation fence diagram for subsurface stratigraphy near LSIT (from Champion et al., 2011). Figure 14. Summary and correlation fence diagram for subsurface stratigraphy near LSIT (from Champion et al., 2011). Figure 15. Complex interbedding of basalt lavas and sediments at the Subsurface Disposal Area (RWMC area) (from Magnuson, 2004). lex interbedding of basalt lavas and sediments at the Subsurface Disposal Area (RWMC area) (from Figure 15. Complex interbedding of basalt lavas and sediments at the Subsurface Disposal Area (RWMC area) (from Magnuson, 2004). Figure 16. Borehole log of NPR-Test (source), and plausible correlations of stratigraphic units between NPR-Test a INTEC. Figure 16. Borehole log of NPR-Test (source), and plausible correlations of stratigraphic units between NPR-Test and INTEC ole log of NPR-Test (source), and plausible correlations of stratigraphic units between NPR-Test and Figure 16. Borehole log of NPR-Test (source), and plausible correlations of stratigraphic units between NPR-Test and INTEC. Nuclear Furnace Option After passing through the reactor, the hydrogen exhaust was sprayed with steam to cool the gas and remove any particulates. The flow then passed through a tube-and-kettle heat exchanger to further reduce the temperature. Next, the gas flowed through a silica gel bed to remove the water and any dissolved fission products. At this point, the only remaining products were the noble gases that were removed by passing the gases through a cryogenically cooled, activated charcoal bed. The result was a hydrogen jet that contained no detectable fission products. Conceptually, a new nuclear furnace reactor could be built today to qualify NTR fuel elements. Then the first test of the NTR would be in space. Potentially, this could reduce program cost but adds program risk. To assess the possibility, difficulties, and ROM cost of building a nuclear furnace, the CSNR discussed the options and possibilities with expert INL staff. The funding level of this project was insufficient to subcontract the INL staff or perform any true cost estimates. The primary issues identified in the discussions were 1) need for a new Environmental Impact Statement, 2) physical security for a new, green-field construction near to existing facilities at the INL site, 3) ability to guarantee that no radioactivity was present in the exhaust, and 4) hydrogen handling and flaring. The reactor power was assumed to be the same as the Nuclear Furnace at 44 MWth. Given these assumptions and the fact that this would be a new Category I facility, the ROM estimate cost is in excess of $250M. This estimate can be refined further and in more detail but further funding to the INL will be required. Conclusions The results show that the INL geology is substantially better suited to the SAFE testing method than the NTS site. The existence of impermeable interbeds just above the sub-surface aquifer ensure that no material from the test, radioactive or not, can enter the water table. Similar beds located just below the surface will prevent any gaseous products from reaching the surface for dispersion. The extremely high permeability of the strata between the interbeds allows rapid dispersion and dilution of the rocket exhaust. Preliminary gas transport calculations, and review of simulations performed for the NTS, indicate that condensation effects will not significantly increase backpressures above those calculated assuming isothermality. This suggests that the greater pressure diffusivity available in the INL subsurface could also allow the injection borehole to be significantly smaller diameter and the depth to be shallower than the holes at the NTS, which could substantially reduce project costs. In addition, the high permeability means a much lower back pressure in the hole against the rocket thrust which increases safety of operations. Finally, because of the highly permeability layered basalt and the presences of protective sedimentary interbeds at a shallow depth, cost of performing a sub-scale, non-nuclear verification experiment at the INL is was determined to be $2,100K. This cost estimate is based on the computation gas injection model and the favorable geology. The assements presented in this report indicates that the non nuclear test could be conducted at a much shallower depth than at the NTS and with a significantly smaller bore hole diameter. Rocketdyne scrubber, Tom hill scrubber, Msfc scrubber Rocketdyne scrubber, Tom hill scrubber, Msfc scrubber References 1-4 ntr benefits studies 5-7 NRC reports Safe howe 1999, Safe DRI 2007 Safe howe 2010 Ackerman, D.J., 1991, Transmissivity of the Snake River Plain aquifer at the Idaho National Engineering Laboratory, Idaho: U. S. Geological Survey Scientific Investigations Report 91-4058, p. 35. Anders, M.H., Geissman, J.W., Piety, L.A., and Sullivan, J.T., 1989, Parabolic distribution of circumeastern Snake River Plain seismicity and latest Quaternary faulting: Migratory pattern and association with the Yellowstone hotspot: Journal of Geophysical Research, v. 94, p. 1589-1621. Anderson, S.R., and Bowers, B., 1995, Stratigraphy of the unsaturated zone and uppermost part of the Snake River Plain aquifer at Test Area North, Idaho National Engineering Laboratory, Idaho: U. S. Geological Survey Water-Resources Investigations Report 95-4130, p. 47. Anderson, SR and MJ Liszewski, 1997, Stratigraphy of the unsaturated zone and the Snake River Plain aquifer at and neare the Idaho National Engineering Laboratory, Idaho. Idaho National Laboratory Report DOE/ID-22142, 65 p. Bartholomay, R.C., Davis, L.C., Fisher, J.C., Tucker, B.J., and Raben, F.A., 2012, Water-Quality Characteristics and Trends for Selected Sites At and Near the Idaho National Laboratory , Idaho , 1949 – 2009: U. S. Geological Survey Scientific Investigations Report 2012-5169, p. 67. Bestland, E.A., Link, P.K., Lanphere, M.A., and Champion, D.E., 2002, Paleoenvironments of sedimentary interbeds in the Pliocene and Quaternary Big Lost Trough (eastern Snake River Plain, Idaho): Geological Society of America Special Paper 353, p. 27–44. Busenberg, E., Plummer, L.N., and Bartholomay, R.C., 2001, Estimated age and source of the young fraction of ground water at the Idaho National Engineering and Environmental Laboratory: U. S. Geological Survey Water-Resources Investigations Report, v. 2001-4265, p. 144. Cahn, Magnuson, Rood, Jolley (2008), Operable Unit 10-08 Sitewide Groundwater and Miscellaneous Sites Remedial Investigation/Baseline Risk Assessment, DOE/ID-11332. Cooper, C.A., and D.L. Decker, 2011, Injection of nuclear rocket exhaust and water into a deep unsaturated zone, Nuclear Technology, 174(3), 452-459. Champion, D.E., Hodges, M.K. V., Davis, L.C., and Lanphere, M.A., 2011, Paleomagnetic Correlation of Surface and Subsurface Basaltic Lava Flows and Flow Groups in the Southern Part of the Idaho National Laboratory , Idaho , with Pleomagnetic Data Tables for Drill Cores Scientific Investigations Report 2011 – 5a049: U. S. Geological Survey Scientific Investigations Report 2011- 5049, p. 33. Champion, D.E., Lanphere, M.A., and Kuntz, M.A., 2002, Accumulation and subsidence of late Pleistocene basaltic lava flows of the eastern Snake River Plain, Idaho, in Link, P.K. and Mink, L.L. eds., Geology, Hydrogeology, and Environmental Remediation: Idaho National Engineering and Environmental Laboratory, Eastern Snake River Plain. Hodges, M.K. V., Orr, S., Potter, K.E., and LeMaitre, T., 2012, Construction Diagrams , Geophysical Logs , and Lithologic Descriptions for boreholes USGS 103, 105, 108, 131, 135, NRF-15 , and NRF-16 , Idaho National Laboratory , Idaho: U. S. Geology Survey Data Series 660, p. 34. Safe howe 1999, Safe DRI 2007 Safe howe 2010 Geological Society of America Special Paper, Geological Society of America, p. 175–192. Davis, L.C., 2010, An Update of Hydrologic Conditions and Distribution of Selected Constituents in Water , Snake River Plain Aquifer and Perched Groundwater Zones , Idaho National Laboratory , Idaho , Emphasis 2006 – 08 Scientific Investigations Report 2010 – 5197: US Geological Survey Scientific Investagations Report 2010-5197, p. 94. Dechert, T. V, Mcdaniel, P.A., Pierce, K.L., Falen, A.L., and Fosberg, M.A., 2006, Late Quaternary Stratigraphy , Idaho National Laboratory , Eastern Snake River Plain , Idaho: Idaho Geological Survey Technical Report 06-1, p. 19. Fisher, J.C., Rousseau, J.P., Bartholomay, R.C., and Rattray, G.W., 2012, A Comparison of U . S . Geological Survey Three-Dimensional Model Estimates of Groundwater Source Areas and Velocities to Independently Derived Estimates , Idaho National Laboratory and Vicinity , Idaho Scientific Investigations Report 2012 – 5152: U. S. Geological Survey Scientific Investigations Report 2012-5152, p. 142. Garabedian, S.P., 1992, Hydrology and digital simulation of the regional aquifer system, Eastern Snake River Plain: U. S. Geological Survey Professional Paper, v. 1408-F, 102 p. Geslin, J.K., Link, P.K., Riesterer, J.W., Kuntz, M.A., and Fanning, C.M., 2002, Pliocene and Quaternary stratigraphic architecture and drainage systems of the Big Lost Trough, northeastern Snake River Plain, Idaho: Geological Society of America Special Paper 353, p. 11–26. Hackett, WP, Smith, RP and S Khericha, 2002. Volcanic hazards of the Idaho National Engineering and Enviromental Laboratory, Southeast Idaho. In, B Bonnichsen, CM White and M McCurry, eds., Tectonic and Magmatic Evolution of the Snake River Plain Volcanic Province, Idaho Geological Survey Bulletin 30:461-482. Hodges, M.K. V., Orr, S., Potter, K.E., and LeMaitre, T., 2012, Construction Diagrams , Geophysical Logs , and Lithologic Descriptions for boreholes USGS 103, 105, 108, 131, 135, NRF-15 , and NRF-16 , Idaho National Laboratory , Idaho: U. S. Geology Survey Data Series 660, p. 34. Holmes, A. a. J., Rodgers, D.W., and Hughes, S.S., 2008, Kinematic analysis of fractures in the Great Rift, Idaho: Implications for subsurface dike geometry, crustal extension, and magma dynamics: Journal of Geophysical Research, v. 113, no. B4, p. 1–15, doi: 10.1029/2006JB004782. Hughes, S.S., Smith, R.P., Hackett, W.R., and Anderson, S.R.,1999, Mafic Volcanism and Environmental Geology of the Eastern Snake River Plain, Idaho, in Hughes, S.S., and Thackray, G.D., eds., Guidebook to the Geology of Eastern Idaho: Idaho Museum of Natural History, p. 143-168. Hughes, S.S., P.H. Wetmore, and J.L. Magnuson, S., 2004, Regulatory Modeling for the Idaho National Engineering and Environmental Laboratory’s Subsurface Disposal Area and Conceptual Model Uncertainty Treatment: Vadose Zone Journal, v. 3, no. 1, p. 59–74, doi: 10.2113/3.1.59. Safe howe 1999, Safe DRI 2007 Safe howe 2010 Casper, 2002, Evolution of Quaternary tholeiitic basalt eruptive centers on the eastern Snake River Plain, Idaho, in Bill Bonnichsen, C.M. White, and Michael McCurry, eds., Tectonic and Magmatic Evolution of the Snake River Plain Volcanic Province: Idaho Geological Survey Bulletin 30: 363-385. Hughes, SS, McCurry, M and DJ Geist, 2002. Geochemical correlations and implications for the magmatic evolution of basalt flow groups at the Idaho National Engineering and Environmental Laboratory. in Link, P.K. and Mink, L.L. eds., Geology, Hydrogeology, and Environmental Remediation: Idaho National Engineering and Environmental Laboratory, Eastern Snake River Plain. Geological Society of America Special Paper, Geological Society of America, p. 151-174. Jackson, S.M., Wong, I.G., Carpenter, G.S., Anderson, D.M., and Martin, S.M., 1993, Contemporary seismicity in the Eastern Snake River Plain, Idaho based on microearthquake monitoring: Bulletin of the Seismological Society of America, v. 83, no. 3, p. 680–695. Jackson, S.M., Carpenter, G.S., Smith, R.P., and Casper, J.L., 2006, Seismic reflectionr project near the southern terminations of the Lost River and Lemhi Faults , Eastern Snake River Plain , Idaho: Idaho National Laboratory Report, v. INL/EXT-06, no. October, p. 16. Kuntz, M.A., Anderson, S.R., Champion, D.E., Lanphere, M.A., and Grunwald, D.J., 2002, Tension cracks, eruptive fissures, dikes, and faults related to late Pleistocene-Holocene basaltic volcanism and implications for the distribution of hydraulic conductivity in the eastern Snake River Plain, Idaho: Geological Society of America Field Guide, v. 353, p. 111–133. Kuntz, M.A., Skipp, B.A., Lanphere, M.A., Scott, W.E., Pierce, K.L., Dalrymple, G. B. Champion, D.E., Embree, G.F., Page, W.R., Morgan, L.A., Smith, R.P., Hackett, W.R., and Rodgers, D.W., 1994, Geologic map of the Idaho National Engineering Laboratory and adjoining areas, eastern Idaho: U. S. Geological Survey IMAP, v. 2330. Kuntz, MA, Geslin, JK, Mark, LE, Hodges, MKV, Kauffman, ME, Champion, DE, Lanphere, MR, Rodgers, DW, Anders, MH, Link, PK and DL Boyack, 2003, Geologic map of the northern and central parts of the Idaho National Engineering and Environmental Laboratory, Eastern Idaho. Idaho Geological Survey Geologic Map 35. Internet reference: http://www.idahogeology.org/PDF/Maps_(M)/Geologic_Maps_(GM)/PDF/TAN_GM-35-m.pdf Magnuson, S., 2004, Regulatory Modeling for the Idaho National Engineering and Environmental Laboratory’s Subsurface Disposal Area and Conceptual Model Uncertainty Treatment: Vadose Zone Journal, v. 3, no. 1, p. 59–74, doi: 10.2113/3.1.59. Magnuson, S.O and A.J. Sondrup, 2006, Subsurface Flow and Transport Model Development for the Operable Unit 7-13/14 Remedial Investigation and Feasibility Study, ICP/EXT-05- 01016. Safe howe 1999, Safe DRI 2007 Safe howe 2010 McCurry, M., Hayden, K.P., Morse, L.H., and Mertzman, S., 2007, Genesis of post-hotspot, A-type rhyolite of the Eastern Snake River Plain volcanic field by extreme fractional crystallization of olivine tholeiite: Bulletin of Volcanology, v. 70, no. 3, p. 361–383, doi: 10.1007/s00445-007-0143- 4. Morse, L.H., 2002, Basalt alteration and authigenic mineralization near the effective base of the Snake River Plain aquifer at the idaho National Engineering and Environmental Laboratory, Idaho: Idaho State University. Mudra, P.J and B.L. Schmalz, 1965, An Appraisal of Gaseous Waste Disposal into the Lithosphere at the National Reactor Testing Station, Idaho, IDO-12024. North Wind, 2011, Idaho National Laboratory comprehensive land use and environmental stewarship report. Idaho National Laboratory Report INL/EXT-0500726, 71 p. Ostenaa, DA, O'Connell, RH, Walter, RA and RJ Creed, 2002. Holocene paleoflood hydrology of the Big Lost River, western Idaho National Engineering and Environmental Laboratory, Idaho. in Link, P.K. and Mink, L.L. eds., Geology, Hydrogeology, and Environmental Remediation: Idaho National Engineering and Environmental Laboratory, Eastern Snake River Plain. Geological Society of America Special Paper, Geological Society of America, p. 91-110. Payne, SJ, Carpenter, NS, Hodges, JM and RG Berg, 2009, INL seismic monitoring annual report: January 1, 2008 - December 31, 2008. Idaho National Laboratory Report INL/EXT-09-16687, 81 p. Pierce, KL and LA Morgan, 1992. The track of the Yellowstone hot spot: volcanism, faulting, and uplift. In, Link, PK, Kuntz, MA and Platt, LB, eds., Regional Geology of Eastern Idaho and Western Wyoming: Geological Society of America Memoir 179: 1-53. Pruess, K., C. Oldenburg, and G. Moridis, 1999, TOUGH2 User’s Guide, Version 2.0, Report LBNL-43134, Lawrence Berkeley National Laboratory.Garabedian, S. P., 1992, “Hydrology and Digital Simulation of the Regional Aquifer System, Eastern Snake River Plain, Idaho,” Professional Paper 1408-F, U.S. Geological Survey. Robertson, J. B., 1974, Digital Modeling of Radioactive and Chemical Waste Transport in the Snake River Plain Aquifer at the National Reactor Testing Station, Idaho – 1952-1970, Open-File Report IDO-22054, U.S. Geological Survey. Schmandt, B., Dueker, K., Humphreys, E., and Hansen, S., 2012, Hot mantle upwelling across the 660 beneath Yellowstone: Earth and Planetary Science Letters, v. 331-332, p. 224–236, doi: 10.1016/j.epsl.2012.03.025. Smith, R.B., Jordan, M., Steinberger, B., Puskas, C.M., Farrell, J., Waite, G.P., Husen, S., Chang, W.-L., and O’Connell, R., 2009, Geodynamics of the Yellowstone hotspot and mantle plume: Seismic and GPS imaging, kinematics, and mantle flow: Journal of Volcanology and Geothermal Research, v. 188, no. 1-3, p. 26–56, doi: 10.1016/j.jvolgeores.2009.08.020. Safe howe 1999, Safe DRI 2007 Safe howe 2010 Smith, R.P., Jackson, S.M., and Hackett, W.R., 1996, Paleoseismology and seismic hazards evaluations in extensional volcanic terrains: Journal of Geophysical Research, v. 101, no. B3, p. 6277–6292. Stroup, C.N., Welhan, J.A., and Davis, L.C., 2008, Statistical Stationarity of sediment interbed thicknesses in a basalt aquifer , Idaho National Laboratory , Eastern Snake River Plain , Idaho. U. S. Geological Survey Scientific Investigations Report 2008–5167. Twining, B. V., Hodges, M.K. V., and Orr, S., 2008, Construction diagrams, geophysical logs, and lithologic descriptions for boreholes USGS 126a, 126b, 127, 128, 129, 130, 131, 132, 133, and 134, Idaho National Laboratory, Idaho: U. S. Department of Energy DOE/ID-222, p. 27. Welhan, J.A., Johannesen, C.M., Reeves, K.S., Clemo, T.M., Glover, J.A., and Bosworth, K.W., 2002, Morphology of inflated pahoehoe lavas and spatial architecture of their porous and permeable zones , eastern Snake River Plain , Idaho: , p. 135–150. Welhan, J.A., Farabaugh, R.L., Merrickj, M.J., and Anderson, S.R., 2006, Geostatistical Modeling of Sediment Abundance in a Heterogeneous Basalt Aquifer at the Idaho National Laboratory , Idaho Scientific Investigations Report 2006-5316: U. S. Geological Survey Scientific Investigations Report 2006-5316, p. 32. Whitehead, RL, 1992, Geohyddrologic framework of the Snake River Plain regional aquifer system, Idaho and Eastern Oregon. U.S. Geological Survey Professional Paper 1408-B, 32 p. Wood, T.R., Bates, D., Bishop, C.W., Heath, G.L., Hubbell, J.M., Hull, L.C., Lehman, R.M., Magnuson, S.O., Mattson, E.D., McCarthy, J.M., Porro, I., Ritter, P.D., Roddy, M., Sisson, J.B., et al., 2000, Deficiencies in vadose zone understanding at the Idaho National Engineering And Environmental Laboratory: Idaho National Engineering and Environmental Laboratory Bechtel BWXT Idahod, LLC Report, v. INEEL/EXT-, p. 191.
https://openalex.org/W3115210829	https://aclanthology.org/2020.semeval-1.291.pdf	English	null	TECHSSN at SemEval-2020 Task 12: Offensive Language Detection Using BERT Embeddings	null	2,020	cc-by	3,496	1 Introduction The usage of offensive or hate words is increasing these days, largely in online communication. People ﬁnd it easier to express their opinions and thoughts in online sources rather than do it personally. The anonymity provided by the online environment encourages many people to express their views in aggressively. The information spread rate is extremely fast in online social media. People, without checking the validity of the information they receive, spread it to others. The text content posted in messages, websites, social media and blogs are highly unstructured, informal, often misspelt, and use shorthanded notations, emojis and emoticons. Getting the meaning in the natural text is a complicated task. There is ongoing research in the ﬁeld of offensive language or hate speech detection, yet it remains only a goal to obtain a full-ﬂedged model. We have participated in the offensive language task conducted in SemEval-2019 by Zampieri et al. (2019a) with various machine learning and deep learning models in Rajalakshmi et al. (2019a). For SemEval-2020 by Zampieri et al. (2020), we have developed and tested deep learning models with different word embeddings. We have used GloVe (Pennington et al., 2014), Word2Vec (Mikolov et al., 2015) and BERT embedding (Devlin et al., 2018) pretrained models to identify the presence of offensiveness in tweets. We have participated in subtask A (OFF/NOT) of classifying whether a tweet is offensive or non-offensive and subtask B (TIN/UNT) of classifying whether an offensive tweet is targeted to anyone or untargeted. The rest of the paper is organized as follows. Section 2 surveys the related work in this ﬁeld. Section 3 describes the methodology used to solve the task. Results are discussed in section 4 and conclusion in section 5. Abstract This paper describes the work of identifying the presence of offensive language in social media posts and categorizing a post as targeted to a particular person or not. The work developed by team TECHSSN for solving the Multilingual Offensive Language Identiﬁcation in Social Media (Task 12) in SemEval-2020 involves the use of deep learning models with BERT embeddings. The dataset is preprocessed and given to a Bidirectional Encoder Representations from Transformers (BERT) model with pretrained weight vectors. The model is retrained and the weights are learned for the offensive language dataset. We have developed a system with the English language dataset. The results are better when compared to the model we developed in SemEval-2019 Task6. 2 Related Work Recently, we have seen great strides in the research on profanity speech detection in social media, which includes hate speech detection, offensive language identiﬁcation, and abusive language detection. Several workshops such as GermEval, SemEval, HatEval, and TRAC gain attention of the researchers in this ﬁeld. Research in hate speech includes work done by Basile et al. (2019), Fortuna and Nunes (2018), Malmasi and Zampieri (2017). Difference between profanity and hate speech, and the challenges involved are discussed in Malmasi and Zampieri (2018). An offensive language detection system is desribed out by This work is licensed under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/. 2190 2190 Proceedings of the 14th International Workshop on Semantic Evaluation, pages 2190–2196 Barcelona, Spain (Online), December 12, 2020. 3 System Methodology We have participated in SemEval-2019 Task 6 to identify and categorize English tweets using machine learning and deep learning models Rajalakshmi et al. (2019a), where it was found that 1D-CNN with GloVe embeddings and 2D-CNN with Word2Vec embeddings have performed better compared to other machine learning and deep learning algorithms. For SemEval-2020 Task 12, in addition to these algorithms, we also used deep learning model with BERT embeddings. The proposed system comprises of the following modules: • Dataset preparation • Preprocessing • Train the model using deep learning techniques • Prediction performance of the model Proceedings of the 14th International Workshop on Semantic Evaluation, pages 2190–2196 Barcelona, Spain (Online), December 12, 2020. Proceedings of the 14th International Workshop on Semantic Evaluation, pages 2190–2196 Barcelona, Spain (Online), December 12, 2020. Davidson et al. (2017) and Mandl et al. (2019). Most of the work use the machine learning and deep learning techniques. The work done by Wu et al. (2019) uses BERT uncased model with an F1 score of 0.8057 for task A and 0.50 for task B. Pavlopoulos et al. (2019) uses perspective API and BERT cased and uncased models to detect the offensive language with an F1 score of 0.7933 for task A and 0.6817 for task B. SemEval-2019 Task6 report of Zampieri et al. (2019b) says that machine learning algorithms such as Support Vector Machine (SVM), logistic regression, and Artiﬁcial Neural Network (ANN), and deep learning techniques such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Bidirectional Long Short-Term Memory network (BiLSTM), Embeddings from Language Models (ELMo), Bidirectional Encoder Representations from Transformers (BERT), Gated Recurrent Unit (GRU) and ensemble techniques have been employed for offensive language detection. Out of the 115 participant teams, 70% of the teams used deep learning techniques. In that, 20% teams used ensemble techniques, 11% used CNN, 13% used LSTM & BiLSTM, 10% used RNN & GRU, 8% used BERT and the remaining used other DL methods. Ensemble and BERT models gave better results for the top teams when compared to the other techniques. Table 1: Data distribution of SOLID dataset for various class labels 3.1 Dataset Preparation The Semi-Supervised Offensive Language Identiﬁcation Datatset (SOLID) used for developing the system comprises 9,089,140 instances in training dataset for subtask A and 3,887 instances in test dataset. It has 188,974 offensive instances in training dataset for subtask B and 1,422 instances in test dataset listed by Rosenthal et al. (2020). The break up of the dataset and its classes are shown in Table 1. Since the available infrastructure in our lab does not support working with the entire dataset for subtask A, we have used 1,000,000 instances to train the model. Each entry in the dataset consists of the features Id, Tweet, Avg conf and Conf std. Avg conf is the average of the conﬁdences predicted by several supervised models for a speciﬁc instance to belong to the positive class for that subtask. The class labels are OFF (offensive) and NOT (not offensive) for subtask A, while TIN (targeted insult and threat) and UNT (untargeted) for subtask B. Conf std is the standard deviation of conﬁdences for a particular instance. We have taken the threshold to be 0.5 for Avg conf and the values greater than the threshold are taken as positive examples and the rest as negative examples. The dataset is prepared in the format Id, Tweet, Label suitable to train the model. Task Label Train Test A OFF 1,448,861 1,080 NOT 7,640,279 2,807 B TIN 149,550 850 UNT 39,424 572 Table 1: Data distribution of SOLID dataset for various class labels Table 1: Data distribution of SOLID dataset for various class labels 2191 3.2 Preprocessing Unstructured tweet data contains a lot of irregularities which will affect the accuracy of the model. Therefore, it is important to preprocess the data before using it to build the model. Data preprocessing is an important step for increasing the performance of the model. The data is preprocessed by removing the irregularities, smoothening and normalizing the dataset. We have used NLTK (Bird et al., 2009) and Spacy toolkits (Honnibal and Montani, 2017) to preprocess the dataset. The preprocessing steps that we have developed in Rajalakshmi et al. (2019a) are listed below. Step e. is omitted for subtask B, since stopwords are signiﬁcant only for target identiﬁcation. a. URL removal b. Emojis and emoticons annotation c. Uppercase to lowercase conversion d. Contractions expansion e. Stopwords removal f. Special characters removal g. Accented characters removal h. Lengthened words reduction i. Text lemmatization j. Extra whitespace removal a. URL removal b. Emojis and emoticons annotation c. Uppercase to lowercase conversion d. Contractions expansion e. Stopwords removal f. Special characters removal g. Accented characters removal h. Lengthened words reduction i. Text lemmatization j. Extra whitespace removal We preprocessed the data with the steps outlined above and built the model using plain text. Furthermore, we can analyze the importance of accented characters, special characters and fully uppercase words and how they affect the performance of the system. 3.3 Model Building Various deep learning techniques with different word embeddings are applied on the SOLID dataset and their performances are analyzed. Our work in SemEval-2019 task 6 showed that 2D-CNN model with Word2Vec embeddings and 1D-CNN model with GloVe embeddings performed better than all other machine learning and deep learning algorithms with different word embeddings. For the present task, we have used LSTM and BERT models in addition to those algorithms. 3.3.3 BiLSTM Recurrent Neural Network (RNN) is especially designed to work with sequential data. Long Short-Term Memory networks (LSTM) (Hochreiter and Schmidhuber, 1997), an extension of RNN, are connected in a special way to avoid vanishing and exploding gradient issues. Bidirectional LSTMs can capture information about the past and future states simultaneously. We have used 2 LSTM layers for bidirection with 150 units and the inputs are trained with a batch size of 128 and dropout value of 0.2. Sigmoid function is used as the activation function in output layer and Adam algorithm is used for optimization. 3.3.2 1D-CNN with GloVe Conventional CNN with one dimensional layer is used with GloVe embeddings with 1 million word vectors of 200 dimensions from twitter data. The embedding layers are used to extract the skip-grams. Convolutional 1D layers use kernel ﬁlters of size 2, 3 and 4. Dropout value is set as 0.2 and 100 ﬁlters are also used. Maxpooling 1D layers are used to select the bigram, trigram and fourgram branches and they are merged for further processing. Softmax function is used in output layer and Relu function is used in all other layers. This model has fewer trainable parameters and takes less time to train. 3.3.4 BERT Bidirectional Encoder Representations from Transformers (BERT) is a deep bidirectional network built using transformers, which is pre-trained to detect a masked word in the given context sentence described by Devlin et al. (2018). It can also be used for text classiﬁcation and semantic relation extraction. We have used the publicly available BERT-Base, Multilingual cased pre-trained model for 104 languages that includes English, Tamil, Telugu, Hindi, Spanish, Arabic, Turkish, Urdu, Danish, Chinese, French and, Greek etc. This model has 12 transformer layers, 768 hidden layers and 12 heads with 110M parameters. This model is well suited for any of the given languages for task 12 given by Zampieri et al. (2020). GloVe and Word2Vec embeddings are context-free, while BERT embeddings use contextual represen- tation for word embeddings. Context-free representation gives a single word embedding for each word irrespective of its preﬁx or sufﬁx. Hence the word “bank” in “bank deposit” and “river bank” has same representation. In contextual representation the word “bank” has different representations based on the context of its nearby words; the contextual representation is considered in both forward and backward directions. Since BERT uses contextual knowledge in decision making, it will provide better interpretation than the context-free GloVe and Word2Vec embeddings as shown in the results. We have used the CoLA (The Corpus of Linguistic Acceptability) dataprocessor given in Warstadt et al. (2018) that is mainly used for single sentence classiﬁcation task. The sequence length is set to 128, and the batch size to 32. Adam optimizer is used and the learning rate is set as 0.2 to minimize the training time since the data is huge. The tweet sequence is tokenized and converted into features. The model is initialized with pretrained weight vectors and retrained to learn the input dataset features. 3.3.1 2D-CNN with Word2Vec Learned Embeddings We have used 2D-Convolutional Neural Network (CNN) model with Google’s Word2Vec pretrained weights as in Rajalakshmi et al. (2019b). The model is then retrained to relearn the weights for OffensE- val2020 SOLID dataset. The structure of the model comprises of the following layers. 1. Input layer 2. Embedding layer 3. Convolutional layer with kernel size 2, 3 and 4 4. Pooling layers for CNN layers 5. Fully connected dense layer 6. Output layer Embedding layer is used to relearn the weights of embedding matrix. Kernel ﬁlters are used to process bigrams, trigrams and fourgrams. Max pooling layer is used to scale down the output vectors to dense feature vectors that are concatenated and ﬂattened in fully connected layer. Output layer consists of 2 units for OFF/NOT or UNT/TIN. The word-grams are concatenated and computed in parallel to extract the possible information from the vectors. This enables the classiﬁer to understand the relationship between the words. The parameters for the model are set as follows: sequence length of the model is 43, learning rate is set as 0.001 and dropout is set as 0.5. Softmax activation function is used for output layer and Relu activation function in other layers. 2192 4 Results and Discussion Various deep learning models with different word embeddings are used to detect the presence of offensive- ness in the given tweet and to identify whether the tweet is targeted or not. Table 2 shows the models used to classify the given tweet into offensive or non-offensive category in subtask A. 1D-CNN model is trained with GloVe pretrained embeddings, 2D-CNN and BiLSTM models with Word2Vec embeddings. Deep network model with BERT embeddings achieves better F1 score when compared to other models. Model used F1 Macro 1D-CNN 0.732 2D-CNN 0.7451 BiLSTM 0.712 BERT 0.86551 Table 2: Results for models used in subtask A Table 2: Results for models used in subtask A Table 2: Results for models used in subtask A 2193 Table 3 shows the results for classifying the tweets into targeted and untargeted. The targeted tweet refers to a particular person or organization or group of people. Results show that BERT model performs better than other models. Among the SemEval-2020 teams participated for task 12, we were ranked 72 in subtask A and 36 in subtask B. Model used F1 Macro 1D-CNN 0.291 2D-CNN 0.3145 BiLSTM 0.2894 BERT 0.38938 Table 3: Results for models used in subtask B Table 3: Results for models used in subtask B 5 Conclusion and Future work There is an increase in the usage of profanity words in online communications due to the ease with which speakers can remain anonymous. People comment about a particular person or an organization or a group of people in an aggressive manner in social media. Due to the fast transmission of online communications, this information is spread rapidly. This has led to the need to detect the offensive tweets and to remove and stop them from spreading further. SemEval-2020 Task 12 involves three subtasks in which we have participated in subtasks A and B. Deep learning models with different word embeddings were used to perform the tasks. Results show that deep network model with BERT embeddings performs better when compared to GloVe and Word2Vec embedding models. Since the BERT model is based on multilingual case-based representation, this can also be applied to other languages like Tamil, Hindi, Telugu, Kannada, Arabic, Danish, Turkish and Greek. We would like to do further investigation on applying this model to other languages. Acknowledgements We would like to thank SSN management for providing us with the infrastructure to carry out this work. This work is carried out in the Machine Learning Research Group (MLRG) GPU server present in the Department of CSE, SSN College of Engineering. References Valerio Basile, Cristina Bosco, Elisabetta Fersini, Debora Nozza, Viviana Patti, Francisco Manuel Rangel Pardo, Paolo Rosso, and Manuela Sanguinetti. 2019. SemEval-2019 task 5: Multilingual detection of hate speech against immigrants and women in twitter. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 54–63. Steven Bird and Edward Loper. 2004. NLTK: The natural language toolkit. In Proceedings of the ACL Interactive Poster and Demonstration Sessions, pages 214–217, Barcelona, Spain, July. Association for Computational Linguistics. Steven Bird, Ewan Klein, and Edward Loper. 2009. Natural language processing with Python: analyzing text with the natural language toolkit. ” O’Reilly Media, Inc.”. Thomas Davidson, Dana Warmsley, Michael Macy, and Ingmar Weber. 2017. Automated hate speech detection and the problem of offensive language. In Proceedings of ICWSM. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirec- tional transformers for language understanding. arXiv preprint arXiv:1810.04805. Darja Fiˇser, Tomaˇz Erjavec, and Nikola Ljubeˇsi´c. 2017. Legal framework, dataset and annotation schema for socially unacceptable on-line discourse practices in Slovene. In Proceedings of the Workshop Workshop on Abusive Language Online (ALW), Vancouver, Canada. 2194 Paula Fortuna and S´ergio Nunes. 2018. A survey on automatic detection of hate speech in text. ACM Computing Surveys (CSUR), 51(4):85. Sepp Hochreiter and J¨urgen Schmidhuber. 1997. Long short-term memory. Neural computation, 9(8):1735–1780. Matthew Honnibal and Ines Montani. 2017. Spacy 2: Natural language understanding with bloom embeddings. convolutional neural networks and incremental parsing, 7(1). Ritesh Kumar, Atul Kr. Ojha, Shervin Malmasi, and Marcos Zampieri. 2018. Benchmarking aggression identiﬁ- cation in social media. In Proceedings of the First Workshop on Trolling, Aggression and Cyberbulling (TRAC), Santa Fe, USA. Ping Liu, Wen Li, and Liang Zou. 2019. Nuli at SemEval-2019 task 6: Transfer learning for offensive language detection using bidirectional transformers. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 87–91. Shervin Malmasi and Marcos Zampieri. 2017. Detecting hate speech in social media. In Proceedings of the International Conference Recent Advances in Natural Language Processing (RANLP), pages 467–472. Shervin Malmasi and Marcos Zampieri. 2018. Challenges in discriminating profanity from hate speech. Journal of Experimental & Theoretical Artiﬁcial Intelligence, 30:1–16. Thomas Mandl, Sandip Modha, Prasenjit Majumder, Daksh Patel, Mohana Dave, Chintak Mandlia, and Aditya Patel. 2019. Overview of the hasoc track at ﬁre 2019: Hate speech and offensive content identiﬁcation in indo- european languages. In Proceedings of the 11th Forum for Information Retrieval Evaluation, pages 14–17. References Tomas Mikolov, Kai Chen, Gregory S Corrado, and Jeffrey A Dean. 2015. Computing numeric representations of words in a high-dimensional space, May 19. US Patent 9,037,464. John Pavlopoulos, Nithum Thain, Lucas Dixon, and Ion Androutsopoulos. 2019. Convai at SemEval-2019 task 6: Offensive language identiﬁcation and categorization with perspective and bert. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 571–576. Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. GloVe: Global vectors for word representa- tion. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, Doha, Qatar, October. Association for Computational Linguistics. S Rajalakshmi, Angel Suseelan, B Logesh, S Harshini, B Geetika, S Dyaneswaran, S Milton Rajendram, and TT Mirnalinee. 2019a. Techssn at SemEval-2019 task 6: Identifying and categorizing offensive language in tweets using deep neural networks. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 753–758. S Rajalakshmi, Angel Suseelan, S Milton Rajendram, and TT Mirnalinee. 2019b. Ssn-sparks at SemEval-2019 task 9: Mining suggestions from online reviews using deep learning techniques on augmented data. In Proceed- ings of the 13th International Workshop on Semantic Evaluation, pages 1237–1241. Sara Rosenthal, Pepa Atanasova, Georgi Karadzhov, Marcos Zampieri, and Preslav Nakov. 2020. A large-scale semi-supervised dataset for offensive language identiﬁcation. In arxiv. Anna Schmidt and Michael Wiegand. 2017. A survey on hate speech detection using natural language process- ing. In Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media. Association for Computational Linguistics, pages 1–10, Valencia, Spain. Alexandra Schoﬁeld and Thomas Davidson. 2017. Identifying hate speech in social media. XRDS: Crossroads, The ACM Magazine for Students, 24(2):56–59. Bhargav Srinivasa-Desikan. 2018. Natural Language Processing and Computational Linguistics: A practical guide to text analysis with Python, Gensim, spaCy, and Keras. Packt Publishing Ltd. Alex Warstadt, Amanpreet Singh, and Samuel R Bowman. 2018. Neural network acceptability judgments. arXiv preprint arXiv:1805.12471. Michael Wiegand, Melanie Siegel, and Josef Ruppenhofer. 2018. Overview of the GermEval 2018 shared task on the identiﬁcation of offensive language. In Proceedings of GermEval. Zhenghao Wu, Hao Zheng, Jianming Wang, Weifeng Su, and Jefferson Fong. 2019. Bnu-hkbu uic nlp team 2 at SemEval-2019 task 6: Detecting offensive language using bert model. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 551–555. 2195 Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Sara Rosenthal, Noura Farra, and Ritesh Kumar. 2019a. Predicting the type and target of offensive posts in social media. References In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), pages 1415–1420. Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Sara Rosenthal, Noura Farra, and Ritesh Kumar. 2019b. SemEval-2019 task 6: Identifying and categorizing offensive language in social media (OffensEval). In Pro- ceedings of the 13th International Workshop on Semantic Evaluation, pages 75–86. Marcos Zampieri, Preslav Nakov, Sara Rosenthal, Pepa Atanasova, Georgi Karadzhov, Hamdy Mubarak, Leon Derczynski, Zeses Pitenis, and C¸ a˘grı C¸ ¨oltekin. 2020. SemEval-2020 Task 12: Multilingual offensive language identiﬁcation in social media (OffensEval 2020). In Proceedings of SemEval. Ziqi Zhang, David Robinson, and Jonathan Tepper. 2018. Detecting hate speech on Twitter using a convolution- GRU based deep neural network. In Lecture Notes in Computer Science. Springer Verlag. 2196
https://openalex.org/W2008690428	https://thescipub.com/pdf/ajassp.2014.1398.1404.pdf	English	null	A REVIEW: INDUSTRIAL CONTROL SYSTEM (ICS) AND THEIR SECUITY ISSUES	American journal of applied sciences	2,014	cc-by	4,360	1.1. Industrial Control System (ICS) Corresponding Author: Shahzad, A., Malaysian Institute of Inf The term Industrial Control System (ICS) is uses for several types of real time infrastructures (systems) such as Distributed Control Systems (DCS), Supervisory Control And Data Acquisition (SCADA) systems and Programmable Logic Controllers (PLC). The broad term Industrial Control System (ICS) had been used for industries or/and real time infrastructures (Stouffer et al., 2006). Industrial Control System (ICS) infrastructures are based on several types of field devices, which are communicating within ICS network for the purposes of message/data and commands delivery and feedback from remote station to master station. The communication between field devices are based on supervisory/automated commands such as an instruction to collect data from sensor connected with remote station, check the alarm status, breakers 1. INTRODUCTION opening and closing status information and time synchronization, transmitted from control station or master station to field devices using Human Machine Interface (HMI). Industrial Control System (ICS) had been designed for critical infrastructure sectors and 90% of these systems are property/owned and operated by private organizations. Industrial Control System (ICS) has been also operated by Federal agencies usually for the purposes of air traffic system control, nuclear plant operations and controlling and uses within oil industry, gas industry, chemical industry, transportation system and pharmaceutical manufacturing (Stouffer et al., 2006). American Journal of Applied Sciences 11 (8): 1398-1404, 2014 ISSN: 1546-9239 © 2014 A. Shahzad et al., This open access article is distributed under a Creative Commons Attribution (CC-BY) 3.0 license doi:10.3844/ajassp.2014.1398.1404 Published Online 11 (8) 2014 (http://www.thescipub.com/ajas.toc) American Journal of Applied Sciences 11 (8): 1398-1404, 2014 ISSN: 1546-9239 © 2014 A. Shahzad et al., This open access article is distributed under a Creative Commons Attribution (CC-BY) 3.0 license doi:10.3844/ajassp.2014.1398.1404 Published Online 11 (8) 2014 (http://www.thescipub.com/ajas.toc) Science Publications ABSTRACT Industrial Control System (ICS) has been designed for critical infrastructure sectors and 90% of these systems are property/owaned and operated by private organizations. Industrial Control System (ICS) were designed to fulfill the basic requirements included system performance and reliability and other basic needs, related with real time transmission, without interlinking with networks (public/private) or/and internet connectivity. In this research, detail review has been conducted which is based on Industrial control system or ICS types and their uses and/or importance within industries or real time industries. The remaining sections highlighted the potential problems or issues which are linked with these systems during communication and detail problem statement has been also conducted and several existing security deployments are reviewed, to find the generic security mechanism that will secure the communication of critical infrastructures. A REVIEW: INDUSTRIAL CONTROL SYSTEM (ICS) AND THEIR SECUITY ISSUES Received 2014-01-14; Revised 2014-05-13; Accepted 2014-06-24 1.3. Distributed Control System (DCS) Distributed Control System (DCS) is another part of Industrial Control System (ICS) and uses to control and monitor industrial production included processing sectors such as water distribution and treatment plants, power generation stations, wastewater collection and treatment plants, fabrication and refining plants, wind farms stations, oil refining station, gas collection and pumping stations, electrical power houses and air conditioning and heating ventilation plants/systems. Distributed Control System (DCS) is usually used control loopor upervisory station also contain control loops and intermediate control for the purpose of tasks distributio, to manage the processes/tasks, that are distributed locally between the controllers within DCS network. Distributed Control System (DCS) gather all information from these localized controllers and then produce whole production or processing execution results. DCS applications are distributed among several controllers (or computers) to minimizes the load on each controller or/and on main controller (or main server). Basic implementation of Distributed Control System (DCS) is comparatively same as SCADA system but in production phase, application or tasks are distributed among several localized controllers such that each controller has assigned function from supervisory controller. Supervisory controller or master controller initial the request and send to field devices. On response, localized controllers generate the results according to supervisory control request and collect data/information from field devices then send response back to main server or supervisory controller (Stouffer et al., 2006). SCADA system provides supervisor control over the field devices and monitors the entire communication from center location, usually by Human Machine Interface (HMI) software. SCADA system has been used various types of media included radio signals, telephone line, cable connection, satellite and micro waves media for communication between field devices located at distance placed. 1.2. Supervisory Control And Data Acquisition (SCADA) System SCADA systems or fields devices are geographical distributed in different locations and monitor/control by centralized control center using human machine interface and utilized for critical processes sectors included water distribution and treatment plants, power generation stations, water collection and treatment plants, fabrication and refining plants, wind farms stations, telecommunication systems, oil refining station, gas collection and pumping stations, electrical power houses, airports monitoring and controlling systems, ships monitoring and controlling systems, space monitoring and controlling stations and air conditioning and heating ventilation plants/systems. In SCADA system, data/information is collected from devices or actuators/sensors connected within network and this information will proceed to master control center for monitoring and controlling purposes. During SCADA communication, information has been visualized in the form of text or graphical representation, thereby visualization facilitates are located and operator at control center for controlling and monitoring the real time environment usually, automatically or by commands operation. In Fig. 1, SCADA system provides communication between the field devices such as Master Terminal Unit (MTU), Remote Terminal Units (RTUs) and Programmable Logic Controllers (PLCs) and entire communication is monitor and operator from center control using various type of communication networks included Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN) and SCADA system has been also deployed wireless technologies such as for communication between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs) or/and field devices (Shahzad et al., 2014a; Musa and Aborujilah, 2013a). The following detail below depicted the services information usually performed by SCADA system. Remote Terminal Units (RTUs) are use to collect information/data from sensors/actuators, that are connected with physical environment and transmits the information to control center or Master Terminal Unit (MTU) for monitoring purposes. Network topology is deployed as static manners in SCADA system and networks nodes are know in advance, for communication between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs). Science Publications 1.2. Supervisory Control And Data Acquisition (SCADA) System ation Technology (MIIT), University Kuala Lumpur, Malaysia Supervisory Control And Data Acquisition (SCADA) system is real time Industrial Control System (ICS), usually uses for monitoring and controlling industrial processes between field devices connected within Science Publications 1398 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 SCADA network. SCADA systems or fields devices are geographical distributed in different locations and monitor/control by centralized control center using human machine interface and utilized for critical processes sectors included water distribution and treatment plants, power generation stations, water collection and treatment plants, fabrication and refining plants, wind farms stations, telecommunication systems, oil refining station, gas collection and pumping stations, electrical power houses, airports monitoring and controlling systems, ships monitoring and controlling systems, space monitoring and controlling stations and air conditioning and heating ventilation plants/systems. In SCADA system, data/information is collected from devices or actuators/sensors connected within network and this information will proceed to master control center for monitoring and controlling purposes. During SCADA communication, information has been visualized in the form of text or graphical representation, thereby visualization facilitates are located and operator at control center for controlling and monitoring the real time environment usually, automatically or by commands operation. In Fig. 1, SCADA system provides communication between the field devices such as Master Terminal Unit (MTU), Remote Terminal Units (RTUs) and Programmable Logic Controllers (PLCs) and entire communication is monitor and operator from center control using various type of communication networks included Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN) and SCADA system has been also deployed wireless technologies such as for communication between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs) or/and field devices (Shahzad et al., 2014a; Musa and Aborujilah, 2013a). The following detail below depicted the services information usually performed by SCADA system. between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs). between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs). Remote Terminal Units (RTUs) are use to collect information/data from sensors/actuators, that are connected with physical environment and transmits the information to control center or Master Terminal Unit (MTU) for monitoring purposes. Network topology is deployed as static manners in SCADA system and networks nodes are know in advance, for communication between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs). between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs). SCADA network. Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 Several solutions were developed to secure SCADA communication, but mostly were based on physical security and limited-communication security using secure socket layer or SSL/internet protocol or IP security. But these solutions have also number of limitations, while deploying within SCADA communication because these solutions have been depending on cryptography algorithms. So, current research proposes the solution that has been developed successfully within SCADA system and successfully secure the SCADA communication between Master Terminal Unit (MTU) and Remote Terminal Units (RTUs) or/and Remote Terminal Units (RTUs) and Master Terminal Unit (MTU). Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 (PLC), for controlling overall network architecture. PLCs are mostly uses for data/information collection from physical environment and upon collection, process the information back to master station based on master station request. Remote Terminal Units (RTUs) within SCADA system are used as PLCs to collect data/information from sensors/actuators and transmit back to master station for the purposes of controlling and monitoring. At other side, field controllers or local controllers perform functions or uses as Programmable Logic Controllers (PLCs) within Distributed Control System (DCS). Local controllers are collect data/information from field devices and send response back to master controller or supervisory controller. Usually, all types of Programmable Logic Controllers (PLCs) have their own memory or storage area for storing information related with instructions being execute or basis on master controller/master station request or functions implementation such as input/output controlling functions, session management, arithmetic and logical functions, alarm controlling function and processing of data/information, (Shahzad et al., 2013; 2014b). connectivity. Traditionally, SCADA systems were connected with proprietary hardware/software and protocols. With the revolution of advance I.T infrastructure, SCADA systems are also move/change from traditional network to advance networks or open standards network protocols, rather than proprietary such as LAN/WAN through internet connectivity significantly increase the performance, reliability and scalability of system (Musa and Aborujilah, 2013a; Raghini et al., 2013). With advance interconnectivity, SCADA platform has been vulnerable from several kinds of communication and cyber attacks and threads. (PLC), for controlling overall network architecture. PLCs are mostly uses for data/information collection from physical environment and upon collection, process the information back to master station based on master station request. Remote Terminal Units (RTUs) within SCADA system are used as PLCs to collect data/information from sensors/actuators and transmit back to master station for the purposes of controlling and monitoring. At other side, field controllers or local controllers perform functions or uses as Programmable Logic Controllers (PLCs) within Distributed Control System (DCS). Local controllers are collect data/information from field devices and send response back to master controller or supervisory controller. Usually, all types of Programmable Logic Controllers (PLCs) have their own memory or storage area for storing information related with instructions being execute or basis on master controller/master station request or functions implementation such as input/output controlling functions, session management, arithmetic and logical functions, alarm controlling function and processing of data/information, (Shahzad et al., 2013; 2014b). Science Publications 1.4. Programmable Logic Controllers (PLC) Control center such as master terminal station is uses as centralized controller to control and monitor the field devices in SCADA network such as communication Supervisory Control And Data Acquisition (SCADA) system and Distributed Control Systems (DCS), both of them have been using Programmable Logic Controllers 1399 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 1.5. Background Problems: SCADA System SCADA systems have been geographical distributed across different locations over the world using Wide Area Network (WAN) technology. SCADA systems have been connected with numbers of remote terminal devices or PLCs through several types of networks such as LAN/WAN, protocols and transmission media such as wire/wireless. The great enhancement within SCADA, connectivity with several advance networks and used of advance I.T infrastructures, brought SCADA communication more demandable for end users. SCADA uses centralized station with advance I.T infrastructure, therefore able to control thousand of remote terminal stations or field devices at the same time without limitation of networks and protocols or open standards. At the other side, large interconnectivity of open standards networks, protocols and uses of open I.T infrastructure within SCADA system, made SCADA platform more vulnerable from several types of threads and attacks (Stouffer et al., 2006; Musa and Aborujilah, 2013b). More detail related with SCADA vulnerabilities and threats are depicted below. Traditionally, SCADA systems have several characteristics, risks and priorities that are quite different from internet based communication systems such as characteristics, risks and priorities and have different specifications for communication such as network and protocol requirements. There are few considerations that must be taking placed, when traditional SCADA infrastructure is replaced with current communication infrastructure by using of open standards protocols and networks such as performance such as session/time management, availability such as expected/unexpected results management, management of risk and disaster, infrastructure security issues, processes consequences, communication response management, operation management, Resource availability and management, protocols and media management and replacement of field devices, devices life session, permission to access and organization support. Supervisory Control And Data Acquisition (SCADA) systems were designed to fulfill the basic requirements such as system performance and reliability and other basic needs related with SCADA system operations of real time industrial infrastructure, without interlinking with networks such as public/private and internet There are numbers of threads that disturb or intercept the SCADA communication such as communication attackers inside/outside organization, attacker or bot- network, attacker using spam, attacker using phishing, attacker using spyware and attacker using malware. 1400 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 Fig. 1. SCADA system communication Fig. 1. 1.5. Background Problems: SCADA System SCADA system communication The vulnerabilities such as installation and configuration of networks or inappropriate, communication architecture, password policy, system authentication and authorization, nonexistence of intrusion detection and prevention system, nonexistence software/hardware firewall and cryptography protection are usually located within SCADA system, that make communication more unsecure. them such as eavesdropping, data modification, data replay, key distribution and other generic attacks (Anandkumar and Jayakumar, 2012; Manikandan and Manimegalai, 2012). Based on review, all existing solutions (generic security solutions) have been based on cryptography techniques such as encryption, digital signature and hashing algorithms, for the purpose of secure data/message communication between SCADA nodes (Hong and Lee, 2008; Bhaya and AlAsady, 2012). After conducting the detail analysis, that has been based on SCADA communication security issues such as threads, vulnerable platforms and nonappropriates security policies and solutions analysis within SCADA communication. The security solution has been developed successfully within SCADA system and successfully secures the SCADA communication and gives research directions to overcome the security issues that are warming SCADA communication (Musa and Aborujilah, 2013a). SCADA communication has been vulnerable from several types of cyber attacks; by rapidly increasing SCADA system connectivity with IP based protocols or open standard protocols. Based on SCADA security, cryptography solutions such as Symmetric and Asymmetric have been deployed to achieve the security services goals such as data confidentiality, data authentication, data integrity and non-repudiation function and secure SCADA communication long run (Drahansky and Balitanas, 2011; Majdalawieh et al., 2006). Science Publications 1.6. Problem Statements: SCADA System The problem statement has been conducted from detail study, based on existing SCADA security issues such as threads/attacks and vulnerabilities. More detail related with SCADA systems and protocols security issues and challenges are depicted below. SCADA systems are vulnerable from cyber attacks. Several existing solutions address the security related with SCADA communication with limitations. Four main components have been highlighted for SCADA security issues such as authenticity, availability, integrity, confidentiality. SCADA system has been reduced the risks, gain control and provides secure communication over attacks/threads using cryptography solution or module (AGA, 2003; Aris et al., 2011 ). Asymmetric and symmetric solutions are deployed within several Several security issues and challenges have been reviewed from existing SCADA implementations. According to the review analysis, there are no proper solutions that SCADA communication and overcome completely secure the the security issues related with 1401 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 networks and successfully achieved the security services included authentication, integrity, non repudiation and confidentiality as main part of SCADA security protection. networks and successfully achieved the security services included authentication, integrity, non repudiation and confidentiality as main part of SCADA security protection. need a solution that significantly secures the critical infrastructure communication, while connected with open standard networks or protocols (Pollet, 2002). SCADA system vendors and developers have been only focusing on functional parts of SCADA system such scalability, reliability, performance and access control without security consideration in mind. There is no generic solution available that fulfill the requirements of SCADA system security. All SCADA functional performances are depending on security issues, if SCADA system is fully secure then whole system performances would absolutely achieve (Rautmare, 2011). SCADA system vendors and developers have been only focusing on functional parts of SCADA system such scalability, reliability, performance and access control without security consideration in mind. There is no generic solution available that fulfill the requirements of SCADA system security. All SCADA functional performances are depending on security issues, if SCADA system is fully secure then whole system performances would absolutely achieve (Rautmare, 2011). SSL/TLS and IP based security solutions have been implemented within several traditional networks applications or/and SCADA communication network or protocols. 1.6. Problem Statements: SCADA System SSL/TLS and IP based solutions have number of issues related with their communication and security, included running on Transport Control Protocol (TCP), base on cryptography algorithms for security purposes and security mechanism is limited for non repudiation function and other unavailable advance security features (Patel and Graham, 2009; Preneel, 1993). SCADA system implementations using control protocols such Distributed Network Protocol (DNP3), fieldbus, modbus and other IP based protocols are harmful and critical for SCADA communication between field devices. These protocols have been designed without any security concerned that fully or partially provides protection against cyber attacks and threads. Several firewalls have been used between SCADA system and corporate networks or internet but unable to fully integrate with SCADA networks, such as in term of SCADA protocols such as DNP3 or Modbus development and configuration. So, lack of security information and protocols configuration unawareness, rapidly increasing more vulnerabilities for SCADA platform and causing major security issues for critical infrastructures (Cai et al., 2008; Shahzad et al., 2013). SCADA security issues such as no proper authentication mechanism for SCADA system in the term of designed and processing or operation, uses of proprietary or vendors protocol with open standards Protocols (TCP/IP), trust on physical security concepts and performance decreases over internet with several vulnerabilities. From these security issues; cryptography solution such as asymmetric using ECC algorithm and symmetric using AES algorithm has been deployed within SCADA communication end-to-end and achieved security services such as data confidentiality, data authentication, data integrity and non-repudiation function. As “Schweitzer Engineering Laboratories, Inc” suggested that cryptography solutions is best approaches to overcome the SCADA security issues during communication (Risley and Ladow, 2003). DNP3 is most important protocol uses within SCADA system. DNP3 is uses almost all over the world; approximately 70% in America within electric and water utilities and remaining 30% in other parts of the world such as Europe, Asia and Australia (TD, 2011). Securing DNP3 protocol or security deployment within DNP3 protocol, significantly enhance the security of SCADA system and reduce the potential attacks and vulnerabilities within communication. Based on SCADA vulnerabilities; homeland security (department) has been used cryptography mechanism (solution) to secure “Nation”s critical infrastructure” from cyber attacks/threads. As conclusion, cryptography solutions are the best approaches to secure or protect SCADA communication over internet and successfully decrease the risks (Shahzad et al., 2014a; Asenjo, 2005; Babu and Singh, 2013). 1.6. Problem Statements: SCADA System Advance Encryption Standard (AES) 256, HMAC and MD5 as part of cryptography solutions have been deployed, to protect SCADA communication, while intrusions (anomaly) are detected by Intelligent Electronic Device (IED) as part of substation controller (Musa and Aborujilah, 2013a; Hong, 2010; AL-Saidi et al., 2011). Science Publications 4. REFERENCES AGA, 2003. Cryptographic protection of SCADA communications. American Gas Association's. pp Manikandan, S.P. and R. Manimegalai, 2012. Survey on mobile ad hoc network attacks and mitigation using routing protocols. Am. J. Applied Sci., 9: 1796- 1801. DOI: 10.3844/ajassp.2012.1796.1801 AL-Saidi, N.M.G., M.R.M. Said and A.M. Ahmed, 2011. Efficiency analysis for public key systems based on fractal functions. J. Comput. Sci., 7: 526- 532. DOI: 10.3844/jcssp.2011.526.532 Musa, S. and A. Aborujilah, 2013a. Simulation base implementation for placement of security services in real time environment. Proceedings of the 7th International Conference on Ubiquitous Information, (CUI ’13), ACM, New York, USA. DOI: 10.1145/2448556.2448587 Anandkumar, K.M. and C. Jayakumar, 2012. Pro-active prevention of clone node attacks in wireless sensor networks. J. Comput. Sci., 8: 1691-1699. DOI: 10.3844/jcssp.2012.1691.1699 Aris, S., A. Messai, M. Benslama, M. Nadjim and M.M- Elharti, 2011. Integration of quantum cryptography through satellite networks transmission. Am. J. Applied Sci., 8: 71-76. DOI: 10.3844/ajassp.2011.71.76 Musa, S. and A. Aborujilah, 2013b. Secure security model implementation for security services and related attacks base on end-to-end, application layer and data link layer security. Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication, (MC’13), ACM, New York, USA. DOI: 10.1145/2448556.2448588 Asenjo, 2005. A retrofit security solution for SCADA communications and maintenance port access Authentication. Encryption Key Manage. Patel, G. and J.H. Graham, 2009. Improving the cyber security of SCADA communication networks. Commun. ACM., 52: 139-142. DOI: 10.1145/1538788.1538820 Babu, A.M. and K.J. Singh, 2013. Performance evaluation of chaotic encryption technique. Am. J. Applied Sci., 10: 35-41. DOI: 10.3844/ajassp.2013.35.41 Pollet, J., 2002. Developing a solid SCADA security strategy. Proceedings of the 2nd IEEE Sensors Conference for Industry, Nov. 19-21, IEEE Xplore Press, pp: 148-156. DOI: 10.1109/SFICON.2002.1159826 Bhaya, W.S. and S.A. AlAsady, 2012. Prevention of spoofing attacks in the infrastructure wireless networks. J. Comput. Sci., 8: 1769-1779. DOI: 10.3844/jcssp.2012.1769.1779 Cai, N., J. Wang and X. Yu, 2008. SCADA system security: Complexity history new develop. Proceedings of the 6th IEEE International Conference on Industrial Informatics, Jul. 13-16, IEEE Xplore Press, Daejeon, pp: 569-574. DOI: 10.1109/INDIN.2008.4618165 Preneel, B., 1993. Cryptographic hash functions: An overview. Proceedings of the 6th International Computer Security and Virus Conference, (VC’ 93), pp: 1-22. Raghini, M., N.U. Maheswari and R. Venkatesh, 2013. Overview on key distribution primitives in wireless sensor network. J. Comput. Sci., 9: 543-550. DOI: 10.3844/jcssp.2013.543.550 Drahansky and M. Balitanas, 2011. Cipher for internet- based supervisory control and data acquisition architecture. J. 2. CONCLUSION The detail literature has been reviewed which is based on Industrial Control System (ICS) deployment, their main architecture and the importance within industries. The security issues have been highlighted that are warming the communication and the existing security mechanisms are also reviewed that are useful to protect the communication from attacks and make Industrial Control System (ICS) platform secure against vulnerabilities. In future work, the The “American National Security Agency” has been suffering from potential attacks and hackers that are serious problems for critical infrastructure sectors. So, 1402 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 Hong, S., 2010. Experiments for embedded protection device for secure SCADA communication. Proceedings of the Power Energy Engineering Conference, Mar. 28-31, IEEE Xplore Press, Chengdu, pp: 1-4. DOI: 10.1109/APPEEC.2010.5448606 cryptography based strong security solution will implement to protect the Industrial Control System (ICS), while connecting with several open networks. 3. ACKNOWLEDGMENT Majdalawieh, F., P. Presicce and D. Wijesekera, 2006. DNPSec: Distributed Network Protocol Version 3 (DNP3) Security Framework. In: Advances in Computer Information and Systems Sciences and Engineering, Khaled E., T. Sobh, A. Mahmood, M. Iskander and M. Karim (Eds.)., Springer, Netherlands, ISBN-10: 978-1-4020-5260-6, pp: 227-234. I would like to thank to my parents and my friend, M. Irfan boosted me morally and provided me great information resources. Science Publications Science Publications 4. REFERENCES Security Eng., 6: 337-348. Rautmare, S., 2011. SCADA system security: Challenges and recommendations. Proceedings of the 1st IEEE India Conference, Dec. 16-18, IEEE Xplore Press, Hyderabad, pp: 1-4. DOI: 10.1109/INDCON.2011.6139567 Hong and S.J. Lee, 2008. Challenges and perspectives in security measures for the SCADA system. Proceedings of the 5th International Conference, Myongji-Tsinghua University Joint Seminar, (IC’ 08). Science Publications 1403 AJAS Shahzad, A. et al. / American Journal of Applied Sciences 11 (8): 1398-1404, 2014 Risley, J. and P. Ladow, 2003. Electronic security of real-time protection and scada communications. Schweitzer Engineering Laboratories Inc. Shahzad, S., S. Musa, A. Aborujilah, M.N. Ismail and M. Irfan, 2013. Conceptual model of real time infrastructure within cloud computing environment. Int. J. Comput. Netw., 5: 18-24. Shahzad, A.A., S. Musa, A. Aborujilah and M. Irfan, 2014a. Industrial Control Systems (ICSs) vulnerabilities analysis and SCADA security enhancement using testbed encryption. Proceedings of the 8th International Conference on Ubiquitous Information Management and Communication, Jan. 09-11, ACM, New York. DOI: 10.1145/2557977.2558061 Stouffer, K., J. Falco, K. Scarfone, K. Stouffer and J. Falco, 2006. Guide to Supervisory Control and Data Acquisition (SCADA) and industrial control systems security. Pennsylvania State University. TD, 2011. Increases in substation automation, integration program spending reported by North American Utilities. Transmission and Distribution. Shahzad, A., A. Aborujilah and M. Irfan, 2014b. A new cloud based supervisory control and data acquisition implementation to enhance the level of security using testbed. J. Comput. Sci., 10: 652-659. DOI: 10.3844/jcssp.2014.652.659 Science Publications 1404 AJAS Science Publications
https://openalex.org/W4320816269	https://zenodo.org/records/7639043/files/WJBPHS-2022-0231.pdf	English	null	Management of acute ischemic stroke	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	3,833	2. Early detection Ischemic stroke can occur both in the community and in the hospital and must be recognized by bystanders and/or providers. Early Recognition activates a stroke-specific chain of survival (3). Stroke is a clinical diagnosis and several features of the patient’s Clinical presentation can be used to identify stroke patients. Emergency Medical Systems are key in detection, triaging, and Transport of stroke patients to receiving facilities.4 Abstract Stroke is a major cause of mortality and morbidity, and thrombolysis has served as a catalyst for major changes in the management of acute ischaemic stroke. Intravenous alteplase (recombinant tissue plasminogen activator) is the only approved thrombolytic agent at present indicated for acute ischaemic stroke. Recombinant tissue plasminogen activator (rt-PA) therapy is effective in reducing early and long-term neurologic disabilities if it is started quickly. This article summarizes the recent advances in thrombolysis for acute ischaemic stroke. Keywords: Cerebral edema; Penumbra; Secondary neuronal injury; Stroke, rt-PA 1. Introduction Stroke is defined by the World Health Organization as a clinical Syndrome consisting of rapidly developing clinical signs of focal (or global in case of coma) disturbance of cerebral function lasting More than 24 hours or leading to death with no apparent cause Other than a vascular origin.1 Stroke is classified broadly into three Categories; ischemic stroke, hemorrhagic stroke and subarachnoid Haemorrhage. Ischemic stroke occurs due to blockage of blood vessel Which limits the blood supply to the brain whereas hemorrhagic Stroke occurs due to rupture of blood vessel leading spillage of Blood in the intracranial cavity.2 Depending on the site of blood Spillage the hemorrhagic stroke could be classified as intracerebral Haemorrhage or subarachnoid haemorrhage.2 Approximately 60–80% Of all strokes is ischemic. This article is dedicated to acute ischemic Strokes and its management.3 Kartik Soni , Rajesh Asija and Rashmi Khanijau Kartik Soni , Rajesh Asija and Rashmi Khanijau Department of Pharmacology, Maharishi Arvind Institute of Pharmacy, Jaipur, Rajasthan, India. World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 Publication history: Received on 26 October 2022; revised on 12 December 2022; accepted on 14 December 2022 Article DOI: https://doi.org/10.30574/wjbphs.2022.12.3.0231 Article DOI: https://doi.org/10.30574/wjbphs.2022.12.3.0231  Corresponding author: Kartik Soni Copyright © 2022 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Management of acute ischemic stroke Kartik Soni , Rajesh Asija and Rashmi Khanijau 4. Prehospital management Workflows and organized systems of care can efficiently reduce Delays in time to treatments. With the deployment of Mobile stroke units (MSUs) equipped with CT scanners and Telemedicine links, recognition of patients and administration of treatments may be more precise and efficient. Recent studies have shown that the implementation of MSUs has led to higher rates and reduced the time to IV-tPA administration and Door-to-needle time compared with regular ambulance trans-Ports to emergency departments (EDs) (4–8). In theory, initiation of therapies for intracerebral haemorrhaged (ICH) such as Blood pressure control and reversal of anticoagulation may also be implemented at the prehospital setting. In addition to clinical Examination with conventional scales such as the Neurological Institutes of Health Stroke Scale (NIHSS), several prehospital Scales and prompt recognition of severe strokes with large vessel Occlusions (LVOs) have successfully been validated.6 3. Epidemiology Stroke is the second most common cause of mortality and third most common cause of disability worldwide.3Globally, 68% of all Strokes are ischemic and 32% are hemorrhagic. Numbers from The USA differ a little with 87% of all strokes being ischemic, 10% Hemorrhagic and about 3% being subarachnoid hemorrhage.5 Data regarding prevalence of stroke in India is lacking however, It can be extrapolated from the data available from the West. In a Study done by Banerjee et al. In 2001 crude prevalence rate of Stroke in India was 147/100,000 and the annual incidence rate Was 36/100,000. Women had substantially higher age-adjusted Prevalence rate (564/100,000 for women vs 196/100,000 for men) And incidence rate (204/100,000 for women vs 36/100,000 for Men). Overall prevalence of stroke ranges from 147– Copyright © 2022 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. or(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 922/100,000 in various studies.8, 9India has the highest burden of acute coronary syndrome (ACS) in the world and the three most common risk factors for ACS are Smoking (40%), high blood pressure (38%), and diabetes (30%). Taking account of the above mentioned data and considering the fact that stroke shares common risk factors with ACS, we can safely Assume that India has a very high incidence of stroke as well.5 5. Ischemic stroke classification7 o the multicentre Trial of Acute Stroke Treatment (TOAST) there are three kinds of ischemic stroke rge vessel stroke According to the multicentre Trial of Acute Stroke Treatment (TOAST) there are three kinds of ischemic stroke:  Large vessel stroke  Large vessel stroke  Small vessel stroke or Lacunar stroke  Cardio embolic stroke large artery strokes could be due to thrombotic or embolic occlusion of the major arteries of the brain like the internal carotid artery, middle cerebral artery, anterior cerebral artery or the vertebrobasilar system. Lacunar strokes are more often due to involvement of smaller or perforating blood vessels supplying the deeper structures of the brain.  Cardio embolic stroke large artery strokes could be due to thrombotic or embolic occlusion of the major arteries of the brain like the internal carotid artery, middle cerebral artery, anterior cerebral artery or the vertebrobasilar system. Lacunar strokes are more often due to involvement of smaller or perforating blood vessels supplying the deeper structures of the brain. 6. Risk factors 8 Non Modifiable Risk Factors include • Age • Race • Sex • Ethnicity • History of migraine headaches • Fibromuscular dysplasia • Heredity: Family history of stroke or transient ischemic attacks (TIAs). • Hypertension • Diabetes mellitus • Cardiac disease (see below) • High cholesterol • Previous stroke • Carotid stenosis • Hyperhomocystinemia • Lifestyle issues: Excessive alcohol intake, tobacco use, illicit drug Use, physical inactivity • Obesity • Oral contraceptive use/postmenopausal hormone use Majority of the ischemic strokes seen in patients with cardiovascular disease are embolic. Embolic strokes may arise directly from the heart or the aorta. Following is the list of conditions that carry a high risk for embolic strokes.9 Majority of the ischemic strokes seen in patients with cardiovascular disease are embolic. Embolic strokes may arise directly from the heart or the aorta. Following is the list of conditions that carry a high risk for embolic strokes.9 211 World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 7. Management of acute stroke • Head/gaze deviation: Absent (0) and present (1). 7.2. Befast, detection of stroke • Time is brain, time to activate stroke system and stroke clock.13 7. Management of acute stroke The most important factor in the management of acute ischemic Stroke is time. The patient with ischemic stroke loses 190, 0000 brain Cells every minute, about 14000,000,000 nerve connections are destroyed every minute and 12 km (7.5 miles) of nerve fibres are lost every minute. The brain ages 3.6 years for every hour it is deprived of blood supply. There are two modalities of treatment available for treatment of acute ischemic stroke. Intravenous thrombolysis and mechanical thrombectomy.10 y mechanical thrombectomy.10 7.1. 8 D’s of stroke care 7.1.1. Detection Involves recognizing the signs and symptoms of an acute stroke 7.1.2. Dispatch Activation of emergency medical services. In most Cases, this involves calling 911 or a stroke team 7.1.3. Delivery Means prompt transport of the patient to a hospital, preferably a stroke center or to a setting in the hospital for Further evaluation by a stroke team. 7.1.4. Door This refers to the arrival of the patient at the ED. According to recommendations from the National Institute of Neurological Disorders and Stroke, an assessment should Be completed by an ED physician within 10min of arriving in the ED 7.1.5. Data Data collection includes results from laboratory tests and Both a physical and a neurologic examination (Neurological Institutes of Health Stroke Scale) 7.1.6. Decision Information, such as the type of stroke, last seen Normal, and time from onset of symptoms, is considered before a treatment decision is made. 7.1.7. Drug/device Fibrinolytic therapy should be administered within 4.5hr of the onset of symptoms. Even if the patient is not a Candidate for fibrinolysis, they may still qualify for endovascular therapy to remove mechanically a clot. 7.1.8. Disposition It is recommended that patients are admitted to An ICU or stroke unit within 3hr of arrival in the ED.11 7.2. Befast, detection of stroke • Balance, acute or sudden onset of loss of balance or Coordination. • Eyes, blurred or unclear vision, double vision, and gaze preference . • Facial weakness or facial asymmetry.12 • Arm and/or leg weakness. • Speech difficulty/slurring of speech. • Time is brain, time to activate stroke system and stroke clock.13 7.3. Arterial Occlusion Evaluation Scale • Facial palsy: Absent (0), mild (1), and moderate (2). • Arm motor impairment: Normal to mild (0), moderate (1), and severe (2). • Leg motor impairment: Normal to mild (0), moderate (1), and severe (2). 7.1. * 8 D’s of stroke care 7.1.1. Detection Involves recognizing the signs and symptoms of an acute stroke 7.1.2. Dispatch Activation of emergency medical services. In most Cases, this involves calling 911 or a stroke team 7.1.3. Delivery Means prompt transport of the patient to a hospital, preferably a stroke center or to a setting in the hospital for Further evaluation by a stroke team. 7.1.4. Door This refers to the arrival of the patient at the ED. According to recommendations from the National Institute of Neurological Disorders and Stroke, an assessment should Be completed by an ED physician within 10min of arriving in the ED 7.1.5. Data Data collection includes results from laboratory tests and Both a physical and a neurologic examination (Neurological Institutes of Health Stroke Scale) 7.1.6. Decision Information, such as the type of stroke, last seen Normal, and time from onset of symptoms, is considered before a treatment decision is made. 7.1.7. Drug/device Fibrinolytic therapy should be administered within 4.5hr of the onset of symptoms. Even if the patient is not a Candidate for fibrinolysis, they may still qualify for endovascular therapy to remove mechanically a clot. 7.1.8. Disposition It is recommended that patients are admitted to An ICU or stroke unit within 3hr of arrival in the ED.11 7.2. Befast, detection of stroke • Balance, acute or sudden onset of loss of balance or Coordination. • Eyes, blurred or unclear vision, double vision, and gaze preference . • Facial weakness or facial asymmetry.12 • Arm and/or leg weakness. • Speech difficulty/slurring of speech. • Time is brain, time to activate stroke system and stroke clock.13 7.3. Arterial Occlusion Evaluation Scale • Facial palsy: Absent (0), mild (1), and moderate (2). • Arm motor impairment: Normal to mild (0), moderate (1), and severe (2). • Leg motor impairment: Normal to mild (0), moderate (1), and severe (2). • Head/gaze deviation: Absent (0) and present (1). 7.1.1. Detection 7.1.1. Detection Involves recognizing the signs and symptoms of an acute stroke Involves recognizing the signs and symptoms of an acute stroke 8. Endovascular treatment for stroke Intravenous tPA has been the backbone of stroke treatment for almost 20 years. Intravenous therapy although effective has specific inclusion and exclusion criteria which limits its use in a large number of patients. Moreover the intravenous tPA cannot be used after the 4.5 hour window and has limited efficacy in patients with large vessel occlusion. Riedel et al showed that it is almost impossible to dissolve large clots with intravenous therapy.15 In another study by Alexandrov and Grotta up to one third of patients had re-occlusion of blood vessels after initial recanalization.The FDA approval of IV-tPA has innovated the entire field of Emergency neurology.16 However, up to 69% of stroke patients Are ineligible to receive IV-tPA due to delayed hospital presentation.16 Over the last 3 years, the time window for AIS Treatment has expanded thanks to EVT and has provided physicians with a stronger therapeutic arsenal. The success of EVT Is measured by the degree or quality of revascularization. The Thrombolysis in Cerebral Infarction (TICI) scale is a tool to standardize the different degrees of reperfusion ranging from no Perfusion (TICI 0) to complete perfusion (TICI 3).17 TICI scores of 2B to 3 are usually regarded as successful reperfusion. Previous studies failed to show improved results with EVT and diminished the initial optimism regarding intervention for AIS. However, the study design of those clinical trials was criticized for not requiring the image proof of LVO, using Older technology for clot retrieval, and having prolonged stroke to puncture times.18 Since, multiple trials have shown the efficacy of EVT in addition to standard medical care in improving the overall outcome of AIS patients with proximal MCA or internal carotid artery (ICA) occlusion when EVT was performed Within either 6 hours, 8 hours (42), or 12 hours Of symptom onset.19 World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 • Aphasia: Performs tasks correctly (0), performs one task correctly (1), and performs neith • Aphasia: Performs tasks correctly (0), performs one task correctly (1), and performs neither task (2). • Aphasia: Performs tasks correctly (0), performs one task correctly (1), and performs neither task (2). • Agnosia: Recognizes his/her arm and deficit (0), does recognize his/her arm but not or deficit (1), and does not recognize his/her arm or deficit (2).14 • Agnosia: Recognizes his/her arm and deficit (0), does recognize his/her arm but not or deficit (1), and does not recognize his/her arm or deficit (2).14 9.2. Blood pressure As part of cerebral autoregulation, blood pressure is commonly elevated during the acute phase of AIS, maximizing perfusion in the ischemic areas. However, severe hypertension can lead to hemorrhagic transformation of the infarct, hypertensive encephalopathy, as well as cardiopulmonary and renal complications. Current AHA/ASA guidelines recommend permissive hypertension with a blood pressure goal of less than or equal to 220/120 mm Hg for the first 24–48 hours. Yet, these blood pressure variables only apply if the patient is not undergoing any acute intervention such as IV-tPA or EVT. If the patient receives IVtPA, the risk of hemorrhagic transformation increases and the blood pressure should be lowered to less than or equal to 185/110 mm Hg prior to IV-tPA administration and to less than or equal to 180/105 mm Hg once IV-tPA has been given. Reperfusion injury and hemorrhagic transformation are of concern in the case of EVT; thus, blood pressure.22 9.1. Oxygenation and ventilation- Supplemental oxygen may be required if a patient’s saturation is Less than 94%. Rapid neurologic Deterioration and ensuing loss of consciousness with impairment of reflexes that maintain the airway mandate definitive Airway control.20 Failure to recognize imminent airway loss may result in complications such as Aspiration, hypoxemia, and hypercapnia, which may result in Secondary neuronal injury. Hyperbaric oxygen was shown to either have no effect or be harmful in AIS patients and should be avoided.21 7.3. * Arterial Occlusion Evaluation Scale • Facial palsy: Absent (0), mild (1), and moderate (2). • Arm motor impairment: Normal to mild (0), moderate (1), and severe (2). • Leg motor impairment: Normal to mild (0), moderate (1), and severe (2). • Head/gaze deviation: Absent (0) and present (1). 212 10. Conclusion Management of acute ischemic stroke is time dependent. Efficient And effective stroke care depends on a well- functioning team from the emergency room to the neurologist and the interventional Neurologist. Accurate diagnosis, emergent management to stabilize the patient and correct choice of imaging can make a lot of difference in the outcome of a patient. Every minute lost in wrong imaging or Lab test results in decrease in functional outcome and ultimately irreversible paralysis. Success of stroke treatment is dependent on the entire team working smoothly and efficiently. 9.4. Cerebral edema Large infarcts of the MCA or ICA are associated with high Morbidity rates of up to 80%. Patients with large hemispheric Infarcts (LHIs) are at increased risk of cerebral edema and fast Neurologic deterioration that led to the term “malignant MCA Infarction”(MMI).24 The ultimate intervention to alleviate increased Intracranial pressure and avoid herniation in LHI with significant edema is surgical decompression with DHC. Three European clinical trials assessed the benefit of DHC in patients 60 years and younger. A pooled analysis of thes e trials showed that DHC does not only reduce mortality by 50% but also improve long-term functional outcome. The NNT to avoid a death is 2 (mRS = 6), whereas the NNT to avoid Death and the most severe to moderately severe disability is 4 (mRS = 4–6). The proportion of patients alive with minimalto-moderate disability (mRS = 0–3) was increased from 21% to 43%. Viewed another way, DHC resulted in a 49% absolute Risk reduction in death, and an absolute increase in the proportion of patients rated as mRS = 2 of 12%, mRS = 3 of 10%, and mRS = 4 of 29%.25 Disclosure of conflict of interest There is no conflict of interest. 9.3. Glycemic control Evidence indicates that persistent in-hospital hyperglycemia During the first 24 hours after AIS is associated with worse outcomes compared with normoglycemia due to multiple potential mechanisms, such as endothelial dysfunction, increased Oxidative stress, and impaired fibrinolysis. However, in the NINDS funded Stroke Hyperglycemia Insulin Network Effort (SHINE) clinical trial, an intensive IV insulin protocol to achieve a systemic glucose between 80 and 130mg/dL was not associated with favourable outcomes at 90 days compared with a Standard regimen of insulin in a “sliding-scale” fashion to keep the glucose between 80 and 180mg/dL. The intensive insulin protocol was associated with significant hypoglycaemic Events and a higher level of care. To this end, it is reasonable to treat hyperglycemia to achieve blood glucose levels in a range of 140–180mg/dL and to monitor closely to prevent hypoglycemia in patients with AIS.23 213 Acknowledgments I would like to thanks Ms Rashmi Khanijau (Associate Professor) and Dr. Rajesh Asija (Principal) for guiding me during this work and drafting the manuscript. I would like to thanks Ms Rashmi Khanijau (Associate Professor) and Dr. Rajesh Asija (Principal) for guiding me during this work and drafting the manuscript. References [1] The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group: Tissue plasminogen activator for acute ischemic Stroke. N Engl J Med 1995; 333:1581–1587 [2] Prabhakaran S, Ruff I, Bernstein RA: Acute stroke intervention: A systematic review. JAMA 2 [3] Jauch EC, Cucchiara B, Adeoye O, et al: Part 11: Adult stroke: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010; 122(18 Suppl 3):S818– S828 [4] Gyrd-Hansen D, Olsen KR, Bollweg K, et al: Cost-effectiveness estimate of prehospital thrombolysis: Results of the PHANTOM-S study. Neurology 2015; 84:1090–1097 [5] Ebinger M, Kunz A, Wendt M, et al: Effects of golden hour thrombolysis: A Prehospital Acute Neurological Treatment and Optimization Of Medical Care in Stroke (PHANTOM-S) substudy. JAMA Neurol 2015; 72:25–30 [6] Kunz A, Ebinger M, Geisler F, et al: Functional outcomes of pre-hospital thrombolysis in a mobile stroke treatment unit compared with conventional care: An observational registry study. Lancet Neurol 2016; 15:1035–1043 [7] Czap AL, Grotta JC, Parker SA, et al: Emergency department door-topuncture time since 2014. Stroke 2019; 50:1774–1780 [8] Ebinger M, Winter B, Wendt M, et al; STEMO Consortium: Effect Of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: A randomized clinical trial. JAMA 2014; 311:1622–1631 [9] Perez de la Ossa N, Carrera D, Gorchs M, et al: Design and validation Of a prehospital stroke scale to predict large arterial occlusion: The Rapid Arterial Occlusion Evaluation scale. Stroke 2014; 45:87–91 214 World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 World Journal of Biology Pharmacy and Health Sciences, 2022, 12(03), 210–215 [10] Higashida R, Alberts MJ, Alexander DN, et al; American Heart Association Advocacy Coordinating Committee: Interactions within Stroke systems of care: A policy statement from the American Heart Association/American Stroke Association. Stroke 2013; 44:2961–2984 [11] Meyer BC, Raman R, Hemmen T, et al: Efficacy of site-independent Telemedicine in the STRokE DOC trial: A randomised, blinded, prospective study. Lancet Neurol 2008; 7:787–795 [12] Goyal M, Menon BK, van Zwam WH, et al; HERMES collaborators: Endovascular thrombectomy after large-vessel ischaemic stroke: A Meta-analysis of individual patient data from five randomised trials. Lancet 2016; 387:1723– 1731 [13] Barlinn J, Gerber J, Barlinn K, et al: Acute endovascular treatment Delivery to ischemic stroke patients transferred within a telestroke Network: A retrospective observational study. References Int J Stroke 2017; 12:502–509 [14] Kepplinger J, Dzialowski I, Barlinn K, et al: Emergency transfer of Acute stroke patients within the East Saxony telemedicine stroke network: A descriptive analysis. Int J Stroke 2014; 9:160–165 [15] Pedragosa A, Alvarez-Sabín J, Rubiera M, et al: Impact of telemedicine on acute management of stroke patients undergoing endovascular procedures. Cerebrovasc Dis 2012; 34:436–442 [16] McAdams M, Murphy J, DePrince M, et al: Assessing the impact of Care in a telemedicine-based stroke network using patient-centered Health-related quality-of-life outcomes. European Stroke J 2016; 1:121 [17] Barber PA, Demchuk AM, Zhang J, et al: Validity and reliability of a Quantitative computed tomography score in predicting outcome of Hyperacute stroke before thrombolytic therapy. ASPECTS Study Group. Alberta Stroke Programme Early CT Score. Lancet 2000; 355:1670–1674 [18] Yu AY, Zerna C, Assis Z, et al: Multiphase CT angiography increases Detection of anterior circulation intracranial occlusion. Neurology 2016; 87:609–616 [19] Albers GW, Marks MP, Kemp S, et al; DEFUSE 3 Investigators: Thrombectomy for stroke at 6 to 16 hours with selection by perfusion Imaging. N Engl J Med 2018; 378:708–71 [20] Campbell BC, Mitchell PJ, Kleinig TJ, et al; EXTEND-IA Investigators: Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015; 372:1009–1018.21. [21] Riedel CH, Zimmermann P, Jensen-Kondering U, Stingele R, Deuschl G, Jansen O. 2011. The importance of size: successful recanalization by intravenous thrombolysis in acute anterior stroke depends on thrombus length. Stroke 42:1775–1777. [22] Alexandrov AV, Grotta J C. 2002. Arterial reocclusion in stroke patients treated with intravenous tissue plasminogen activator. Neurology 59:862–867. [23] Del Zoppo G, Higashida R, Furlan A, Pessin MS, Rowley HA, Gent M. PROACT: a phase II randomized trial of recombinant Pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke. Stroke. 1998;29:4– 11. [24] Furlan A, Higashida R, Wechsler L, Gent M, Rowley H, Kase C, et al. Intra-arterial prourokinase for acute ischemic stroke – the PROACT II study: a randomized controlled trial. JAMA. 1999;282:2003–2011. [25] Smith W, Sung G, Starkman S, Saver JL, Kidwell CS, Gobin YP, et al. Safety and efficacy of mechanical embolectomy in acute ischemic stroke: results of the MERCI trial. Stroke. 2005;36:1432–1438 215
https://openalex.org/W2086653737	https://europepmc.org/articles/pmc2360540?pdf=render	English	null	Evaluation of models to predict BRCA germline mutations	British journal of cancer	2,006	cc-by	7,592	Received 19 May 2006; revised 16 August 2006; accepted 22 August 2006 Correspondence: Professor R Ward, Department of Medical Oncology, St Vincent’s Hospital, Victoria St, Darlinghurst, NSW 2010, Australia; E-mail: r.ward@garvan.unsw.edu.au This work is original. It has been presented by Rachel Williams at the following conference: Familial Cancer 2005: Research and Practice, Couran Cove Qld, August 2005. Evaluation of models to predict BRCA germline mutations doi:10.1038/sj.bjc.6603358 www.bjcancer.com & 2006 Cancer Research UK Molecular Diagnostics Keywords: breast cancer; models; BRCA; risk; mutations restricted to individuals affected by breast or epithelial ovarian cancer, and whose family history as a whole satisfies a ‘potentially high-risk category’ (NBCC Genetics Working Group, 2000; Nelson et al, 2005). The prevalence of germline BRCA mutations in such high-risk families is estimated at 3.4–15.5% (Frank et al, 2002; Huang et al, 2002; Antoniou et al, 2003; Myriad, 2004; Nelson et al, 2005). These low figures have led to the development of models that can more accurately assess the pre-test probability of identifying a BRCA1/2 germline mutation (Couch et al, 1997; Parmigiani et al, 1998; Berry et al, 2002; Frank et al, 2002; Evans et al, 2004). Mendelian models like BRCAPRO (Parmigiani et al, 1998; Berry et al, 2002) evaluate the probability that an individual is a gene mutation carrier, while other models such as Penn (Couch et al, 1997), Manchester (Evans et al, 2004) and Frank-Myriad (Frank et al, 2002) determine the likelihood of identifying a mutation on the basis of known family history. Although the performance of some of these models has been previously examined, it is perhaps telling that no single model has been universally adopted. Previous model validation studies have considered only a subset of the available models, have not compared the results with germline testing, and have not considered the barriers to the use of models in clinical practice (Berry et al, 2002; Euhus et al, 2002; de la Hoya et al, 2003; Marroni et al, 2004; James et al, 2006; Barcenas et al, 2006). Germline mutations in BRCA1 and BRCA2 confer an estimated 65 and 45% cumulative lifetime risk of developing breast cancer, and an ovarian cancer risk of 39 and 11%, respectively (Antoniou et al, 2003). Identification of these individuals before they present with cancer is important since prophylactic surgery can reduce morbidity and mortality in these individuals (Struewing et al, 1995; Rebbeck et al, 1999; Hartmann et al, 1999; Hartmann et al, 2001; Meijers-Heijboer et al, 2001; Kauff et al, 2002; Rebbeck et al, 2004). Unfortunately, genetic testing is expensive, family history of breast cancer is common and BRCA mutations are rare (Ford et al, 1994; Peto et al, 1999; Antoniou et al, 2002). A triage tool to identify those families most likely to benefit from germline testing would, therefore, be very useful. British Journal of Cancer (2006) 95, 914 – 920 & 2006 Cancer Research UK All rights reserved 0007 – 0920/06 $30.00 British Journal of Cancer (2006) 95, 914 – 920 & 2006 Cancer Research UK All rights reserved 0007 – 0920/06 $30.00 British Journal of Cancer (2006) 95, 914 – 920 British Journal of Cancer (2006) 95, 914 – 920 & 2006 Cancer Research UK All rights reserved 0007 – 0920/06 $30.00 www.bjcancer.com Evaluation of models to predict BRCA germline mutations ang1, R Williams1, J Leary2, kConFab Investigators3, C Ringland4, J Kirk2 and R Ward,1,4,5 HH Kang1, R Williams1, J Leary2, kConFab Investigators3, C Ringland4, J Kirk2 and R Ward*,1,4,5 1Department of Medical Oncology, St Vincent’s Hospital, Sydney, New South Wales, Australia; 2Familial Cancer Service, Westmead Institute for Cancer Research at Westmead Millennium Institute, University of Sydney, Westmead, Sydney 2052, Australia; 3Kathleen Cuningham Consortium for Research into Familial Breast Cancer, Peter MacCallum Cancer Institute, St Andrews Place, East Melbourne, Victoria 3002, Australia; 4School of Medical Sciences, University of NSW, Sydney 2052, Australia; 5St Vincent’s Clinical School, University of NSW, Sydney 2052, Australia The selection of candidates for BRCA germline mutation testing is an important clinical issue yet it remains a significant challenge. A number of risk prediction models have been developed to assist in pretest counselling. We have evaluated the performance and the inter-rater reliability of four of these models (BRCAPRO, Manchester, Penn and the Myriad-Frank). The four risk assessment models were applied to 380 pedigrees of families who had undergone BRCA1/2 mutation analysis. Sensitivity, specificity, positive and negative predictive values, likelihood ratios and area under the receiver operator characteristic (ROC) curve were calculated for each model. Using a greater than 10% probability threshold, the likelihood that a BRCA test result was positive in a mutation carrier compared to the likelihood that the same result would be expected in an individual without a BRCA mutation was 2.10 (95% confidence interval (CI) 1.66–2.67) for Penn, 1.74 (95% CI 1.48–2.04) for Myriad, 1.35 (95% CI 1.19–1.53) for Manchester and 1.68 (95% CI 1.39–2.03) for BRCAPRO. Application of these models, therefore, did not rule in BRCA mutation carrier status. Similar trends were observed for separate BRCA1/2 performance measures except BRCA2 assessment in the Penn model where the positive likelihood ratio was 5.93. The area under the ROC curve for each model was close to 0.75. In conclusion, the four models had very little impact on the pre-test probability of disease; there were significant clinical barriers to using some models and risk estimates varied between experts. Use of models for predicting BRCA mutation status is not currently justified for populations such as that evaluated in the current study British Journal of Cancer (2006) 95, 914–920. doi:10.1038/sj.bjc.6603358 www.bjcancer.com & 2006 Cancer Research UK British Journal of Cancer (2006) 95, 914–920. Patient cohort Family cancer clinics at St Vincent’s and Westmead Hospitals, Sydney, provided pedigrees of families who had undergone BRCA1/2 mutation analysis in the period 1998–2004. BRCA1/2 testing was performed using the DNA of an affected individual, usually the youngest living available affected person, or an obligate gene mutation carrier. For the purposes of the study, this individual was defined as the proband. Families in which no affected individuals were available for BRCA1/2 testing were not included in the study. To be eligible for germline testing, at least one unaffected family member must have had a life time risk of breast cancer of 1 : 4 or greater as defined by the Australian National Breast Cancer (NBCC) guidelines (NBCC Genetics Working Group, 2000). This included individuals with at least two first- or second-degree relatives on one side of the family diagnosed with breast or ovarian cancer, together with additional features on the same side of the family. These features included an additional relative with breast or ovarian cancer; breast cancer diagnosed before the age of 40 years, ovarian cancer before 50 years, bilateral breast cancer, breast and ovarian cancer in the same woman, Jewish ancestry or breast cancer in a male relative. Mutation screening for BRCA1 and BRCA2 The extent to which BRCA1/2 were screened reflects the availability of testing methodologies over the study period. Full DNA sequence analysis of BRCA1/2 together with a screen for the BRCA1 exon 13 duplication (common in families of British descent) (Puget et al, 1999) was used as the minimum testing regime in 89 families. Mutation analysis on the remaining 291 families was performed by a screening strategy that involved the use of a number of different techniques. The protein truncation test (PTT) was performed for exon 11 of BRCA1 and exons 10, 11 and 27 of BRCA2. This test covers 65% of the coding region for each gene and identifies the most common types of pathogenic mutation in BRCA1/2 (Garvin, 1998). Sequence analysis or heteroduplex analysis of exons 2 and 20 of BRCA1 using either polyacrylamide gel electrophoresis or denaturing high-perfor- mance liquid chromatography were used to detect nonsense, frameshift and missense mutations within these exons. All samples were screened for the BRCA1 duplication exon 13 mutation and a subset of 125 families from the entire cohort were also screened for rare large genomic rearrangements in BRCA1 using multiplex ligation-dependent probe amplification (MLPA) (Woodward et al, 2005). Each pedigree was de-identified, with only the sex, birthday, date of death, type of cancer and age at diagnosis retained for the analysis. Not all cancers could be verified because of the requirement for client consent in Australia. Pedigrees were viewed with Progeny software v5.4.05 (Progeny Software, LLC, South Bend, IN, USA). Furthermore, families of Ashkenazi Jewish ancestry were not included in this study. This study was approved by the St Vincents Hospital Human Research Ethics Committee. Molecular Diagnostics Evaluation of models to predict BRCA germline mutations Like most countries (summarized by Nelson et al, 2005), Australia has developed guidelines (NBCC Genetics Working Group, 2000) for the referral of individuals to a family cancer clinic for counselling and further evaluation. Under these guidelines, full BRCA1/2 mutation screening is usually This study sought to compare the ability of four models (BRCAPRO, Manchester, Penn and the Myriad-Frank) to deter- mine the likelihood of finding a BRCA gene mutation in at-risk Models and germline BRCA mutations HH Kang et al Models and germline BRCA mutations 915 individuals using a large group of patients who had undergone germline testing. ambiguous or incomplete, prespecified rules were used for the entry of information into each model (Table 1). Determination of mutation likelihood using the Manchester model was performed by assigning a score on the basis of malignancy type and age of diagnosis (Evans et al, 2004). Scores obtained do not equate directly to probabilities; however, the authors suggest that a cut- off score of 10 points for BRCA1 and BRCA2 separately, or a combined score of 15 points correspond to an approximate mutation probability of 10% (Evans et al, 2004; Evans et al, 2005). The Myriad model was applied using on-line tables (Myriad, 2004) and an on-line questionnaire was used to obtain independent probabilities for BRCA1 and BRCA2 mutations using the Penn model (University of Pennsylvania Abramson Cancer Center, 2005). The latter model requires the presence of breast cancer in the family lineage and thus six families with only ovarian cancer could not be analysed. The overall chance of a mutation in either BRCA1 or BRCA2 was calculated as the probability of mutation in BRCA1 plus the probability of mutation in BRCA2 assuming there was no mutation in BRCA1. For the BRCAPRO model, carrier probabilities were calculated by entering information on the proband’s first- and second-degree relatives into CancerGene software (CaGene version 3.3, supplied by Assistant Professor D Euhus, UT Southwestern Medical Center at Dallas, Texas, USA). Use of models to identify mutation carriers Each of the four risk assessment models (Manchester (Evans et al, 2004; Evans et al, 2005), Myriad (Frank et al, 2002), Penn (Couch et al, 1997) and BRCAPRO (Parmigiani et al, 1998; Berry et al, 2002)) were applied to each pedigree by two investigators (HK and RWi). All models except Myriad allow calculations for combined BRCA mutation status, as well as BRCA1 and BRCA2 indepen- dently. Risk assessments were based on the proband, and maternal or paternal inheritance was assessed using the Progeny program (Progeny Software, LLC, South Bend, IN, USA). If the potential carrier side was not obvious, the assessments were conducted for both paternal and maternal relatives and the highest score was included in the analysis. As data on some pedigrees was BRCA1/2 sequence variants of uncertain clinical significance were identified in nine cases and these families were excluded from the analysis; however, families with nondeleterious, common polymorphisms were included in the no-mutation group. Table 1 Conventions used for interpreting ambiguous pedigrees Issue Resolution Applicable model Half siblings Considered full siblings M, P, B, My Age of cancer diagnosis unknown Assigned lowest age possible consistent with current age M Bilateral breast cancer Counted as two different cases M, My Bilateral ovarian cancer Counted as one case M, P, B, My Age of living relative unknown Estimated as 25 years younger than parent or 2 years difference from sibling B Age of death not specified Estimated as 70 years or approximated from age of death of siblings B Malignancy age 470 years, age of death unknown Death estimated as 2 years post diagnosis if cancer is the cause of death B Death in infancy Age of death recorded as 1 year B Age of cancer diagnosis unknown Estimated as 2 years before death for deceased relatives and 2 years younger than current age for living relatives B M ¼ Manchester, P ¼ Penn, B ¼ BRCAPRO, My ¼ Myriad. Use of models to identify mutation carriers British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK Table 1 Conventions used for interpreting ambiguous pedigrees Issue Resolution Applicable model Half siblings Considered full siblings M, P, B, My Age of cancer diagnosis unknown Assigned lowest age possible consistent with current age M Bilateral breast cancer Counted as two different cases M, My Bilateral ovarian cancer Counted as one case M, P, B, My Age of living relative unknown Estimated as 25 years younger than parent or 2 years difference from sibling B Age of death not specified Estimated as 70 years or approximated from age of death of siblings B Malignancy age 470 years, age of death unknown Death estimated as 2 years post diagnosis if cancer is the cause of death B Death in infancy Age of death recorded as 1 year B Age of cancer diagnosis unknown Estimated as 2 years before death for deceased relatives and 2 years younger than current age for living relatives B M M h t P P B BRCAPRO M M i d Table 1 Conventions used for interpreting ambiguous pedigrees Table 1 Conventions used for interpreting ambiguous pedigrees Proportion of positive BRCA tests varied according to carrier probability The study cohort included pedigrees from 380 families, of whom 52 (13.7%) carried deleterious mutations in either BRCA1 (34 subjects) or BRCA2 (18 subjects). The mutation frequency was higher (22.5%) in individuals subjected to complete DNA sequencing of BRCA1/2 (20 of 89 individuals) compared with those in whom sequencing had not been performed (11%, 32 of 291 subjects). j A total of 45 different mutations were recorded among the BRCA mutation positive families, whereas duplication of BRCA1 exon 13 occurred in a further five families. The mean carrier probabilities for mutation-positive individuals were 53% using BRCAPRO, 26% for Myriad and 32% using the Penn model. The probabilities for mutation-negative individuals were 24, 14, and 12%, respectively. The mean score for mutation-positive individuals using the Manchester model was 37; the mean score for mutation-negative individuals was 21. All models showed a relationship between the carrier probability and the proportion of positive tests (Table 2). Issue M ¼ Manchester, P ¼ Penn, B ¼ BRCAPRO, My ¼ Myriad. British Journal of Cancer (2006) 95(7), 914 – 920 British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK Models and germline BRCA mutations HH Kang et al Models and germline BRCA mutations HH Kang et al 916 Discrimination of BRCA mutation carriers from noncarriers Table 3 shows the sensitivity, specificity, positive and negative predictive values for each risk model at the 10% threshold. The negative predictive value of all four models was remarkably consistent (range ¼ 0.93–0.96) in that only 4–6% of families assigned a mutation probability of 10% or less actually carried a Table 2 The proportion of BRCA mutation carriers according to carrier probability as determined by the four combination models Category of carrier probability (%)a o10 10–20 20–40 40–60 60–80 480 Manchester 0/19(0.0%) 12/176(6.8%) 22/147(15.0%) 12/31(38.7%) 2/3(66.7%) 4/4(100%) BRCAPRO 12/190(6.3%) 6/48(12.5%) 4/36(11.1%) 3/24(12.5%) 9/39(27.3%) 18/43(41.9%) Myriad 8/176(4.6%) 14/115(12.2%) 17/61(27.9%) 13/26(50.0%) 0/2(0.0%) N/A Penn 16/232(6.9%) 13/74(17.6%) 8/37(21.6%) 2/12(16.7%) 6/8(75.0%) 7/11(63.6%) Six families with only ovarian cancer could not be analysed by the Penn model. N/A ¼ not applicable. aScore, rather than probability, for Manchester model. Molecular Diagnostics Table 3 Performance measures for each model at the 10% threshold Proportion of carriers by model probability (%) Test parameters at 10% threshold (95% CI) o10 X10% Sensitivity Specificity PPV NPV Combined Manchester 6/119 46/261 0.89(0.77,0.95) 0.35(0.30,0.40) 0.18(0.14,0.23) 0.95(0.89,0.98) BRCAPRO 12/190 40/190 0.77(0.64,0.86) 0.54(0.49,0.60) 0.21(0.16,0.27) 0.94(0.89,0.96) Myriad 8/176 44/204 0.85(0.73,0.92) 0.51(0.46,0.57) 0.22(0.17,0.28) 0.96(0.91,0.98) Penn 16/232 36/142 0.69(0.56,0.80) 0.67(0.62,0.72) 0.25(0.19,0.33) 0.93(0.89,0.96) BRCA1 Manchester 4/187 30/193 0.88(0.73,0.95) 0.53(0.48,0.58) 0.16(0.11,0.21) 0.98(0.95,0.99) BRCAPRO 7/225 27/155 0.79(0.63,0.90) 0.63(0.58,0.68) 0.17(0.12,0.24) 0.97(0.94,0.99) Penn 14/281 20/93 0.58(0.42,0.74) 0.79(0.74,0.83) 0.22(0.14,0.31) 0.95(0.92,0.97) BRCA2 Manchester 6/189 12/191 0.67(0.44,0.84) 0.51(0.45,0.56) 0.06(0.04,0.11) 0.97(0.93,0.99) BRCAPRO 12/308 6/72 0.33(0.16,0.56) 0.82(0.78,0.85) 0.08(0.04,0.17) 0.96(0.93,0.98) Penn 12/348 6/26 0.33(0.16,0.56) 0.94(0.92,0.96) 0.23(0.11,0.42) 0.97(0.94,0.98) PPV ¼ positive predictive value, how likely the patient is to have a mutation given that the model predicts carrier status. NPV ¼ negative predictive value, how likely mutation is not present, given that the model does not predict mutation status. A Manchester score of 10 for BRCA1 and BRCA2 or a combined score of 15 corresponds with a mutation probability of 10% (Evans et al, 2004). Table 3 Performance measures for each model at the 10% threshold Table 3 Performance measures for each model at the 10% threshold PPV ¼ positive predictive value, how likely the patient is to have a mutation given that the model predicts carrier status. NPV ¼ negative predictive value, how likely mutation is not present, given that the model does not predict mutation status. A Manchester score of 10 for BRCA1 and BRCA2 or a combined score of 15 corresponds with a mutation probability of 10% (Evans et al, 2004). Statistical methods All generated scores were analysed using SPSS statistical software V13.0 (SPSS Inc., Chicago, IL, USA). Sensitivity, specificity, positive and negative predictive values, and positive and negative likelihood ratios were calculated for each risk model at the 10% threshold. For the Manchester model, this threshold corresponded with a cutoff score of 10 points for BRCA1 and BRCA2 separately, and a score of 15 points for the combined model. The positive likelihood ratio of a model was calculated as the proportion of individuals with a BRCA1/2 mutation who have a model probability 10% or greater divided by the proportion of individuals without a mutation who have a model probability 410%. Similarly, the negative likelihood ratio was calculated as the ratio of the proportion of individuals with a BRCA1/2 mutation who have a model probability o10% to the proportion of individuals without a mutation who have a model probability o10%. Receiver operator characteristics (ROC) curves were constructed by plotting the sensitivity (or true positive fraction) against 1 minus specificity (or false positive fraction) for all possible values of the mutation probability. The area under the ROC curve (C-statistic) was calculated as a measure of the accuracy of the model for discriminating between mutation carriers and those without a BRCA mutation. It represents the fraction of all probands with identified family mutations that have a detection probability higher than a proband with no mutation identified in the family. The k scores were used to assess the measure of agreement between 100 independent risk calculations from 100 pedigrees chosen at random. In this analysis, a score of 0–0.2 describes slight, 0.2–0.4 fair, 0.4–0.6 moderate, 0.6–0.8 substantial and over 0.8 almost perfect agreement (Koch et al, 1977). Proportion of positive BRCA tests varied according to carrier probability Differentiating mutations in BRCA1 and BRCA2 The Manchester, BRCAPRO and Penn models allow separate calculations of the probability of BRCA1 and BRCA2 mutations (Tables 3 and 5). The positive likelihood ratios for BRCA1 mutation carriers were 2.74 (95% CI 1.94–3.88) for Penn, 1.87 (95% CI 1.59–2.21) for Manchester and 2.15 (95% CI 1.72–2.67) for BRCAPRO. The negative likelihood ratios were 0.52, 0.22 and 0.36, respectively. The Penn model could potentially be used to rule in BRCA2 mutations, in that the positive likelihood ratio was 5.93 (95% CI 2.72–12.94) although the negative likelihood ratio The accuracy of the four models was also compared by examining the areas under the ROC curve (Figure 2 and Table 4). A model that correlates perfectly with BRCA mutation status would have an area under the curve of 1.0 while an area of 0.5 indicates that the model has no discriminatory value. Using the scores for detecting either a BRCA1 or BRCA2 mutation, the area Table 4 Area under the ROC curve (C-statistics) for each model 95% confidence interval Test result variable C-statistics Lower bound Upper bound BRCA1 models Manchester 0.808 0.738 0.879 BRCAPRO 0.802 0.731 0.874 Penn 0.808 0.740 0.876 BRCA2 models Manchester 0.660 0.523 0.797 BRCAPRO 0.626 0.500 0.752 Penn 0.703 0.569 0.838 Combination Manchester 0.759 0.688 0.831 BRCAPRO 0.743 0.672 0.814 Myriad 0.753 0.680 0.827 Penn 0.757 0.686 0.827 ROC ¼ receiver operator characteristic. Table 4 Area under the ROC curve (C-statistics) for each model 0.8 1.0 Penn BRCA Myriad Manchester 0.8 1.0 0.6 0.6 0.4 0.4 0.2 0.2 0 Post-test probability Pretest probability Figure 1 Post-test probability for mutation carrier status is shown for each model using the positive and negative likelihood ratios. 0.8 1.0 Penn BRCA Myriad Manchester 0.8 1.0 0.6 0.6 0.4 0.4 0.2 0.2 0 Post-test probability Pretest probability Molecular Diagnostics Figure 1 Post-test probability for mutation carrier status is shown for each model using the positive and negative likelihood ratios. ROC ¼ receiver operator characteristic. Combined Sensitivity 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 Penn BRCAPRO BRCA1 BRCA2 Myriad Manchester Penn BRCAPRO Manchester Penn BRCAPRO Manchester Figure 2 Receiver operator characteristics (ROC) curves for models comprising BRCA1, BRCA2 and the combination. Diagonal segments are produced by ties. Discrimination of BRCA mutation carriers from noncarriers The likelihood that the model probability was 410% in a mutation carrier was 2.10 (95% CI 1.66–2.67) for Penn, 1.74 (95%CI 1.48–2.04) for Myriad, 1.35 (95% CI 1.19–1.53) for Manchester and 1.68 (95% CI 1.39–2.03) for BRCAPRO. Applica- tion of these models, therefore, does not rule in BRCA mutation carrier status. The negative likelihood ratios were consistently greater than 0.1 (BRCAPRO 0.43, Manchester 0.34, Myriad 0.30 and Penn 0.46), which indicated that the models are also unable to rule out mutation carrier status. The post-test probability for mutation carrier status (calculated by multiplying pre-test odds of disease by the positive or negative likelihood ratio) is shown in Figure 1. It is apparent that the estimated pre-test probability of a mutation is not significantly altered by calculating the mutation risk estimate using each of the models. For example, a prediction of a germline mutation for the Penn model would change an individual’s pre-test mutation probability from 40 to 58%, whereas using the Manchester model it would increase from 40 to 47%. under the curve for each model was close to 0.75 (Table 4). By this measure, the models were equal in their ability to discriminate between mutation carrying and noncarrying probands. y g y g p To evaluate the possibility that the mutation-testing strategy influences the performance of the models, we stratified the data according to type of gene sequencing results. Although the frequency of mutation carriers was higher in the group tested by BRCA1/2 sequencing, the models were found to perform similarly for those with and without complete DNA sequencing results. While positive predictive values of models tended to be higher for those with complete DNA sequencing, there were no significant differences in mean model scores or probabilities, sensitivity, specificity, predictive values, likelihood ratios and C-statistics (data not presented). Discrimination of BRCA mutation carriers from noncarriers British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK Models and germline BRCA mutations HH Kang et al Models and germline BRCA mutations 917 mutation in BRCA1 or BRCA2 (Table 3). Given the relatively low prevalence of BRCA mutations in the study population, it is not surprising that the positive predictive value at the 10% threshold was low (range ¼ 0.18–0.25). The predictive value improved as the threshold probability was increased; however, 58% of probands assigned an 80% or greater mutation probability using BRCAPRO did not have a mutation detected (Table 2). All probands with a score of 80 or more according to the Manchester model carried a mutation. The likelihood that the model probability was 410% in a mutation carrier was 2.10 (95% CI 1.66–2.67) for Penn, 1.74 (95%CI 1.48–2.04) for Myriad, 1.35 (95% CI 1.19–1.53) for Manchester and 1.68 (95% CI 1.39–2.03) for BRCAPRO. Applica- tion of these models, therefore, does not rule in BRCA mutation carrier status. The negative likelihood ratios were consistently greater than 0.1 (BRCAPRO 0.43, Manchester 0.34, Myriad 0.30 and Penn 0.46), which indicated that the models are also unable to rule out mutation carrier status. The post-test probability for mutation carrier status (calculated by multiplying pre-test odds of disease by the positive or negative likelihood ratio) is shown in Figure 1. It is apparent that the estimated pre-test probability of a mutation is not significantly altered by calculating the mutation risk estimate using each of the models. For example, a prediction of a germline mutation for the Penn model would change an individual’s pre-test mutation probability from 40 to 58%, whereas using the Manchester model it would increase from 40 to 47%. mutation in BRCA1 or BRCA2 (Table 3). Given the relatively low prevalence of BRCA mutations in the study population, it is not surprising that the positive predictive value at the 10% threshold was low (range ¼ 0.18–0.25). The predictive value improved as the threshold probability was increased; however, 58% of probands assigned an 80% or greater mutation probability using BRCAPRO did not have a mutation detected (Table 2). All probands with a score of 80 or more according to the Manchester model carried a mutation. British Journal of Cancer (2006) 95(7), UK British & 2006 Cancer Research UK Validation of risk calculation estimates Given the inherent ambiguities in interpreting pedigrees, two of the investigators (HK and RWi) specified rules for dealing with various clinical scenarios (see methods and Table 1). At the completion of the study, a k score of mutation-risk estimates using the models was determined for 100 randomly selected pedigrees (25 cases for each risk model). Overall, the k score was 0.82 reflecting excellent agreement between observers when calculating the mutation risk for each proband. The measure of agreement differed between models in that perfect agreement was noted for Penn (k ¼ 1.0) and Manchester (k ¼ 0.932), whereas only sub- stantial agreement was found for Myriad (k ¼ 0.714) and for BRCAPRO (k ¼ 0.60). The areas of disagreement in applying the BRCAPRO model were related to clinical judgment on choice of proband, estimation of age of relatives, and inclusion of maternal and paternal relatives. Both the Penn model and BRCAPRO required computer access. In the case of BRCAPRO, the time taken to enter family trees was a major impediment to routine use. BRCAPRO only incorporates first- and second-degree relatives and therefore cousins of the proband who are affected with cancer will not be used to generate a probability score unless the counselor changes the proband. This scenario was in part responsible for the low k scores associated with the use of BRCAPRO. The on-line Penn model was a very rapid and efficient mode of calculating not only the individual but also the family’s probability of carrying a BRCA1/2 mutation. The model also provided question prompts with helpful explanatory comments; this feature significantly reduced the chance of disagreement between the independent risk assessors. Unfortu- nately, the Penn model also restricted questions to three generations, and did not include ovarian cancer only families or mother–daughter ovarian–breast cancer inheritance patterns. Molecular Diagnostics In comparing the accuracy of the models, we found that they had an equal ability to discriminate between mutation carrying and noncarrying probands (C-statistic 0.743–0.759 for combined analysis using the ROC curves) when considering the entire range of possible test thresholds. These results are highly comparable to those of an Italian study of BRCA mutation carriers where risk was assessed using the Myriad tables, BRCAPRO and an old version of the Penn model (Marroni et al, 2004). Differentiating mutations in BRCA1 and BRCA2 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 BRCA2 Penn BRCAPRO Manchester Combined Sensitivity 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 Penn BRCAPRO Myriad Manchester 1.0 1-specificity 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0 0.2 BRCA1 Penn BRCAPRO Manchester Figure 2 Receiver operator characteristics (ROC) curves for models comprising BRCA1, BRCA2 and the combination. Diagonal segments are produced by ties. British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK British Journal of Cancer (2006) 95(7), 914 – 920 British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK Models and germline BRCA mutations HH Kang et al 918 Table 5 The proportion of BRCA1 and BRCA2 mutation carriers according to carrier probability as determined by Manchester, BRCAPRO and Penn Category of carrier probability (%)a o10 10–20 20–40 40–60 60–80 480 BRCA1 Manchester 4/187(2.1%) 15/150(10.0%) 12/39(30.8%) 3/3(100.0%) 0/1(0.0%) N/A BRCAPRO 7/225(3.1%) 4/43(9.3%) 3/37(8.1%) 3/26(11.5%) 6/25(24.0%) 11/24(45.8%) Penn 14/281(5.0%) 4/42(9.5%) 4/25 (16.0%) 1/7(14.3%) 5/8(62.5%) 6/11(54.6%) BRCA2 Manchester 6/189(3.2%) 6/152(3.9%) 5/37(13.5%) 0/1(0.0%) 1/1(100.0%) N/A BRCAPRO 12/308(3.9%) 1/35(2.9%) 2/19(10.5%) 1/9(11.1%) 2/7(28.6%) 0/2(0.0%) Penn 12/348(3.5%) 5/20(25.0%) 1/6(16.7%) N/A N/A N/A N/A ¼ not applicable. aScore, rather than probability, for Manchester model. 8 Table 5 The proportion of BRCA1 and BRCA2 mutation carriers according to carrier probability as determined by Manchester, BRCAPRO and Penn on the ‘number of cases per family’, yet the definition of a family can be highly subjective. For instance, large pedigrees with many elderly female subjects with breast cancer generated a cumulative score above 10% using the Manchester model. On the other hand, the Myriad tables only allowed inclusion of a maximum of three members of the family, including the patient. Other disconcerting features of the Myriad model were that breast cancers diagnosed above 50 years were ignored, whereas for those diagnosed before 50 years there was no stratification according to the age of diagnosis. Further deficiencies included the equal weighting given to male and female breast cancers and the inability to input bilateral breast cancer or other tumours associated with BRCA1/2 mutation, namely prostate and pancreatic cancers (Lynch et al, 1999; Tulinius et al, 2002; Liede et al, 2004). was unacceptably high at 0.71 (95% CI 0.51–0.98). The Manchester and BRCAPRO had likelihood ratios around 1, indicating that these models provided no additional information in determining BRCA2 mutation carrier status. Differentiating mutations in BRCA1 and BRCA2 A comparison of the respective ROC curves shows that the Manchester model, BRCAPRO and Penn performed similarly for BRCA1, but that BRCAPRO was the least accurate model for BRCA2 identification (Figure 2, Table 4). The proportions of BRCA1 and 2 mutation carriers at each threshold for the Manchester, BRCAPRO and Penn models are shown in Table 5. Validation of risk calculation estimates The initial report of the Manchester model claimed superiority over BRCAPRO on the basis of a larger area under the ROC curve (0.772 vs 0.596) (Evans et al, 2004). We were unable to confirm these findings and note that the area under the curve reported by Evans et al for BRCAPRO was considerably lower than that found in the current study as well as a number of other studies (Euhus et al, 2002; Marroni et al, 2004). The source of these discrepancies may relate to the limitations of comparing a scoring system (Manchester) with probability calculations (BRCAPRO). & 2006 Cancer Research UK British Journal of Cancer (2006) 95(7), 914 – 920 REFERENCES Antman EM, Grudzien C, Sacks DB (1995) Evaluation of a rapid bedside assay for detection of serum cardiac troponin T. JAMA 273: 1279–1282 Eng C, Brody LC, Wagner TM, Devilee P, Vijg J, Szabo C, Tavtigian SV, Nathanson KL, Ostrander E, Frank TS (2001) Interpreting epidemiolo- gical research: blinded comparison of methods used to estimate the prevalence of inherited mutations in BRCA1. J Med Genet 38: 824–833 y p Antoniou A, Pharoah PD, Narod S, Risch HA, Eyfjord JE, Hopper JL, Loman N, Olsson H, Johannsson O, Borg A, Pasini B, Radice P, Manoukian S, Eccles DM, Tang N, Olah E, Anton-Culver H, Warner E, Lubinski J, Gronwald J, Gorski B, Tulinius H, Thorlacius S, Eerola H, Nevanlinna H, Syrjakoski K, Kallioniemi OP, Thompson D, Evans C, Peto J, Lalloo F, Evans DG, Easton DF (2003) Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 72: 1117–1130 Euhus DM (2001) Understanding mathematical models for breast cancer risk assessment and counseling. Breast J 7: 224–232 Euhus DM, Smith KC, Robinson L, Stucky A, Olopade OI, Cummings S, Garber JE, Chittenden A, Mills GB, Rieger P, Esserman L, Crawford B, Hughes KS, Roche CA, Ganz PA, Seldon J, Fabian CJ, Klemp J, Tomlinson G (2002) Pretest prediction of BRCA1 or BRCA2 mutation by risk counselors and the computer model BRCAPRO. J Natl Cancer Inst 94: 844–851 Antoniou AC, Pharoah PD, McMullan G, Day NE, Stratton MR, Peto J, Ponder BJ, Easton DF (2002) A comprehensive model for familial breast cancer incorporating BRCA1, BRCA2 and other genes. Br J Cancer 86: 76–83 Evans DG, Eccles DM, Rahman N, Young K, Bulman M, Amir E, Shenton A, Howell A, Lalloo F (2004) A new scoring system for the chances of identifying a BRCA1/2 mutation outperforms existing models including BRCAPRO. J Med Genet 41: 474–480 Barcenas CH, Hosain GM, Arun B, Zong J, Zhou X, Chen J, Cortada JM, Mills GB, Tomlinson GE, Miller AR, Strong LC, Amos CI (2006) Assessing BRCA carrier probabilities in extended families. J Clin Oncol 24: 354–360 Evans DG, Lalloo F, Wallace A, Rahman N (2005) Update on the Manchester Scoring System for BRCA1 and BRCA2 testing. DISCUSSION In this study we used the family histories of a large tested cohort, some with known BRCA germline mutations, to evaluate the clinical effectiveness of four risk prediction models for BRCA mutations. Estimation of pre-test mutation probability is an important clinical issue, particularly in view of the expense, legal implications and other difficulties associated with ambiguous germline BRCA testing results. Germline testing is currently recommended solely on the basis of clinical suspicion, yet even within this group of patients few will harbour a mutation (13% in the current study). While prediction models (Euhus, 2001; Domchek et al, 2003) offer an opportunity to improve clinical decision-making, to be effective in clinical practice they must be practical as well as accurate. Our study highlighted certain aspects of the models, which are likely to detract from their effective use in family cancer clinics. The Manchester model (Evans et al, 2004) and the Myriad mutation prevalence tables (Myriad, 2004) did not require computer access and were extremely rapid methods of assessing risk. Like most empiric models, the risk estimates relied Although the area under the ROC curves was comparable between models, this does not mean that they are equally accurate in predicting germline mutation status at the well-accepted 10% & 2006 Cancer Research UK & 2006 Cancer Research UK British Journal of Cancer (2006) 95(7), 914 – 920 Models and germline BRCA mutations HH Kang et al Models and germline BRCA mutations 919 counselor may still interpret the pedigree in a way that alters the risk estimates. BRCA germline mutation testing represents a further source of significant error. The true prevalence of BRCA mutations is often underestimated because of the limitations of molecular testing (sensitivity of molecular techniques 70%) (Eng et al, 2001). Furthermore, models are often derived on the basis of mutation testing results from one uncharacterised individual in a high-risk family. Exclusion of a BRCA mutation in one individual does not necessarily indicate that the family is mutation negative. As uncontrollable factors such as cost, death and unavailability often dictate the choice of individual within a family for mutation testing, it is clear that BRCA mutational status is not a ‘gold standard test’. Given these limitations in developing and applying risk models, we advocate the development of risk prediction models that are less reliant on clinical history. probability threshold. ACKNOWLEDGEMENTS We thank Heather Thorne and Eveline Niedermayer for supply of data from kConFab for this project. We are grateful to the physicians, surgeons and oncologists who endorsed this project, the interviewing staff and the many families who participated in this research. kConFaB has been funded by the Kathleen Cuningham Foundation, National Breast Cancer Foundation, National Health and Medical Research Council (NHMRC), Cancer Council of Victoria, Cancer Council of South Australia, Queens- land Cancer Fund, Cancer Council of New South Wales, Cancer Foundation of Western Australian and Cancer Council of Tasmania. This work is supported by National Health and Medical Research Council. Overall, our study suggests that routine use of Penn, Manche- ster, BRCAPRO or Myriad for predicting BRCA mutation status in clinical practice is not currently justified. Furthermore, as the performance of the models was not influenced by the type of mutation testing, we should consider other sources of error. Risk models are developed and applied on the basis of pedigrees constructed from clinical histories. Yet, these histories are themselves often inaccurate in that cancer verification is often impossible and age at diagnosis is frequently an estimate. We have shown that even where prespecified criteria are established to address the nuances of various patient scenarios, the expert Molecular Diagnostics DISCUSSION Examination of the likelihood ratios allows such a comparison and also provides an assessment of the value of the models in relation to other tools. Penn was the most discriminatory model, in that individuals with BRCA mutation were twice as likely to have a positive prediction as those without a germline mutation. In contrast, the Manchester model had a ratio closer to unity indicating that it is unhelpful in clinical decision-making. In other areas of medicine, likelihood ratios greater than 5 are generally a prerequisite for the adoption of a clinical test or procedure. By way of comparison, the positive likelihood ratio for mammography is 14 (Kerlikowske et al, 1995), for bedside cardiac-specific troponin T 6.3 (Antman et al, 1995) and for ultrasonography for deep venous thrombosis 47.8 (Wells et al, 2000). The separate BRCA1/2 performance measures for Manchester, BRCAPRO and Penn displayed similar trends to the combined values, with the only exception being the use of Penn for assessment of BRCA2 risk, where the positive likelihood ratio was 5.9. A positive result in Penn for BRCA2 mutation should confidently warrant a mutation screen. REFERENCES J Natl Cancer Inst 93: 1633–1637 Nelson HD, Huffman LH, Fu R, Harris EL (2005) Genetic risk assessment and BRCA mutation testing for breast and ovarian cancer susceptibility: systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med 143: 362–379 Huang P, Leary J, Smith M, Kirk J (2002) Determining the probability of finding a breast cancer susceptibility gene mutation in clinic based families. In Fifth Australasian Conference on Familial and Genetic Aspects of Cancer, Anon (ed). South Australia: Barossa Valley Parmigiani G, Berry D, Aguilar O (1998) Determining carrier probabilities for breast cancer-susceptibility genes BRCA1 and BRCA2. Am J Hum Genet 62: 145–158 James PA, Doherty R, Harris M, Mukesh BN, Milner A, Young MA, Scott C (2006) Optimal selection of individuals for BRCA mutation testing: a comparison of available methods. J Clin Oncol 24: 707–715 Peto J, Collins N, Barfoot R, Seal S, Warren W, Rahman N, Easton DF, Evans C, Deacon J, Stratton MR (1999) Prevalence of BRCA1 and BRCA2 gene mutations in patients with early-onset breast cancer. J Natl Cancer Inst 91: 943–949 Kauff ND, Satagopan JM, Robson ME, Scheuer L, Hensley M, Hudis CA, Ellis NA, Boyd J, Borgen PI, Barakat RR, Norton L, Castiel M, Nafa K, Offit K (2002) Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 346: 1609–1615 Puget N, Sinilnikova OM, Stoppa-Lyonnet D, Audoynaud C, Pages S, Lynch HT, Goldgar D, Lenoir GM, Mazoyer S (1999) An Alu-mediated 6-kb duplication in the BRCA1 gene: a new founder mutation? Am J Hum Genet 64: 300–302 g Kerlikowske K, Grady D, Rubin SM, Sandrock C, Ernster VL (1995) Efficacy of screening mammography. A meta-analysis. JAMA 273: 149–154 Rebbeck TR, Friebel T, Lynch HT, Neuhausen SL, van ’t Veer L, Garber JE, Evans GR, Narod SA, Isaacs C, Matloff E, Daly MB, Olopade OI, Weber BL (2004) Bilateral prophylactic mastectomy reduces breast cancer risk in BRCA1 and BRCA2 mutation carriers: the PROSE Study Group. J Clin Oncol 22: 1055–1062 Koch GG, Landis JR, Freeman JL, Freeman Jr DH, Lehnen RC (1977) A general methodology for the analysis of experiments with repeated measurement of categorical data. Biometrics 33: 133–158 Liede A, Karlan BY, Narod SA (2004) Cancer risks for male carriers of germline mutations in BRCA1 or BRCA2: a review of the literature. REFERENCES J Med Genet 42: e39 Berry DA, Iversen Jr ES, Gudbjartsson DF, Hiller EH, Garber JE, Peshkin BN, Lerman C, Watson P, Lynch HT, Hilsenbeck SG, Rubinstein WS, Hughes KS, Parmigiani G (2002) BRCAPRO validation, sensitivity of genetic testing of BRCA1/BRCA2, and prevalence of other breast cancer susceptibility genes. J Clin Oncol 20: 2701–2712 Ford D, Easton DF, Bishop DT, Narod SA, Goldgar DE (1994) Risks of cancer in BRCA1-mutation carriers. Breast Cancer Linkage Consortium. Lancet 343: 692–695 Frank TS, Deffenbaugh AM, Reid JE, Hulick M, Ward BE, Lingenfelter B, Gumpper KL, Scholl T, Tavtigian SV, Pruss DR, Critchfield GC (2002) Clinical characteristics of individuals with germline mutations in BRCA1 and BRCA2: analysis of 10 000 individuals. J Clin Oncol 20: 1480–1490 Couch FJ, DeShano ML, Blackwood MA, Calzone K, Stopfer J, Campeau L, Ganguly A, Rebbeck T, Weber BL (1997) BRCA1 mutations in women attending clinics that evaluate the risk of breast cancer. N Engl J Med 336: 1409–1415 Garvin AM (1998) A complete protein truncation test for BRCA1 and BRCA2. Eur J Hum Genet 6: 226–234 de la Hoya M, Diez O, Perez-Segura P, Godino J, Fernandez JM, Sanz J, Alonso C, Baiget M, Diaz-Rubio E, Caldes T (2003) Pre-test prediction models of BRCA1 or BRCA2 mutation in breast/ovarian families attending familial cancer clinics. J Med Genet 40: 503–510 Hartmann LC, Schaid DJ, Woods JE, Crotty TP, Myers JL, Arnold PG, Petty PM, Sellers TA, Johnson JL, McDonnell SK, Frost MH, Jenkins RB (1999) Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med 340: 77–84 y g Hartmann LC, Sellers TA, Schaid DJ, Frank TS, Soderberg CL, Sitta DL, Frost MH, Grant CS, Donohue JH, Woods JE, McDonnell SK, Vockley CW, Deffenbaugh A, Couch FJ, Jenkins RB (2001) Efficacy of bilateral Domchek SM, Eisen A, Calzone K, Stopfer J, Blackwood A, Weber BL (2003) Application of breast cancer risk prediction models in clinical practice. J Clin Oncol 21: 593–601 British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK Models and germline BRCA mutations HH Kang et al Models and germline BRCA mutations HH K l 920 prophylactic mastectomy in BRCA1 and BRCA2 gene mutation carriers. J Natl Cancer Inst 93: 1633–1637 prophylactic mastectomy in BRCA1 and BRCA2 gene mutation carriers. REFERENCES J Clin Oncol 22: 735–742 Rebbeck TR, Levin AM, Eisen A, Snyder C, Watson P, Cannon-Albright L, Isaacs C, Olopade O, Garber JE, Godwin AK, Daly MB, Narod SA, Neuhausen SL, Lynch HT, Weber BL (1999) Breast cancer risk after bilateral prophylactic oophorectomy in BRCA1 mutation carriers. J Natl Cancer Inst 91: 1475–1479 Lynch HT, Watson P, Tinley S, Snyder C, Durham C, Lynch J, Kirnarsky Y, Serova O, Lenoir G, Lerman C, Narod SA (1999) An update on DNA- based BRCA1/BRCA2 genetic counseling in hereditary breast cancer. Cancer Genet Cytogenet 109: 91–98 Struewing JP, Watson P, Easton DF, Ponder BA, Lynch HT, Tucker MA (1995) Prophylactic oophorectomy in inherited breast/ovarian cancer families. J Natl Cancer Inst Monogr 17: 33–35 Marroni F, Aretini P, D’Andrea E, Caligo MA, Cortesi L, Viel A, Ricevuto E, Montagna M, Cipollini G, Ferrari S, Santarosa M, Bisegna R, Bailey-Wilson JE, Bevilacqua G, Parmigiani G, Presciuttini S (2004) Evaluation of widely used models for predicting BRCA1 and BRCA2 mutations. J Med Genet 41: 278–285 Tulinius H, Olafsdottir GH, Sigvaldason H, Arason A, Barkardottir RB, Egilsson V, Ogmundsdottir HM, Tryggvadottir L, Gudlaugsdottir S, Eyfjord JE (2002) The effect of a single BRCA2 mutation on cancer in Iceland. J Med Genet 39: 457–462 Meijers-Heijboer H, van Geel B, van Putten WL, Henzen-Logmans SC, Seynaeve C, Menke-Pluymers MB, Bartels CC, Verhoog LC, van den Ouweland AM, Niermeijer MF, Brekelmans CT, Klijn JG (2001) Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 345: 159–164 University of Pennsylvania Abramson Cancer Center (2005) The Penn II BRCA1 and BRCA2 Mutation Risk Evaluation Model. http://acgh.afcri. upenn.edu/cgi-bin/bcm.pl (accessed 20 November 2005) Wells PS, Anderson DR, Ginsberg J (2000) Assessment of deep vein thrombosis or pulmonary embolism by the combined use of clinical model and noninvasive diagnostic tests. Semin Thromb Hemost 26: 643–656 Myriad (2004) Mutation prevalence tables for BRCA1/2 genes. http:// www.myriadtests.com/provider/mutprev.htm (accessed 20 November 2005) Molecular Diagnostics Woodward AM, Davis TA, Silva AG, Kirk JA, Leary JA (2005) Large genomic rearrangements of both BRCA2 and BRCA1 are a feature of the inherited breast/ovarian cancer phenotype in selected families. J Med Genet 42: e31 NBCC Genetics Working Group (2000) Advice about familial aspects of breast cancer and ovarian cancer – a guide for health professionals. British Journal of Cancer (2006) 95(7), 914 – 920 REFERENCES http://www.genetics.com.au/pdf/pubs/boguidelines.pdf (accessed 20 November 2005) British Journal of Cancer (2006) 95(7), 914 – 920 & 2006 Cancer Research UK
https://openalex.org/W2128041310	https://www.scielo.br/j/lajss/a/MMT4DkDQX9rXSDmdfz7v7Lc/?lang=en&format=pdf	English	null	An analytical model of a clamped sandwich beam under low-impulse mass impact	Latin American Journal of Solids and Structures	2,014	cc-by	11,920	1651 1651 Keywords sandwich beam, mental foam core, analytical model, low impulse, projectile impact An analytical model of a clamped sandwich beam un- der low-impulse mass impact Wen-zheng Jianga Ying Liua, * Yu Gua G X Lub Abstract An analytical model is developed to examine a low impulsive projectile impact on a fully clamped sandwich beams by consider- ing the coupled responses of the core and the face sheets. Firstly, based on the dynamic properties of foam cores, the sandwich beam is modeled as two rigid perfectly-plastic beams connected by rigid perfectly-plastic springs. Different from the previous sandwich beam model, the transverse compression and bending effects of the foam core are considered in the whole deformation process. Based on this model, different coupling mechanism of sandwich beams are constructed so that an analytical solution considering small deformation is derived. The coupled dynamic responses of sand- wich beams with different core strengths are investigated. The results indicate that this model improves the prediction accuracy of the responses of the sandwich beams, and is available for the situation when the sandwich beam undergoes moderate global deformation. aDepartment of Mechanics, School of Civil Engineering, Beijing Jiaotong University, Beijing, 100044 bSchool of Mechanical and Aerospace Engi- neering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore bSchool of Mechanical and Aerospace Engi- neering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore *Author e-mail: yliu5@bjtu.edu.cn aDepartment of Mechanics, School of Civil Engineering, Beijing Jiaotong University, Beijing, 100044 1 INTRODUCTION As one kind of representative structures widely used in the design of commercial and military vehicles, such as aircrafts, spacecrafts, vehicles, ships, and high speed train carriages, responses of sandwich structures subjected to impact at a wide range of velocities have attracted great atten- tion and been extensively studied. How to accurately predict the dynamic behaviors of sandwich structures is of current interest. For the shock resistance of clamped sandwich beams, Fleck and Deshpande (2004) separate the responses of these beams into three stages, that is, fluid-structure interaction stage; the core compression stage and a combined beam bending and stretching stage. By decoupling these three stages, they develop an analytical model for understanding the blast responses of sandwich beams. In a parallel work, Xue and Hutchinson (2004) compare the blast resistance of clamped sandwich beams to that of monolithic beams of the same mass via three-dimensional finite ele- Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1652 ment simulations. Then, some finite element (FE) simulations by Rabczwk et al . (2004) and Liang et al . (2007) suggest that the model of Fleck and Deshpande may over-estimate or under- estimate the deflection of sandwich beams under blasting loading. This discrepancy indicates that coupling between stages of responses can influence the deflections. Deshpande and Fleck (2005) examine the coupling between the fluid-structure interaction stage I and the core compression stage II. Their results show that the Taylor (1963) analysis based on a free-standing front face- sheet underestimate the transmitted momentum by 20-40% for sandwich beams comprising high strength cores, which explains the discrepancies between the FE simulations of Rabczuk et al . (2004) and the analytical predictions of Fleck and Deshpande (2005). In order to explain the discrepancy between the FE simulation of Liang et al . (2007) and the analytical predictions of Fleck and Deshpande (2004), Tilbrook et al . (2007) developed an analyt- ical model based on time-scales of core compression and the bending/stretching response of the sandwich beam. In their model, four regimes of behavior, that is, decoupled responses with the sandwich core densification partially or completely, and coupled responses with partial or full core compression, are defined. 1 INTRODUCTION However, during their formulation, they made one critical assumption, that is, neglecting the shear strength of the core prior to equalization of the velocities of the front and back faces, and the transverse compression strength loads the back and front faces simulta- neously. Hence, before the velocity equalization, the sandwich beam is treated approximately as a pressure cavity, which obviously has different deformation mechanism from the structures with foam cores. Following their treatments, some theoretical and experimental works are carried out (Qin, et al ., 2009; 2011; Wang, et al ., 2011). Although this model to some extend accounts for the coupling between the core compression and the beam bending/stretching phases, their analytical results under-predict the peak back face deflection and over-predict the support reactions, espe- cially for sandwich beams with high strength cores (Tilbrook, et al ., 2007). Structurally, a typical sandwich beam consists of two thin face sheets bonded to a core made from low- density materials. The deformation of the front face under initial dynamic loading, then the core compression and the deflection of the back face, is a series of coupled responding process. The foam core provides not only the transverse pressure, but also the bending/shear strength during the coupled deformation between the core compression and the beam bend- ing/stretching phases. Moreover, researches show that the responses of the foam at the impact or distal ends are different for different impacting conditions (Liu and Zhang, 2009; Liu, et al ., 2012), which indicates that the pressures loads on the back and front faces due to the transverse compression of the core may be different, and asynchronous. As a result, how to consider the bending/shear effect of the foam core during the deformation, and describe the transverse pres- sures on the face sheets caused by the core compression, are the key points in the analytical mod- eling of sandwich beams, and have not been absolutely solved to per author’s knowledge. Aiming to these problems, an abstracted beam-spring model for the sandwich beam is firstly established by considering the transverse compression and bending effect of the foam core in the whole deformation process. Different from beam-spring system on-rigid foundation (Chen and Yu, 2002; Yu et al ., 2002), the coupled motion of the front and back beam is considered. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 ANALYTICAL MODEL OF A SANDWICH BEAM UNDER MASS IMPACTING As shown in Fig. 1a, a fully clamped slender metal foam core sandwich beam impacted by a striker with an initial velocity at the midpoint is considered herein. The mass and the initial velocity of the striker are m0 and V0, respectively. The beam length is 2L. Two face-sheets with thickness hf and hb are perfectly bonded to the foam core with the core thickness being C. It is assumed that the face sheet metals obey the rigid-perfectly plastic law with the yield strength is ! y . Figure 1 (a) A sandwich beam impacted by a mass; (b) Simplified rigid-perfectly plastic beam-spring model. Figure 1 (a) A sandwich beam impacted by a mass; (b) Simplified rigid-perfectly plastic beam-spring model. In the dynamic analysis we assume that displacements occur only in a direction transverse to the original axis of the beam. Small transverse deflections are considered, such that the deflection w at the mid-span of the beam is assumed to be small compared to the beam length and the rotational inertia is not included into the moment balance equations. In the dynamic analysis we assume that displacements occur only in a direction transverse to the original axis of the beam. Small transverse deflections are considered, such that the deflection w at the mid-span of the beam is assumed to be small compared to the beam length and the rotational inertia is not included into the moment balance equations. Lopatnikov et al . (2007) distinguished the impact/shock loaded cellular solids deformation pat- terns as two modes, i.e., (a) homogeneous deformation, that is, cellular medium deforms homogene- ously over the entire volume of the sample; and (b) progressive collapse, during which the same deformation is reached by complete densification of the portion of the cellular material adjacent to the location where the load applies, while the rest is assumed to be un-deformed. In the present discussion, since small deflection assumption is made and low impulsive loading is considered, ho- mogeneous deformation is assumed which implies the same stress responses at the impact and distal ends of the core. Figure 2 displays a typical stress-strain curve of the cellular material under the dynamic/shock loading, in which σ 0 is the initial force peak, and ε D and ! p are densification strain and plateau stress, respectively. 1 INTRODUCTION Based on this model, a projectile impact on a fully clamped sandwich beam with a foam core is investigated based on small deflection theory. The coupling deformation mechanism of the beam-spring system Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1677 is formulated for sandwich structures with soft or intermediate strength cores. Then the analyti- cal predictions are compared with experimental results and finite element (FE) simulations of the mass impacted sandwich beams to show the availability of this model. At last, the conclusion is given. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 ANALYTICAL MODEL OF A SANDWICH BEAM UNDER MASS IMPACTING Considering that the total outer input energy is greater than the energy absorbed due to the elastic deformation, the rigid perfectly-plastic-locking (RPPL) mod- el is adopted (red line in Fig. 2), and after the densification, the core is treated as rigid. Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1654 Figure 2 Typical dynamic stress-strain curves for the foam. The red line corresponds to rigid perfectly-plastic-locking (RPPL) model. Figure 2 Typical dynamic stress-strain curves for the foam. The red line corresponds to rigid perfectly-plastic-locking (RPPL) model. Based on this homogeneous deformation assumption of the foam core, rigid-perfectly plastic springs, whose responses obey RPPL model given in Fig. 2, are introduced to establish the relation between the pressures acting on the front and back faces due to the transverse compression of the foam core. Employing the lumped mass approximation, the mass of the core is equally distributed to the front and back beams, that is, the mass per unit length of the front and back faces are (Fleck and Deshpande, 2004) mf = hf ρ f + ! cC / 2 (1) (1) and mb = hb! b + ρcC / 2 (2) (2) where ρ f , ! b , and ! c are the densities of the front face, the back face and the foam core, respec- tively. In Fig. 3, during the bending of the face sheets, the foam core is also bended. Consequently, ex- cept for the transverse pressure, the bending effect of the core should also be considered. Different from Tilbrook et al . (2007), in which the shear/bending effect of the core is ignored before the end of the core compression, a simple but critical equivalence is made, that is, the bending moment of the core is divided equally by considering the homogeneous deformation assumption, and added to the corresponding values of the front and back faces to account for the bending effect provided by the foam core. This treatment effectively accounts for the bending effect of the core during the de- formation and the validation of this assumption is confirmed via full FE simulation in Section 3. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 ANALYTICAL MODEL OF A SANDWICH BEAM UNDER MASS IMPACTING Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1656 2.1 Intermediate strength foam core ( 1 3 γ ≤ ≤) When the front beam is impacted, the midpoint of the beam moves with the velocity V0 at the instantaneous impact, whilst the left part keeps stationary. In order to keep balance, a disturbance is propagated along the beam from the midpoint to the fixed ends and the core is gradually com- pressed. This includes two different kinematical phases: a transient phase (0 ! t ≤tI , phase I, Fig. 4a), and a modal phase (tI < t ≤tII , phase II, Fig. 4b). At the same time, the back beam will also deform since the plateau stress of the core is assumed to be greater than its yield stress (γ>1). Cor- responding to the kinematical phases of the front beam, the deformation of the back beam includes the deflection under a moving and then a uniform loading (Figs. 5a and 5b), respectively. When the velocities of the front and back beams become the same (t=teq), the beams deform as a system, as shown in Fig. 6 (t ≥teq , phase III). Moreover, if the densification (t=tD) is reached before the equivalent velocity is obtained (tD ≤t ! teq ), the front and back beams also begin to deform as a system (phase III). ANALYTICAL MODEL OF A SANDWICH BEAM UNDER MASS IMPACTING Then we have 2 f 0 f / 2 / 4 / 2 C y f C M M M h M σ = + = + (3) 2 f 0 f / 2 / 4 / 2 C y f C M M M h M σ = + = + ( 2 f 0 f / 2 / 4 / 2 C y f C M M M h M σ = + = + (3) Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1655 1655 and Mb0 = M b + MC / 2 = ! yhb 2 / 4 + MC / 2 (4) (4) where Mf, Mb, and MC are the sectional limit bending moments of the front face, the back face and the foam core, respectively, in which MC = σ yCC 2 1−ε m ( ) 2 / 4 (5) (5) where ! yC is the yield strength of the foam core, εm is the middle point compression strain which is defined as εm =\|Wf -Wb\|/C with Wf and Wb the mid-span deflections of the front and back beams, respectively. Figure 3 Experimental results of impact failure modes with respect to load intensity (Wang, et al., 2011). Figure 3 Experimental results of impact failure modes with respect to load intensity (Wang, et al., 2011). By now, the sandwich beam is modeled as two beams with larger mass and bending moments (compared to the face sheets) connected by rigid-perfectly plastic springs, which is shown in Fig. 1b. After the deformation of the front beam, the forces due to the compression of the foam core will act on the back beam. If the core is sufficiently strong, it will decelerate the front beam and simul- taneously accelerate the back beam; otherwise, the back beam will keep un-deformed at the plateau stage of the core compression. The response of the clamped beam depends on the transverse pres- sure sp acting from the core onto the back face. Following Jones (1989), by taking γ=! p /! bs with ! bs =4Mb0/L2 the static extreme pressure of the back beam with "m=0, the sandwich beam with intermediate or low strength cores are considered in the following section. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 2.1.1.1 Front beam In the first phase, a disturbance propagates from the central point to the fixed ends. Because of symmetry, the right half beam 0 x L ! ! is considered. Its transverse velocity field is assumed to be (Figs. 4a and 4b) !w f 1 = !Wf 1(1−x / ξ) 0 ! x ≤ξ , !wf 1 = 0 ξ ! x ≤L , (6) (6) where w is the transverse displacement, # is the position of the plastic hinge which dependents on the time t. The upper dot means the differentiating with respect to the time t, and the suffix f rep- resents the front beam and the number 1 the first phase. At the moving plastic hinges (x=±#), the bending moment Mf1 is maximum and the transverse force Q=0. The force balance between the two moving plastic hinges along the transverse direction yields where w is the transverse displacement, # is the position of the plastic hinge which dependents on the time t. The upper dot means the differentiating with respect to the time t, and the suffix f rep- resents the front beam and the number 1 the first phase. At the moving plastic hinges (x=±#), the bending moment Mf1 is maximum and the transverse force Q=0. The force balance between the two moving plastic hinges along the transverse direction yields m0 !! Wf 1 + 2 mf !!wf 1 0 !∫ dx+ 2 q(x)dx 0 ξ∫ = 0 (7) (7) where q(x) is the dynamic pressure caused by the compression of the foam core, which keeps con- stant before densification as shown in Fig.2 (red line). where q(x) is the dynamic pressure caused by the compression of the foam core, which keeps con- stant before densification as shown in Fig.2 (red line). Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1657 1657 Figure 4 (a) Velocity profile of the front beam in phase I; (b) a free body diagram of the left half front beam in phase I; (c) velocity profile of the front beam in phase II; (d) a free body diagram of the left half front beam in phase II. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 2.1.1.1 Front beam Figure 4 (a) Velocity profile of the front beam in phase I; (b) a free body diagram of the left half front beam in phase I; (c) velocity profile of the front beam in phase II; (d) a free body diagram of the left half front beam in phase II. Substitution Eq. (6) into Eq. (7) leads to Substitution Eq. (6) into Eq. (7) leads to Substitution Eq. (6) into Eq. (7) leads to m0 !! Wf 1 + 2mf !! Wf 1(1! x / ξ)d 0 ξ∫ x + 2 q(x)dx 0 !∫ = 0 (8) (8) that is, m0 !! Wf 1 + mf !! Wf 1ξ + !Wf 1 !! ( ) = −2 q(x)d 0 !! x (9) (9) Considering the moment balance between x=0 and x=# to the beam midpoint, and the boundary conditions: Mf1=Mf0 at x=0; M f1= -Mf0 and Q=0 at x=L, we have 2Mf 0 − mf !!wf 1 0 ξ∫ xdx − q(x)xdx 0 ξ∫ = 0 (10) (10) Substitution Eqs. (6) into Eq. (10) and considering q(x)=σ p yields 2M f 0 −mf !! Wf 1ξ 2 / 6 + !Wf 1 !ξξ / 3 ( ) −σ pξ 2 / 2 = 0 (11) (11) Then we have d !Wf 1ξ 2 ( ) d t = (12M f 0 ! 3! p" 2) / mf (12) (12) Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 8 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1658 Integrating Eq. (12) and considering when t=0, #=0, we have t = mf !Wf 1! 2 + 3σ p ξ 2 dt 0 t∫ 12M f 0 (13) (13) Integrating Eq. (9) and considering when t=0, #=0, and !Wf 1 =V0 , we have Integrating Eq. (9) and considering when t=0, #=0, and !Wf 1 =V0 , we have !Wf 1 = m0V0 −2 σ pξ d 0 t! t m0 + mfξ (14) (14) Combination Eqs. (13) and (14) we have the time-position relation of the dynamic plastic hinge, that is t = mf! 2 m0V0 ! 2σ p ξ d 0 t∫ t ⎛ ⎝ ⎞ ⎠ 12M f 0 m0 + mf! ( ) + σ p ξ 2 d 0 t! t 4M f 0 (15) (15) Differentiating Eq. 2.1.1.1 Front beam w f 1 is the relative deflection of the front beam at point x. where ! w f 1 is the relative deflection of the front beam at point x. 2.1.1.1 Front beam (15) with respect to the time t, we obtain the travelling velocity of the plastic hinge, that is, Differentiating Eq. (15) with respect to the time t, we obtain the travelling velocity of the plastic hinge, that is, !ξ = 2mfσ pξ 3 m0 + mfξ ( ) + 12Mf 0 −3! pξ 2 ( ) m0 + mfξ ( ) 2 mfξ 2m0 + mf! ( ) m0V0 −2! p ! d 0 t∫ t ⎛ ⎝ ⎞ ⎠ (16) (16) When the plastic hinge arrives at the fixed end, we have #=L, that is, t=tI, the first phase ends. According to Eqs. (14) to (16), we can obtain the related deformation parameters of the front beam in the first phase. It is seen that x is time and position dependent. The direct integrals of Eqs. (14) to (16) are impossible. Numerical integration using MatLab is performed. Moreover, as expected when ! p=0, Eqs. (14) to (16) degenerate to the equations for monoclinic beams given in Appendix A. Comparison between Eq. (14) and Eq. (A6) indicates that the velocity of the mid-span point of the front beam is smaller than that of the monolithic beam due to the existence of the foam core. In the first phase, the kinetic energy of the mass and the front beam is the kinetic energy of the mass and the front beam is Ef I = Emass + E front = m0 !Wf 1 2 2 + mf !Wf 1 2! 3 (17) (17) When the moving plastic hinge length is #, the dissipated plastic bending energy is When the moving plastic hinge length is #, the dissipated plastic bending energy is Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1659 1659 U Bending-FI = 4Mf 0Wf 1 ξ (18) U Bending-FI = 4Mf 0Wf 1 ξ (18) U Bending-FI = 4Mf 0Wf 1 ξ (18) and the dissipated spring potential energy is USpring-FI = 2 ! p! w f 1d x 0 ξ" USpring-FI = 2 ! p! w f 1d x 0 ξ" (19) (19) where ! w f 1 is the relative deflection of the front beam at point x. where ! w f 1 is the relative deflection of the front beam at point x. where ! 2.1.1.2 Back beam Because γ>1, during the deformation of the front beam, the back beam also deforms. According to experimental and numerical examination, the transverse velocity field of the back beam in the first phase is assumed as (Figs. 5a and 5b) !wb1 = !Wb1(1−x / ! ) 0 ≤x ≤ξ , !wb1 = 0 ! ! x ! L , (20) (20) and the governing equations of the beam are and the governing equations of the beam are and the governing equations of the beam are ! 2Mb1 ! 2x = ! ! p + mb ∂2wb1 ! 2t 0 ! x ! " (21a) ! 2Mb1 ! 2x = 0 ! < x " L (21b) ! 2Mb1 ! 2x = ! ! p + mb ∂2wb1 ! 2t 0 ! x ! " (21a) (21a) ! 2Mb1 ! 2x = 0 ! < x " L (21b) (21b) where Mb1 is the bending moment of the back beam at the first stage. where Mb1 is the bending moment of the back beam at the first stage. where Mb1 is the bending moment of the back beam at the first stage. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Figure 5 (a) Velocity profile of the back beam in phase I; (b) a free body diagram of the left half back beam in phase I; (c) velocity profile of the back beam in phase II; (d) a free body diagram of the left half back beam in phase II. a) Velocity profile of the back beam in phase I; (b) a free body diagram of the left half back beam in phase I; (c) velocity profile of the back beam in phase II; (d) a free body diagram of the left half back beam in phase II. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 ng Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1660 Substitution Eq. (20) into Eq. (21a) leads to Substitution Eq. (20) into Eq. (21a) leads to ! 2M b1 ! 2x = ! σ p + mb(1! x / ξ) d 2Wb1 d 2t 0 ≤x ≤ξ (22) (22) Integrating Eq. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 2.1.1.2 Back beam wb1 is the relative deflection of the back beam at point x. 2.1.2 Coupled deformation in the second phase, tI<t≤ tII 2.1.2 Coupled deformation in the second phase, tI<t≤ tII In the second phase, the plastic hinges are fixed during which the remaining energy is dissipated by the plastic deformation of the plastic hinges (Fig. 4c), the foam core and the deformation of the back beam (Fig. 5c). 2.1.1.2 Back beam (22) and considering Mb1 = Mb0 , Q = ∂M b1 / ∂x = 0 at x=0, we have M b1 = −! px2 / 2 + mb(x2 / 2 ! x3 / 6! ) d 2Wb1 d 2t + M b0 0 ! x ! ! (23) (23) Since at x=#, Mb1=-Mb0, according to Eq. (23), we have Since at x=#, Mb1=-Mb0, according to Eq. (23), we have d 2Wb1 d 2t = (3! p" 2 / 2 ! 6Mb0) / mb! 2 (24) (24) It should be pointed out that for beams under the moving loading, the assumed velocity field is only valid for ! p ! 4Mb0 L" (25) which leads to ξ ! 4Mb0 L! p (26) ! p ! 4Mb0 L" (25) (25) which leads to ξ ! 4Mb0 L! p (26) (26) As a result, the back beam does not deform simultaneously with the front beam. Only after Eq. (26) is satisfied does the back beam begins to deform. Integration Eq. (24) leads to As a result, the back beam does not deform simultaneously with the front beam. Only after Eq. (26) is satisfied does the back beam begins to deform. ation Eq. (24) leads to !Wb1 = 3σ pξ 2 / 2 −6M b0 ( )t / mbξ 2 (27) (27) In Fig. 5a, when the moving hinges arrive at the fixed ends, that is, ξ = L , the moving forces arrive at the fixed ends and the first stage ends. The kinetic energy of the back beam in the first phase is EBI = mb !Wb1 2ξ 3 (28) (28) Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1661 1661 When the length of the moving pressure x=x, the dissipated plastic bending energy is U Bending-BI = 4Mb0Wb1 ξ (29) (29) and the dissipated spring potential energy is U BSpringI = 2 ! p! wb1d x 0 !" (30) U BSpringI = 2 ! p! wb1d x 0 !" (30) where ! wb1 is the relative deflection of the back beam at point x. where ! wb1 is the relative deflection of the back beam at point x. where ! wb1 is the relative deflection of the back beam at point x. where ! 2.1.2.1 Front beam where !Wf I is the mid-span velocity of the front beam at the end of the first phase. In the second phase, the kinetic energy of the front beam and the mass is In the second phase, the kinetic energy of the front beam and the mass is EFII = Emass + E front = m0 !Wf 2 2 2 + mf !Wf 2 2 L 3 (37) (37) For the front beam, the dissipated plastic bending energy is For the front beam, the dissipated plastic bending energy is For the front beam, the dissipated plastic bending energy is U Bending-FII = 4M f 0Wf 2 L (38) (38) and the dissipated spring potential energy is and the dissipated spring potential energy is and the dissipated spring potential energy is U FSpringII = 2 ! pΔwf 2d x 0 L∫ (39) U FSpringII = 2 ! pΔwf 2d x 0 L∫ (39) where ! w f 2 is the relative deflection of the front beam at point x. where ! w f 2 is the relative deflection of the front beam at point x. where ! w f 2 is the relative deflection of the front beam at point x. 2.1.2.1 Front beam Seen as Fig. 4c, we assume the transverse velocity field of the front beam in the second phase is !wf 2 = !Wf 2(1! x / L) 0 " x " L (31) !wf 2 = !Wf 2(1! x / L) 0 " x " L (31) The governing equation of the beam is The governing equation of the beam is ! 2M f 2 ! 2x = ! p + mf 1" x / L ( ) d 2Wf 2 d 2t 0 # x # L (32) (32) Integrating Eq. (32) and considering M f 2 = Mf 0 , Q = m0 !! Wf 2 / 2 at x=0, we have ∂M f 2 ! x = ! px+ mf x ! x2 2L ! "# $ % & d 2Wf 2 d 2t + m0 2 d 2Wf 2 d 2t 0 ≤x ≤L (33) ∂M f 2 ! x = ! px+ mf x ! x2 2L ! "# $ % & d 2Wf 2 d 2t + m0 2 d 2Wf 2 d 2t 0 ≤x ≤L (33) (33) and M f 2 = σ px2 / 2 + mf x2 2 −x3 6L ⎛ ⎝⎜ ⎞ ⎠⎟ d 2Wf 2 d 2t + m0 2 d 2Wf 2 d 2t x+ Mf 0 0 ≤x ≤L (34) (34) Considering M f2= -Mf0 at x=L, we have ring M f2= -Mf0 at x=L, we have Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1662 !! Wf 2 = ! 12M f 0 + 3σ pL2 2mf L2 + 3m0L (35) !! Wf 2 = ! 12M f 0 + 3σ pL2 2mf L2 + 3m0L (35) Integrating Eq. (35) leads to Integrating Eq. (35) leads to !Wf 2 = − 12Mf 0 + 3σ pL2 2mf L2 + 3m0L t + !Wf I (36) (36) where !Wf I is the mid-span velocity of the front beam at the end of the first phase. In the second phase, the kinetic energy of the front beam and the mass is where !Wf I is the mid-span velocity of the front beam at the end of the first phase. 2.1.3 Coupled deformation in the third phase, ( t ≥teq , or tD ! t ! teq ) Coupled deformation in the third phase, ( t ≥teq , or tD ! t ! teq ) 2.1.2.2 Back beam As shown in Fig. 5c, we assume the transverse velocity field of the back beam in the second phase is is !wb2 = !Wb2(1−x / L) (40) (40) and the governing equation of the beam is and the governing equation of the beam is and the governing equation of the beam is ! 2M b2 ! 2x = "! p + mb 1−x / L ( ) d 2Wb2 d 2t 0 ! x ! L (41) (41) Integrating Eq. (41) and considering M b2 = M b0 , Q = ∂Mb2 / ∂x = 0 at x=0, we have Integrating Eq. (41) and considering M b2 = M b0 , Q = ∂Mb2 / ∂x = 0 at x=0, we have Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1663 1663 ∂Mb2 ∂x = −! px+ mb x −x2 2L ⎛ ⎝⎜ ⎞ ⎠⎟ d 2Wb2 d 2t 0 ≤x ≤L (42) (42) and and Mb2 = ! ! px2 / 2 + mb x2 2 ! x3 6L " #$ % &’ d 2Wb2 d 2t + Mb0 0 ( x ( L (43) (43) Considering M b2= -Mb0 at x=L, we have Considering M b2= -Mb0 at x=L, we have Considering M b2= -Mb0 at x=L, we have Considering M b2= -Mb0 at x=L, we have !! Wb2 = 3! pL2 ! 12Mb0 ( ) 2mbL2 ( ) (44) (44) Integrating Eq. (44) leads to Integrating Eq. (44) leads to Integrating Eq. (44) leads to Integrating Eq. (44) leads to !Wb2 = 3! pL2 ! 12M b0 ( )t 2mbL2 ( ) + !WbI (45) (45) where !WbI is the velocity of the middle point of the back beam at the end of the first phase. Then the kinetic energy of the back beam in the second phase is Then the kinetic energy of the back beam in the second phase is EBII =mbL !Wb2 2 3 (46) (46) U Bending-BII = 4Mb0Wb2 L (47) (47) The dissipated spring potential energy is The dissipated spring potential energy is The dissipated spring potential energy is U BSpringII = 2 σ p! wb2d x 0 L" (48) (48) where ! 2.1.2.2 Back beam wb2 is the relative deflection of the back beam at point x in the phase II. 2.1.3 Coupled deformation in the third phase, ( t ≥teq , or tD ! t ! teq ) In the first and second phases, the front beam makes decelerated motion, while the back beam has accelerated motion. If core densification is not reached, the front and back beams will arrive at the same velocity at time t=teq, after which the sandwich beam deforms as one single beam as shown in Fig. 6, which is referred as phase III. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 1664 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1664 Figure 6 (a) Velocity profile of the beams in phase III- system deformation; (b) a free body diagram of the left half beams in phase III- system deformation. Figure 6 (a) Velocity profile of the beams in phase III- system deformation; (b) a free body diagram of the left half beams in phase III- system deformation. At that time, we assume the velocity field of the sandwich beam with the form !wIII = !WIII(1! x / L) 0 " x " L (49) (49) and the governing equation and the governing equation ! 2M III ! 2x = mf + mb ( )(1−x / L) d 2WIII d 2t + m0 d 2WIII d 2t 0 ≤x ≤L (50) (50) Integrating Eq. (50) and considering M eq = M f0 + M b0 , Q = ∂MIII / ∂x = 0 at x=0, we have Integrating Eq. (50) and considering M eq = M f0 + M b0 , Q = ∂MIII / ∂x = 0 at x=0, we have M III = mf + mb ( )(x2 ! x3 / 6L) d 2Weq d 2t + m0x2 2 d 2Weq d 2t + M eq 0 " x " L (51) (51) Considering M III= -Meq at x=L, we have Considering M III= -Meq at x=L, we have Considering M III= -Meq at x=L, we have !! Weq = − 12M eq 2 mf + mb ( )L2 + 3m0L2 (52) Integration Eq. (52) yields !WIII = !Weq − 12Meq 2 mf + mb ( )L2 + 3m0L2 t (53) !! 2.1.3 Coupled deformation in the third phase, ( t ≥teq , or tD ! t ! teq ) Weq = − 12M eq 2 mf + mb ( )L2 + 3m0L2 (52) (52) !WIII = !Weq − 12Meq 2 mf + mb ( )L2 + 3m0L2 t (53) (53) If the densification occurs at t=tD before the common velocity of the faces is reached, the sandwich beam directly deforms as in phase III, which is governed by Eqs. (52) and (53). At that time, !Weq is determined by the momentum balance. If the densification occurs in the first phase of the front and back beams, we have Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1665 1665 m0 !Wf + 2mf !w f t=tD 0 ξ∫ d x + 2mb !wb t=tD d x 0 !∫ = m0 !Weq + 2 mf + mb ( ) !weq 0 ξ∫ d x (54a) m0 !Wf + 2mf !w f t=tD 0 ξ∫ d x + 2mb !wb t=tD d x 0 !∫ = m0 !Weq + 2 mf + mb ( ) !weq 0 ξ∫ d x (54a) (54a) and in the second phase, we have and in the second phase, we have and in the second phase, we have m0 !Wf + 2mf !w f t=tD 0 L! d x + 2mb !wb t=tD d x 0 L! = m0 !Weq + 2 mf + mb ( ) !weq 0 L! d x (54b) (54b) The kinetic energy of system in the phase III is The kinetic energy of system in the phase III is The kinetic energy of system in the phase III is EIII = Emass + E front + Eback = m0 !Weq 2 2 + mf + mb ( )L !Weq 2 3 (55) (55) The dissipated plastic bending energy is The dissipated plastic bending energy is The dissipated plastic bending energy is U Bending-III = 4 M f 0 + M b0 ( )Weq L (56) (56) Since the core is not compressed in the Phase III, the energy dissipated by springs is zero. 2.2 Low strength core ( γ <1 ) In this case, if all of the energy is dissipated before the core densification is reached, the back beam will not deform and the final deflection of the upper beam could be obtained according to the equa- tions given in section 2.1.1.1 and 2.1.2.1 for the front beam. If the densification is reached, the de- formation of the beams will follow Eqs. (52), (53), and (54). Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1666 For the impulse, it is defined as For the impulse, it is defined as I = I L ρ fσ y (59) (59) In addition, the dependent non-dimensional groups are In addition, the dependent non-dimensional groups are In addition, the dependent non-dimensional groups are In addition, the dependent non-dimensional groups are t = t τ , w f t( ) = w f L , wb t( ) = wb L , Wf = Wf L , Wb = Wb L , v f t( ) = v f ρ f hf I , and vb t( ) = vbρ f hf I (60) (60) where τ = L ! y " f is the response time of a plastic string of length 2L made from a material with yield strength σ y and density ρ f . 2.3 Non-dimensional groups Following Tilbrook et al. (2007), we use the same independent and dependent non-dimensional groups. The non-dimensional geometric variables of the sandwich beam are C = C L , h = hf + hb 2C , !h = hb hf (57) (57) and the non-dimensional core properties are ! D , ρ = ρc ρ f , σ yC = σ yC ρσ y (58) (58) Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION For the sandwich beams under blasting loading, Liang et al . (2007) divides their behaviors into three regimes by comparing various time-scales in the responses, which is given in Fig. 7. In the figures, tbd is the time at which the back face begins to decelerate, and teq is the time that the mid- span velocities of the front and back faces are equal. Following their definition, explicit finite ele- ment analysis is performed and the results are compared with the analytical study. In the FE mod- el, the face sheets and the core are assumed perfectly bonded together. Element type Solid164, an 8- node solid structural element provided by the code (LS-DYNA, 2007) is used to discretize the face sheets and the foam core. There are 4050 and 10125 elements for the face sheets and the foam core of the beam, respectively. The vertical, horizontal and rotational displacements of nodes at the ends of the beam are zero. Appropriate mesh refinements near the impact point and the end support are included. Mesh sensitivity is checked before calculations. The automatic time step calculation op- tion from the code is chosen. The contact between the beam and the roller is modeled by using a contact pair surface algorithm with a frictionless contact option (LS-DYNA, 2007). Figure 7 Velocity versus time histories of the mid-span of the front and back faces in three regimes: regime A- strong core; regime B- densification reaches before the velocities equalization; regime C- soft core [4]. Figure 7 Velocity versus time histories of the mid-span of the front and back faces in three regimes: regime A- strong core; regime B- densification reaches before the velocities equalization; regime C- soft core [4]. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1 1667 Firstly the analytical prediction and experimental results are compared. Following Wang et al . (2011), the material and geometrical parameters of sandwich beams are listed in Table 1. The strik- er mass is equal to the mass of the foam projectile used in experiments. Fig. 8 shows the experi- mental results and our analytical prediction for the normalized maximum middle point deflection of the back face. 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION It is seen that when the core makes the large global deformation, our prediction agrees well with the experimental ones. But when the core shear or fractures was present in the experiments of the sandwich structures, which is not considered in the analytical modeling, the deviation of our predictions from the experimental results increases. Nevertheless, during the global deformation stage, our model could well predict the experimental results. Table 1 Material parameters of sandwich beams Figure 8 Comparison between the analytical predictions and experimental results by Wang et al. [10] for (a) Metal foam core, and (b) aluminum honeycomb core. Table 1 Material parameters of sandwich beams Figure 8 Comparison between the analytical predictions and experimental results by Wang et al. [10] for (a) Metal foam core, and (b) aluminum honeycomb core. L ti A i J l f S lid d St t 11 (2014) 1651 1678 Figure 9 shows the contours of plastic strain in the right part of sandwich beam obtained by FE simulation since the deformation is symmetry. Fig. 9a shows the mesh of the model. The impulse I = 4.91×10−6 . Beam geometrical and material parameters are: hf=hb=0.5mm, C=10mm, Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Figure 9 shows the contours of plastic strain in the right part of sandwich beam obtained by FE simulation since the deformation is symmetry. Fig. 9a shows the mesh of the model. The impulse I = 4.91×10−6 . Beam geometrical and material parameters are: hf=hb=0.5mm, C=10mm, Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1668 ! p=0.1MPa ($=2.08), "D=0.7, ! y =75.8MPa. It is seen that during the deformation of the sand- wich beam, the foam core is not only compressed, but also bended as the striker velocities decrease to zero, which may be of significant effect on the overall behavior of the sandwich beam. Therefore, the effect of compression and bending of the foam core on global deformation should be considered in the theoretically analysis. In the following discussion, we compare the finite element results with the aforementioned analytical solutions. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Figure 9 Numerical results for the impact response of the sandwich beam. The impulse 6 4.91 10 I ! = " , V=20m/s. Beam geometri- cal and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION (a) Mesh of the sand- wich beam; (b) Distribution of plastic strain when 0.15 t = ; (c) Distribution of plastic strain when 0.5 t = ; (d) Distribution of plastic strain when 1.5 t = . tin American Journal of Solids and Structures 11 (2014) 1651 1678 gure 9 Numerical results for the impact response of the sandwich beam. The impulse 6 4.91 10 I ! = " , V=20m/s. Beam geometri- al and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. (a) Mesh of the sand- wich beam; (b) Distribution of plastic strain when 0.15 t = ; (c) Distribution of plastic strain when 0.5 t = ; (d) Distribution of plastic strain when 1.5 t = . Figure 9 Numerical results for the impact response of the sandwich beam. The impulse 6 4.91 10 I ! = " , V=20m/s. Beam geometri- cal and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. (a) Mesh of the sand- wich beam; (b) Distribution of plastic strain when 0.15 t = ; (c) Distribution of plastic strain when 0.5 t = ; (d) Distribution of plastic strain when 1.5 t = . Figure 9 Numerical results for the impact response of the sandwich beam. The impulse 6 4.91 10 I ! = " , V=20m/s. Beam geometri- cal and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. (a) Mesh of the sand- wich beam; (b) Distribution of plastic strain when 0.15 t = ; (c) Distribution of plastic strain when 0.5 t = ; (d) Distribution of plastic strain when 1.5 t = . Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1669 Figure 10 shows the comparison of the predicted results from the present model and that of Til- brook et al . (2007), as well as the numerical ones from FE simulation. It is seen that because the shear/bending strength of the core prior to the equalization of the velocities of the front and back faces (that is, in the first and second phases) is ignored, the velocity of the front beam, predicted by Tilbrook et al . 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION (red dash-dot lines), is a little slower than the FE results (blue dash lines), whilst the velocity of the back beam is larger than those given by FE simulation. The velocity equivalent reaches in the first phase, and the responses of the sandwich beam goes directly into phase III, which corresponds to Regime A in Fig. 7a. It is seen that the front and back faces reach a higher equal velocity much earlier than the numerical one, which results in the over-estimation of the mid- span deflection of the face-sheets. This great discrepancy further indicates the necessity of consider- ing the shear/bending strength of the core even prior to the equalization of the face velocities. In our modeling (black solid lines), the velocities of the front and back beams are coincided well with those given by FE simulation. The core densification occurs before the equivalence of beam velocities in the first phase, and the responses of the sandwich beam follow directly into phase III, where the sandwich beam deforms as a single beam, corresponding to Regime B given in Fig. 7b. In phase III, the velocity deviation from FE simulations is due to the assumption made for the core in the analytical modeling, that is, the core is treated rigid after densification. In fact, in FE simula- tion, at time t=tD, the core densification at the mid-span occurs. However, the other parts of the core have not yet been densified. This is just why no obvious velocity impact discontinuity is ob- served in the FE results. Comparison between the results given by our model and FE simulation shows that our model can well predict the mid-span deflection of the sandwich beam. It is also no- ticed that the response of the back beam is postponed in our modeling. This is partly due to the increase of the bending moment of the back beam in our model since the bending strength of the core is considered (Eq. 4). The comparison indicates that our model in treatment of the core defor- mation is reasonable and could well describe the dynamic responses of sandwich beams under a projectile impact. Figure 10 Comparison of analytical predictions of our and Tilbrook models with respect to the data given by FE simulations for normal- ized mid-span velocity and deflection of front and back face-sheets. The impulse I = 9.82 ! 10"6 , V=40m/s. 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION Beam geometrical and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Figure 10 Comparison of analytical predictions of our and Tilbrook models with respect to the data given by FE simulations for normal- ized mid-span velocity and deflection of front and back face-sheets. The impulse I = 9.82 ! 10"6 , V=40m/s. Beam geometrical and material parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1670 Figs. 11 to 13 give the analytical and FE predictions of the normalized mid-span velocity and deflection of front and back beams under different impulse loadings. The results before the mid- span velocity reaches zero for the first time are given. Seen as Fig. 11, under low impulse loading ( 6 6.63 10 I − = × ), the mid-span velocities of the front and back beams reach the equal value in the first phase, and then goes into phase III- system deformation directly, which has the same behavior as that shown in Fig. 7a. The mid-span deflection predicted by our model and FE simulation coin- cides well with each other. In this situation, the velocity deviation in phase III is due to the theoret- ical assumption that the beams deform as a system after the velocity equalization. Actually, in the FE simulation, before the velocity equivalence, the front beam makes the decelerated motion, while the back beam accelerated one. At time t=teq, the equalization of the mid-span velocities reaches. Then at the next time, the velocity of the back face will be larger than that of the front face, which results in the traction of the core. At that time, the directions of the pressures act on the face sheets have changed. The back face will make the decelerated motion whilst the front face acceler- ated one, which results in the equal face velocity a few times later. Then the front and back faces will make the same motion again until all of the kinematic energy is dissipated. This process also indicates that at the first velocity equalization, due to the direction inverse of the face pressures and the large acceleration, the delimitation of the back face with the core maybe happen. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION Figure 11 Analytical and FE predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=30m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Figure 11 Analytical and FE predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=30m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Along with the increase of the loading impulse ( I = 9.82 ×10−6 , Fig. 12), the densification reaches before the equalization of the beam mid-span velocities, then the responses of the beams follows phase III, as shown in Fig. 7b. It is noticed that the predicted velocity of the back face is the same as that given by FE simulation. But the responses of the front and back beams are not synchro- nous. The responses of the back beam are postponed. According to Eqs. (26) and (16), the delay time could be determined. However, the mid-span deflection predicted by our model agrees well with those given by FE simulation. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1671 Figure 12 Analytical and FEM predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=40m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Figure 12 Analytical and FEM predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=40m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Along with the further increase of the loading impulse ( I =13.2 ×10−6 , Fig. 13), due to the post- poned responses of the back beam, the densification occurs before the motion of the back beam, and the responses of the sandwich beam goes directly into phase III. It is seen that the discrepancy be- tween the analytical predictions of mid-span normalized deflection and the numerical ones could be ignored. Figure 13 Analytical and FEM predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=60m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. Figure 13 Analytical and FEM predictions of normalized mid-span velocity and deflection of front and back face-sheets with V=60m/s. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, σp=0.1MPa (γ=2.08), εD=0.7, σy =75.8MPa. 3 DYNAMIC RESPONSE IN TERMS OF DEFLECTION Figure 14 shows the comparison between the analytical predictions and numerical simulations for the normalized kinematic energy and plastic energy with respect to the time. The loading im- pulse I = 9.82 ×10−6 (V=40m/s, Fig. 12). It is seen that in phase I, the analytical and numerical ones agree well with each other. In the third phase, the differences are due to the continuously cou- pling deformation of the face sheets and the core, which is ignored in the analytical modeling. But our model predicts well the final responses of the sandwich beams. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1672 Figure 14 The time histories of normalized kinetic and plastic energies in the face sheets and the foam core of sandwich beams. Figure 14 The time histories of normalized kinetic and plastic energies in the face sheets and the foam core of sandwich beams. Figure 15 gives the normalized mid-span velocity with respect to the time under the same load- ing impulse but different impacting mass. For light mass impacting, the difference between the initial mid-span velocity of the front beam Vf \|t =0 and the equivalent velocity of the face sheets is great, which results in larger mid-span deflection of the front beam. It is seen that under the same impulsive loading, densification at the mid-span of the beam occurs when m0=0.0125kg (red dash dot lines), which is due to the formation of the concave dent on the front beam in phase I, with the dent width inverse proportion to the impact mass (Eq. 16, mass increasing causing the decreased plastic hinge velocity). So for certain loading impulse, under light mass impacting, the local defor- mation phenomenon near the impacting zone is predominating. Along with the increase of the im- pact mass, the value of the equivalent velocity and the time teq are both decreased. The difference between the initial front beam mid-span velocity Vf \|t =0 and the equivalent velocity of the face sheets is decreased, which results in the decreased mid-span deflection of the front beam. That is, along with the increase of impacting mass, the deformation of the sandwich beam is global bending dominated. Figure 15 gives the normalized mid-span velocity with respect to the time under the same load- ing impulse but different impacting mass. For light mass impacting, the difference between the initial mid-span velocity of the front beam Vf \|t =0 and the equivalent velocity of the face sheets is atin American Journal of Solids and Structures 11 (2014) 1651-1678 Figure 15 Comparison of normalized front and back face-sheet mid-span velocities versus time under the same impulse loading ( I = 9.82 ! 10"6 ) but different impact mass. Figure 15 Comparison of normalized front and back face-sheet mid-span velocities versus time under the same impulse loading ( I = 9.82 ! 10"6 ) but different impact mass. Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1673 Figure 16 displays the normalized mid-span velocity and deflection with respect to the time his- tories for different sandwich beams under the same impulse ( I = 6.63! 10"6 ). For soft core (! <1 , Figure 16 displays the normalized mid-span velocity and deflection with respect to the time his- tories for different sandwich beams under the same impulse ( I = 6.63! 10"6 ). For soft core (! <1 , Fig. 16a), the densification occurs in the phase II and then the front and back beams deform sys- tematically in phase III. During this process, the back beam is assumed un-deformed until densifica- tion occurs (our model, solid lines). It is noticed that in the FE simulation, the back face sheet is not stationary although the final deflection of the back beam could be ignored (FE results, dash lines). Along with the increase of the core strength (! >1), the equalization of mid-span velocities of beams may happen in the first or second phase, which is then followed by the systematical defor- mation of the sandwich beam (Figs. 16b and 16c). It is seen that along with the increase of the core strength, the time to the equalization of beam mid-span velocity is decreased, which is due to the accelerated motion of the back beam, but decelerated motion of the front beam. It is seen in all of these situations, our model can well predict the mid-span deflection of the beams. Figure 16 Analytical and FEM predictions of the normalized front and back face-sheet mid-span velocity and deflection versus time for sandwich beams with different strength foam core: (a) sp =0.01MPa; (b) sp =0.05MPa; (c) sp =0.1MPa. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, εD=0.7, σy =75.8MPa. Figure 16 Analytical and FEM predictions of the normalized front and back face-sheet mid-span velocity and deflection versus time for sandwich beams with different strength foam core: (a) sp =0.01MPa; (b) sp =0.05MPa; (c) sp =0.1MPa. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, εD=0.7, σy =75.8MPa. Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1674 Figure 16 (continued) Analytical and FEM predictions of the normalized front and back face-sheet mid-span velocity and deflection versus time for sandwich beams with different strength foam core: (a) sp =0.01MPa; (b) sp =0.05MPa; (c) sp =0.1MPa. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, εD=0.7, σy =75.8MPa. Figure 16 (continued) Analytical and FEM predictions of the normalized front and back face-sheet mid-span velocity and deflection versus time for sandwich beams with different strength foam core: (a) sp =0.01MPa; (b) sp =0.05MPa; (c) sp =0.1MPa. Material and geometrical parameters are: hf=hb=0.5mm, C=10mm, εD=0.7, σy =75.8MPa. 4 CONCLUDING REMARKS In this paper, a modified beam-spring model is proposed to investigate the dynamic responses of a clamped sandwich beam under impulsive projectile impact. Different from Tilbrook et al . (2007), that is, the bending/shear effect of the core is ignored before the velocity equivalence of the face sheets, the bending effects of the core are considered in the whole deformation process of the sand- wich beams. Different from beam-spring system on-rigid foundation (Chen and Yu, 2000; Yu et al ., 2002), the coupling motion of the front and back beams is considered. The comparison among ana- lytical predictions, experimental data, and numerical simulations shows that our model enables acceptable predictions of the responses of the sandwich beams under mass impact when the foam core makes large global deformation. It should be pointed out that although the present model is based on small deflection, it has po- tential to be further developed by incorporating more factors. For instance, large deflections and local deformation characteristic of the foam core are expectable to be incorporated into the model, which will be given in the forthcoming papers. Acknowledgements The financial supports from the National Science Foundation of China (No. 11272046, 11172033), the National High Technology Research and Development Program of China (863 Program, No. 2013AA030901), the National Basic Research Program of China (973 Program) (2010CB7321004), and 111 pro- ject, are acknowledged. The first author also would thanks for the support by a fund from Defense Research and Technology Office, NTU, Singapore, and the Program for New Century Excellent Talents in University (NCET). Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 References Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1676 References Fleck, N.A., Deshpande, V.S. (2004). The resistance of clamped sandwich beams to shock loading. ASMA J Appl Mech 71: 386-401. Xue, Z., Hutchinson, J.W. (2004). A comparative study of blast-resistant metal sandwich plates. Inter J Impact Eng 30: 1283-1305. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1675 Rabczuk, T., Kim, J.Y., Samaniego, E., et al . (2004). Homogenization of sandwich structures. Inter J Num Methods Eng 7: 1009-1027. uk, T., Kim, J.Y., Samaniego, E., et al . (2004). Homogenization of sandwich structures. Inter J Num ods Eng 7: 1009-1027. Liang, Y., Spuskanyuk, A.V., Flores, S.E., et al . (2007). The response of metallic sandwich panels of water blast. J Appl Mech 74: 81-99. Deshpande, V.S., Fleck, N.A. (2005). A one-dimensional response of sandwich plates to underwater shock load- ing. J Mech Phys Solids 53: 2347-2383. Taylor, G.I. (1963). The pressure and impulse of submarine explosion waves on plates. The Scientific Papers of G I Taylor, vol. III: Cambridge University Press. Tilbrook, M., Deshpande, V.S., Fleck, N.A. (2007). The impulsive response of sandwich beams: Analytical and numerical investigation of regimes of behavior. J Mech Phys Solids 54: 2242-2280. Qin, Q.H., Wang, T.J., Zhao, S.Z. (2009). Large deflections of metallic sandwich and monolithic beams under lo- cally impulsive loading. Inter J Mech Sci 51: 752-773. Qin, Q.H., Wang, T.J. (2011). Low-velocity heavy-mass impact response of slender metal foam core sandwich beam. Compos Struct 93: 1526-1537. Wang, Z.H., Lin, J., Ning, J.G., Zhao, L.M. (2011). The structural response of clamped sandwich beams subject to impact loading. Compos Struct 93: 1300-1308. Liu, Y., Zhang, X.C. (2009). The influence of cell micro-topology on the in-plane dynamic crushing of honey- combs. Int J Impact Eng 36: 98-109. Liu, Y., Wu, H.X., Wang, B. (2012). Gradient design of metal hollow sphere (MHS) foams with density gradient. Compos part B 43: 1346-1352. Chen, X.W., Yu, T.X. (2000). Elastic–plastic beam-on-foundation under quasi-static loading. Int J Mech Sci 42: 2261–2281. Yu, T.X., Chen, X.W., Chen, Y.Z. (2002). Elastic–plastic beam-on-foundation subjected to mass impact or im- pulsive loading. Comp Struct 80: 1965-1973. Lopatnikov, S.L., Gama, B.A., Gillespie, Jr J.W. (2007). Modeling the progressive collapse behavior of metal foams. Int J Impact Eng 34: 587-95. Jones N (1989). Structural Impact. Cambridge: Cambridge University Press9. LSTC (2007). LS-DYNA keyword user’s manual. Livermore Software Technology Corporation. Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Appendix A: Clamped monolithic beam under mass impact (Jones, 1989) W(1−x / ξ) + !Wx !ξ / ξ 2]xd 0 ξ∫ x = 0 (A8) that is, 2M 0 −m( !! Wξ 2 / 6 + !W !ξξ / 3) = 0 (A9) 2M 0 −m( !! Wξ 2 / 6 + !W !ξξ / 3) = 0 (A9) 2M 0 −m( !! Wξ 2 / 6 + !W !ξξ / 3) = 0 (A9) Then Eq. (A9) is rewritten as Then Eq. (A9) is rewritten as d( !Wξ 2) / dt =12M 0 / m (A10) d( !Wξ 2) / dt =12M 0 / m d( !Wξ 2) / dt =12M 0 / m (A10) (A10) Integrating Eq. (A10) and considering when t=0, #=0, we have t = m !Wξ 2 /12M 0 (A11) (A11) Combining with Eq. (A6), we have t = mm1V0ξ 2 /[12M 0(m1 + mξ)] (A12) (A12) Differentiating Eq. (A12) with respect to the time t, we obtain the traveling velocity of the plas- tic hinges, that is Differentiating Eq. (A12) with respect to the time t, we obtain the traveling velocity of the plas- tic hinges, that is !ξ =12M 0(m1 + mξ)2 /[mm1V0ξ(2m1 + mξ)] (A13) (A13) Since no transverse displacement occurs before the moving plastic hinge arrives at t(x), when t ≥t(x) , the transverse displacement is determined by w = !w t(x) t∫ dt (A14) (A14) where t(x) is obtained according to Eq. (A12) by letting #=x. ! x) is obtained according to Eq. (A12) by letting #=x. !ξ = dξ / dt combining with Eq (A1) Eq (A14) is rewritten as where t(x) is obtained according to Eq. (A12) by letting #=x. Let !ξ = dξ / dt , combining with Eq. (A1), Eq. (A14) is rewritten as where t(x) is obtained according to Eq. (A12) by letting #=x. Let !ξ = dξ / dt , combining with Eq. (A1), Eq. (A14) is rewritten as Let !ξ = dξ / dt , combining with Eq. (A1), Eq. (A14) is rewritten as w = !w x ξ∫ dξ / !ξ (A15) w = !w x ξ∫ dξ / !ξ (A15) ution of Eqs. (A1), (A6), (A13) into (A15) yields Substitution of Eqs. (A1), (A6), (A13) into (A15) yields Substitution of Eqs. Appendix A: Clamped monolithic beam under mass impact (Jones, 1989) In the first phase, considering the symmetric deformation, the right half beam 0 ≤x ≤L is consid- ered. Its transverse velocity is expressed as (Fig. 4a) !w = !W(1! x / ! ) 0 ! x ≤! (A1a) !w = 0 ξ ≤x ≤L (A1b) !w = !W(1! x / ! ) 0 ! x ≤! (A1a) (A1a) !w = 0 ξ ≤x ≤L (A1b) (A1b) where # is dependent on the time t and the position x, and the upper dot means the differential with respect to the time t. At the moving plastic hinges (x=±#), the bending moment M0 is the maximum and the transverse force Q=0. The force balance between the two moving plastic hinges along the transverse direction yields m0 !! W + 2 m1!!w 0 ξ∫ dx = 0 (A2) (A2) where m1 is the mass per length of the beam. Substitution Eq. (A1) into Eq. (A2) leads to m0 !! W + 2m1 [ !! W(1−x / ξ) + !Wx !ξ / ξ 2]d 0 ξ∫ x = 0 (A3) (A3) that is, that is, m0 !! W + m1( !! Wξ + !W !ξ) = 0 (A4) m0 !! W + m1( !! Wξ + !W !ξ) = 0 (A4) Considering t=0, !W = 0 , and ! = 0 , the time integral of Eq. (A4) leads to Considering t=0, !W = 0 , and ! = 0 , the time integral of Eq. (A4) leads to m1 !W + m !Wξ = m1V0 (A5) m1 !W + m !Wξ = m1V0 (A5) that is, that is, !W =V0 / (1+ m0ξ / m1) (A6) (A6) Considering the moment balance between x=0 and x=# to the beam midpoint, and the boundary conditions: M=M01 at x=0; M= -M01 and Q=0 at x=#, we have 2M0 − m0 !!w 0 ξ∫ xdx = 0 (A7) (A7) Substitution Eq. (A1) into Eq. (A7) yields Substitution Eq. (A1) into Eq. (A7) yields n Eq. (A1) into Eq. (A7) yields Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1677 1677 2M 0 −m [ !! W(1−x / ξ) + !Wx !ξ / ξ 2]xd 0 ξ∫ x = 0 (A8) 2M 0 −m [ !! Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Appendix A: Clamped monolithic beam under mass impact (Jones, 1989) (A1), (A6), (A13) into (A15) yields w = V0(1−x / ξ)mm1V0ξ(2m1 + mξ)dξ (1+ mξ / m1)12M 0(m1 + mξ)2 x ξ∫ (A16) (A16) Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Latin American Journal of Solids and Structures 11 (2014) 1651-1678 Ying Liu et al./An analytical model of a clamped sandwich beam under low-impulse mass impact 1678 Integration the above equation, we have Integration the above equation, we have w = m1 2V0 2 24mM0 [ 1+ ! (1+" )2 −1+ 2! 1+ ! + 2! 1+" + 2ln1+" 1+ ! ] (A17) (A17) where α = mξ / m1 , β = mx / m1. where α = mξ / m1 , β = mx / m1. In the second phase (Fig. 4c), it assumes that the corresponding transverse displacement field obeys w2 = L −x ( )θ (A18) (A18) The energy conversation leads to 4M 0θ = m0V0 2 1+ 2mf L / 3m0 −x ( ) / 2 1+ mf L / m0 ( ) 2 ⎡ ⎣⎢ ⎤ ⎦⎥ (A19) (A19) According to Eqs. (A18) and (A19), we have According to Eqs. (A18) and (A19), we have w2 = m0V0 2L 1+ 2mf L / 3m0 ( ) 1−x / L ( ) / 8M 0 1+ mf L / m0 ( ) 2 ⎡ ⎣⎢ ⎤ ⎦⎥ (A20) w2 = m0V0 2L 1+ 2mf L / 3m0 ( ) 1−x / L ( ) / 8M 0 1+ mf L / m0 ( ) 2 ⎡ ⎣⎢ ⎤ ⎦⎥ (A20) (A20)
https://openalex.org/W2924245642	https://europepmc.org/articles/pmc6522648?pdf=render	English	null	Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging	Analytical and bioanalytical chemistry/Analytical & bioanalytical chemistry	2,019	cc-by	7,435	Introduction and not well understood [2, 3]. One approach to understanding disease pathophysiology is to study the chemical alterations resulting from the diseased state. An increasingly popular group of molecules to study for understanding pathophysiol- ogy is small metabolites. Small metabolites are found as in- termediate and final products of all chemical processes in living systems and metabolic profiling of biological samples has become a viable approach due to recent technological advancements [4]. Diabetic kidney disease (DKD) is a common complication of diabetes that can lead to end-stage renal disease (ESRD). ESRD is a fatal condition requiring dialysis and ultimately renal transplantation [1]. Despite decades of research, the un- derlying molecular mechanisms behind DKD are still debated Parts of this work were presented at XIXth Euroanalysis 2017, Stockholm (Sweden) in August/September 2017, and have been awarded with an ABC Best Poster Award. Hilde-Marléne Bergman and Lina Lindfors contributed equally to this work. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00216-019-01721-5) contains supplementary material, which is available to authorized users. * Ingela Lanekoff Ingela.Lanekoff@kemi.uu.se 1 Department of Chemistry-BMC, Uppsala University, Box 599, 751 24 Uppsala, Sweden 2 Department of Medical Cell Biology, Uppsala University, Box 571, 751 23 Uppsala, Sweden Parts of this work were presented at XIXth Euroanalysis 2017, Stockholm (Sweden) in August/September 2017, and have been awarded with an ABC Best Poster Award. Hilde-Marléne Bergman and Lina Lindfors contributed equally to this work. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00216-019-01721-5) contains supplementary material, which is available to authorized users. * Ingela Lanekoff Ingela.Lanekoff@kemi.uu.se 1 Department of Chemistry-BMC, Uppsala University, Box 599, 751 24 Uppsala, Sweden 2 Department of Medical Cell Biology, Uppsala University, Box 571, 751 23 Uppsala, Sweden Parts of this work were presented at XIXth Euroanalysis 2017, Stockholm (Sweden) in August/September 2017, and have been awarded with an ABC Best Poster Award. Metabolic profiling provides detailed analysis of small me- tabolites in a sample and can be used to understand altered metabolic pathways in disease states. Mass spectrometry (MS) is a common technique for metabolic profiling for both targeted and untargeted analyses [5]. One main advantage of using MS for metabolomics lies in its untargeted nature and ability to detect hundreds of chemical species simultaneously without labelling or preselection. Abstract Diabetic kidney disease is a serious complication of diabetes that can ultimately lead to end-stage renal disease. The pathogenesis of diabetic kidney disease is complex, and fundamental research is still required to provide a better understanding of the driving forces behind it. We report regional metabolic aberrations from an untargeted mass spectrometry imaging study of kidney tissue using an insulinopenic rat model of diabetes. Diabetes was induced by intravenous injection of streptozotocin, and kidneys were harvested 2 weeks thereafter. Imaging was performed using nanospray desorption electrospray ionization connected to a high- mass-resolving mass spectrometer. No histopathological changes were observed in the kidney sections; however, mass spec- trometry imaging revealed a significant increase in several 18-carbon unsaturated non-esterified fatty acid species and monoacylglycerols. Notably, these 18-carbon acyl chains were also constituents of several increased diacylglycerol species. In addition, a number of short- and long-chain acylcarnitines were found to be accumulated while several amino acids were depleted. This study presents unique regional metabolic data indicating a dysregulated energy metabolism in renal mitochondria as an early response to streptozotocin-induced type I diabetes. Keywords nano-DESI . Acylcarnitine . Fatty acid oxidation . Branched-chain amino acid . Streptozotocin . Type I diabetes Analytical and Bioanalytical Chemistry (2019) 411:2809–2816 https://doi.org/10.1007/s00216-019-01721-5 Analytical and Bioanalytical Chemistry (2019) 411:2809–2816 https://doi.org/10.1007/s00216-019-01721-5 RESEARCH PAPER Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging Hilde-Marléne Bergman1 & Lina Lindfors1 & Fredrik Palm2 & Jan Kihlberg1 & Ingela Lanekoff1 Received: 7 December 2018 /Revised: 12 February 2019 /Accepted: 26 February 2019 /Published online: 20 March 2019 # The Author(s) 2019 * Ingela Lanekoff Ingela.Lanekoff@kemi.uu.se 1 Department of Chemistry-BMC, Uppsala University, Box 599, 751 24 Uppsala, Sweden 2 Department of Medical Cell Biology, Uppsala University, Box 571, 751 23 Uppsala, Sweden Introduction The features of MS have been important in studies screening biofluids—for metabolites as biomarkers for diagnosis of DKD, as well as to predict progression to ESRD [6]. Hilde-Marléne Bergman and Lina Lindfors contributed equally to this work. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00216-019-01721-5) contains supplementary material, which is available to authorized users. 1 Department of Chemistry-BMC, Uppsala University, Box 599, 751 24 Uppsala, Sweden In addition to analysing biofluids, MS can be used to ac- quire data directly from defined locations on the surface of thin tissue sections using mass spectrometry imaging (MSI) 2810 H.-M. Bergman et al. [7, 8]. By collecting spatially resolved MS data, the distribu- tion and relative abundance of detected ions can be visualised in two-dimensional (2D) maps. These 2D maps provide unique insights into chemical microenvironments at selected locations and thereby the biochemical system. Previous MSI studies on kidneys of diabetic rodent models indicate the importance of lipids and energy metabolism in disease pathology [9, 10]. Specifically, the kidney of db/db mice with DKD 16 weeks after diabetes onset showed anatomically specific increases of four lipid classes: gangliosides, sulfoglycosphingolipids, lysophospholipids, and phosphatidylethanolamines [9]. In ad- dition, the level of sphingomyelin 18:1/16:0 and the ATP/AMP ratio were increased in the glomeruli of Akita mice with DKD 19 weeks after diabetes onset [10]. These studies show that MSI can be used to register metabolic alterations in DKD > 16 week after diabetes onset and advance knowledge on dis- ease progression. 24 h after injection using a test reagent strip from blood sam- ples obtained from a cut of the tip of the tail. Animals were considered diabetic if blood glucose concentrations increased to ≥15 mmol/l. Age-matched normoglycaemic rats were used as controls. Two weeks after induction of diabetes, animals were anaesthetised (n = 3/group) with thiobutabarbital (i.p., 120 and 80 mg/kg bw for controls and diabetic animals, re- spectively, due to their different responses to anaesthetics), and kidneys were rapidly dissected and frozen in liquid nitro- gen. One kidney per animal was sectioned into 12-μm-thick sections (Leica CM3000, Leica Biosystems, Nussloch, Germany) and thaw mounted onto regular glass slides. Water was used to hold the fresh frozen tissue during section- ing, and all sections were stored at −80 °C prior to nano-DESI MSI analysis. nano-DESI MSI A custom-built nano-DESI MSI source was assembled as pre- viously described [11], and a capillary holder was used to fix the primary and secondary capillaries at an angle [18]. The nano- DESI solvent consisted of a 9:1 solution of methanol:water (v/v), with an addition of 3 μmol/l lysophosphatidylcholine 19:0, and was supplied at a rate of 0.5 μl/min. Imaging was performed by moving the sample in lines under the nano-DESI probe at a speed of 60 μm/s in the x-direction and a step size of 200 μm between the lines in the y-direction. The average acquisition time of 0.7 s per spectrum resulted in pixel sizes of ~ 42 × 200 μm. Three tissue sections from one kidney each of 3 control and 3 STZ-treated rats were analysed by nano-DESI MSI in a random order. Introduction Kidney function (data presented in the Electronic Supplementary Material (ESM) Table S1) was determined in parallel in both control and diabetic animals (n = 4/group) as previously described [16]. In brief, anaesthesia was induced using thiobutabarbital and body temperature maintained at 37.5 °C using a servo-controlled heating pad. Tracheostomy was done, and catheters were inserted into the femoral artery for monitoring of the mean arterial blood pressure, into the femoral vein for infusion of 3H-inulin (185 kBq h−1 kg−1; controls 5 ml h−1 kg−1 and diabetics 10 ml h−1 kg−1), and into the bladder to allow drainage. The left kidney was immobilised in a plastic cup and the left ureter was catheter- ised in order to collect urine for subsequent measurements of left kidney function. Glomerular filtration rate was calculated from the urinary excretion rate of 3H-inulin, using a standard liquid scintillation technique [17]. Urinary protein concentra- tion was analysed according to manufacturer’s protocol (DC protein assay, Bio-Rad Laboratories, Sundbyberg, Sweden). In the present study, we employ nanospray desorption electrospray ionization (nano-DESI) MSI to investigate meta- bolic alterations in thin kidney tissue sections from the insulinopenic streptozotocin (STZ) type 1 diabetes rat model, 2 weeks after disease onset [11, 12]. nano-DESI is a liquid extraction technique where analytes are desorbed from the sample surface in ambient conditions and with minimal sam- ple preparation. For imaging, the sample is continuously moved under the nano-DESI probe while data is continuously acquired. This generates a matrix of mass spectra where each spectrum corresponds to a specific point on the sample sur- face. While previous metabolomic studies focused on investi- gating chemical changes in tissue of diabetic kidney or DKD after > 16 weeks [13, 14], this study focuses on metabolic alterations that occur only 2 weeks after disease onset. We report that despite the lack of histological alterations at this early stage, the metabolism has already shifted. In particular, there is a dysregulation of the citric acid cycle, fatty acid oxidation (FAO), and branched-chain amino acid (BCAA) catabolism within the renal cortex. These results provide new insights into the complex metabolic consequences of di- abetes and disease progression. Data analysis After nano-DESI MSI, the analysed tissue sections were stained by haematoxylin & eosin (H&E). The protocol is de- scribed in the ESM. Regions of interest (ROIs) of the cortex together with the outer strip of the outer medulla, and the inner strip of the outer medulla together with the inner medulla, were manually defined based on optical images of the stained tissue sections. Microscopy images of H&E-stained tissue sections were used for histological evaluation. Data containing m/z values and intensities were extracted from Xcalibur raw files using Decon2LS [19]. Following this, data matrices were generated and mass spectra were extracted from defined ROIs using an in-house script [20]. For further comparisons, all intensities were normalised to the total ion current (TIC) and increased intensities were interpreted as increased abundances. Welch’s t test was used to select m/z values with significantly (p < 0.05) different relative intensi- ties between control and diabetic tissues. Only m/z values that were present in > 5% of the pixels in each ROI and in > 25% of all tissue sections were chosen for further investigation. In addition, only m/z values showing significant differences in both the [M+Na]+ and [M+K]+ ion channels were selected. All abundances are interpreted from TIC-normalised data. Ion images were generated using MSIQuickView, and the localisation of all biologically relevant peaks to the kidney tissue was verified by manual inspection [11]. Animals and sample preparation All animal procedures were approved by the local animal ethics committee in Uppsala and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Diabetes mellitus was induced by injection of STZ (55 mg/kg bw, Sigma-Aldrich, St. Louis, MO, USA) in the tail vein of male Sprague-Dawley rats (Charles River, Sulzfeld, Germany) weighing approximately 300 g [15]. Blood glucose concentrations were determined All mass spectrometric analyses were performed in an untargeted fashion on a Q-Exactive™Plus Orbitrap™ (Thermo Fisher Scientific, Bremen, Germany). MSI data ac- quisition was performed in positive mode with a scan window Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging 2811 medulla (OS), inner strip of the outer medulla (IS), and inner medulla (IM). Ion images generated with nano- DESI MSI reflect these anatomical regions and reveal their differences in chemical composition. More than 250 ion images of low molecular weight ions with unique chemical formulas were acquired from kidney tissue sections with nano-DESI MSI. Of all the detected ions, the majority lo- calise to the OS and/or the cortex, while 50 ions are dis- tributed evenly over the tissue section and ~ 75 ions are localised to the IS. Methylhistidine (Fig. 1b) is, for exam- ple, more abundant in the OS and propionylcarnitine (C3) is mainly localised to the cortex (Fig. 1c). While these metabolites show complementary distributions, the mem- brane lipid sphingomyelin 34:1 localises to both of these regions (Fig. 1d). Further, betaine [21] (Fig. 1e) is mainly localised to the IS and sorbitol (Fig. 1f) mainly to the IM. The large amount of metabolites detected and imaged with nano-DESI MSI can provide novel insights into localised metabolism and biological function in kidney tissue. of m/z 100–1000, using a mass resolution of 140,000 (m/Δm at m/z 200). The instrument was externally calibrated, the spray voltage was set to 3 kV, and the heated capillary tem- perature was set to 300 °C. Metabolism is altered in the diabetic kidney tissue Two weeks after STZ treatment, rats were deemed diabetic with high glucose blood levels. In addition, they showed signs of kidney dysfunction, such as increased glomerular filtration rate, urine flow, and urinary protein excretion (ESM Table S1). However, at this early stage of disease, there were no patho- logical changes correlating to DKD observed in the cortex during histological evaluation of the tissue (Fig. 2). The lack of histological changes in the tissue, despite clear signs of kidney dysfunction, suggests that metabolic alterations are in- volved in early disease pathogenesis. Metabolite distributions and abundances in thin kidney tissue sections were imaged by nano-DESI MSI using three sections from one kidney each of 3 age-matched controls and 3 STZ diabetes model rats, 2 weeks after STZ treatment. The results show a significant change in abundance of 38 annotated endogenous metabolites, including glucose, in the cortex and OS of the kidney (Table 1, details on annotation in ESM Tables S2-S4 and S6-S11). Specifically, the abundance of 24 metabolites was significantly increased while the abundance of 14 metabolites, including several amino acids, was significantly decreased (p value < 0.05, ESM Table S2). This suggests a highly altered metabolism in the kidney tissue already 2 weeks after diabetes onset. The altered metabolites in the cortex and OS anatomical regions can be divided into six metabolite groups: non-esterified fatty acids (NEFAs); monoacylglycerols (MG); diacylglycerols (DG); short-chain and long-chain acylcarnitines; and amino acids. Species of NEFA, MG, DG, and short-chain and long-chain acylcarnitines had higher abundance in the diabetic kidney while amino acids had lower abundance in the diabetic tissue (ESM Table S5 and Fig. S1). Analyte identification The total number of endogenous compounds detected in a control tissue section was estimated by searching all de- tected m/z values in the human metabolome database (http://www.hmdb.ca) and Metlin (https://metlin.scripps. edu) to exclude biologically irrelevant peaks. The number of detected endogenous compounds was determined by removing all duplicate hits with the same elemental composition in addition to ions detected as several adduct ions. Analyte identification strategies are further described in the ESM. Acylcarnitines accumulate in diabetic kidney tissue Both short- and long-chain acylcarnitine species were found significantly increased (p value < 0.01) in the cortex of diabetic kidney tissue compared to control. The large differences in sig- nal intensities of the short-chain and long-chain acylcarnitines are displayed on a logarithmic scale in Fig. 4a, clearly showing the accumulation of acylcarnitines in the diabetic kidney. While the data in Fig. 4a is specific to the cortex and OS, the ion images in Fig. 4b–i show the distribution in the entire tissue section. The ion images reveal that the distribution of short-chain acylcarnitines is highly dependent on the acyl chain of the re- spective species (Fig. 4b–f). Specifically, all short-chain acylcarnitines are detected with high intensity in the cortex but show variable localisation to the OS and the medulla. While the localisation of C3 and isohydroxybutyrylcarnitine (C4- OH) to the medulla is minor (Fig. 4c, e), acetylcarnitine (C2), isobutyrylcarnitine (C4), and isovaleryl carnitine (C5) show some localisation in medullary regions (Fig. 4b, d, f). In contrast, all elevated long-chain acylcarnitines show more dispersed localisation across the diabetic kidney tissue section (Fig. 4g–i and ESM Fig. S4) and an almost complementary distribution to the altered NEFA, MG, and DG species that primarily localise to the OS of the kidney (Fig. 3b–e). Fig. 1 Anatomical regions of kidney tissue have distinct molecular composition. (a) Optical image of kidney section with an overlay highlighting anatomical regions. (b) Ion image of [methylhistidine+H]+ (m/z 170.0923). (c) Ion image of [propionylcarnitine+H]+(m/z 218.1386). (d) Ion image of [sphingomyelin 34:1+K]+ (m/z 741.5307). (e) Ion image of [betaine+Na]+ (m/z 140.0681. (f) Ion image of [sorbitol+Na]+ (m/z 205.0681) Results Specifically, all short-chain acylcarnitines are detected with high intensity in the cortex but show variable localisation to the OS and the medulla. While the localisation of C3 and isohydroxybutyrylcarnitine (C4- OH) to the medulla is minor (Fig. 4c, e), acetylcarnitine (C2), isobutyrylcarnitine (C4), and isovaleryl carnitine (C5) show some localisation in medullary regions (Fig. 4b, d, f). In contrast, all elevated long-chain acylcarnitines show more dispersed localisation across the diabetic kidney tissue section (Fig. 4g–i and ESM Fig. S4) and an almost complementary distribution to the altered NEFA, MG, and DG species that primarily localise to the OS of the kidney (Fig. 3b–e). Fig 2 Representative optical Fig. 1 Anatomical regions of kidney tissue have distinct molecular composition. (a) Optical image of kidney section with an overlay highlighting anatomical regions. (b) Ion image of [methylhistidine+H]+ (m/z 170.0923). (c) Ion image of [propionylcarnitine+H]+(m/z 218.1386). (d) Ion image of [sphingomyelin 34:1+K]+ (m/z 741.5307). (e) Ion image of [betaine+Na]+ (m/z 140.0681. (f) Ion image of [sorbitol+Na]+ (m/z 205.0681) increased NEFA and MG species were 18:1, 18:2, and 18:3 (number of carbons:number of double bonds). In addition, the major constituents of the elevated DG species 32:0, 34:1, 34:2, and 36:4 also contained acyl chains 18:1, 18:2, and 18:3 (Fig. 3, ESM Table S7). These results suggest an impor- tant role of the metabolism for these moieties in DKD. However, while their abundances are significantly different (p value < 0.05) in the cortex and OS, their spatial distribu- tions are conserved independent of diabetic state (ESM Fig. S2). In both control and STZ-treated tissue, the NEFA, MG, and DG species co-localise in the junction between the cortex and the medulla (OS, IS, and IM) with minor localisation to the IS and IM regions, suggesting the importance of this re- gional interface in DKD (Fig. 3b–d, ESM Fig. S3). Fig. 2 Representative optical images of H&E-stained control and diabetic kidney sections, 2 weeks post STZ treatment, at four different magnifications Results Kidney tissue contains distinct anatomical regions responsible for activities such as filtration of blood and formation of urine. Figure 1a highlights the four major anatomical regions in a transverse kidney section: cortex, outer strip of the outer H.-M. Bergman et al. 2812 Fig. 1 Anatomical regions of kidney tissue have distinct molecular composition. (a) Optical image of kidney section with an overlay highlighting anatomical regions. (b) Ion image of [methylhistidine+H]+ (m/z 170.0923). (c) Ion image of [propionylcarnitine+H]+(m/z 218.1386). (d) Ion image of [sphingomyelin 34:1+K]+ (m/z 741.5307). (e) Ion image of [betaine+Na]+ (m/z 140.0681. (f) Ion image of [sorbitol+Na]+ (m/z 205.0681) Specific lipid species accumulate in diabetic kidney tissue The altered lipid species NEFA, MG, and DG in the cortex and OS regions of diabetic kidney all contained 18 carbon unsaturated acyl chains. Specifically, the acyl chains of increased NEFA and MG species were 18:1, 18:2, and 18:3 (number of carbons:number of double bonds). In addition, the major constituents of the elevated DG species 32:0, 34:1, 34:2, and 36:4 also contained acyl chains 18:1, 18:2, and 18:3 (Fig. 3, ESM Table S7). These results suggest an impor- tant role of the metabolism for these moieties in DKD. However, while their abundances are significantly different (p value < 0.05) in the cortex and OS, their spatial distribu- tions are conserved independent of diabetic state (ESM Fig. S2). In both control and STZ-treated tissue, the NEFA, MG, and DG species co-localise in the junction between the cortex and the medulla (OS, IS, and IM) with minor localisation to the IS and IM regions, suggesting the importance of this re- gional interface in DKD (Fig. 3b–d, ESM Fig. S3). Acylcarnitines accumulate in diabetic kidney tissue Both short- and long-chain acylcarnitine species were found significantly increased (p value < 0.01) in the cortex of diabetic kidney tissue compared to control. The large differences in sig- nal intensities of the short-chain and long-chain acylcarnitines are displayed on a logarithmic scale in Fig. 4a, clearly showing the accumulation of acylcarnitines in the diabetic kidney. While the data in Fig. 4a is specific to the cortex and OS, the ion images in Fig. 4b–i show the distribution in the entire tissue section. The ion images reveal that the distribution of short-chain acylcarnitines is highly dependent on the acyl chain of the re- spective species (Fig. 4b–f). Specific lipid species accumulate in diabetic kidney tissue The altered lipid species NEFA, MG, and DG in the cortex and OS regions of diabetic kidney all contained 18 carbon unsaturated acyl chains. Specifically, the acyl chains of Fig. 2 Representative optical images of H&E-stained control and diabetic kidney sections, 2 weeks post STZ treatment, at four different magnifications g. 2 Representative optical mages of H&E-stained control d diabetic kidney sections, weeks post STZ treatment, at ur different magnifications Table 1 Molecules detected using nano-DESI MSI analysis that have significantly altered signal intensity in rat kidney sec- tions from STZ-treated rats (2 weeks post treatment) com- pared to control. C2, acetylcarnitine; C3, propionylcarnitine; C4, isobutyrylcarnitine; C4-OH, hydroxyisobutyrylcarnitine; C5, 2-methylbuturylcarnitine/ isovalerylcarnitine; C16:0, palmitoylcarnitine; C18:2, linoleylcarnitine; C18:0, stearoylcarnitine. Mass error < 5 ppm Increased signal in diabetes Decreased signal in diabetes Chemical formula Compound Chemical formula Compound C4H6O3 b) 2-Ketobutyric acid* C3H7NO3 Serine b) C6H8O4 b) 3-Hexenedioic acid* C2H7NO2S Hypotaurine b) C6H10O5 b) 3-Hydroxymethylglutaric acid* C3H7N3O2 Guanidinoacetic acid b) C6H12O6 b) Glucose* C5H9NO3 b) Hydroxyproline* C9H17NO4 C2 b) C7H7NO2 b) Anthranilic acid* C10H19NO4 C3 b) C7H13NO2 Proline betaine b) C11H21NO4 C4 b) C5H11N3O2 4-Guanidinobutanoic acid b) C12H23NO4 C5 b) C7H15NO2 b) Dehydroxycarnitine* C11H21NO5 C4-OH c) C5H9NO4 Glutamate a) C18H30O2 NEFA 18:3 b) d) C6H9N3O2 Histidine a) C18H32O2 NEFA 18:2 b) d) C7H11N3O2 Methylhistidine b) C18H34O2 NEFA 18:1 b) d) C9H11NO3 Tyrosine a) C19H34O2 Methyl linoleate b) d) C11H12N2O2 Tryptophan b) C10H14N5O7P AMP b) C38H76NO8P Phosphatidylcholine 30:0 b) C21H36O4 MG 18:3 b) d) C21H38O4 MG 18:2 b) d) C21H40O4 MG18:1 b) d) C23H45NO4 C16:0 b) C25H45NO4 C18:2 b) C25H49NO4 C18:0 b) C35H68O5 DG 32:0 d) C37H68O5 DG 34:2 b) d) C37H70O5 DG 34:1 b) d) C39H68O5 DG 36:4 b) d) *The most likely isomer a) Level 1 identification through tandem mass spectrometry [22] b) Level 2 identification through tandem mass spectrometry [22] c) Level 3 identification through tandem mass spectrometry [22] d) Presence or absence of double bond(s) confirmed by Ag+ -adduct formation Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging 2813 Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging 2813 y a) Level 1 identification through tandem mass spectrometry [22] b) Level 2 identification through tandem mass spectrometry [22] c) Level 3 identification through tandem mass spectrometry [22] d) Presence or absence of double bond(s) confirmed by Ag+ -adduct formation Discussion (a) Heat map representing the mean signal intensities of three non-esterified fatty acids (NEFA), three monoacylglycerol (MG), and four diacylglycerol (DG) species. The rows represent diabetic rats (D1–D3) and control rats (C1–C3). The colour gradient spans from highest to lowest mean relative intensity across the 6 kidneys for each molecular species. Ion images of (b) [NEFA 18:2+K]+ (m/z 319.2035), (c) [MG 18:2+K]+ (m/z 393.2402), (d) [DG 34:2+K]+ (m/z 631.4699), and (e) [DG 36:4+K]+ (m/z 655.4699) Fig. 3 Untargeted nano-DESI MSI reveals increased NEFA, MG, and DG species in the cortex of diabetic kidney (p value < 0.05). (a) Heat map representing the mean signal intensities of three non-esterified fatty acids (NEFA), three monoacylglycerol (MG), and four diacylglycerol (DG) species. The rows represent diabetic rats (D1–D3) and control rats (C1–C3). The colour gradient spans from highest to lowest mean relative intensity across the 6 kidneys for each molecular species. Ion images of (b) [NEFA 18:2+K]+ (m/z 319.2035), (c) [MG 18:2+K]+ (m/z 393.2402), (d) [DG 34:2+K]+ (m/z 631.4699), and (e) [DG 36:4+K]+ (m/z 655.4699) Fig. 4 Accumulation of acylcarnitine species in the combined cortex and OS of diabetic kidney. (a) Graph displaying mean signal intensities of 8 acylcarnitine species in diabetic and control kidney on a logarithmic scale. Grey, diabetes; white, control. n = 3 each for diabetes and control. Error bars represent + 1 standard deviation. Ion images display (b) [C2+ H]+ (m/z 204.1228), (c) [C3+H]+ (m/z 218.1386), (d) [C4+H]+ (m/z 232.1540), (e) [C4-OH+H]+(m/z 248.1490), (f) [C5+H]+ (m/z 246.1697), (g) [C16:0+H]+ (m/z 400.3422), (h) [C18:2+H]+ (m/z 424.3422), and (i) [C18:0+H]+ (m/z 428.3735); p < 0.01 except for C3 where p < 0.05 Fig. 3 Untargeted nano-DESI MSI reveals increased NEFA, MG, and DG species in the cortex of diabetic kidney (p value < 0.05). (a) Heat map representing the mean signal intensities of three non-esterified fatty acids (NEFA), three monoacylglycerol (MG), and four diacylglycerol (DG) species. The rows represent diabetic rats (D1–D3) and control rats (C1–C3). The colour gradient spans from highest to lowest mean relative intensity across the 6 kidneys for each molecular species. Ion images of (b) [NEFA 18:2+K]+ (m/z 319.2035), (c) [MG 18:2+K]+ (m/z 393.2402), (d) [DG 34:2+K]+ (m/z 631.4699), and (e) [DG 36:4+K]+ (m/z 655.4699) note that the 18:2 and 18:3 acyl chains, which are the sole precursors of other NEFAs [32], are essential fatty acids that cannot be synthesised by mammals [33]. Discussion Thus, the accumulation of 18:1, 18:2, and 18:3 NEFA and DG species likely plays a key role in the progres- sion of DKD. The kidney has a high-energy demand for sustaining active transport, of small and large chemical substances throughout the nephron, which is further increased to maintain electrolyte and volume homeostasis following the elevated glomerular filtration rate in early diabetes. A common source for energy Fig. 4 Accumulation of acylcarnitine species in the combined cortex and OS of diabetic kidney. (a) Graph displaying mean signal intensities of 8 acylcarnitine species in diabetic and control kidney on a logarithmic scale. Grey, diabetes; white, control. n = 3 each for diabetes and control. Error bars represent + 1 standard deviation. Ion images display (b) [C2+ H]+ (m/z 204.1228), (c) [C3+H]+ (m/z 218.1386), (d) [C4+H]+ (m/z 232.1540), (e) [C4-OH+H]+(m/z 248.1490), (f) [C5+H]+ (m/z 246.1697), (g) [C16:0+H]+ (m/z 400.3422), (h) [C18:2+H]+ (m/z 424.3422), and (i) [C18:0+H]+ (m/z 428.3735); p < 0.01 except for C3 where p < 0.05 Fig. 3 Untargeted nano-DESI MSI reveals increased NEFA, MG, and DG species in the cortex of diabetic kidney (p value < 0.05). (a) Heat map representing the mean signal intensities of three non-esterified fatty acids (NEFA), three monoacylglycerol (MG), and four diacylglycerol (DG) species. The rows represent diabetic rats (D1–D3) and control rats (C1–C3). The colour gradient spans from highest to lowest mean relative intensity across the 6 kidneys for each molecular species. Ion images of (b) [NEFA 18:2+K]+ (m/z 319.2035), (c) [MG 18:2+K]+ (m/z 393.2402), (d) [DG 34:2+K]+ (m/z 631.4699), and (e) [DG 36:4+K]+ (m/z 655.4699) 2814 H.-M. Bergman et al. Fig. 3 Untargeted nano-DESI MSI reveals increased NEFA, MG, and DG species in the cortex of diabetic kidney (p value < 0.05). (a) Heat map representing the mean signal intensities of three non-esterified fatty acids (NEFA), three monoacylglycerol (MG), and four diacylglycerol (DG) species. The rows represent diabetic rats (D1–D3) and control rats (C1–C3). The colour gradient spans from highest to lowest mean relative intensity across the 6 kidneys for each molecular species. Ion images of (b) [NEFA 18:2+K]+ (m/z 319.2035), (c) [MG 18:2+K]+ (m/z 393.2402), (d) [DG 34:2+K]+ (m/z 631.4699), and (e) [DG 36:4+K]+ (m/z 655.4699) g Fig. 3 Untargeted nano-DESI MSI reveals increased NEFA, MG, and DG species in the cortex of diabetic kidney (p value < 0.05). Discussion abundances to cellular regions can be correlated to known biological activities. The localisation of individual metabolites to anatomical re- gions of the kidney suggests their respective importance for cellular and biological function in these regions. For example, the wide distribution of sphingomyelin in the tissue displays its general importance as a membrane lipid in the plasma membrane. Other localisations are, however, less known. For example, the distribution of methylhistidine and C3 to either side of the cortex-OS border is not known, but the minimal overlap of the respective ion images indicates that metabolic activities within these regions are different. Further, the medullary region is known to have a high salt concentration, and the high abundance of the osmoprotective compounds betaine and sorbitol evidences their importance in protecting cells from osmotic stress in this region [23]. Thus, the numerous metabolites detected by MSI and their The altered metabolite abundances in the kidney 2 weeks after STZ treatment show that despite no obvious histopatho- logical changes, there are significant metabolic alterations within the tissue at this stage. An important region for DKD is the cortex as this is the site for blood filtration through the glomeruli. Further, a key part of disease progression is fibrosis of the tubuli, which occurs in the cortex. Here, we find that several lipid species accumulate to the cortex, which is in accordance with previous reports of renal lipid accumulation in response to hyperglycaemia [24–27]. However, the acyl chain composition of accumulating lipids and the pathophys- iological role of the acyl chains 18:2 and 18:3 are not fully known. Overall, it is a common view that the mechanisms and possible toxic effects of renal lipid accumulation have been under-investigated [27–31]. In this context, it is of interest to 2814 H.-M. Bergman et al. note that the 18:2 and 18:3 acyl chains, which are the sole precursors of other NEFAs [32], are essential fatty acids that cannot be synthesised by mammals [33]. Further, dietary stud- ies report that an increased intake of NEFA 18:2 and 18:3 is negatively associated with DKD, which supports their impor- tance in disease development [34, 35]. One possible implica- tion of the importance of NEFA 18:1, 18:2, and 18:3 and DG 18:2/18:2 is that they promote activation of protein kinase C (PKC) [36, 37]. PKC is highly involved in a number of hyperglycaemia-induced cellular responses and in DKD [38–40]. Discussion Another major contributor of C2 is BCAA catabolism, and many of the short-chain acylcarnitines (C4, C4-OH, and C5) found in- creased in the kidney 2 weeks after disease onset are carni- tine esters of key intermediates in the catabolism of leucine, isoleucine, and valine [43]. The BCAA catabolism product, C3, is converted to succinyl-CoA, another substrate of the citric acid cycle. Our results of apparent increases in both intermediates and products from catabolism of all three BCAAs in conjunction with decreases in several amino acids (AA) suggest an overall increase in AA metabolism. It is worth noting that the accumulation of C2 in diabetic kidney could be a result of either increased formation or reduced utilisation in the tissue. Regardless, the combined findings show a significant increase in substrates for the citric acid cycle, a requirement to meet the energy demands of hyperglycaemia. is the citric acid cycle; however, the kidney is mainly fuelled by the more ATP-efficient long-chain FAO in the mitochon- drial matrix [41]. The rate-controlling step of FAO is esterifi- cation of fatty acids with carnitine, forming long-chain acylcarnitines [42]. This transformation is required for active transport of the fatty acids across the inner mitochondrial membrane. In addition to the elevated abundance of long- chain acylcarnitines, accumulation of the key component C2 further indicates a dysregulated FAO already at this early stage of diabetes. The final product of FAO is acetyl-CoA, which can be transesterified to C2 and either transported out of the kidney or used as a substrate in the citric acid cycle. C2 is the universal end product of mitochondrial metabolism; its in- creased abundance is therefore indicative of a perturbance of both the FAO and the citric acid cycle. Another major contributor of C2 is BCAA catabolism, and many of the short-chain acylcarnitines (C4, C4-OH, and C5) found in- creased in the kidney 2 weeks after disease onset are carni- tine esters of key intermediates in the catabolism of leucine, isoleucine, and valine [43]. The BCAA catabolism product, C3, is converted to succinyl-CoA, another substrate of the citric acid cycle. Our results of apparent increases in both intermediates and products from catabolism of all three BCAAs in conjunction with decreases in several amino acids (AA) suggest an overall increase in AA metabolism. References 1. Packham DK, Alves TP, Dwyer JP, Atkins R, de Zeeuw D, Cooper M, et al. Relative incidence of ESRD versus cardiovascular mortal- ity in proteinuric type 2 diabetes and nephropathy: results from the DIAMETRIC (Diabetes Mellitus Treatment for Renal Insufficiency Consortium) database. Am J Kidney Dis. 2012;59(1):75–83. https://doi.org/10.1053/j.ajkd.2011.09.017. 2. Nishikawa T, Brownlee M, Araki E. Mitochondrial reactive oxygen species in the pathogenesis of early diabetic nephropathy. J Diabetes Investig. 2015;6(2):137–9. https://doi.org/10.1111/jdi.12258. 2. Nishikawa T, Brownlee M, Araki E. Mitochondrial reactive oxygen species in the pathogenesis of early diabetic nephropathy. J Diabetes Investig. 2015;6(2):137–9. https://doi.org/10.1111/jdi.12258. 3. Sun Y-M, Su Y, Li J, Wang L-F. Recent advances in understanding the biochemical and molecular mechanism of diabetic nephropathy. Biochem Biophys Res Commun. 2013;433(4):359–61. https://doi. org/10.1016/j.bbrc.2013.02.120. 3. Sun Y-M, Su Y, Li J, Wang L-F. Recent advances in understanding the biochemical and molecular mechanism of diabetic nephropathy. Biochem Biophys Res Commun. 2013;433(4):359–61. https://doi. org/10.1016/j.bbrc.2013.02.120. Compliance with ethical standards All animal procedures were approved by the local animal ethics commit- tee in Uppsala and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Conflict of interest The authors declare that they have no conflict of interest. Conflict of interest The authors declare that they have no conflict of interest. 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. Discussion It is worth noting that the accumulation of C2 in diabetic kidney could be a result of either increased formation or reduced utilisation in the tissue. Regardless, the combined findings show a significant increase in substrates for the citric acid cycle, a requirement to meet the energy demands of hyperglycaemia. Acknowledgements A. Fasching and L. Sagulin, Department of Medical Cell Biology, Uppsala University, Sweden, are acknowledged for techni- cal assistance. Funding information This work was funded by the Swedish Foundation for Strategic Research, the Swedish Research Council, the Swedish Diabetes Foundation, AstraZeneca Gothenburg, and Jeansson Foundation. Discussion Further, dietary stud- ies report that an increased intake of NEFA 18:2 and 18:3 is negatively associated with DKD, which supports their impor- tance in disease development [34, 35]. One possible implica- tion of the importance of NEFA 18:1, 18:2, and 18:3 and DG 18:2/18:2 is that they promote activation of protein kinase C (PKC) [36, 37]. PKC is highly involved in a number of hyperglycaemia-induced cellular responses and in DKD [38–40]. Thus, the accumulation of 18:1, 18:2, and 18:3 NEFA and DG species likely plays a key role in the progres- sion of DKD. Fig. 4 Accumulation of acylcarnitine species in the combined cortex and OS of diabetic kidney. (a) Graph displaying mean signal intensities of 8 acylcarnitine species in diabetic and control kidney on a logarithmic scale. Grey, diabetes; white, control. n = 3 each for diabetes and control. Error bars represent + 1 standard deviation. Ion images display (b) [C2+ H]+ (m/z 204.1228), (c) [C3+H]+ (m/z 218.1386), (d) [C4+H]+ (m/z 232.1540), (e) [C4-OH+H]+(m/z 248.1490), (f) [C5+H]+ (m/z 246.1697), (g) [C16:0+H]+ (m/z 400.3422), (h) [C18:2+H]+ (m/z 424.3422), and (i) [C18:0+H]+ (m/z 428.3735); p < 0.01 except for C3 where p < 0.05 The kidney has a high-energy demand for sustaining active transport, of small and large chemical substances throughout the nephron, which is further increased to maintain electrolyte and volume homeostasis following the elevated glomerular filtration rate in early diabetes. A common source for energy Metabolite aberrations in early diabetes detected in rat kidney using mass spectrometry imaging 2815 is the citric acid cycle; however, the kidney is mainly fuelled by the more ATP-efficient long-chain FAO in the mitochon- drial matrix [41]. The rate-controlling step of FAO is esterifi- cation of fatty acids with carnitine, forming long-chain acylcarnitines [42]. This transformation is required for active transport of the fatty acids across the inner mitochondrial membrane. In addition to the elevated abundance of long- chain acylcarnitines, accumulation of the key component C2 further indicates a dysregulated FAO already at this early stage of diabetes. The final product of FAO is acetyl-CoA, which can be transesterified to C2 and either transported out of the kidney or used as a substrate in the citric acid cycle. C2 is the universal end product of mitochondrial metabolism; its in- creased abundance is therefore indicative of a perturbance of both the FAO and the citric acid cycle. Conclusion Reducing VEGF-B signaling ameliorates renal lipotoxicity and protects against diabetic kidney disease. Cell Metab. 2017;25(3):713–26. https://doi.org/10.1016/j.cmet.2017. 01.004. 12. Roach PJ, Laskin J, Laskin A. Nanospray desorption electrospray ionization: an ambient method for liquid-extraction surface sam- pling in mass spectrometry. Analyst. 2010;135(9):2233–6. https:// doi.org/10.1039/c0an00312c. 28. Bobulescu IA. Renal lipid metabolism and lipotoxicity. Curr Opin Nephrol Hypertens. 2010;19(4):393–402. https://doi.org/10.1097/ MNH.0b013e32833aa4ac. 13. Wu T, Qiao S, Shi C, Wang S, Ji G. Metabolomics window into diabetic complications. J Diabetes Investig. 2018;9(2):244–55. https://doi.org/10.1111/jdi.12723. 29. Kang HM, Ahn SH, Choi P, Ko Y-A, Han SH, Chinga F, et al. Defective fatty acid oxidation in renal tubular epithelial cells has a key role in kidney fibrosis development. Nat Med. 2014;21:37. https://doi.org/10.1038/nm.3762. 14. Gonzalez Suarez ML, Thomas DB, Barisoni L, Fornoni A. Diabetic nephropathy: is it time yet for routine kidney biopsy? World J Diabetes. 2013;4(6):245–55. https://doi.org/10.4239/wjd.v4.i6. 245. 30. Scerbo D, Son N-H, Sirwi A, Zeng L, Sas KM, Cifarelli V, et al. Kidney triglyceride accumulation in the fasted mouse is dependent upon serum free fatty acids. J Lipid Res. 2017;58(6):1132–42. https://doi.org/10.1194/jlr.M074427. 15. Wei M, Ong L, Smith MT, Ross FB, Schmid K, Hoey AJ, et al. The streptozotocin-diabetic rat as a model of the chronic complications of human diabetes. Heart Lung Circ. 2003;12(1):44–50. https://doi. org/10.1046/j.1444-2892.2003.00160.x. 31. Kimmelstiel P, Wilson C. Intercapillary lesions in the glomeruli of the kidney. Am J Pathol. 1936;12(1):83–98.7. 16. Franzén S, Pihl L, Fasching A, Palm F. Intrarenal activation of endothelin type B receptors improves kidney oxygenation in type 1 diabetic rats. Am J Physiol Renal Physiol. 2018;314(3):F439– F44. https://doi.org/10.1152/ajprenal.00498.2017. 32. Cunnane SC, Manku MS, Horrobin DF. Abnormal essential fatty acid composition of tissue lipids in genetically diabetic mice is partially corrected by dietary linoleic and γ-linolenic acids. Br J Nutr. 2007;53(3):449–58. https://doi.org/10.1079/BJN19850054. 33. Ramsammy LS, Haynes B, Josepovitz C, Kaloyanides GJ. Mechanism of decreased arachidonic acid in the renal cortex of rats with diabetes mellitus. Lipids. 1993;28(5):433–9. 17. Nordquist L, Brown R, Fasching A, Persson P, Palm F. Proinsulin C-peptide reduces diabetes-induced glomerular hyperfiltration via efferent arteriole dilation and inhibition of tubular sodium reabsorp- tion. Am J Physiol Renal Physiol. 2009;297(5):F1265–F72. https:// doi.org/10.1152/ajprenal.00228.2009. 34. Dos Santos ALT, Duarte CK, Santos M, Zoldan M, Almeida JC, Gross JL, et al. Low linolenic and linoleic acid consumption are associated with chronic kidney disease in patients with type 2 dia- betes. PLoS One. 2018;13(8):e0195249. https://doi.org/10.1371/ journal.pone.0195249. 18. Conclusion 4. Putri SP, Yamamoto S, Tsugawa H, Fukusaki E. Current metabolo- mics: technological advances. J Biosci Bioeng. 2013;116(1):9–16. https://doi.org/10.1016/j.jbiosc.2013.01.004. 4. Putri SP, Yamamoto S, Tsugawa H, Fukusaki E. Current metabolo- mics: technological advances. J Biosci Bioeng. 2013;116(1):9–16. https://doi.org/10.1016/j.jbiosc.2013.01.004. In this study, we provide new insights into local alterations of renal metabolism in early-stage insulinopenic diabetes. The results show significant lipid and acylcarnitine accu- mulation, indicating a dysregulation of the citric acid cy- cle, FAO, and BCAA metabolism in the cortex of diabetic kidney already 2 weeks after disease onset and before dis- tinguishable histopathological changes. These results indi- cate that alterations in mitochondrial energy metabolism are associated with tubular and glomerular damage as a result of hyperglycaemia and insulin deficiency. Having demonstrated the suitability of nano-DESI MSI for detect- ing localised metabolic changes in kidney tissue, future studies should monitor the identified metabolite classes through the progression of diabetes and kidney disease. Comparative studies on different models of diabetes will also provide a better understanding of the origins and con- sequences of these metabolic alterations. 5. Gowda GAN, Djukovic D. Overview of mass spectrometry-based metabolomics: opportunities and challenges. Methods Mol Biol. 2014;1198:3–12. https://doi.org/10.1007/978-1-4939-1258-2_1. 6. Zhang Y, Zhang S, Wang G. Metabolomic biomarkers in diabetic kidney diseases—a systematic review. J Diabetes Complicat. 2015;29(8):1345–51. https://doi.org/10.1016/j.jdiacomp.2015.06.016. 7. McDonnell LA, Heeren RMA. Imaging mass spectrometry. Mass Spectrom Rev. 2007;26(4):606–43. https://doi.org/10.1002/mas.20124. 8. Wu C, Dill AL, Eberlin LS, Cooks RG, Ifa DR. Mass spectrometry imaging under ambient conditions. Mass Spectrom Rev. 2013;32(3):218–43. https://doi.org/10.1002/mas.21360. 9. Grove KJ, Voziyan PA, Spraggins JM, Wang S, Paueksakon P, Harris RC, et al. Diabetic nephropathy induces alterations in the glomerular and tubule lipid profiles. J Lipid Res. 2014;55(7): 1375–85. https://doi.org/10.1194/jlr.M049189. 10. Miyamoto S, Hsu C-C, Hamm G, Darshi M, Diamond-Stanic M, Declèves A-E, et al. Mass spectrometry imaging reveals elevated glomerular ATP/AMP in diabetes/obesity and identifies 2816 H.-M. Bergman et al. sphingomyelin as a possible mediator. EBioMedicine. 2016;7:121– 34. https://doi.org/10.1016/j.ebiom.2016.03.033. 26. Herman-Edelstein M, Scherzer P, Tobar A, Levi M, Gafter U. Altered renal lipid metabolism and renal lipid accumulation in hu- man diabetic nephropathy. J Lipid Res. 2014;55(3):561–72. https:// doi.org/10.1194/jlr.P040501. 11. Lanekoff I, Heath BS, Liyu A, Thomas M, Carson JP, Laskin J. Automated platform for high-resolution tissue imaging using nanospray desorption electrospray ionization mass spectrometry. Anal Chem. 2012;84(19):8351–6. https://doi.org/10.1021/ ac301909a. 27. Falkevall A, Mehlem A, Palombo I, Heller Sahlgren B, Ebarasi L, He L, et al. Conclusion Lanekoff I, Burnum-Johnson K, Thomas M, Cha J, Dey S, Yang P, et al. Three-dimensional imaging of lipids and metabolites in tissues by nanospray desorption electrospray ionization mass spectrometry. Anal Bioanal Chem. 2015;407(8):2063–71. https://doi.org/10. 1007/s00216-014-8174-0. 35. Barcelli UO, Weiss M, Beach D, Motz A, Thompson B. High linoleic acid diets ameliorate diabetic nephropathy in rats. Am J Kidney Dis. 1990;16(3):244–51. 19. Jaitly N, Mayampurath A, Littlefield K, Adkins JN, Anderson GA, Smith RD. Decon2LS: an open-source software package for auto- mated processing and visualization of high resolution mass spec- trometry data. BMC Bioinformatics. 2009;10:87. https://doi.org/10. 1186/1471-2105-10-87. 36. Shinomura T, Asaoka Y, Oka M, Yoshida K, Nishizuka Y. Synergistic action of diacylglycerol and unsaturated fatty acid for protein kinase C activation: its possible implications. PNAS. 1991;88(12):5149–53. https://doi.org/10.1073/pnas.88.12.5149. 20. Bergman HM, Lundin E, Andersson M, Lanekoff I. Quantitative mass spectrometry imaging of small-molecule neurotransmitters in rat brain tissue sections using nanospray desorption electrospray ionization. Analyst. 2016;141(12):3686–95. https://doi.org/10. 1039/c5an02620b. 37. Kishimoto A, Takai Y, Mori T, Kikkawa U, Nishizuka Y. Activation of calcium and phospholipid-dependent protein kinase by diacyl- glycerol, its possible relation to phosphatidylinositol turnover. J Biol Chem. 1980;255(6):2273–6. 38. Brownlee M. The pathobiology of diabetic complications a unify- ing mechanism. Diabetes. 2005;54(6):1615–25. https://doi.org/10. 2337/diabetes.54.6.1615. 21. Duncan KD, Lanekoff I. Spatially resolved surface sampling capil- lary electrophoresis mass spectrometry. Manuscript in revision. 39. Geraldes P, King GL. Activation of protein kinase C isoforms & its impact on diabetic complications. Circ Res. 2010;106(8):1319–31. https://doi.org/10.1161/circresaha.110.217117. 22. Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA, et al. Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomics. 2007;3(3):211–21. https://doi.org/10.1007/s11306-007-0082-2. p g 40. Noh H, King GL. The role of protein kinase C activation in diabetic nephropathy. Kidney Int. 2007;72:S49–53. https://doi.org/10.1038/ sj.ki.5002386. 23. Schmolke M, Bornemann A, Guder WG. Site-specific regulation of organic osmolytes along the rat nephron. Am J Physiol Renal Physiol. 1996;271(3):F645–F52. 41. Nieth H, Schollmeyer P. Substrate-utilization of the human kidney. Nature. 1966;209:1244. https://doi.org/10.1038/2091244a0. 24. Proctor G, Jiang T, Iwahashi M, Wang Z, Li J, Levi M. Regulation of renal fatty acid and cholesterol metabolism, inflammation, and fibrosis in Akita and OVE26 mice with type 1 diabetes. Diabetes. 2006;55(9):2502. 42. Schooneman MG, Vaz FM, Houten SM, Soeters MR. Acylcarnitines: reflecting or inflicting insulin resistance? Diabetes. 2013;62(1):1–8. https://doi.org/10.2337/db12-0466. p g 43. Manoli I, Venditti CP. Disorders of branched chain amino acid metabolism. Conclusion Transl Sci Rare Dis. 2016;1(2):91–110. https://doi. org/10.3233/trd-160009. 25. Sun L, Halaihel N, Zhang W, Rogers T, Levi M. Role of sterol regulatory element-binding protein 1 in regulation of renal lipid metabolism and glomerulosclerosis in diabetes mellitus. J Biol Chem. 2002;277(21):18919–27. https://doi.org/10.1074/jbc. M110650200. 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.
https://openalex.org/W3003365453	https://www.nature.com/articles/s41598-020-58417-5.pdf	English	null	Understanding hyaluronic acid induced variation of water structure by near-infrared spectroscopy	Scientific reports	2,020	cc-by	5,940	Qin Dong1, Xueping Guo2, Lian Li1, Chen Yu1, Lei Nie1, Weilu Tian1, Hui Zhang1, Siling Huang2 & Hengchang Zang1,3,4* In order to understand the hydration effect of hyaluronic acid (HA) in aqueous solution, near-infrared (NIR) spectroscopy was used to investigate the HA aqueous solutions at different concentrations and temperature. As HA concentration was raised, there was a nonlinear change in absorption value in the first overtone region of OH, indicating the changes of hydration water. A reconstructed spectrum based on principal component analysis (PCA) was established and analyzed with the concept of aquaphotomics. The results showed that HA acted as a structure maker to make water molecules arranged in order. Water species with two hydrogen bonds (S2) and three hydrogen bonds (S3) showed the decrease at low concentration range of 0–40 mg/mL, but increased at higher concentration, indicating the difference in water species at different HA concentration. Meanwhile, HA had the ability to improve the thermal stability of water structure, suggesting a potential bio-protective function. This study provides a unique perspective on the molecular interactions between HA and water molecules, which is helpful for understanding the role of HA in life process and may serve as the basis for HA applications. Hyaluronic acid (HA) is a natural polysaccharide and one of the main component of the extracellular matrix1. It is widely distributed in various tissues and organs of animals as well as the capsules of some bacteria, and plays a sig- nificant role in life process1. Due to its high water retaining capacity, biocompatibility and non-immunogenicity, HA is an attractive material for various cosmetic, food and medical applications2–4.l Hyaluronic acid (HA) is a natural polysaccharide and one of the main component of the extracellular matrix1. It is widely distributed in various tissues and organs of animals as well as the capsules of some bacteria, and plays a sig- nificant role in life process1. Due to its high water retaining capacity, biocompatibility and non-immunogenicity, HA is an attractive material for various cosmetic, food and medical applications2–4.l pp It is known that the structure and morphology of polysaccharides are strongly influenced by water mole- cules via controlling their conformations, various forms of aggregation, thermal properties and kinetics5,6. In this respect, the knowledge of the interaction between water and polysaccharide and the hydrated structure is of high importance for understanding its performance in the application. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 OPEN Qin Dong1, Xueping Guo2, Lian Li1, Chen Yu1, Lei Nie1, Weilu Tian1, Hui Zhang1, Siling Huang2 & Hengchang Zang1,3,4* Qin Dong1, Xueping Guo2, Lian Li1, Chen Yu1, Lei Nie1, Weilu Tian1, Hui Zhang1, Siling Huan & Hengchang Zang1,3,4* Qin Dong1, Xueping Guo2, Lian Li1, Chen Yu1, Lei Nie1, Weilu Tian1, Hui Zhang1, Siling Huang2 & Hengchang Zang1,3,4* Studies on HA hydration have been reported by using various experimental techniques, including nuclear magnetic resonance (NMR)7, viscometry8, ultrasonic and densitometry analyses9, thermal analysis10,11, spectroscopic techniques12–14, and by means of computer sim- ulations15. A complex relationship between HA and water molecules was partly discovered that the macroscopic properties of HA are significantly dependent on its degree of hydration and the function of HA relies on the hydration capacity4,16–18. y p y Analysis of the hydration water in aqueous solution is challenging, and the structural changes in water induced by HA are yet to be fully understood. Near-infrared (NIR) spectroscopy, as a powerful analytical method, has unique advantages in studies of molecular structure and interaction19–22. It mainly reflects the overtones and combination modes of functional groups containing hydrogen atoms, such as CH, OH and NH, which play a significant role in chemical bonding and other chemical phenomena such as hydrogen transfer, hydrogen bond- ing23. The vibrations of these groups greatly reflect even subtle changes in molecular interactions. Therefore, NIR spectroscopy has become a powerful spectroscopic technique in structural biology.h The purpose of this study is to continue and extend the research on the hydration behaviors of HA, but par- ticular attention is paid on investigating the water structure changes induced by HA. NIR spectra of water and aqueous HA solutions with a range of concentration were measured at different temperatures. Principal com- ponent analysis (PCA) was conducted on the spectra data to extract the information of different water species. 1School of Pharmaceutical Sciences, Shandong University, Wenhuaxi Road 44, Jinan, 250012, China. 2Bloomage Biotechnology Corporation Limited, Tianchen Street 678, Jinan, 250012, China. 3Key Laboratory of Chemical Biology (Ministry of Education), Shandong University, Wenhuaxi Road 44, Jinan, 250012, China. 4National Glycoengineering Research Center, Shandong University, Binhai Road 72, Qingdao, 266200, China. email: zanghcw@126.com Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ Figure 1. NIR spectra of water, HA aqueous solution (100 mg/mL) from 25 °C to 55 °C (a) and the second derivatives of the selected spectra in Fig. 1 (b). Figure 1. NIR spectra of water, HA aqueous solution (100 mg/mL) from 25 °C to 55 °C (a) and the second derivatives of the selected spectra in Fig. 1 (b). Figure 2. Average absorption of HA solution at 25 °C in the range of 7930–5930 cm−1 at different concentrations. Figure 2. Qin Dong1, Xueping Guo2, Lian Li1, Chen Yu1, Lei Nie1, Weilu Tian1, Hui Zhang1, Siling Huang2 & Hengchang Zang1,3,4 Average absorption of HA solution at 25 °C in the range of 7930–5930 cm−1 at different concentrations. Aquaphotomics was employed to analyze the effect of HA on water structure. The results showed that HA acted as a structure maker to make water structure ordered in aqueous solutions and different water network was promoted at different HA concentrations. Meanwhile, HA had the function of improving the thermal stability of water structures. The results provide new information of HA hydration and possibly shed more light on the function of water in bio-systems. Results NIR NIR spectra of water and HA solutions. Figure 1(a) showed NIR spectra of water, HA aqueous solution (100 mg/mL) from 25 °C to 55 °C. It could be seen the addition of HA caused a decrease in the intensity of the absorption around 6900 cm−1, which is an overlap of the absorption mainly due to the combinations of OH sym- metric and antisymmetric stretching modes of various water species24. Figure 1(b) displayed the second deriv- atives of the spectra in Fig. 1(a), showing peaks at 7072, 6941 and 6821 cm−1. These three peaks can be assigned as S0, S1 and S2, representing water molecules with no, one and two hydrogen bonds and the result was consistent with those reported before25,26.f p To study the effect of HA concentration on the absorption, the absorption band of 7930–5930 cm−1 at 25 °C as the region of the absorption peak between two minima was selected to make further analysis26,27. The absorption values in this region were averaged to reduce the noise effect on the data and plotted against HA concentration in Fig. 2. Consistently, the absorption dropped almost linearly when adding HA to water. However, it should be noted that from the concentration of 40 mg/mL onward, the slope starts to decrease slowly. PCA analysis on HA concentration dependent NIR spectra. To figure out the water molecular struc- tures at different HA concentration, PCA of HA solution spectra had been performed. Figure 3(a),(b) showed the first two loadings and scores of the spectra of HA aqueous solutions, respectively. The first principal component accounted for 99.4% of the total spectral change, and its loading was characterized by a broad negative peak centered at 6900 cm−1. With a variance explanation of 99.4%, the first principal component can be considered to explain almost all spectral changes of importance. The second principal component described 0.53% of the total variance, and its loading showed a dominating positive peak at 7089 cm−1 and a negative characteristic at 6634 cm−1, characterized by less hydrogen bonded OH and more hydrogen bonded OH. The first two principal components had explained 99.93% spectral variation. The score of the remaining principal components fluctu- ated around zero and could be considered as the change due to noise. Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ Figure 3. Results NIR The first two principal components of the spectra of HA solutions with different concentrations at 25 °C extracted by PCA: the loadings (a1, a2), scores (b1, b2). The first principal component accounts for 99.4% of the total variance and the second principal component accounts for 0.53% of the total variance. Figure 3. The first two principal components of the spectra of HA solutions with different concentrations at 25 °C extracted by PCA: the loadings (a1, a2), scores (b1, b2). The first principal component accounts for 99.4% of the total variance and the second principal component accounts for 0.53% of the total variance. Figure 4. Concentration dependency of water spectral changes depicted by aquagram patterns at 25 °C. Figure 4. Concentration dependency of water spectral changes depicted by aquagram patterns at 25 °C. Analysis of different water species in HA aqueous solutions with different concentration. In order to further investigate the water structure changes caused by HA, the reconstructed spectra from the first two principal components were calculated. Then, a radar chart named aquagram was established to visualize water spectral pattern at different concentrations. Twelve characteristic water wavenumbers have been defined by Tsenkova et al.28–30, which cover various conformations of water molecules and can be used to character the spectral patterns in the first overtone region of water. Figure 4 displayed the concentration dependency of water spectral changes depicted by aquagram patterns. Figure 5 showed the intensity variation of each water species with increasing HA concentration. It can be seen that with the increase of HA concentration, the water species of C1-C8 showed a linear downward trend, while the C11 and C12 showed an upward trend. The intensities of C9 (S2) and C10 (S3) decreased firstly, and when the HA concentration was higher than 40 mg/mL, the intensity began to increase. PCA analysis of HA aqueous solution induced by temperature. In order to further explore the HA effect on water structure, PCA was performed on spectra data of HA solution at different temperatures. Figure 6 showed the first two loadings and scores of the spectra of 100 mg/mL HA solutions from 25 to 55 °C. The first principal component explained 99.8% of the total spectral variance, and its loading was dominated by a positive peak at 7091 cm−1 and a broader negative peak centered at 6725 cm−1. Discussionh The perturbation of hydration water structure due to the presence of HA was investigated by NIR spectroscopy. The decrease in the intensity with increasing HA concentration in Fig. 1 could be explained by the reduction of water. Furthermore, as the temperature increases, the peak position shifts to high frequency which may be due to an increase in weak hydrogen bonded OH and a decrease in strong hydrogen bonded OH32. The intensities of the three water species S0, S1 and S2 in Fig. 1(b) changed with increasing temperature, indicating the temperature had the effects on water structure.h f The calculation of the absorption values in the region of 7930–5930 cm−1 in Fig. 2 proved the nonlinear decreas- ing behavior of HA aqueous solutions. The reason for the decreasing generally comes from two possibilities. One possibility was that HA absorbed less NIR light than water in the OH stretching region, and therefore, an increase in the concentration of HA would result in a decrease in absorption. It can be considered that although HA has a long chain structure, its contribution to NIR absorption is different from that of pure water. Another possibility was that water molecules were affected by HA, which finally changed the molecular vibration. If the change was entirely caused by the absorption of HA, the absorption value should be linear with HA concentration. Therefore, the changes should also come from the third substance, that was, the contribution of hydration water.hi PCA was performed on HA concentration dependent NIR spectra. The score of the first component in Fig. 3 was simply a straight line, suggesting that the spectral changes extracted by the first component varied linearly over the entire concentration range, mainly reflecting the decreasing of water. The score vector was parabolic with a top position of 40 mg/mL, which indicated that the spectral changes captured by the second loading had undergone a transitional change around this concentration. This indicated that the loadings plot of the second principal component reflected the hydrogen bond change of water caused by the increase in HA concentration. Aquagram was plotted based on the reconstructed spectra of PCA in Fig. 4. As a result, quite a characteristic water absorbance pattern was biased towards the low frequency region with increasing HA concentration, indi- cating the transformation from less hydrogen bonded water structure to more hydrogen bonded water cluster. Results NIR With a variance explanation of 99.8%, the first principal component can be considered to explain all spectral changes of importance. The second principal component accounted for 0.16% of the total variance. Although its loading showed a dominating positive peak at 7018 cm−1, its score vector gave a fluctuation around zero, indicating little more than a noise. Therefore, only the first principal component was used to reconstruct the spectra. Then the reconstructed spectra were depicted by aquagram, as shown in Fig. 7. Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ Figure 5. Intensity variation of different water species with different HA concentrations at 25 °C (a) for C1-C8 and (b) for C9-C12. Figure 5. Intensity variation of different water species with different HA concentrations at 25 °C (a) for C1-C8 and (b) for C9-C12. Figure 6. The first two principal components of the spectra of 100 mg/mL HA solutions at different temperatures extracted by PCA: the loadings (a1, a2), scores (b1, b2). The first principal component accounts for 99.8% of the total variance and the second principal component accounts for 0.16% of the total variance. Figure 6. The first two principal components of the spectra of 100 mg/mL HA solutions at different temperatures extracted by PCA: the loadings (a1, a2), scores (b1, b2). The first principal component accounts for 99.8% of the total variance and the second principal component accounts for 0.16% of the total variance. Figure 7. Temperature dependency of water spectral changes depicted by aquagram patterns in 100 mg/mL HA aqueous solution. Figure 7. Temperature dependency of water spectral changes depicted by aquagram patterns in 100 mg/mL HA aqueous solution. Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ Figure 8. Variation of the peak intensity of S0 (a), S1 (b) and S2 (c) in the reconstructed spectra of water and HA solution (100 mg/mL) with temperature. The solid lines were obtained by linear fitting. Figure 8. Variation of the peak intensity of S0 (a), S1 (b) and S2 (c) in the reconstructed spectra of water and HA solution (100 mg/mL) with temperature. The solid lines were obtained by linear fitting. Analysis of different water species in HA aqueous solutions with different temperature. In order to get more information of water structure changes with temperature in the HA solution, the reconstructed spectra from the first component were calculated. Results NIR Figure 8 showed the intensity variation for the spectral com- ponents of S0, S1 and S2, which could be clearly identified in Fig. 1(b). As illustrated in Fig. 8 that, with increasing temperature, the intensity for the spectra of S0 and S1 increased but that for S2 decreased, indicating clearly that more OH of S0 and S1 was formed but that of S2 was dissociated. The result was consistent with the previous studies, that was the larger water clusters dissociated into smaller ones with the increasing temperature25,26,31. The temperature effect on S0, S1 and S2 could be obtained by comparing the slope in the Fig. 8. The slope indicates the variation rate of the intensity with temperature, showing the sensitivity of the water species to temperature. The squared correlation R2 was obtained from linear fitting between the intensity of the spectral components and the temperature. Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 Materials and Methods Reagents and sample preparation. HA was provided by Bloomage Biotechnology (Jinan, China) with Mw of 7775 Da by using gel permeation chromatography coupled with multiple-angle laser light scattering (GPC- MALS). Continuous concentration of HA solutions (1 mg/mL, 5 mg/mL and 10–100 mg/mL with a step of 10 mg/ mL) were prepared by dissolving HA powder in double distilled water (Milli-Q, Millipore, resistivity ≥18.5 MΩ cm). Spectral measurement. All NIR spectra were recorded using Antaris II FT-NIR spectrometer (Thermo Fisher scientific Inc., American) equipped with a 1 mm cuvette, tungsten-halogen light source, and InGaAs detec- tor. The spectral range is from 10000 to 4000 cm−1 and the spectra are digitalized with 4 cm−1 intervals in Fourier transform. The temperature-dependent spectra of water and HA solutions were collected between 25 °C and 55 °C with 5 °C interval. In order to increase the signal to noise ratio, the air reference was measured and subtracted from sample spectrum and the scan number was both set to 32. The spectrum of each sample was measured three times and averaged as a final spectrum. Data processing. Spectra were imported into Matlab R2015a (The Math Works Inc., Natick, MA, USA), which was used for data transformation and processing. PCA was performed on spectral data of HA aqueous solution to extract information related to water changes and the reconstructed spectra were calculated accord- ingly. Then, a radar chart named aquagram was established to visualize water spectral pattern at different concen- trations. Discussionh The intensity variation of each water species in Fig. 5 showed that when the HA concentration was lower than 40 mg/mL, only the formation of S4 was promoted. Probably because HA is a long-chain polysaccharide rich in hydroxyl, it can directly combine a large number of water molecules into a tight network structure. The addition of a low concentration of HA may promote the conversion of S1, S2 and S3 into S4, the strongly hydrogen bonded water species. However, as the concentration was further increased above 40 mg/mL, S4 continued to increase, while more HA molecules began to promote the formation of S2 and S3. When HA concentration was lower than 40 mg/mL, the strong hydrogen bonded water structure was preferentially formed and when the concentration continued to rise, the medium hydrogen bonded water network began to form. Moreover, it has been found that in dilute solution, HA behaves as a stiffened random coil, while in concentrated solutions, stiffened random coils show entanglement33,34. Therefore, water may be involved in HA entanglement. Previous studies have shown that biomacromolecules normally promote the formation of hydrogen bonds in solution, such as the formation of S2, Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ S3, and S4 22,26. HA solution exhibits different types of water networks at different HA concentration, which may be related to its long extending structure in water.h g g PCA analysis was carried out on the spectra of 100 mg/mL HA solutions from 25 to 55 °C.The score of the first principal component in Fig. 6(a) changed almost linearly with the temperature, indicating that the spectra that were extracted by this component changed a linear trend over the whole temperature region. Based on the reconstructed spectra, the aquagram showed quite a characteristic trend that the absorbance was biased towards to high frequency, indicating a dissociation of water cluster and the formation of less hydrogen bonded water structure32. The peak intensity of S0, S1 and S2 in Fig. 8 were reduced, which may be due to the volume effect. Meanwhile, the slope R2 of S0, S1 and S2 with temperature in the HA solution is less than its slope in pure water, indicating that HA had the effect of improving the stability of water structure and prevent the changes with tem- perature. The same conclusion was found in the study of other concentrations of HA aqueous solution. Discussionh Since a large number of hydrogen bonds can be formed between the hydroxyl groups of HA and water1, it is speculated that when the temperature rises, the intermolecular hydrogen bonds need to be broken first, which increases the resistance to the vibration of water molecules. This result can be interpreted that the thermal stability of HA solu- tion was achieved by the stability of water35 and further provided an evidence of the biological protective function of carbohydrates in aqueous solutions by reducing the sensitivity of water to temperature36–38. Materials and Methods The ranges of peaks for aquagram depicting were described as follows28–30: C1 (7440 cm−1: 2 close ν3: H2O asymmetric stretching vibration.), C2 (7349 cm−1: OH-(H2O)1,2,4: water solvation shell.), C3 (7270 cm−1: ν1 + ν3: H2O symmetrical stretching vibration and H2O asymmetric stretching vibration.), C4 (7234 cm−1: OH-(H2O)1,4: water solvation shell, O2-(H2O)4: hydrated superoxide clusters, 2 close ν1: H2O symmetrical stretching vibration.), C5 (7072 cm−1: water confined in a local field of ions, S0: free water, water with free OH-.), C6 (7037 cm−1: water hydration band, H-OH bend and OH…O.), C7 (6941 cm−1: S1: water molecules with one hydrogen bond.), C8 (6906 cm−1: OH-(H2O)4,5: water solvation shell.), C9 (6821 cm−1: S2: water molecules with two hydrogen bonds, 2 close ν2 + ν3: H2O bending and asymmetrical stretching vibration.), C10 (6784 cm−1: S3: water molecules with three hydrogen bonds.), C11 (6705 cm−1: S4: water molecules with four hydrogen bonds.), C12 (6607 cm−1: ν1: H2O symmetrical stretching vibration, ν2: H2O bending vibration, strongly bound water.). Twelve characteristic water absorption bands mentioned above covering the most unique water species termed as water matrix coordinates (WAMACS) were used as axes and the values for aquagram were derived from the following equation39: μ σ = − λ λ λ ′ λ A A ( )/ μ σ = − λ λ λ ′ λ A A ( )/ (1) (1) where Aλ, μλ and σλ are absorbance after multiplicative scatter correction (MSC), the average of all spectra and the standard deviation of all spectra, respectively, at wavelength λ (convert to wavenumber in this paper). where Aλ, μλ and σλ are absorbance after multiplicative scatter correction (MSC), the average of all spectra and he standard deviation of all spectra, respectively, at wavelength λ (convert to wavenumber in this paper). Received: 8 November 2019; Accepted: 3 January 2020; Published: xx xx xxxx Received: 8 November 2019; Accepted: 3 January 2020; Published: xx xx xxxx Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 References Secondary and tertiary structures of hyaluronan in aqueous solution, investigat rotary shadowing-electron microscopy and computer simulation. Biochem. J. 274, 699–705 (1991). d k l l k Th l d d f h h l h l l d y g py p 35. Mondek, J., Kalina, M., Simulescu, V. & Pekař, M. Thermal degradation of high molar mass hyaluronan in solution and in powder; comparison with BSA. Polym. Degrad. Stab. 120, 107–113 (2015). p y g 6. Momany, F. & Schnupf, U. DFT optimization and DFT-MD studies of glucose, ten explicit water molecules enclosed by an implici solvent, COSMO. Comput. Theor. Chem. 1029, 57–67 (2014). ph 37. Cocinero, E. J. et al. Free fructose is conformationally locked. J. Am. Chem. Soc. 135, 2845–2852 (2013).hi 37. Cocinero, E. J. et al. Free fructose is conformationally locked. J 38. Suzuki, T. The hydration of glucose: the local configurations in sugar–water hydrogen bonds. Phys. Chem. Chem. Phys. 10, 96 (2008). 9. Tsenkova, R., Iordanova, I. K., Toyoda, K. & Brown, D. R. Prion protein fate governed by metal binding. Biochem. Biophys. Res Commun. 325, 1005–1012 (2004). Acknowledgementsh g This work was supported by National Natural Science Foundation of China (81703403) and Major Innovation Project of Shandong Province (2018CXGC1405). The author would like to thank Bloomage biotechnology for free providing HA samples. j g free providing HA samples. References 1. Kuo, J. W. Practical aspects of hyaluronan based medical products 5–24 (CRC/Taylor & Francis 2006). 2. Bukhari, S. N. A. et al. Hyaluronic acid, a promising skin rejuvenating biomedicine: a review of recent updates and pre-clinical and clinical investigations on cosmetic and nutricosmetic effects. Int. J. Biol. Macromol. 120, 1682–1695 (2018). gf ( ) 3. Salwowska, N. M., Bebenek, K. A., Żądło, D. A. & Wcisło-Dziadecka, D. L. Physiochemical properties and application of hyaluronic acid: a systematic review. J. Cosmet. Dermatol. 15, 520–526 (2016). gf 3. Salwowska, N. M., Bebenek, K. A., Żądło, D. A. & Wcisło-Dziadecka, D. L. Physiochemical properties and application of hyaluronic acid: a systematic review. J. Cosmet. Dermatol. 15, 520–526 (2016). 4. Di Cerbo, A., Laurino, C., Palmieri, B. & Iannitti, T. A. dietary supplement improves facial photoaging and skin sebum, hydration and tonicity modulating serum fibronectin, neutrophil elastase 2, hyaluronic acid and carbonylated proteins. J. Photochem. Photobiol. B. 144, 94–103 (2015). 4. Di Cerbo, A., Laurino, C., Palmieri, B. & Iannitti, T. A. dietary supplement improves facial photoaging and skin sebum, hydration and tonicity modulating serum fibronectin, neutrophil elastase 2, hyaluronic acid and carbonylated proteins. J. Photochem. Photobiol. B. 144, 94–103 (2015). 5. Laage, D., Elsaesser, T. & Hynes, J. T. Water dynamics in the hydration shells of biomolecules. Chem. Rev. 117, 10694–10725 (2017). 6. Biswas, R. & Bagchi, B. Anomalous water dynamics at surfaces and interfaces: Synergistic effects of confinement and surface interactions. J. Phys. Condens. Matter 30, 013001 (2018).l 5. Laage, D., Elsaesser, T. & Hynes, J. T. Water dynamics in the hydration shells of biomolecules. Chem. Rev. 117, 10694–10725 (2017). 6. Biswas, R. & Bagchi, B. Anomalous water dynamics at surfaces and interfaces: Synergistic effects of confinement and surface interactions. J. Phys. Condens. Matter 30, 013001 (2018).l Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ . Cowman, M. K. & Matsuoka, S. Experimental approaches to hyal 9. Davies, A., Gormally, J., Wyn-Jones, E., Wedlock, D. J. & Phillips, G. O. A study of hydration of sodium hyaluronat compressibility and high precision densitometric measurements. Int. J. Biol. Macromol. 4, 436–438 (1982).h avies, A., Gormally, J., Wyn-Jones, E., Wedlock, D. J. & Phillips, G. O. A study of hydration of sodium hyaluronate from ompressibility and high precision densitometric measurements. Int. J. Biol. Macromol. 4, 436–438 (1982).h k y , k y , y p p yh , ( ) 11. Liu, J. References & Cowman, M. Thermal analysis of semi-dilute hyaluronan solutions. J. Therm. Anal. Calorim. 59, 547–557 (2000). h y yh 2. Haxaire, K., Maréchal, Y., Milas, M. & Rinaudo, M. Hydration of polysaccharide hyaluronan observed by IR spectrometry. Preliminary experiments and band assignments. Biopolym. 72, 10–20 (2003). y p g p y 13. Haxaire, K., Maréchal, Y., Milas, M. & Rinaudo, M. Hydration of hyaluronan polysaccharide observed by IR spe Definition and quantitative analysis of elementary hydration spectra and water uptake. Biopolym. 72, 149–161 (2003) y p g p y 13. Haxaire, K., Maréchal, Y., Milas, M. & Rinaudo, M. Hydration of hyaluronan polysaccharide observed by IR spectrometr 13. Haxaire, K., Maréchal, Y., Milas, M. & Rinaudo, M. Hydration of hyaluronan polysaccharide observed by IR spectrometry. II. Definition and quantitative analysis of elementary hydration spectra and water uptake. Biopolym. 72, 149–161 (2003). i q y y y p p p y 4. Maréchal, Y., Milas, M. & Rinaudo, M. Hydration of hyaluronan polysaccharide observed by IR spectrometry. III. Structure and mechanism of hydration. Biopolym. 72, 162–173 (2003).h y p y 15. Mazeau, K. & Rinaudo, M. The prediction of the characteristics of some polysaccharides from molecular modeling. Comparison with effective behavior. Food Hydrocolloids 18, 885–898 (2004). ff y 6. Servaty, R., Schiller, J., Binder, H. & Arnold, K. Hydration of polymeric components of cartilage — an infrared spectroscopic study on hyaluronic acid and chondroitin sulfate. Int. J. Biol. Macromol. 28, 121–127 (2001). y 7. Kargerová, A. & Pekař, M. Densitometry and ultrasound velocimetry of hyaluronan solutions in water and in sodium chloride solution. Carbohydr. Polym. 106, 453–459 (2014).t y y 18. Korogiannaki, M., Jones, L. & Sheardown, H. Impact of a Hyaluronic Acid-Grafted Layer on the Surface Properties of Model Silicone Hydrogel Contact Lenses. Langmuir 35, 950–961 (2019). y g g 19. Ozaki, Y. Near-infrared spectroscopy — its versatility in analytical chemistry. Anal. Sci. 28, 545–563 (2012). h b b f l l b 20. Cui, X., Liu, X., Yu, X., Cai, W. & Shao, X. Water can be a probe for sensing glucose in aqueous solutions by temperature depen near infrared spectra. Anal. Chim. Acta 957, 47–54 (2017). near infrared spectra. Anal. Chim. Acta 957, 47–54 (2017). 21. Dong, Q. Near-infrared spectroscopic study of molecular interaction in ethanol-water mixtures. Spectrochim. Acta Part A 222, 117183 (2019). ( ) 2. Dong, Q. References Analysis of hydration water around human serum albumin using near-infrared spectroscopy. Int. J. Biol. Macromol. 138 927–932 (2019). 23. Czarnecki, M. A., Morisawa, Y., Futami, Y. & Ozaki, Y. Advances in Molecular Structure and Interaction Studies Using Near-Infrared Spectroscopy. Chem. Rev. 115, 9707–9744 (2015). p py 24. Workman, J. & Weyer, L. Practical Guide to Interpretive Near-Infrared Spectroscopy 57–58 (CRC Press, 2007). 25. Maeda, H., Ozaki, Y., Tanaka, M., Hayashi, N. & Kojima, T. Near infrared spectroscopy and chemometrics studies of temperature- dependent spectral variations of water: relationship between spectral changes and hydrogen bonds. J. Near Infrared Spectrosc. 3, 191–201 (1995). 26. Cui, X., Cai, W. & Shao, X. Glucose induced variation of water structure from temperature dependent near infrared spectra. RSC Adv. 6, 105729–105736 (2016). 27. McCabe, W. C. & Fisher, H. F. Near-infrared spectroscopic method for investigating the hydration of a solute in aqueous solution. J. Phys. Chem. 74, 2990–2998 (1970). 8. Tsenkova, R. Aquaphotomics: dynamic spectroscopy of aqueous and biological systems describes peculiarities of water. J. Nea Infrared Spectrosc. 17, 303–313 (2009).ii 29. Kojić, D., Tsenkova, R., Tomobe, K., Yasuoka, K. & Yasui, M. Water confined in the local field of ions. ChemPhysChem 15, 4077–4086 (2014). 30. Gowen, A. A., Tsenkova, R., Esquerre, C., Downey, G. & O’Donnell, C. P. Use of near infrared hyperspectral imaging to identify water matrix co-ordinates in mushrooms (agaricus bisporus) subjected to mechanical vibration. J. Near Infrared Spectrosc. 17, 363–371 (2009). ( ) 31. Cheng, D., Cai, W. & Shao, X. Understanding the interaction between oligopeptide and water in aqueous solution using tempera dependent near-infrared spectroscopy. Appl. Spectrosc. 72, 1354–1361 (2018). 32. Segtnan, V. H., Šašić, Š., Isaksson, T. & Ozaki, Y. Studies on the structure of water using two-dimensional near-infrared correl spectroscopy and principal component analysis. Anal. Chem. 73, 3153–3161 (2001). p py p p p y 33. Garg, H. G. & Hales, C. A. Chemistry and biology of hyaluronan 3–4 (Elsevier Ltd 2004). 33. Garg, H. G. & Hales, C. A. Chemistry and biology of hyaluronan 3 4 (Elsevier Ltd 2004). 34. Scott, J. E., Cummings, C., Brass, A. & Chen, Y. Secondary and tertiary structures of hyaluronan in aqueous solution, investigated by rotary shadowing electron microscopy and computer simulation Biochem J 274 699 705 (1991) 34. Scott, J. E., Cummings, C., Brass, A. & Chen, Y. Competing interests Th h d l g The authors declare no competing interests. The authors declare no competing interests. Author contributions Q.D. and H.Za. conceived the study and supervised all experiments. X.G. and S.H. were responsible for HA related technical consultation. Q.D., L.L. and C.Y. carried out all experiments. Q.D. contributed to data analysis, result interpretation, figure preparation and wrote the manuscript. L.N., W.T. and H.Zh. participated in the discussions. All the authors reviewed and approved the manuscript. Additional information Correspondence and requests for materials should be addressed to H.Z. Reprints and permissions information is available at www.nature.com/reprints. Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5 www.nature.com/scientificreports/ 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. 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/. fi Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. 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/. Additional information The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2020 Scientific Reports \| (2020) 10:1387 \| https://doi.org/10.1038/s41598-020-58417-5
https://openalex.org/W3014288215	https://www.cambridge.org/core/services/aop-cambridge-core/content/view/0F1AAAD41732E64F963313A2E91D1B0C/S0033291720000604a.pdf/div-class-title-translating-the-promise-of-5ht-span-class-sub-4-span-receptor-agonists-for-the-treatment-of-depression-div.pdf	English	null	Translating the promise of 5HT<sub>4</sub> receptor agonists for the treatment of depression	Psychological medicine	2,020	cc-by	10,921	Author for correspondence: Author for correspondence: Susannah E Murphy, E-mail: susannah. murphy@psych.ox.ac.uk Author for correspondence: Susannah E Murphy, E-mail: susannah. murphy@psych.ox.ac.uk Invited Review Cite this article: Murphy SE, de Cates AN, Gillespie AL, Godlewska BR, Scaife JC, Wright LC, Cowen PJ, Harmer CJ (2021). Translating the promise of 5HT4 receptor agonists for the treatment of depression. Psychological Medicine 51, 1111–1120. https://doi.org/ 10.1017/S0033291720000604 1University Department of Psychiatry, Warneford Hospital, University of Oxford, OX3 7JX, UK and 2Oxford Health NHS Foundation Trust, Warneford Hospital, Oxford, UK Introduction 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. More selective activation of specific 5-HT receptors has the potential to produce better tol- erated, faster acting treatments. Preclinical studies suggest that the 5-HT4 receptor is a particu- larly promising therapeutic target for this approach. In animal models, 5-HT4 receptor activation has antidepressant effects across a range of behavioural paradigms. Intriguingly, the onset of antidepressant effects following 5-HT4 receptor agonism is more rapid than that seen with SSRIs, and is accompanied by early molecular and cellular changes that mirror those seen following chronic administration of conventional antidepressants. This profile of effects suggests that 5-HT4 receptor agonists may hold potential as novel antidepressants, Abstract Animal experimental studies suggest that 5-HT4 receptor activation holds promise as a novel target for the treatment of depression and cognitive impairment. 5-HT4 receptors are post- synaptic receptors that are located in striatal and limbic areas known to be involved in cog- nition and mood. Consistent with this, 5-HT4 receptor agonists produce rapid antidepressant effects in a number of animal models of depression, and pro-cognitive effects in tasks of learn- ing and memory. These effects are accompanied by molecular changes, such as the increased expression of neuroplasticity-related proteins that are typical of clinically useful antidepressant drugs. Intriguingly, these antidepressant-like effects have a fast onset of their action, raising the possibility that 5-HT4 receptor agonists may be a particularly useful augmentation strategy in the early stages of SSRI treatment. Until recently, the translation of these effects to humans has been challenging. Here, we review the evidence from animal studies that the 5-HT4 recep- tor is a promising target for the treatment of depression and cognitive disorders, and outline a potential pathway for the efficient and cost-effective translation of these effects into humans and, ultimately, to the clinic. Received: 13 September 2019 Revised: 21 February 2020 Accepted: 28 February 2020 First published online: 3 April 2020 Translating the promise of 5HT4 receptor agonists for the treatment of depression Susannah E Murphy1,2, Angharad N de Cates1,2, Amy L Gillespie1,2, Beata R Godlewska1,2, Jessica C Scaife1,2, Lucy C Wright1,2, Philip J Cowen1,2 and Catherine J Harmer1,2 Introduction The potentiation of monoamine activity, and in particular serotonergic neurotransmission, has been the dominant target of antidepressant treatments since the serendipitous discovery of the antidepressant effects of monoamine oxidase inhibitor and tricyclics in the 1950s. Whilst drugs such as selective serotonin reuptake inhibitors (SSRIs) are more effective than placebo (Cipriani et al., 2018), there are significant limitations to current treatment approaches. Only about one third of patients achieve remission after 14 weeks of SSRI treatment and around 30% of patients remain ‘treatment resistant’, even after trying multiple treatment approaches (Rush et al., 2009; Trivedi et al., 2006). When SSRIs are effective, they require chronic administration to achieve therapeutic efficacy and it can take several weeks for depres- sive symptoms to fully resolve (Frazer & Benmansour, 2002; Mitchell, 2006). SSRI treatment is also associated with adverse effects which can compromise treatment adherence, including gastrointestinal symptoms, drowsiness, weight gain, sexual dysfunction and emotional blunt- ing (Goethe, Woolley, Cardoni, Woznicki, & Piez, 2007; Goodwin, Price, De Bodinat, & Laredo, 2017). Given that depression is now ranked as the single largest contributor to global disability (World Health Organisation, 2017), there is a pressing need for novel treatment approaches that are more effective, better tolerated and faster acting. pp g SSRIs selectively block serotonin (5-hydroxytryptamine, 5-HT) transporters, which results in an increase in 5-HT levels at all post-synaptic serotonin receptor subtypes. This indirect activation of a broad range of receptors is likely to contribute not only to the therapeutic action, but also to the unwanted aspects of SSRIs. In particular, early activation of 5-HT1A autoreceptors by SSRIs results in a paradoxical decrease in 5-HT cell firing, and is thought to contribute to the slow onset of the clinical action of SSRIs (Artigas, 2013). Equally, activa- tion of post-synaptic 5-HT2C receptors has been associated with the increase in anxiety that can characterise initiation of SSRI treatment (Burghardt, Bush, McEwen, & LeDoux, 2007). l f f h h l d b l © The Author(s) 2020. 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. © The Author(s) 2020. Key words: Antidepressants; cognition; emotion; serotonin Antidepressants; cognition; emotion; serotonin Psychological Medicine cambridge.org/psm Psychological Medicine 5-HT4 receptors are expressed in the basal ganglia and limbic system Since its first description 30 years ago (Dumuis, Bouhelal, Sebben, Cory, & Bockaert, 1988), the 5-HT4 receptor has been discovered both in the brain and in various sites in the periphery. Peripherally, 5-HT4 receptors are found in the heart (Kaumann, 1990), the intestine (Craig & Clarke, 1990), the adrenal glands (Idres, Delarue, Lefebvre, & Vaudry, 1991) and the bladder (Tonini & Candura, 1996). In the brain, 5-HT4 receptors are located post-synaptically and are expressed in the hypothalamus, hippocampus, basal ganglia, amygdala, olfactory bulb, septal area and neocortex (Cai, Flores-Hernandez, Feng, & Yan, 2002; Compan et al., 1996; Roychowdhury, Haas, & Anderson, 1994; Vilaró, Cortés, & Mengod, 2005; Waeber, Sebben, Nieoullon, Bockaert, & Dumuis, 1994). In rodents, 5-HT4 receptors are mainly situated on gamma-aminobutyric acid (GABA) neurones in the striatum and nucleus accumbens (Compan et al., 1996), glutamate pyramidal cells in the medial prefrontal cortex (PFC) and hippocampus (Cai et al., 2002; Vilaró et al., 2005) and hippo- campal granule cells (Vilaró et al., 2005). 5-HT4 receptors have also been described in the rat dorsal raphe nucleus (DRN) in autoradiographic studies (Vilaró et al., 2005; Waeber et al., 1994), although 5-HT4 receptor mRNA has not been found sug- gesting that the 5-HT4 receptors are located on DRN afferent neu- rons rather than on DRN 5-HT cells (Vilaró et al., 2005). Importantly, 5-HT4 receptors in the PFC control the firing rate of midbrain serotonergic cells. Consistent with this, the basal fir- ing of 5-HT cells in the DRN is reduced by over 50% in 5-HT4 receptor knockout mice compared with wildtypes (Conductier et al., 2006). The facilitatory control exerted by 5-HT4 receptors on serotonergic firing is thought to operate via a positive PFC– DRN feedback loop in which the activation of postsynaptic 5-HT4 receptors located on medial PFC pyramidal cells has an excitatory effect on serotonin neurons in the DRN dorsal raphe, resulting in an increase in the release of serotonin in target struc- tures such as the PFC and hippocampus. Consistent with this model, overexpression of 5-HT4 receptors in the mPFC, but not Evidence from PET studies in humans confirms that 5-HT4 receptors have a high density in the caudate, putamen and nucleus accumbens, with lower expression across a broad range of cortical and limbic areas, including the hippocampus and amygdala (Beliveau et al., 2017). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 1112 Susannah E Murphy et al. & Sanger, 2001). Post-mortem human brain studies have also con- firmed a high density of the 5-HT4 receptor in the basal ganglia (caudate nucleus, putamen, nucleus accumbens, globus pallidus and substantia nigra), as well as the hippocampus (particularly in the CA1 region) and an intermediate density in the neocortex and amygdala (Bonaventure et al., 2000; Varnas, Halldin, Pike, & Hall, 2003). In addition, low levels of 5-HT4 binding sites have been reported in the human thalamus and raphe nuclei (Beliveau et al., 2017; Varnas et al., 2003). and may be useful adjuncts to SSRI treatment to increase the speed of onset of therapeutic effects. In addition, 5-HT4 receptor agonists produce pro-cognitive effects in tasks of learning and memory in rodents, highlighting a potential role in targeting cog- nitive deficits associated with depression, which are not well addressed by current antidepressant medications (Shilyansky et al., 2016). To date, the evidence that the 5-HT4 receptor might be a use- ful treatment target mainly comes from preclinical studies, and it is important that these findings are translated into human studies. Human PET imaging studies using the [11C]-SB 207145 ligand have provided some initial support for a role of the 5-HT4 recep- tor in the pathophysiology of depression (Madsen et al., 2014). However, it remains to be established whether activation of 5-HT4 receptors in humans has effects that mirror the antidepres- sant and pro-cognitive effects seen in animals, and therefore whether this is a viable antidepressant target. Until recently, such experimental studies in humans have not possible due to a lack of suitable compounds. With the licensing of the 5-HT4 par- tial receptor agonist, prucalopride (Resolor), for the treatment of constipation, there is a timely opportunity to establish the effects of 5-HT4 activation on mood and cognition in humans, using an experimental medicine framework. Here, we review the evidence from animal studies that the 5-HT4 receptor is a promising target for the treatment of depression and cognitive disorders, and out- line a potential pathway for the efficient translation of these effects into humans and, ultimately, to the clinic. The neurochemical and molecular effects of 5-HT4 receptor activation 5HT4 receptors are transmembrane-spanning G protein-coupled receptors, and their activation results in increased neuronal excit- ability that is mediated through the stimulation of adenylyl cyclase and an increase in cyclic adenosine monophosphate (Bockaert, Claeysen, Compan, & Dumuis, 2008; Dumuis et al., 1988). In addition, 5-HT4 receptor activation initiates important G protein-independent pathways, including the activation of tyro- sine kinase Src and, in turn, neuronal extracellular-signal-regu- lated kinase (ERK) (Barthet et al., 2007), which is known to play a critical role in hippocampal synaptic plasticity (Norman, Thiels, Barrionuevo, & Klann, 2000). 5-HT4 receptors also indirectly affect neuronal function through their modulation of neurotransmitter release, including GABA (Bijak & Misgeld, 1997), acetylcholine (Consolo, Arnaboldi, Giorgi, Russi, & Ladinsky, 1994; Johnson et al., 2012; Siniscalchi, Badini, Beani, & Bianchi, 1999), dopamine (Bockaert, Claeysen, Compan, & Dumuis, 2004; Bockaert et al., 2008; Bonhomme, De Deurwaerdere, Le Moal, & Spampinato, 1995), histamine (Johnson et al., 2012) and serotonin (Ge & Barnes, 1996; Licht et al., 2010). Of particular interest is the enhanced release of acetylcholine in the frontal cortex and hippo- campus following 5-HT4 receptor activation, which has been linked to its facilitatory role in memory and cognition (Hagena & Manahan-Vaughan, 2017; King, Marsden, & Fone, 2008). In support of this 5-HT4 receptor agonists have been shown to increase electrically-evoked [3H]choline efflux in guinea pig brain slices in the cerebral cortex, hippocampus and nucleus basa- lis magnocellularis (NbM) (Siniscalchi et al., 1999), and facilitate the release of ACh selectively in the frontal cortex (Consolo et al., 1994). Importantly, both of these effects are blocked by 5-HT4 receptor antagonists, which further supports a specific role for the 5-HT4 receptor in cholinergic neurotransmission. Given that 5-HT4 receptors are not present on cholinergic basal forebrain neurons, it is thought that the enhancement of ACh release induced by 5-HT4 receptor agonists is likely to be mediated by GABAergic and/or glutamatergic cell populations (Penas-Cazorla & Vilaro, 2015). 5-HT4 receptors are expressed in the basal ganglia and limbic system https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 5-HT4 receptors are expressed in the basal ganglia and limbic system 5-HT4 receptor mRNA and protein has also been shown to be down-regulated in rats subjected to maternal deprivation (but not chronic unpredictable stress), with a strong correlation between levels of hippocampal 5-HT4 receptor mRNA and anhedonic-like behaviour in these animals (Bai et al., 2014). Short-term 5-HT4 receptor activation has also been shown to increase hippocampal cell proliferation and the expression of neuroplasticity-related proteins such as CREB and BDNF (Ishizuka, Goshima, Ozawa, & Watanabe, 2014; Lucas et al., 2007; Pascual-Brazo et al., 2012). These mirror those seen with chronic antidepressant administration (Malberg, Eisch, Nestler, & Duman, 2000; Warner-Schmidt & Duman, 2006), but occur more rapidly. For example, 3 days administration of the 5-HT4 receptor agonist, RS67333, increases hippocampal cell prolifer- ation in the dentate gyrus and the expression of neuroplasticity-related proteins, such as BDNF and CREB (Lucas et al., 2007; Pascual-Brazo et al., 2012). After 7 days of administration the level of these neuroplastic changes are compar- able with those seen after 2–3 weeks of SSRI administration (Malberg et al., 2000; Pascual-Brazo et al., 2012; Samuels et al., 2016). Further, long-term 5-HT4 receptor activation (4 weeks) facilitates the maturation of newborn neurons in the adult rat hippocampus (Mendez-David et al., 2014). ( ) 5-HT4 receptor knock out mice do not show a clear depressed- like phenotype, with no differences in behaviour on the forced swim test, which may be due to the considerable adaptations in the serotoninergic system that occur across development in these animals (Amigo et al., 2016; Conductier et al., 2006). However, the knock out animals do show some evidence of anhedonic- and apathetic-like behaviour, with a reduction in sucrose consumption and nesting behaviour, and decreased cen- tral time in the open field (OF), which is suggestive of increased anxiety (Amigo et al., 2016). Interestingly, a recent study suggests that focal overexpression of the 5-HT4 receptor in the medial PFC produces an ‘anti-depressed’ and anxiolytic behavioural pheno- type, which is consistent with the central role ascribed to descend- ing inputs from the mPFC to the DRN in mediating increased raphe firing (Castello et al., 2018). pp p Furthermore, 5-HT4 receptors are involved in fine-tuning of the key cellular processes underlying synaptic plasticity, long-term potentiation (LTP) and long-term depression (LTD), which is likely to play a key role in their effects on long term hippocampal- dependent memory. 5-HT4 receptors are expressed in the basal ganglia and limbic system Similarly, the highest levels of human 5-HT4 receptor mRNA have been found in the caudate and puta- men, with lower levels in the amygdala temporal cortex and hippocampus (Medhurst, Lezoualc’h, Fischmeister, Middlemiss, 1113 Psychological Medicine in the striatum or hippocampus, increases DRN 5-HT neuronal activity (Lucas et al., 2005). In addition, pharmacological studies have demonstrated that administration of the selective 5-HT4 receptor agonists, RS67333 and prucalopride, increases DRN 5-HT cell firing (Faye et al., 2019; Lucas et al., 2005; Lucas & Debonnel, 2002; Moha ou Maati et al., 2016). Intriguingly, this increase in firing can be seen within 30 min of drug administra- tion, suggesting that 5-HT4 receptor agonists hold potential as rapid onset antidepressant agents (Lucas et al., 2005; Lucas & Debonnel, 2002). This is in stark contrast to the decrease in raphe cell firing that is characteristic of SSRIs, and which has been implicated in their delayed onset of action (Blier, Piñeyro, el Mansari, Bergeron, & de Montigny, 1998). Importantly, the excitatory effect of 5-HT4 receptor agonists on the firing of mid- brain 5-HT cells is abolished by bilateral electrolytic lesions of the mPFC, demonstrating a causal role of the mPFC in the facilitation of 5-HT cell firing by 5-HT4 receptor agonists (Moha ou Maati et al., 2016). encoding (Hagena & Manahan-Vaughan, 2017; Twarkowski et al., 2016). The effect of 5-HT4 receptor activation on dentate gyrus plasticity is particularly interesting in light of recent evi- dence that LTP induction in the dentate gyrus is sufficient to induce antidepressant-like effects in rat models within a rapid timeframe (Kanzari et al., 2018), suggesting that the facilitatory effect of 5-HT4 receptor agonists on DG plasticity may, at least in part, account for their fast-acting antidepressant-like properties. Preclinical studies support a role for the 5-HT4 receptor in the aetiology of depression and anxiety Preclinical studies provide support for the involvement of the 5-HT4 receptor in the aetiology of depression. Changes in 5-HT4 receptor expression have been reported in a number of rodent models of depression, although the direction of effect var- ies by model. In the olfactory bulbectomy (OBX) and glucocortic- oid receptor heterozygous [GR(±)] models, an increase in 5-HT4 receptor binding has been reported in the ventral hippocampus (OBX) and striatum [GR(±)] (Licht et al., 2010). In contrast, in the Flinders-sensitive line rat model of depression, a decrease in hippocampal 5-HT4 receptor binding was reported (Licht et al., 2009). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 5-HT4 receptors are expressed in the basal ganglia and limbic system Activation of the 5-HT4 receptor has been shown to prevent LTD in all the main hippocampal subfields (Kemp & Manahan-Vaughan, 2005; Twarkowski, Hagena, & Manahan-Vaughan, 2016). The effect of 5-HT4 receptor activa- tion on LTP is complex and specific to individual hippocampal subfields, which may relate to differences in the expression of dif- ferent receptor isoforms (Hagena & Manahan-Vaughan, 2017). A number of studies have demonstrated that stimulation of 5-HT4 receptors in freely behaving rats promotes synaptic information- encoding through LTP at both the CA1 region (Kemp & Manahan-Vaughan, 2005) and dentate gyrus (Kulla & Manahan-Vaughan, 2002; Twarkowski et al., 2016). Conversely, LTP at mossy fibre synapses of the CA3 region is suppressed by 5-HT4 receptor activation (Twarkowski et al., 2016). The effect of 5-HT4 receptor activation on LTP has also been shown to be time-dependent; for example, the long lasting sustainment of LTP (i.e. after 48 h) in the dentate gyrus can be impaired by 5-HT4 receptor stimulation (Twarkowski et al., 2016). Taken together, this pattern suggests that 5-HT4 receptor activation acts to promote information encoding through LTP in CA1 and dentate gyrus at the expense of information coding through LTD, which is likely to impact upon the content of information Short-term administration of 5-HT4 receptor agonists has antidepressant and anxiolytic-like effects in animals Studies in rodents have demonstrated that 5-HT4 receptor ago- nists exert antidepressant-like effects on behavioural paradigms that mirror those seen with conventional SSRIs, but with a more rapid onset of action (Lucas et al., 2007; Mendez-David et al., 2014; Pascual-Brazo et al., 2012). Using the classic forced swim paradigm, the 5-HT4 receptor partial agonists RS67333 and prucalopride have been shown to reduce immobility follow- ing acute (Lucas et al., 2007) and 3–5 day administration (Gomez-Lazaro et al., 2012; Pascual-Brazo et al., 2012). These effects parallel those seen with acute SSRI administration, which also induces a reduction in immobility on this task, although the effect of the 5-HT4 agonists was almost double that seen with acute citalopram (Lucas et al., 2007). y The effects of 5HT4 receptor antagonism on animal depres- sion- or anxiety-like behaviour are less well established. Acute administration of the 5-HT4 receptor antagonists SDZ205-557, GR113808 and SB204070 does not influence anxiety-related behaviour in the light/dark choice test in mice (Costall & Naylor, 1997). Four weeks administration of the antagonist GR125487 has also been shown not to have an effect on behav- ioural models of anxiety in a corticosterone model of depres- sion/anxiety, although the antagonist did block the effects of an SSRI in this model (Mendez-David et al., 2014). In contrast, two studies have reported an anxiolytic effect of acute administra- tion of SB204070, GR113808 (Silvestre, Fernandez, & Palacios, 1996) and SB207266A (Kennett, Bright, Trail, Blackburn, & Sanger, 1997; Silvestre et al., 1996) on the EPM test. It is unclear what explains these contradictory findings, although there are dif- ferences in dosage, timing of antagonist administration and behavioural outcome measures used in the studies, which may be important. p ( ) Short-term administration of 5-HT4 agonists also induces behavioural changes in paradigms that require chronic adminis- tration of SSRIs to shift. For example, 3 days administration of RS67333 was sufficient to increase sucrose consumption in the chronic mild stress model of depression (Lucas et al., 2007) and in CORT rats (Pascual-Brazo et al., 2012), and decrease hyperlo- comotion in the OBX model (Lucas et al., 2007). Within 7 days of RS67333 a complete reversal of the effects of corticosterone on sucrose consumption was seen, paralleling the effect of 21-day treatment with fluoxetine (Pascual-Brazo et al., 2012), and sug- gesting a rapid reversal of anhedonic behaviour following 5-HT4 receptor agonism. 5-HT4 receptors are involved in the mechanism of action of SSRIs There is also convincing evidence that 5-HT4 receptors play a role in the action of SSRIs in animal models. Chronic administration of monoaminergic antidepressants, such as fluoxetine, paroxetine and venlafaxine (but not reboxetine), decreases 5-HT4 receptor density in the basal ganglia and hippocampus in the rat brain (Licht et al., 2009; Vidal, Valdizan, Mostany, Pazos, & Castro, 2009; Vidal, Valdizan, Vilaró, Pazos, & Castro, 2010). In contrast, expression of the 5-HT4 receptor has been shown to be upregu- lated after chronic fluoxetine administration in corticostriatal pyr- amidal cells expressing the adaptor protein p11, located in layer 5 of the cerebral cortex (Schmidt et al., 2012). More direct evidence for a role for the 5-HT4 receptor in the action of conventional antidepressants comes from studies in which 5-HT4 receptor ligands have been shown to modify the effects of monoaminergic antidepressants. For example, adminis- tration of the 5-HT4 receptor agonist RS67333 has been shown to 1114 Susannah E Murphy et al. augment the acute effect of paroxetine on extracellular serotonin levels in the rat ventral hippocampus (Licht et al., 2010). Perhaps the most convincing evidence comes from studies which demon- strate that disrupting 5-HT4 receptor function can abolish the behavioural effects of SSRIs. For example, the 5-HT4 receptor antagonist GR125487 blocks the effects of fluoxetine on a range of depression- and anxiety-related paradigms in a corticosterone model (CORT) of anxiety/depression (Mendez-David et al., 2014), suggesting that 5-HT4 receptor activation is necessary for the antidepressant and anxiolytic effects of SSRIs (although see Cryan and Lucki, 2000). A similar effect is found in 5-HT4 recep- tor knock out mice, where chronic fluoxetine administration has been shown to reverse the behavioural effects of OBX in wild type but not knock out animals (Amigo et al., 2016). Intriguingly, co-administration of 5-HT4 receptor agonists with SSRIs prevents the attenuation in raphe firing that is typically seen at the onset of SSRI administration (Lucas et al., 2010), sug- gesting that 5-HT4 receptor agonists may have promise as an aug- mentation strategy for SSRI treatment. In support of this proposal, co-administration of RS67333 and the SSRI citalopram led to a synergistic inhibition of hippocampal firing, while the same combination potentiated the ability of RS67333 to phos- phorylate CREB in the hippocampus and reduce immobility in the forced swim test (Lucas et al., 2010). ( ) 5-HT4 receptor agonists have also been associated with rapid effects on anxiety. Short-term administration of 5-HT4 receptor agonists has antidepressant and anxiolytic-like effects in animals These effects are consistent with the neurobio- logical adaptations that have been reported following 3 days of 5-HT4 receptor agonism, including desensitisation of 5-HT1A autoreceptors, increased CREB phosphorylation and increased hippocampal cell proliferation (Lucas et al., 2007), and suggest that 5-HT4 receptor agonists have a unique, rapid-acting profile of effects. 5-HT4 receptors are involved in the mechanism of action of SSRIs Seven days administration of RS67333 reduces latency to feed in the novelty suppressed feeding (NSF) paradigm, whereas 14 days administration of fluoxetine is required to see the same effect (Pascual-Brazo et al., 2012). A similar effect was seen in the mouse CORT model of anxiety/depression, where 7 day administration of RS67333, but not fluoxetine, reduced anxiety-related behaviour in the OF test and the elevated plus maze (EPM), although the effects of subchronic RS67333 on the NSF were not replicated in this model (Mendez-David et al., 2014). Interestingly, the effects of RS67333 on the OF and EPM tests were also evident in animals who had undergone hip- pocampal X-irradiation, suggesting that they are not dependent on hippocampal neurogenesis (Mendez-David et al., 2014). g Interestingly, there is also evidence to suggest that the 5-HT4 receptor contributes to the neurotrophic effects of chronic SSRI treatment. For example, the usual increase in cell proliferation in the dentate gyrus following chronic administration of fluoxet- ine is not present in 5-HT4 receptor knock out mice (Imoto et al., 2015), and is partially blocked by 5-HT4 receptor antagonism (Mendez-David et al., 2014), suggesting that the neurogenic effects of SSRIs may be partly mediated by the 5-HT4 receptor. Further support for this comes from a study which demonstrates that the induction of immature like granule cells by chronic fluox- etine treatment (‘dematuration’) is blocked in 5-HT4 receptor knock-out mice (Kobayashi et al., 2010). Consistent with a fast onset of action, acute administration of RS67333 is sufficient to induce an anxiolytic-like effect on the ele- vated zero-maze (Bell, Duke, Gilmore, Page, & Begue, 2014), and reverse the anxiogenic effects of chronic cannabinoid exposure during adolescence (Abboussi et al., 2016). A recent report con- firmed an anxiolytic-like effect of acute systemic infusion of RS67333 on the EPM, the OF test and the NSF test (Faye et al., 2019). Intriguingly, a similar profile was seen when RS67333 was directly infused into the mPFC, and the anxiolytic-like effects of RS67333 were attenuated by optogenetic inhibition of the mPFC terminals in the DRN, suggesting that direct activation of 5-HT4 receptors in the mPFC is responsible for the fast anxio- lytic effects of 5-HT4 receptor agonists, via the mPFC-DRN cir- cuit (Faye et al., 2019). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 5-HT4 Receptor activation has a facilitatory effect on rodent tests of learning and memory y y The timing of 5-HT4 receptor activation seems to be critical to the profile of effects on memory processes, with a particularly clear role for the receptor in learning and memory acquisition (Hagena & Manahan-Vaughan, 2017). Consistent with this, stud- ies in which an agonist is administered prior to training have typ- ically shown a facilitatory effect of 5-HT4 receptor activation (Lamirault & Simon, 2001; Meneses & Hong, 1997; Orsetti, Dellarole, Ferri, & Ghi, 2003), whereas the effect of post-training agonist administration is less clear, with reports of both enhanced (Lamirault & Simon, 2001; Orsetti et al., 2003) and impaired (Meneses & Hong, 1997; Nasehi, Tabatabaie, Khakpai, & Zarrindast, 2015) memory consolidation. ) y Interestingly, the cognition enhancing effects of 5-HT4 recep- tor activation have also been demonstrated in the chronic cortico- sterone model of anxiety/depression (Darcet, Gardier, David, & Guilloux, 2016), suggesting that this may be a promising strategy for treating the cognitive impairments associated with stress- related disorders. Chronic administration of the 5-HT4 receptor agonist RS67333 restored corticosterone-induced deficits on a broad range of tests of learning and memory, in contrast to chronic fluoxetine, which had a much narrower profile of effects. The 5-HT4 receptor agonist RS67333 has also been shown to restore emotional memory performance in a passive avoidance test in the Flinders sensitive line rat model of depression (Eriksson et al., 2012). Together these findings suggest that 5-HT4 receptor agonists may be a potential target for the improvement of the cognitive and emotional processing deficits associated with depression. 5-HT4 Receptor activation has a facilitatory effect on rodent tests of learning and memory A large body of evidence demonstrates a facilitatory effect of 5-HT4 receptor activation on rodent and primate behavioural https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 1115 Psychological Medicine tests of cognition, and in particular learning and memory (Hagena & Manahan-Vaughan, 2017; King et al., 2008). Early studies using relatively non-selective compounds, such as the mixed 5-HT4 receptor agonist/5-HT3 receptor antagonist BIMU1 demonstrated a facilitation of autoshaping (Meneses & Hong, 1997) and associative memory (Marchetti-Gauthier, Roman, Dumuis, Bockaert, & Soumireu-Mourat, 1997), which was attributed to 5-HT4 receptor activation since both effects were blocked by co-administration of a 5-HT4 receptor antagon- ist. Subsequent studies using more selective agonists confirmed that acute 5-HT4 receptor activation improves a wide of range of behavioural tests of learning and memory, including place rec- ognition (Lamirault & Simon, 2001), object recognition (Lamirault & Simon, 2001; Levallet, Hotte, Boulouard, & Dauphin, 2009; Moser et al., 2002), delayed matching-to-sample (Terry et al., 1998), Morris water maze (Lelong, Dauphin, & Boulouard, 2001), delayed spontaneous alternation (Mohler et al., 2007) and olfactory association learning (Marchetti et al., 2004; Marchetti et al., 2011) tasks. The positive effect of 5-HT4 receptor agonism on object recognition is also seen after chronic (14 day) activation of the receptor (Quiedeville et al., 2015). Critically, 5-HT4 receptor antagonism has consistently been shown to block these agonist-induced pro-cognitive effects, con- firming a specific role for the 5-HT4 receptor in learning and memory (Fontana, Daniels, Wong, Clark, & Eglen, 1997; Lamirault & Simon, 2001; Mohler et al., 2007; Moser et al., 2002). Whilst 5-HT4 receptor activation has particularly strong effects on hippocampal-dependent learning and memory, it also facilitates other forms of cognition, including attention (as mea- sured by the five choice serial reaction time task, Hille, Bate, Davis, and Gonzalez, 2008) and social memory (Letty et al., 1997). SDZ205557 and RS67532 have been shown to impair olfactory association memory formation and passive memory formation (Galeotti, Ghelardini, & Bartolini, 1998; Marchetti, Dumuis, Bockaert, Soumireu-Mourat, & Roman, 2000), other studies report no effect of 5-HT4 receptor antagonism on memory processes (Moser et al., 2002; Orsetti et al., 2003). It has been sug- gested that this may be due to a low level of behaviourally- mediated exogenous serotonin release, which would lessen the effects of receptor blockade compared with receptor activation (Orsetti et al., 2003). 5-HT4 Receptor activation has a facilitatory effect on rodent tests of learning and memory Consistent with this, stud- ies in which an agonist is administered prior to training have typ- ically shown a facilitatory effect of 5-HT4 receptor activation (Lamirault & Simon, 2001; Meneses & Hong, 1997; Orsetti, Dellarole, Ferri, & Ghi, 2003), whereas the effect of post-training agonist administration is less clear, with reports of both enhanced (Lamirault & Simon, 2001; Orsetti et al., 2003) and impaired (Meneses & Hong, 1997; Nasehi, Tabatabaie, Khakpai, & Zarrindast, 2015) memory consolidation. p g The memory enhancing properties of 5-HT4 receptor activa- tion are often attributed to a facilitation of central cholinergic activity. This is supported by studies demonstrating that 5-HT4 receptor agonists enhance the brain ACh output under both base- line conditions (Consolo et al., 1994) and during mnemonic demands (Mohler et al., 2007), and that 5-HT4 receptor agonists reverse the cognitive impairments induced by muscarinic antago- nists (Cachard-Chastel et al., 2008; Fontana et al., 1997; Lo et al., 2014; Marchetti-Gauthier et al., 1997; Moser et al., 2002). For example, acute 5-HT4 receptor activation reverses the impair- ments in spatial learning and memory that are induced by atro- pine and scopolamine on the Morris water maze (Fontana et al., 1997; Lo et al., 2014) and the deficits in passive avoidance behaviour that are induced by scopolamine on the passive avoid- ance test (Galeotti et al., 1998; Lo et al., 2014). More specifically, local infusion of the 5-HT4 receptor agonist RS67333 into the NbM enhances the acquisition and consolidation of place recog- nition memory, which has been suggested to be due to a potenti- ation of the activity of the cholinergic NbM-cortical pathways (Orsetti et al., 2003; although see Penas-Cazorla & Vilaro, 2015). Further support for the idea that the memory-enhancing effects of 5-HT4 receptor activation are mediated via an inter- action with cholinergic systems comes from studies that demon- strate that acetylcholinesterase inhibitors (AChEI) potentiate the effects of 5-HT4 agonists on cognition (Cachard-Chastel et al., 2007; Lamirault, Guillou, Thal, & Simon, 2003; Mohler et al., 2007; Moser et al., 2002). For example, sub-efficacious doses of the 5-HT4 receptor agonist prucalopride and the AChEI donepezil reverse scopolamine-induced deficits in spatial learning and memory when given in combination (Cachard-Chastel et al., 2007). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 5-HT4 Receptor activation has a facilitatory effect on rodent tests of learning and memory Likewise, 5-HT4 receptor knock-out mice do not have clear behavioural deficits in learning and memory (Segu et al., 2010). However, these animals do show an impair- ment in long-term spatial memory on the Morris water maze fol- lowing scopolamine administration at a dose ineffective in wild type controls, suggesting that developmental adaptations in the cholinergic system may compensate for the absence of 5-HT4 receptors under baseline conditions (Segu et al., 2010). tests of cognition, and in particular learning and memory (Hagena & Manahan-Vaughan, 2017; King et al., 2008). Early studies using relatively non-selective compounds, such as the mixed 5-HT4 receptor agonist/5-HT3 receptor antagonist BIMU1 demonstrated a facilitation of autoshaping (Meneses & Hong, 1997) and associative memory (Marchetti-Gauthier, Roman, Dumuis, Bockaert, & Soumireu-Mourat, 1997), which was attributed to 5-HT4 receptor activation since both effects were blocked by co-administration of a 5-HT4 receptor antagon- ist. Subsequent studies using more selective agonists confirmed that acute 5-HT4 receptor activation improves a wide of range of behavioural tests of learning and memory, including place rec- ognition (Lamirault & Simon, 2001), object recognition (Lamirault & Simon, 2001; Levallet, Hotte, Boulouard, & Dauphin, 2009; Moser et al., 2002), delayed matching-to-sample (Terry et al., 1998), Morris water maze (Lelong, Dauphin, & Boulouard, 2001), delayed spontaneous alternation (Mohler et al., 2007) and olfactory association learning (Marchetti et al., 2004; Marchetti et al., 2011) tasks. The positive effect of 5-HT4 receptor agonism on object recognition is also seen after chronic (14 day) activation of the receptor (Quiedeville et al., 2015). Critically, 5-HT4 receptor antagonism has consistently been shown to block these agonist-induced pro-cognitive effects, con- firming a specific role for the 5-HT4 receptor in learning and memory (Fontana, Daniels, Wong, Clark, & Eglen, 1997; Lamirault & Simon, 2001; Mohler et al., 2007; Moser et al., 2002). Whilst 5-HT4 receptor activation has particularly strong effects on hippocampal-dependent learning and memory, it also facilitates other forms of cognition, including attention (as mea- sured by the five choice serial reaction time task, Hille, Bate, Davis, and Gonzalez, 2008) and social memory (Letty et al., 1997). The timing of 5-HT4 receptor activation seems to be critical to the profile of effects on memory processes, with a particularly clear role for the receptor in learning and memory acquisition (Hagena & Manahan-Vaughan, 2017). Establishing the effects of 5-HT4 receptor activation in humans Newer, more selective partial agonists, such as prucalo- pride which is licensed for the treatment of constipation, have good cardiac safety profiles. Prucalopride has good brain penetration (Johnson et al., 2012), has pro-cognitive and antidepressant-like effects in animal models (Cachard-Chastel et al., 2007; Lucas et al., 2007) and therefore offers a unique opportunity to investigate the role of the 5-HT4 receptor in human cognition. g We recently explored the effect of a single dose of prucalo- pride in healthy volunteers and found consistent pro-cognitive effects across measures of learning memory including declara- tive memory, reward learning and emotional encoding (Murphy, Wright, Browning, Cowen, & Harmer, 2019). These findings replicate the findings in preclinical studies and sup- port further clinical assessment of these compounds in depres- sion and for disorders involving memory impairment. In line with data in animal models that 5HT4 agonists may be beneficial in depression, and also in combination with an SSRI, we are cur- rently exploring the effects of a 5HT4 partial agonist (PF-04995274) on emotional and cognitive function in unmedi- cated depressed patients (NCT03516604) and patients who have not showed a sufficient response to current SSRI treatment (NCT03515733). This kind of experimental medicine approach can provide early proof-of-concept evidence to support larger clin- ical trials, and may have value in selecting the best treatment, for the right subgroup of depressed patients, prior to full randomised clinical trials of efficacy. ( ) These divergent correlational findings highlight the import- ance of further investigation in humans using an experimental medicine approach. Establishing the cognitive effects of com- pounds targeting the 5-HT4 receptor is particularly important in light of the limited predictive validity of animal models and the disappointing failure to predict the efficacy of pro-cognitive and antidepressant drugs in humans from drug discovery studies in rodents (Hyman, 2012; Millan et al., 2012). We have previously demonstrated that acute administration of clinically active antide- pressants in healthy volunteers and depressed patients produces changes in information processing that can be measured using behavioural and neuroimaging paradigms (Godlewska & Harmer, 2020). Importantly, these changes are predictive of later therapeutic effects and these measures can therefore be used as surrogate markers of clinical effects to screen and under- stand the mechanisms of candidate agents early in development. Establishing the effects of 5-HT4 receptor activation in humans Taken together, this preclinical evidence suggests that the 5-HT4 receptor may be a promising target for the treatment of depres- sion and cognitive disorder in humans. However, there have been relatively few studies in humans to establish the translation of these effects, and to characterise the profile of 5-HT4 receptor activation on human cognition and emotion. There is some initial evidence to support a role for the 5-HT4 receptor in the neurobiology of depression. Healthy individuals with a family history of depression have been shown to have lower 5-HT4 receptor binding in the striatum, and this reduction p In contrast to the clear pro-cognitive effects of 5-HT4 receptor activation, the effect of 5-HT4 receptor antagonism is less well characterised and findings are mixed. Whilst the antagonists 1116 Susannah E Murphy et al. is most pronounced in people with more than one first degree relative with a history of depression (Madsen et al., 2014). Conversely, an increase in the density and functionality of 5-HT4 receptors in cortical and striatal areas has been reported in an analysis of post-mortem brains from depressed violent sui- cide victims (Rosel et al., 2004). Polymorphisms in the splice vari- ant region of the gene encoding the 5-HT4 receptor have been reported to be associated with unipolar depression (Ohtsuki et al., 2002), although this was not confirmed in a recent genome wide association study (Wray et al., 2018). An involvement of the 5-HT4 receptor in human memory is also supported by a PET study that reported an inverse correlation of 5-HT4 receptor bind- ing in the hippocampus and performance on the Auditory Verbal Learning Task (Haahr et al., 2013). Interestingly, the direction of this effect is opposite to the memory enhancing effects of 5-HT4 receptor agonism seen in preclinical studies, which may be due to the binding potential representing a composite measure of recep- tor density and affinity and/or that lower 5-HT4 receptor binding could also be reflective of higher chronic endogenous serotonin levels (Haahr et al., 2014). side effects. Indeed, two 5-HT4 agonists, cisapride and tegaserod, that were developed for the treatment of gastrointestinal problems have both been restricted in use or withdrawn due to concerns about cardiac side effects. However, the particular concern with these drugs was risk of ventricular arrhythmias, which is probably due to their interaction with the human ether-a-go-go related gene potassium channel (De Maeyer, Lefebvre, & Schuurkes, 2008). Conclusions There has been much interest in the psychotropic effects of 5-HT4 receptor activation, stimulated by consistent evidence that 5-HT4 receptor agonists have antidepressant-like and pro-cognitive effects on a range of animal models. However, the translation of these effects to humans has been slow, and, to date, no 5-HT4 receptor agonists have been licensed for any indication other than gastrointestinal disorders. Whilst there had been some con- cern about the safety of 5-HT4 receptor agonists, due to this receptor’s important roles outside of the central nervous system (for example in the heart), prucalopride has proved to be a safe approach to the treatment of chronic constipation. Experimental medicine offers a cost-effective way to efficiently translate these effects into humans and explore the possibility that 5-HT4 recep- tor activation might be a useful adjunct to antidepressant treat- ment, both to speed the onset of clinical antidepressant effects and to target the cognitive symptoms that are not well addressed by current treatments. Experimental medicine often takes a stepwise approach, first testing the effects of the compound in healthy participants to gain information on relevant dosing, duration of treatment and specific cognitive and emotional processes which are sensitive to this manipulation. This information can then be used to guide the most appropriate design for testing in clinical groups includ- ing those with depression, anxiety and cognitive impairment. Again, experimental medicine markers (such as measures of learning or emotional processing) can be useful prior to clinical testing of efficacy (symptom improvement) to characterise likely mechanisms, target and patient groups. Such a stepwise approach may improve decision making about a compound’s development and reduce the high failure rate associated with drug development in this area (Wong, Siah, & Lo, 2019). g Until recently, further elucidation of the role of the 5-HT4 receptor in humans has been difficult because of the lack of suit- able compounds to manipulate 5-HT4 receptor function. The development of 5-HT4 receptor agonists for use as antidepres- sants in humans is complicated by the important role played by the 5-HT4 receptor in the periphery, in particular in the gastro- intestinal tract and the heart, and the potential for peripheral Acknowledgements. The study was funded by the Medical Research Council and supported by the NIHR Oxford Health Biomedical Research Centre and NIHR Oxford Cognitive Health Clinical Research Facility. Establishing the effects of 5-HT4 receptor activation in humans Testing the effects of compounds targeting the 5HT4 receptor in humans using this approach could help clarify the role of the 5HT4 system in cognition and emotion and translate the poten- tially promising results from preclinical models to clinical application. https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press References doi:10.1038/ sj.bjp.0701513. Beliveau, V., Ganz, M., Feng, L., Ozenne, B., Hojgaard, L., Fisher, P. M., … Knudsen, G. M. (2017). A high-resolution in vivo atlas of the human brain’s serotonin system. Journal of Neuroscience, 37(1), 120–128. doi:10.1523/ jneurosci.2830-16.2016. Craig, D. A., & Clarke, D. E. (1990). Pharmacological characterization of a neuronal receptor for 5-hydroxytryptamine in Guinea pig ileum with prop- erties similar to the 5-hydroxytryptamine receptor. Journal of Pharmacology and Experimental Therapeutics, 252(3), 1378–1386. Bell, R., Duke, A. A., Gilmore, P. E., Page, D., & Begue, L. (2014). Anxiolytic-like effects observed in rats exposed to the elevated zero-maze following treatment with 5-HT2/5-HT3/5-HT4 ligands. Scientific Reports, 4, 3881. doi:10.1038/srep03881. Cryan, J. F., & Lucki, I. (2000). 5-HT4 receptors do not mediate the antidepressant-like behavioral effects of fluoxetine in a modified forced swim test. European Journal of Pharmacology, 409(3), 295–299. doi:10.1016/s0014-2999(00)00858-x. Bijak, M., & Misgeld, U. (1997). Effects of serotonin through serotonin1A and serotonin4 receptors on inhibition in the Guinea-pig dentate gyrus in vitro. Neuroscience, 78(4), 1017–1026. doi:10.1016/s0306-4522(96)00666-5. Darcet, F., Gardier, A. M., David, D. J., & Guilloux, J. P. (2016). Chronic 5-HT4 receptor agonist treatment restores learning and memory deficits in a neuroendocrine mouse model of anxiety/depression. Neuroscience Letters, 616, 197–203. doi:10.1016/j.neulet.2016.01.055. Blier, P., Piñeyro, G., el Mansari, M., Bergeron, R., & de Montigny, C. (1998). Role of somatodendritic 5-HT autoreceptors in modulating 5-HT neuro- transmission. Annals of the New York Academy of Sciences, 861, 204–216. Bockaert, J., Claeysen, S., Compan, V., & Dumuis, A. (2004). 5-HT4 receptors. Current Drug Targets: CNS and Neurological Disorders, 3(1), 39–51. De Maeyer, J. H., Lefebvre, R. A., & Schuurkes, J. A. (2008). 5-HT4 Receptor agonists: Similar but not the same. Neurogastroenterology and Motility, 20 (2), 99–112. doi:10.1111/j.1365-2982.2007.01059.x. Bockaert, J., Claeysen, S., Compan, V., & Dumuis, A. (2008). 5-HT(4) receptors: History, molecular pharmacology and brain functions. Neuropharmacology, 55(6), 922–931. doi:10.1016/j.neuropharm.2008.05.013. Dumuis, A., Bouhelal, R., Sebben, M., Cory, R., & Bockaert, J. (1988). A non- classical 5-hydroxytryptamine receptor positively coupled with adenylate cyclase in the central nervous system. Molecular Pharmacology, 34(6), 880–887. Bonaventure, P., Hall, H., Gommeren, W., Cras, P., Langlois, X., Jurzak, M., & Leysen, J. E. (2000). Mapping of serotonin 5-HT(4) receptor mRNA and lig- and binding sites in the post-mortem human brain. Synapse (New York, N.Y.), 36(1), 35–46. doi:10.1002/(sici)1098-2396(200004)36:1<35:: aid-syn4>3.0.co;2-y. Eriksson, T. M., Delagrange, P., Spedding, M., Popoli, M., Mathe, A. A., Ogren, S. O., & Svenningsson, P. (2012). References Castello, J., LeFrancois, B., Flajolet, M., Greengard, P., Friedman, E., & Rebholz, H. (2018). CK2 regulates 5-HT4 receptor signaling and modulates depressive-like behavior. Molecular Psychiatry, 23(4), 872–882. doi:10.1038/ mp.2017.240. Abboussi, O., Said, N., Fifel, K., Lakehayli, S., Tazi, A., & El Ganouni, S. (2016). Behavioral effects of D3 receptor inhibition and 5-HT4 receptor activation on animals undergoing chronic cannabinoid exposure during adolescence. Metabolic Brain Disease, 31(2), 321–327. doi:10.1007/s11011-015-9753-2. Cipriani, A., Furukawa, T. A., Salanti, G., Chaimani, A., Atkinson, L. Z., Ogawa, Y., … Geddes, J. R. (2018). Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: A systematic review and network meta-analysis. Lancet (London, England), 391(10128), 1357–1366. doi:10.1016/ S0140-6736(17)32802-7. Amigo, J., Diaz, A., Pilar-Cuellar, F., Vidal, R., Martin, A., Compan, V., … Castro, E. (2016). The absence of 5-HT4 receptors modulates depression- and anxiety-like responses and influences the response of fluoxetine in olfactory bulbectomised mice: Adaptive changes in hippocampal neuroplas- ticity markers and 5-HT1A autoreceptor. Neuropharmacology, 111, 47–58. doi:10.1016/j.neuropharm.2016.08.037. Compan, V., Daszuta, A., Salin, P., Sebben, M., Bockaert, J., & Dumuis, A. (1996). Lesion study of the distribution of serotonin 5-HT4 receptors in rat basal ganglia and hippocampus. European Journal of Neuroscience, 8 (12), 2591–2598. Artigas, F. (2013). Serotonin receptors involved in antidepressant effects. Pharmacology and Therapeutics, 137(1), 119–131. doi:10.1016/ j.pharmthera.2012.09.006. Bai, M., Zhu, X. Z., Zhang, Y., Zhang, S., Zhang, L., Xue, L., … Zhang, X. (2014). Anhedonia was associated with the dysregulation of hippocampal HTR4 and microRNA Let-7a in rats. Physiology and Behavior, 129, 135– 141. doi:10.1016/j.physbeh.2014.02.035. Conductier, G., Dusticier, N., Lucas, G., Cote, F., Debonnel, G., Daszuta, A., … Compan, V. (2006). Adaptive changes in serotonin neurons of the raphe nuclei in 5-HT(4) receptor knock-out mouse. European Journal of Neuroscience, 24(4), 1053–1062. doi:10.1111/j.1460-9568.2006.04943.x. Barthet, G., Framery, B., Gaven, F., Pellissier, L., Reiter, E., Claeysen, S., … Dumuis, A. (2007). 5-hydroxytryptamine 4 receptor activation of the extra- cellular signal-regulated kinase pathway depends on Src activation but not on G protein or beta-arrestin signaling. Molecular Biology of the Cell, 18(6), 1979–1991. doi:10.1091/mbc.e06-12-1080. Consolo, S., Arnaboldi, S., Giorgi, S., Russi, G., & Ladinsky, H. (1994). 5-HT4 Receptor stimulation facilitates acetylcholine release in rat frontal cortex. Neuroreport, 5(10), 1230–1232. Costall, B., & Naylor, R. J. (1997). The influence of 5-HT2 and 5-HT4 receptor antagonists to modify drug induced disinhibitory effects in the mouse light/ dark test. British Journal of Pharmacology, 122(6), 1105–1118. Conclusions The views expressed are those of the authors and not necessarily those of the National Health Service (NHS), the NIHR or the Department of Health. The research materials supporting this publication can be accessed by contact- ing the corresponding author. 1117 Psychological Medicine to reverse scopolamine-induced memory deficit in C57Bl/6j mice. Behavioural Brain Research, 187(2), 455–461. doi:10.1016/j.bbr.2007.10.008. to reverse scopolamine-induced memory deficit in C57Bl/6j mice. Behavioural Brain Research, 187(2), 455–461. doi:10.1016/j.bbr.2007.10.008. Disclosures. CJH has received consultancy fees from P1vital Ltd., Janssen Pharmaceuticals, Sage Therapeutics, Pfizer and Lundbeck. SEM has received consultancy fees from P1vital Ltd. and Janssen Pharmaceuticals. CJH and SEM hold grant income from UCB Pharma and Janssen Pharmaceuticals. CJH, SEM and PJC hold grant income from a collaborative research project with Pfizer. ANdC is funded by a Wellcome Trust Clinical Doctoral Research Fellowship. Disclosures. CJH has received consultancy fees from P1vital Ltd., Janssen Pharmaceuticals, Sage Therapeutics, Pfizer and Lundbeck. SEM has received consultancy fees from P1vital Ltd. and Janssen Pharmaceuticals. CJH and SEM hold grant income from UCB Pharma and Janssen Pharmaceuticals. CJH, SEM and PJC hold grant income from a collaborative research project with Pfizer. ANdC is funded by a Wellcome Trust Clinical Doctoral Research Fellowship. Cachard-Chastel, M., Lezoualc’h, F., Dewachter, I., Deloménie, C., Croes, S., Devijver, H., … Gardier, A. M. (2007). 5-HT4 receptor agonists increase sAPPalpha levels in the cortex and hippocampus of male C57BL/6j mice. British Journal of Pharmacology, 150(7), 883–892. doi:10.1038/sj.bjp.0707178. Cai, X., Flores-Hernandez, J., Feng, J., & Yan, Z. (2002). Activity-dependent bidirectional regulation of GABA(A) receptor channels by the 5-HT(4) receptor-mediated signalling in rat prefrontal cortical pyramidal neurons. Journal of Physiology, 540(Pt 3), 743–759. https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press References 5-HT4 receptors improve social olfactory memory in the rat. Neuropharmacology, 36(4-5), 681–687. Hagena, H., & Manahan-Vaughan, D. (2017). The serotonergic 5-HT4 recep- tor: A unique modulator of hippocampal synaptic information processing and cognition. Neurobiology of Learning and Memory, 138, 145–153. doi:10.1016/j.nlm.2016.06.014. Levallet, G., Hotte, M., Boulouard, M., & Dauphin, F. (2009). Increased par- ticulate phosphodiesterase 4 in the prefrontal cortex supports 5-HT4 receptor-induced improvement of object recognition memory in the rat. Psychopharmacology, 202(1–3), 125–139. doi:10.1007/s00213-008-1283-8. Hille, C., Bate, S., Davis, J., & Gonzalez, M. I. (2008). 5-HT4 Receptor agonism in the five-choice serial reaction time task. Behavioural Brain Research, 195 (1), 180–186. doi:10.1016/j.bbr.2008.08.007. Hyman, S. E. (2012). Revolution stalled. Science Translational Medicine, 4 (155), 155cm111. doi:10.1126/scitranslmed.3003142. Licht, C. L., Kirkegaard, L., Zueger, M., Chourbaji, S., Gass, P., Aznar, S., & Knudsen, G. M. (2010). Changes in 5-HT4 receptor and 5-HT transporter binding in olfactory bulbectomized and glucocorticoid receptor heterozy- gous mice. Neurochemistry International, 56(4), 603–610. doi:10.1016/ j.neuint.2010.01.003. Idres, S., Delarue, C., Lefebvre, H., & Vaudry, H. (1991). Benzamide derivatives provide evidence for the involvement of a 5-HT4 receptor type in the mech- anism of action of serotonin in frog adrenocortical cells. Brain Research: Molecular Brain Research, 10(3), 251–258. doi:10.1016/0169-328x(91) 90068-9. Licht, C. L., Marcussen, A. B., Wegener, G., Overstreet, D. H., Aznar, S., & Knudsen, G. M. (2009). The brain 5-HT4 receptor binding is down- regulated in the flinders sensitive line depression model and in response to paroxetine administration. Journal of Neurochemistry, 109(5), 1363– 1374. doi:10.1111/j.1471-4159.2009.06050.x. Imoto, Y., Kira, T., Sukeno, M., Nishitani, N., Nagayasu, K., Nakagawa, T., … Segi-Nishida, E. (2015). Role of the 5-HT4 receptor in chronic fluoxetine treatment-induced neurogenic activity and granule cell dematuration in the dentate gyrus. Molecular Brain, 8, 29. doi:10.1186/s13041-015-0120-3. Lo, A. C., De Maeyer, J. H., Vermaercke, B., Callaerts-Vegh, Z., Schuurkes, J. A., & D’Hooge, R. (2014). SSP-002392, a new 5-HT4 receptor agonist, dose- dependently reverses scopolamine-induced learning and memory impair- ments in C57Bl/6 mice. Neuropharmacology, 85, 178–189. doi:10.1016/ j.neuropharm.2014.05.013. Ishizuka, T., Goshima, H., Ozawa, A., & Watanabe, Y. (2014). Stimulation of 5-HT4 receptor enhances differentiation of mouse induced pluripotent stem cells into neural progenitor cells. Clinical and Experimental Pharmacology and Physiology, 41(5), 345–350. doi:10.1111/1440-1681. 12224. Lucas, G., Compan, V., Charnay, Y., Neve, R. L., Nestler, E. J., Bockaert, J., … Debonnel, G. (2005). Frontocortical 5-HT4 receptors exert positive feed- back on serotonergic activity: Viral transfections, subacute and chronic treatments with 5-HT4 agonists. References Emotional memory impairments in a genetic rat model of depression: Involvement of 5-HT/MEK/Arc sig- naling in restoration. Molecular Psychiatry, 17(2), 173–184. doi:10.1038/ mp.2010.131. Bonhomme, N., De Deurwaerdere, P., Le Moal, M., & Spampinato, U. (1995). Evidence for 5-HT4 receptor subtype involvement in the enhancement of striatal dopamine release induced by serotonin: A microdialysis study in the halothane-anesthetized rat. Neuropharmacology, 34(3), 269–279. doi:10.1016/0028-3908(94)00145-i. Faye, C., Hen, R., Guiard, B., Denny, C., Gardier, A., Mendez-David, I., & David, D. (2019). Rapid anxiolytic effects of RS67333, a serotonin type 4 receptor agonist, and diazepam, a benzodiazepine, are mediated by projections from the prefrontal cortex to the dorsal raphe nucleus. Biological Psychiatry 87(6):514-525. doi:https://doi.org/ 10.1016/j.biopsych.2019.08.009. Burghardt, N. S., Bush, D. E., McEwen, B. S., & LeDoux, J. E. (2007). Acute selective serotonin reuptake inhibitors increase conditioned fear expression: Blockade with a 5-HT(2C) receptor antagonist. Biological Psychiatry, 62 (10), 1111–1118. doi:10.1016/j.biopsych.2006.11.023. Fontana, D. J., Daniels, S. E., Wong, E. H., Clark, R. D., & Eglen, R. M. (1997). The effects of novel, selective 5-hydroxytryptamine (5-HT)4 receptor ligands in rat spatial navigation. Neuropharmacology, 36(4-5), 689–696. Cachard-Chastel, M., Devers, S., Sicsic, S., Langlois, M., Lezoualc’h, F., Gardier, A. M., & Belzung, C. (2008). Prucalopride and donepezil act synergistically https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press Susannah E Murphy et al. 1118 Frazer, A., & Benmansour, S. (2002). Delayed pharmacological effects of anti- depressants. Molecular Psychiatry, 7 (Suppl 1), S23–S28. doi:10.1038/ sj.mp.4001015. Kemp, A., & Manahan-Vaughan, D. (2005). The 5-hydroxytryptamine4 recep- tor exhibits frequency-dependent properties in synaptic plasticity and behavioural metaplasticity in the hippocampal CA1 region in vivo. Cerebral Cortex, 15(7), 1037–1043. doi:10.1093/cercor/bhh204. Galeotti, N., Ghelardini, C., & Bartolini, A. (1998). Role of 5-HT4 receptors in the mouse passive avoidance test. Journal of Pharmacology and Experimental Therapeutics, 286(3), 1115–1121. Kennett, G. A., Bright, F., Trail, B., Blackburn, T. P., & Sanger, G. J. (1997). Anxiolytic-like actions of the selective 5-HT4 receptor antagonists SB 204070A and SB 207266A in rats. Neuropharmacology, 36(4-5), 707–712. doi:10.1016/s0028-3908(97)00037-3. Ge, J., & Barnes, N. M. (1996). 5-HT4 receptor-mediated modulation of 5-HT release in the rat hippocampus in vivo. British Journal of Pharmacology, 117 (7), 1475–1480. doi:10.1111/j.1476-5381.1996.tb15309.x. King, M. V., Marsden, C. A., & Fone, K. C. (2008). A role for the 5-HT(1A), 5-HT4 and 5-HT6 receptors in learning and memory. Trends in Pharmacological Sciences, 29(9), 482–492. Godlewska, B. R., & Harmer, C. J. (2020). References Cognitive neuropsychological theory of antidepressant action: A modern-day approach to depression and its treatment. Psychopharmacology. doi:10.1007/s00213-019-05448-0. Kobayashi, K., Ikeda, Y., Sakai, A., Yamasaki, N., Haneda, E., Miyakawa, T., & Suzuki, H. (2010). Reversal of hippocampal neuronal maturation by seroto- nergic antidepressants. Proceedings of the National Academy of Sciences of the United States of America, 107(18), 8434–8439. doi:10.1073/ pnas.0912690107. Goethe, J. W., Woolley, S. B., Cardoni, A. A., Woznicki, B. A., & Piez, D. A. (2007). Selective serotonin reuptake inhibitor discontinuation: Side effects and other factors that influence medication adherence. Journal of Clinical Psychopharmacology, 27(5), 451–458. doi:10.1097/jcp.0b013e31815152a5. Gomez-Lazaro, E., Garmendia, L., Beitia, G., Perez-Tejada, J., Azpiroz, A., & Arregi, A. (2012). Effects of a putative antidepressant with a rapid onset of action in defeated mice with different coping strategies. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 38(2), 317–327. doi:10.1016/j.pnpbp.2012.04.019. Kulla, A., & Manahan-Vaughan, D. (2002). Modulation by serotonin 5-HT(4) receptors of long-term potentiation and depotentiation in the dentate gyrus of freely moving rats. Cerebral Cortex, 12(2), 150–162. doi:10.1093/cercor/ 12.2.150. Lamirault, L., Guillou, C., Thal, C., & Simon, H. (2003). Combined treatment with galanthaminium bromide, a new cholinesterase inhibitor, and RS 67333, a partial agonist of 5-HT4 receptors, enhances place and object rec- ognition in young adult and old rats. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 27(1), 185–195. doi:10.1016/s0278-5846(02) 00351-2. Goodwin, G. M., Price, J., De Bodinat, C., & Laredo, J. (2017). Emotional blunt- ing with antidepressant treatments: A survey among depressed patients. Journal of Affective Disorders, 221, 31–35. doi:10.1016/j.jad.2017.05.048. Haahr, M. E., Fisher, P., Holst, K., Madsen, K., Jensen, C. G., Marner, L., & Hasselbalch, S. (2013). The 5-HT4 receptor levels in hippocampus corre- lates inversely with memory test performance in humans. Human Brain Mapping, 34(11), 3066–3074. doi:10.1002/hbm.22123. Lamirault, L., & Simon, H. (2001). Enhancement of place and object recogni- tion memory in young adult and old rats by RS 67333, a partial agonist of 5-HT4 receptors. Neuropharmacology, 41(7), 844–853. Haahr, M. E., Fisher, P. M., Jensen, C. G., Frokjaer, V. G., Mahon, B. M., Madsen, K., & Knudsen, G. M. (2014). Central 5-HT4 receptor binding as biomarker of serotonergic tonus in humans: A [11C]SB207145 PET study. Molecular Psychiatry, 19(4), 427–432. doi:10.1038/mp.2013.147. Lelong, V., Dauphin, F., & Boulouard, M. (2001). RS 67333 And D-cycloserine accelerate learning acquisition in the rat. Neuropharmacology, 41(4), 517–522. Letty, S., Child, R., Dumuis, A., Pantaloni, A., Bockaert, J., & Rondouin, G. (1997). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press References Biological Psychiatry, 57(8), 918–925. doi:10.1016/j.biopsych.2004.12.023. Johnson, D. E., Drummond, E., Grimwood, S., Sawant-Basak, A., Miller, E., Tseng, E., … Siok, C. J. (2012). The 5-hydroxytryptamine4 receptor agonists prucalopride and PRX-03140 increase acetylcholine and histamine levels in the rat prefrontal cortex and the power of stimulated hippocampal theta oscillations. Journal of Pharmacology and Experimental Therapeutics, 341 (3), 681–691. doi:10.1124/jpet.112.192351. Lucas, G., & Debonnel, G. (2002). 5-HT4 Receptors exert a frequency-related facilitatory control on dorsal raphé nucleus 5-HT neuronal activity. European Journal of Neuroscience, 16(5), 817–822. Kanzari, A., Bourcier-Lucas, C., Freyssin, A., Abrous, D. N., Haddjeri, N., & Lucas, G. (2018). Inducing a long-term potentiation in the dentate gyrus is sufficient to produce rapid antidepressant-like effects. Molecular Psychiatry, 23(3), 587–596. doi:10.1038/mp.2017.94. Lucas, G., Du, J., Romeas, T., Mnie-Filali, O., Haddjeri, N., Piñeyro, G., & Debonnel, G. (2010). Selective serotonin reuptake inhibitors potentiate the rapid antidepressant-like effects of serotonin4 receptor agonists in the Rat. PLoS ONE, 5(2):e9253. Kaumann, A. J. (1990). Piglet sinoatrial 5-HT receptors resemble human atrial 5-HT4-like receptors. Naunyn-Schmiedebergs Archives of Pharmacology, 342(5), 619–622. doi:10.1007/bf00169055. Lucas, G., Rymar, V. V., Du, J., Mnie-Filali, O., Bisgaard, C., Manta, S., … Debonnel, G. (2007). Serotonin(4) (5-HT(4)) receptor agonists are putative https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press 1119 Psychological Medicine antidepressants with a rapid onset of action. Neuron, 55(5), 712–725. doi:10.1016/j.neuron.2007.07.041. Ohtsuki, T., Ishiguro, H., Detera-Wadleigh, S. D., Toyota, T., Shimizu, H., Yamada, K., … Arinami, T. (2002). Association between serotonin 4 recep- tor gene polymorphisms and bipolar disorder in Japanese case-control sam- ples and the NIMH genetics initiative bipolar pedigrees. Molecular Psychiatry, 7(9), 954–961. doi:10.1038/sj.mp.4001133. Madsen, K., Torstensen, E., Holst, K. K., Haahr, M. E., Knorr, U., Frokjaer, V. G., … Knudsen, G. M. (2014). Familial risk for major depression is asso- ciated with lower striatal 5-HT(4) receptor binding. International Journal of Neuropsychopharmacology, 18(1) pii: pyu034. doi:10.1093/ijnp/pyu034. Orsetti, M., Dellarole, A., Ferri, S., & Ghi, P. (2003). Acquisition, retention, and recall of memory after injection of RS67333, a 5-HT(4) receptor agonist, into the nucleus basalis magnocellularis of the rat. Learning and Memory, 10(5), 420–426. doi:10.1101/lm.67303. Malberg, J. E., Eisch, A. J., Nestler, E. J., & Duman, R. S. (2000). Chronic anti- depressant treatment increases neurogenesis in adult rat hippocampus. Journal of Neuroscience, 20(24), 9104–9110. Marchetti-Gauthier, E., Roman, F. S., Dumuis, A., Bockaert, J., & Soumireu-Mourat, B. (1997). BIMU1 Increases associative memory in rats by activating 5-HT4 receptors. Neuropharmacology, 36(4–5), 697–706. References Pascual-Brazo, J., Castro, E., Díaz, A., Valdizán, E. M., Pilar-Cuéllar, F., Vidal, R., … Pazos, A. (2012). Modulation of neuroplasticity pathways and antidepressant-like behavioural responses following the short-term (3 and 7 days) administration of the 5-HT₄ receptor agonist RS67333. International Journal of Neuropsychopharmacology, 15(5), 631–643. doi:10.1017/S1461145711000782. Marchetti, E., Chaillan, F. A., Dumuis, A., Bockaert, J., Soumireu-Mourat, B., & Roman, F. S. (2004). Modulation of memory processes and cellular excit- ability in the dentate gyrus of freely moving rats by a 5-HT4 receptors par- tial agonist, and an antagonist. Neuropharmacology, 47(7), 1021–1035. doi:10.1016/j.neuropharm.2004.06.033. Penas-Cazorla, R., & Vilaro, M. T. (2015). Serotonin 5-HT4 receptors and forebrain cholinergic system: Receptor expression in identified cell popula- tions. Brain Structure & Function, 220(6), 3413–3434. doi:10.1007/ s00429-014-0864-z. Marchetti, E., Dumuis, A., Bockaert, J., Soumireu-Mourat, B., & Roman, F. S. (2000). Differential modulation of the 5-HT(4) receptor agonists and antag- onist on rat learning and memory. Neuropharmacology, 39(11), 2017–2027. doi:10.1016/s0028-3908(00)00038-1. Quiedeville, A., Boulouard, M., Hamidouche, K., Da Silva Costa-Aze, V., Nee, G., Rochais, C., & Bouet, V. (2015). Chronic activation of 5-HT4 receptors or blockade of 5-HT6 receptors improve memory performances. Behavioural Brain Research, 293, 10–17. doi:10.1016/j.bbr.2015.07.020. Marchetti, E., Jacquet, M., Escoffier, G., Miglioratti, M., Dumuis, A., Bockaert, J., & Roman, F. S. (2011). Enhancement of reference memory in aged rats by specific activation of 5-HT(4) receptors using an olfactory associative dis- crimination task. Brain Research, 1405, 49–56. doi:10.1016/ j.brainres.2011.06.020. Rosel, P., Arranz, B., Urretavizcaya, M., Oros, M., San, L., & Navarro, M. A. (2004). Altered 5-HT2A and 5-HT4 postsynaptic receptors and their intra- cellular signalling systems IP3 and cAMP in brains from depressed violent suicide victims. Neuropsychobiology, 49(4), 189–195. doi:10.1159/ 000077365. Medhurst, A. D., Lezoualc’h, F., Fischmeister, R., Middlemiss, D. N., & Sanger, G. J. (2001). Quantitative mRNA analysis of five C-terminal splice variants of the human 5-HT4 receptor in the central nervous system by TaqMan real time RT-PCR. Brain Research: Molecular Brain Research, 90(2), 125–134. Roychowdhury, S., Haas, H., & Anderson, E. G. (1994). 5-HT1A and 5-HT4 receptor colocalization on hippocampal pyramidal cells. Neuropharmacology, 33(3–4), 551–557. Mendez-David, I., David, D. J., Darcet, F., Wu, M. V., Kerdine-Römer, S., Gardier, A. M., & Hen, R. (2014). Rapid anxiolytic effects of a 5-HT₄ recep- tor agonist are mediated by a neurogenesis-independent mechanism. Neuropsychopharmacology, 39(6), 1366–1378. doi:10.1038/npp.2013.332. Rush, A. J., Warden, D., Wisniewski, S. R., Fava, M., Trivedi, M. H., Gaynes, B. N., & Nierenberg, A. A. (2009). References STAR*D: Revising conventional wisdom. CNS Drugs, 23(8), 627–647. doi:10.2165/00023210-200923080-00001. Meneses, A., & Hong, E. (1997). Effects of 5-HT4 receptor agonists and antagonists in learning. Pharmacology, Biochemistry and Behavior, 56(3), 347–351. doi:10.1016/s0091-3057(96)00224-9. Samuels, B. A., Mendez-David, I., Faye, C., David, S. A., Pierz, K. A., Gardier, A. M., … David, D. J. (2016). Serotonin 1A and serotonin 4 receptors: Essential mediators of the neurogenic and behavioral actions of antidepres- sants. Neuroscientist, 22(1), 26–45. doi:10.1177/1073858414561303. Millan, M. J., Agid, Y., Brune, M., Bullmore, E. T., Carter, C. S., Clayton, N. S., … Young, L. J. (2012). Cognitive dysfunction in psychiatric disorders: Characteristics, causes and the quest for improved therapy. Nature Reviews. Drug Discovery, 11(2), 141–168. doi:10.1038/nrd3628. Schmidt, E. F., Warner-Schmidt, J. L., Otopalik, B. G., Pickett, S. B., Greengard, P., & Heintz, N. (2012). Identification of the cortical neurons that mediate antidepressant responses. Cell, 149(5), 1152–1163. doi:10.1016/ j.cell.2012.03.038. Mitchell, A. J. (2006). Two-week delay in onset of action of antidepressants: New evidence. British Journal of Psychiatry, 188, 105–106. England. Segu, L., Lecomte, M. J., Wolff, M., Santamaria, J., Hen, R., Dumuis, A., … Compan, V. (2010). Hyperfunction of muscarinic receptor maintains long- term memory in 5-HT4 receptor knock-out mice. PLoS ONE, 5(3), e9529. doi:10.1371/journal.pone.0009529. Moha ou Maati, H., Bourcier-Lucas, C., Veyssiere, J., Kanzari, A., Heurteaux, C., Borsotto, M., … Lucas, G. (2016). The peptidic antidepressant spadin interacts with prefrontal 5-HT(4) and mGluR(2) receptors in the control of serotonergic function. Brain Structure & Function, 221(1), 21–37. doi:10.1007/s00429-014-0890-x. Shilyansky, C., Williams, L. M., Gyurak, A., Harris, A., Usherwood, T., & Etkin, A. (2016). Effect of antidepressant treatment on cognitive impair- ments associated with depression: A randomised longitudinal study. The Lancet. Psychiatry, 3(5), 425–435. doi:10.1016/s2215-0366(16)00012-2. Mohler, E. G., Shacham, S., Noiman, S., Lezoualc’h, F., Robert, S., Gastineau, M., … Ragozzino, M. E. (2007). VRX-03011, a novel 5-HT4 agonist, enhances memory and hippocampal acetylcholine efflux. Neuropharmacology, 53(4), 563–573. doi:10.1016/j.neuropharm.2007.06.016. Silvestre, J. S., Fernandez, A. G., & Palacios, J. M. (1996). Effects of 5-HT4 receptor antagonists on rat behaviour in the elevated plus-maze test. European Journal of Pharmacology, 309(3), 219–222. doi:10.1016/ 0014-2999(96)00457-8. Moser, P. C., Bergis, O. E., Jegham, S., Lochead, A., Duconseille, E., Terranova, J. P., … Scatton, B. (2002). SL65.0155, A novel 5-hydroxytryptamine(4) receptor partial agonist with potent cognition-enhancing properties. Journal of Pharmacology and Experimental Therapeutics, 302(2), 731–741. doi:10.1124/jpet.102.034249. Siniscalchi, A., Badini, I., Beani, L., & Bianchi, C. (1999). https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press References 5-HT4 Receptor modulation of acetylcholine outflow in Guinea pig brain slices. Neuroreport, 10(3), 547–551. Murphy, S. E., Wright, L. C., Browning, M., Cowen, P. J., & Harmer, C. J. (2019). A role for 5-HT4 receptors in human learning and memory. Psychological Medicine, 1–9. doi:10.1017/s0033291719002836. Terry, A. V., Jr., Buccafusco, J. J., Jackson, W. J., Prendergast, M. A., Fontana, D. J., Wong, E. H., … Eglen, R. M. (1998). Enhanced delayed matching per- formance in younger and older macaques administered the 5-HT4 receptor agonist, RS 17017. Psychopharmacology, 135(4), 407–415. doi:10.1007/ s002130050529. Nasehi, M., Tabatabaie, M., Khakpai, F., & Zarrindast, M. R. (2015). The effects of CA1 5HT4 receptors in MK801-induced amnesia and hyperloco- motion. Neuroscience Letters, 587, 73–78. doi:10.1016/j.neulet.2014.12.019. Tonini, M., & Candura, S. M. (1996). 5-HT4 Receptor agonists and bladder disorders. Trends in Pharmacological Sciences, 17, 314–316. England. Norman, E. D., Thiels, E., Barrionuevo, G., & Klann, E. (2000). Long-term depression in the hippocampus in vivo is associated with protein phosphatase-dependent alterations in extracellular signal-regulated kinase. Journal of Neurochemistry, 74(1), 192–198. doi:10.1046/ j.1471-4159.2000.0740192.x. Trivedi, M. H., Fava, M., Wisniewski, S. R., Thase, M. E., Quitkin, F., Warden, D., … Rush, A. J. (2006). Medication augmentation after the failure of SSRIs for depression. New England Journal of Medicine, 354(12), 1243–1252. doi:10.1056/NEJMoa052964. https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press Susannah E Murphy et al. 1120 neurotoxic lesions. Journal of Comparative Neurology, 484(4), 418–439. doi:10.1002/cne.20447. Twarkowski, H., Hagena, H., & Manahan-Vaughan, D. (2016). The 5-hydro- xytryptamine4 receptor enables differentiation of informational content and encoding in the hippocampus. Hippocampus, 26(7), 875–891. doi:10.1002/hipo.22569. Waeber, C., Sebben, M., Nieoullon, A., Bockaert, J., & Dumuis, A. (1994). Regional distribution and ontogeny of 5-HT4 binding sites in rodent brain. Neuropharmacology, 33(3–4), 527–541. doi:10.1016/0028-3908(94) 90084-1. Varnas, K., Halldin, C., Pike, V. W., & Hall, H. (2003). Distribution of 5-HT4 receptors in the postmortem human brain – an autoradiographic study using [125I]SB 207710. European Neuropsychopharmacology, 13(4), 228–234. Warner-Schmidt, J. L., & Duman, R. S. (2006). Hippocampal neurogenesis: Opposing effects of stress and antidepressant treatment. Hippocampus, 16 (3), 239–249. doi:10.1002/hipo.20156. Vidal, R., Valdizan, E. M., Mostany, R., Pazos, A., & Castro, E. (2009). Long-term treatment with fluoxetine induces desensitization of 5-HT4 receptor-dependent signalling and functionality in rat brain. Journal of Neurochemistry, 110(3), 1120–1127. doi:10.1111/j.1471-4159.2009.06210.x. Wong, C. H., Siah, K. W., & Lo, A. W. (2019). Estimation of clinical trial suc- cess rates and related parameters. Biostatistics, 20(2), 273–286. https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press References World Health Organisation (2017). Depression and other common mental disorders. Global Health Estimates. Vidal, R., Valdizan, E., Vilaró, M., Pazos, A., & Castro, E. (2010). Reduced sig- nal transduction by 5-HT4 receptors after long-term venlafaxine treatment in rats. British Journal of Pharmacology, 161(3), 695–706. doi:10.1111/ j.1476-5381.2010.00903.x. Wray, N. R., Ripke, S., Mattheisen, M., Trzaskowski, M., Byrne, E. M., Abdellaoui, A., … Consortium, M. D. D. W. G. O. T. P. G. (2018). Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major depression. Nature Genetics, 50(5), 668–681. doi:10.1038/s41588-018-0090-3. Vilaró, M. T., Cortés, R., & Mengod, G. (2005). Serotonin 5-HT4 receptors and their mRNAs in rat and Guinea pig brain: Distribution and effects of https://doi.org/10.1017/S0033291720000604 Published online by Cambridge University Press
W4323074714.txt	https://drpress.org/ojs/index.php/fbem/article/download/5598/5412	en		Research on Digital Maturity of Enterprises in Yangtze River Delta	Frontiers in business, economics and management	2,023	cc-by	2,921	Frontiers in Business, Economics and Management ISSN: 2766-824X \| Vol. 7, No. 3, 2023 Research on Digital Maturity of Enterprises in Yangtze River Delta Long Yang, Yue Hu Information management department, School of management science and engineering, Anhui University of Finance and economics, Bengbu, Anhui 233000, China Abstract: in recent years, the digital economy market has a very broad prospect, and its vigorous development in recent years has also proved the importance of digital transformation, and the digital road is the only way for enterprises to realize the due economic laws. The digital transformation of enterprises in the Yangtze River Delta is an important means for them to enter the digital economy market. The digital economy is not only conducive to eliminating the collaborative obstacles of the development of enterprises in the Yangtze River Delta, but its huge population base, market scale and economic volume also provide the basic and necessary conditions for the development of regional digital economy. This study is more targeted to study the Yangtze River Delta region, which has considerable research value. At the same time, it actively responds to China's strategy of building a "Digital China". Based on the actual situation of the Yangtze River Delta region, it evaluates the digital maturity of enterprises, and promotes the research on the shortcomings and advantages of the digital transformation of enterprises, which is conducive to promoting the improvement of the operation efficiency of enterprises in this region to a great extent, And it can effectively drive enterprises in other regions to carry out digital transformation, provide reference and experience for enterprises in other regions, and achieve win-win results between enterprises, and generate synergy benefits. Key words: Yangtze River Delta; Digitalization; Enterprise digital maturity; Evaluation index. 1. Introduction reflect the digital maturity of Chinese enterprises to a certain extent. Nowadays, the era is in the era of great change in the transition to the digital economy, and all kinds of cutting-edge information technologies are accelerating the iteration. These cutting-edge information technologies have greatly reduced the costs of the three elements of links, information and computing required by enterprise operations. The reduction of factor costs is bound to bring great changes to various industries in enterprises. Digital economy is becoming the core element of global industrial change and economic growth, and is also a new type of production factor, which drives the original factors to produce new value. The digital transformation of enterprises has also become a bridge for enterprises to transition from industrial economy to digital economy. Boston Corporation's "digital drive: a thousand miles a day" in 2018 showed that nearly 60% of the world's top 100 enterprises are digital enterprises. Compared with 2012, only 17% of enterprises apply digital technology extensively [1]. The Yangtze River Delta region is geographically connected and culturally interlinked. It is the most active region for small and medium-sized enterprises in China and one of the most economically developed regions in China. From the perspective of national strategy, the Yangtze River Delta region also occupies an important position in the national modernization drive and is an important engine of economic development. The enterprise construction in the Yangtze River Delta region can also reflect the overall enterprise outlook of China. From an international perspective, the essence of international competition is a comprehensive competition based on economy and technology. For enterprises in the Yangtze River Delta region, digital transformation can not only improve efficiency and reduce costs, but also enhance the country's international competitiveness [2]. Selecting enterprises in the Yangtze River Delta as the research object of digital maturity can 2. Research Status of Enterprise Digital Maturity The digital transformation, upgrading, maturity of enterprises and the development level of national digital economy have been the hot topics of scholars' discussion and research since the 14th five year plan. According to the research object and content, the relevant research can be divided into three categories. The first category is the definition discussion, trigger mechanism and Policy Research of digital transformation at home and abroad. For example, George Westerman believes that digital transformation is the use of technology by enterprises to completely change the operation efficiency and performance of enterprises, and provide a reliable basis for the company's decision-making; Gregory vial (2019) believes that digital transformation should be seen as a process that will trigger those seeking to change their value creation path strategies [3]. Zhang Yi (2021) and lishuqin (2020) studied the digital transformation of enterprises from the perspective of policies at home and abroad [4-5]. The second is to study the digital transformation capability / maturity evaluation model of a specific industry in China by using intelligent analysis methods. For example, Wang Rui, Dongming and houwenhao (2019) built the digital maturity evaluation model of manufacturing enterprises from four dimensions based on the analytic hierarchy process (AHP) - decision test and Evaluation Laboratory (DEMATEL) method [6]; Xujinghan (2020) constructed the evaluation index system of digital transformation ability of manufacturing enterprises from the perspective of technological change, organizational change and management change, and verified the applicability of the evaluation index system by applying triangular fuzzy number and its closeness degree, and grey fixed weight clustering 315 different to some extent. Although there is no overall definition to summarize all aspects of digital transformation, the influencing factors of digital transformation are relatively fixed. According to the current situation of global digital transformation development, the influencing factors of digital transformation from any perspective can be summarized as five aspects: digital strategy and ecology, digital organization and management, digital technology capability, digital business and performance, and digital talents. This paper studies the digital maturity of enterprises in the Yangtze River Delta from these five aspects. "Digital strategy and ecology" can be divided into two modules: digital strategy and digital ecosystem. The concept of the former is relatively large, including both its own content and its relationship with other strategies. From the content point of view, it is mainly a collection of several different types of digital strategies. From the relationship point of view, The most important is the relationship between the general strategy of enterprise development; The latter is mainly built from internal and external aspects, so this paper uses seven indicators to measure the number of short-term and long-term digital strategies, the proportion of personalized strategies for customers, sustainability, the percentage of integration with the overall strategy, the degree of internal efficient collaboration, the number of external affiliated enterprises, and the number of external active customers. Digital organization and management" is one of the cores of enterprise digital transformation. In the process of digital transformation, the enterprise organization mode is constantly developing towards liquid organization and intelligent organization, and the main factors to realize these two modes are members, architecture and culture; The digital organization framework must be combined with intelligent management to play its significant advantages, and the core elements of intelligent management are team, system and decision-making; Therefore, this paper uses seven indicators to measure the degree of "comprehensive data", "the percentage of all staff sharing and governance", "the percentage of member awareness and self driving", "the degree of digital culture construction", "the digital leadership of management team", "the application of intelligent management" and "the efficiency of scientific decisionmaking". Digital technology capability" is an important scale to reflect the digital level of enterprises. The foundation of technology is the support of equipment, and the guarantee of technology is the establishment of security system. The technological breakthroughs and achievements are directly related to the investment of enterprises. This paper uses "it infrastructure", "it system architecture capability", "digital security system soundness", "security protection level", "number of digital technologies" "Digital technology R & D investment" and "number of digital products" are seven indicators to measure. Digital business and performance" refers to the process of business upgrading and income growth brought by digital transformation to enterprises. In this process, no matter how the enterprise business is expanded, it is closely surrounded by two core contents: efficiency and revenue. Efficiency is the method and revenue is the result. Seven indicators are used to measure it, including "percentage of AI supported business", "continuous realization of business value", "process efficiency", "number of digital channels", "proportion of method [7]. The third category is quantitative research on enterprise digital transformation and suggestions and Countermeasures for accelerating development. Enterprise digital transformation is a new exploration process. Scholars use various analysis methods and quantitative indicators, combined with the current development status of enterprise digital transformation in China, put forward methods and Countermeasures Based on actual and huge benefits, and find an optimal development path. The above research can help people comprehensively understand the enterprise digital transformation and its evaluation system from different perspectives. However, these studies do not cover all industries, and from a national perspective, the digital transformation of enterprises does not take geographical factors into account, so the proposed evaluation model and index system have some shortcomings. This paper selects the Yangtze River Delta region with certain representativeness and influence to avoid the influence of regional factors on the comprehensive score of enterprises due to inaccurate parameter estimation. At the same time, it establishes a perfect comprehensive evaluation index system, seizing the common ground of different industries. The index value can reflect the real level of corresponding enterprises, and has certain comparative significance and persuasiveness. 3. Construction of Enterprise Digital Maturity Evaluation Index System in the Yangtze River Delta 3.1. Principles for selecting evaluation indicators In order to build a comprehensive and effective index evaluation system to evaluate the digital maturity of enterprises, and ensure that the results are more accurate and practical, the following principles are followed when selecting indicators: (1) Scientific principle. The selection of digital maturity evaluation index must be objective and representative, which can fully reflect the degree, characteristics, period and other information of enterprise digitization. Indicators should be scientific from both qualitative and quantitative perspectives. (2) Principle of effectiveness. The collection of indicator data must be real, controllable, accurate and effective, which can reflect the development of enterprise digital transformation from different levels and effectively assess the level of enterprise digital maturity. (3) The principle of hierarchy. The selected indicators must be hierarchical, appropriate in number and able to be clustered. According to the characteristics of indicators, they are interrelated to form multi-dimensional primary and secondary indicators, so as to build a clear hierarchy and logical evaluation index system. (4) Independence principle. The index system should fully reflect the digital maturity level of the enterprise, so it must be related, but each index must also be independent and distinguishable. 3.2. Construction and interpretation of evaluation index system Scholars have different opinions on the definition of digitalization and digital transformation. From different perspectives, the essence of digital transformation will be 316 more and more extensive. Enterprise digitalization is also a dynamic process, which makes it more suitable for the capability maturity model. Enterprise digitalization can realize the transformation and upgrading of enterprises by continuously optimizing industrial resources from the initial state, so as to achieve the evolution and development of specific capabilities or specific goals. The capability maturity model constructs a convenient self-assessment tool for digital level assessment and analysis, and is an important method to realize enterprise digitization. It can help enterprises understand their digital construction level more comprehensively, find problems in digital construction in time, and then evaluate the current digital construction level objectively and scientifically. digital business profits and revenues", "KPI value" and "sustainable investment". Talents are the foundation and soul of an enterprise at any time. The driving force of digital transformation is digital talents, so the construction of enterprise digital talents team is the top priority of digital transformation. The number of digital talents owned by an enterprise, the investment in the construction of digital talents and the benefits provided to them are important manifestations of the digital strength of the enterprise. This paper adopts the "proportion of digital talents" "Digital level of high-level talents", "digital talent training investment", "incentive mechanism investment", "average salary" and "other welfare benefits". According to the above ideas, the enterprise maturity evaluation index system with 12 secondary indicators and 33 tertiary indicators as shown in Figure 1 is constructed. 4.1. Construction of judgment matrix AIJ in matrix M represents the comparison result of index I relative to index J. the judgment matrix is as follows: ⋯ 𝑎 𝑎 ⋮ 𝑎 ⋮ a M ⋯ 𝑎 𝑎 In order to reduce the uncertainty of results caused by human subjective influence, all elements at each level are compared under the same standard. The quantitative comparison scoring standard is shown in Table 1. 4. Enterprise Digital Maturity Evaluation Model Capability Maturity Model (CMM) is used to guide software development organizations to improve the process of software testing and upgrading. It has established a dynamic evaluation standard, and its application scope has gradually expanded, and its application field has become Table 1. Index comparison scale and its description AIJ assignment meaning aij=1 aij=3 aij=5 aij=7 aij=9 Aij=2, 4, 6, 8 Element I and element J are of the same importance to the factors at the upper level Element I is slightly more important than element J Element I is obviously more important than element J Element I is much more important than element J Element I is more important than element J Scale at the time of compromise between two adjacent scales marriage. In order to ensure the scientificity of the results, we check the consistency of the matrix to ensure the rationality of the indicators. λ n CI n 1 n λ CR n 1 RI If cr<0.1, it indicates that the matrix meets the requirements without modification; Otherwise, experts should be invited to revise the judgment matrix again and repeat steps 1 and 2 to make the calculation result cr<0.1. 4.2. Weight calculation The vectors of matrix M are geometrically averaged, and the average vector w obtained after normalization is the determined comparison weightω ω W / W 4.3. Inspection consistency In the actual operation process, the subjective wishes of experts may lead to some deviation in the comparison of Table 2. Reference standard value of average random consistency index Order reference value 1 0.00 2 0.00 3 0.58 4 0.90 5 1.12 6 1.24 description of capability levels reflecting the digital level. Based on the existing research model, this study divides the enterprise digital maturity into four levels through the digital maturity evaluation index system, expert scoring, and the corresponding weight ratio. 4.4. Classification of enterprise digital maturity Digital maturity related models generally adhere to the basic principles and framework of CMM, but differ in the 317 Figure 1. Classification of enterprise digital maturity Unit level: enterprises start the stage of digital transformation, divide their businesses into different units, and gradually realize the application of digital technology. At this time, enterprises lack digital technology talents, there is no reliable security system, and digital benefits are low. Process level: focusing on the core business process, the enterprise has made clear the strategic goal of digital transformation, has had preliminary digital facilities and talent base, and can achieve certain digital benefits. Ecological level: the digital transformation of enterprises has been relatively mature, which can realize the data sharing and prediction of core businesses, support cross business exchanges, and integrate digital equipment of relatively large scale. The enterprises have formed a good digital ecology and have considerable digital benefits. Leader: the digital maturity of enterprises has become the leader in the industry, which can realize the collaborative innovation of all business activities and continuously derive new digital management and business models. Acknowledgment This paper is funded by the undergraduate innovation and entrepreneurship training program of Anhui University of Finance and Economics (Project No.: s202110378548), and the ownership of the research results belongs to Anhui University of Finance and economics. References [1] Litingting. Research on digital transformation maturity evaluation of small and medium-sized manufacturing enterprises [j]. science and technology entrepreneurship monthly, 2023,36 (01): 110-114. [2] Wangsiwei Research on digital maturity evaluation system of manufacturing enterprises [d]. Hangzhou University of Electronic Science and technology, 2020.doi:10.27075/d.cnki.ghzdc.2020.000070 [3] Gregory vial Understanding Digital transformation: a review and a research agenda[j] Journal of strategic information systems, 2019, 28 (2): 118-144. 5. Conclusion [4] Zhang Yi. Digitalization and digital transformation of intelligent manufacturing enter a new stage -- Viewing the development trend of enterprise digitalization transformation from the perspective of policy [j]. lifting and transportation machinery, 2021 (11): 28-29. Digital technology has digitally transformed most businesses and scenarios, which has played a positive role in reducing costs and increasing efficiency. At the same time, it has brought changes to the operation mode of enterprises, governments, scientific research institutions and other industries, and has empowered and reshaped their working ability and service ability. From five dimensions, this study established 12 secondary indicators and 33 tertiary indicators to build an enterprise digital maturity evaluation index system, which helps enterprises better find their own problems from data, identify the direction of future digital transformation, guide enterprises to expand the scope of manufacturing resource allocation from traditional elements to data elements, and improve the development level of digital transformation. [5] Lishuqin. European Union's support for the development of SMEs' digital transformation policy proposition and Enlightenment [j]. management modernization, 2020,40 (05): 65-68. [6] Wang Rui, Dongming, houwenhao. Research on digital maturity evaluation model and method of manufacturing enterprises [j]. science and technology management research, 2019,39 (19): 57-64. [7] Xujinghan Research on Evaluation of digital transformation capability of manufacturing enterprises [d]. Hangzhou University of Electronic Science and technology, 2020. 318
https://openalex.org/W1978046486	https://zenodo.org/records/1769514/files/article.pdf	English	null	Volcanic origin of oil	Journal of the Franklin Institute	1,904	public-domain	5,249	This/ournal, eliv, 143, 225, 263. + " Volcanic Origin of Natural Gas and Petroleum," ./our. Can. Minin¢ Inst., Vol. VI :i: " Oil Fields of the Texas-Louisiana Gulf Coastal Plain." Bulletin No. 212, 1". S. Geologieal Survey, Washington, D. C., 19o 3. Volcanic Orzgin oj Otl. Volcanic Orzgin oj Otl. June. 19o4. ] 443 Mining and Metallurgical Section. Volcanic Origin of Oil. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. ECommunication to the Secretary, in discussion of the paper of Robert T. Hill, Washington, D. C., on "The Beau_ mont Oil Fields, with Notes on other Oil Fields of the Texas Region, ''~ presented at a joint session of the Franklin Insti- tute and of the American Institute of Mining Engineers, Philadelphia, May I5, I9O2.1 In a recent papert I took exception to the opening para- graph of the above paper of Mr. Hill, in which he says : "In endeavoring to interpret the geological occurrence of oil, the geologist is confronted by the fact that science has not yet solved the problem of its origin, which lies at the root of the subject. For the present we must consider oil as a material in the rocks, the origin of which is still unex- plained." I take this second occasion to re-enter our objection to the above statement, as well as to the following similar statement of Messrs. Hayes and Kennedy in their recent bul letin + on these same oil fields: "The origin of petroleum is one of the most obscure problems by which geologists are confronted." I claim that these statements, that we know so little to- day about the origin of petroleum, are not warranted in the present state of geological science, and I do not hesitate to state that, on the contrary, geology can to-day most clearly prove that the origin of oil is inorganic and the result of solfatarie volcanic emanations. Coste : 444 [j. F. I., [j. F. I., It has been so long assumed that oil or bitumen is due to the decomposition of the organic remains of the sedimentary rocks, that the words organic and bituminous are used syn- onymously, and accordingly, without attempt at explanation, the origin of oil is in every case ascribed to some bitumi- nous shale horizon, more or less full of organic matter. Such fallacious reasoning attributing the origin of oil to bitu- minous shales or to shales containing oil--that is to say to oil--of course, proves nothing except that this question is prejudged and entirely misunderstood by a great many geologists who seem to regard the abundant geological proofs that oil did not originate according to the above axiom as extraordinary phenomena and profound mysteries not yet solved by science. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. y y Chemistry can extract hydrocarbons from organic remains and can also produce hydrocarbons from simple mineral reactions (as in the commercial manufacture of acetylene, for instance), but the question before us is not : What are all the numerous chemical reactions capable of producing hydrocarbons or bitumens ? but only: How does nature do it ? What are the geological facts in this connection ? If in chemistry hydrocarbons are generally grouped and classed among the organic compounds, it is not a proof that the natural hydrocarbons have geologically an organic origin and that, in geology, organic and bituminous are synony- mous terms, to be exchanged and used indifferently, one for the other. The geological evidence is, on the contrary, very clear and very strong in disproving absolutely the organic origin of bitumens or hydrocarbons found in the earth's strata (excepting some marsh gas); and further, it shows plainly that the natural geological process of to-day and of ages past in the formation of these products is a mineral or inorganic process. This geological evidence is so well known and so indisputable that a simple enumeration of it seems to us sufficient: (i) Animal organic remains or bodies are never entombed in the rock formations and, therefore, cannot there produce oil or anything else. y g (2) Vegetable organic remains in the rock formations (2) Vegetable organic remains in the rock formations Volcanic Orzzin of Oil. 445 June, I9O4.] always decompose into carbonaceous matter, that is, peat, lignite and coal, with a small production of marsh gas, which, however, either escapes in the atmosphere or remains in the coal, and has evidently nothing whatever to do with the natural gas and oil of the gas and oil fields. g g (3) Further distillation of carbonaceous matter has not taken place in nature, except in very local and very rare instances, as proved by all the lignite and coal beds of the sedimentary strata. g g (3) Further distillation of carbonaceous matter has not taken place in nature, except in very local and very rare instances, as proved by all the lignite and coal beds of the sedimentary strata. " Volcanic Origin of Natural Gas and Petroleum," flour. Can. gllining Inst., Vol. VI. t [don. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. y (4) As reviewed elsewhere, ~ and as many very able geologists and scientists have repeatedly observed and recorded, in the volcanic districts of the earth, not only gas- eous, liquid and solid hydrocarbons or bitumens are among the most important products of the solfataric volcanic emanations, but also carbonic acid, chlorides (mostly com- mon salt), sulphuretted hydrogen, sulphur, gypsum and hot calcareous and siliceous waters are always the other con- spicuous products of these emanations, and all these asso- ciated products together stamp the solfataric volcanic phenomena with a unique and unmistakable seal. (4) As reviewed elsewhere, ~ and as many very able geologists and scientists have repeatedly observed and recorded, in the volcanic districts of the earth, not only gas- eous, liquid and solid hydrocarbons or bitumens are among the most important products of the solfataric volcanic emanations, but also carbonic acid, chlorides (mostly com- mon salt), sulphuretted hydrogen, sulphur, gypsum and hot calcareous and siliceous waters are always the other con- spicuous products of these emanations, and all these asso- ciated products together stamp the solfataric volcanic phenomena with a unique and unmistakable seal. That this volcanic phenomenon is the normal and orderly process of petroleum production is to us also a most clearly established geological fact for the following reasons, which we have discussed at length in another paper : ~- That this volcanic phenomenon is the normal and orderly process of petroleum production is to us also a most clearly established geological fact for the following reasons, which we have discussed at length in another paper : ~- g p p (a) It is the only geological process of petroleum produc- tion to be witnessed in active operation to-day in nature. (b) In all the oil and gas fields or petroleum deposits the gaseous products are under a strong pressure which is not artesian or hydrostatic, which increases with depth and which cannot be anything else than a volcanic pressure. (c) In some of the oil and gas fields heated waters, oils and gases are met with. g (d) All the oil and gas fields bear imprinted, largely through the products associated with the oil and gas, the seal we have referred to above as the distinct characteristic of solfataric volcanic emanations. Coste : 446 [J. F. " Contributions to Economic Geology," Bulletin No 2r3, U. S. Geol. Surv., pp. 345, ~47, 34,". BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. I., (e) The oil and gas fields are located along the faulted and fissured zones of the crust of the earth parallel to the great orogenic and volcanic dislocations. (f) Oil, gas and bitumens are never indigenous to the strata in which they are found; they are secondary products impregnating and cutting porous rocks of all ages, exactly as volcanic products alone could do. (g) Oil and gas are stored products, in great abundance in certain localities, while neighboring localities often are entirely barren ; and many of the strata among which they are found are so impervious that the source of these hydro- carbons must be the volcanic source below, which alone is abundant enough and alone possesses sufficient energy to force and accumulate such large quantities of these and associated products in so many spots through such imper- vious strata. No other oil or gas fields conform to what has just been written more closely and more plainly than some of the fields under review in Mr. Hill's paper. Indeed, Mr. Hill shows that in the Texas Louisiana Coast Prairie : No other oil or gas fields conform to what has just been written more closely and more plainly than some of the fields under review in Mr. Hill's paper. Indeed, Mr. Hill shows that in the Texas Louisiana Coast Prairie : (I) The oil and gas are always found under small mounds or salt islands, which are gentle recent quaquaversal uplifts or uplifted domes. This is fully confirmed by C. ~V. Hayes in a recent paper * and by others. (2) The salt water and the oils under these mounds are still, in some cases, hot. (3) The oil and gas under these mounds do not occur in any definite stratum, but in spots of many strata and in very large quantities in these limited areas. (4) The same may be said of the products associated with the oil or gas under these mounds, viz., sulphur, sulphuret- ted hydrogen, salt, gypsum, limestones and dolomites, which form, under these mounds, many irregular masses and pockets without any stratigraphical order of any kind. (5) The above associated products are not found in the wells drilled outside of the mounds any more than the oil or Ualcanic Origin of Oil. ~See also Capt. Lucas's paper, Trans. ~tin. Inst. 3!in. Eng., Vol. XXIX, and the " Oil Fields of the Texas-Louisiana Gulf Coastal Plain," Bulletin No. 212, U. S. Geol. Surv., p. i26. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. 447 Juue, i9o4.] gas, except in very small quantity, while under the mounds they often form very thick vertical masses, hundreds and thousands of feet in thickness.* If we keep these now well established facts carefully in mind we cannot come to any other eonclusion but the one adopted by Mr. Hill, that "these materials (of the mounds associated with the oil and including the oil)have origi- nated by seeondary replacement, and that they may be of post tertiary age." But, in his explanation of these secondary replacements under the mounds, Mr. Hill certainly does not go far enough. In my opinion, his explanatory hypothesis which he offers on page 35 as a "basis for discussion" and not " as a final explanation" should be altered as follows: p The oil, sulphur, salt, natural gas and hydrogen-sulphide pockets of the Texas Coastal Plain are not indigenous to the strata in which they are found, but are the resultant products of columns of hot saline, silicious, calcareous, magnesian and sulphur water and vapors, associated with sulphur and hydrocarbon gases and vapors, which have ascended, under volcanic pressure, at points along lines of structural weak- ness, and have disseminated also, in more or less minute quantities, a little oil, gas, sulphur and salt through thou- sands of feet of shales, sand and marine littoral sediments of the Coastal Plain section. These lines of structural weakness or fissures were par. tially sealed by the deposition of the latter overlapping strata now capping the oil pools, but the spasmodic recur- rences of the volcancic emanations kept raising and par- tially replacing the sands and clays of these latter strata to form in spots along these fissures the present mounds and salt islands. It will be observed that Mr. Hill's "hydrostatic pressure" is here changed to a volcanic pressure, and that the other important change we suggest is that the saline waters as Coste : 448 [J. F. I.. [J. F. I.. well as the hydrocarbon and sulphur gases and vapors are also of volcanic origin, instead of being derived from mete- oric sources and from the strata and sediments beneath the Coast Prairie. Before giving my reasons for the above suggestions, I wish to remark that they form a complete geological expla- nation of the origin of the oil and associated products of the mounds, while Mr. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. Hill's explanation is only partial and does not in any way solve the problem of the final origin of the oil, sulphur and salt which Mr. Hill simply admits, without explanation, to be disseminated in more or less minute quantities through the " bituminous" sediments of the Coastal Plain section. Nor does Mr. Hill's explanation account at all for the natural gas and for the sulphuretted- hydrogen gas which are two of the most important of the associated products of Mr. Hill's "oil-phenomena " of the mounds: it cannot, indeed, be admitted that the descend- ing meteoric waters leach or gather also these gases down- ward through the sediments to the places of structural weakness or fissures, and then float these gases upward at a few spots along these fissures My main reasons for the above suggested explanation of the origin of the oil phe- nomena of the mounds are based on the following simple facts : (I) The oils, waters and gases under these mounds are not under an hydrostatic pressure, small or great. This is amply demonstrated by the fact that the once famous gushers of Spindle Top are already gushing no more and have now to be pumped. If the pressure there was artesian or hydrostatic they would be gushing yet, the same as at first, and, if the oil had exhausted in some of these wells, these would be gushing water out of the supposed artesian water column behind them. It is certainly now a well- known and admitted geological fact, that, in all the oil and gas fields, the rock pressure of the gas is a stored energy continually decreasing as the gas is used, so much so that the gas itself has to be pumped to-day in many fields where its pressure was too strong at first ; and, surely, such a phe- nomenon cannot possibly have anything to do with an Volca~zic Origin of OiL June, I9O4. ] 449 artesian or hydrostatic water pressure. If not hydrostatic, then, what is it ? artesian or hydrostatic water pressure. If not hydrostatic, then, what is it ? Messrs. Hayes and Kennedy, in their recent report on t/tese all fields, say ." * " It appears highly probable that the pres- sure in the oil reservoir is due largely to the expansive force aJ the associated~as." Quite so ! "Oil Fields of the Texas-Louisiana Gulf Coastal Plain," U. S. Geol. Survey, Bulletin No. 212, p. i35 ; also Bulletin No. 213, p. 35o. t In Mr. Hill's paper under discussion, p. 4o. :~ Mr. Hill's paper, p. 34. VOL. CLVII. NO. 942. 2 9 BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. But this is hardly sufficient, and the question remains: What is the expansive force of the gas due to ? It remains unanswered and cannot be answered unless we go to its real volcanic source. This one word "volcanic" explains it all! And the invocation of this word is here a forced conclusion. We have indeed seen that there is no sign of the "tremendous hydrostatic pres- sure "~ of Mr. Hill in the wells of Spindle Top, which have to be pumped now; yet there is no doubt Mr. Hill is right in his conclusion that the oil and associated products have originated by secondary replacements caused by circulating waters (accompanied by vapors and gases), and if these waters and gases were not pushed by an hydrostatic pres- sure, that is, did not come from above, then they must surely have come from below, pushed by a volcanic pressure. y p y p (2) In full support of this I must now call your attention particularly to what Mr. Hill calls ~ "the association of oil, sulphur, sulphuretted hydrogen gas, gypsum, dolomite and salt, constituting collectively what might be termed the oil phenomena" of the mounds. Is not that association the unmistakable seal I have referred to above as belonging to the solfataric volcanic phenomena? And how could that unique stamp be found under these mounds unless they had been percolated by precisely the same hot waters, vapors and gases which are emitted in great abundance, and at times with disastrous violence and pressure, along the lines of structural weakness or fissures of the volcanic districts. of the earth ? y p y p (2) In full support of this I must now call your attention particularly to what Mr. Hill calls ~ "the association of oil, sulphur, sulphuretted hydrogen gas, gypsum, dolomite and salt, constituting collectively what might be termed the oil phenomena" of the mounds. Is not that association the unmistakable seal I have referred to above as belonging to the solfataric volcanic phenomena? And how could that unique stamp be found under these mounds unless they had been percolated by precisely the same hot waters, vapors and gases which are emitted in great abundance, and at times with disastrous violence and pressure, along the lines of structural weakness or fissures of the volcanic districts. of the earth ? C. \V. Hayes, Bulletin No. 2r3, U. S. Geol. Survey, p. 347. t " Volcanic Origin of Natural Gas and Petroleum," flour. Ca~. Min. Inst, Vol. VI, p. I4. Stienlific American, October :% :9o3. i[ "The Nature and Origin of Asphalt," Long Island City, N. V., 1898. Mr. Hill's paper under discussion. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. Therefore, along lines of faults or of structural weakness~ of the Texas-Louisiana Gulf Coastal Plain, volcanic gas- p p p VOL. CLVII. NO. 942. 2 9 Coste .. [l. V. I., 450 aqueous emanations have issued in spots, like volcanoes do along volcanic ranges, constituting at each spot "an elevat- ing force acting vertically and lifting a small portion of the earth's crust," as Mr. Hayes * puts it, but without explaining what the force was. This force, as Mr. Hayes also reeog- nizes, "was active down to a very recent date ; " in fact, it may be said to be active now, as in places the waters and oils are hot yet and the gases are still issuing. Thus the mounds and salt-islands of that region were formed exactly as sulfiones, salses and mud volcanoes form to-day similar deposits in many typical volcanic districts of the earth, such, for instance, as in Italy+ and in Java,** "where clouds of steam also issue from the vents and the heat is intense," and where sulphur, gypsum and salt are also abundantly associated. The only difference is one of intensity, which is less pronounced in the Texas-Louisiana region more remote from the centers of volcanic activity. An in- termediate case between the mounds and salt islands of Texas-Louisiana and the Java mud volcanoes is the Trinidad Pitch Lake, which the able and most careful researches of Clifford Richardson [I have determined to be unquestionably "the crater of an old mud volcano or geyser" with an influx of soft pitch at the center of the lake, amount- ing yet to thousands of tons yearly. g y y y There is one more point in Mr. Hill's paper which I con- sider important to discuss in this connection of the origin of oil. Throughout the paper Mr. Hill divides the Texas oil deposits into two classes : the sheet-oil deposits and the pocket-oil deposits. In the first class he conceives the oil to be indigenous or nearly so, and to come from "bitumi- nous" shales in close proximity to the sands. The bitumen (or oil) in the shales Mr. Hill admits and believes is due to the decomposition of organic matter.§ But even if that [~lcanic Orz~bl of OiL 45 I June, I9o4. ~ were so (which of course we only admit now for the sake ,,f argument), Mr. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. Hill does not explain how the oil was able to travel out of the impervious pores of the shales into the sands. No matter how close the proximity, we think that such a movement of the oil out of the shales would be im- possible, as the characteristic condition of shales is to be absolutely impervious. Even under great pressure, as there is abundant evidence in the oil and gas fields, a few feet or even a few inches of shales prevent all leakages, so much so that shales are generally considered as forming, and often do form, the impervious covers or cap rocks of the sand reservoirs, and covers or cap rocks sealing tightly and hermeticaily the oil from below, cannot also logically be regarded as the source from which other sand reservoirs above have been filled. In regard to the second-class or pocket-oil depositsl the foreign or adventitious nature of the oil and associated pro- ducts is so glaring that for these Mr. Hill gives up the old idea of an indigenous oil made by decomposition of organic matter ; but evidently this old idea is not going to be given up without a struggle, as it is retained for all the other deposits of Texas except for the ones of the Post-Eocene under the mounds, and even these are made to originate eventually from the same organic source, but in a less direct way through the intervention of circulating meteoric waters. The only attempt Mr. Hill makes to prove this indige- nous organic origin of the so-called sheet-oil is in these words : ~ " Of the 22,000 feet of sedimentaries in the Texas section, all but 2,000 feet are unconsolidated clays and sands, which may 3e more or less bituminous. Some of the lime-stones are also bituminous. Oil or bitumen has been found in at least a dozen horizons of the section In two instances in Texas bituminous matter (asphaltum)is faun# in apparently indigenous beds of lime-shell conglom- erate. The oil of the Corsicana field and that of San Antonio is derived from the shales of the Upper Cretaceous. The strata on the Eocene Tertiary present every The only attempt Mr. ~:- Mr. Hill's paper under discussion, p. 3 8. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. Hill makes to prove this indige- nous organic origin of the so-called sheet-oil is in these words : ~ " Of the 22,000 feet of sedimentaries in the Texas section, all but 2,000 feet are unconsolidated clays and sands, which may 3e more or less bituminous. Some of the lime-stones are also bituminous. Oil or bitumen has been found in at least a dozen horizons of the section In two instances in Texas bituminous matter (asphaltum)is faun# in apparently indigenous beds of lime-shell conglom- erate. The oil of the Corsicana field and that of San Antonio is derived from the shales of the Upper Cretaceous. The strata on the Eocene Tertiary present every pp The strata on the Eocene Tertiary present every Caste ." [J. F. I., [J. F. I., 45 z favorable condition for the generation of petroleum, whether this material be derived from the decomposition of marine organisms, as alleged by some, or from the hydrocarbons generated by vegetable matter, as bdieved by others. There are vast accumulations of both materials in Eastern Texas. Between the Eocene or Claiborne stratum of Nacog- doches and the uppermost Pleistocene stratum of Beaumont, there are thousands of feet of ffypsiferous clays and sands in which nature may now be distilling oiL" As will readily be seen, all the above proofs of Mr. Hill are simply assertions that bitumen or oil exists there (and, of course, that proves nothing as to its origin) or are sup- positions against the known facts, viz., "that the oil is derived from the decomposition of marine organisms or of vegetable matter, of which there are vast accumulations in Eastern Texas, or is distilled from gypsiferous clays and sands." Is it not, indeed, known without the shadow of a doubt ? (i) That no bodies of marine organisms are or were ever entombed in these sedimentaries; shells or bones of mil- lions of organisms were, but not one body. Just as it is in the sedimentaries being formed to-day along thousands of miles of coast lines, from which not one geologist has ever produced one entombed single body of marine or any other organism. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. So much for the "vast accumulations of marine organisms in Eastern Texas." g (2) Is it not also an undisputed recognized geological fact that the vegetable matter entombed in these sedimentaries (including the g-ypsiferous clays and sands) is to be found, and is found to-day as carbonaceous " undistilled " matter, containing yet the bitumens or oils which it would have pro- duced had the temperature necessary for distillation ever been attained ? Either there was distillation of these sedi- ments or there was not, and the lignites, coals and other carbonaceous matter found abundantly in them to-day are most convincing proofs that there is not, and was not, suffi- cient heat there to cause distillation and':to thus produce petroleum. g (2) Is it not also an undisputed recognized geological fact that the vegetable matter entombed in these sedimentaries (including the g-ypsiferous clays and sands) is to be found, and is found to-day as carbonaceous " undistilled " matter, containing yet the bitumens or oils which it would have pro- duced had the temperature necessary for distillation ever been attained ? Either there was distillation of these sedi- ments or there was not, and the lignites, coals and other carbonaceous matter found abundantly in them to-day are most convincing proofs that there is not, and was not, suffi- cient heat there to cause distillation and':to thus produce petroleum. These simple geological considerations, then, lead us to Yo•anic Origin of OiL 453 June, I9O¢. ] the safe conclusion that the so-called sheet-oil is not an indigenous product of decomposition of organic matter. That it is also, like the so-called pocket.oil, a secondary pro- duct of impregnation and replacement becomes quite clear when we remember that these so-called sheet-oil deposits are found not only in a dozen horizons, but in hundreds, from the oldest Paleozoic to the alluvial gravels and sands of to-day ; and that nowhere do they spread like sheets ; but that, on the contrary, they are always found in relatively very small pools in porous reservoir rocks, which are barren of oil outside of the little pools (exactly as in the case of the mounds), these pools forming only a very small percentage of the area of the numerous rock-sheets in which they are only occasionally found. That these accidental pools are only receptacles or reservoirs is admitted to-day by all who have studied the question. * "Natural Gas in Ontario,"Jour. Ca~z. 3[*n. lust., Vol. III. AN OPEN-HEARTH FURNACE MAKING STEEL BY A CONTINUOUS PROCESS. AN OPEN-HEARTH FURNACE MAKING STEEL BY A CONTINUOUS PROCESS. In a recent number of Slam und Eisen, Dr. Surzycki describes an open- hearth furnace that has been working satisfactorily since September, 19o2 , at the works of B. Hantke, at Czenstocha. In this furnace, which is of 3o-ton capacity, two tap-holes situated one over the other, but not in line, lead into a double spout by which the whole or any part of the contents of the furnace are easily tapped at any time. The furnace is charged with cold scrap to which, when melted, molten pig iron from a blast furnace or a mixer is poured in. When the bath is quiet, iron ore and mill-scale are added, and a further amount of pig iron. The charging lasts until the furnace is quite filled. The charge is de-phosphorized by adding lime, and when de-carboni- zation has gone far enough the furnace is tapped. The de-oxidation of the steel is carried out by adding wood carbon and ferro-manganese in a ladle during the tapping of the charge. After tapping, the upper tap-hole of the furnace is closed and the furnace repaired. A calculated amount of ore and roll-scale and a proportionate amount of pig iron is then run in. The prac- tice is carried out uninterruptedly for one to two weeks. If, from any cause, the remainder of the furnace contents must be cast, it is easily done by open- ing the lower tap-hole. The method, while based on the Talbot process, has the advantage that it may be worked in an ordinary fixed furnace, if not too small. BY EUGENE COSTE, E.M., Toronto, Ontario, Canada. That these reservoirs were filled also from the great volcanic tank below, in a manner exactly similar to the case of the Texas mounds, is plain when all the evidence enumerated above, which geology can bring forward to-day, is considered together. The above views on the origin of the oil phenomena, not only of the mounds and salt islands of the Texas-Louisiana district, but of all other oil deposits, lead one to a simple interpretation of the geological occurrence of oil which should be a guide to important results in the practical development of new oil and gas fields in the United States, as they have already been such a valuable guide to us in the developments of large new gas fields in Canada. ~ These views lead, indeed, to the following important conclusions: g p Oil a~d gas were only supplied along some of the lilzes of structural weakness, or along some of the fractured zones of the crust of the earth, and therefore the new fields are to be found only along these zones or belts. The numerous oil and gas fields known to.clay indicate plainly quite a number of these oil belts, but more remain yet to be discovered, and new ones are coming rapidly to the front, especially in the United States. That this is the Notes and Comments. 454 [J. F. I., [J. F. I., solution of the problem of the geological occurrence of oil and of oil developments and explorations, the writer has long been convinced on the considerations and for the rea- sons given above. Our suggestions are therefore that, as far as practical results are concerned, the important point is to accurately trace these fissured zones or belts on good maps and to drill in the localities thus indicated. p I have been at work for years, ever since i888, on maps of this character embracing North America, and I hope to be able to publish these before very long, and as soon as our present knowledge of these most important structural dis- locations is a little more complete. TORONTO, OI~TARIO, CANADA, December 5, 19o3- RAILWAYS OF THE WORLD IN 19o2. At the end of 19o2 the railway mileage of the world was 352,5oo miles, of which the United States had 2o2,471 miles, or 38 per cent., and Europe had I8O,7O8 miles, or 34 per cent. The British Empire had 91,485 miles, or I7 per cent. of the total ; the German Empire had 32,753 miles, and .'.he Russian Empire, 31,945 miles. The United States had, therefore, six times as much mileage as either the Russian Empire or the German Empire.
https://openalex.org/W4246459746	https://publikasi.polije.ac.id/index.php/j-remi/article/download/2228/pdf	Indonesian	null	Analisis Beban Kerja Petugas Rekam Medis Dengan Menggunakan Metode Wisn dan Fishbone di Puskesmas Ambulu Tahun 2019	Jurnal Rekam Medik dan Informasi Kesehatan	2,020	cc-by-sa	6,184	J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Raisa Putri Ramadhani1, Rinda Nurul Karimah2, Nugroho Setyo Wibowo3, Andri Permana Wicaksono4 Program Studi Rekam Medik, Jurusan Kesehatan, Politeknik Negeri Jember123,4, *e-mail: raisaputrirr@yahoo.co.id1, rindanurul2019@mail.ugm.ac.id2, nugroho@polije.ac.id3, andriperman4@gmail.com4 Abstrak Permasalahan yang ada terkait beban kerja dari petugas di unit rekam medis di Puskesmas Ambulu ialah tidak pernah dilakukannya analisis dan evaluasi, selain itu semenjak tahun 2018 unit rekam medis dan unit pendaftaran terpecah menjadi dua unit yang berdiri sendiri. Penelitian ini bertujuan untuk melakukan analisis beban kerja petugas rekam medis sehingga diketahui tingkat beban kerja dari petugas menggunakan metode WISN (workload indicator staff need), jika telah diketahui beban kerja petugas rekam medis selanjutnya dilakukan analisis faktor penyebab beban kerja dengan menggunakan metode fishbone dengan pendekatan 5M (man, money, methode, material, mechine) lalu dilanjutkan dengan FGD (focus group discusion) untuk menentukan prioritas utama penyebab beban kerja sehingga dapat dilakukan upaya perbaikan beban kerja. Penelitian ini merupakan jenis penilitan kualitatif dengan teknik pengumpulan data berupa observasi, wawancara, dokumentasi, dan FGD. Subyek penelitian ini ada 3 petugas rekam medis yaitu 1 perekam medis dan 2 pembantu perekam medis. Berdasarkan penelitian ini analisis beban kerja yaitu jobd escription sudah terlaksana dengan baik dikarenakan dalam job description masih belum terdapat pembagian tugas berdasarkan jabatan sehingga penyelesaian tugas dilakukan bersama-sama. Hasil perhitungan beban kerja mengetahui jumlah kebutuhan petugas didapatkan hasil jumlah ideal petugas yang dibutuhkan yaitu 4,85 sehingga dibulatkan menjadi 5 sehingga dapat disimpulkan bahwa beban kerja di unit rekam medis tinggi dengan jumlah 0,6. Upaya perbaikan beban kerja diantaranya dilakukan penambahan jumlah sumber daya manusia serta diadakan pelatihan sehingga menambah pengetahuan dan keterampilan dari petugas. Kata Kunci : Beban kerja, Faktor penyebab beban kerja, Workload indicator staff need Keywords : Workload, the causative factor of workload, Workload indicator staff need ANALISIS BEBAN KERJA PETUGAS REKAM MEDIS DENGAN MENGGUNAKAN METODE WISN DAN FISHBONE DI PUSKESMAS AMBULU TAHUN 2019 Raisa Putri Ramadhani1, Rinda Nurul Karimah2, Nugroho Setyo Wibowo3, Andri Permana Wicaksono4 1. Pendahuluan Berdasarkan Menkes (2014), pusat kesehatan masyarakat yang biasa disingkat menjadi puskesmas merupakan salah satu jenis fasilitas pelayanan kesehatan primer yang menyelenggarakan upaya kesehatan masyarakat yang lebih mengutamakan pemberian pelayanan pada aspek promotif dan preventif dibandingkan dengan aspek kuratif maupun aspek rehabilitatif agar tercapai derajat kesehatan masyarakat yang setinggi-tingginya. Fasilitas pelayanan yang disediakan oleh puskesmas terdiri dari unit medis maupun unit penunjang medis. Unit medis yang terdapat pada puskesmas terdiri dari unit rawat jalan, unit rawat inap, unit rawat darurat dan lain sebagainya, sementara untuk unit penunjang medis pada puskesmas diantaranya terdiri dari unit pendaftaran dan unit rekam medis, Unit pdiantaranya unit rawat jalan yang terdiri dari berbagai macam poli, unit rawat inap, unit gawat darurat serta unit lainnya termasuk unit rekam medis. Menurut Menkes (2008) rekam medis merupakan berkas yang berisikan catatan pelayanan yang diberikan kepada pasien selama berada di pusat kesehatan masyarakat. Isi dari catatan berkas rekam medis dapat dibedakan menjadi 2 jenis yaitu data sosial serta data medis pasien. data sosial pasien tersebut antara lain identitas pasien, sementara data medis pasien yaitu segala tindakan pelayanan kesehatan yang telah diberikan kepada pasien. Dalam pengisian isi dari berkas rekam medis yang bertanggung jawab yaitu dokter, serta tenaga kesehatan baik medis maupun non medis diantaraya perekam medis Menurut Pranoto (2015), analisis beban kerja merupakan suatu upaya untuk mengetahui waktu yang digunkan oleh petugas untuk menyelesaikan suatu tugas tertentu yang diharapkan dapat mengetahui proporsi jumlah petugas yang diperlukan dalam suatu unit kerja tertentu dalam suatu instansi termasuk puskesmas. Analisis beban kerja harus dilakukan rutin guna mendapatkan informasi mengenai gambaran beban kerja dari unit-unit pada suatu instansi. Hal ini dikarenakan gambaran-gambaran tersebut pada umumnya seirama dengan perkembangan kinerja dari suatu organisasi yang ada pada instansi tersebut. Berdasarkan studi pendahuluan yang telah dilakukan dapat diketahui bahwa analisis beban kerja petugas rekam medis di Puskesmas Ambulu masih belum pernah dilakukan sebelumnya. Jumlah petugas yang terdapat di unit rekam medis berjumlah 3 orang dimana hanya terdapat satu orang yang merupakan lulusan dari rekam medis sementara 2 orang lainnya merupakan lulusan SMA. Berdasarkan hasil observasi yang telah dilakukan pada 4 Mei 2019, diketahui rata-rata angka kunjungan pasien per harinya yang didapat dengan cara melakukan perhitungan jumlah seluruh pasien dibagi dengan jumlah hari aktif kerja petugas selama sebulan. Abstract The existing problem that related to workload of medical records officer in Primary Health Care of Ambulu there are never done the analysis and evaluation, more over since 2018 medical record unit and registration unit splinted into two unit to be on one’s own. This research that intend to do the analysis workload of medical record officers therfore be discovered the level of workload of officer with WISN (workload indicator staff need) methode. If ther is already known the workload of medical record officers the next is an analysis of causative factors of workload was carried out with fishbone methode with approach of 5M (man, money, methode, material, mechine) and then the next with FGD (focus group discusion) to determine the main priority that caused workload so that improvements of workload can be made. The type of this research is a qualitative with collecting data tecnic in the form of observation, interview, documentation, and FGD. The subject of this research are 3 medical record officer, there are 1 medical recorder and 2 medical recorder helper. The based on this research workload analysis the job description is well done because in the job description there is no division of tasks according to posisition so completion of the task have done together. The result of the calculation of workload is knowing the amount of officer needs the results obtined ideal number of officers there are 4,85 there for be rounded to 5 so the conclusion is the workload in the unit medical record is high with rhe total 0,6. The efforts to improve the workload including addition of human resources and the held of training thus increasing the knowledge and skills of officers. 582 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Vol. 1 No. 4, September 2020 1. Pendahuluan Berdasarkan data tersebut maka dapat diketahui bahwa angka kunjungan pasien di Puskesmas Ambulu menunjukkan adanya kenaikan jumlah pada bulan Juli sebanyak 1 orang, Oktober sebanyak 13 orang, Desember sebanyak 3 orang, Januari sebanyak 9, Februari sebanyak 8 orang, dan Maret sebnayak 2 orang. Serta menunjukkan penurunan angka kunjungan pasien pada bulan Juni sebanyak 2 orang, Agustus sebanyak 19 orang, September sebanyak 8 orang, November sebanyak 4 orang, dan April sebanyak 22 orang. Dapat diketahui bahwa kenaikan jumlah pasien tertinggi terdapat pada bulan Oktober, serta terjadi penurunan angka kunjungan pasien terendah pada bulan April. Jumlah kunjungan pasien merupakan salah satu dari faktor penyebab terjadinya beban kerja dari petugas rekam medis yang nantinya diabandingkan dengan jumlah petugas, selain jumlah kunjungan pasien, masih terdapat faktor-faktor lain yang dapat menyebabkan beban kerja dari petugas rekam medis. Faktor-fakror yang mempengaruhi beban kerja petugas rekam medis di Puskesmas Ambulu dapat dibagi menjadi 2 antara lain, faktor internal dan faktor eksternal. Faktor internal yang mempengaruhi beban kerja petugas rekam medis di Puskesmas Ambulu cenderung diakibatkan, motivasi, dan persepsi kerja dari petugas yang menganggap pekerjaan mereka lebih banyak dibandingkan jumlah tenaga kerjanya sehingga menyebabkan petugas tersebut memilah dan memilih pekerjaan dan beberapa pekerjaaan yang seharusnya menjadi tanggung jawab perekam medis menjadi tidak dikerjakan. Faktor eksternal yang mempengaruhi beban kerja petugas rekam medis di puskesmas ambulu antara lain lingkungan kerja beserta 583 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Vol. 1 No. 4, September 2020 sarana dan prasarana, angka kunjungan pasien, serta job description dimana didalamnya terdapat rincian tugas serta wewenang yang harus dilakukan oleh petugas di unit rekam medis. Berdasarkan penjabaran tersebut maka perlu dilakukan suatu analisis beban kerja untuk mengetahui tingkat beban kerja petugas rekam medis serta menentukan faktor utama penyebab beban kerja dari petugas sehingga nantinya dapat dilakukan upaya perbaikan dengan mencari solusi agar kualitas pelayanan yang dihasilkan dapat maksimal maka dari itu dengan adanya anilisis ini diharapkan dapat menjadi masukan bagi puskesmas dalam membentuk susunan anggota di unit rekam medis dengan tugas dan wewenang yang sesuai dengan kebutuhan supaya menghasilkan pelayanan yang berkualitas serta memperbaiki isi dari job description agar semua kompetensi dasar yang harus dimiliki oleh perekam medis dapat terealisasi dengan sepenuhnya di Puskesmas Ambulu sehingga didapatkan kualitas pelayanan yang paripurna. 2.1 Jenis/desainPenelitian Penelitian analisis beban kerja yang dilakukan kepada petugas rekam medis Puskesmas Ambulu merupakan jenis penelitian kualitatif dengan menggunakan metode WISN dan digram fishbone. 2. Metode Penelitian Metode penelitian merupakan suatu cara yang ilmiah untuk mendapatkan data dengan tujuan dan kegunaan tertentu (Elisanti, A.D Ardianto, E. T 2020). Jenis penelitian yang digunakan merupakan penilitan kualitatif dengan menggunakan metode pengumpulan data antara lain observasi langsung, wawancara, dan studi pustaka. Sehingga didapatkan data-data yang nantinya akan diolah dengan menggunakan metode WISN yang akhirnya didapat tingkat beban kerja dari petugas rekam medis di Puskesmas Ambulu dan kemudian dari hasil tersebut dicari faktor utama dari beban kerja petugas dengan menggunakan metode Fishbone. j g g gg Jenis data yang dikumpulkan peniliti dapat dibagi menjadi 2 macam yaitu data primer dan sekunder, untuk data primer yaitu data yang dihasilkan dari observasi dan wawancara yang didapatkan hasil seluruh aktivitas dari petugas, sementara untuk data sekunder yaitu data kehadiran petugas serta daftar kunjungan pasien di Puskesmas Ambulu. 1. Pendahuluan Analisis beban kerja petugas rekam medis di puskesmas ambulu dilakukan dengan menggunakan metode WISN, dimana dengan menggunakan metode WISN maka dapat diketahui jumlah staf yang dibutuhkan serta tekanan beban kerja dari petugas. Setelah melakukan perhitungan dengan menggunakan metode WISN maka selanjutnya akan dilakukan analisis faktor penyebab yang dapat memengaruhi beban kerja dari petugas rekam medis di Puskesmas Ambulu dengan menggunakan metode fishbone. Metode fishbone meupakan suatu metode sebab akibat yang digunakan untuk menentukan suatu permasalahan utama dengan cara melakukan diskusi kelompok. Metode fishbone yang digunakan pada penelitian ini dengan cara pengelompokkan 5M (man, money methode, material, dan machine) terhadap permasalahan yang ditemukan.. Berdasarkan paparan dari permasalahan tersebut, maka peneliti berkeinginan untuk meniliti tentang ‘Analisis Beban Kerja Petugas Rekam Medis dengan menggunakan Metode WISN dan fishbone di Puskesmas Ambulu Tahun 2019’ kelebihan analisis beban kerja petugas ini yaitu untuk mengetahui jumlah petugas rekam medis yang dibutuhkan, selanjutnyamenentukan faktor utama penyebab beban kerja agar dicari solusinya, faktor penyebab beban kerja petugas ditinjau dari faktor internal yaitu motifasi dan persepsi petugas terhadap pekerjaan yang dikerjakan serta faktor eksternal sarana dan prasarana dan tingkatan banyaknya pekerjaan . 2.3 Metode Pengumpulan Data Metode pengumpulan data untuk menganalisis beban kerja petugas rekam medis menggunakan wawancara yang disertai pedoman wawancara yang berisi daftar pertanyaan yang telah dipersiapkan sebelumnya dan akan disajikan ke informan yang berkaitan dengan beban kerja petugas rekam medis bagian, observasi yang beserta pedoman observasi yang berisi hal- hal yang perlu diamati yang berkaitan dengan kinerja petugas rekam medis dokumentasi seperti halnya merupakan bukti penunjang dari wawancara dan observasi yang telah dilakukan. 2.2 SubjekPenelitian Subjek penelitian yang dilakukan di Puskesmas Ambulu yaitu 3 orang narasumber yang merupakan petugas rekam medis serta 1 orang petugas perekam medis serta 2 orang pembantu perekam medis yang berdasarkan job description dari petugas tersebut. Dari ketiga petugas 584 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Vol. 1 No. 4, September 2020 rekam medis, satu diantaranya yang merupakan lulusan dari D-III Rekam Medis, sementara 2 lainnya merupakan lulusan sekolah mengah atas (SMA). rekam medis, satu diantaranya yang merupakan lulusan dari D-III Rekam Medis, sementara 2 lainnya merupakan lulusan sekolah mengah atas (SMA). 2.4 Metode Analisis Data Metode analisis data pada penelitian ini dianalisis dengan menggunakan analisis kualitatif yaitu dengan mengolah data yang diperoleh serta memaparkan hasil penelitian analisis beban kerja petugas rekam medis. Pengolahan data disini menggunaakan metode WISN (Workload Indicator Staff Need) dengan mengidentifikasi daftar hadir dari petugas serta waktu kerja dari petugas yang nantinya dapat dihitung dan diketahui tingkat beban kerja dari petugas rekam medis, kemudian dari hasil tersebut dianalisis untuk diketahui faktor utama penyebab dari beban kerja petugas dengan menggunakan metode fishbone.. “ jobdesc disini sudah ada, dan isinya juga cukup jelas, di jobdescnya kan masik jadi satu sama pendaftaran jadi yang dibagi sendiri tugasnya” 3.3. Memperkirakan Waktu Kerja Yang Tersedia Perhitungan waktu kerja yang tersedia dapat dilihat dari lampiran 8. Waktu kerja yang tersedia merupakan waktu seorang pekerja melakukan pekerjaanya yang dikaitkan dengan data kehadiran. Hal ini dikarenakan setiap orang berhak untuk libur dari melakukan aktivitas pekerjaan. Memperkirakan waktu kerja yang tersedia dapat dilakukan dengan menjumlahkan hari kerja dalam satu tahun kemudian dikurangi dengan hari libur, hari cuti tahunan, hari cuti lainnya seperti pelatihan dan lain sebagainya serta ketidakhadiran petugas yang disebakan oleh kepentingan pribadi. untuk hal ini Puskesmas Ambulu memiliki hari kerja dimulai dari hari senin hingga hari sabtu sehingga total hari kerjanya berjumlah 6 hari yaitu hari senin dimulai dari pukul 07.00 s/d 14.00, untuk hari jum’at dimali dari pukul 07.00 s/d 11.00 sementara untuk hari sabtu dimali dari pukul 07.00 s/d 12.30. Hal ini didukung dari kutipan wawancara dari informasi berikut ini: Hari yang dibutuhkan untuk menghitug waktu kerja petugas rekammedis di Puskesmas Ambulu yaitu 313 hari aktif kerja selama satu tahun, 16 hari libur nasionla berdsarakan tanggalan, 8 hari cuti nasional berdasarkan tanggalan, serta rata-rata waktu kerja dalam satuan jam yaitu 6,2 jam yang dimulai dari hari senin hingga sabtu. Tabel 1 Waktu Kerja Yang Tersedia Petugas Rekam Medis No Job Description A B C D E F AWT 1 Paetugas Rekam Medis 313 16 8 0 0 6 1791,8 2 Pembantu Rekam Medis 1 313 16 8 0 0 6 1791,8 3 Pembantu Rekam Medis 2 313 16 8 0 0 6 1791,8 Sumber: Data Primer Yang diolah Peneliti (( Tabel 1 Waktu Kerja Yang Tersedia Petugas Rekam Medis Perhitungan WISN yang terdapat pada tabel 3.1 menjelaskan bahwa hasil perhitungan waktu kerja yang tersedia petugas rekam medis di Puskesmas Ambulu selama setahun yaitu 1791,8 jam. 3.2. Identifikasi Prioritas Jenis Fasilitas Kesehatan Menurut Retnowati (2019) analisis beban kerja merupakan suatu aktivitas yang perlu dilakukan secara berkala, selain itu analaisis ini juga perlu dilakukan apabila terdapat perubahan dalam sruktur organisasi sehingga diperlukan rencana promosi dan rotasi karyawan. Metode WISN yang digunakan untuk mengukur kesesuaian kebutuhan jumlah petugas tidak dapat diterapkan sekaligus dalam satu waktu terhadap seluruh unit kerja yang ada di puskesmas hal ini dikarenakan sumber daya yang terbatas, maka dari itu perlu diadakan prioritas untuk menentukan unit kerja yanga akan menjadi target untuk penerapan metode WISN. Penambahan jumlah petugas rekam medis berguna untuk meratakan sumber daya manusia agar dapat terlaksana pelayanan yang berkualitas, namun dengan adanya distribusi petugas ke rekam medis maka perlu adanya pengukuran kesesuain jumlah petugas dengan menggunakan metode WISN mengingat masih terdapat tugas dan wewenang yang masih belum dapat dilakukan oleh petugas rekam medis. 3.1. Identifikasi Job Desccription Petugas Rekam Medis di Puskesmas Ambulu Jember Tahun 2019. Menurut Elbadiansyah (2019) menyatakan bahwa job description atau yang biasa disebut dengan deskripsi pekerjaan merupakan suatu catatan yang berisikan tugas serta tanggung jawab pekerja terhadap suatu jabatan tertentu. Isi dari job description ini telah disesuiakan dengan kondisi pekerjaan di lapangan sehingga dengan adanya job description dapat digunakan dalam perekrutan dan seleksi tenaga kerja baru, penempatan posisi jabatan, penentuan pendidikan dasar serta kemampuan yang dimiliki oleh petugas, pelatihan serta perbaikan syarat-syarat dalam pekerjaan, perencanaan organisasi dan promosi jabatan petugas. Isi dari job description harus dipaparkan secara rinci dan menggunakan bahasa yang mudah dipahami sehingga dapat dijadikan pedoman bagi petugas agar kedepannya segala aktifitas di unit rekam medis terkait dengan tugas dan wewenang dari petugas dapat terkoordinir dengan baik sehingga petugas tidak merangakap pekerjaan. Berdasarakan hasil wawancara dan observasi yang telah dilakukan oleh penulis di Puskesmas Ambulu tahun 2019 menghasilkan informasi bahwa pada unit rekam medis di Puskesmas Ambulu sudah terdapat job description. Namun job description tersebut diperuntukkan unit kerja rekam medis dan loket pendaftaran. Pada tahun 2018 terjadi pemisahan unit kerja, antara loket pendaftaran dan rekam medis sementara untuk job description yang ada masih belum dilakukan pembaharuan sehingga pembagian tugas dan tanggung jawab antara petugas loket pendaftaran dengan petugas rekam medis yang dilakukan secara manual yaitu dengan cara didiskusikan mengenai apa saja tugas dan wewenang dari masing-masing unit. Hal ini didukung dengan kutipan wawancara dari informasi berikut ini: Dapat disimpulkan bahwa jobdescription dari petugas rekam medis di Puskesmas Ambulu sudah ada namun isinya belum diperbaruhi mengigat struktur organisasi yang berubah dikarenakan terbentukna unit kerja baru yaitu rekam medis. 585 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Vol. 1 No. 4, September 2020 3.4. Identidikasi Komponen Beban Kerja Menurut WHO (2010:18) menyataka komponen beban kerja merupakan aktivitas kerja yang dilakukan oleh petugas yang menghabiskan waktu kerja dari petugas selama seharian. Komponen beban kerja yang ada pada suatu instansi kesehatan dapat dbagi menjadi 3 jenis yaitu kegiatan pelayanan kesehatan, kegiatan pendukung dan kegiatan tambahan. 586 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 3.5. Standar Aktivitas Beban Kerja Tabel 2 Standar Pelayanan Rekam Medis Kategori staf : perekam medis di puskesmas ambulu Kelompok beban kerja Komponen beban kerja Satuan waktu/Tingkatan Kerja Aktivitas pelayanan kesehatan semua perekam medis Pembuatan berkas rekam medis pasien baru 1 menit per klien Pengambilan berkas 4 menit per klien Pembuatan Tracer 0,5 menit per klien Pengembalian Berkas 1 menit per klien Penyusunan Formulir 2 menit per klien Pengecekan kelengkapan formulir 2 menit per klien Pengecekan Kelengkapan Isi 4 menit per klien Pelaporan RM 6 menit per klien Sumber: Data Primer Yang diolah Peneliti J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Dapat diketahui bahwa aktivitas pelayanan rekam medis yang membutuhkan waktu paling banyak yaitu assembling, hal ini dikarenakan petugas selain mengurutkan formulir sesuai tanggal juga mengecek kelengkapan formulir, sehingga jika terdapat formulir yang tidak lengkap maka petugas akan meminta pihak terkait untuk melengkapinya. Tabel 3 Standar Kegiatan Pendukung Kategori staf : perekam medis di puskesmas ambulu Rata-rata jam kerja yang tersedia dalam sehari: 6,2 Jam Hari kerja dala seminggu: 6 Hari Hari kerja dalam setahun: 289 Hari Jam kerja dalam setahun: 1.791,8 Jam Kelompok beban kerja Komponen beban kerja CAS (waktu kerja yang sebenarnya) CAS% (presentasee waktu kerja) Aktivitas pendukung semua perekam medis Distribusi 8 menit perhari 154 jam setahun 8,6%=(154/1791,8)x100 Perencanaan RM 120 Menit per tahun 6,7%=(120/1791,8)x100 Rapat 12 jam per bulan 8,0%=[(12x12)/1791,8]x100 Jumlah CAS% 23,3% Sumber: Data Primer Yang diolah Peneliti Tabel 3 Standar Kegiatan Pendukung Tabel 3 Standar Kegiatan Pendukung Dari tabel tersebut dapat diketahui standar kegiatan pendukung atau CAS sebesar 23,4% dimana angka persentase terbesar terdapat pada aktivitas rapat yang diadakan setiap bulan. 3.4. Identidikasi Komponen Beban Kerja Tabel 4 Standar Kegiatan Tambahan t f k di di k b l Tabel 4 Standar Kegiatan Tambahan Tabel 4 Standar Kegiatan Tambahan Kategori staf : perekam medis di puskesmas ambulu Kelompok beban kerja Komponen beban kerja Jumlah Staf IAS (Waktu kerja aktual per orang) IAS Tahunan Aktivitas tambahan tertentu perekam medis Pengawasan Mahasiswa Perekam medis yang Magang 1 10 jam per tahun 10 jam setahun Total IAS dalam setahun 3.4. am Sumber: Data Primer Yang diolah Peneliti Sumber: Data Primer Yang diolah Peneliti Dari tabel tersebut dapat diketahui bahwa total IAS yang merupakan waktu kerja aktual per orang dalam kurun waktu satu tahun untuk petugas rekam medis sebanyak 10 jam. 587 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 3.6. Standar Beban Kerja Petugas rekam Medis di Puskesmas Ambulu Perhitungan standar beban kerja jumlah waktu kerja petugas (AWT) yaitu 21.108 jam, sedangkan satandar pelayanan rekam medis berdasarakan tabel 9.1 aktivitas filing sebanyak 5 menit per pasien, aktivitas asembling selama 8 per pasien, dan aktivitas pelapporan selama 6 menit per pasien Tabel 5 Standar beban kerja Tabel 5 Standar beban kerja Kategori staf : perekam medis di puskesmas ambulu AWT dalam setahun: 1.791,8 jam Kelompok beban kerja Komponen beban kerja Satuan waktu/Tingkatan Kerja Aktivitas pelayanan kesehatan semua perekam medis Pembuatan berkas baru (1.791,8x60) = 107.508 Pengambilan berkas (1.791,8x15) = 26.877 Pembuatan tracer (1.791,8x120) = 26.877 Pengembalian berkas (1.791,8x60) = 107.508 Penyusunan formulir (1.791,8x30) = 53.754 Pengecekan kelengkapan lembar formulir (1.791,8x30) = 53.754 Pengecekan kelengkapan isi (1.791,8x15) = 26.877 Pelaporan RM (1.791,8x10) = 17.918 Sumber: Data Primer Yang diolah Peneliti Dapat diektahui bahwa satuan waktu pada standar beban kerja paling tinggi yaitu aktivitas filing dengan jumlah 21.501,6 jam, sedangkan satuan waktu untuk standar beban kerja paling rendah yaitu aktivitas assembling dengan jumlah waktu 13.918 jam. 3.7. Menetukan Kebutuhan staf Menggunakan Metode WISN Menganalisis dan Menafsirkan Hasil Perhitungan Dengan Menggunakan Metode WISN 3.7. Menetukan Kebutuhan staf Menggunakan Metode WISN 3.7. Menetukan Kebutuhan staf Menggunakan Metode WISN 3.7. Menetukan Kebutuhan staf Menggunakan Metode WISN Menurut WHO (2010:26) menentukan kebutuhan staf merupakan perhitungan kebutuhan staf sesuai dengan semua pekerjaan yang dilakukan oleh petugas mulai dari kegiatan pelayanan, keiatan pendukung maupun kegiatan tambahan. Tabel 6 Menentukan kebutuhan staf Kategori staf : perekam medis di Puskesmas Ambulu AWT dalam setahun: 1.584 jam Kelompok beban kerja Komponen beban kerja Beban kerja tahunan Standar beban kerja Jumlah staf yang dibutuhkan Aktivitas pelayanan kesehatan semua perekam medis Pembuatan berkas baru 21.108 107.508 0,2 Pengambilan berkas 21.108 26.877 0,8 Pembuatan tracer 21.108 215.016 0.1 Pengembalian berkas 21.108 107.508 0,2 Penyusunan formulir 21.108 53.754 0,4 Pengecekan kelengkapan lembar formulir 21.108 53.754 0,4 Pegecekan kelengkapan isi 21.108 26.877 0.8 Pelaporan RM 21.108 17.918 1.2 A. Jumlah staf yag dibutuhkan kegiatan pelayanan kesehatan 4.1 Kelompok beban kerja Komponen beban kerja CAS (waktu kerja yang sebenarnya) CAS% (presentasee waktu kerja) Aktivitas pendukung semua perekam medis Perencanaan RM 120 Menit per tahun 6,7%=(120/1791,8)x100 Rapat 12 jam per bulan 8,0%=[(12x12)/1791,8]x100 Distribusi 147 jam setahun 8,6%=(154/1791,8)x100 Jumlah CAS% 23,3% B. Faktor kategori tunjangan {1/[1-(presentase total CAS/100)]} 1,3 Tabel 6 Menentukan kebutuhan staf edis di Puskesmas Ambulu Tabel 6 Menentukan kebutuhan staf i t f k di di P k A b l 588 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Kelompok beban kerja Komponen beban kerja Jumlah Staf IAS (Waktu kerja aktual per orang) IAS Tahunan Aktivitas tambahan tertentu perekam medis Pengawasan Mahasiswa Perekam medis yang Magang 1 10 jam per tahun 10 jam setahun Total IAS dalam setahun 10 jam C. Faktor Penyisihan Individu (Total IAS Tahunan/AWT) 0,0056 Total jumlah staf yang diperlukan berdasarkan WISN (AxB+C) 5,33 Sumber: Data Primer Yang diolah Peneliti J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Dapat diketahui bahwa kebutuhan petugas berdasarkan metode WISN berjumlah 4,85 untuk mnegetahui angka jumlah kebutuhan petugas dengan pastinya maka perlu dibulatkan terlebih dahulu. Menurut WHO (2010) pembulatan angka dapat terjadi dibulatkan keata maupun kebawah tergantung bersarnya angka setelah koma seperti jika angka dibelakang koma 0 sampai1 maka akan dibulatkan kebawah namun jika angak dibelakan koma 2 sampai 9 maka akan dibulatkan keatas. Jadi jumlah kebutuhan petugas berdasarkan metode WISN yang berjumlah 4,85 dapat dibulatkan keatas sehingga menjadi 5, sehingga petugas yang dibutuhkan menurut metode WISN berjumlah 4 orang 3.8. 3.10. Menentukan Faktor Utama Penyebab Beban Kerja Serta solusi Denga Menggunakan FGD 3.10. Menentukan Faktor Utama Penyebab Beban Kerja Serta solusi Dengan Menggunakan FGD Berdasarkan hasil kegiatan FGD (focus group discussion) pada Sabtu 14 Desember 2019 didapatkan hasil analisis faktor utama penyebab tingginya beban kerja petugas rekam medis di Puskesmas Ambulu serta solusi untuk mengatasi tingginya beban kerja tersebut. Hasil dari FGD yang telah dilakukan yaitu dapat diketahui bahwa faktor utama penyebab dari tingginya beban kerja merupakan ketidak seimbangan antara jumlah tenaga kerja perekam medis dengan tugas yang dikerjakan. Dapat dibuktikan seperti kutipan berkiut: “Penyebab utamanya ya dari manusianya, petugasnya kurang Cuma 3 orang sedangkan tugasnya banyak, solusianya ya perlu adanya penambahan petugas, kalo pelatihan sepertinya untuk sekarang ndak bisa sooalnya tidak ada danany, mungkin nanti bisa diadakan pelatihan” Rekam medis di Puskesmas Ambulu terkait beban kerja perlu adanya penamnbahan petugas sehingga beban kerja dari petugas normal jadi pelayanan dapat dilakukan secara maksimal. Selain itu tindakan lainnya yng dapat dilakukan adalah memperbarui struktur organisasi dan job description agar sesuai. 4. Simpulan dan Saran 4.1 Simpulan 3.8. Menganalisis dan Menafsirkan Hasil Perhitungan Dengan Menggunakan Me Kalo material ya yang pasti jumlah raknya kurang jadi mapnya ditaruk dikerdus gitu, terus mapnya yang putih yang udah disediain kurang jadi beberapa rekam medis yang ndak pake map jadi ditumpuk jadi satu gitu. Kalo jobdesc nanti mau di rubah masih, itu kan masik pake yang lama isisnya jadi antara rekam medis dengan pendaftaran masih jadi satu, jadi nanti bisa diatur ulang, Kalo struktur organisasi ya sama kaya jobdesc, nanti diatur ulang, kan ini masih baru jadi masih perlu banyak perbaikan dan penyesuaian” J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 medis Cuma satu yang lain kan bukan sama kalo bisa ada tunjangan uang lembur dah. Kalo method ya ini kan masik baru jadi jobdesc masih ndak ada pembagian tugas terus rekam medis di struktur organisasinya masih belum tercantum. Kalo material ya yang pasti jumlah raknya kurang jadi mapnya ditaruk dikerdus gitu, terus mapnya yang putih yang udah disediain kurang jadi beberapa rekam medis yang ndak pake map jadi ditumpuk jadi satu gitu. Kalo jobdesc nanti mau di rubah masih, itu kan masik pake yang lama isisnya jadi antara rekam medis dengan pendaftaran masih jadi satu, jadi nanti bisa diatur ulang, Kalo struktur organisasi ya sama kaya jobdesc, nanti diatur ulang, kan ini masih baru jadi masih perlu banyak perbaikan dan penyesuaian” Berdasarkan kutipan diatas maka dapat disusun diagram ikan (fishbone) seperti di bawah ini : Gambar 1. Fishbond Chart Berdasarkan kutipan diatas maka dapat disusun diagram ikan (fishbone) seperti di bawah ini : Gambar 1. Fishbond Chart 3.8. Menganalisis dan Menafsirkan Hasil Perhitungan Dengan Menggunakan Me 3.8. Menganalisis dan Menafsirkan Hasil Perhitungan Dengan Menggunakan Metode WISN Menganalisis dan menafsirkan hasil perhitungan sehingga dapat diketahui tingkat beban kerja dari petugas yang terkait. Kebutuhan petugas rekam medis menurut metode WISN dapat diketahui berjumlah 5 orang, sementara jumlah petugas yang ada berjumlah 3 orang, sehingga terdapat selisih kekurangan petugas sebanyak dua orang. untuk mengetahui tekanan beban kerja dari petugas rekam medis di Puskesmas Ambulu. j p g Tabel 7 Hasil Analisa dan Menafsilkan WISN Kategori Staf : Rekam mMedis di Puskesmas Ambulu Jumlah Saat ini Jumlah Yang dibutuhkan Berdasarkan WISN Kekurangan /Kelebihan Masalah Tenaga Kerja Rasio WISN Tekanan Beban Kerja 3 5 -2 Kekurangan 0,6 Tinggi Sumber: Data Primer Yang diolah Peneliti Dapat disimpulkan bahwa tekanan beban kerja dari petugas rekam medis di Puskesmas Ambulu tinggi dengan jumlah 0,6 3.9. Analisis Faktor Penyebab Beban Kerja Dengan Menggunakan Metode Fishbone Diagram fishbone merupakan suatu metode untuk mengidentifikasi faktor-faktor penyebab dari permasalah yang ada. Penelitian analisis faktor penyebab beban kerja ini menggunakan pendekatan 5 M (man, money, method, material, machine). Berdasarkan hasil wawancara dan observasi yang telah dilakukan penyebab dari beban kerja petugas rekam medis berdasarkan faktor manusia (man) yaitu terdiri dari kurangnya motivasi petugas terhadap suatu pekerjaan serta kurangnya jumlah petugas, untuk faktor machine yaitu kurangnya jumlah PC serta tidak tersedianya printer, sementara untuk money yaitu tidak tersedianya anggaran untuk peplaihan bagi petugas serta tidak terdapat anggaran tunjangan bagi petugas yang melakukan kerja lembur, untuk method yaitu kurang jelasnya strktur organisasi serta pada job description tidak terdapat pembagian tugas menurut jabatan, untuk material yaitu kurangnya jumlah rak, kurangnya jumlah kursi, kurangnya jumlah map. Dibuktikan dengan kutipan wawancara berikut: “kalo kendalanya dari faktor manusianya ya kurangnya jumlah petugas juga petugas sini kan Cuma dikit jadi ndak mau ribet ya kerjakan sebisanya dah, sementara kalo dari peralatan mesin ya komputernya kurang disini Cuma ada satu sama ndak ada printer jadi kalo formulirnya habis minta dulu kertas formulir ke TU jadi nanti mereka yang ngasik, Sementara kalo uang sih sebenarnya ndak ada tapi kalo misalnya bisa ya butuh anggaran buat pelatihan kan disini petugas yang lulusan rekam 589 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 medis Cuma satu yang lain kan bukan sama kalo bisa ada tunjangan uang lembur dah. Kalo method ya ini kan masik baru jadi jobdesc masih ndak ada pembagian tugas terus rekam medis di struktur organisasinya masih belum tercantum. 4. Simpulan dan Saran 4.1 Simpulan p Berdasarkan penjabaran hasil serta pembahasan dari penelitian mengenai analisis beban kerja petugas rekam medis serta mengetahui faktor penyebab dari beban pekerja petugas dapat disimpulkan sebagai berikut: 1. Berdsarkan identifikasi job description didapatkan hasil bahwa pelaksanaannya sudah terlaksanan dengan baik, namun dalam job description masih belum tertera pembagian tugas sesuai dengan jabatan. 590 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Vol. 1 No. 4, September 2020 2. Berdsarakan identifikasi prioritas jenis fasilitas pelayanan, terpilihnya unit rekam medis disebabkan pada unit tersebut belum pernah dilakaukan analisis beban kerja serta rekam medis termasuk unit kejabaru hal ini dikarenakan baru tahun 2018 rekam medis menjadi unit kerja sendiri, maka perlu diadakannya analisis beban kerja 2. Berdsarakan identifikasi prioritas jenis fasilitas pelayanan, terpilihnya unit rekam medis disebabkan pada unit tersebut belum pernah dilakaukan analisis beban kerja serta rekam medis termasuk unit kejabaru hal ini dikarenakan baru tahun 2018 rekam medis menjadi unit kerja sendiri, maka perlu diadakannya analisis beban kerja j p y j 3. Berdasarkan hasil perhitungan waktu kerja yang tersedia didapatkan hasil bahwa dalam setahun petugas rekam medis memiliki waktu kerja sebanyak 1.584 jam 3. Berdasarkan hasil perhitungan waktu kerja yang tersedia didapatkan hasil bahwa dalam setahun petugas rekam medis memiliki waktu kerja sebanyak 1.584 jam 4. Berdasarakan identifikasi komponen bebankerja pada unit rekam medis didapatkan hasil bahwa aktivitas pada kegiatan pelayanan terdiri dari filing, asembling, dan pelaporan. Sementara itu untuk kegiatan tambahan terdapat aktivitas ditribusi, perencanaan rekam medis, serta rapat, serta pada kegiatan tambahan terdapat aktivitas pengawasan terhadap mahasiswa yang melakukan kegiatan magang maupun penelitian di Puskesmas Ambulu 5. Berdsarakan penetapan standar aktivitas didapatkan hasil, standara pelayanan yang terdiri dari kegiatan filing selama 5 menit, asembling selama 8 menit, dan pelaporan selama 6 menit. Sementara untuk standar kegiatan pendukung didapatkan hasil didaptakan hasil presentase CAS sebesar 23,4 % dan untuk kegiatan tambahan didapatkan hasil total IAS sebesar 10 jam j 6. Berdasrakan hasil perhitunga standar beban kerja didaptakan hasil bahwa kegiatan filing memliki satuan waktu sebesar 19.008, kegiatan asembling sebesar 11.880 dan kegiatan pelaporan sebesar 15.840 j 6. Berdasrakan hasil perhitunga standar beban kerja didaptakan hasil bahwa kegiatan filing memliki satuan waktu sebesar 19.008, kegiatan asembling sebesar 11.880 dan kegiatan pelaporan sebesar 15.840 p p 7. 4. Simpulan dan Saran 4.1 Simpulan Berdasarakan penentuan kebutuhan staf dengan cara mengalikan standar beban kerja dengan standar kegiatan pendukung yang hasilnya ditambahkan dengan standar kegiatan tambahan didaptakan hasil bahwa jumla petugas sesuai dengan metode WISN sebanyak 3,5163 yang diblatkan menmjadi 4 orang y g j g 8. Berdasarkan hasil analisis dan menfasirkan hasil perhitungan dengan menggunkana metode WISN didapatkan hasil bahwa beban kerja dari petugas rekam medis di Puskesmas Ambulu tinggi dengan jumlah 0,75 9. gg g j 9. Berdasrkan analisis faktor penyebab beban kerja dari petugas yang di kelompokkan menjadi 5 pendekatan yaitu man yang terdiri dari kurangnya petugas serta kurangnya motivasi kerja, semntara untuk machine didapatkan hasil tidak terdapatnya printer serta kurangnya jumla PC, untuk money yaitu tidak tersedianya dana untuk pelatihan petugas serta tidak terdapat dana tunjangan untuk petugas, untuk methode tidak terdapatnya pembagian jabatan pada job description serta struktur organisasi yang ada masi belum jelas, yang terakhir material yang diapatkan hasil kurangnya jumlah rak, kurangnya jumlah kursi, kuranya jumlah map. Hasil penentuan faktor utama penyebab beban kerja serta solusi dengan menggunakan FGD (focus group discussion) didapatkan hasil bahwa bena kerja yang tinggi pada rekam medis disebabkan kurangnya jumlah petugas serta didapatk hasil solusi berupa usulan penambahan petugas. 4.2 Saran Berdasarakan hasil penelitian yang telah dilakukan menenai beban kerja dari petuga rekam medis di Puskesmas Ambulu maka peneliti menyarankan beberapa saran sebagai beriku J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 4.2.1 Bagi Puskesmas g a. Perlu adanya petimbangan bagi pihak puskesmas untuk mengadakan rotasi petugas atau perekrutan tugas baru untuk menjadi petuags di unit rekam medis dikarenakan untuk memenuhi jumlah idealnya masih kurang 1 orang petugas b. Perlu adanya petimbanan untuk diadakannya redesain ruang rekam medis serta penambahan jumlah rak berkas agar jumlah rak sesuai dengan kebutuhan sehingga semua berkas rekam medis dapat disimpan kedalam rak sebagaimana mestinya p p g y c. Perlu adanya evaluasi ulang pada job description di unit rekam medis sehingga terdapat penjelasan mengenai tugas dan wewenang petugas sesuai dengan jabatan yang dimiliki 4.2.2 Bagi Peneliti Selanjutnya a. Iharapkan untuk penelitian selanjutnya untuk melakukan penelitian lanutan mengenai beban kerja mengingt 7 komptensi poko rekam medis salah satunya kegiatan coding masih 591 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 belum dilakukan oleh petugas rekam medis, serta carilah akar masalah penyebab kegiatan tersebut tidak dilakasanakan sebagaimana mestinya g y b. Diharapkan untuk penelitian selanjutnya melakukan penelitian lanjutan mengenai beban keja petugas rekam medis yang dikaitkan dengan stress kerja mengingat unit rekam medis termasuk unit yang baru berdiri semnetraa petugas rekam medis yang tersedia yang merupakan lulusan rekam medi hanya terdapat satu orang sementara sisanya tidak memiliki latar belakang pendidikan rekam medis. b. Diharapkan untuk penelitian selanjutnya melakukan penelitian lanjutan mengenai beban keja petugas rekam medis yang dikaitkan dengan stress kerja mengingat unit rekam medis termasuk unit yang baru berdiri semnetraa petugas rekam medis yang tersedia yang merupakan lulusan rekam medi hanya terdapat satu orang sementara sisanya tidak memiliki latar belakang pendidikan rekam medis. Daftar Pustaka Budi C.S. 2011. Manajemen Unit Kerja Rekam Medis. Yogyakarta. Quantum Sinergis Media Departemen Kesehatan RI. Pedoman Penyusunan Perencanaan SDM kesehatan di Tingkat Provinsi, Kab/Kota serta Rumah Sakit. Jakarta: Direktorat Jenderal Pelayanan Medik, 2004. Garmelia E. 2010. Pedoman Penyelenggaraan Rekam Medis dan Informasi Kesehatan di Rumah Sakit. Direktorat Bina Upaya Pelayanan dan KKMF Kementrian Kesehatan Republik Indonesia. Hattta R.G. 2014. Pedoman Manajemen Informasi Kesehatan di Sarana Pelayanan Kesehatan. Jakarta. Universitas Indonesia Mar’ih R.H. 2017. Panduan Praktis Menyusun Analisis Beban Kerja. Jakarta. Raih Asa Sukses. Menkes. 2008. Permenkes Nomor 269/MENKES/PER/III/2008 Tentang Rekam Medis. Jakarta : Menteri Kesehatan RI. Menkes. 2014. Permenkes Nomor 75 Tentang Pusat Kesehatan Masyarakat. Jakarta : Menteri Kesehatan RI Notoatmodjo, Dr.Soekidjo. Metodelogi Penelitian Kesehatan. Jakarta: Rhineka Cipta, 2002. Retnowati Pranoto H.L. 2015. Analisis Beban Kerja Sumber Daya Manusia Perusahaan. Jakarta. . PPM Manajemen. Rianti Alicia F,L., E. M. (2015). Analisis Beban Kerja Berdasarkan Metode WISN Petugas Assembling di RSDU Tugurejo Semarang Tahun 2015, 1–11. http://eprints.binus.ac.id diakses pada tanggal 25 maret 2019 Rustiyanto, Ery. Etika Profesi Perekam Medis dan Informasi Kesehatan. Jogyakarta: Graha Ilmu, 2009). Setiawan D dan Prasetyo H. 2015. Metodologi Penelitian Kesehatan. Yokyakarta. Graha Ilmu Setiawan D dan Prasetyo H. 2015. Metodologi Penelitian Kesehatan. Yokyakarta. Graha Ilmu S Indradi R 2017 Rekam Medis Banten Universitas Terbuka S Indradi R. 2017. Rekam Medis. Banten. Universitas Terbuka 592 592 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 World Health Organization.2010. Workload Indicators of Staffing Need User’s Manual. France: Ganefa Suryani, F. (2018). Penerapan Metode Diagram Sebab Akibat (Fish None Diagram) dan J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 World Health Organization.2010. Workload Indicators of Staffing Need User’s Manual. France: Ganefa Suryani, F. (2018). Penerapan Metode Diagram Sebab Akibat (Fish None Diagram) dan FMEA (Failure Mode And Effect) Dalam Menganalisa Resiko Kecelakaan Kerja di PT. Pertamina Talisman Jambi Merang. Journal Industrial Servicess, 3(2), 64–69. http://sinta2.risetdikti.go.id/authors/detail?id=6109217&view=overview diakses pada 20 maret 2019 Wardanis, T. . (2018). Analisis Beban Kerja Tenaga Rekam Medis Rumah Sakit Bedah Surabaya Menggunakan Metode FTE. Administrasi Kesehatan Indonesia, 6(1), 43– 60. https://docplayer.info/amp/35796207-Analisis -beban-kerja-dan-kebutuhan- sumber-daya-manusia-oleh-teguh-setiyawan-.html diakses pada 20 maret 2019 Wibawa Nanda, P.R., Sugiono., Remba Yanuar, E. (n.d.). Analisis Beban Kerja Dengan Metode Workload Analysis Sebagai Pertimbangan Pemberian Insentif Pekerja (Studi Kasus di Bidang PPIP PT Barata Indonesia (Persero) Gresik), 672–683. J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 Daftar Pustaka http://srmsi.studentjournal.uci.ac.id/ diakses pada 27 maret 2019 J-REMI : Jurnal Rekam Medik Dan Informasi Kesehatan E-ISSN: 2721-866X Vol. 1 No. 4, September 2020 World Health Organization.2010. Workload Indicators of Staffing Need User’s Manual. France: Ganefa Suryani, F. (2018). Penerapan Metode Diagram Sebab Akibat (Fish None Diagram) dan FMEA (Failure Mode And Effect) Dalam Menganalisa Resiko Kecelakaan Kerja di PT. Pertamina Talisman Jambi Merang. Journal Industrial Servicess, 3(2), 64–69. http://sinta2.risetdikti.go.id/authors/detail?id=6109217&view=overview diakses pada 20 maret 2019 Wardanis, T. . (2018). Analisis Beban Kerja Tenaga Rekam Medis Rumah Sakit Bedah Surabaya Menggunakan Metode FTE. Administrasi Kesehatan Indonesia, 6(1), 43– 60. https://docplayer.info/amp/35796207-Analisis -beban-kerja-dan-kebutuhan- sumber-daya-manusia-oleh-teguh-setiyawan-.html diakses pada 20 maret 2019 y g y p Wibawa Nanda, P.R., Sugiono., Remba Yanuar, E. (n.d.). Analisis Beban Kerja Dengan Metode Workload Analysis Sebagai Pertimbangan Pemberian Insentif Pekerja (Studi Kasus di Bidang PPIP PT Barata Indonesia (Persero) Gresik), 672–683. http://srmsi.studentjournal.uci.ac.id/ diakses pada 27 maret 2019 593
https://openalex.org/W2004450415	https://publications.polymtl.ca/3491/1/2015_Plantard_Pose_Estimation_Kinect_Ergonomic_Studies.pdf	English	null	Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin	Sensors	2,015	cc-by	9,521	Titre: Title: Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin Auteurs: Authors: Pierre Plantard, Edouard Auvinet, Anne-Sophie Pierres, & Franck Multon Date: 2015 Type: Article de revue / Article Référence: Citation: Plantard, P., Auvinet, E., Pierres, A.-S., & Multon, F. (2015). Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin. Sensors, 15(1), 1785-1803. https://doi.org/10.3390/s150101785 Document en libre accès dans PolyPublie Open Access document in PolyPublie URL de PolyPublie: PolyPublie URL: https://publications.polymtl.ca/3491/ Version: Version officielle de l'éditeur / Published version Révisé par les pairs / Refereed Conditions d’utilisation: Terms of Use: CC BY Document publié chez l’éditeur officiel Document issued by the official publisher Titre de la revue: Journal Title: Sensors (vol. 15, no. 1) Maison d’édition: Publisher: MDPI URL officiel: Official URL: https://doi.org/10.3390/s150101785 Mention légale: Legal notice: Titre: Title: Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin Auteurs: Authors: Pierre Plantard, Edouard Auvinet, Anne-Sophie Pierres, & Franck Multon Date: 2015 Type: Article de revue / Article Référence: Citation: Plantard, P., Auvinet, E., Pierres, A.-S., & Multon, F. (2015). Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin. Sensors, 15(1), 1785-1803. https://doi.org/10.3390/s150101785 Document en libre accès dans PolyPublie Open Access document in PolyPublie URL de PolyPublie: PolyPublie URL: https://publications.polymtl.ca/3491/ Version: Version officielle de l'éditeur / Published version Révisé par les pairs / Refereed Conditions d’utilisation: Terms of Use: CC BY Document publié chez l’éditeur officiel Document issued by the official publisher Titre de la revue: Journal Title: Sensors (vol. 15, no. 1) Maison d’édition: Publisher: MDPI URL officiel: Official URL: https://doi.org/10.3390/s150101785 Mention légale: Legal notice: Titre: Title: Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin Auteurs: Authors: Pierre Plantard, Edouard Auvinet, Anne-Sophie Pierres, & Franck Multon Date: 2015 Type: Article de revue / Article Référence: Citation: Plantard, P., Auvinet, E., Pierres, A.-S., & Multon, F. (2015). Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin. Sensors, 15(1), 1785-1803. https://doi.org/10.3390/s150101785 Document en libre accès dans PolyPublie Open Access document in PolyPublie Document publié chez l’éditeur officiel Document issued by the official publisher Titre de la revue: Journal Title: Sensors (vol. 15, no. Sensors 2015, 15, 1785-1803; doi:10.3390/s150101785 sensors ISSN 1424-8220 www.mdpi.com/journal/sensors Article Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin Pierre Plantard 1,2,, Edouard Auvinet 3,4, Anne-Sophie Le Pierres 2 and Franck Multon 1,5 1 M2S Laboratory, University Rennes 2, ENS Rennes, Avenue Robert Schuman, 35170 Bruz, France; E-Mail: Franck.Multon@irisa.fr 2 FAURECIA Automotive Seating, ZI de Brières les Scellés, B.P. 89 91152 Etampes, France; E-Mail: anne-sophie.lepierres@faurecia.com 3 Ecole Polytechnique de Montréal, C.P. 6079, Succursale Centre-ville, Montréal, H3C 3A7 QC, Canada; E-Mail: edouard.auvinet@polymtl.ca 4 CHU Sainte Justine de Montréal, 3175, Chemin de la Côte-Sainte-Catherine, Montréal, H3T1C5 QC, Canada 5 INRIA, MimeTIC team, Campus Universitaire de Beaulieu, 35042 Rennes, France Author to whom correspondence should be addressed; E-Mail: pierre.plantard@irisa.fr; Tel.: +33-290-091-577. Academic Editor: Alexander Star Received: 7 November 2014 / Accepted: 7 January 2015 / Published: 15 January 2015 Abstract: Analyzing human poses with a Kinect is a promising method to evaluate potentials risks of musculoskeletal disorders at workstations. In ecological situations, complex 3D poses and constraints imposed by the environment make it difficult to obtain reliable kinematic information. Thus, being able to predict the potential accuracy of the measurement for such complex 3D poses and sensor placements is challenging in classical OPEN ACCESS Sensors 2015, 15, 1785-1803; doi:10.3390/s150101785 sensors ISSN 1424-8220 www.mdpi.com/journal/sensors OPEN ACCESS sensors ISSN 1424-8220 www.mdpi.com/journal/sensors OPEN ACCESS 1) Maison d’édition: Publisher: MDPI URL officiel: Official URL: https://doi.org/10.3390/s150101785 Mention légale: Legal notice: Ce fichier a été téléchargé à partir de PolyPublie, le dépôt institutionnel de Polytechnique Montréal This file has been downloaded from PolyPublie, the institutional repository of Polytechnique Montréal https://publications.polymtl.ca 1. Introduction Analyzing human postures and movements at workstations is a key issue in order to evaluate potentials risks of musculoskeletal disorders. To this end, methods have been developed to evaluate exposure to risk factors in the workplace. They can be divided into three groups according to the accuracy of the data collection and the measurement technique that is used: self-report, observational methods, and direct measurement [1]. The first group, self-report methods, can take many various forms such as rating scales, questionnaires, checklists, or interviews. This kind of approach focuses on the assessment of physical workload, body discomfort, or work stress, which are difficult to objectively measure. Therefore, although this method is easy-to-use, it is not reliable enough and could result in misleading interpretations [2,3]. The second group, observational methods, consist of direct observation of the worker and his tasks. These are widely used in industry. One of the most well-known methods is the Rapid Upper Limb Assessment (RULA) [4]. This kind of tool requires to rate the pose of the worker, generally based on an estimation of joint angles. These methods are easy-to-use, generally do not require complex setups, and can be used to evaluate a wide range of working tasks [5]. However, since data collection is obtained through subjective observation or simple estimation of projected angles in videos/pictures, the method would be subject to inaccuracy or bias across different observers [6]. Video-based systems have been introduced to partly overcome these limitations, especially to increase accuracy and robustness of joint angle estimation. They do not restrict or disturb the natural motions of the workers [7], but it remains difficult and tedious to obtain 3D information and to place cameras at suitable/unobstructed positions in congested workplaces. Thirdly, direct methods, unlike the others, collect data directly from sensors attached to the worker’s body. They are preferred in a research context, but are difficult to implement in real work situations [1]. Posture assessment with goniometric devices are widely used in ergonomics and provide high accuracy for epidemiologic studies [8]. However, wearing these devices may cause discomfort and affect postural behavior. Moreover, this tool is limited to planar movements, and will become problematic for complex joints, such as the shoulder. Magnetic systems are also used for motion tracking [9]. Pose Estimation with a Kinect for Ergonomic Studies: Evaluation of the Accuracy Using a Virtual Mannequin 2 FAURECIA Automotive Seating, ZI de Brières les Scellés, B.P. 89 91152 Etampes, France; E-Mail: anne-sophie.lepierres@faurecia.com 3 Ecole Polytechnique de Montréal, C.P. 6079, Succursale Centre-ville, Montréal, H3C 3A7 QC, Canada; E-Mail: edouard.auvinet@polymtl.ca 4 CHU Sainte Justine de Montréal, 3175, Chemin de la Côte-Sainte-Catherine, Montréal, H3T1C5 QC, Canada 5 INRIA, MimeTIC team, Campus Universitaire de Beaulieu, 35042 Rennes, France 5 INRIA, MimeTIC team, Campus Universitaire de Beaulieu, 35042 Rennes, France * Author to whom correspondence should be addressed; E-Mail: pierre.plantard@irisa.fr; Tel.: +33-290-091-577. Academic Editor: Alexander Star Academic Editor: Alexander Star Received: 7 November 2014 / Accepted: 7 January 2015 / Published: 15 January 201 Abstract: Analyzing human poses with a Kinect is a promising method to evaluate potentials risks of musculoskeletal disorders at workstations. In ecological situations, complex 3D poses and constraints imposed by the environment make it difficult to obtain reliable kinematic information. Thus, being able to predict the potential accuracy of the measurement for such complex 3D poses and sensor placements is challenging in classical experimental setups. To tackle this problem, we propose a new evaluation method based on a virtual mannequin. In this study, we apply this method to the evaluation of joint positions (shoulder, elbow, and wrist), joint angles (shoulder and elbow), and the corresponding RULA (a popular ergonomics assessment grid) upper-limb score for a large set of poses and sensor placements. Thanks to this evaluation method, more than 500,000 configurations have been automatically tested, which would be almost impossible to evaluate with classical protocols. The results show that the kinematic information obtained by the Kinect software is generally accurate enough to fill in ergonomic assessment grids. However inaccuracy strongly increases for some specific poses and sensor positions. Using this Sensors 2015, 15 1786 evaluation method enabled us to report configurations that could lead to these high inaccuracies. As a supplementary material, we provide a software tool to help designers to evaluate the expected accuracy of this sensor for a set of upper-limb configurations. Results obtained with the virtual mannequin are in accordance with those obtained from a real subject for a limited set of poses and sensor placements. Keywords: Kinect; accuracy; virtual mannequin Keywords: Kinect; accuracy; virtual mannequin 1. Introduction They can continuously measure joint motion in six degrees of freedom, but are difficult to use on site because of perturbations of the magnetic fields due to machines and ferromagnetic objects. Finally, inertial sensors (such as accelerometers or gyroscope sensors) can accurately assess human pose [10,11]. However, these devices can be disturbed by the environment conditions of the real work Sensors 2015, 15 Sensors 2015, 15 1787 situations (such as vibrations). All these wearing sensors are sensitive to placement on the body, making their use in the workplace difficult. Advances in the field of markerless motion capture can potentially offer an opportunity to overcome these limitations of disturbing the natural motion of workers. Recent development of low-cost depth cameras such as the Microsoft® Kinect sensor provides easy-to-use, markerless, calibration-free and cheap alternatives. This device (based on PrimeSense Technology, Tel Aviv, Israel) [12–14] is composed of an infrared projector of structured light and an infrared camera that returns a depth image of the scene at 30 Hz. The associated software enables us to identify pixels that belong to the body parts by the use of a randomized decision forest algorithm. This body part classification finally leads to the estimation of twenty 3D joint positions of the main joints in the human body [15]. Several authors have studied the accuracy of the kinematic data provided by the Kinect in various application domains [16–19]. The hardware accuracy has been investigated through the study of the depth map [20–22]. The results show that the hardware component of the Kinect is accurate enough to capture 3D objects in a workplace environment [20]. The accuracy and reliability of the kinematic data provided by the Shotton’s algorithm have also been investigated, especially in rehabilitation. The joint positions [23,24] and the resulting joint angle errors [16,18,24,25] look promising for most clinical uses. However, only a few postures were studied in previous works, mostly in clinical analysis for very limited and standardized postures performed in 2D anatomical plans. Although the Kinect looks promising for capturing 3D data of work activities over periods of time, it raises new questions that have not been addressed yet. While several Kinect-based applications were developed to improve the postural assessment of observational methods [26–28], the reliability of the angular joint values delivered by the systems have not been accurately evaluated in ergonomic applications yet. 1. Introduction Indeed, the workplace environment is often crowded with many potential occlusions that block a video-based system from seeing the subject’s body properly, leading to potential critical errors [29]. Moreover, placing the Kinect at an ideal position, i.e., 2 m away from the subject at a convenient height, is usually impossible. According to the Kinect recommendations (NUI Kinect for Windows 1.8) the accuracy of the sensor decreases when these ideal conditions are not satisfied. Thus, it also seems necessary to determine the influence of the Kinect view angle relative to the worker, as suggested in [26]. Evaluating a wide range of poses and sensor configurations is a difficult task, especially if we need to consider several repetitions for each condition. To address this problem, [30] suggests using simulation with an anthropometric 3D mesh. They proposed to deform the anthropometric 3D mesh to mimic the desired poses and simulate the resulting depth image of the scene, as a Kinect would do under the same conditions. Using a simulated depth map representing a 3D mesh instead of real humans before running the Shotton’s algorithm enables us to test a large set of configurations and repetitions, as Shotton did in [15]. It also enables us to test the kinematic data reconstruction algorithm independently from the hardware device. The quality of the depth images measured with a Kinect has been extensively evaluated in previous works [20–22] and the new version of the Kinect offers better quality. However the accuracy of the kinematic reconstruction software is less clear whatever the quality of the sensor is, especially for complex 3D poses in constrained environments, such as the workplaces. Using rendered virtual mannequins as an input of this kinematic reconstruction software Sensors 2015, 15 Sensors 2015, 15 1788 will enable us to clarify the reliability of this estimation in realistic conditions, independently from the sensor’s accuracy. In this work, we consequently propose to evaluate the accuracy of kinematic data computed by the Kinect software according to simulated depth images associated with a wide set of poses and sensor positions. The results of this work aims at providing the reader with details about the accuracy of the Kinect to help experimenters use this sensor in an optimal way under workplace conditions. Section 2 provides details about the material and methods used to achieve this goal. 2. Material and Methods In this section, we describe the methods used to evaluate the accuracy of the Kinect measurements with numerical mannequins. One of the main difficulties is to accurately control, maintain, and repeat the poses performed by the subjects. Moreover it is difficult to accurately standardize the environment, which would also be captured by the sensor. Finally, designing experimental protocols to differentiate errors due to the sensor from those due to the pose estimation method is almost impossible. To overcome these limitations we propose to simulate numerical mannequins: (1) it provides us with accurate control of the poses and reliable ground truth data; (2) it enables us to focus on the errors caused by the pose estimation algorithm, assuming that the sensor is noise-free; (3) a given pose can be generated and tested several times to obtain more reliable statistical analysis; and (4) several sensor placements can be tested, which is complex to carry-out in real situations in an actual standardized manner for each tested human pose. Using a virtual mannequin instead of real subjects could introduce biases in the results, especially if the skinning of the character is not totally in accordance with a real human silhouette. Thus, to ensure that results obtained with the virtual mannequin could be used to estimate errors in real humans, we carried out an experiment with a real subject. The pose of the virtual mannequin is set to match its joint centers positions to those of the real subject to compare (1) the joint angles measured with the virtual mannequin and the simulated Kinect; (2) the actual joint angles of the skeleton; and (3) the joint angles provided by the actual Kinect sensor and software. 1. Introduction Section 3 reports the results, i.e., the accuracy of the reconstructed poses using a Kinect thanks to virtual mannequins. Results obtained with the virtual mannequin have been compared to actual measurements from a real subject for a limited set of poses in order to ensure that the virtual mannequin and the sensor simulation did not lead to unrealistic results. Section 4 discusses the result before a global conclusion. 2.1. Method Overview The pipeline of the method can be divided in three parts, as shown in Figure 1. In the first part, an anthropometric 3D mesh representing the surface of the human body is generated using MakeHuman software [31], as suggested in [30]. A skeleton is associated with the mesh to control the pose of this virtual human. As a consequence, changes in joint configuration leads to an adaptation of the 3D mesh thanks to linear blend skinning method available in the Blender software [32]. The resulting 3D mesh is exported in a standard obj file and the corresponding joint positions are exported in a separated text file. 1789 Sensors 2015, 15 Sensors 2015, 15 Figure 1. Overall method pipeline. (a) 3D meshes in specific poses, joint position values of reference (xref, yref, zref in meters), computed joint angle values of reference (αref, βref, γref in degrees), and computed RULA score (Rularef); (b) Depth images of the mesh in specific sensor positions; (c) Analysis of the depth images with the Kinect software for joint localization [15]; (d) Estimated joint position values (x’, y’, z’ in meters), computed estimated joint angle values (α’, β’, γ’ in degrees), and computed estimated RULA score (Rula’); (e) Measure error between values of reference and estimated values. Figure 1. Overall method pipeline. (a) 3D meshes in specific poses, joint position values of reference (xref, yref, zref in meters), computed joint angle values of reference (αref, βref, γref in degrees), and computed RULA score (Rularef); (b) Depth images of the mesh in specific sensor positions; (c) Analysis of the depth images with the Kinect software for joint localization [15]; (d) Estimated joint position values (x’, y’, z’ in meters), computed estimated joint angle values (α’, β’, γ’ in degrees), and computed estimated RULA score (Rula’); (e) Measure error between values of reference and estimated values. ure 1. Overall method pipeline. (a) 3D meshes in specific poses, joint position values In the second part, the behavior of the Kinect sensor is simulated by transforming the 3D mesh into depth images according to the imposed virtual sensor position and orientation in space, and the camera-intrinsic parameters. This way, we assume that the sensor is noise-free and is able to deliver accurate depth maps. The resulting 3D mesh is transformed into a depth map that is used as an input of the Kinect software. 2.1. Method Overview This software estimates the 3D position of the joint centers according to the Shotton’s algorithm [15]. Let us consider now the method used to compare the results obtained with the Kinect software to the reference data (i.e., the joint angles applied to the virtual mannequin). Sensors 2015, 15 Sensors 2015, 15 1790 RULA score. Consequently, in this paper, we focused on three main parameters: joint position (shoulder, elbow, and wrist joint), joint angles (shoulder and elbow joint angles), and the corresponding RULA upper-limb score. According to the estimated joint positions, joint angles should be computed using the ISB recommendation [33]. However the skeleton estimated by the Shotton’s method is not compatible with this recommendation except for the humerus coordinate system. For the other joint coordinate systems, we have to proceed slightly differently according to the available points. For the trunk, the X-axis is defined as the normal of the plan formed by the spine joint (placed at the pelvis), the basis of the neck, and the right shoulder joint. The right shoulder joint is chosen to compute the trunk reference frame because it does not move significantly when performing left-arm motions. As the Kinect software does not deliver all the required kinematic information to compute the forearm reference frame according to the ISB recommendations, we alternatively use the vector convention detailed by [18]. Once the joint angles are known it is possible to compute RULA upper-limb scores [4]. For the shoulder joint, we need to isolate the flexion/extension angle and the abduction/adduction angles. The ISB guidelines propose to use three Eulerian angles to define the shoulder motion. The first one is the plane of elevation, which indicates if the elevation is an abduction (0°) or a forward flexion (90°). The second one is the elevation value of the shoulder. With these two Euler angles, we can compute the flexion and the abduction angles. As there is no available information for the wrist joint, the wrist, and wrist twist RULA scores are set the minimum values. Briefly, RULA is an assessment grid that consists of associating risks of musculoskeletal disorders to joint angles, repetition, and loads. A score is associated with joint angle intervals for each joint (such as 1 score for an elbow flexion in the [60°–100°] interval). Tables are used to compute a unique score (between 1 and 7) for each upper-limb and for the entire body taking repetition and loads into account. Hence RULA scores vary in a discontinuous manner between 1 and 7 according to continuous joint angles. 2.2. Kinematic Parameters Used to Estimate the Accuracy of the Kinect Software In this subsection, we describe the method used to compute joint angles according to the joint positions estimated by the Kinect software. First of all, one has to notice that if a pose sent to the Kinect is too different from the previous one, the system may need several frames to converge to a stable estimation. In this work we wish to send poses without any continuity in time and we consequently have to let the system converge to this stable solution. In the context of this experiment we have shown that the system converges after a maximum of 5 images. To take a security margin into account, we sent each pose to the Kinect software 15 times to guarantee reaching a stable estimation. In the end, only the final pose (the 15th one) is used in the following to analyze the accuracy of the system. Once a stable pose is obtained, we have to compare the results to the reference values that were used to generate the 3D mesh of the virtual mannequin. The Kinect software only returns joint positions. An error in joint position would have a direct impact on the computation of joint angles and 2.3. Protocol In order to assess the accuracy of the Kinect in simulated workplace conditions, a large set of work poses (left-arm motions) in different sensor positions have been investigated. In this paper, we simulate poses limited to the reachable workspace frequently used in ergonomics [34]. Hence, Figure 2a shows that each pose can be defined by the relative position of the hand in the shoulder reference frame inside this reachable workspace. This position is defined by three parameters: azimuth (0° to 110° with a 10° step), elevation (−45° to 45° with a 10° step), and depth (associated with the elbow flexion ranging from 0° to 110° with a 10° step). If the hand position is fixed, the elbow is still free to swivel about a circular arc whose normal vector is parallel to the axis from the shoulder to the hand. Hence, we have chosen to sample the swivel angle in three main values: 0°, 90°, and 135° which corresponds to the main kinds of grips one can see in industrial work (see Figure 2b). Our parameters produce 4752 various poses. In this study, we have focused on the left elbow and shoulder joints, as the upper-limb is one of the most studied body parts in ergonomics (i.e., in [35]). 1791 Sensors 2015, 15 Figure 2. (a) Parameters of the experimental setup, the left hand reaching volume in blue (azimuth, elevation and elbow flexion) and the positions and orientations of the sensor in dark (azimuth and elevation); (b) The type of grip parameter: Grip from below at the left (0° swivel angle), grip from the side at the middle (90° swivel angle) and grip form above at the right (135° swivel angle). Figure 2. (a) Parameters of the experimental setup, the left hand reaching volume in blue (azimuth, elevation and elbow flexion) and the positions and orientations of the sensor in dark (azimuth and elevation); (b) The type of grip parameter: Grip from below at the left (0° swivel angle), grip from the side at the middle (90° swivel angle) and grip form above at the right (135° swivel angle). Another important parameter is the placement of the Kinect as it is almost impossible to satisfy the constructor recommendation (placing the Kinect exactly 2 m in front of the subject) in actual industrial conditions. This is mainly due to occlusions in the workplace and the displacement of operators in this area. 2.3. Protocol Thus the accuracy measured in perfect situation may underestimate the actual inaccuracy in workplace conditions. Consequently, we also tested the impact of the sensor’s location on the accuracy of joint angle measurements. Thus, for each pose, the sensor position is defined relatively to the worker’s position by two parameters: azimuth (−50° to 50° with a 10° step) and elevation (−50° to 50° with a 10° step), as described in Figure 2a. In this figure, the large set of Kinect positions is depicted in black. In each configuration, the virtual Kinect sensor is oriented to always look at the middle of the worker. 121 different sensor positions were consequently tested for each pose. Combining the number of poses with the number of sensor positions produces more than 500,000 configurations, which is impossible to evaluate in real situations. For each of these configurations we compute the root mean square error (RMSE) between the resulting parameters (joint position, angle, and RULA score) and reference values applied to the virtual mannequin. Sensors 2015, 15 Sensors 2015, 15 To this end, we asked a subject (age: 30, height: 175 cm, mass: 70 kg, right-handed) to perform a set of poses similar to those described above (i.e., sampling the reachable space of the left arm). The subject was asked to make circles with his hand with three different elevations relative to the horizontal plane (−45°, 0°, 45°). These movements involve all the degrees of freedom of the left shoulder and elbow (see Figure 3). In this study, we only tested the configurations with a 90° swivel angle (see Figure 2b). Each circle for the given arm elevation was repeated five times. Figure 3. Examples of movement performed by the subject with 0° of elevation. Figure 3. Examples of movement performed by the subject with 0° of elevation. Reflective markers were placed over standardized anatomical landmarks, in accordance with the International Society of Biomechanics recommendations [33]. The 3D positions of the reflective markers was collected with a 100 Hz Vicon-MX system (product of Oxford Metrics, Oxford, UK) composed of nine 4-Mpixels cameras. The Kinect sensor was placed 2 m in front of the subject, as recommended by [20]. The Kinect skeleton model was directly obtained at 30 Hz from the official Microsoft software development kit (Microsoft Kinect SDK v1.8) based on the Shotton’s algorithm [15]. Let us denote the data provided by the Kinect as RK, and data provided by the Vicon motion capture system as MBS. The two systems were synchronized thanks to the inter-correlation method suggested in [36]. Data were also resampled to obtain the same timestamp to facilitate comparison. A fourth order low-pass Butterworth filter with a cut-off frequency of 6 Hz was applied on the angle data for both systems [18]. The root mean square error (RMSE) and the Pearson correlation between the three estimation methods is performed for the upper-limb joint angles. Each reference pose measured on the subject with (MBS) was then retargeted to drive the numerical mannequin in a similar joint configuration using inverse kinematics, as reported in [37]. The position of the elbow and wrist joint centers were used as constraints of the inverse kinematics algorithm. This optimization method returns the joint angles minimizing the distance between virtual and real joint centers for the numerical mannequin. Finally we applied the method described in Section 2.1 to compute expected outputs of the Kinect software based on the resulted virtual mannequin. 2.4. Comparison with Real Human Using numerical mannequins to evaluate the accuracy of human limb motions is very interesting, as it enables us to control the test conditions with high precision for a huge set of configurations. However, it is natural to question if evaluations based on simplified numerical mannequins instead of real humans can lead to realistic results. To evaluate if the results obtained from the numerical mannequin are similar to those obtained from a real human, we carried out a complementary pilot study for a limited set of body configurations. We assume that similar joint angles in simulated and real measurements with an actual Kinect for a subset of configurations could help to ensure that the protocol based on the virtual mannequin is reliable. 1792 Sensors 2015, 15 Let us denote the data provided by this simulation of the Kinect sensor with the virtual mannequin as VK. The overall process is depicted in Figure 4. nsors 2015, 15 1 Figure 4. Experimental setup of the comparison of joint angles estimated with an optoelectronic motion capture system (MBS), the corresponding actual Kinect measurements (RK), and the simulated outputs using a virtual mannequin (VK). ensors 2015, 15 17 Figure 4. Experimental setup of the comparison of joint angles estimated with an optoelectronic motion capture system (MBS), the corresponding actual Kinect measurements (RK), and the simulated outputs using a virtual mannequin (VK). 1793 Sensors 2015, 15 Sensors 2015, 15 Figure 4. Experimental setup of the comparison of joint angles estimated with an optoelectronic motion capture system (MBS), the corresponding actual Kinect measurements (RK), and the simulated outputs using a virtual mannequin (VK). 3.1. Evaluation of the Pose Estimation Accuracy for a Kinect Placed in Front of the Character The first part of the results shows the accuracy of the estimated kinematic data when only the pose is changing, with a Kinect placed in front of the subject. Out of 4752 tested poses, we report the highlight of the results that seem representative of the global results (all the results are available as supplementary material: an executable program providing accuracy for all the possible upper-limb configurations). Figure 5 depicts the estimation error for the 135° swivel angle for all the azimuth and elevation configurations, with no elbow flexion. The first row depicts the error for the shoulder, elbow, and wrist joint positions. The second row depicts the error for the shoulder and elbow joint angles. The resulting RULA upper-body score is shown in the last row. These results show that the accuracy of the shoulder joint position decreases with the growth of the azimuth pose parameter (Figure 5a left). However, the position error for the shoulder remains more stable than other joints (0.019 ± 0.009 m). For the elbow and wrist joint positions, a peak error is found in the same body configuration (azimuth near to 90° and elevation near to −10°) whereas there is no peak for the shoulder joint. The elbow and wrist errors increase to 0.16 m and 0.27 m respectively for this pose. For the rest of the pose configurations, the accuracy of the joint positions is acceptable; the overall average is 0.018 ± 0.023 m for the elbow joint and 0.024 ± 0.038 m for the wrist joint. As a consequence, the same peak error occurs for the shoulder and elbow angles (Figure 5b). The peak error is greater than 40° and the overall average error is 4.5° ± 8.9° and 12.6° ± 17.2° for the shoulder and elbow angles respectively. The same type of error is found for the RULA score in the last row of Figure 5c. Some other error peaks were measured for other body configurations, but they were due to 1794 Sensors 2015, 15 limitations of the RULA score. It should be noted that RULA scores were attributed to ranges of angular values. Hence, two very close angular values near the boundaries of these ranges may lead to two different scores. For example, the peak error found at 40° of azimuth and 10° of elevation resulted from the reference shoulder flexion of 46°, which leads to a RULA score of 3, while the estimated shoulder flexion is 44° leads to a score of 2. Figure 5. Accuracy of the Kinect measurement of the 135° swivel angle poses relative to azimuth and elevation pose parameters and with a zero elbow flexion (a) Error distribution of the shoulder (left), elbow (center) and wrist (right) joint positions estimated; (b) Error distribution of the shoulder (left) and elbow (right) joint angles calculated; (c) Error distribution of the resulting upper-body RULA score. Figure 5. Accuracy of the Kinect measurement of the 135° swivel angle poses relative to azimuth and elevation pose parameters and with a zero elbow flexion (a) Error distribution of the shoulder (left), elbow (center) and wrist (right) joint positions estimated; (b) Error distribution of the shoulder (left) and elbow (right) joint angles calculated; (c) Error distribution of the resulting upper-body RULA score. Figure 6 depicts the evolution of the shoulder angle error according to three parameters: azimuth, elevation, and elbow flexion. The error increases with the elbow flexion parameter for an azimuth values close to 0° and an elevation value close to −50°. The error peak measured in Figure 5b also appears when elbow flexion is null (azimuth close to 90° and elevation close to −10°). However, this error area disappears when the elbow flexion increases. For little elbow flexion (bottom part of Figure 6), except for this particular area, the shoulder angle is accurately estimated (less than 8°). On the contrary, for high azimuth values (close to 110°), elevation (close to 50°), and elbow flexion (close to 110°), inaccuracy remains significant (values up to 51°). Finally the estimation error of the shoulder angle is 1795 Sensors 2015, 15 Sensors 2015, 15 more important when the elbow flexion parameter increases in the left part of the Figure 6 (azimuth close to 0° and elevation close to −50°). Figure 6. Accuracy of the Kinect measurement of the shoulder angle (0° swivel angle) according to three parameters: azimuth, elevation, and elbow flexion. Figure 6. Accuracy of the Kinect measurement of the shoulder angle (0° swivel angle) di t th t i th l ti d lb fl i Figure 6. Accuracy of the Kinect measurement of the shoulder angle (0° swivel angle) according to three parameters: azimuth, elevation, and elbow flexion. Figure 7 illustrates several examples of misestimated poses for a Kinect placed in front of the mannequin. For each selected pose, the figure depicts its error map and the corresponding mesh. Each row depicts a pose associated with a specific swivel angle (a: 0°, b: 90°, and c: 135°). The left column of this figure depicts the shoulder joint angle error and the right column depicts the elbow joint angle error. For this figure, only poses with error values greater than 30° were selected. Pose estimation dropouts are clearly visible for all the examples and are highlighted with dotted square. Sensors 2015, 15 1796 Figure 7. Example of misestimated poses with a Kinect placed in front of the subject for the shoulder joint (left column) and the elbow joint (right column). Dotted squares show the graphic areas selected. (a) Swivel angle at 0°; (b) Swivel angle at 90°; (c) Swivel angle at 135°. Figure 7. Example of misestimated poses with a Kinect placed in front of the subject for the shoulder joint (left column) and the elbow joint (right column). Dotted squares show the graphic areas selected. (a) Swivel angle at 0°; (b) Swivel angle at 90°; (c) Swivel angle at 135°. 3.2. Evaluation of the Pose Estimation Accuracy for Various Kinect Placements The second part of the results shows the impact of the Kinect placement relative to the subject on pose estimation accuracy. The RMSE is computed for all the poses relative to the azimuth and elevation of the sensor placement (see Figure 8). Row (a) depicts the RMSE distribution of the estimated joint positions of the shoulder, elbow, and wrist. Row (b) illustrates the RMSE for the shoulder and elbow joint angles. Finally the impact of the error on the upper-body RULA score is shown in Row (c). First of all, results show that the RMSE is more important when the sensor is placed at a low position (sensor elevation = −50°), especially for large azimuth values. Indeed, the sensor placed on the opposite side of the arm measured (sensor azimuth = −50°) leads to misestimation of the shoulder, elbow, and wrist joint positions (Figure 8a), up to 0.18 m, 0.18 m, and 0.15 m respectively. However, one can notice that when the sensor is placed on the same side as the measured arm (sensor azimuth = 50°) it leads to a significant RMSE for the shoulder position (Figure 8a left), up to 0.17 m. Sensors 2015, 15 1797 This shoulder position error consequently leads to high errors in joint angles (shoulder and elbow errors up to 56° and 41° respectively) and RULA score up to 1.16 (Figure 8b,c). This shoulder position error consequently leads to high errors in joint angles (shoulder and elbow errors up to 56° and 41° respectively) and RULA score up to 1.16 (Figure 8b,c). Figure 8. Root mean square error (RMSE) of all poses relative to the sensor placement (azimuth and elevation). (a) RMSE distribution of the shoulder (left), elbow (center), and wrist (right) joint positions estimated; (b) RMSE distribution of the shoulder (left) and elbow (right) joint angles calculated; (c) RMSE distribution of the resulting upper-body RULA score. Figure 8. Root mean square error (RMSE) of all poses relative to the sensor placement (azimuth and elevation). (a) RMSE distribution of the shoulder (left), elbow (center), and wrist (right) joint positions estimated; (b) RMSE distribution of the shoulder (left) and elbow (right) joint angles calculated; (c) RMSE distribution of the resulting upper-body RULA score. 4. Discussion In this study, we have estimated the accuracy of the Kinect in simulated workplace conditions for a large set of work poses and with different sensor positions. This paper aims at providing the reader with details about the accuracy of the Kinect software to provide optimal methods for its use in workplace conditions. Our results are in accordance with previous works, especially for the accuracy of the estimated joint position [23,24] and joint angles [24,25]. It remains difficult to compare our work with others because most of them generally measure planar gestures in 2D anatomical plans. Previous studies were also limited to a Kinect placed in front of the subject, and the impact of the device position relative to the subject was generally not evaluated. Only one work measured the accuracy of the Kinect for different orientations of the subject, ranging from the frontal to the side view [23]. However, only four orientations were tested, which is insufficient to evaluate the impact of the device position relative to the subject. As the method presented in this paper uses virtual mannequins to simulate the depth images, on can consider that the evaluation method is independent of the hardware device and focuses on the error about the pose estimation method. The first part of the results showed that the pose estimation algorithm was acceptable for most applications (for example the shoulder position error is 0.025 m), except for a few specific body configurations. For these poses, the error can grow up to more than 40° for the shoulder joint angle (see Figures 5–7). In these cases, some body parts are partially occluded by other body segments in the Kinect axis (Figure 7) and it is almost impossible for the Kinect to correctly perceive the arm and the forearm. For example, in Figure 7b left, the error of the shoulder joint angle is high when the arm is straight and aligned with the Kinect axis. In this body configuration the segments are not well measured and the estimation of the joint center location is not accurate. In the second part of the results, we find that that the sensor position could have an important impact on the pose estimation accuracy. Specifically, the elevation of the Kinect has a greater impact compared to azimuth (Figure 8). 3.3. Comparison to Results Obtained with Real Human Results are summarized in Table 1. When comparing motion capture data (MBS) to real (RK) and virtual Kinect (VK), one can see that error with the virtual Kinect is lower than those of the real Kinect. This is compatible with the fact that the virtual Kinect is noise-free compared to the real one, leading to lower error. For the shoulder elevation angle, comparison between RK and VK exhibits an 8.5° (±3.6) RMSE, and a Pearson correlation of p = 0.86. For the elbow joint angle, comparison between RK and VK exhibits a 13° (±2.3) RMSE, and a Pearson correlation of p = 0.96. The results of this pilot study tend to show that our evaluation method based on a virtual mannequin exhibit similar results than when using a real Kinect. 1798 Sensors 2015, 15 Sensors 2015, 15 Table 1. Root mean square error (RMSE) is expressed in degree (st. dev.), p = Pearson coefficient of correlation. Table 1. Root mean square error (RMSE) is expressed in degree (st. dev.), p = Pearson coefficient of correlation. Table 1. Root mean square error (RMSE) is expressed in degree (st. dev.), p = Pearson coefficient of correlation. Table 1. Root mean square error (RMSE) is expressed in degree (st. dev.), p = Pearson coefficient of correlation. MBS-RK MBS-VK RK-VK Shoulder joint angle RMSE 9.5° (±2.9) 5.2° (±1.5) 8.5° (±3.6) p 0.83 0.92 0.86 Elbow joint angle RMSE 11.4° (±2.0) 8.2° (±1.3) 13.0° (±2.3) p 0.97 0.98 0.96 MBS-RK MBS-VK RK-VK Shoulder joint angle RMSE 9.5° (±2.9) 5.2° (±1.5) 8.5° (±3.6) p 0.83 0.92 0.86 Elbow joint angle RMSE 11.4° (±2.0) 8.2° (±1.3) 13.0° (±2.3) p 0.97 0.98 0.96 Di i 4. Discussion Indeed, a Kinect placed at a low position (sensor elevation = −50°) leads to a large error of the pose estimation (up to 0.169 m for the shoulder position error), especially for large azimuth values (sensor azimuth = −50° and 50°). This particular sensor placement lead to partial occlusions of body parts of interest. In other cases, the kinematic information provided by the Kinect software is acceptable for most applications (0.052 ± 0.03 m for the shoulder position error). One has to consider that results could be worse than those reported in this paper when using a real Kinect because of noise that was not taken into account here. In the third part of the results, the measured joint angles of a real human were correlated with those obtained with a real Kinect and method simulated Kinect based on the virtual mannequin. In this pilot Sensors 2015, 15 1799 Sensors 2015, 15 study, the errors reported when using our method were similar to those obtained with a real Kinect. However, one can notice that the results obtained with the simulated Kinect were slightly better than those obtained with the real Kinect, which is in accordance to the fact that the virtual sensor is noise-free while the actual sensor is noisy. The results of this study highlight the importance of the design of the motion capture protocol with a Kinect. Unlike other works, this study aims to evaluate both complex poses and various sensor placements. On one hand, these situations are far from ideal conditions for Kinect measurements. On the other hand, they tend to mimic natural situations encountered in actual workplaces. The results presented in this paper can help experimenters to resolve constraints imposed by these ecological situations. In this paper almost 500,000 configurations (body pose and Kinect position) have been tested. Thus, this evaluation method provides a huge amount of data, which makes it difficult to present in a compact manner in this contribution. We have chosen to highlight some of the extreme values and an overall average error value. In this study, the RULA evaluation is limited to the shoulder and elbow joints. It will be interesting to measure the other body parts requested by the RULA method (trunk, neck, wrist, and leg), to quantify their impacts on the resulting RULA score. 5. Conclusions The study presented in this work suggests that the Kinect software can be a useful motion capture tool for ergonomic evaluation. In most of the results reported in this work, the accuracy is enough to correctly fill-in ergonomic assessment grids like the RULA grid. However large errors can occur in specific cases that have been reported in this paper, such as when the arm is aligned with the Kinect sensor. We also noticed that sensor placement may lead to bad estimations. Hence, an experimenter has to carefully design his protocol according to these results. He has to deal with constraints of the environment (such as occlusions or possible sensor placements) while taking these recommendations into account. The results reported in this study have been obtained with a new systematic method based on a virtual mannequin. It allows us to automatically estimate the accuracy of the system for numerous complex poses and sensor positions, which would not be possible with classical methods and protocols. Moreover, this method focuses on the accuracy of the pose estimation algorithm provided by the official Kinect software. It assumes that the device is noise-free and delivers perfect depth images. Hence results with an actual Kinect sensor would certainly be worse than those reported in this paper. However, we assumed that being able to separate errors due to the sensor from those due to the method is a strong advantage, even for future upgrades of the measurement systems, such as the new Kinect 2. It would be possible to add noise to the simulated depth images used as an input in this work to simulate potential noise of the sensor. However, in this paper, we reported numerous results about the impact of experimental setups on the skeleton reconstruction, assuming perfect 3D data provided by the Kinect. Dealing with depth images inaccuracies would again increase the complexity of the results reported in this paper. In the future, it would be interesting to develop an online test bench based on the virtual mannequin and sensors. It would help experimenters to test the preliminary expected accuracy for their recordings according to the experimental conditions. It would also enable us to extend this accuracy analysis to the entire body. From the ergonomic point of view, correctly using a Kinect in actual workplace conditions would enable ergonomists to analyze motions instead of isolated poses. 5. Conclusions It provides supplementary temporal information, such as the time spent above a given RULA score. It could also be used as real-time feedbacks, as suggested in [39]. 4. Discussion To this end it would be necessary to apply the same method to these joint angles in a new protocol. The Kinect software does not deliver all the required kinematic information to compute the reference frames according to the ISB recommendations. In this work we slightly adapted the trunk reference frame to compute the joint angles of the shoulder. We also used an alternative computation for the elbow joint, as proposed by [18]. This lack of information would be a problem if we implement a biomechanical model from the Kinect data, such as using inverse dynamics to estimate joint torques and forces. To tackle this problem, [38] developed a model-based approach to enhance the anatomical accuracy of the standard skeleton provided by the Kinect. It could also be applied to the method developed in this paper in order to satisfy the ISB recommendations. The method presented in this paper relies on a virtual mannequin, which strongly influenced the results. Hence the shape and the anatomical model of this mannequin could be slightly different from that of a real human, and thus could drive the evaluation to inappropriate results. In the future, it would be essential to evaluate the sensibility of the evaluation method to the choice of the mannequin. For example, we could compare the results obtained with a large set of virtual mannequins. Moreover, it would be interesting to compare results obtained from a real human with those obtained from the virtual mannequin, with a larger set of subjects. This method allowed us to evaluate specific poses and sensor configurations in an accurate, reliable, and standardized manner, even if the intrinsic noise of the device were not considered here. The process is fully automatic. It enables an experimenter to quickly perform an evaluation. In addition, it is also possible to simulate occlusions that could occur in real situations by adding virtual objects in the environment. For example, we could add clothing to the virtual mannequin, or a specific workstation in front of him. Sensors 2015, 15 Sensors 2015, 15 1800 Acknowledgments This work was partially funded by the Cifre convention N° 1222/2012 and Faurecia Company. References 1. Li, G.; Buckle, P. Comparison of methods for the assessment of postural load on the back. Ergonomics 1999, 42, 674–695. 1. Li, G.; Buckle, P. Comparison of methods for the assessment of postural load on the back. Ergonomics 1999, 42, 674–695. 2. Burdorf, A.; Laan, J. Current techniques for assessing physical exposure to work-related musculoskeletal risks, with emphasis on posture-based methods. Scand. J. Work. Environ. Health 1991, 17, 425–429. 2. Burdorf, A.; Laan, J. Current techniques for assessing physical exposure to work-related musculoskeletal risks, with emphasis on posture-based methods. Scand. J. Work. Environ. Health 1991, 17, 425–429. 3. Wiktorin, C.; Karlqvist, L.; Winkel, J. Validity of self-reported exposures to work postures and manual materials handling. Stockholm MUSIC I Study Group. Scand. J. Work. Environ. Health 1993, 19, 208–214. 3. Wiktorin, C.; Karlqvist, L.; Winkel, J. Validity of self-reported exposures to work postures and manual materials handling. Stockholm MUSIC I Study Group. Scand. J. Work. Environ. Health 1993, 19, 208–214. 4. McAtamney, L.; Corlett, E.N. RULA: A survey method for the investigation of work-related upper limb disorders. Appl. Ergon. 1993, 24, 91–99. 4. McAtamney, L.; Corlett, E.N. RULA: A survey method for the investigation of work-related upper limb disorders. Appl. Ergon. 1993, 24, 91–99. 5. David, G.C. Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders. Occup. Med. 2005, 55, 190–199. 6. Burdorf, A.; Derksen, J.; Naaktgeboren, B.; van Riel, M.; Measurement of trunk bending during work by direct observation and continuous measurement. Appl. Ergon. 1992, 23, 263–267. 7. Chang, C.C.; Hsiang, S.; Dempsey, P.G.; McGorry, R.W. A computerized video coding system for biomechanical analysis of lifting tasks. Int. J. Ind. Ergonom. 2003, 32, 239–250. 8. Smutz, P.; Serina, E.; Rempel, D. A system for evaluating the effect of keyboard design on force, posture, comfort, and productivity. Ergonomics 1994, 37, 1649–1660. 9. Hindle, R.J.; Pearcy, M.J.; Cross, A.T.; Miller, D.H.T. Three-dimensional kinematics of the human back. Clin. Biomech. 1990, 5, 218–228. 10. Hansson, G.A.; Asterland, P.; Holmer, N.G.; Skerfving, S. Validity and reliability of triaxial accelerometers for inclinometry in posture analysis. Med. Biol. Eng. Comput. 2001, 39, 405–413. accelerometers for inclinometry in posture analysis. Med. Biol. Eng. Comput. 2001, 39, 405–413. 11. Bernmark, E.; Wiktorin, C. A triaxial accelerometer for measuring arm movements. Appl. Ergon. 2002, 33, 541–547. 11. Bernmark, E.; Wiktorin, C. A triaxial accelerometer for measuring arm movements. Appl. Ergon. 2002, 33, 541–547. 12. Sensors 2015, 15 Sensors 2015, 15 1801 Sensors 2015, 15 1801 Conflicts of Interest The authors declare no conflict of interest. Author Contributions PP and EA devised the protocol and collected the data. PP performed the parameter identification, data analysis and drafted the manuscript. EA contributed to the method development. FM intensively reviewed the manuscript. FM and AL guided the discussion on the protocol and the interpretation of the results and gave the final approval for submission. All authors read, revised and approved the final manuscript. Sensors 2015, 15 Sensors 2015, 15 1802 17. Clark, R.A.; Bower, K.J.; Mentiplay, B.F.; Paterson, K.; Pua, Y. Concurrent validity of the Microsoft Kinect for assessment of spatiotemporal gait variables. J. Biomech. 2013, 46, 2722–2725. 17. Clark, R.A.; Bower, K.J.; Mentiplay, B.F.; Paterson, K.; Pua, Y. Concurrent validity of the Microsoft Kinect for assessment of spatiotemporal gait variables. J. Biomech. 2013, 46, 2722–2725. 18. Bonnechère, B.; Jansen, B.; Salvia, P.; Bouzahouene, H.; Omelina, L.; Moiseev, F.; Sholukha, V.; Cornelis, J.; Rooze, M.; Jan, S.V.S. Validity and Reliability of the Kinect within Functional Assessment Activities: Comparison with Standard Stereophotogrammetry. Gait Posture 2014, 39, 593–598. 19. Gonzáles, A.; Hayashibe, M.; Bonnet, V.; Fraisse, P. Whole Body Center of Mass Estimation with Portable Sensors: Using the Statically Equivalent Serial Chain and a Kinect. Sensors 2014, 14, 16955–16971. 20. Dutta, T. Evaluation of the Kinect sensor for 3-D kinematic measurement in the workplace. Appl. Ergon. 2012, 43, 645–649. 21. Rafibakhsh, N.; Gong, J.; Siddiqui, M.K.; Gordon, C.; Lee, H.F. Analysis of XBOX Kinect Sensor Data for Use on Construction Sites: Depth Accuracy and Sensor Interference Assessment. In Proceedings of the Construction Research Congress, West Lafayette, IN, USA, 21–23 May 2012; pp. 848–857. 22. Khoshelham, K.; Elberink, S.O. Accuracy and resolution of Kinect depth data for indoor mapping applications. Sensors 2012, 12, 1437–1454. 23. Obdržálek, Š.; Kurillo, G.; Ofli, F.; Bajcsy, R.; Seto, E.; Jimison, H.; Pavel, M. Accuracy and Robustness of Kinect Pose Estimation in the Context of Coaching of Elderly Population. In Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), San Diego, CA, USA, 28 August–1 September 2012; pp. 1188–1193. 24. Kurillo, G.; Chen, A.; Bajcsy, R.; Han, J.J. Evaluation of upper extremity reachable workspace using Kinect camera. Technol. Health Care 2013, 21, 641–656. 25. Fernández-Baena, A.; Susín, A.; Lligadas, X. Biomechanical Validation of Upper-Body and Lower-Body Joint Movements of Kinect Motion Capture Data for Rehabilitation Treatments. In Proceedings of the 4th International Conference on Intelligent Networking and Collaborative Systems (INCoS), Bucharest, Romania, 19–21 September 2012; pp. 656–661. 26. Diego-Mas, J.A.; Alcaide-Marzal, J. Using Kinect™ sensor in observational methods for assessing postures at work. Appl. Ergon. 2013, 45, 976–985. 27. Martin, C.C.; Burkert, D.C.; Choi, K.R.; Wieczorek, N.B.; McGregor, P.M.; Herrmann, R.A.; Beling, P.A. A Real-Time Ergonomic Monitoring System Using the Microsoft Kinect. References Freedman, B.; Shpunt, A.; Machline, M.; Arieli, Y. Depth Mapping Using Projected Patterns. U.S. Patent 2010/0118123, 13 May 2010. 13. Shpunt, A. Depth Mapping Using Multi-Beam Illumination. U.S. Patent 2010/0020078, 28 January 2010. 14. Spektor, E.; Mor, Z.; Rais, D. Integrated Processor for 3D Mapping. U.S. Patent 2010/0007717, 14 January 2010. 15. Shotton, J.; Fitzgibbon A.; Cook, M.; Sharp, T.; Finocchio, M.; Moore, R.; Kipman, A.; Blake, A. Real-time human pose recognition in parts from single depth images. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI, USA, 20–25 June 2011; pp. 1297–1304. 16. Clark, R.A.; Pua, Y.H.; Fortin, K.; Ritchie, C.; Webster, K.E.; Denehy, L.; Bryant, A.L. Validity of the Microsoft Kinect for assessment of postural control. Gait Posture 2012, 36, 372–377. Sensors 2015, 15 In Proceedings of the IEEE Systems and Information Engineering Design Symposium (SIEDS), Charlottesville, VA, USA, 27 April 2012; pp. 50–55. 28. Haggag, H.; Hossny, M.; Nahavandi, S.; Creighton, D. Real Time Ergonomic Assessment for Assembly Operations Using Kinect. In Proceedings of the 15th International Conference on Computer Modelling and Simulation (UKSim), Cambridge, UK, 10–12 April 2013; pp. 495–500. 29. Bó, A.P.L.; Hayashibe, M.; Poignet, P. Joint Angle Estimation in Rehabilitation with Inertial Sensors and its Integration with Kinect. In Proceedings of the 33th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Boston, MA, USA, 30 August–3 September 2011; pp. 3479–3483. Sensors 2015, 15 1803 30. Penelle, B.; Debeir, O. Human Motion Tracking for Rehabilitation Using Depth Images and Particle Filter Optimization. In Proceedings of the 2nd International Conference on Advances in Biomedical Engineering (ICABME), Tripoli, Lebanon, 11–13 September 2013; pp. 211–214. 31. MakeHuman. Available online: http://www.makehuman.org (accessed on 13 Jan 32. Blender. Available online: http://www.blender.org (accessed on 13 January 2015). 33. Wu, G.; van der Helm, F.C.; Veeger, H.E.; Makhsous, M.; van Roy, P.; Anglin, C.; Nagels, J.; Karduna, A.R.; McQuade, K.; Wang, X.; et al. International Society of Biomechanics. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion—Part II: Shoulder, elbow, wrist and hand. J. Biomech. 2005, 38, 981–992. 34. Abdel-Malek, K.; Yang, J.; Brand, R.; Tanbour, E. Towards understanding the workspace of human limbs. Ergonomics 2004, 47, 1386–1405. 35. Bernard, B.P. (Ed.) Musculoskeletal Disorders and Workplace Factors: A Critical Review of Epidemiologic Evidence for Work-Related Musculoskeletal Disorders of the Neck, Upper Extremity, and Low Back; US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health: Cincinnati, OH, USA, 1997. 36. Auvinet, E.; Multon, F.; Aubin, C.E.; Meunier, J.; Raison, M. Detection of gait cycles in treadmill walking using a Kinect. Gait Posture 2014, in press. 37. Monzani, J.S.; Baerlocher, P.; Boulic, R.; Thalmann, D. Using an Intermediate Skeleton and Inverse Kinematics for Motion Retargeting. Comput. Graph. Forum. 2000, 19, 11–19. 38. Bonnechère, B.; Sholukha, V.; Moiseev, F.; Rooze, M.; van Sint Jan, S. From Kinect™ to anatomically-correct motion modelling: Preliminary results for human application. In Proceedings of 3rd European Conference on Gaming and Playful Interaction in Health Care, Games for Health, Amsterdam, The Netherlands, 27–29 October 2013; pp. 15–26. 39. © 2015 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 license (http://creativecommons.org/licenses/by/4.0/). Sensors 2015, 15 Vignais, N.; Miezal, M.; Bleser, G.; Mura, K.; Gorecky, D.; Marin, F. Innovative system for real-time ergonomic feedback in industrial manufacturing. Appl. Ergon. 2012, 44, 566–574. © 2015 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 license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2970457044	https://europepmc.org/articles/pmc6712022?pdf=render	English	null	Cystic transformation of focal lesions after therapy is associated with remission but adverse outcome in myeloma	Blood cancer journal	2,019	cc-by	4,221	© The Author(s) 2019 OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproduction 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/. Cystic transformation of focal lesions after therapy is associated with remission but adverse outcome in myeloma Maximilian Merz1,2, Thomas Hielscher 3, Elias Karl Mai 1, Anja Seckinger1, Dirk Hose1, Anna Jauch4, Sandra Sauer1, Steffen Luntz5, Uta Bertsch1,6, Marc S. Raab1,7, Kai Neben1, Hans Salwender8, Igor W. Blau9, Hans-Walter Lindemann10, Jan Dürig11, Christof Scheid12, Mathias Haenel13, Katja Weisel14, Tim Weber15, Stefan Delorme16, Hartmut Goldschmidt1,6 and Jens Hillengass1,2 second ASCT was recommended. The study was per- formed in accordance with the Declaration of Helsinki, the European Clinical Trial Directive (2005) and was approved by the local ethics committees. Inclusion and exclusion criteria as well as primary end points of the study have been reported3–5. MRI was performed at pri- mary diagnosis and inclusion into the GMMG MM5 trial and repeated after the last ASCT before the start of consolidation therapy (Supplemental Fig. 1). Whole-body imaging was performed using unenhanced T1-weighted turbo-spin echo sequences as well as T2-weighted short- tau inversion recovery (STIR) sequences and analyzed as described previously6. All images were assessed by two experienced investigators blinded to outcome. In accor- dance with the IMAJEM study of the IFM and previous analyses from our group, response to treatment was deﬁned as signal decrease in T2-weighted images as well as signal recovery in T1-weighted images7,8. At inclusion into the trial, interphase ﬂuorescence in situ hybridization (iFISH) and gene expression proﬁling (GEP) were per- formed on CD138-puriﬁed plasma cells to identify cyto- genetic abnormalities as well as proliferation activity, as described previously9,10. Description of statistical analyses can be found in the supplemental material. A total of 167 patients were enrolled in the GMMG MM5 trial at the University Hospital Heidelberg between July 2010 and October 2012. Eighty-three of the 167 patients received a MRI before starting the therapy, and 77 patients after the last ASCT before consolidation treatment (median 98 days after last ASCT, interquartile range: 20 days). Thirty-four patients (44.2%) were treated with tandem ASCT and at the time of the second MRI, 41 patients Correspondence: Maximilian Merz (maximilian.merz@med.uni-heidelberg.de) 1Department of Internal Medicine V, University Hospital Heidelberg, Heidelberg, Germany 2Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA Full list of author information is available at the end of the article. These authors contributed equally: Hartmut Goldschmidt, Jens Hillengass Parts of this work were presented at the annual meetings of the European Hematology Association in Stockholm as well as the European Myeloma Network in Torino in 2018 and awarded with travel grants at both meetings. Merz et al. Blood Cancer Journal (2019) 9:71 https://doi.org/10.1038/s41408-019-0235-3 Merz et al. Blood Cancer Journal (2019) 9:71 https://doi.org/10.1038/s41408-019-0235-3 Blood Cancer Journal O p e n A c c e s s C O R R E S P O N D E N C E Dear Editor, Response assessment in multiple myeloma (MM) is based on measurements of monoclonal proteins and minimal residual disease (MRD) in bone marrow samples from the iliac crest1. Since discrete areas of plasma cell accumulation can be visualized as focal lesions (FL) by magnetic resonance imaging (MRI) as well as positron- emission computed tomography (PET/CT), the Interna- tional Myeloma Working Group (IMWG) has included MRI and PET/CT in their updated guidelines for primary diagnosis and follow-up1,2. We analyzed conventional MRI at primary diagnosis and after ASCT in newly diagnosed patients enrolled in the prospective MM5 trial (EudraCT No. 2010-019173-16) of the German-Speaking Myeloma Multicenter Group (GMMG). Treatment of newly diagnosed, symptomatic MM patients within the GMMG MM5 trial consisted of three cycles PAd (borte- zomib, doxorubicin, dexamethasone) or VCD (bortezo- mib, cyclophosphamide, dexamethasone) induction therapy; high dose melphalan followed by ASCT as well as consolidation and maintenance therapies with lenalido- mide for 2 years or until complete response (CR) (Sup- plemental Fig. 1). For patients not achieving a near complete response (nCR) or CR after the ﬁrst ASCT a 2Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA Full list of author information is available at the end of the article. These authors contributed equally: Hartmut Goldschmidt, Jens Hillengass Parts of this work were presented at the annual meetings of the European Hematology Association in Stockholm as well as the European Myeloma Network in Torino in 2018 and awarded with travel grants at both meetings. Correspondence: Maximilian Merz (maximilian.merz@med.uni-heidelberg.de) 1Department of Internal Medicine V, University Hospital Heidelberg, Heidelberg, Germany 2 © The Author(s) 2019 Disease grow- ing beyond cortical bone was detected in 21 patients (25.3%). After ASCT, residual FLs were found in 62 patients (80.5%), and residual diffuse inﬁltration in 47 patients (61.0%). Response to treatment of FL is usually characterized by signal normalization in T1- as well as T2-weighted images. However, in 28 patients (45.2% of patients with residual FLs) we observed in a subset of FL a cystic transformation after ASCT. Cystic transformation of FL was characterized by signal intensity similar to cerebrospinal ﬂuid on T2- and T1-weighted images (Fig. 1). When analyzing cystic FL after therapy, no distinct anatomical location, no morphologic feature or size at baseline could be identiﬁed that would have predicted cystic transformation. In 4 of the 28 patients with cystic FL a PET/CT had been performed at the same time point. None of the respective lesions showed increased FDG uptake (Fig. 1). No signiﬁcant associations between the presence of residual FLs and diffuse inﬁltration with rates of nCR/CR after ASCT were observed. Patients with residual FLs and a cystic transformation showed sig- niﬁcantly higher rates of nCR/CR (75%) compared to patients without the respective changes after therapy (41%, p = 0.005). Analyses of baseline characteristics revealed that patients with cystic lesions after therapy harbored more often a deletion of chromosome 13q14 (61.5 vs. 33.3%, p = 0.03), disease exceeding bone (48.0 vs. 7.1%, p < 0.001) as well as a medium/high proliferation index as assessed by GEP at baseline (92 vs. 67%, p = 0.03). No associations between MRI ﬁndings and the other tested cytogenetic abnormalities were found, espe- cially not for high-risk abnormalities (del17p, t(4; 14), gain1q21) or a hyperdiploid karyotype. Presence of bone- exceeding disease at baseline was associated with shorter OS (5 year OS 59%, 95% conﬁdence interval (CI) [40%; 87%] vs 83% [73%; 93%], p log-rank: 0.03, Fig. 2b) while no signiﬁcant differences for PFS were found (Fig. 2a). When analyzing the entire cohort, the disappearance or persis- tence of FLs or diffuse inﬁltration was not associated with signiﬁcant survival differences. However, patients not achieving a CR after ASCT had a shorter PFS (Fig. 2c) if diffuse marrow inﬁltration was present at the second MRI (Median PFS: 26 months 95%CI [24; 34] vs. 54 months [32, not reached], p log-rank: 0.03). © The Author(s) 2019 The Author(s) 2019 OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproduction 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/. Blood Cancer Journal Page 2 of 5 Merz et al. Blood Cancer Journal (2019) 9:71 Fig. 1 Cystic transformation of FL. Representative images from a patient with a large focal lesion in the sacrum with extramedullary growth. Images at baseline in the top row with common features of a focal myeloma lesion (red arrows): Decreased signal intensity on T1-weighted images (a) and increased intensity on T2-weighted images (b) as well as DWI (c). PET/CT proved increased FDG uptake at primary diagnosis. Upon achieving remission after ASCT (bottom row), the lesion decreased signiﬁcantly in size as depicted on T1- (e) and T2-weighted (f) images. The lesion showed a hyperintense signal on T2-weighted images but no increased FDG uptake compared to the surrounding tissue. MRI images in coronar orientation, PET/CT in transversal orientation Fig. 1 Cystic transformation of FL. Representative images from a patient with a large focal lesion in the sacrum with extramedullary growth. Images at baseline in the top row with common features of a focal myeloma lesion (red arrows): Decreased signal intensity on T1-weighted images (a) and increased intensity on T2-weighted images (b) as well as DWI (c). PET/CT proved increased FDG uptake at primary diagnosis. Upon achieving remission after ASCT (bottom row), the lesion decreased signiﬁcantly in size as depicted on T1- (e) and T2-weighted (f) images. The lesion showed a hyperintense signal on T2-weighted images but no increased FDG uptake compared to the surrounding tissue. MRI images in coronar orientation, PET/CT in transversal orientation (53.2%) had achieved a near CR/CR. Patient character- istics are summarized in supplemental table 1. FL were found in 76 patients (91.6%) at primary diagnosis, diffuse marrow inﬁltration in 81 patients (97.6%). Blood Cancer Journal © The Author(s) 2019 Also, patients with cystic transformation of FLs after therapy had a shorter PFS than patients without such signal alterations (Median PFS: 17 months 95%CI [14; 34] vs. 45 months [29, not reached], p log-rank: 0.014, Fig. 2e), but this did not apply for OS which was identical for both groups (Fig. 2f). The presence of cystic lesions retained their negative prog- nostic impact after adjustment for ISS, treatment arm, (53.2%) had achieved a near CR/CR. Patient character- istics are summarized in supplemental table 1. FL were found in 76 patients (91.6%) at primary diagnosis, diffuse marrow inﬁltration in 81 patients (97.6%). Disease grow- ing beyond cortical bone was detected in 21 patients (25.3%). After ASCT, residual FLs were found in 62 patients (80.5%), and residual diffuse inﬁltration in 47 patients (61.0%). Response to treatment of FL is usually characterized by signal normalization in T1- as well as T2-weighted images. However, in 28 patients (45.2% of patients with residual FLs) we observed in a subset of FL a cystic transformation after ASCT. Cystic transformation of FL was characterized by signal intensity similar to cerebrospinal ﬂuid on T2- and T1-weighted images (Fig. 1). When analyzing cystic FL after therapy, no distinct anatomical location, no morphologic feature or size at baseline could be identiﬁed that would have predicted cystic transformation. In 4 of the 28 patients with cystic FL a PET/CT had been performed at the same time point. None of the respective lesions showed increased FDG uptake (Fig. 1). No signiﬁcant associations between the presence of residual FLs and diffuse inﬁltration with rates of nCR/CR after ASCT were observed. Patients with residual FLs and a cystic transformation showed sig- niﬁcantly higher rates of nCR/CR (75%) compared to patients without the respective changes after therapy (41%, p = 0.005). Analyses of baseline characteristics Blood Cancer Journal Merz et al. © The Author(s) 2019 Blood Cancer Journal (2019) 9:71 Page 3 of 5 months since randomisation OS probability 0 12 24 36 48 60 72 0 % 25 % 50 % 75 % 100 % I IIIIIIIIIIIIIIIIIIIIIIIII I I IIIIIIIII I II IIIIIIII II III IIII II I 62 62 60 56 54 26 9 21 20 19 18 16 4 2 Log−rank: p= 0.0331 no yes months since randomisation PFS probability 0 12 24 36 48 60 72 0 % 25 % 50 % 75 % 100 % I I II I III I I I I I I III IIII I I II II I I 62 55 48 39 27 15 6 21 17 14 10 7 2 1 Log−rank: p= 0.348 no yes A Bone-exceeding disease B C D F E 2 Survival according to MRI ﬁndings before and after ASCT. Progression-free (PFS) and overall survival (OS) analysis in patients with bone- eeding disease (a and b) as well as patients not achieving CR and with residual diffuse inﬁltration after ASCT (a and b). PFS (c) and OS (d) for ents with cystic lesion after ASCT months since randomisation OS probability 0 12 24 36 48 60 72 0 % 25 % 50 % 75 % 100 % I IIIIIIIIIIIIIIIIIIIIIIIII I I IIIIIIIII I II IIIIIIII II III IIII II I 62 62 60 56 54 26 9 21 20 19 18 16 4 2 Log−rank: p= 0.0331 no yes months since randomisation PFS probability 0 12 24 36 48 60 72 0 % 25 % 50 % 75 % 100 % I I II I III I I I I I I III IIII I I II II I I 62 55 48 39 27 15 6 21 17 14 10 7 2 1 Log−rank: p= 0.348 no yes A Bone-exceeding disease B C D Bone-exceeding disease A B C D F E F E g. 2 Survival according to MRI ﬁndings before and after ASCT. Progression-free (PFS) and overall survival (OS) analysis in patients with bone- ceeding disease (a and b) as well as patients not achieving CR and with residual diffuse inﬁltration after ASCT (a and b) PFS (c) and OS (d) for E F E Fig. 2 Survival according to MRI ﬁndings before and after ASCT. © The Author(s) 2019 Progression-free (PFS) and overall survival (OS) analysis in patients with bone- exceeding disease (a and b) as well as patients not achieving CR and with residual diffuse inﬁltration after ASCT (a and b). PFS (c) and OS (d) for patients with cystic lesion after ASCT Francophone du Myélome showed for the ﬁrst time that return of MRI to normal after ASCT does not allow predicting outcome7. Normalization of MRI is rather infrequent after therapy. It occurred in 11%, 13%, and 20% in the IMAJEM trial, a previous analysis of our group and the current study, respectively7,8. In the current analysis, Francophone du Myélome showed for the ﬁrst time that return of MRI to normal after ASCT does not allow predicting outcome7. Normalization of MRI is rather infrequent after therapy. It occurred in 11%, 13%, and 20% in the IMAJEM trial, a previous analysis of our group and the current study, respectively7,8. In the current analysis, performance of single or tandem ASCT as well as response after ASCT (Hazard ratio, 95% CI: 2.47 [1.25; 4.91], p = 0.0097). Changes of MRI before and after ASCT have been associated with outcome in retrospective ana- lyses of newly diagnosed MM8. However, the recent prospective IMAJEM study by the Intergroupe Blood Cancer Journal Page 4 of 5 Merz et al. Blood Cancer Journal (2019) 9:71 In summary, return of MRI ﬁndings to normal was of prognostic signiﬁcance in patients without CR after ASCT. MRI identiﬁes a subgroup of patients with high- quality responses but adverse outcome after therapy. residual FL or diffuse inﬁltration were not associated with adverse outcome when analyzing the entire cohort, which is also in line with the IMAJEM trial7. Additionally, we demonstrated that the presence of disease exceeding cortical bone is a negative prognostic factor, even in the era of proteasome inhibitor-based induction therapy and lenalidomide maintenance7,11. Furthermore, we observed that almost half of our patients with residual FL showed a mixed pattern of signal alterations. In general, signal intensity of responding FL or diffuse inﬁltration would be expected to level up with the surrounding unaffected bone marrow. However, T2 hyperintense transformation upon successful therapy has been described especially for bone metastases from solid tumors12. Liquefaction of necrotic tumor tissue is thought to be the reason for this effect that might mimic active disease also in diffusion-weighted images (DWI). Authors’ contributions H.G. is the principal investigator of the study. U.B. and H.G. wrote the study protocol. M.M., J.H., T.H. and H.G. analyzed and interpreted the data. M.M., J.H. and H.G. wrote the manuscript. T.H. performed statistical analysis. S.L. and U.B. were responsible for data acquisition and monitoring. M.M., H.S., M.H., E.K.M., U.B., I.W.B., C.S., D.H., J.H., M.S.R., M.M., I.G.H.S.W., C.G., H.W.L., M.Z., K.W. and J.D. contributed with the inclusion of patients. D.H., A.S., and A.J. performed plasma cell sorting and FISH analysis. T.W. and S.D. were responsible for magnetic resonance imaging. Manuscript editing and writing, and ﬁnal approval of the manuscript was done by all authors. Author details 1D f 1Department of Internal Medicine V, University Hospital Heidelberg, Heidelberg, Germany. 2Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA. 3Division of Biostatistics, German Cancer Research Center (DKFZ), Heidelberg, Germany. 4Institute of Human Genetics, University Heidelberg, Heidelberg, Germany. 5Coordination Center for Clinical Trials, University Hospital Heidelberg, Heidelberg, Germany. 6National Center for Tumor Diseases (NCT), Heidelberg, Germany. 7Max-Eder Research Group Experimental therapies for hematologic malignancies, DKFZ, Heidelberg, Germany. 8Asklepios Klinik und St. Georg, Altona, Hamburg, Germany. 9Department of Internal Medicine III, Charité Campus Benjamin Franklin, Berlin, Germany. 10Department of Hematology and Oncology, Kath. Krankenhaus Hagen gem. GmbH - St.-Marien-Hospital, Hagen, Germany. 11Department of Hematology and Oncology, University Hospital of Essen, Essen, Germany. 12Department of Internal Medicine I, University of Cologne, Cologne, Germany. 13Department of Internal Medicine III, Klinikum Chemnitz gGmbH, Chemnitz, Germany. 14II. Medizinische Klinik und Poliklinik, Universitätsklinikum Hamburg-Eppendorf (UKE), Hamburg, Germany. 15Department of Radiology, University Hospital Heidelberg, Heidelberg, Germany. 16Department of Radiology, German Cancer Research Center DKFZ, Heidelberg, Germany Acknowledgements The authors thank the investigators, the study nurses and all members of the study teams at the participating GMMG trial sites, the teams of the myeloma research laboratory, the FISH laboratory and the central laboratory at the University Hospital Heidelberg, the coordination centers for clinical trials (KKS) in Heidelberg and Leipzig, the pharmacies at the trial sites and, most importantly, the participating patients and their families. © The Author(s) 2019 In order to conﬁrm the hypothesis that the observed effects were caused by rapid tumor cell decay in proliferative disease, we analyzed baseline characteristics of patients who after therapy would have cystic FL. We found that patients with such changes were less likely to harbor slowly proliferative disease as assessed by GEP from random bone marrow aspirates. Our assumption was furthermore supported by the fact that patients with cystic FL achieved higher rates of nCR/CR after ASCT and none of the lesions showed FDG uptake in the small group of patients with available PET/CT at the same time point. Regardless of deep remissions after ASCT, patients with cystic lesions showed shorter PFS, even after adjustment for treatment and response. We hypothesize that these ﬁndings also in part explain the adverse prog- nosis of patients who lose an initially deep response shortly after high dose chemotherapy13. However, no signiﬁcant effect on OS was observed, which might be a result of the limited number of patients or limited follow- up. Additionally, patients without CR after ASCT and residual diffuse MRI signal alterations had an adverse outcome. Diffuse bone marrow inﬁltration as detected by MRI at baseline has been associated with high-risk disease and adverse outcome after ASCT14,15. Importantly, patients not achieving a CR upon treatment within the GMMG MM5 trial were treated with lenalidomide maintenance for 2 years. This explains the separation of PFS curves at approximately 24 months and might sup- port longer or even continuous treatment in patients with suboptimal response and residual diffuse inﬁltration in MRI. To address limitations of the current study in future trials, ﬁndings from MRI need to be correlated with PET/ CT in a larger number of patients to conﬁrm that cystic FL are uniformly PET negative. Furthermore, imaging ﬁndings should be correlated to MRD assessment to clarify the discrepancies between PET/CT and MRD negativity by ﬂowcytometry found in the IMAJEM trial by Conﬂict of interest The GMMG MM5 Trial (EudraCT no. 2010–019173–16) is supported by grants from Janssen-Cilag, Celgene, Chugai and The Binding Site. H.G.: Celgene: Consultancy, Honoraria, Research Funding; Chugai: Research Funding; Janssen Cilag: Consultancy, Honoraria, Research Funding. J.D.: Janssen Cilag: Honoraria; Celgene: Honoraria. E.K.M.: Janssen: honoraria, travel grant; Takeda: honoraria, travel grant, advisory board; Celgene: travel grant; Onyx: travel grant; Munidpharma: travel grant. H.-W.L. Honoraria: Amgen, Novartis, Takeda und Celgene M.M. AMGEN: advisory board and travel grants. Takeda: advisory board, travel grants and research support; Celgene: travel grant Schmidt-Wolf: Janssen Cilag: Honoraria; Novartis: Honoraria. K.W.: Janssen Cilag: Consultancy, Honoraria; Celgene: Consultancy, Honoraria, Research Funding. C.S.: Janssen Cilag: Honoraria; Celgene: Honoraria; Novartis: Honoraria. H.S.: Janssen Cilag: Honoraria; Celgene: Honoraria. All other authors declare that they have no conﬂict of interest. Received: 7 June 2019 Revised: 1 July 2019 Accepted: 5 August 2019 Received: 7 June 2019 Revised: 1 July 2019 Accepted: 5 August 2019 Patients With Multiple Myeloma Included in the IFM/DFCI 2009 Trial: Results of the IMAJEM Study. J. Clin. Oncol. 35, 2911–2918 (2017). 8. Hillengass, J. et al. Changes in magnetic resonance imaging before and after autologous stem cell transplantation correlate with response and survival in multiple myeloma. Haematologica 97, 1757–1760 (2012). References 9. Hose, D. et al. Proliferation is a central independent prognostic factor and target for personalized and risk-adapted treatment in multiple myeloma. Haematologica 96, 87–95 (2011). 1. Kumar, S. et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 17, e328–e346 (2016). 1. Kumar, S. et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 17, e328–e346 (2016). 10. Neben, K. et al. Administration of bortezomib before and after autologous stem cell transplantation improves outcome in multiple myeloma patients with deletion 17p. Blood 119, 940–948 (2012). 2. Rajkumar, S. V. et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 15, e538–e548 (2014). 2. Rajkumar, S. V. et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 15, e538–e548 (2014). 3. Mai, E. K. et al. Phase III trial of bortezomib, cyclophosphamide and dex- amethasone (VCD) versus bortezomib, doxorubicin and dexamethasone (PAd) in newly diagnosed myeloma. Leukemia 29, 1721–1729 (2015). 3. Mai, E. K. et al. Phase III trial of bortezomib, cyclophosphamide and dex- amethasone (VCD) versus bortezomib, doxorubicin and dexamethasone (PAd) in newly diagnosed myeloma. Leukemia 29, 1721–1729 (2015). 11. Usmani, S. Z. et al. Extramedullary disease portends poor prognosis in multiple myeloma and is over-represented in high-risk disease even in the era of novel agents. Haematologica 97, 1761–1767 (2012). 4. Goldschmidt, H. et al. Response-Adapted Lenalidomide Maintenance in Newly Diagnosed, Transplant-Eligible Multiple Myeloma: Results from the Multicenter Phase III GMMG-MM5 Trial. Blood 130, 400–400 (2017). 4. Goldschmidt, H. et al. Response-Adapted Lenalidomide Maintenance in Newly Diagnosed, Transplant-Eligible Multiple Myeloma: Results from the Multicenter Phase III GMMG-MM5 Trial. Blood 130, 400–400 (2017). 12. Padhani, A. R. et al. Therapy monitoring of skeletal metastases with whole- body diffusion MRI. J Magn Reson Imaging 39, 1049–1078 (2014). 13. Barlogie, B. et al. Complete remission sustained 3 years from treatment initiation is a powerful surrogate for extended survival in multiple myeloma. Cancer 113, 355–359 (2008). 5. Merz, M. et al. Subcutaneous versus intravenous bortezomib in two different induction therapies for newly diagnosed multiple myeloma: an interim ana- lysis from the prospective GMMG-MM5 trial. Haematologica 100, 964–969 (2015). 5. Merz, M. et al. Publisher’s note S i N Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. Supplementary Information accompanies this paper at (https://doi.org/ 10.1038/s41408-019-0235-3). Blood Cancer Journal Page 5 of 5 Page 5 of 5 Page 5 of 5 Merz et al. Blood Cancer Journal (2019) 9:71 Merz et al. Blood Cancer Journal (2019) 9:71 Received: 7 June 2019 Revised: 1 July 2019 Accepted: 5 August 2019 6. Hillengass, J. et al. Prognostic signiﬁcance of focal lesions in whole-body magnetic resonance imaging in patients with asymptomatic multiple mye- loma. J. Clin. Oncol. 28, 1606–1610 (2010). References Subcutaneous versus intravenous bortezomib in two different induction therapies for newly diagnosed multiple myeloma: an interim ana- lysis from the prospective GMMG-MM5 trial. Haematologica 100, 964–969 (2015). 14. Moulopoulos, L. A. et al. Diffuse pattern of bone marrow involvement on magnetic resonance imaging is associated with high risk cytogenetics and poor outcome in newly diagnosed, symptomatic patients with multiple myeloma: a single center experience on 228 patients. Am. J. Hematol. 87, 861–864 (2012). 6. Hillengass, J. et al. Prognostic signiﬁcance of focal lesions in whole-body magnetic resonance imaging in patients with asymptomatic multiple mye- loma. J. Clin. Oncol. 28, 1606–1610 (2010). 6. Hillengass, J. et al. Prognostic signiﬁcance of focal lesions in whole-body magnetic resonance imaging in patients with asymptomatic multiple mye- loma. J. Clin. Oncol. 28, 1606–1610 (2010). 15. Mai, E. K. et al. A magnetic resonance imaging-based prognostic scoring system to predict outcome in transplant-eligible patients with multiple myeloma. Haematologica 100, 818–825 (2015). 7. Moreau, P. et al. Prospective Evaluation of Magnetic Resonance Imaging and [18F]Fluorodeoxyglucose Positron Emission Tomography-Computed Tomo- graphy at Diagnosis and Before Maintenance Therapy in Symptomatic Blood Cancer Journal
https://openalex.org/W3207555796	https://bovine-ojs-tamu.tdl.org/bovine/article/download/1240/1231	English	null	Medical management of dairy heifers from birth to breeding	The Bovine practitioner	1,986	cc-by-sa	4,444	Introduction Cows are watched closely for “making up.” Just before parturition the cow is put into a thoroughly cleaned and disinfected box stall, freshly bedded with straw. Not only will this help prevent disease in the newborn calf, but will also reduce mastitis and nonspecific genital-tract infections. In our herd of 400 cows, an individual assigned to the maternity barn monitors cows routinely and, hence, is available to assist cows during parturition, resulting in fewer calves dying during birth. Cows are allowed to labor for approximately one hour. When calving assistance is needed, the cow’s tail is secured with a cord around her neck. The perineal area of the cow and the assistant’s hands are washed before entering the vagina. Immediately after delivery into a bed of clean straw, the calFs navel is disinfected with 7% tincture of iodine and a navel clip applied. The cow is provided with as much warm water as she will drink. Within the first two hours of life the newborn calf is fed two quarts of high-quality colostrum from a nipple bottle. Another half gallon is provided within the next eight hours. The calf remains with the cow in the box stall for approximately 24 hours, at which time the calf is taken to one of the research facilities. Within the first 24 hours of life calves are injected with Z2 cc of MuSe®. The navel is redipped at 4-5 days of age and the umbilical clip removed. Medical Management of Dairy Heifers from Birth to Breeding Erich Studer, D.V.M. Director of Veterinary Medicine Carnation Research Farm Washington 98014-0500 Paper presented at the Inter Mountain Veterinary Medical Association meeting, Las Vegas, February, 1986. Introduction fetus. In an attempt to control milk fever, cows are limited to 75 gms of calcium and 25-40 gms of phosphorus daily. fetus. In an attempt to control milk fever, cows are limited to 75 gms of calcium and 25-40 gms of phosphorus daily. Two to three weeks prior to parturition, dry cows are brought to close-up quarters and the ration changed to oat hay, corn silage, concentrates, and the dry cow supplement. The amount of grain is gradually increased to about 1% of the body weight or 15 lbs. The udders are retreated with a lactating infusion product, 7-ml of Mu-Se® is injected parenterally, and a booster E. coli vaccination given. Cows are watched closely for “making up.” Just before parturition the cow is put into a thoroughly cleaned and disinfected box stall, freshly bedded with straw. Not only will this help prevent disease in the newborn calf, but will also reduce mastitis and nonspecific genital-tract infections. The future of a dairy operation depends on the successful raising or procurement of heifer calves that will potentially equal and preferably exceed the milk production of the existing herd. In a modern dairy where rigid culling is practiced, 20-30% of the milking herd is usually replaced annually. Since mortality rates for young calves often exceed 20% under some management conditions, it is imperative that as many calves as possible be raised to maturity. This requires various disease control measures, including proper dry cow management, suitable calving facilities, colostrum feeding and navel disinfection in newborn calves, appropriate housing and adequate nutrition in growing calves. The following management procedures, adopted over many years of experience and research at Carnation Research Farm have been effective in reducing calfhood mortality to less than 3%. fetus. In an attempt to control milk fever, cows are limited to 75 gms of calcium and 25-40 gms of phosphorus daily. Two to three weeks prior to parturition, dry cows are brought to close-up quarters and the ration changed to oat hay, corn silage, concentrates, and the dry cow supplement. The amount of grain is gradually increased to about 1% of the body weight or 15 lbs. The udders are retreated with a lactating infusion product, 7-ml of Mu-Se® is injected parenterally, and a booster E. coli vaccination given. General Procedures The mean weight gain in calves fed different qualities of calf starters along with whole milk was compared (Table III) FIGURE 1. Mean Body Weight Gain in Two MR With Soy isolate 22/10 Compared to Whole Milk. FIGURE 1. Mean Body Weight Gain in Two MR With Soy isolate 22/10 Compared to Whole Milk. 2. Calf Manna® (25% protein, 3% fat and less than 6% fiber) is made available to calves at 2-3 days of age. When calves consume 1 lb of Calf Manna, they are switched to an 18%-protein starter with 25% Calf Manna. At 10 weeks of age the calf starter is replaced by a 16%-protein concentrate. Whole Milk MR 1 22/10 MR 2 22/10 Soylsolate Soylsolate 3. A high-quality alfalfa hay is made available ad libitum at two weeks of age. 4. Water is made available at all times. 4. Water is made available at all times. 4. Water is made available at all times. 5. Weaning is scheduled for five weeks if calves are consuming at least I Vi lbs of calf starter. Calves remain in individual pens for 1-2 weeks after weaning, before moving to community pens. Comparisons of milk and replacers revealed that calves fed whole milk outperformed calves fed milk replacers of any kind (Table I, Fig I). Milk replacers with 22% milk protein and 20% animal fat were superior to replacers with soy isolate containing 22% protein and 10% fat. However, feeding milk replacer of acceptable quality, such as those with 20% protein, with some soy isolate, 10% animal fat and less than 0.5% fiber provides adequate growth of calves when they are raised in environmentally-controlled facilities with optimum disease prevention. FIGURE 2. Change in PCV in Four High Quality Milk Replacers Compared To Whole Milk FIGURE 2. Change in PCV in Four High Quality Milk Replacers Compared To Whole Milk TABLE 1. Mean Weight in Replacer. Calves Fed Different Qualities of Milk TABLE 1. Mean Weight in Replacer. General Procedures Calves Fed Different Qualities of Milk Mean Weight Gain (Lbs.) Milk Based MR with Soy Age Whole Milk MR 22/20 Isolate 22/10 (Weeks) N = 195 N = 75 N = 75 1 4.08 2.60 2.56 2 8.69 5.75 4.10 3 15.71 12.37 9.71 4 25.55 20.73 18.06 5* 35.96 31.64 27.26 6 48.37 44.69 40.21 7 62.55 59.28 54.15 8 77.40 74.61 68.56 9 91.13 88.40 82.80 10 106.46 102.78 96.37 * Time of weaning high quality milk replacers generally outperformed calves fed a poorer quality milk replacer, the mortality was at least as low in calves fed milk replacers with some soy isolate in trials over a three-year period (Table II). Performance of calves fed whole milk and milk replacers were compared using hematocrit, serum protein, serum cholesterol, serum triglyceride and blood glucose (Fig 2-6). These clinical measurements are good means for evaluating the performance of calves fed milk replacers. The values correlate directly with the quality of the milk replacer. Calf starters also can affect the growth of calves. The mean weight gain in calves fed different qualities of calf starters along with whole milk was compared (Table III). Calves fed high-quality calf starter of 25% protein, beginning at two days old, outperformed calves fed calf starters containing 16% and 18% protein. The high-quality calf starter, Calf Manna®, reversed the drop in hematocrit that occurred in milk-fed calves (Fig VIII). Although calves fed whole milk generally outperform calves fed milk replacers, they have a drop in hematocrit compared to milk replacer fed calves (Fig II). The incidence of calf scours associated with different liquid feeding regimens is illustrated in Fig 7. Milk replacers with some soy isolate were associated with greater frequency of scours earlier in life, as compared to higher-quality milk replacers or whole milk. Although calves fed whole milk or Nutrition of the Older Calf At Carnation Research Farm, calves are weaned at 5 General Procedures Raising calves to healthy, mature heifers begins with proper management of the pregnant dry cow. At drying off, after cows are milked for the last time, the udder is infused with a long-acting cloxacillin, mammary-infusion product. Teats are dipped with an organic iodine preparation and E. coli bacterin (K99) is administered. A systemic insecticide like Warbex® is also applied to the dorsum to control external parasites and warbels. As an alternative, Ivermectin can be given to treat both external and internal parasites. Dry cows are maintained in a large lot where they can exercise freely and be fed feed-lot style. Alternatively, cows could be turned out to pasture; however, letting them forage, dietary control is lost, aggravating postpartum diseases like milk fever and retained placentas. In the lot cows are provided a ration of oat hay, grass silage, beet pulp and a dry cow supplement consisting of three pounds of concentrates with vitamins and minerals added. The vitamins and minerals are thus force fed to meet or exceed NRC recommendations. This ration provides only sufficient nutrients for maintenance of the cow and development of the All calves are assigned to nutritional research studies, evaluating milk replacers and calf starters. All milk-replacer and calf-starter trials are compared to the following baseline protocol: 1. Two quarts of whole milk are fed twice daily. Alternatively, all milk-based milk replacers with 20% protein, 20% fat and less than 0.2% fiber can be fed. Milk replacer is fed at a rate of Z2 lb dry powder twice a day, with an appropriate amount of water, usually equal to two quarts per feeding. Paper presented at the Inter Mountain Veterinary Medical Association meeting, Las Vegas, February, 1986. THE BOVINE PRACTITIONER — NO. 21 14 FIGURE 1. Mean Body Weight Gain in Two MR With Soy isolate 22/10 Compared to Whole Milk. Whole Milk MR 1 22/10 MR 2 22/10 Soylsolate Soylsolate FIGURE 2. Change in PCV in Four High Quality Milk Replacers Compared To Whole Milk Whole Milk MR 1 22 15/ MR 2 22 20/ MR 3 20 20/ MR 4 21 20/ high quality milk replacers generally outperformed calves fed a poorer quality milk replacer, the mortality was at least as low in calves fed milk replacers with some soy isolate in trials over a three-year period (Table II). Calf starters also can affect the growth of calves. Nutrition of the Older Calf Compared To Whole Milk Whole Milk MR 1 22/10 Soylsolate MR 2 22/10 Soylsolate MR 3 22/10 Soylsolate FIGURE 4. Mean Cholesterol in Three MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 MR 3 22/10 Soylsolate Soylsolate Soylsolate weeks old. This is not a problem if calves have been free of diseases and are consuming l'A lbs of an 18% protein calf starter. However, some calves, not at this level, are kept on liquid feed in order to get optimum growth. As a calf’s demand for nutrients increases, grain consumption increases as the amount of liquid feed is kept constant. The transition from liquid feed to dry feed is thus more easily accomplished. Hay fed to calves should be of excellent quality and alfalfa with crude protein of >18%, a TDN value of >50% and a crude fiber value of <25% is preferred. About 6 months of the amount should be reduced, especially when the roughage is of high quality. Free choice minerals providing calcium, phosphorous, magnesium, trace minerals and salt, should be made available. FIGURE 3. Mean Total Protein Change in 3 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 Soylsolate MR 2 22/10 Soylsolate MR 3 22/10 Soylsolate FIGURE 3. Mean Total Protein Change in 3 MR with Soy Isolate. Compared To Whole Milk FIGURE 5. Mean Triglyceride Change in 3 MR with Soy Isolate. Compared To Whole Milk FIGURE 5. Mean Triglyceride Change in 3 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 MR 3 22/10 Soylsolate Soylsolate Soylsolate Weeks On Test FIGURE 4. Mean Cholesterol in Three MR with Soy Isolate. Compared To Whole Milk FIGURE 4. Mean Cholesterol in Three MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 MR 3 22/10 Soylsolate Soylsolate Soylsolate FIGURE 6. Mean Glucose Change in 2 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 Soyisolate Soyisolate —351---------------------------------------------------------------- 2 4 5 6 8 10 Weeks On Test FIGURE 6. Mean Glucose Change in 2 MR with Soy Isolate. Compared To Whole Milk weeks old. This is not a problem if calves have been free of diseases and are consuming l'A lbs of an 18% protein calf starter. Nutrition of the Older Calf At Carnation Research Farm, calves are weaned at 5 NOVEMBER, 1986 15 FIGURE 3. Mean Total Protein Change in 3 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 Soylsolate MR 2 22/10 Soylsolate MR 3 22/10 Soylsolate FIGURE 4. Mean Cholesterol in Three MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 MR 3 22/10 Soylsolate Soylsolate Soylsolate the amount should be reduced, especially when the roughage is of high quality. Free choice minerals providing calcium, phosphorous, magnesium, trace minerals and salt, should be made available. FIGURE 5. Mean Triglyceride Change in 3 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 MR 3 22/10 Soylsolate Soylsolate Soylsolate Weeks On Test weeks old. This is not a problem if calves have been free of diseases and are consuming l'A lbs of an 18% protein calf starter. However, some calves, not at this level, are kept on liquid feed in order to get optimum growth. As a calf’s demand for nutrients increases, grain consumption increases as the amount of liquid feed is kept constant. The transition from liquid feed to dry feed is thus more easily accomplished. Hay fed to calves should be of excellent quality and alfalfa with crude protein of >18%, a TDN value of >50% and a crude fiber value of <25% is preferred. About 6 months of age, calves can be switched to a 14-16% protein dairy concentrate, with a good quality roughage, such as alfalfa, and possibly, corn silage. Up to 6 lbs of grain should be fed to ensure a gain of about 1 /i lbs per day. As heifers get older. FIGURE 6. Mean Glucose Change in 2 MR with Soy Isolate. Compared To Whole Milk Whole Milk MR 1 22/10 MR 2 22/10 Soyisolate Soyisolate —351---------------------------------------------------------------- 2 4 5 6 8 10 Weeks On Test TABLE 2. Calf Mortality in Calves Fed Whole Milk, High Quality and Acceptable MR over a 3-Year- Period. No. Removed Product No. Calves Mortality from Trial Acceptable M.R. 337 1 (0.3%) 22 ( 6.5%) High Q. M.R. 238 2 (0.8%) 11 ( 4.6%) Whole Milk 286 3(1.0%) 29 (10.1%) FIGURE 3. Mean Total Protein Change in 3 MR with Soy Isolate. Disease Control The two most important diseases encountered in raising calves are scours and pneumonia. Some of the important preventive measures utilized at Carnation Research Farm, including immunization of dry cows with E. coli (K99) vaccine, cleaning and disinfecting box stalls for newborn calves, colostrum feeding, navel disinfection and using high- quality milk replacers, have already been mentioned. g y y Control MR 122/10 MR 2 22/10 MR 3 22/10 MR 4 22/10 W-Milk Soyisolate Soyisolate All Milk All Milk Weeks On Test In addition, the type of housing is important. Because of the location of facilities on a hillside, heavy precipitation in Reseach Farm are housed in barns with individual tie stalls. Tying calves separately with complete separation between stalls reduces the spread of disease-causing agents. An outbreak of scours can often be stopped by placing calves at some distance from calves that are experiencing disease. Feeding utensils can be an important fomite for transmitting infectious agents. Hence, each calf has its own feed pan and bucket. Liquid feed is provided only by nipple bottles. Nipples are soaked in a disinfectant between feedings, and the bottles scrubbed and disinfected. TABLE 3. Mean Weight Gain in Calves Fed Different Qualities of Calf Starters Along With Whole Milk. At Carnation Research Farm, calf scours is treated according to the following protocol: The feeding regime is maintained as usual and liquid feed is provided morning and night regardless of the severity of scours. Oral electrolytes with high dextrose concentrations, lycine as an amino acid, and acetate as the buffer are provided once midway between the milk or milk-replacer feedings. Ten mis of oral spectinomycin are added to the electrolyte solution. The rectal temperature of young calves is taken daily. If rectal temperatures exceed 103°F, calves are given parenteral spectinomycin at 5 mg per lb. If calves become dehydrated, oral electrolytes are provided again in the middle of the night. With this regime, we find that calves continue to gain weight in spite of scouring for several days. Mean Weight Gain (Lbs.) Age 16% Prot. C.S. 18% Prot. C.S. C.M. 25% Prot. Nutrition of the Older Calf However, some calves, not at this level, are kept on liquid feed in order to get optimum growth. As a calf’s demand for nutrients increases, grain consumption increases as the amount of liquid feed is kept constant. The transition from liquid feed to dry feed is thus more easily accomplished. weeks old. This is not a problem if calves have been free of diseases and are consuming l'A lbs of an 18% protein calf starter. However, some calves, not at this level, are kept on liquid feed in order to get optimum growth. As a calf’s demand for nutrients increases, grain consumption increases as the amount of liquid feed is kept constant. The transition from liquid feed to dry feed is thus more easily accomplished. TABLE 2. Calf Mortality in Calves Fed Whole Milk, High Quality and Acceptable MR over a 3-Year- Period. TABLE 2. Calf Mortality in Calves Fed Whole Milk, High Quality and Acceptable MR over a 3-Year- Period. Hay fed to calves should be of excellent quality and alfalfa with crude protein of >18%, a TDN value of >50% and a crude fiber value of <25% is preferred. About 6 months of age, calves can be switched to a 14-16% protein dairy concentrate, with a good quality roughage, such as alfalfa, and possibly, corn silage. Up to 6 lbs of grain should be fed to ensure a gain of about 1 /i lbs per day. As heifers get older. No. Removed Product No. Calves Mortality from Trial Acceptable M.R. 337 1 (0.3%) 22 ( 6.5%) High Q. M.R. 238 2 (0.8%) 11 ( 4.6%) Whole Milk 286 3(1.0%) 29 (10.1%) THE BOVINE PRACTITIONER — NO. 21 16 FIGURE 7. Mean No. of Days Treated for Scours Comparing. High Quality MR, MR with Soy Isolate and Whole Milk FIGURE 7. Mean No. of Days Treated for Scours Comparing. High Quality MR, MR with Soy Isolate and Whole Milk Control MR 122/10 MR 2 22/10 MR 3 22/10 MR 4 22/10 W-Milk Soyisolate Soyisolate All Milk All Milk Weeks On Test TABLE 3. Mean Weight Gain in Calves Fed Different Qualities of Calf Starters Along With Whole Milk. Immunization Schedule Lepto 5-way to unfreshened heifers > six months old. IBR, BVD, PI3, Lepto-5-way and Hemophilus somnus vaccine to cows at 30-days postpartum. Lepto-5-way to cows at the time of pregnancy rechecking three months after conception. E. coli bacterin at drying off, E. coli bacterin and Hemophilus somnus vaccine at retreatment of udder, three weeks prepartum. 6. Evaluation of commercial and autogenous Pasteurella multocida and Pasteurella haemolytica bacterins revealed neither was effective in preventing calf pneumonia. Vaccination with either bacterin resulted in increased occurrence of fever after vaccination, and an increase in pneumonia after booster innoculation with the autogenous bacterin. Disease Control (Weeks) N = 195 N = 30 N = 75 1 4.08 4.10 4.88 2 8.69 9.85 9.13 3 15.71 15.55 16.44 4 25.55 24.50 27.63 5* 35.96 33.15 38.98 6 48.37 46.85 52.78 7 62.55 61.50 71.14 8 77.40 80.05 88.30 9 91.13 92.90 99.22 10 106.146 106.00 112.37 * Weaning was at 5 weeks of age. Environmental factors are very important in relation to calf pneumonia; however, the subject of housing and environment is beyond the scope of this paper. For much of the country, calf hutches without artificially-controlled environments are the best and most inexpensive way to house calves. When calves are housed in barns, environmen tal control becomes more critical, especially for the prevention of calf pneumonia. Environmental factors to be considered are: FIGURE 8. Mean Packed Cell Volume Change Comparing Calves Fed Calf Manna and a 16% Protein Starter. FIGURE 8. Mean Packed Cell Volume Change Comparing Calves Fed Calf Manna and a 16% Protein Starter. FIGURE 8. Mean Packed Cell Volume Change Comparing Calves Fed Calf Manna and a 16% Protein Starter. 6 5 4 3 2 1 0 —1 —2 —3 —4 — 5 — 6 0 2 4 6 8 10 12 14 16 26 PCV (%) Change Calf Manna 16% Protein 25% Protein Starter mm 1. Wall and ceiling insulation 2. Ambient temperature 3. Relative humidity 4 V il i 3. Relative humidity 4. Ventilation 5. Population density 6. Pen construction and bedding 7. Sanitation and disinfection Calf pneumonia is a serious disease problem at Carnation Research Farm. Several years of research in calf pneumonia have revealed the following: 1. The incidence of calf pneumonia is loosely related to 1. The incidence of calf pneumonia is loosely related to NOVEMBER, 1986 17 ambient weather and conditions with more cases occurring in the fall, winter and sometime spring. However, there has been no consistent pattern from year to year. ambient weather and conditions with more cases occurring in the fall, winter and sometime spring. However, there has been no consistent pattern from year to year. Immunization Schedule An important aspect of disease prevention in dairy calves is a sound immunization program. At Carnation Research Farm the immunization schedule is as follows: 2. The major viral infections of cattle, including IBR, BVD and P13 do not have a major role in the etiology of calf pneumonia. 2. The major viral infections of cattle, including IBR, BVD and P13 do not have a major role in the etiology of calf pneumonia. Two Weeks Old Hemophilus somnus vaccine, Pasteurella multo cida / Pasteurella haemolytica vaccine and Bovine Respiratory Syncitial Virus Vaccine (experimental investigation). Two Weeks Old Hemophilus somnus vaccine, Pasteurella multo cida / Pasteurella haemolytica vaccine and Bovine Respiratory Syncitial Virus Vaccine (experimental investigation). 3. Chlamydial agents do not play a major role in the etiology of calf pneumonia. 3. Chlamydial agents do not play a major role in the etiology of calf pneumonia. Two Months Old Hemophilus somnus and wart vaccine Four to Five Months Old Two Months Old Hemophilus somnus and wart vaccine Four to Five Months Old 4. Transtracheal lavages from pneumonia have demon strated a 75% recovery rate of mycoplasma, but only a 5% recovery rate of ureaplasma. This suggests that myco-plasma species may be associated with disease. 4. Transtracheal lavages from pneumonia have demon strated a 75% recovery rate of mycoplasma, but only a 5% recovery rate of ureaplasma. This suggests that myco-plasma species may be associated with disease. Brucella vaccine and wart vaccine Six-to- Seven Months Old Males and Females—Clostridial disease 7-way. Females only—IBR, BVD, PI3, Lepto-5-way & Hemophilus somnus vaccine. Annually Lepto 5-way to unfreshened heifers > six months old. IBR, BVD, PI3, Lepto-5-way and Hemophilus somnus vaccine to cows at 30-days postpartum. Lepto-5-way to cows at the time of pregnancy rechecking three months after conception. E. coli bacterin at drying off, E. coli bacterin and Hemophilus somnus vaccine at retreatment of udder, three weeks prepartum. Six-to- Seven Months Old Males and Females—Clostridial disease 7-way. Females only—IBR, BVD, PI3, Lepto-5-way & Hemophilus somnus vaccine. Annually 5. Of bacterial agents recovered from transtracheal lavages, only Pasteurella multocida and Pasteurella haemolytica were frequently isolated. Pasteurella multocida was isolated from the trachea of about 2/3 of the cases, while Pasteurella haemolytica was isolated from the trachea of about 1 / 5 of the cases. In calves that died or were euthanized, Pasteurella haemolytica was isolated from the lungs of 3/4 of the cases. Other Management Procedures A. At birth, calves are given x/i ml of Mu-Se® subcutaneously. A. At birth, calves are given x/i ml of Mu-Se® subcutaneously. 7. We have evidence that the Bovine-Respiratory-Syncitial Virus is involved in the etiology of calf pneumonia. We have positive isolations and seroconversions. 7. We have evidence that the Bovine-Respiratory-Syncitial Virus is involved in the etiology of calf pneumonia. We have positive isolations and seroconversions. B. At two months of age, 1 ml of Mu-Se® is administered. B. At two months of age, 1 ml of Mu-Se® is administered. B. At two months of age, 1 ml of Mu-Se® is administered. C. At six weeks of age, calves are dehorned by electro cautery and supernumerary teats removed surgically. C. At six weeks of age, calves are dehorned by electro cautery and supernumerary teats removed surgically. D. Internal and external parasites are controlled by periodic treatment. Panacur® is used twice in the spring on pastured heifers and again in the fall. A systemic organophosphate like Warbex® is applied to the dorsum of the animals in the spring and fall. The protocol for treatment of calf pneumonia is as follows: Initial treatment consists of penicillin, 30,000 units per lb of body weight subcutaneously once daily. If, after two days the response to penicillin is poor, spectinomycin, 5 mg per lb of body weight subcutaneously, twice daily, is substituted. If spectinomycin is not effective within two days, Ditrim® or Trimethiprim®, 1 ml per 20 lbs of body weight subcutaneously, once daily of the 28% solution, or 1 ml per 40 lbs of bodyweight, of the 48% solution is given. Parenteral antibiotics are given for two days after the body temperature is below 103°F and then a similar antibiotic is given orally for another three days. E. Coccidiosis is prevented by using Deccox® either top dressed or mixed in the grain and fed to weaned calves. F. Ringworm is treated topically with fungicidal ointments when only one or two calves are affected. For group treatment, Orthocide Garden Fungicide®, 50%- Chevron Chemical Co., is applied with a pressure sprayer at 400 to 450 lbs pressure. The appropriate mixture for cattle is .45-.50 lbs per 20 gallons of water. This concentration is doubled for facilities, fences, mangers, etc. Transtracheal lavages are performed periodically to identify causative agents and perform sensitivity testing. THE BOVINE PRACTITIONER — NO. 21 18
https://openalex.org/W2602207157	http://www.scielo.br/pdf/rceres/v64n1/2177-3491-rceres-64-01-00098.pdf	Portuguese	null	Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha	Revista Ceres	2,017	cc-by	6,719	RESUMO A análise do acúmulo e da exportação de nutrientes pela cultura de feijão-vagem é fundamental para programa de adubação mais sustentável, pois a definição das doses de adubos orgânicos com base apenas na produção máxima estimada não garante a manutenção da fertilidade do solo. Objetivou-se, com este trabralho, avaliar o efeito da adubação com esterco de galinha sobre a produtividade, o acúmulo e a exportação de nutrientes pelas vagens de feijão-vagem. Utilizou-se o delineamento em blocos casualizados, com cinco doses de esterco de galinha (0; 5; 10; 20 e 40 t ha-1) e quatro repetições. Foram constatados maiores teores de P e Mg nas folhas com aplicação da dose de 40 t ha-1 de esterco. O comprimento máximo da vagem foi 14,47 cm, estimado com a dose de 33,33 t ha-1 de esterco. Os maiores valores de diâmetro, número de vagens por planta e produtividade de vagens foram observados na maior dose de esterco aplicada. Em termos relativos, ou seja, total exportado em relação ao total extraído pela parte aérea, o fósforo é o nutriente mais exportado pelas vagens, em média de 58%, seguidos pelo N (55%), K (43%), Mg (40%), S (38%) e pelo Ca (17%). Na maior dose, embora o acúmulo de Ca tenha ocorrido em grande quantidade (31,3 kg ha-1), apenas 13% dele foram exportados pelas vagens. A adubação do feijão-vagem com esterco de galinha supre os nutrientes essenciais e aumenta a produti- vidade de vagens, de 7,2 (sem adubação), para 16,3 t ha-1 (adubação com 40 t ha-1 de esterco de galinha). Os restos vegetais do feijão-vagem constituem importante fonte de nutrientes, sendo obtidas na maior dose de esterco aplicada (40 t ha-1) as seguintes quantidades de macronutrientes (kg ha-1): N (51,4); P (5,1); K (27,6); Ca (27,1); Mg (8,2); S (5,1), que poderão retornar ao solo, com a incorporação das plantas. Palavras-chave: Phaseolus vulgaris; conteúdo de nutrientes; cultivo orgânico; nutrição de plantas. Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha Ivan de Paiva Barbosa Magalhães1, Maria Aparecida Nogueira Sediyama2, Fred Denilson Barbosa da Silva3, Sanzio Mollica Vidigal4, Cláudia Lúcia Oliveira Pinto4, Iza Paula Carvalho Lopes3 Ivan de Paiva Barbosa Magalhães1, Maria Aparecida Nogueira Sediyama2, Fred Denilson Barbosa Sanzio Mollica Vidigal4, Cláudia Lúcia Oliveira Pinto4, Iza Paula Carvalho Lopes3 10.1590/0034-737X201764010014 Submetido em 11/06/2015 e aprovado em 17/10/2016. 1 Universidade Federal de Viçosa, Viçosa, Minas Gerais, Brasil. Bolsista PIBIC FAPEMIG/EPAMIG. ivan.barbosa@ufv.br 2Pesquisadora/Bolsista CNPq/EPAMIG. Viçosa, Minas Gerais, Brasil. marians@epamig.br 3Bolsista FAPEMIG/EPAMIG. Viçosa, Minas Gerais, Brasil. izzaagro@yahoo.com.br; freddenilson@gmail.com 4Pesquisador/Bolsista BIPDT FAPEMIG/EPAMIG. Viçosa, Minas Gerais, Brasil. sanziomv@gmail.com clucia@epamig.ufv.br Autor para correspondência: ivan.barbosa@ufv.br Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Submetido em 11/06/2015 e aprovado em 17/10/2016. 1 Universidade Federal de Viçosa, Viçosa, Minas Gerais, Brasil. Bolsista PIBIC FAPEMIG/EPAMIG. ivan.barbosa@ufv.br 2Pesquisadora/Bolsista CNPq/EPAMIG. Viçosa, Minas Gerais, Brasil. marians@epamig.br 3Bolsista FAPEMIG/EPAMIG. Viçosa, Minas Gerais, Brasil. izzaagro@yahoo.com.br; freddenilson@gmail.com 4Pesquisador/Bolsista BIPDT FAPEMIG/EPAMIG. Viçosa, Minas Gerais, Brasil. sanziomv@gmail.com clucia@epamig.ufv.br Autor para correspondência: ivan.barbosa@ufv.br Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 INTRODUÇÃO mente durante todo o ciclo da cultura, mas a época de maior exigência, quando a velocidade de absorção é máxi- ma, ocorre dos 35 aos 50 dias da emergência da planta, coincidindo com a época do florescimento. Neste período, o feijoeiro absorve de 2,0 a 2,5 kg N/ha dia (Rosolem & Marubayashi, 1994). O feijão-vagem (Phaseolus vulgaris L.) está entre as hortaliças mais consumidas no Brasil, com produção de 56.776 t de vagens (IBGE, 2006). Nesse último censo agropecuário, Minas Gerais apresentou-se com 27,8% da produção nacional, com 15.501 t de vagens. Parte da pro- dução do Estado é comercializada nas Unidades Centrais de Revenda e Abastecimento, com a movimentação média mensal de 712 t de vagens (CEASAMINAS, 2013). Outro fator importante que afeta a produtividade de vagens é que as quantidades de N absorvidas pelo feijoeiro, oriundas da mineralização da matéria orgânica do solo e da simbiose com as bactérias do gênero Rhizobium, não são suficientes para se alcançarem altas produtividades (Brito et al., 2011; Souza et al., 2011; Perez et al., 2013). A aplica- ção de adubo orgânico a partir da produtividade esperada é uma estratégia utilizada no cultivo do feijão-vagem para suprir adequadamente a demanda de N da cultura. A produção de feijão-vagem é realizada, principalmen- te, em pequenas propriedades rurais (Peixoto et al., 2002). No entanto, o parco desenvolvimento de tecnologias direcionadas para o cultivo orgânico dessa hortaliça limita a obtenção de produções sustentáveis. Uma das estraté- gias que se destaca no cultivo orgânico é o aproveitamen- to de resíduos orgânicos na adubação, usados especial- mente por agricultores familiares, por sua adequação às características das pequenas propriedades (Sediyama et al., 2014). Alguns trabalhos sugerem que a aplicação de cama de frango eleva o rendimento do cultivo de feijão-vagem. San- tos et al. (2001) aplicaram esterco de galinha na dose de 13 t ha-1 e obtiveram a produtividade máxima de 26,3 t ha-1 de vagens com o cultivar Macarrão Favorito, tipo trepador. Em outro trabalho, a aplicação de 26 t ha-1 de cama de aviário com 16,3% de umidade proporcionou produtivida- de máxima de 20 t ha-1 de vagens, com o cv. Alessa (Olivei- ra et al., 2006). Apesar do aumento da produtividade de vagens, outras informações técnicas são necessárias para melhorar o manejo com os adubos orgânicos. Yield and export of nutrients in chicken manure-fertilized snap bean The analysis of nutrient accumulation and exportation in snap bean is fundamental to help develop a more sustainable fertilization, since defining the organic fertilizer doses needed for snap bean cultivation based only on an estimated maximum production does not guarantee that soil fertility is maintained. The objective of this study was to evaluate the effect of chicken manure fertilizer on snap bean yield, accumulation, and nutrient exportation. The experiment was arranged in a randomized block design, with five chicken manure fertilizer doses (0; 5; 10; 20, and 40 t ha-1) and four repetitions. Higher P and Mg contents were verified in the leaves following application of chicken manure fertilizer dose of 40 t ha-1. Maximum pod length was 14.47 cm, estimated with chicken manure fertilizer dose of 33.33 t ha-1. The highest diameter, number of pods per plant, and pod yield were verified at the highest chicken manure fertilizer dose Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha 99 applied (40 t ha-1). In relative terms, i.e., total exported in relation to total extracted of the aerial part, phosphorus is the most exported nutrient in the pods, at an average of 58%, followed by N (55%), K (43%), Mg (40%), S (38%), and Ca (17%). Although a large amount of Ca accumulation (31.3 kg ha-1) was verified at the highest dose, only 13% of it was exported by the pods. Chicken manure fertilizer supplies the essential nutrients and increases pod yield from 7.2 to 16.3 t ha-1, when comparing between no fertilization and the highest chicken manure fertilizer dose. The cultural residues of snap beans are an important source of Ca and Mg, with the following nutrient amounts (kg ha-1): N (51.4); P (5.1); K (27.6); Ca (27.1); Mg (8.2); and S (5.1) obtained at the highest chicken manure fertilizer dose applied (40 t ha-1), which can return to soil through incorporation by the plants. Key words: Phaseolus vulgaris; nutrient content; organic cultivation; plant nutrition. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 MATERIAL E MÉTODOS O experimento foi conduzido no Campo Experimental da Empresa de Pesquisa Agropecuária de Minas Gerais - EPAMIG, em Oratórios, MG, de 23/04 a 03/07/2012. A área experimental está situada em 20°25’49" S e 42°48’20" O, em altitude média de 500 m. O clima da região varia do tipo Cwa, tropical úmido, a Aw, semiúmido de verões quentes; a vegetação natural é de floresta tropical semidecidual ou ombrófila mista (Cunha et al., 2000). A precipitação média anual é de 1.250 mm, a temperatura máxima média anual de 21,8 °C e, a mínima média anual, de 19,5 °C. A colheita foi iniciada aos 58 dias após a semeadura, quando as vagens estavam tenras. Quatro coletas por semana foram realizadas durante 22 dias, para evitar va- gens com características fora do ponto ideal de colheita. No momento da coleta, as vagens foram separadas em comerciais e não comerciais, em cada parcela. As vagens comercializáveis coletadas foram pesadas, contadas e medidas em cada colheita. A pesagem e a con- tagem das vagens foram realizadas imediatamente após a classificação, para quantificar a massa da matéria fresca e o número de vagens, respectivamente. O diâmetro foi me- dido na posição central e, o comprimento, entre as extremi- dades da vagem. Ao final das colheitas, somaram-se os valores obtidos em cada coleta e dividiu-se o valor total pelo número de plantas correspondente à área útil de cada parcela. Os resultados foram expressos em gramas por plan- ta, para massa de vagens frescas; número de vagens por planta; comprimento (cm) e diâmetro (mm) de vagens. O solo, Argissolo Vermelho-Amarelo câmbico, fase ter- raço, apresentou na camada de 0-20 cm: pH (H2O)= 6,0; P= 13,4 mg/dm3; K= 142 mg/dm3; Ca+2= 2,0 cmol/dm3; Mg+2= 1,0 cmol/dm3; Al+3= 0,0; V= 58% e matéria orgânica= 2,1 dag/kg (Embrapa, 2013). Utilizou-se o delineamento experimental em blocos casualizados, com cinco tratamentos e quatro repetições. Os tratamentos foram constituídos pelas doses do esterco de galinha curtido: 0, 5, 10, 20 e 40 t ha-1. O esterco de galinha foi armazendo por seis meses e na época da sua aplicação apresentou as seguintes caracterís- ticas (g/kg): N=24,1; P=21,6; K=1,0; Ca=115,4; Mg=6,7 e S=4,0; para C.O.=19,97 dag/kg, pH=8,20, C/N= 8,28 e teor de umidade=13,44%, em estufa a 75 °C. INTRODUÇÃO Minas Gerais está entre os principais Estados produ- tores de aves, atividade que gera grande volume de resí- duos orgânicos, em especial as camas de galinhas poedeiras e de frangos de corte (UBA, 2014). Em geral, esses adubos orgânicos são muito utilizados na produção de hortaliças, por seus teores de nutrientes, especialmente nitrogênio, cálcio e fósforo, mais altos que os do esterco bovino, do húmus de minhoca e dos compostos orgânicos (Abreu et al., 2010). Quantificar a extração e a exportação de nutrientes no cultivo de feijão-vagem pode fornecer dados importantes para a definição das doses de cama de frango mais susten- táveis, ou seja, que utilizem o mínimo de adubo, sem redu- zir o potencial de ganho na produtividade. Embora essas informações sejam importantes para o programa de aduba- ção e manejo do resíduo, há carência de estudos para o cultivo dessa hortaliça, tanto com adubação química, quan- to com orgânica. A cultura do feijão-vagem é conhecida por ser responsiva ao fósforo, com aumento significativo de va- gens por planta e, consequentemente, maior produtivida- de (Oliveira et al., 2005). Essa característica, juntamente com o baixo teor de fósforo e sua alta adsorção nos solos de Minas Gerais, exige que a adubação seja realizada cons- tantemente (Santos et al., 2008). Além disso, o feijoeiro é cultura exigente em nutrientes, em função do pequeno e pouco profundo sistema radicular e do ciclo curto. A ab- sorção de nutrientes, especialmente o N, ocorre pratica- No cultivo de Phaseolus vulgaris para produção de grãos, a aplicação de 120 kg ha-1 de N proporcionou a extração de 84,2 e 40 kg ha-1 e exportação de 80,2 e 55 kg ha- Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Ivan de Paiva Barbosa Magalhães et al. 100 1 de N, no sistema de plantio direto, nos anos agrícolas 2007/08 e 2008/09, respectivamente (Perez et al., 2013). É provável que o cultivo de feijoeiro, em relação à extração e à exportação de nutrientes, apresente comportamentos diferentes, quanto aos objetivos de comercializar grãos ou vagens. Assim, objetivou-se, neste trabalho, avaliar o efeito da adubação orgânica com esterco de galinha sobre o es- tado nutricional, o acúmulo e a exportação de nutrientes e a produtividade de feijão-vagem, cv. Macarrão Favorito, tipo trepador. feitas pelo método de gotejamento, usando-se fitas perfu- radas com intervalos de 10 cm, dispostas em cada fileira de plantas. INTRODUÇÃO Foram feitas pulverizações quinzenais com urina de vaca fermentada a 1,0%, até o florescimento das plan- tas. A análise química da urina indicou as seguintes carac- terísticas, em %: N=6,96; P=0,0; K=0,89; Ca=0,00; Mg=0,04; S=0,03; C. Org.=0,17; em mg/kg Zn=0,0; Fe=1,0; Mn=0,0; Cu=0,0 e pH=8,5. Quando as plantas estavam em pleno florescimento, aos 40 dias de idade, realizou-se a coleta da quarta folha, indicadora do estado nutricional, em seis plantas de cada parcela. O material foi secado em estufa com circulação de ar, à temperatura de 65 °C, por 72 h, foi moído e amostras foram enviadas para determinação dos teores de N, P, K, Ca, Mg e S, de acordo com procedimento da Embrapa (2009). Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 MATERIAL E MÉTODOS A aplicação de plantio, de metade da dose definida para cada tratamento, foi realiza- da na profundidade de 0-15 cm e incorporada com enxada, uma semana antes da semeadura (18/04/2012). O restante foi aplicado em cobertura e incorporado à profundidade 0-5 cm, aos 30 dias após a semeadura (23/05/2012). Para determinação da umidade foi retirada uma amos- tra de 20 vagens, ao acaso, de cada parcela. As vagens foram submetidas à secagem em estufa de circulação for- çada de ar, com temperatura de 65 ºC, por 72 h. Pesou-se e calculou-se a umidade das vagens para, finalmente, calcu- lar a massa das vagens secas. A produtividade foi obtida a partir do valor total da massa de vagens frescas de cada parcela, obtida ao longo das colheitas realizadas. O resultado foi expresso em tone- ladas de vagens por hectare. Cada parcela continha 40 plantas, em quatro fileiras de 3,0 m de comprimento, espaçadas de 1,0 m. O espaçamento entre plantas foi de 0,3 m. Foram colhidas 16 plantas cen- trais, em cada parcela. A semeadura foi realizada com as sementes de feijão-vagem do cultivar Macarrão Favorito, tipo trepador. Foram semeadas duas sementes por cova e aos 15 dias após a semeadura foi realizado o desbaste, deixando-se uma planta por cova. Após a última coleta das vagens (80 dias), as folhas, ramos e os caules das plantas foram coletados, pesados e picados separadamente. Em seguida, coletou-se uma amos- tra deste material picado para secagem em estufa com cir- culação de ar, à temperatura de 65 °C, por 72 h. Após essa secagem, o material foi moído e as amostras foram envia- das ao laboratório, para determinar os teores de N, P, K, Ca, Mg e S, de acordo com procedimento da Embrapa (2009). O manejo de plantas daninhas foi realizado por meio de duas capinas com enxadas, nas linhas, e três roçadas na parte externa das linhas de plantio. As irrigações foram Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha 101 nar aumento do teor nas folhas indicadoras. É provável que o fornecimento de Ca pelo esterco de galinha, por sua baixa translocação na planta, não tenha favorecido maior absorção desse nutriente (Brackmann et al., 2010). MATERIAL E MÉTODOS No entanto, o baixo teor de cálcio nas folhas não refletiu em queda da produtividade; além disso, a planta não ex- pressou características morfológicas que mostrassem deficiência desse elemento. Miranda et al. (2010), em es- tudo da resposta de feijão-de-corda decorrente da omis- são de macro e micronutrientes, observaram que o Ca é o primeiro nutriente cuja deficiência manifesta-se em sinto- mas visíveis, além de sua ausência ser fator responsável pelo maior prejuízo ao desenvolvimento e à produção da biomassa das plantas. O acúmulo de nutrientes foi calculado a partir do teor do nutriente das folhas, dos ramos e das vagens e conver- tido para kg ha-1, de acordo com a massa da matéria seca de cada componente. A exportação de nutrientes foi de- terminada a partir do teor de nutrientes nas vagens e con- vertida em kg ha-1, de acordo com a média da massa de matéria seca das vagens, obtida ao longo das coletas rea- lizadas no experimento. Os dados foram analisados, utilizando-se o programa estatístico SAEG (SAEG, 2007), sendo submetidos à análi- se de variância e de regressão polinomial, a 5% de proba- bilidade. Os modelos foram escolhidos com base no signi- ficado biológico, maior significância dos regressores e coeficiente de determinação. A aplicação de esterco de galinha proporcionou au- mento do teor de fósforo e de magnésio na folha indicadora. O teor foliar de P respondeu de forma quadrática, e o de Mg de forma linear, às doses de esterco. Os maiores teores foliares, 5,53 g kg-1 de P e 4,4 g kg-1 de Mg, foram obtidos na maior dose de esterco aplicada (Figuras 1A e 1B). O fornecimento equilibrado de fósforo, desde o início do desenvolvimento vegetativo, estimula o crescimento radicular logo na fase inicial e melhora a formação dos primórdios das partes reprodutivas e dos frutos (Raij, 1991). O fósforo é o nutriente que possibilita respostas mais acen- tuadas na produção de feijão-vagem (Filgueira, 2008), en- quanto o magnésio, como parte integrante da molécula da clorofila, é essencial para o crescimento das plantas e a base para realização da fotossíntese (Fontes, 2011). RESULTADOS E DISCUSSÃO 102 enchimento e no crescimento de grãos e frutos, melhoran- do a qualidade do produto e o valor de mercado (Filgueira, 2008). Oliveira et al. (2008) observaram o aumento do nú- mero de vagens por planta de feijão-vagem,em função de doses de K2O, mostrando que o K desempenha importan- te papel na produção de vagens. Esse resultado pode ser associado ao maior estímulo ao desenvolvimento do sistema radicular, à formação dos primórdios reprodutivos e dos frutos, proporcionados pelo aumento da absorção de fósforo, em função das doses de esterco aplicadas (Ishimura et al., 1985). Araújo et al. (2001) trabalharam com o mesmo cultivar de feijão- vagem e não encontraram resposta em diâmetro de va- gens em função das doses de esterco suíno, na presença e na ausência da adubação mineral, com valores médios de 10,8 e 10,6 mm, muito próximos dos obtidos neste tra- balho. Esse resultado pode ser associado ao maior estímulo ao desenvolvimento do sistema radicular, à formação dos primórdios reprodutivos e dos frutos, proporcionados pelo aumento da absorção de fósforo, em função das doses de esterco aplicadas (Ishimura et al., 1985). Araújo et al. (2001) trabalharam com o mesmo cultivar de feijão- vagem e não encontraram resposta em diâmetro de va- gens em função das doses de esterco suíno, na presença e na ausência da adubação mineral, com valores médios de 10,8 e 10,6 mm, muito próximos dos obtidos neste tra- balho. Observou-se também resposta significativa, de forma linear decrescente, dos teores de Ca nas vagens, em res- posta ao aumento das dose de esterco de galinha (Figura 2B). Para o N, P, Mg e S, os teores médios nas vagens foram iguais a 33,9; 4,2; 2,8 e 2,0 g kg-1, respectivamente. O aumento das doses de esterco proporcionou maior cresci- mento da parte aérea e reduziu a concentração de Ca nas vagens, provavelmente, por causa da baixa mobilidade do Ca nos tecidos, que depende da transpiração, sendo acu- mulado em tecidos que transpiram mais facilmente, princi- palmente nas folhas (Beninni et al., 2003). Além disso, o Ca é pouco móvel no floema, pois uma vez assimilado, não apresenta redistribuição para as outras partes da planta. Com isso, observou-se menor valor de Ca no fruto com o aumento das doses de esterco. O maior número de vagens por planta (65) foi obtido com a maior dose de esterco (Figura 3C). RESULTADOS E DISCUSSÃO Esse valor é mai- or do que o relatado por Oliveira et al. (2007), com o mes- mo cultivar, na região de Areia, PB, de 20 vagens por plan- ta, com a adição de 145 kg ha-1 de K2O em solo com 53,8 mg dm-3 de potássio residual. O maior número de vagens por planta, observado nesta pesquisa, provavelmente, foi oca- sionado pelo equilíbrio entre os nutrientes, que, por sua vez, é mais importante para o ganho em produtividade do que o aumento das quantidades de macronutrientes, iso- ladamente. É provável também que as características físi- cas do solo tenham favorecido o melhor desenvolvimento das raízes das plantas e a absorção de nutrientes, pois a área experimental tem sido cultivada organicamente desde 2003 (Sediyama et al., 2014). Não houve respostas significativas da massa da ma- téria fresca e da percentagem de matéria seca de vagens às doses de esterco de galinha, cujos valores médios foram 7,64 g e 8,99%, respectivamente. Entretanto, hou- ve respostas significativas do comprimento e do diâme- tro da vagem e do número de vagens por planta (Figura 3). O máximo comprimento da vagem (14,47cm) foi obser- vado para a dose de 33,33 t ha-1 de esterco de galinha (Figura 3A). A aplicação de esterco proporcionou o au- mento linear do diâmetro da vagem, sendo o maior valor (10,29 mm) observado na maior dose (Figura 3B). Resul- tado semelhante ocorreu no número de vagens por plan- ta, o qual foi maior com aplicação de 40 t ha-1 de esterco (Figura 3C). Esse aumento nos componentes da produ- ção proporcionou resposta significativa da produtivida- de de vagens do cultivar Macarrão Favorito, tipo trepador. O número médio de vagens por planta foi o princi- pal componente responsável pelo aumento da produtivi- dade (r > 0,99), visto que a massa média da vagem não apresentou respostas significativas, em função das doses de esterco. Resultados semelhantes foram obtidos por Ishimura et al. (1985), que observaram, ao utilizarem dife- rentes doses de NPK na produção de feijão-vagem, au- mento significativo da produtividade e do número de va- gem por planta, ausência de diferença entre as massas médias das plantas e enfatizaram o fósforo como o nutri- ente de grande importância na produção de frutos. Figura 2: Teores de potássio (A) e de cálcio (B) em vagens de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. RESULTADOS E DISCUSSÃO Na avaliação do estado nutricional das plantas, as do- ses de esterco de galinha influenciaram positivamente nos teores foliares de fósforo e magnésio, mas não apresenta- ram efeito significativo sobre os teores de N, K, Ca e S, cujos valores médios foram: 44,7, 17,5, 10,2 e 2,2 g kg-1, respectivamente. À exceção do K e do Ca, os teores de nutrientes apresentaram-se dentro do intervalo conside- rado adequado para a cultura: N (40-60 g kg-1), P (3-7 g kg- 1), Mg (3-8 g kg-1) e S (2-5 g kg-1), de acordo com Raij et al. (1997). Os teores médios de K e de Ca encontrados nas fo- lhas foram inferiores aos valores de referência: 25-40 g kg-1 de K e 15-30 g kg-1 de Ca (Raij et al. 1997). O baixo teor foliar de K deve-se, provavelmente, ao baixo teor desse nutriente no esterco de galinha (1,0 g kg-1 de K), além da maior quantidade de Ca aplicada com o esterco (115,4 g kg-1 de Ca), que pode ter levado a uma inibição da absorção de K, por competição, pelo excesso de Ca, como observado para algumas plantas (Malavolta et al., 1997). Por outro lado, o esterco apresentou alto teor de Ca (115,4 g kg-1), mas não foi suficiente para proporcio- Embora o K não tenha apresentado teor adequado na folha indicadora, observou-se aumento significativo do teor desse nutriente nas vagens, com o aumento das do- ses de esterco aplicadas. O máximo teor de K nas vagens foi estimado com a dose de 31,39 t ha-1 de esterco de gali- nha (Figura 2A). O K favorece a formação e translocação das reservas na planta, constituídas por carboidratos, au- menta a eficiência do uso da água pela planta e auxilia no Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Figura 1: Teores de fósforo (A) e de magnésio (B) em folhas de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * significativo a 5% de probabilidade pelo teste F, respectivamente. Figura 1: Teores de fósforo (A) e de magnésio (B) em folhas de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * significativo a 5% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Ivan de Paiva Barbosa Magalhães et al. RESULTADOS E DISCUSSÃO , 0, significativo a 5 e 10% de probabilidade pelo teste F, respectivamente. Figura 2: Teores de potássio (A) e de cálcio (B) em vagens de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. , 0, significativo a 5 e 10% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha 103 dutividades alcançadas nesta pesquisa são próximas das obtidas por Peixoto et al. (2002), que estudaram a adapta- bilidade e a estabilidade de 15 genótipos de feijão-vagem, de crescimento indeterminado, em oito ambientes, e obti- veram valores entre 15,16 a 17,68 t ha-1, com uma popula- ção de 50.000 plantas por ha. A maior produtividade de vagens (16,3 t ha-1) foi alcançada com a dose de 40 t ha-1 de esterco (Figura 3D). É provável que a suplementação de nutrientes pelo esterco de galinha tenha favorecido o desenvolvimento das plan- tas nas fases vegetativa e reprodutiva. Santos et al. (2001) também constataram aumento de 11,3 t ha-1 da produtivi- dade de vagens por meio da aplicação de esterco de gali- nha, obtendo-se produtividade máxima de 26,3 t ha-1 de vagens, com 13,0 t ha-1 do esterco. De modo geral, as pro- As massas de matérias fresca e seca da parte aérea das plantas de feijão-vagem responderam de forma linear cres- cente ao aumento das doses de esterco de galinha. Na Figura 3: Comprimento médio das vagens (A), diâmetro médio das vagens (B), número médio das vagens (C), número médio das vagens por planta (D) e produtividade de vagens (E) das plantas de feijão-vagem, cv. Macarrão Favorito, cultivadas com esterco de galinha. , 0 significativo a 5 e 10% de probabilidade pelo teste F, respectivamente. Figura 3: Comprimento médio das vagens (A), diâmetro médio das vagens (B), número médio das vagens (C), número médio das vagens por planta (D) e produtividade de vagens (E) das plantas de feijão-vagem, cv. Macarrão Favorito, cultivadas com esterco de galinha. , 0 significativo a 5 e 10% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Figura 4: Matérias fresca (A) e seca (B) da parte aérea de plantas de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * significativo a 5% de probabilidade pelo teste F. RESULTADOS E DISCUSSÃO Figura 4: Matérias fresca (A) e seca (B) da parte aérea de plantas de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * significativo a 5% de probabilidade pelo teste F. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Figura 4: Matérias fresca (A) e seca (B) da parte aérea de plantas de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * significativo a 5% de probabilidade pelo teste F. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Ivan de Paiva Barbosa Magalhães et al. 104 maior dose de esterco de galinha (40 t ha-1), foram obtidas 9,72 t ha-1 de matéria fresca e 2,05 t ha-1 de matéria seca, ou seja, aumentos de 6,72 t de matéria fresca e 1,31 t de maté- ria seca, ao comparar-se com as das plantas cultivadas sem adubação (Figuras 4A e 4B). As massas de matérias fresca e seca encontradas nos tecidos das plantas de fei- jão-vagem têm grande importância residual para o sistema de produção, pois retornam nutrientes e matéria orgânica, com a incorporação das plantas ao solo, e assim contribu- em para o aumento e a manutenção da fertilidade do solo e, consequentemente, da produtividade . Este comportamento pode ser atribuído também ao au- mento da biomassa da planta, pois as doses crescentes de esterco não influenciaram os teores foliares desses nutri- entes. Este resultado sugere que a concentração de nutri- entes no solo estava em condições adequadas para suprir as necessidades das plantas. Houve aumento linear para acúmulo de N, P, Mg e S pelas plantas (massa de matéria fresca de plantas + va- gens), em função das doses de esterco aplicadas. No caso do N, o maior valor obtido foi de 101,1 kg ha-1 de N (Figura 5A). A maior produtividade de vagens e o maior número de vagens por planta, além dos teores de nutrientes na plan- ta, favoreceram o acúmulo de nutrientes. Neste trabalho, a maior produção de massa de matéria fresca da planta pode A aplicação de esterco de galinha aumentou de forma linear os acúmulos de K, Ca e S pelas plantas (massa de matéria fresca de plantas + vagens) (Figuras 5B, 5E e 5F). RESULTADOS E DISCUSSÃO Ceres Viçosa v 64 n 1 p 098-107 jan/fev 2017 ra 5: Acúmulos de nitrogênio (A), fósforo (B), potássio (K), cálcio (Ca) magnésio (Mg) e enxofre (S) pelas plantas de feijão- m (plantas + vagens), cv. Macarrão favorito, adubadas com esterco de galinha. *, e 0 significativo a 1, 5 e 10% de probabilidade este F, respectivamente. Figura 5: Acúmulos de nitrogênio (A), fósforo (B), potássio (K), cálcio (Ca) magnésio (Mg) e enxofre (S) pelas plantas de feijão- vagem (plantas + vagens), cv. Macarrão favorito, adubadas com esterco de galinha. *, e 0 significativo a 1, 5 e 10% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha 105 ter sido o principal fator para maior acúmulo, pois o teor de N na folha não foi influenciado pelas doses crescentes de esterco de galinha. ficativas em seu teor nas folhas. Esse comportamento pode ser atribuído à competição pelos sítios de adsorção de P no solo, condição que ocasiona o aumento da dis- ponibilidade desse nutriente para as plantas (Hue, 1991). Além disso, a área sob manejo orgânico apresenta me- lhor estrutura físico-química e melhor mobilidade de água, condições fundamentais para a absorção de P, que ocor- re por difusão (Novais & Smyth, 1999). Essa maior dispo- nibilidade favoreceu maior teor de P nas folhas, consequentemente, maior acúmulo com o aumento da dose de esterco. O maior acúmulo de P (11,3 kg ha-1) ocorreu na maior dose de esterco aplicada. O acúmulo e a exportação de P são quantitativamente mais baixos que os dos outros macronutrientes, como N e K, mas isso não implica que seu fornecimento seja menos importante. Pelo contrário, corresponde ao nutriente com maior resposta à aduba- ção da cultura, pois à medida que aumentaram as doses de esterco de galinha, o acúmulo desse nutriente também aumentou (Figura 5B). É provável que o esterco de gali- nha, mesmo com teores de P relativamente baixos, em comparação com os teores das doses normalmente usa- das na adubação mineral, proporcionou respostas signi- Com a maior produtividade (16,3 t ha-1), alcançada com a maior dose de esterco de galinha, as quantidades de macronutrientes exportadas pelas vagens, em ordem de- crescente, foram, em kg ha-1, N (49,7) > K (18,7) > P (6,2) > Rev. Ceres, Viçosa, v. 64, n.1, p. REFERÊNCIAS Abreu IMO, Junqueira AMR, Peixoto JR & Oliveira AS (2010) Qualidade microbiológica e produtividade de alface sob aduba- ção química e orgânica. Ciência e Tecnolologia de Alimentos, 30:108-118. As doses de esterco não influenciaram nas propor- ções de nutrientes exportados, sendo estas iguais à mé- dia para todos os tratamentos. Em todas as doses, os nutrientes exportados em maior quantidade foram o N e K, com valores consideravelmente superiores aos dos demais nutrientes. No entanto, em termos relativos, ou seja, quantidade exportada em relação a total acumulado na parte aérea (massa de matéria fresca de plantas + va- gens), o P é o nutriente mais exportado com a colheita das vagens, em média, 58%, seguido de N (55%), K (43%), Mg (40%), S (35%) e Ca (17%). Na maior dose, observou- se que, embora o acúmulo de Ca na planta tenha sido de 31,3 kg ha-1, apenas 13% deste foi exportado pelas va- gens. Esse menor teor de Ca nas vagens pode ser expli- cado pela baixa mobilidade deste nutriente, com tendên- cia a acumular na parte aérea e pouca translocação para os frutos. Araújo JS, Oliveira AP, Silva JAL, Ramalho CI & Neto FLC (2001) Rendimento do feijão-vagem cultivado com esterco suíno e adubação mineral. Revista Ceres, 48:501-510. Beninni ERY, Takahashi HW & Neves CSVJ (2003) Manejo do cálcio em alface de cultivo hidropônico. Horticultura Brasilei- ra, 21:605-610. Brackmann A, Schorr MRW, Pinto JAV & Venturini TL (2010) Aplicações pré-colheita de cálcio na qualidade pós-colheita de maçãs ‘Fuji’. Ciência Rural, 40:1435-1438. Brito MMP, Muraoka T & Silva EC (2011) Contribuição da fixa- ção biológica de nitrogênio, fertilizante nitrogenado e nitrogê- nio do solo no desenvolvimento de feijão e caupi. Bragantia, 70:206-215. CEASAMINAS (2013) Centro de Abastecimento de Minas Gerais. Informações Nutricionais e Informações de Mercado. Disponí- vel em: <http://www.ceasaminas.com.br> Acessado em: 5 de dezembro de 2014. A incorporação da parte aérea das plantas de feijão- vagem adubadas com 40 t ha-1 de esterco de galinha pode- rá devolver ao solo as seguintes quantidades de macronutrientes (kg ha-1): N (51,4); P (5,1); K (27,6); Ca (27,1); Mg (8,2); S (5,1). Assim, os restos vegetais do fei- jão-vagem constituem importante fonte de nutrientes, que podem retornar ao solo no final do ciclo da cultura e con- tribuir para a manutenção da sua fertilidade. RESULTADOS E DISCUSSÃO 098-107, jan/fev, 201 Figura 6: Exportação dos macronutrientes pelas vagens de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. * e 0 significativo a 1, 5 e 10% de probabilidade pelo teste F, respectivamente. Figura 6: Exportação dos macronutrientes pelas vagens de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. , e 0 significativo a 1, 5 e 10% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Figura 6: Exportação dos macronutrientes pelas vagens de feijão-vagem, cv. Macarrão favorito, adubadas com esterco de galinha. *, e 0 significativo a 1, 5 e 10% de probabilidade pelo teste F, respectivamente. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Ivan de Paiva Barbosa Magalhães et al. 106 AGRADECIMENTOS Mg (4,3) > Ca (4,2) > S (2,9). Assim, as quantidades expor- tadas pelas vagens apresentaram resposta significativa às doses aplicadas (Figura 6). Essas informações são funda- mentais para auxiliar na indicação da melhor adubação da cultura, pois conhecendo a quantidade exportada pelas vagens é possível quantificar o que deve ser reposto ao solo antes de cada cultivo, em função da eficiência do fertilizante, para a manutenção da fertilidade e a garantia do potencial produtivo da cultura. Os autores agradecem à Fundação de Amparo à Pes- quisa do Estado de Minas Gerais (FAPEMIG) e ao Conse- lho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), pelo auxílio financeiro ao projeto e pelas bolsas PIBIC, BIPDT e PQ. REFERÊNCIAS Cunha TJF, Blancaneaux P, Calderano Filho B, Carmo CAFS, Garcia NCP & Lima BEM (2000) Influência da diferenciação pedológica no desenvolvimento da seringueira no município de Oratórios, MG. Pesquisa Agropecuária Brasileira, 35:145-155. Embrapa - Empresa Brasileira de Pesquisa Agropecuária (2009) Manual de análises químicas de solos, plantas e fertilizantes. Brasília, Embrapa Informação Tecnológica. 627p. Embrapa - Empresa Brasileira de Pesquisa Agropecuária (2013) Sistema Brasileiro de Classificação de Solos. 3ª ed. Brasília, Embrapa. 353p. Filgueira FAR (2008) Novo Manual de Olericultura: Agrotecnologia moderna na produção e comercialização de hortaliça. Viçosa, UFV. 421p. Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha 107 Novais RF & Smyth TJ (1999) Fósforo em solo e planta em condições tropicais. Viçosa, UFV. 399p. Novais RF & Smyth TJ (1999) Fósforo em solo e planta em condições tropicais. Viçosa, UFV. 399p. Raij B Van (1991) Fertilidade do solo e adubação. Piracicaba, Potafós. 343p. Raij B Van (1991) Fertilidade do solo e adubação. Piracicaba, Potafós. 343p. Oliveira AP, Cardoso MMO, Barbosa LJN, Silva JEL & Morais MS (2005) Resposta do feijão-vagem a P2O5 em solo arenoso com baixo teor de fósforo. Horticultura Brasileira, 23:128-132. Raij B Van, Cantarella H, Quaggio JA & Furlani AMC (1997) Recomendações de adubação e calagem para o Estado de São Paulo. 2ª ed. Campinas, IAC. 285p. (Boletim Técnico, 100). Rosolem CA & Marubayashi OM (1994) Seja o doutor do seu feijoeiro. In: Informações Agronômicas, 68:01-16. Oliveira NG, De-Polli H, Almeida DL & Guerra JGM (2006) Fei- jão-vagem semeado sobre cobertura viva perene de gramínea e leguminosa e em solo mobilizado, com adubação orgânica. Pes- quisa Agropecuária Brasileira, 41:1361-1367. SAEG (2007) SAEG: Sistema para análises estatísticas, versão 9.1. Viçosa, Fundação Arthur Bernardes. CD-ROM. Oliveira AP, Silva JA, Alves AU, Dorneles CSM, Alves AU, Olivei- ra ANP, Cardoso EA & Silva Cruz IS (2007) Rendimento de feijão-vagem em função de doses de K2O. Horticultura Brasilei- ra, 25:29-33. Santos GM, Oliveira AP, Silva JAL, Alves EU & Costa CC (2001) Characteristics and yield of snap-bean pod in function of sources and levels of organic matter. Horticultura Brasileira, 19:30-35. Santos DR, Gatiboni LC & Kaminski J (2008) Fatores que afetam a disponibilidade do fósforo e o manejo da adubação fosfatada em solos sob sistema plantio direto. Ciência Rural, 38:576-586. Oliveira FL, Guerra JGM, Almeida DL, Ribeiro RLD, Silva ED, Silva VV & Espindola JAA (2008) Desempenho de taro em função de doses de cama de aviário, sob sistema orgânico de produção. Horticultura Brasileira, 26:149-153. Sediyama MAN, Santos IC & Lima PC (2014) Cultivo de hortali- ças no sistema orgânico. Revista Ceres, 61:829-837. Peixoto N, Braz LT, Banzatto DA & Oliveira AP (2002) Adapta- bilidade e estabilidade em feijão-vagem de crescimento indeterminado. Horticultura Brasileira, 20:616-618. Souza EFC, Soratto RP & Pagani FA (2011) Aplicação de nitrogê- nio e inoculação com rizóbio em feijoeiro cultivado após milho consorciado com braquiária. Pesquisa Agropecuária Brasileira, 46:370-377. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 CONCLUSÕES A adubação do solo com esterco de galinha para o cultivo de feijão-vagem cv. Macarrão Favorito melhora o estado nutricional da planta e aumenta a sua produtivida- de de 7,2 para 16,3 t ha-1, com aplicação de 40 t ha-1 de esterco. Fontes PCR (2011) Nutrição mineral de plantas: avaliação e diagnose. Viçosa, Arka. 296p. Hue NV (1991) Effects of organic acids/anions on phosphorus sorption and phytoavailability in soils with different mineralogies. Soil Science, 152:463-471. IBGE (2006) Censo Agropecuário Brasil. Grandes Regiões e Uni- dades da Federação. Disponível em: <www.ibge.gov.br>. Acessado em: 23 de janeiro de 2015. As quantidades de macronutrientes (kg ha-1) exporta- das pelas vagens, em ordem decrescente, foram: N (49,7) > K (18,7) > P (6,2) > Mg (4,3) > Ca (4,2) > S (2,9), consideran- do-se uma produtividade de vagens de 16,3 t ha-1 e aduba- ção com 40 t ha-1 de esterco de galinha. Ishimura I, Feitosa CT, Lisbão RS, Passos FA, Fornasier JB & Noda M (1985) Diferentes doses de N, P, K na produção do feijão-vagem em solo orgânico álico do Vale do Ribeira (SP). Bragantia, 44:429-436. Os restos vegetais do feijão-vagem constituem impor- tante fonte de nutrientes, sendo obtidas, na maior dose de esterco (40 t ha-1), as seguintes quantidades de macronu- trientes (kg ha-1): N (51,4); P (5,1); K (27,6); Ca (27,1); Mg (8,2); S (5,1), que poderão retornar ao solo, com a incorpo- ração das plantas. Malavolta E, Vitti GC & Oliveira AS (1997) Avaliação do estado nutricional das plantas: princípios e aplicações. 2ª ed. Piracicaba, Potafos. 319p. Miranda RS, Suderio FB, Sousa AF & Gomes Filho E (2010) Defi- ciência nutricional em plântulas de feijão-de-corda decorrente da omissão de macro e micronutrientes. Revista Ciência Agro- nômica, 41:326-333. Rev. Ceres, Viçosa, v. 64, n.1, p. 098-107, jan/fev, 2017 Produtividade e exportação de nutrientes em feijão-vagem adubado com esterco de galinha Perez AAG, Soratto RP, Manzatto NP & Souza EFC (2013) Ex- tração e exportação de nutrientes pelo feijoeiro adubado com nitrogênio, em diferentes tempos de implantação do sistema plantio direto. Revista Brasileira de Ciência do Solo, 37:1276- 1287. UBA - União Brasileira de Avicultura (2014) Relatório Anual. Disponível em: <http://www.ubabef.com.br/publicacoes> Acessado em: 23 de janeiro de 2015.
https://openalex.org/W4283713922	https://scindeks-clanci.ceon.rs/data/pdf/1452-662X/2022/1452-662X2201011H.pdf	English	null	A predictive value of early clinical parameters for abnormal brain MRI scan in neonates treated with therapeutic hypothermia	Sanamed	2,022	cc-by	3,516	INTRODUCTION The neuronal injury after perinatal asphyxia is a process that lasts, rather than being an isolated, one- time event (1). The primary cell injury is a direct con sequence of hypoxia and ischemia. During the prima ry injury the cerebral concentrations of energy-rich compounds, adenosine triphosphate, and phosphorus creatine are decreased, which causes cytotoxic edema and neuronal injury. The next stage is the reperfusion stage, where cellular edema subsides in approximately thirty minutes. Next comes the latent phase, in which oxidative cerebral energy metabolism is very close to normal and lasts around 6 hours after acute asphyx ia. After the latent phase, secondary injury develops with extracellular accumulation of excitatory amino acids, free radicals, and induction of apoptosis and inflammatory activity with a final breakdown of ox idative metabolism and neuronal death. Therapeutic hypothermia is a standard neuroprotective therapy in asphyxiated neonates (2). The therapeutic window for the implementation of hypothermia corresponds to the latent phase of hypoxic-ischemic brain injury (3). Aim: To identify which of three selected clinical parameters (oral feeding ability, muscle tone, history of seizure) evaluated 10 days after therapeutic hypo thermia could predict the primary outcome of an ab normal brain MRI. Methods: We reviewed the medical records of neonates ≥ 36 completed weeks of gestation consec utively treated with therapeutic hypothermia who underwent brain MRI. Clinical parameters on day 10 after therapeutic hypothermia were correlated with brain MRI findings in the first 7-14 days of life. Logic regression analysis was performed using all three co variates of the clinical status, with an abnormal MRI as the primary outcome. Results: Brain MRI was abnormal in 42 (51.85 %) neonates with the following distribution of brain injury patterns: abnormal signal in the basal nuclei in 6, an abnormal signal in the cortex in 16, an abnor mal signal both in the cortex and basal nuclei in 20 neonates. Out of three analyzed clinical parameters, feeding difficulty (P < 0.001, OR 8.3, 95% CI 2.9 - 28.9) and a history of seizures (P < 0.001, OR 11.95, 95% CI 3 - 44.5) were significantly associated with an abnormal MRI. The brain magnetic resonance imaging (MRI) findings after perinatal hypoxia have a predictive val ue for the neurodevelopmental outcome. 2022; 17(1): 11–15 ISSN-1452-662X 2022; 17(1): 11–15 ISSN-1452-662X DOI: 10.5937/sanamed17-36698 UDK: 616.831-073.7; 615.832.9.065 ID: 67536137 Original article DOI: 10.5937/sanamed17-36698 UDK: 616.831-073.7; 615.832.9.065 ID: 67536137 Original article Primljen/Received 28. 02. 2022. god. Primljen/Received 28. 02. 2022. god. abnormal MRI. This may be used in selective planning of pre-discharge MRI in asphyxiated neonates. Abstract: Introduction: Brain MRI scans can pre dict neurodevelopmental outcomes in neonates treated with therapeutic hypothermia. It is a common clinical practice to perform brain MRI before discharge, but brain MRI scans performed at around four months of age have a better prognostic value for a long-term neu rological outcome in asphyxiated neonates. Keywords: therapeutic hypothermia, clinical pa rameters, feeding difficulties, seizures, brain MRI. Prihvaćen/Accepted 18. 04. 2022. god. Prihvaćen/Accepted 18. 04. 2022. god. Prihvaćen/Accepted 18. 04. 2022. god. Hadzimuratovic Emina,1 Hadzimuratovic Admir,1 Pokrajac Danka,1 Selimovic Amina,1 Muhasilovic Senad2 Hadzimuratovic Emina,1 Hadzimuratovic Admir,1 Pokrajac Danka,1 Selimovic Amina,1 Muhasilovic Senad2 1 Pediatric Clinic, University Medical Center of Sarajevo, Sarajevo, Bosnia and Herzegovina 2 Sarajevo School of Science and Technology Program Coordinator, Sarajevo, Bosnia and Herzegovina RESULTS We analyzed an MRI scan and three selected clin ical parameters (oral feeding ability, muscle tone, his tory of seizure) in 81 neonates treated with therapeutic hypothermia. MRI scans were performed at median postnatal age of 9 days (range 6 to 24 and 50% IQR 6 to 10). Ten days after completion of treatment with therapeutic hypothermia 45 neonates (55.55%) failed to achieve full oral feeding, 24 (29.63%) had a devia tion in muscle tone (hypotonia or hypertonia) and 53 (65.43%) had a history of one or more seizures which required treatment with anticonvulsants. Brain MRI was abnormal in 42 (51.85%) neonates. 35 out of 45 neonates with feeding difficulties, 20 out of 24 neo nates with a deviation in muscle tone, and 38 out of 53 neonates who had clinical seizures by day 10 after completion of therapeutic hypothermia had an abnor mal MRI. Table 1. shows the predictive values of these three selected clinical findings evaluated for an abnor mal MRI.i Both three selected clinical parameters (oral feed ing ability, muscle tone, history of seizure) and brain MRI scans were analyzed in 81 of the 88 neonates treated with therapeutic hypothermia. The medical re cords of the seven neonates who died were excluded. The therapeutic hypothermia for 72 h, initiated before 6 h of life was performed according to Bristol Royal Hospital’s guidelines for therapeutic hypothermia in neonates. In all cases, prior to the initiation of thera peutic hypothermia parental consent was obtained. All neonates underwent a brain MRI before dis charge. The brain MRI was performed using a 1,5-T magnet with T1- and T2- weighted imaging and the scans were interpreted by neuroradiologists. We used the categorization of the brain injury similar to the categorization described by Barkovich et al (6). This The brain MRI findings for all 81 neonates were: normal scan in 35, an abnormal signal in the basal nu Table 1. INTRODUCTION The common clinical practice is to perform brain MRI before dis charge, but the studies of serial MRI findings follow ing therapeutic hypothermia showed that brain MRI findings at around four months of age have the highest Conclusion: Neonates who were capable of full oral feeding by day 10 after therapeutic hypothermia and had no history of seizures were unlikely to have an Hadzimuratovic Emina, Hadzimuratovic Admir, Pokrajac Danka, Selimovic Amina, Muhasilovic Senad 12 categorization is based on the recognition of two basic imaging patterns of hypoxic-ischemic brain injury, one in which the primary damage is to the basal nuclei and another in which the damage is primarily to the cor tex (6–9). Injuries to both basal nuclei and cortex were considered ‘severely abnormal’. correlation with a long-term neurodevelopmental out come (4). Thus, in some cases, it might be better to prolong brain MRI to that age. This study aimed to analyze a correlation between the clinical findings ten days after therapeutic hypo thermia and the brain MRI findings which might lead to identifying neonates who should have a brain MRI performed before discharge and neonates in whom it is better to postpone the brain MRI scan to be more in formative. This selective and targeted approach to the brain MRI evaluation of asphyxiated infants might be more useful. The three selected clinical parameters (oral feed ing ability, muscle tone, history of seizure) were eval uated 10 days after therapeutic hypothermia. The three selected clinical parameters (oral feed ing ability, muscle tone, history of seizure) were eval uated 10 days after therapeutic hypothermia. Data were analyzed by PASW Statistics 18. The clinical parameters between the groups were compared by Χ² test or t-test or Mann - Whitney test, as appro priate. The multivariate analysis was performed to de termine which of the selected clinical parameters could predict an abnormal or severely abnormal MRI scan. Two-sided P-values ≤ 0.05 were regarded as significant. MATERIAL AND METHODS The medical records of 88 neonates ≥ 36 com pleted weeks of gestation consecutively treated with therapeutic hypothermia between December 2017 and December 2021 at Pediatric Clinic University Medi cal Center Sarajevo, Bosnia and Herzegovina were reviewed. Our research was conducted by the ethical standards of the Declaration of Helsinki and the ethical standards of the University Medical Center of Saraje vo, Bosnia and Herzegovina. Abbrevations: TH, therapeutic hypothermia; PPV, positive predictive value; NPV, negative predictive value, CI, confidence interval; OR odds ratio ons: TH, therapeutic hypothermia; PPV, positive predictive value; NPV, negative predictive value, CI, confidence OR odds ratio A PREDICTIVE VALUE OF EARLY CLINICAL PARAMETERS FOR ABNORMAL BRAIN MRI SCAN IN NEONATES TREATED... 13 Table 2. Significance of clinical parameters for prediction of ‘severely abnormal’ brain MRI Clinical parameter Sensitivity (%) Specificity (%) PPV (%) NPV (%) P-value (QR, 95%CI) Inability of oral feeding 83 54 38 92 0.013 5.5,1.3-21.5 Hystory of clinical seizure 92 36 39 95 0.032 9.5,1.2-78.5 Abbrevations: PPV, positive predictive value; NPV, negative predictive value, CI, confidence interval; OR odds ratio Table 2. Significance of clinical parameters for prediction of ‘severely abnormal’ brain MRI bbrevations: PPV, positive predictive value; NPV, negative predictive value, CI, confidence interval; OR odds ratio orrhage, ischemia, congenital anomalies,…) and have a prognostic value. Early MRI abnormalities may not represent the final pathology, but the injury that is still evolving and studies show that delaying brain MRI scans to 4 months of age has the highest correlation with a long-term neurodevelopmental outcome (4). In our study, the negative predictive values of oral feed ing ability and a history of seizures ten days after the completion of therapeutic hypothermia for prediction of ‘severely abnormal’ MRI were 92% and 95%. re spectively. This signifies that neonates who attain oral feeding and have no history of seizures are unlikely to have severe changes on brain MRI. According to our results, brain MRI scans before discharge in such cases are not necessary and can be postponed to a period of life when MRI scans are more informative regarding the long-term neurodevelopmental outcome. clei in 7, abnormal signal in the cortex in 16, abnormal signal both in the cortex and in the basal nuclei in 23 neonates. MRI scans showing injury of both basal nu clei and cortex were considered ‘severely abnormal’. Univariate analysis showed that all three selected clinical parameters evaluated 10 days after completion of therapeutic hypothermia were significantly associat ed with abnormal brain MRI scans the inability of oral feeding (P < 0.001, OR 10.2, 95% CI 3.4 – 25.7), de viation in muscle tone (P < 0.001, OR 8.9, 95% CI 2.5 – 29.7) and a history of clinical seizures (P < 0.001, OR 12.5, 95% CI 3.4 – 41.7). When we performed the mul tivariate analysis, only feeding difficulty (P < 0.001, OR 8.3, 95% CI 2.9 - 28.9) and a history of seizure (P < 0.001, OR 11.95, 95% CI 3 - 44.5) was linked with an increased risk of abnormal brain MRI findings. A PREDICTIVE VALUE OF EARLY CLINICAL PARAMETERS FOR ABNORMAL BRAIN MRI SCAN IN NEONATES TREATED... Out of 23 neonates with ‘severely abnormal’ MRI, 20 neonates had difficulties with oral feeding, and 22 had a history of seizures. When we repeated multivar iate regression analysis with all three clinical parame ters for the outcome of ‘severely abnormal’ brain MRI, feeding difficulty and a history of seizures were again associated with the outcome, with negative predictive values of 92% and 95%, respectively (Table 2). In conclusion, oral feeding ability and history of clinical seizures 10 days after completion of therapeu tic hypothermia may be used for a selective and tar geted approach to the brain MRI evaluation of the as phyxiated neonates. This approach may be more useful since delayed brain MRI scans are more informative in terms of long-term neurodevelopmental outcomes. Abbreviations Currently, there are several clinical scoring sys tems in the early neonatal period proposed for the prediction of neurological outcomes after therapeutic hypothermia (10, 11). Many of these tests incorporate feeding ability and history of seizures, which appear to be significant predictors of neurodevelopmental out comes following perinatal asphyxia (12, 13). The clin ical assessment immediately after the therapeutic hy pothermia treatment may result in false prediction due to the postponed removal of medication or impaired neuronal activity because of a briefly preceding hypox ic-ischemic incident and delaying clinical examination is more informative (14). Our study suggests that defer ring clinical evaluation until ten days after therapeutic hypothermia completion provides significant prognos tic information on neurodevelopmental outcomes. MRI — magnetic resonance imaging CI — confidence interval NPV — negative predictive value OR — odds ratio PPV — positive predictive value TH — therapeutic hypothermia; MRI — magnetic resonance imaging CI — confidence interval NPV — negative predictive value OR — odds ratio PPV — positive predictive value TH — therapeutic hypothermia; MRI — magnetic resonance imaging CI — confidence interval NPV — negative predictive value OR — odds ratio PPV — positive predictive value TH — therapeutic hypothermia; TH — therapeutic hypothermia; Acknowledgment None.l Conflict of Interests: The authors declare no conflicts of interest related to this article. Funding: None Funding: None RESULTS Predictive value of three clinical parameters (oral feeding ability, deviation in muscle tone, history of clinical seizures) for abnormal brain MRI in 81 neonates treated with TH Clinical parameter evaluated 10 days after TH Abnormal MRI scan Sensitivity (%) Specificity (%) PPV (%) NPV (%) P-value (OR,95%Cl) Inability of oral feeding (n = 45) 35 81 72 75 72 < 0.001 (8.3, 2.9 - 28.9) Deviation in muscle tone (hypotonia/hypertonia) (n = 24) 20 48 87 83 55 0.004 (3.9, 1.6 - 9.8) Hystory of clinical seizure (n = 53) 38 90 55 69 80 < 0.001 (11.9, 3 - 44.5) Abbrevations: TH, therapeutic hypothermia; PPV, positive predictive value; NPV, negative predictive value, CI, confidence interval; OR odds ratio 1. Predictive value of three clinical parameters (oral feeding ability, deviation in muscle tone, history of clinical seizures) for abnormal brain MRI in 81 neonates treated with TH Abbrevations: TH, therapeutic hypothermia; PPV, positive predictive value; NPV, negative predictive value, CI, confidence interval; OR odds ratio REDICTIVE VALUE OF EARLY CLINICAL PARAMETERS FOR ABNORMAL BRAIN MRI SCAN IN NEONATES TREATED... Hadžimuratović Emina,1 Hadžimuratović Admir,1 Pokrajac Danka,1 Selimović Amina,1 Muhasilović Senad2 1 Pediatric Clinic, University Medical Center of Sarajevo, Sarajevo, Bosnia and Herzegovina Sarajevo School of Science and Technology Program Coordinator, Sarajevo, Bosnia and Herzegovina Rezultati: MRI mozga je bio abnormalan kod 42 (51,85%) novorođenčadi sa sledećom distribucijom obrazaca oštećenja mozga: abnormalni signal u bazal nim jedrima kod 6, abnormalni signal u korteksu kod 16, abnormalni signal i u korteksu i bazalnim jedrima kod 20 novorođenčadi. Od tri analizirana klinička pa rametra, samo poteškoće pri hranjenju (P < 0,001, OR 8,3, 95% CI 2,9 - 28,9) i konvulzije (P < 0,001, OR 11,95, 95% CI 3 - 44,5) su bili značajno povezani sa abnormalnim MRI nalazom. Uvod: Nalaz MRI mozga ima prediktivnu vred nost za neurorazvojni ishod kod novorođenčadi leče nih terapijskom hipotermijom. Uobičajena klinička praksa je da se uradi MRI mozga pre otpusta, ali MRI mozga u dobi od oko 4 meseca života ima bolju pro gnostičku vrednost za dugoročni neurološki ishod kod novorođenčadi sa asfiksijom.i Cilj: Identifikovati koji od odabranih kliničkih parametara (sposobnost oralnog hranjenja, mišićni to nus, konvulzije) procenjeni 10 dana nakon terapijske hipotemije mogu predvideti primarni ishod abnormal nog MRI mozga. Zaključak: Novorođenčad koja su bila sposobna za potpuno oralno hranjenje do 10. dana života nakon terapijske hipotermije i nisu imala konvulzije, imala su malu verovatnost da će imati abnormalni nalaz MRI mozga. Ovo se može koristiti u selektivnom planiranju MRI mozga pre otpusta kod asfiksiranih novorođen čadi. Metode: Pregledali smo medicinsku dokumenta ciju novorođenčadi ≥ 36 navršenih sedmica gestacije koja su uzastopno lečena terapijskom hipotermijom i bila podvrgnuta MRI mozga. Klinički parametri 10 dana nakon terapijske hipotermije bili su u korelaciji sa nalazima MRI mozga učinjenim u prvih 7-14 dana života. Urađena je logička regresiona analiza korište njem sve tri kovarijante kliničkog statusa, sa abnor malnim MRI nalazom kao primarnim ishodom. Ključne reči: terapijska hipotermija, klinički pa rametri, poteškoće pri hranjenju, napadi, magnetna re zonanca mozga. REFERENCES athic newborns. Acta Paediatr. 2011; 100(10): 1344-9. doi: 10.1111/j.1651-2227.2011.02327.x. 1. Annink KV, de Vries LS, Groenendaal F, Vijlbrief DC, Weeke LC, Roehr CC et al. The development and validation of a cerebral ultrasound scoring system for infants with hypox ic-ischaemic encephalopathy. Pediatr Res. 2020; 87(Suppl 1): 59-66. doi: 10.1038/s41390-020-0782-0. 6. Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Par tridge JC, Allen F et al. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. AJNR Am J Neuroradiol. 1998; 19(1): 143-9. 6. Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Par tridge JC, Allen F et al. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. AJNR Am J Neuroradiol. 1998; 19(1): 143-9. 7. de Vries LS, Groenendaal F. Patterns of neonatal hypoxic-ischaemic brain injury. Neuroradiology. 2010; 52(6): 555-66. doi: 10.1007/s00234-010-0674-9. 2. Alderliesten T, de Vries LS, Staats L, van Haastert IC, Weeke L, Benders MJ, et al. MRI and spectroscopy in (near) term neonates with perinatal asphyxia and therapeutic hypo thermia. Arch Dis Child Fetal Neonatal Ed. 2017; 102(2): 147- 52. doi: 10.1136/archdischild-2016-310514. 8. Tann CJ, Nakakeeto M, Hagmann C, Webb EL, Nyom bi N, Namiiro F et al. Early cranial ultrasound findings among in fants with neonatal encephalopathy in Uganda: an observational study. Pediatr Res. 2016; 80(2): 190-6. doi: 10.1038/pr.2016.77. 8. Tann CJ, Nakakeeto M, Hagmann C, Webb EL, Nyom bi N, Namiiro F et al. Early cranial ultrasound findings among in fants with neonatal encephalopathy in Uganda: an observational study. Pediatr Res. 2016; 80(2): 190-6. doi: 10.1038/pr.2016.77. 3. Alderliesten T, Nikkels PG, Benders MJ, de Vries LS, Groenendaal F. Antemortem cranial MRI compared with post mortem histopathologic examination of the brain in term infants with neonatal encephalopathy following perinatal asphyxia. Arch Dis Child Fetal Neonatal Ed. 2013; 98(4): 304-9. doi: 10.1136/archdischild-2012-301768. 9. Bonifacio SL, de Vries LS, Groenendaal F. Impact of hypothermia on predictors of poor outcome: how do we decide to redirect care? Semin Fetal Neonatal Med. 2015; 20(2): 122- 7. doi: 10.1016/j.siny.2014.12.011. with neonatal encephalopathy following perinatal asphyxia. Arch Dis Child Fetal Neonatal Ed. 2013; 98(4): 304-9. doi: 10.1136/archdischild-2012-301768. 10. Hadžimuratović E, Skokić F, Hadžimuratović A, Hadžipasić-Nazdrajić A, Mujić M, Hadžimuratović A. Acute renal failure in term newborn following perinatal asphyxia. Sanamed. 2017; 12(1): 11-4. doi: 10.24125/sanamed.v1i1.162. 4. Licensing Licensing This work is licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) License. The brain MRI scans after therapeutic hypother mia assure that there is no other pathology (e.g. hem 14 Hadzimuratovic Emina, Hadzimuratovic Admir, Pokrajac Danka, Selimovic Amina, Muhasilovic Senad 8. Tann CJ, Nakakeeto M, Hagmann C, Webb EL, Nyom bi N, Namiiro F et al. Early cranial ultrasound findings among in fants with neonatal encephalopathy in Uganda: an observational study. Pediatr Res. 2016; 80(2): 190-6. doi: 10.1038/pr.2016.77. ZNAČAJ RANIH KLINIČKIH PARAMETARA U PREDIKCIJI MRI NALAZA NA MOZGU KOD NOVOROĐENČADI LEČENIH TERAPIJSKOM HIPOTERMIJOM Hadžimuratović Emina,1 Hadžimuratović Admir,1 Pokrajac Danka,1 Selimović Amina,1 Muhasilović Senad2 11. Liu W, Yang Q, Wei H, Dong W, Fan Y, Hua Z. Prog nostic value of clinical tests in neonates with hypoxic-ischemic encephalopathy treated with therapeutic hypothermia: a sys 7. de Vries LS, Groenendaal F. Patterns of neonatal hypoxic-ischaemic brain injury. Neuroradiology. 2010; 52(6): 555-66. doi: 10.1007/s00234-010-0674-9. REFERENCES Weeke LC, Groenendaal F, Mudigonda K, Blennow M, Lequin MH, Meiners LC, et al A novel magnetic resonance im aging score predicts neurodevelopmental outcome after perina tal asphyxia and therapeutic hypothermia. J Pediatr. 2018; 192: 33-40.e2. doi: 10.1016/j.jpeds.2017.09.043. 11. Liu W, Yang Q, Wei H, Dong W, Fan Y, Hua Z. Prog nostic value of clinical tests in neonates with hypoxic-ischemic encephalopathy treated with therapeutic hypothermia: a sys 11. Liu W, Yang Q, Wei H, Dong W, Fan Y, Hua Z. Prog nostic value of clinical tests in neonates with hypoxic-ischemic encephalopathy treated with therapeutic hypothermia: a sys 5. Elstad M, Whitelaw A, Thoresen M. Cerebral Resis tance Index is less predictive in hypothermic encephalop 5. Elstad M, Whitelaw A, Thoresen M. Cerebral Resis tance Index is less predictive in hypothermic encephalop 5. Elstad M, Whitelaw A, Thoresen M. Cerebral Resis tance Index is less predictive in hypothermic encephalop 13. Locci E, Bazzano G, Demontis R, Chighine A, Fanos V, d’Aloja E. Exploring perinatal asphyxia by metabolomics. Metabolites. 2020; 10(4): 141. doi: 10.3390/metabo10040141. 14. Thoresen M. Patient selection and prognostication with hypothermia treatment. Semin Fetal Neonatal Med. 2010; 15(5): 247-52. doi: 10.1016/j.siny.2010.05.008. A PREDICTIVE VALUE OF EARLY CLINICAL PARAMETERS FOR ABNORMAL BRAIN MRI SCAN IN NEONATES TREATED... 15 13. Locci E, Bazzano G, Demontis R, Chighine A, Fanos V, d’Aloja E. Exploring perinatal asphyxia by metabolomics. Metabolites. 2020; 10(4): 141. doi: 10.3390/metabo10040141. 14. Thoresen M. Patient selection and prognostication with hypothermia treatment. Semin Fetal Neonatal Med. 2010; 15(5): 247-52. doi: 10.1016/j.siny.2010.05.008. tematic review and meta-analysis. Front Neurol. 2020; 11: 133. doi: 10.3389/fneur.2020.00133. tematic review and meta-analysis. Front Neurol. 2020; 11: 133. doi: 10.3389/fneur.2020.00133. 13. Locci E, Bazzano G, Demontis R, Chighine A, Fanos V, d’Aloja E. Exploring perinatal asphyxia by metabolomics. Metabolites. 2020; 10(4): 141. doi: 10.3390/metabo10040141. 12. Hadders-Algra M. Early diagnostics and early inter vention in neurodevelopmental disorders-age-dependent chal lenges and opportunities. J Clin Med. 2021; 10(4): 861. doi: 10.3390/jcm10040861. 14. Thoresen M. Patient selection and prognostication with hypothermia treatment. Semin Fetal Neonatal Med. 2010; 15(5): 247-52. doi: 10.1016/j.siny.2010.05.008. Correspondence to/Autor za korespondenciju Correspondence to/Autor za korespondenciju Emina Hadžimuratović, MD, Ph.D., Assistant Professor Pediatric Clinic, University Medical Center of Sarajevo Patriotske lige 81, 71 000 Sarajevo, Bosna i Herzegovina Email: eminahadzimuratovic@yahoo.com , y j Patriotske lige 81, 71 000 Sarajevo, Bosna i Herzegovina Email: eminahadzimuratovic@yahoo.com How to cite this article. Hadzimuratovic E, Hadzimuratovic A, Pokrajac D, Selimovic A, Muhasilovic S. A predictive value of early clinical parameters for abnormal brain MRI scan in neonates treated with therapeutic hypothermia. Sanamed.2022;17(1): 11-15.
https://openalex.org/W2048762179	https://erem.ktu.lt/index.php/erem/article/download/7438/3804	English	null	Landfill mining: challenges and perspectives	Aplinkos tyrimai, inžinerija ir vadyba	2,014	cc-by	1,371	Landfill Mining: Challenges and Perspectives Prof. Dr. Gintaras Denafas Department of Environmental Technologies, Faculty of Chemical Technology, Kaunas University of Technology, Kaunas, Lithuania. E-mail: gintaras.denafas@ktu.lt http://dx.doi.org/10.5755/j01.erem.68.2.7438 Landfilling has been one of the most common methods to dispose waste in many regions worldwide. However, from the environmental and economic point of view such approach of waste management can be considered unsustainable since materials and energy that can be recycled and recovered are disposed and not used further. Landfills are well-known sources of contaminants released into the air, surface waters, groundwater and soil through long-term landfill gas emissions and leaching of hazardous substances. It is estimated that approximately 300 million tons of copper are currently disposed in landfills and other waste repositories (e.g. tailings and slag heaps) all over the world) corresponding to more than 30% of the remaining reserves of copper in known ores. Landfills contain significant amounts of combustibles and recyclable materials. It is estimated that there are between 150,000 to 500,000 old and still active landfills throughout the European Union (EU) representing the total volume of 30-50 billion m3 of waste. These facts confirm the need to identify the potential recovery of secondary materials and energy in both, short and long-term, perspectives through the implementation of landfill mining projects. Furthermore, landfill mining has received insufficiently attention till now. In Europe and Asia, the growing need for remediation of old landfills and removal of deposits obstructing urban development were important drivers for the increased interest in landfill mining since the end of 20th century. Despite of such keen interest, the research activities focused on landfill mining suddenly decreased in 2000, and since then, only sporadic initiatives have been reported in the scientific literature. Economic downturns, difficulties in extraction of high-quality, marketable recyclables from the deposits, and lower demand for landfill space in certain regions of the world due to introduction of more sophisticated waste treatment and recycling programs have been some of the responsible factors. During the period of last five years, performed research projects have shown that landfills contain large volume of valuable materials such as metals and plastics that constitute an important driving force for resource recovery. In contrast, remediation projects are always related to considerable societal costs and the implementation of integrated approach, while landfill mining involves both, profitable resource recovery and remediation, and, therefore, it offers more viable and efficient strategy than conduction of projects with remediation objectives solely. Aplinkos tyrimai, inžinerija ir vadyba, 2014. Nr. 2 (68), P. 3-4 ISSN 1392-1649 (print) Environmental Research, Engineering and Management, 2014. No. 2 (68), P. 3-4 ISSN 2029-2139 (online) http://erem.ktu.lt Editorial Aplinkos tyrimai, inžinerija ir vadyba, 2014. Nr. 2 (68), P. 3-4 ISSN 1392-1649 (print) Environmental Research, Engineering and Management, 2014. No. 2 (68), P. 3-4 ISSN 2029-2139 (online) http://erem.ktu.lt Landfill Mining: Challenges and Perspectives Also, the HELCOM recommendation 24/5 has clearly stated and redefined the needs for proper handling of waste and landfilling, and relevant measures to remediate and prevent future pollution with the consequent restoration of the ecological status of the Baltic Sea Region, including Lithuania. The prevention of contaminants in the Baltic Sea originated from landfills and dumpsites has also been emphasized. The research and international cooperation project “Closing the Life Cycle of Landfills - Landfill Mining in the Baltic Sea Region for Future” of Linnaeus University, Sweden was approved by Swedish Institute. The project is a joint venture of research groups and institutes of the participating countries (Sweden, Latvia, Norway, Italy, Denmark, Estonia, Lithuania and Ukraine). Partnership in this project have emphasized the importance of the cross-sector cooperation among the experts from universities, waste management associations, municipalities and companies working towards achieving the goals established by Swedish Environment Protection Agency, HELCOM and the EU Waste Framework Directive. The main objective of this project is to initiate new full-scale landfill mining projects in the Baltic Sea Region and compare the obtained results with landfill mining projects in different countries. The resource recovery as raw materials from excavated landfill waste as well as refuse-derived fuel (RDF) for energy recovery both are of the beneficial for generating revenues for the success of the project. The full-scale excavation and remediation of the Kudjape Landfill (Estonia) has been started in September, 2012, and it was finished in September, 2013. The manual sorting provides necessary information regarding the suitable techniques for material recovery from different fractions, however, due to the safety and efficiency issues in order to avoid any risk on human heath (workers) during waste separation by handpicking, automated sorting of waste using intelligent machinery (e.g., robots) were decided as future requirements for safe and effective LFM. Further research is needed towards the possible treatment of the fine fractions to remove the organic matter and to recover the valuable metals such as Cu, Al and Fe. Pilot studies have proven also the presence of other heavy and rare earth elements in the landfill mass. Last but not the least, cost-effective and environmentally-friendly washing techniques for cleaning the LFM waste and treatment of generated washing wastewaters should be the focus of future research. Landfill Mining: Challenges and Perspectives Such findings arouse the challenge of the current view on landfills as final disposal site of waste and indicate the emergence of shifting the old model of waste management into a landfill mining perspective in which the extraction of valuable material and energy resources is considered. There are still a number of challenges to reach the full potential of material recovery/recycling, energy utilization and nature restoration to be overcome mainly in terms of technological development. Gintaras Denafas On-going researches in many countries provide also comprehensive and systematic analysis to evaluate the suitability of closed landfill sites as power parks. The results of life cycle assessment show that both, solar and wind, turbine technologies are favorable with regard to energy consumption and greenhouse gas emissions. Synergies among these two technologies and landfill gas extraction in the closed landfill sites are studied. In principle the fourth technology, i.e., solid refuse fuel (SRF) production from excavated waste can be a part of such investigations. g One concept to be applied in Italy consists of cyclical process with especially designed steps like: Landfill cell preparation;  Landfill cell preparation;  Landfill cell preparation;  Waste disposal and realization of leachate recirculation, and biogas collection systems in the prepared cells; S i l l i i f ll i ll h b d ib d i  Waste disposal and realization of leachate recirculation, and biogas collection systems in the prepared cells;  Final stabilization of waste in each cell through air ventilation; l stabilization of waste in each cell through air ventila  Removal of temporary covering in order to allow the LandFill Mining (LFM) operations; p y g g ( ) p ;  Reuse of cells once completely empty, until the establishment of a cyclical process of landfilling, emptying and processing/recovery.  Reuse of cells once completely empty, until the establishment of a cyclical process of landfilling, emptying and processing/recovery. During the past decade, the EU Council Directive 1999/31/EC and the EU Council Directive 2008/98/EC, regarding landfilling have established that organic wastes and recyclables produced in the EU must be sent to incineration plants, composting plants and recycling units instead of disposing in landfills. Landfill Mining: Challenges and Perspectives Lithuania counted up to 843 of closed landfills of different sizes, where during their operation time about 3.4 million tons of waste have been landfilled (apart the old Kariotiškės landfill, where municipal waste from Lithuanian capital city Vilnius was formerly disposed). Regarding the EU support many of these landfills have been already remediated; the largest of which have installed landfill gas collection systems. Thus, a minimum for five-year period the mining activities in such landfills are in principle impossible; however, the above-described problems and opportunities do not disappear in this situation and will be important later. On the other hand, currently operated modern regional landfills will be fulfilled soon due to the inefficient separate waste collection and recycling. In this way, by implementation of mining projects for currently still used but almost fulfilled landfills the Lithuanian landscape would be improved and environmental impact of landfills through emitted landfill gas and generated leachate would be reduced. 4
https://openalex.org/W3133996838	https://tahiti.journal.fi/article/download/103186/60224	Finnish	null	Transsituationaaliset esineet ja asiat osallistavassa taiteessa	Tahiti	2,021	cc-by-sa	7,455	Taiteen esineistä ja asioista Taiteen esineistä ja asioista Lähtökohtanani tälle tekstille on ajatusleikki siitä, mitä tapahtuisi jos osallistavaa taidet ta tarkastelisikin ihmisosallistujien rinnal la myös esineiden näkökulmasta. Pohdin esineiden ja muiden ainesten toimijuuksia taiteessa kahdesta eri suunnasta. Yhtääl tä otan tarkasteluuni tanskalaisen taiteilija ryhmä Superflexin Free Shop -intervention, jonka tapahtumapaikkana on kuluttamisen – esineiden ja ihmisten kanssaolemisen – tyy pillisin näyttämö, kauppa.4 Ja toisaalta tar kastelen Tatu Gustafssonin soap wi-fi stilts -näyttelyä (Titanik, Turku, 2019), joka osal taan jatkoi Hueblerin ajatusta taiteesta, joka ei lisäisi uusia esineitä maailmaan. Transsituationaaliset esineet ja asiat osallistavassa taiteessa Aineeton ja esineetön taide2 voisi olla yksi keino lähestyä esineiden kumulatiivi sen kasvun kysymyksiä. Esimerkiksi osal listavaa taidetta eli taidetta, jonka ennakko ehtona on yleisön osallistuminen teoksen tuotantoprosesseihin, onkin usein koros tuneesti tarkasteltu aineettomana sosiaa listen sidosten taiteena.3 Tämä osallista vien taideteosten oletettu aineettomuus on kuitenkin tutkimuksellinen haaste. Osallis tavia prosesseja ei koskaan voi nähdä tai tuntea kokonaan. Ne ovat annetusti epä määräisiä ja monitulkintaisia: hetkellisiä, hajanaisia, jälkikäteisiä ja yhteismitatto mia. Niissä katsoja-osallistujan ja katso misen kohteen eli taideteoksen väliselle etäisyydelle perustuva representatiivinen suhde on korvautunut monenlaisten ja satunnaisten ainesten läsnäololla. Miten käsitellä tätä sekavaa ja pakenevaa mo ninaisuutta? Riikka Haapalainen Vuonna 1969 yhdysvaltalainen käsitetaiteilija Douglas Huebler kirjoitti taiteellisista lähtö kohdistaan tähän tapaan: ”Maailma on täynnä esineitä, enemmän tai vähemmän kiinnosta via; en halua lisätä niitä enempää.”1 Hueblerin ratkaisu esineiden kylläännyttämässä maail massa oli dematerialisoida taiteellinen työs kentely ja dokumentoida ihmislajia ja -eloa sanojen, karttojen, piirrosten sekä valokuvien kautta. Reilu viisikymmentä vuotta Hueblerin kannanoton jälkeen sen sisältöä on helppo lukea kulutus- ja ilmastokriittisesti. Kun maat ja meret täyttyvät tavarapaljoudesta, voi olla eettisesti vaikea perustella luonnonvaroja ku luttavia uusien – kiinnostavien tai vähemmän kiinnostavien – (taide-)esineiden välttämättö myyttä. Ilmastokriisin aikana taiteen kysymyk set eivät ole vain taiteen kysymyksiä. Siirryn ensin Free Shopiin Itäkeskuksen Ruohonjuuri-kauppaan ja vuoteen 2011 (Ku vat 1–3). Tilanne vaikuttaa jokapäiväiseltä ja arkiselta kaupassa käynniltä. Mutta silti kaikki ei ihan ole kuten tavallisesti. Teoksen 156 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T Kuva 1. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-ny kytaidefestivaali/Veikko Somerpuro. Kuva 3. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nyky taidefestivaali/Veikko Somerpuro. Kuva 2. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nykytaidefestivaali/Veikko Somerpuro. Kuva 3. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nyky taidefestivaali/Veikko Somerpuro. Kuva 1. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-ny kytaidefestivaali/Veikko Somerpuro. Kuva 2. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nykytaidefestivaali/Veikko Somerpuro. Kuva 1. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-ny kytaidefestivaali/Veikko Somerpuro. Kuva 1. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-ny kytaidefestivaali/Veikko Somerpuro. Kuva 1. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-ny ytaidefestivaali/Veikko Somerpuro. Kuva 2. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nykytaidefestivaali/Veikko Somerpuro. Kuva 2. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nykytaidefestivaali/Veikko Somerpuro. Kuva 3. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nyky taidefestivaali/Veikko Somerpuro. Kuva 2. Superflex, Free Shop (Ruohonjuuri, Itäkeskus, Helsinki), 2011. Kuva: IHME-nykytaidefestivaali/Veikko Somerpuro. 157 T Kuva 4. Tatu Gustafsson, Wi-fi from so meone to everyone, 2019. Soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 5. Tatu Gustafsson, Deodorant from someone who researches the future, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 6. bes from -näyttely, Selin. tafsson, Wi-fi from so e, 2019. Soap wi-fi stilts galleria. Kuva: Susanna Kuva 5. Tatu Gustafsson, Deodorant from someone who researches the future, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 6. Riikka Haapalainen Tatu Gustafsson, Fluorescent tu bes from the gallery, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 4. Tatu Gustafsson, Wi-fi from so meone to everyone, 2019. Soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 6. Tatu Gustafsson, Fluorescent tu bes from the gallery, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 5. Tatu Gustafsson, Deodorant from someone who researches the future, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 4. Tatu Gustafsson, Wi-fi from so meone to everyone, 2019. Soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 5. Tatu Gustafsson, Deodorant from someone who researches the future, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. Kuva 6. Tatu Gustafsson, Fluorescent tu bes from the gallery, 2019. soap wi-fi stilts -näyttely, Titanik-galleria. Kuva: Susanna Selin. 158 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T mottona on: ”Kaikki, mitä asiakas haluaa ostaa, on ilmaista. Free Shop voi tapahtua missä tahansa, milloin tahansa, kenelle ta hansa.” Tietyn ajan kaikki se, mitä asiakas tuo kaupan kassahihnalle maksettavaksi onkin ilmaista. Kuvassa 2 teostapahtumaan tietämättään joutunut osallistuja ei vielä ole täysin oivaltanut, ettei ostoksia vastaan täl lä kertaa vaaditakaan rahaa. Siksi huomio kuvassa kiinnittyy erityisesti teoksen piilo kameramaiseen asetelmallisuuteen, kassan ja asiakkaan ennakoivaan jännitteeseen ja sen yllätykselliseen purkuun. Onko ostos ten ilmaisluonteen paljastuminen myös se momentti, jossa teoksen ’taide’ tapahtui? Vai piileekö se kassan ja asiakkaan välises sä vuorovaikutuksessa? Mikä interventios sa paikantuu varsinaiseksi taiteeksi ja mikä on sen suhde erilaisiin jälkiin, tallenteisiin, kokemuksiin ja teksteihin, joita prosessista samanaikaisesti ja viiveellä väistämättä syn tyy? lua; sitä, kuinka Superflex halusi tutkia mah dollisuuksia vastatalouteen ja rahan vallan kyseenalaistamiseen.5 Lisäksi voi taustoittaa teoksen tuotantoprosessiin liittyneitä käy tännöllisempiä ennakkovalmisteluja, joita Free Shopin tapauksessa olivat esimerkik si neuvottelut interventioon osallistuneiden myymälöiden kanssa. Toiseksi tarkastelussa voi kiinnittyä teoksen varsinaiseen tapah tumiseen ja siinä tiivistyvään, arjen rutiinit katkaisevaan sosiaaliseen yhdessäoloon. Free Shopissa tämä merkitsisi kaikkea sitä, mitä kaupassa ihmisten välillä tapahtui in tervention kestäessä. Kolmas ajallinen taso on intervention jälkeiseen re-presentaatioon, uudelleen esittämiseen ja tulkintaan liittyvät sisällöt ja käytännöt, kuten esimerkiksi teok sen näytteilleasettamisen ratkaisut ja osallis tujien kertomukset. Free Shopin jälkituotan toon sisältyy myös teoksen dokumentaatio, joka toteutettiin tarkasti Superflexin ennalta määrittelemien ehtojen mukaisesti. yksittäisen vastaanottotilanteen ylittävää moninaisuutta.6 Transsituationaalisina teos ten vastaanotto ei koskaan voi olla kattava, kokonainen tai lopullinen: niiltä puuttuu kes kiö. Aina joku, jokin tai jotain jää teoksesta tavoittamatta. Riikka Haapalainen Siksi myös teoksen tapahtu mis- ja tarkastelukenttää sekä sen dualistisia lähtökohtia on hajautettava. Taiteen tarkas telu transsituationaalisena kutsuu avartavaa, moninaistavaa ja -arvoistavaa luentatapaa suhteessa teoksen aineellisuuteen, ajalli suuteen ja sosiaalisuuteen: moninlukemista ohi yleisen ja ihmiskeskeisen ymmärryksen taideteoksen luonteesta.7 Taidehistorian tutkimus oli pitkään objek tikeskeistä: sen tutkimuskohteita, esineitä, ei tarvinnut kyseenalaistaa. Taidehistorian esine oli singulaari; kestävä, kiinteä ja erityi syydellään kaikesta muusta rajautuva. Sa malla taidehistorian esineistö tuotti induktii vista havaintoa maailmasta: taiteen kautta voidaan löytää esteettisen ja historiallisen tarkastelun rinnalla laajempi ymmärrys yh teiskunnasta ja kulttuurista. Tämän erityi sen ja tunnistettavan objektin lähtökohta kuitenkin ylläpitää ja luo yleisyyttä ja yksiai kaisuutta; hegemonisia ja hierarkkisia ra kenteita taiteen kentälle. Free Shopin kaltaiselle osallistavalle teok selle voi pyrkiä luomaan muotoa tai raken netta ainakin kolmella ajallisella tarkaste lu- ja toimintatavalla. Ensinnäkin voi tuoda näkyville interventiota valmistelevan työn taustoittamalla Superflexin taiteellista ajatte Osallistavat taideteokset ovat aina samaan ja eri aikaan moniaalla, muualla ja tässä – joko prosesseina tai tilanteeseen liittyvinä tallenteina; erilaisina narratiiveina, muistoina ja laajenevina verkostoina. Siksi väitän, että teokset ovat annetusti transsituationaalisia, 159 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T Myös taide-esineiden ja ihmisten välinen suhde on ollut selkeän asetelmallinen ja ky seenalaistamaton: ihmiset (katsojat) ovat suhteessa taide-esineisiin vastakkaisissa tai projisoivissa positioissa: tekemässä tulkinto ja, synnyttämässä merkityksiä, nimeämässä ja luokittelemassa. Tätähän taidehistorialli seen tutkimukseen – tarkkaan katsomiseen – liittyvä taideteoksen kontemplaatio paradig maattisesti on.8 Taidehistorian objektit on en sin katseen kohdistamisella eriytetty muusta ympäristöstä, minkä jälkeen niille on kielen, tulkintojen ja teorioiden, kautta luotu uudel leen yhteys takaisin maailmaan tai vähin täänkin vallitseviin taiteen diskursseihin. Myös taide-esineiden ja ihmisten välinen suhde on ollut selkeän asetelmallinen ja ky seenalaistamaton: ihmiset (katsojat) ovat suhteessa taide-esineisiin vastakkaisissa tai projisoivissa positioissa: tekemässä tulkinto ja, synnyttämässä merkityksiä, nimeämässä ja luokittelemassa. Tätähän taidehistorialli seen tutkimukseen – tarkkaan katsomiseen – liittyvä taideteoksen kontemplaatio paradig maattisesti on.8 Taidehistorian objektit on en sin katseen kohdistamisella eriytetty muusta ympäristöstä, minkä jälkeen niille on kielen, tulkintojen ja teorioiden, kautta luotu uudel leen yhteys takaisin maailmaan tai vähin täänkin vallitseviin taiteen diskursseihin. tavoin teoksen ’sisään’, vaan he aina jäävät etäisemmiksi tarkkailijoiksi, vastakkaiseen suhteeseen sekä teokseen että taiteilijaan. yhteen ja ainakin hetkellisesti myös pitävät heitä yhdessä. Riikka Haapalainen Ja juuri tämän jo-siellä-olemisen takia osallistavan taiteen esineet eivät tutkimuk sellisesti ole olleet kiinnostavia.11 Free Sho pin kulutushyödykkeet ovat jääneet anonyy meinä, tunnistamattomina ja toisinnettavina teosluennan marginaaleihin. Tämä osoittaa taidekirjoittamisen ihmiskeskeisyyden, jota muun muassa posthumanistinen ja uusma terialistinen tutkimus on kritisoinut. Osallistavan taiteen oletettu immateriaalisuus Taiteen niin sanottua osallistavaa käännettä on 1990-luvulta alkaen luonnehdittu para digmaattisena siirtymänä esineistä ihmisiin, materiaalisesta immateriaaliseen.10 Kun tai deteos syntyy ja toteutuu osallistumisen pro sesseissa – sosiaalisessa vaihdossa ja vuo rovaikutuksessa – on houkuttelevaa jättää sen arkisen aineellinen puoli vähemmälle huomiolle. Esineet ja asiat ovat osallistavien prosessien sokea piste tai tiedostamaton, ne ovat epäkiinnostavia ja ei-erityisiä. Tämä onkin ymmärrettävää, sillä esimerkiksi Free Shopin kaltaisen teoksen esinemaailma on kovin satunnainen. Teoksen eri toteutusver sioiden valokuvatallenteissa näkyy muun muassa apteekkituotteita, päivittäistavaroi ta, äänilevyjä, juustoja ja sesonkikasviksia. Esineet ja asiat ovat jo kaupan hyllyillä ja vitriineissä niin kuin aina muulloinkin, ilman taiteilijan valintaa tai kontrollia. Ne lavasta vat ja kehystävät teostapahtumaa, kuljetta vat ihmisiä toistensa luokse, liittävät heidät – liittyvä taideteoksen kontemplaatio paradig maattisesti on.8 Taidehistorian objektit on en sin katseen kohdistamisella eriytetty muusta ympäristöstä, minkä jälkeen niille on kielen, tulkintojen ja teorioiden, kautta luotu uudel leen yhteys takaisin maailmaan tai vähin täänkin vallitseviin taiteen diskursseihin. Osallistaviin prosesseihin päätyvien esinei den materiaalinen toimijuus eroaa sattuman varaisuudessaan esimerkiksi readymadejen ehdottamasta esineisyydestä. Readymaden ajatus edellyttää taiteilija-tekijän kontrollia: kohottaessaan readymaden taiteeksi taiteili ja valitsee sille samalla myös tilan ja paikan. Free shopin kaltaisessa esinekoosteessa taiteilija on jo ennalta luopunut kontrollistaan. Esineet jo-siellä sallivat sekä sattumanvarai suuden että yllättävät ja ei-toivotut yhteydet. Modernin eriyttävä ja rajaava paradigma murtuu kaiken sallimisella ja hierarkioiden kumoutumisella. Taidehistorioitsija Claire Bishop on esi tellyt myös osallistavan taiteen keskeisenä toimijuuskategoriana katsojuuden (specta torship).9 Siten Bishop jatkaa sekä singu laarin yleisön ajatusta että moderniteetille tunnusomaista ajatusta katseen kautta hah mottuvasta ja haltuun otettavasta maailmas ta. Katsojuus pitää sisällään samanaikaisen ajatuksen sekä osallisuudesta että vääjää mättömästi ulkopuolisesta tarkkailijaposi tiosta. Tämä positio alleviivaa taiteilijan ja maallikon välistä erottelua: maallikoilla – kat sojilla – ei ole katsoessaan pääsyä taiteilijan Suhteessa moderniin taiteeseen osal listavalla taiteella ajatellaan olevan kaksi peruslähtökohtaa: yhtäältä osallistava tai 160 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T linjoilla. Hänkin korostaa taiteen sosiaalisten suhteiden, relaatioiden merkittävyyttä. Hä nelle taiteilija ei ole tekijä vaan pikemminkin mahdollistaja tai alullepanija. Siksi taiteessa ei korostuisi materiaalinen työ vaan teoksen synnyttämiin sosiaalisiin suhteisiin perustu vat esteettiset muodot.15 Huomio keskittyy ihmisten väliseen yhdessäoloon, sosiaalisiin yhteenliittymiin aineellisten sijaan. Osallistavan taiteen oletettu immateriaalisuus Samal la relationaalisen estetiikan vaalima ajatus yhdessäolosta menettää kriittisen potenti aalinsa ja pelkistyy tai pikemminkin yleistyy arvovapaaksi, porvarilliseksi urbaaniksi viih tymiseksi – vaikka juuri tämä kriittinen po tentiaali on ollut Bourriaud’n relationaalisen estetiikan ytimessä. Hänelle taiteen tärkeä lähtökohta on sen kyky häiritä arkea tuomal la kaupallistuneisiin, spektaakkelin läpäise miin ihmistenvälisiin suhteisiin uusia ”sosiaa lisia säröjä”, epätavanomaisuutta.16 Säröissä korostuvat vaihtoehtoiset tavat kommunikoi da ja synnyttää vaihtoa ihmisten välille. Silti Bourriaud’n ajatus sosiaalisista halkeamista tarkoittaa lähinnä korjaavaa toimintaa, ei uu distavaa tai täysin uutta tuottavaa. sa muihin ihmisiin ja asioihin: taideteokseen osallistuessaan jokainen osaltaan tuottaa ja muokkaa uudelleen teoksen merkityksiä ja sisältöjä – myös tahattomasti ja passiivise na. Siten toimijuus ja osakkuus ovat käsit teinä transsituationaalisia: ne ylittävät sen, mihin kulloinkin aina osallistutaan. Ne avaa vat suhteita yli varsinaisen taideteoksen ja ovat siten myös ajallisesti ja toiminnallisesti avoimempia ja enemmän kytköksissä muu hun. Samalla kun taideprosessi liittää yksilöt yhteen, aktivoituu myös heidän suhteensa elämään, yhteisöön ja olemiseen. de purkaa taiteen tekemisen ja vastaanoton kaksinapaisuutta ääripäinään aktiivinen tai teilija ja passiiviseksi oletettu vastaanotta ja-katsoja, ja toisaalta se korostaa taiteen poliittista ja sosiaalista kontekstia yli sen materiaalisen olomuodon. Kaikkialle levit täytyvä ja monenlaisen muodon saava taide siirtää katsojan teoksen vastaanottajasta tai osallistujasta tilanteeseen, jossa teos on ja tapahtuu. Se aktivoi katsojan suhteen maail maan. Osallistava taide – kuten arkinen elä mäkin – osallistaa kokonaisvaltaiseen, moni aistiseen tilanteeseen. Maria Lind on luonnehtinut osallistumista siirtymänä taiteilija–katsoja-erottelusta taitei lijan ja muiden väliseksi yhteistoiminnaksi: mahdollisuutena osallistua johonkin, jonka joku toinen on luonut tai käynnistänyt mut ta jonka sisältöihin voi itse silti vaikuttaa.14 Osallistavissa prosesseissa se, mitä taiteilija luo, ei ole enää taideteos tai taide-esine pe rinteisessä mielessä, sillä siltä puuttuu lopul linen muoto ja keskiö, hierarkkisesti jäsenty nyt ja tarkastelua ohjaava suunta. Teos on pikemminkin väline tai tilanne, jossa taiteilija on yhdessä yleisönsä kanssa jakamassa ja saamassa. Myös Nicolas Bourriaud on rela tionaalisen estetiikan ajattelussaan samoilla Taiteen osallistavasta käänteestä on kir joittanut muun muassa Suzana Milevska, jonka mukaan käänteeseen vaikutti 1990-lu vun filosofinen ja sosiologinen keskustelu, erityisesti poststrukturalismin esille tuomat ajatukset tekijän kuolemasta ja relationaa lisen sosiologian ymmärrys yhteiskunnan yhteiskunnallistumisesta dynaamisesti erilai sissa vuorovaikutussuhteissa.12 Tässä dyna miikassa moni asia määrittyy toisin. Esimer kiksi Irit Rogoffille taiteeseen osallistuminen näyttäytyy aktiivisena ja hierarkiattomana toi mijuutena (agency) ja osallisuutena tai osak kuutena (stakeholder).13 Osakkuuden käsite osoittaa sidoksen, joka kaikilla on suhtees Claire Bishopin positio osallistavan taiteen keskusteluissa on ollut taide-erityisyydestä kiinnipitävää. Osallistavan taiteen oletettu immateriaalisuus Hän on kritisoinut voimakkaas 161 Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T Mutta ehkä sosiaalisuutta, esteettisyyt tä ja aineellisuutta ei tulisikaan nähdä vas takkaisina arvoina. Ne ovat kaikki erilaisilla painotuksilla aina läsnä taiteessa ja sen vas taanottotilanteissa. Ja vielä enemmänkin, Bruno Latouria20 seuraten, on materiaalisen ja sosiaalisen erottelu mahdotonta, sillä ih misten keskinäiset suhteet ovat aina esinei den ja asioiden välittämiä. Osallistuminen voi olla vuorovaikutusta teoksen kanssa, osallistumista teoksen toteuttamiseen, mut ta myös muuntuvia sosiaalisia tai seuralli sia suhteita ihmisten, asioiden ja ilmiöiden välillä. Siksi niitä voi tarkastella ainoastaan prosessuaalisesti ja relationaalisesti, suh teissa ja kytköksissä aina johonkin toiseen. Ja siksi osallistava taidekin on perustavan laatuisesti materiaalista ja esinelähtöistä.21 Teosten materiaalisuus – teoksen mahdollis tavat arkipäiväiset ja satunnaiset esineet ja asiat – on vain erilaisia transsituationaalisia suhteita. Näitä suhteita luonnehtii jatkuva keskinäisten voimasuhteiden koettelu, joka esimerkiksi Free Shopissa näyttäytyy aina kin ihmisten ja tavaroiden välisen konsume ristisen suhteen neuvotteluna. ti relationaalisen estetiikan esiin nostamia taiteen sosiaalisia ja eettisiä päämääriä.17 Hänen mukaansa taiteen sosiaaliset sisällöt ovat ylikorostuneet esteettisen kustannuk sella. Osallistavassa taiteessa on keskityt ty liikaa ihmisten ja yhteisöjen sosiaaliseen muutokseen ja osallistamisen prosesseihin, jolloin taiteen erityisyys – teosten esteet tinen laatu – on unohdettu. Bishopille kes keistä osallistavassa taiteessa onkin koros taa teosta ’taiteena’ ja esteettisenä ilmiönä. Tämän vuoksi hän itse erottaa dualistisesti taiteen sosiaalisen ja esteettisen sisällön toi sistaan.18 tä välittävistä toimijuuksista ovat yksinker taisimmillaan ihmiskeskeisiä, samalla tapaa kuin on sosiaalisesta estetiikasta kirjoitta neen Michel Maffesolin näkemys kuvista ja objekteista. Maffesolin mukaan objektien kautta voidaan sulautua yhteisölliseen maa ilmaan (samalla tavoin kuin esimerkiksi ka tolisessa kirkossa pyhimyskuvaa palvottaes sa liitytään uskonnolliseen yhteisyyteen).22 Siten aineellinen todellisuus tulee määritty neeksi ja muotoutuneeksi ihmisen kautta – ei itsenäisesti, ennakoimattoman omalakisesti tai edes sekoittuen. Tätä ihmiskeskeisyyttä jatkavat myös osallistavasta taiteesta kirjoi tetut osallistumisen topologiat, jotka kartoit tavat teoksen ihmisosallistujien osallistumi sen tapoja ja rajoja.23 Bishop on tunnistanut osallistavan tai teen teoksista kaksi erilaista osallistumisen ulottuvuutta: yhtäältä osallistuminen voi tar koittaa katsojan osallistumista yksin tai seu rassa niin sanottuun interaktiiviseen taidete okseen ja installaatioon, ja toisaalta se voi olla kollektiivista osallistumista sosiaalisen elämyksen tuottamiseen.19 Ensimmäinen ulottuvuus tarkoittaisi katsojan fyysistä kiin nittymistä teokseen, joka on aineellisesti ra jattavissa. Jälkimmäinen puolestaan viittaisi niihin taiteen osallistaviin aineettomiin ja yh teisöllisiin elämyksiin, joihin ei yhtä suoraan voi fyysisesti kiinnittyä. Osallistavan taiteen oletettu immateriaalisuus Osallistavan taiteen erityisyys sosiaalis ten suhteiden esille tuojana ei siten tarjoa radikaalisti uutta varhaisempaan taidekä sitykseen; se ainoastaan ylläpitää taiteen ihmiskeskeisyyttä. Onhan taiteen vastaan otto ollut aina annetusti sosiaalista ja katso jaansa osallistavaa, eikä tähänkään taiteen osallistava käänne ole tuonut muutosta. Se on ainoastaan laajentanut ja moninaistanut taiteen käsitettä. Siksi osallistava taide ei it sessään tuo vaihtoehtoisia ja kestävämpiä Sosiologi Latourin esittämät ajatukset ai neellisista yhteisöistä ja esineiden yhteisyyt 162 alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T toiminnoista tai interventioista.26 Mitä täl lainen esineiden ja asioiden materiaalinen toimijuus, ihmisestä riippumaton oleminen voisi tarkoittaa taiteessa? Esimerkiksi Free Shopissa keskeisenä teostapahtuman mää rittäjänä on nollasummainen ostoskuitti (Kuva 3). Sen kädestä käteen siirtyminen merkitsee teokseen käsikirjoitettua toiminnallista huip pukohtaa: hetkeä, jolloin tuotteiden ilmaisuus paljastuu ja asiakas saa tuta mahdottoman mahdolliseksi tulemisen, rahaan perustuvan markkinatalouden hetkellisen ja onnekkaan kumoutumisen. Kassakuitti paljastaa inter vention taideluonteen. toimintatapoja maailmaan, joka jo on kyl lääntynyt tavaroista ja taide-esineistä. teen julkisiin esittämiskäytäntöihin liittyvästä uutuuden vaateesta, minkä voi nähdä ekolo gisesti kestämättömäksi. Ihmisten, esineiden ja asioiden kanssatoimijuus Nimensä soap wi-fi stilts mukaisesti näyt telytilassa oli esillä muun muassa saip puapala, langattoman verkon mokkula ja puujalat. Silti näyttelyn esineet eivät olleet klassisia duchampilaisia teollisesti valmis tettuja readymadeja, jotka taiteilijan tietoi nen valinta oli kohottanut taideteoksiksi. Tässä muodonmuutoksessa teosesineen materiaalisuuden on nähty heikkenevän tai ohenevan laadultaan käsitteellisemmäksi, aineettomammaksi. Soap wi-fi stilts -näyt telyn materiaalinen esineistö oli arkisesta toiminta- ja käyttöympäristöstään irroitettua ja uudelleen sijoitettua lainatavaraa. Siksi esineiden taidestatus oli läpi näyttelyn häi lyvä. Gustafsson oli pyytänyt esinelainoja itselleen kiinnostavilta ihmisiltä tai instituu tioilta, taikurista ja tulevaisuudentutkijasta museoon ja automarkettiin. Valikoimaan sisältyi myös Gustafssonin henkilökohtai sia tavaroita. Tällä tavoin Gustafsson pyrki kiertämään taiteilijuudelle asetetun odotus arvon uutuudesta ja uusien teosten valmis tamisesta: luomaan näyttelyn, joka ei edel lyttäisi mitään uutta. Taiteen osallistava käänne on merkinnyt vahvojen, singulaarien (taide-)esineiden puuttuessa käännöstä ihmisten ja yhteisö jen välisiin suhteisiin. Tässä käännöksessä esineet ja asiat hukkuivat ihmistenvälisen osallistumisen sosiaalisen kohinan alle. Esi neiden ja asioiden maailma oli liian jokapäi väistä, arkista ja vailla suoraa representatii vista suhdetta johonkin toiseen. Entä muuttuisiko Free Shopin esineis tön toimijuus, jos ne siirrettäisiin pois kau pan hyllyiltä tunnuskuvallisemmalle taiteen areenalle, näyttelykontekstiin? Miten niiden toimijuuteen ja materiaalisuuteen vaikut taisi paikka, jossa taiteen ja taide-esineen vastaanottotapaa ohjataan koodatummin kontemplatiiviseen ja käsitteellisempään tar kasteluun? Yhden alustan tälle pohdinnalle tarjoaa esimerkiksi Tatu Gustafssonin ke sällä 2019 Turun Titanik-galleriaan kutsuma soap wi-fi stilts -näyttely.27 Näyttely esitti ky symyksen siitä, tarvitseeko taidenäyttelyyn tuottaa uusia tavaroita. Kysymys nousi tai Kuten todettua, Latourille toimijuus on ih misten lisäksi myös esineille kuuluva omi naisuus: esineet määrittelevät ja säätelevät ihmisten välistä toimintaa.24 Muiden muassa filosofi Graham Harman on korostanut esi neiden materiaalista voimaa Latouriakin radi kaalimmin. Harmanin mukaan esineitä tulisi tarkastella itsenäisinä, ihmisten vuorovai kutuksesta tai projektioista riippumattomi na toimijoina.25 Myös uusmaterialisti Jane Bennett, jonka ajatukset materiaalisesta todellisuudesta ovat Harmanille vastak kaiset, on korostanut esineiden ja asioi den olevaisuutta riippumattomana ihmisen 163 Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T Gustafssonin näyttelyssä vierailleet esi neet osoittivat myös sen, kuinka valmiiksi (ja ahtaaksi) kooditettua näyttelykonteks ti on. Taidenäyttely rakenteena ja asioiden esittämisen muotona on niin vakiintunut ja pysyvä, että taiteen vastaanoton kaltainen suuntautuminen voidaan synnyttää käytän nössä mille tahansa ilmiölle. Soap wi-fi stilts -näyttelyn esineistö ei kuitenkaan asettunut näyttelykoodeihin hangoittelematta vas taan. Ihmisten, esineiden ja asioiden kanssatoimijuus Ne koettelivat ainakin kolmella tasolla gallerianäyttelyä taiteen esille saattamisen muotona: ne synnyttivät kitkaa suhteessa luonnollistuneisiin ajatuksiin tekijyyydestä (omistajuudesta), rahasta ja ajallisuudesta. Ensinnäkin Gustafssonin tekijyys oli näytte lyssä häivytetty verrattain vähäiseksi. Hän ei ollut valmistanut tai edes valinnut esillä olevia teosesineitä. Taiteilijan läsnäoloon teosprosessissa ei viitannut myöskään gal leriassa ollut teosluettelo: Gustafsson kuten muutkin näyttelyn ihmistoimijat eli esineiden lainaajat olivat luettelossa nimettöminä. Esineet eivät olleet galleriassa myynnissä, jolloin niiden arvo syntyi irrallaan kaupallisen galleriatilan yhdestä määrittävästä voimas ta, rahasta. Raha on suhde, joka yhteismi tallistaa – tekee vertailukelpoiseksi – lähes kaiken. Silti sekä Free Shopissa että soap wi-fi stilts -näyttelyssä esineiden ostaminen ja omistaminen onnistuu ainoastaan kulu tushyödykkeinä, ei taideteoksina. Molem pien esineistöstä puuttuu taiteilijan tietoi nen ja päämääräsuuntautunut esteettinen valinta. Tämä eroaa lähtökohtaisesti siitä, kun Marcel Duchamp 1914 kävi ostamassa pariisilaisen tavaratalon kotitalousosastolta pullonkuivaustelineen. Sillä ostotapahtuman jälkeen Duchamp nimesi ja kohotti teline-rea dymadensa taiteeksi. Tämän nimeämisen jälkeen pullonkuivaustelineellä ei ollut enää pääsyä takaisin ”vain” kotitaloustuotteeksi. Hierarkkinen ajatus esineen kohottamisesta taide-esineeksi tarkoittaa esineen materiaa lisuuden muutosta: siitä irrotetaan sen alku peräinen aineellinen yhteys työhön ja joka päivään, jolloin jäljelle jää sen käsitteellinen puoli, näennäisesti immateriaalinen ajatus katsojan mielessä. Readymaden tarkastelu immateriaalisena käsitteenä irrottaa esineen sen alkuperäisestä tuotanto- ja toimintaket esine oli galleriassa vain sen ajan minkä teok sen omistaja halusi. Laina-ajan jälkeen esi ne palautui omistajalleen, todennäköisesti jatkamaan toimijuuttaan alkuperäisessä ar kisessa käyttötarkoituksessaan, mutta kan taen itsessään uusia merkityskerrostumia. Näyttelyn ajallinen liikkuvuus ja epäsään nöllisyys siirsi teoksia osin samankaltaiseen näyttelyvieraan rooliin kuin missä näyttelys sä vierailleet ihmisetkin olivat. He ja ne olivat läsnä kukin tavallaan aineellisena ja mate riaalisena vieraana, jonka läsnäolo osaltaan vaikutti, muutti ja mukautti näyttelyä. Esineet näyttäytyivät siten prosessuaalisessa tilas sa ilman representatiivisia ja etäännyttäviä yhteyksiä johonkin ei-läsnäolevaan toiseen. Tämä ajallinen liikkuvuus tarjosi mahdolli suuden tuoda esiin esineiden merkitysten jatkuva muutos: ne eivät ole koskaan ai noastaan jotain ja yhtä – joko taideteoksia tai arkiesineitä – vaan sekä–että koko ajan. Tämä sekoittunut moninaisuus – transsitua tionaalisuus – merkitsee sitä, että pysyvien määritelmien sijaan teos kirjoittuu aina ja joka hetki uudelleen. Uusien taide-esineiden vastustuksen rin nalla Gustafssonin esille saattama esineistö vastusti myös perinteiseen näyttelyrakentee seen liittyvää ajallista kestoa. Näyttelyn teos- tai esine-elementtien ajallisuus vaihteli: kukin Gustafssonin näyttelyssä esineet esitettiin yhteydessä omistajiinsa, mikä ylläpiti niiden materiaalisia, historiallisia ja arkisia juuria. Ihmisten, esineiden ja asioiden kanssatoimijuus 164 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa taidekaupalle). Nykyaikaiselle esinesuhteel le on sosiologi Turo-Kimmo Lehtosen mu kaan tyypillistä, että kaikkiin asioihin voidaan suhtautua ja myös suhtaudutaan kauppata varoina.28 Tämä suhtautumistapa on läpäis syt myös taiteen, kuten John Berger totesi jo 1970-luvulla.29 Jos taide on lähtökohtaisesti ostettavaksi tarkoitettu kulutushyödyke, ku ten Lehtonen ja Berger väittävät, ovat taiteen esittämis- ja tapahtumispaikatkin annetusti ei-neutraaleja kaupallisia tiloja. Kuluttamisen paradigmaattisuus ja rahan määrittämien suhteiden kaikkiallisuus tulee sekä Free Shopissa että soap wi-fi stilts -näyttelyssä käänteisesti esille. Ne ovat kulutustavarois ta koostuvia teoksia, joita ei silti voi teoksina ostaa eikä myydä. Niiden toiminnallisuus on ainakin hetkelliseksi irrottautunut rahan pyö rittämästä vaihtotaloudesta – ne ovat onnis tuneet sata vuotta varhaisemman dada-liik keen päämäärässä tuhota ajatus taiteesta kulutushyödykkeenä tai luksustavarana. justa ja mahdollistaa teoksen tarkastelun näennäisen neutraalina ja historiattomana – samaan tapaan on kritisoitu Bourriaud’n relationaalista estetiikkaa siitä, että siinä arki näyttäytyy yhtenä ja liian yleisenä. Tämä yleisyys mahdollistaa etäisyydenoton ja re presentatiivisen suhteen aineelliseen todelli suuteen, ja samalla se riisuu teokselta poliit tisen ja aktivistisen voiman. oikeuskäytäntöihin kytkeytynyttä talouspoli tiikkaa.30 Rowland väittää, että orjuus ei ole koskaan poistunut Yhdysvalloista, se on vain muuttunut epäsuoremmaksi. Yhdeksi esimer kiksi epäsuorasta orjuudesta Rowland antaa vankiloihin suljettujen ihmisten pakkotyövoi man, joka on julkisen hallinnon toimivuuden kannalta rakenteellisesti välttämätöntä: eri laiset kätketyt hyöty- ja intressiketjut edellyt tävät valta- ja alistussuhteiden pysyvyyttä. Rowlandin näyttelykokonaisuudessa oli esil lä sellaisia arkisia esineitä kuten oikeussa lien yleisöpenkkejä, viemärikaivojen tukiren kaita tai poliisien toimistopöytiä, jotka kaikki oli valmistettu yhdysvaltalaisten vankiloihin suljettujen ihmisten palkattomana työvelvoit teena. Rowland esittää historiallisten doku menttien avulla tämän pakkotyövelvoitteen – jolla on leimallisesti rasistisia juuria – jatku mona orjatyölle. justa ja mahdollistaa teoksen tarkastelun näennäisen neutraalina ja historiattomana – samaan tapaan on kritisoitu Bourriaud’n relationaalista estetiikkaa siitä, että siinä arki näyttäytyy yhtenä ja liian yleisenä. Tämä yleisyys mahdollistaa etäisyydenoton ja re presentatiivisen suhteen aineelliseen todelli suuteen, ja samalla se riisuu teokselta poliit tisen ja aktivistisen voiman. Free Shopin ja Gustafssonin näyttelyn esi neet eivät kuitenkaan ole määritelmällisesti readymadeja, sillä niitä ei uudelleennime tä tai uudelleenasemoida joksikin muuksi. Läpi teos- ja näyttelyprosessin esineet ja asiat toimivat myös kuten ennenkin. Free Shopissa ostokset ja kuitti liikkuvat sujuvasti taideobjektin ja kulutushyödykkeen katego rioiden välissä; suolasaippua Gustafssonin taidenäyttelyssä on samaan aikaan jalustal le asetettu näyttelyesine ja pianistin kosme tiikkatuote. Ihmisten, esineiden ja asioiden kanssatoimijuus Ne eivät ole joko–tai vaan koko ajan sekä–että. Koska esineet ovat julkishallintoa varten ja julkishallinnon tilauksesta valmistettuja, ne eivät koskaan tule myyntiin. Rahalla niihin ei pääse käsiksi. Voidakseen esitellä esineis töä ja mahdollistaakseen, ettei niistä kukaan enää edes epäsuorasti hyödy, on Rowland perustanut yleishyödyllisen säätiön, joka on esineet ensin normaalien protokollien mu Vielä tarkemmin taiteen kauppatavarasuh teen purkamisen on toteuttanut esimerkiksi Cameron Rowland, jonka Indirect Benefit -te oskokonaisuus oli 2016 esillä Fri Art -taide museossa Sveitsissä. Indirect Benefit esitteli nyky-yhdysvaltalaista orjuuteen ja rasistisiin Osallistava taide onkin koettu taiteentutki mukselle ongelmallisena myös sen vuoksi, että se ei synnytä tai luo esineitä – siitä ei jää juuri mitään aineellista, jota voisi ostaa ja omistaa (ovathan taidehistoria ja taiteen ken tän rakenteet pohjanneet vahvassa mielessä 165 alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T kaan hankkinut ja sen jälkeen näyttelyyn lai nannut. Enää niitä ei voi ostaa eikä niitä voi siirtää alkuperäisiin käyttötarkoituksiinsa. Ne ovat erkaantuneet rahan ja hyödyn logiikas ta todistamaan juuri tätä samaa logiikkaa, muistuttamaan yhteiskunnan rakenteellises ta ja rasistisesta sorrosta. Esimerkiksi Free Shopin kaltaiset interve ntiot toimivat osallistujiinsa nähden ainakin kahdella ajallisuuden tasolla: välittömyyden ja viiveen kautta.31 Välittömyys syntyy osal listavan intervention suorasta luonteesta. Teos toteutuu samalla kun siihen osallistu taan. Tämän välittömyyden ajatellaan usein olevan esimerkillinen tai oikea tapa osallis tua teokseen. Toisaalta teos tapahtuu myös viiveellä32: muut kuin teokseen välittömästi osallistuneet pääsevät siihen jälkikäteen osallisiksi teosprosessista syntyneiden do kumenttien, tallenteiden, suullisten tiedonan tojen sekä muiden epäsuorien aineellisten materiaalien välityksellä. välille, mikä ylläpitää etuoikeuksia ja portin vartijuutta suhteessa teoksen synnyttämään tietoon ja kokemukseen. Tämän rakenteelli sen portinvartijuuden transsituationaalisuu den ajatus kumoaa. Teosten transsituationaalisuus merkitsee taiteentutkimuksessa myös dualistisen ma teriaalinen–immateriaalinen-asetelmallisuu den purkamista. Transsituationaalisuutta leimaa moninaisten toimijuuksien epämää räisyys, epälineaarisuus ja tunnistamatto muus. Se myös aktivoi teoksen katsomis- ja vastaanottokokemuksen kriittisenä harjoit teena, jossa ei ole yksiselitteisesti oikeaa positiota suhteessa teokseen.34 Siksi trans situationaalisuuden ajatus, taideteoksen ym märtäminen liikkuvina suhteina korostaa tai teesta kirjoittamisen poliittisuutta: siirtymää läsnäoloon ja nykyhetkestä käsin syntyvän teosluennan eettisyyteen ja vastuullisuuteen. Tällöin tutkimuksellista mielenkiintoa ja kat setta teokseen ja sen prosessiin ei ainoas taan hajauteta vaan palautetaan teoksesta näennäisesti puuttuvien keskiöiden väliin ja liitoksiin, esineisiin ja asioihin; niihin mo ninaisiin aineellisiin toimijoihin, jotka tuovat moneuden yhteen. Teokseen on aina aktiivi sesti rakennettava uusi suhde nykyhetkestä Taideteoksen transsituationaaliset suhteet Kuten todettua, taiteen toimijuuksia voi tar kastella ihmisosallistujien ohella myös esi neiden näkökulmasta. Teoksissa paljastuvat esinemaailmat ja aineelliset jäljet hajautta vat tarkastelun yli paikkojen ja tilanteiden, transsituationaalisiin suhteisiin. Transsitua tionaalisessa tarkastelussa teoksella ei ole yksiselitteisesti määrittyvää päättymispistet tä, päämäärää tai tavoitetta. Teos luo ym pärilleen erilaisia ja osin sattumanvaraisia osallisuus- ja muistikertoimia. Ymmärrän nämä transsituationaaliset suhteet hie rarkkisesti litteinä. Kenelläkään tai millään ei koskaan ole täyttä tai ensisijaisempaa pääsyä taideteoksen ehdottamiin maail moihin. Kaikki teoksessa ja siihen sekoit tuvissa aineksissa on nykyhetkestä käsin yhtäläisen tärkeää. Tämä viive tuo teoksen tarkasteluun epä varmuuksia. Esimerkiksi Martha Buskirk toteaa, kuinka teokseen viiveellä osallistu vien pitää ”luottaa” taiteilijan kertomukseen teoksesta.33 Hän siis painottaa luottamusta siihen, että taiteilija kykenisi tavoittamaan teoksesta jonkin kiinteän teoshahmon, jo hon kokemus voitaisiin viiveelläkin palauttaa – vaikka tuo kokemus onkin välittyneempi ja siten sekä laadullisesti heikompi että ko kemuksellisesti ohuempi. Buskirk rakentaa näin hierarkkisen suhteen teoksen välittö män ja viiveellä tapahtuvan vastaanoton Tällöin tutkimuksellista mielenkiintoa ja kat setta teokseen ja sen prosessiin ei ainoas taan hajauteta vaan palautetaan teoksesta näennäisesti puuttuvien keskiöiden väliin ja liitoksiin, esineisiin ja asioihin; niihin mo ninaisiin aineellisiin toimijoihin, jotka tuovat moneuden yhteen. Teokseen on aina aktiivi sesti rakennettava uusi suhde nykyhetkestä 166 alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T esimerkiksi Black Lives Matter -liikkeen yh teydessä, tulee teosta tarkastellessa kysyä ja uudelleenkysyä, minkälaisiin arvoihin ja poliittisiin ja yhteiskunnallisiin sisältöihin teos ja teoksen tarkastelijan toimijuus paikantu vat. Taide, kuten elämäkin, on aina kaksija koista, muttei perustuen joko–tai-ajatteluun vaan ja-sanan aktiiviseen käyttöön. Taide on sekä–että. Samanaikaisuuksina, muutoksi na, liikkeinä ja vaihteluina hetkellisyyden ja ikuisuuden, kiinteän ja häilyvän, yksilöllisen ja yhteisöllisen seassa. Ei yksiä ilman toisia, eikä toisia ilman kolmansia; ei teosta ilman määrittäjiä, eikä määrittäjiä ilman erilaisia transsituationaalisia konteksteja. käsin: se on kirjoitettava uudelleen. Trans situationaalisuus ja siihen sisältyvä kirjoitta misen poliittisuus avaavat mahdollisuuden kritisoida ja purkaa toimimattomia taiteentut kimisen protokollia. Siten tieto ei ole kumula tiivista ja karttuvaa – tai edes itsekorjaavaa. Transsituationaalisuus harjoitteena tuottaa (taide-)esineille kerronnallisen korpuksen, joka ei ole lineaarista, kronologista eikä yk sinomaan ihmis- tai instituutiokeskeistä. taiteen transsituationaalisissa tarkasteluis sa. Luonnehtiihan nykytaidetta usein toimi juuksien epämääräisyys, epälineaarisuus ja tunnistamattomuus sekä teoksen aineellis ten jälkien ja suhteiden kanssatoimijuus. Taiteen binääriset rakenteet ilmaisevat keinotekoista universalismia, yleisyyttä joka taiteen käytännössä johtaa helposti normatii visuuteen ja totalisoivaan diskurssiin. Taideteoksen transsituationaaliset suhteet Näen näisen neutraali yleinen diskurssi ylläpitää juuri niitä hegemonisia rakenteita ja alistavia tai sortavia käytäntöjä, joita taiteen tulisi jat kuvasti kritisoida ja haastaa. Se estää uuden ja moniarvoisemman tiedon esilletulon. Insti tutionaalinen taidehistoria on toistuvasti otta nut tehtäväkseen luoda pysyvyyttä ja arvoa taiteelle ja teoksille – keinotekoisen illuusion taiteen ikuisuudesta ja yleisyydestä. Tämä taiteen ylläpitävä ja rajaava funktio samalla ylläpitää taiteeseen – ja siten koko yhteiskun taan – sisäänrakennettuja väkivaltaisia ja toi seuttavia valtarakenteita. Teoksia tarkastel laan eheinä tarinoina irti koko kontekstistaan ja materiaalisista ympäristöistään, joihin an netusti sisältyy häiriöitä, epätäsmällisyyksiä ja katseelle kitkaa tuottavia sisältöjä. Siinä missä orjakauppiaiden patsaita kaadetaan ja taiteen kaanoneita uudelleenarvioidaan Binäärisyyden purku monin- ja uudelleenkirjoittaen Modernistinen taide perustui selkeään kak sinapaisuuteen. Taideteoksen vahva ja yk siääninen läsnäolo edellytti etäisyyttä teok sen ja sen tarkastelijan välille sekä erillisyyttä muusta maailmasta,35 mikä voimakkaasti eriarvoistaa ja ylläpitää taiteen vastaanotto tilanteen valtasuhteita. Tämä ajatus taiteen ja kaiken muun erillisyydestä on kiinteä ja fundamentalistinen puhdasoppisuuden ra kenne, jota tulee moniarvoisessa yhteiskun nassa aktiivisesti kritisoida. Kiinteät jaottelut esimerkiksi taiteilijan ja katsojan, subjektin ja objektin, taideteoksen ja arkiesineen, taiteen ja ei-taiteen sekä materiaalisen ja immate riaalisen välillä kyseenalaistuvat viimeistään Osallistavan taiteen aineellinen moninai suus kutsuu ja tallentaa itseensä vieraan varaisesti kaiken, niin kutsutut kuin kutsu mattomatkin. Niitä voi ajatella esimerkkeinä feministifyysikko Karen Baradin36 toimijuus realismista: ne ovat erilaisten ja yhteismitat tomien asioiden, esineiden ja ihmisten muo dostamaa sekoittunutta toimijuutta. Tämä yhteistoimijuus pohjaa horjuviin ja huojuviin suhteisiin, joissa kieli ja normatiiviset kate goriat eivät tarjoa varmuutta ja turvaa. Se itsessään on uutta, huokoista ja poeettista ilmausta. Erilaisuuden salliessaan teokset 167 alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T asettuvat alttiiksi tiedoille ja prosesseille, joi den sisällöistä ja suunnista ei ole varmuutta ja jotka hyväksyvät sotkuisuuden ja tilan nesidonnaisuuden positivistisen selkeyden laatimisen sijaan. Taideteoksen määritelmä ei ole sama ja pysyvä vaan määritelmät liik kuvat ja vaihtuvat eri tilanteissa ja erilaisissa suhteissa. Se mahdollistaa taiteen kehitys kertomuksen sijaan ei-loogista ja ei-lineaa rista, sotkua ja epäyhteneväisyyttä; lumou tumisen ja magian sallimaa ruumiillisuutta, kielen ja sanojen ilmaisukyvyn koko kapa siteetin käyttöönottamaa moninaisuutta. Se hyväksyy myös ei-tietämisen ja ei-tuntemi sen: sen, että teokseen aina jää asioita, joita ei voi tietää eikä tuntea kokonaan. leista, pysyvyydestä tai paradigmoista. Särö luo ja rakentaa uusia topologioita ja monin kirjoittamisen tapoja.38 nä soap wi-fi stilts -näyttelyn taide- ja arkie sineistön keskinäisessä vaihdettavuudessa ilman uutuuden vaatimusta. Kolonialismin ja orjakaupan sietämättömän väkivaltaiset kuvat tulevat traagisesti esiin esimerkiksi Indirect Benefit -näyttelyn viimeistellyn tyy likkäissä ja funktionaalisissa toimistohuone kaluissa. Transsituationaalisuuden tunnistaminen ei muuta tapaa tarkastella taidetta. Se vain muuttaa ja osoittaa sen eettisen välttämät tömyyden, että kaikki tulee arvioida aina uudestaan nykyhetkestä käsin. Vain silloin teokset tulevat liitetyiksi myös laajempiin eet tisiin yhteyksiinsä ja taidehistorian tieto säi lyy elinvoimaisena. Teosten materiaalisuus luo pohjan, jolla ne kiinnittyvät toiminnalli sesti poliittisiin, taloudellisiin ja sosiaalisiin kysymyksiin. Tällöin (taiteen) historioitsijan tehtäväksi tulee tuoda takaisin tietoisuuteen yhteiskunnassa jo unohdetut mutta edel leen vallitsevat asiat – jotta voidaan mennä eteenpäin. Lopuksi Taide usein mielletään harvinaisten, singu laarien, yksittäisten ja autenttisten esineiden kokoelmaksi. Tällaista ilmiselvää erityisyyttä ei Free Shopin tai soap wi-fi stilts -näyttelyn esineillä ole. Ne ovat kiistatta jokapäiväisiä ja yleisiä. Monesti osallistamiseen painottunei den teosten materiaalisuus onkin banaalia, satunnaista eikä millään tavalla huomionar voista tai edes esteettistä: ne olivat jo-siellä. Ne edelsivät interventiota ja jäivät sinne in tervention jälkeen. Siitäkin huolimatta tällai silla mahdollisesti arkipäiväisillä esineillä on keskeinen rooli ihmisten yhteen tuomisessa ja kommunikaation luojina: ne toimivat ih mistenvälisen yhdessäolemisen aineellisina välittäjinä. Silti esineiden korostaminen ei merkitse paluuta taiteen formaaleihin esteet tisiin tai historiallisiin sisältöihin vaan ennen Taide usein mielletään harvinaisten, singu laarien, yksittäisten ja autenttisten esineiden kokoelmaksi. Tällaista ilmiselvää erityisyyttä ei Free Shopin tai soap wi-fi stilts -näyttelyn esineillä ole. Ne ovat kiistatta jokapäiväisiä ja yleisiä. Monesti osallistamiseen painottunei den teosten materiaalisuus onkin banaalia, satunnaista eikä millään tavalla huomionar voista tai edes esteettistä: ne olivat jo-siellä. Kun tunnustaa teosten ja toimijuuk sien ambivalenssin, avautuu mahdollisuus nyt-hetkestä ja siihen sisältyvistä kiireelli sistä asioista nousevaan uuteen tietoon, säröihin tai deleuzelaisittain ajateltuna paon viivoihin ohi vallitsevien selitys- ja tulkinta ketjujen.37 Silloin teoksen kanssaoleminen tarjoaa pääsyn myös yllättävään ja enna koimattomaan eettiseen suhteeseen. Särön mahdollisuus tarkoittaa jonkin särkymistä, rakoilua ja rikkoutumista tai irti päästämistä – oli kyse sitten uskomuksista, ajattelumal Tällaisia nykyisyyden luonnetta voimak kaasti määrittäviä mutta jo tapahtuneita asioita ovat esimerkiksi 1800-luvun lopun teollisen vallankumouksen synnyttämä ja nyt luonnollistunut ylikulutus ja luonnonva rojen tuhlaaminen, joka ilmastokatastrofista huolimatta kasvaa edelleen.39 Teollistumisen ja siihen liittyvän kapitalistisen edistysuskon kuvat näkyvät muun muassa Free Shopin ilmaisen ostamisen utopiassa ja käänteise Ne edelsivät interventiota ja jäivät sinne in tervention jälkeen. Siitäkin huolimatta tällai silla mahdollisesti arkipäiväisillä esineillä on keskeinen rooli ihmisten yhteen tuomisessa ja kommunikaation luojina: ne toimivat ih mistenvälisen yhdessäolemisen aineellisina välittäjinä. Silti esineiden korostaminen ei merkitse paluuta taiteen formaaleihin esteet tisiin tai historiallisiin sisältöihin vaan ennen 168 Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T Goldsmith (2011)40 on viitannut Huebleriin luonnehtiessaan epäluovaa kirjoittamista (uncreative writing). Goldsmithiä soveltaen epäluova taide ei olisi luovuuden puutetta vaan ennen kaikkea kieltäytymistä tuotta masta maailmaan lisää sitä, mikä jo lähtö kohtaisesti on ongelma ja haaste: esineitä, yleisyyttä, samuutta. Juuri goldsmithiläiseen epäluovuuteen myös Tatu Gustafssonin näyttelyynsä lainaamat kulutushistorialliset esineet transsituationaalisesti tarttuvat. kaikkea esineiden ja asioiden huomioimi seen osallistavan taiteen aktiivisina toimijoi na. Siten ymmärrän materiaalisuuden roolin eräänlaisena synnyttäjänä, muokkaajana ja itseään laajemman muodon antajana. Lopuksi Esi neet kuljettavat, muokkaavat ja vaikuttavat. Viitteet Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press, 2007), 205. rogoff/. Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press, 2007), 205. 24 Ks. Latour, ”From Realpolitik to Dingpolitik, or How to Make Things Public”, in Making Things Public: Atmospheres of Democracy, ed. Bruno Latour & Peter Weibel (Cambridge, MA: The MIT Press, 2005), 14–41. rogoff/. Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press, 2007), 205. 16). Laajemmin särö merkitsee minkä tahansa uuden mahdollisuuden ilmaantumista vallitsevien uskomusten ja ideologioiden takaa. y ) 14 Maria Lind, ”The Collaborative Turn”, in Taking the Matter into the Common Hands: Contemporary Art and Collaborative Practices, ed. Johanna Billing, Maria Lind & L. Nilsson (London: Black Dog, 2007), 17. y ) 14 Maria Lind, ”The Collaborative Turn”, in Taking the Matter into the Common Hands: Contemporary Art and Collaborative Practices, ed. Johanna Billing, Maria Lind & L. Nilsson (London: Black Dog, 2007), 17. y ) 14 Maria Lind, ”The Collaborative Turn”, in Taking the Matter into the Common Hands: Contemporary Art and Collaborative Practices, ed. Johanna Billing, Maria Lind & L. Nilsson (London: Black Dog, 2007), 17. 25 Harman, Object-Oriented Ontology. 26 Jane Bennett, Vibrant Matter: A Political Ecology of Things (Durham, NC: Duke University Press, 2010). 27 soap wi-fi stilts 2.–25.8.2019, Titanik-galleria, lehdistötiedote, luettu 15.6.2020, http://www.titanik.fi/ 38 Luomallani moninkirjoittamisen käsitteellä hahmotan teosluentaa, joka pitää sisällään erilaisten kirjoittamistapojen moninaisuuden ja tapojen välisen vaihdettavuuden. Se siten sallii vastuullisen moneuden synnyttämisen ja ylläpitämisen taiteessa ja sen luentatavoissa. Huom. moninkirjoittaminen ei ole toisinkirjoittamista. Ajatus toisinkirjoittamisesta pitää yllä hierarkkista suhdetta suhteessa alkuperäiseen tai vallitsevaan kirjoittamisen tai luennan tapaan. Siten se implikoi vain vaihtoehtoisuutta ja myöntyy keskiön ja marginaalin väliseen binääriseen suhteeseen. 39 Vrt. Susan Sontag on rinnastanut kuluttamisen syövän ja tuberkuloosin kaltaisiin tauteihin (tautien 38 Luomallani moninkirjoittamisen käsitteellä hahmotan teosluentaa, joka pitää sisällään erilaisten kirjoittamistapojen moninaisuuden ja tapojen välisen vaihdettavuuden. Se siten sallii vastuullisen moneuden synnyttämisen ja ylläpitämisen taiteessa ja sen luentatavoissa. Huom. moninkirjoittaminen ei ole toisinkirjoittamista. Ajatus toisinkirjoittamisesta pitää yllä hierarkkista suhdetta suhteessa alkuperäiseen tai vallitsevaan kirjoittamisen tai luennan tapaan. Siten se implikoi vain vaihtoehtoisuutta ja myöntyy keskiön ja marginaalin väliseen binääriseen suhteeseen. 39 Vrt. Viitteet 1 ”The world is full of objects, more or less interesting; I do not wish to add any more.” Douglas Huebler, Statement, 1969, luettu 2.7.2020, , http://www.ubu. com/papers/huebler_statements.html. 2 Niin sanotusta aineettomasta taiteesta käytetään kontekstista riippuen erilaisia termejä, kuten esimerkiksi ephemeral, intangible, dematerialized, immaterial, objectless ja conceptual art. Keskustelun aineettomasta taiteesta käynnistivät Lucy Lippard ja John Chandler 1968 julkaistulla artikkelillaan ”The Dematerialization of Art”, Art International 12, no. 2 (February 1968): 31–36. Siksi ehdotan tässä tekstissä taiteen eri tyiseen rajaamiseen ja kategorioiden ylläpi tämiseen pohjautuvan neuvottelun rinnalle moninaisuutta ja sallivuutta: ja-sanan käyt töä dualismien sijaan. Taiteen transsitua tionaalisuus – ja siten myös epävarmuus ja ambivalenssi – luo marginaalien dyna miikkaa, muutosta ja liikettä, joka tasa-ar voistaa ja moninaistaa myös taidehistorian tutkimusta. Epävarmuuden tunnustaminen tekee taiteesta kirjoittamisesta poliittisesti ja eettisesti vastuullisen harjoitteen, jolla etsiä kestävää suhdetta myös vieraisiin tai taiteek si tunnistamattomiinkin aineellisiin element teihin. 3 Muun muassa Claire Bishop on määritellyt osallistavan taiteen kolmen tunnusomaisen piirteen mukaan: se 1) aktivoi katsojaa, 2) haastaa perinteisen tekijyyden tasa-arvoistamalla taideteoksen luomisprosessia sekä 3) muokkaa yhteisöjä. Claire Bishop, ”The Social Turn: Collaboration and its Discontents”, Artforum (February 2006). 4 Free Shopin kaksi ensimmäistä toteutuskertaa olivat vuonna 2003 Tokiossa ja Bremenissä, minkä jälkeen se on uusittu eri puolilla maailmaa erilaisten taidetapahtumien ja -festivaalien yhteydessä. Ks. esim. Riikka Haapalainen, Utopioiden arkipäivää: Osallistumisen ja muutoksen paikkoja nykytaiteessa 1980–2011 (Helsinki: Unigrafia, 2018), 125, 148; Superflex, Free Shop: Anything the Customer Wants to Purchase is Free (Copenhagen: Pork Salad Press, 2009).l Aloitin tekstini siteeraamalla Douglas Huebleria ja hänen havaintoaan siitä, kuin ka maailma on täyttymässä tavaroihin. Free Shopin dokumenttikuvat osoittavat juuri tä män tavaroiden määrättömyyden: niissä paljastuu kulutettavaksi tarkoitettujen tava roiden houkutus ja runsaus. Myös Kenneth 5 Superflex, Free Shop, 2009. 6 Termi on omani, ks. lisää Haapalainen, ”Utopioiden arkipäivää”, 13, 29. 7 Moninlukeminen on kehittämäni käsite, jolla viittaan samanaikaisesti vaihtoehtoisiin teoksen luentatapoihin. 8 Klassinen esimerkki taiteen kontemplatiivisesta vastaanotosta ks. Johann Joachim Winckelmann, Jalosta yksinkertaisuudesta: Kirjoituksia antiikin taiteesta ja arkkitehtuurista, suom. Vesa Oittinen 169 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa T alainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa (Helsinki: VAPK-kustannus, 1992). 9 Bishop, ”The Social Turn: Collaboration and its Discontents”, 8. 10 Suzana Milevska, ”Participatory Art: A Paradigm Shift from Objects to Subjects”, Springerin 6, no. 2 (Feb. 2006), luettu 1.8.2020, https://www.springerin. Viitteet at/en/2006/2/partizipatorische-kunst/ 11 Myös Latour luonnehtii ihmisten ja asioiden välittämää sosiaalista vuorovaikutuksen maailmaa jo valmiina olevana: already-there, ks. Bruno Latour, ”On Interobjectivity”, Mind, Culture, and Activity 3, no. 4 (1996): 234, luettu 1.8.2020, http://www.bruno- latour.fr/sites/default/files/63-INTEROBJECTS-GB. pdf. 19 Bishop, ”The Social Turn”, 10. 20 Latour, ”On Interobjectivity”, 1996, passim. Ks. myös Turo-Kimmo Lehtosen luonnehdinta materiaalisuuden merkityksestä yhteisöjen ylläpitämisessä ja luomisessa. Turo-Kimmo Lehtonen, Aineellinen yhteisö (Tutkijaliitto: Helsinki, 2008). 21 Ylipäätään taiteen jaottelu puhtaasti aineettomaan ja aineelliseen on ongelmallinen. Esimerkiksi Graham Harmanin esittelemä ooo-teoria (object- oriented-ontology) väittää, että mikään ei lopulta ole aineetonta, vaan kaikella on aineellinen perusta. Graham Harman, Object-Oriented Ontology: A New Theory of Everything (London: Pelican Books, 2018), 43. 31 Martha Buskirk, The Contingent Object of Contemporary Art (Cambridge, MA: The MIT Press, 2003), 217. Viive ja välittömyys luennassani eivät sisällä hierarkkista tai muutoin arvottavaa eroa. 32 Ibid. ( ) 9 Bishop, ”The Social Turn: Collaboration and its Discontents”, 8. , p y g Shift from Objects to Subjects”, Springerin 6, no. 2 (Feb. 2006), luettu 1.8.2020, https://www.springerin. at/en/2006/2/partizipatorische-kunst/ 11 Myös Latour luonnehtii ihmisten ja asioiden välittämää sosiaalista vuorovaikutuksen maailmaa jo valmiina olevana: already-there, ks. Bruno Latour, ”On Interobjectivity”, Mind, Culture, and Activity 3, no. 4 (1996): 234, luettu 1.8.2020, http://www.bruno- latour.fr/sites/default/files/63-INTEROBJECTS-GB. pdf. 33 Ibid., 218. 34 Tämä merkitsee Donna Harawayn paikallistuneen tiedon ajatuksen aktiivista soveltamista. Ks. Donna Haraway, ”Situated Knowledges: The Science Question in Feminism and the Priviledge of Partial Perspective”, Feminist Studies 14, no. 3 (1988): 575–599. ”On Interobjectivity”, Mind, Culture, and Activity 3, no. 4 (1996): 234, luettu 1.8.2020, http://www.bruno- latour.fr/sites/default/files/63-INTEROBJECTS-GB. pdf. 35 Ks. esim. Clement Greenberg, ”’Amerikkalaistyyppisestä’ maalauksesta”, suom. Leevi Lehto, teoksessa Modernin ulottuvuuksia. Fragmentteja modernista ja postmodernista, toim. Jaakko Lintinen (Helsinki: Kustannusosakeyhtiö Taide, 1989), 104–132. ”’Amerikkalaistyyppisestä’ maalauksesta”, suom. Leevi Lehto, teoksessa Modernin ulottuvuuksia. Fragmentteja modernista ja postmodernista, toim. Jaakko Lintinen (Helsinki: Kustannusosakeyhtiö Taide, 1989), 104–132. 12 Ks. Roland Barthes, Tekijän kuolema, tekstin syntymä, suom. Lea Rojola & Pirjo Thorel (Tampere: Vastapaino, 1993); Olli Pyyhtinen, ”Geometrista sosiologiaa”, Tiede ja Edistys 2, nro 27 (2002): 161–162. 22 Michel Maffesoli, Maailman mieli. Yhteisöllisen tyylin muodoista, suom. Mika Määttänen (Helsinki: Gaudeamus, 1995), 140–144. ) 23 Ks. esimerkiksi Pablo Helguera, Education for Socially Engaged Art: A Materials and Techniques Handbook (New York: Jorge Pinto Books, 2011); Kaija Kaitavuori, Art of Engagement: Audience Participation and Contemporary Art (London: Courtauld Institute of Art, University of London, 2015). 24 Ks. Viitteet Latour, ”From Realpolitik to Dingpolitik, or How to Make Things Public”, in Making Things Public: Atmospheres of Democracy, ed. Bruno Latour & Peter Weibel (Cambridge, MA: The MIT Press, 2005), 14–41. 23 Ks. esimerkiksi Pablo Helguera, Education for Socially Engaged Art: A Materials and Techniques Handbook (New York: Jorge Pinto Books, 2011); Kaija Kaitavuori, Art of Engagement: Audience Participation and Contemporary Art (London: Courtauld Institute of Art, University of London, 2015). 36 Karen Barad, “Posthumanist Performativity: Toward an Understanding of How Matter Comes to Matter”, Journal of Women in Culture and Society 28, no. 3 (2003): 801–831. 6 6 13 Irit Rogoff, ”Looking Away – Participating Singularities, Ontological Communities” (IKKM Research project, 2011), 133, luettu 15.6.2020, http:// www.ikkm-weimar.de/en/fellows/former-fellows/irit- rogoff/. Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press, 2007), 205. 14 Maria Lind, ”The Collaborative Turn”, in Taking the Matter into the Common Hands: Contemporary Art and Collaborative Practices, ed. Johanna Billing, Maria Lind & L. Nilsson (London: Black Dog, 2007), 17. 15 Nicolas Bourriaud, Relational Aesthetics (Dijon- Quetigny: Les presses du reel, 2002). 16 Bourriaud, Relational Aesthetics, 45. 17 Claire Bishop,”Antagonism and Relational Aesthetics”, October 110 (Fall 2004): 64–65. 18 Claire Bishop, Artificial Hells: Participatory Art and the Politics of Spectatorship (London: Verso, 2012), passim. 13 Irit Rogoff, ”Looking Away – Participating Singularities, Ontological Communities” (IKKM Research project, 2011), 133, luettu 15.6.2020, http:// www.ikkm-weimar.de/en/fellows/former-fellows/irit- rogoff/. Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press 2007) 205 13 Irit Rogoff, ”Looking Away – Participating Singularities, Ontological Communities” (IKKM Research project, 2011), 133, luettu 15.6.2020, http:// www ikkm-weimar de/en/fellows/former-fellows/irit- 13 Irit Rogoff, ”Looking Away – Participating Singularities, Ontological Communities” (IKKM Research project, 2011), 133, luettu 15.6.2020, http:// www.ikkm-weimar.de/en/fellows/former-fellows/irit- rogoff/. Katso myös Latourin ajatus toimijuudesta aktiivisesti todellisuutta muokkaavana, koettelevana elementtinä. Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press, 2007), 205. 14 Maria Lind, ”The Collaborative Turn”, in Taking the Matter into the Common Hands: Contemporary Art and Collaborative Practices, ed. Johanna Billing, Maria Lind & L Nilsson (London: Black Dog 2007) 37 Karl Marx kuvasi säröllä tai raolla kapitalistiselle taloudelle vaihtoehtoista kommunikatiivista suhdetta arkeen (Bourriaud, Relational Aesthetics, 45, 37 Karl Marx kuvasi säröllä tai raolla kapitalistiselle taloudelle vaihtoehtoista kommunikatiivista suhdetta arkeen (Bourriaud, Relational Aesthetics, 45, rogoff/. Viitteet Susan Sontag on rinnastanut kuluttamisen syövän ja tuberkuloosin kaltaisiin tauteihin (tautien 15 Nicolas Bourriaud, Relational Aesthetics (Dijon- Quetigny: Les presses du reel, 2002). 16 Bourriaud, Relational Aesthetics, 45. 17 Claire Bishop,”Antagonism and Relational Aesthetics”, October 110 (Fall 2004): 64–65. 18 Claire Bishop, Artificial Hells: Participatory Art and the Politics of Spectatorship (London: Verso, 2012), passim. 40 Kenneth Goldsmith, Uncreative Writing: Managing Language in the Digital Age (New York: Columbia University Press, 2011). tapaa näivettää ihmistä kuvailtiin juuri kuluttamisena). Susan Sontag, Sairaus vertauskuvana & Aids ja sen vertauskuvat, suom. Osmo Saarinen (Hämeenlinna: Karisto, 1991). soap-wi-fi-stilts/. 28 Lehtonen, ”Aineellinen yhteisö”, 84. 28 Lehtonen, ”Aineellinen yhteisö”, 84. 29 John Berger, The Ways of Seeing (London: Penguin Books, 1971). i 30 Cameron Rowland, Indirect Benefit, exhibition booklet (Fri Art, 2016), luettu 7.7.2020, https://www. fri-art.ch/sites/default/files/2016-09/Indirect%20 Benefit%20Booklet_1.pdf. 170 T Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja as tapaa näivettää ihmistä kuvailtiin juuri kuluttamisena). Susan Sontag, Sairaus vertauskuvana & Aids ja sen vertauskuvat, suom. Osmo Saarinen (Hämeenlinna: Karisto, 1991). tapaa näivettää ihmistä kuvailtiin juuri kuluttamisena). Susan Sontag, Sairaus vertauskuvana & Aids ja sen vertauskuvat, suom. Osmo Saarinen (Hämeenlinna: Karisto, 1991). FT Riikka Haapalainen on taidehisto rioitsija ja kasvatustieteilijä, joka toimii vanhempana yliopistonlehtorina Aal to-yliopiston taiteen laitoksella. Haapalai sen tutkimus- ja opetusaloina ovat muun muassa avantgarden ja nykytaiteen teoriat, taiteen osallistavat ja sosiaaliset prosessit sekä näyttelypedagogiikka ja kriittinen mu seotutkimus. 40 Kenneth Goldsmith, Uncreative Writing: Managing Language in the Digital Age (New York: Columbia University Press, 2011). 171 Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa Tahiti 4/2020 \| Artikkelit \| Haapalainen: Transsituationaaliset esineet ja asiat osallistavassa taiteessa
https://openalex.org/W3207970508	https://www.frontiersin.org/articles/10.3389/fpubh.2021.732539/pdf	English	null	Mental Health Needs Assessment During the COVID-19 Pandemic: Consensus Based on Delphi Study	Frontiers in public health	2,021	cc-by	5,704	ORIGINAL RESEARCH published: 20 October 2021 doi: 10.3389/fpubh.2021.732539 Mental Health Needs Assessment During the COVID-19 Pandemic: Consensus Based on Delphi Study Irena Makivi ´c, Vesna Švab and Špela Selak Prevention and Promotion Programmes Management, National Institute of Public Health, Ljubljana, Slovenia The COVID-19 pandemic has revealed signiﬁcant gaps in mental health in terms of unrecognized and unmet needs. The goal was to accurately assess the needs and identify gaps in this area during the epidemiological crisis. A Delphi study to identify the needs was conducted with a group of decision-makers, experts, and users of mental health services. A starting point of the Delphi study was prepared in two working groups, based on recognizable international recommendations and experiences of the practitioners from the ﬁeld situation. This initial set of emergency measures was supplemented through the ﬁrst Delphi round, and consensus about the importance was reached in the second round. A total of 41 activities were derived, the vast majority of which were rated with a score of 4 or more. Mental health activities, which should be addressed in terms of needs, can be divided into systemic measures and service measures. This study recognizes a need to reorganize services in the direction of improving local accessibility and strengthening the network of services for immediate responses to the psychological, health, and social needs of individuals, including those arising from crisis situations, such as COVID-19 pandemic. The results of this study are in line with the international recommendations and also inﬂuenced the formulation of the Action Plan of the National Mental Health Program, while some of the measures were already implemented during the publication of the research results. INTRODUCTION Specialty section: This article was submitted to Public Mental Health, a section of the journal Frontiers in Public Health The COVID-19 pandemic has aﬀected the mental health of populations around the world. Evidence from the outbreaks of SARS, Ebola, Inﬂuenza A H1N1, and MERS has shown their impact on psychological distress (1–5), which was also recognized from the beginning of the COVID-19 pandemic (6–12). Feelings of bereavement, isolation, and fear during pandemics, as well as loss of income, increased the risk of developing insomnia, anxiety, and depressive symptoms. People at the risk of mental disorders may experience a relapse with regard to a range of mental and substance- use disorders (6, 10, 13). Pre-existing mental disorders increase the risk of becoming severely physically ill or even dying, as well as the risk of having long-term complications due to COVID-19 (14, 15). Widespread psychological distress was reported in many populations during the reported period, and a long-term increase in the number and severity of mental health problems is possible (16). The eﬀects of the COVID-19 pandemic on mental health are as important as understanding its clinical features, transmission patterns, and the overall management of the outbreak (17). Received: 29 June 2021 Accepted: 17 September 2021 Published: 20 October 2021 Keywords: needs assessment, mental health, health plan implementation, COVID-19, pandemic Edited by: Sharon Lawn, Flinders University, Australia Edited by: Sharon Lawn, Flinders University, Australia Reviewed by: Kristiana Siste, University of Indonesia, Indonesia Preethy Kathiresan, All India Institute of Medical Sciences Jodhpur, India Correspondence: Irena Makivi ´c irena.makivic@nijz.si MATERIALS AND METHODS To gain the highest level of consensus among professionals working in the ﬁeld of mental health, the Delphi method for gathering data from respondents focusing on their domain of expertise (38) and achieving consensus on measures was used. Two interdisciplinary working groups, formed within the implementation of the Resolution on the Mental Health Program 2018–2028, were asked to suggest potential participants to take part in the Delphi study. The participants, invited from all key stakeholder groups, were chosen due to their reputation, skills, and role in the ﬁeld of mental health. The selection of the participants was additionally supplemented by the snowball technique to prepare the ﬁnal list. Therefore, we invited 128 participants from whom the majority (53.9%) were service providers, and the rest were decision-makers (25.8%) and service users (20.3%). None of the invited participants were excluded, only those who did not want to participate and did not sign the agreement were not included in the ﬁrst round of the Delphi study. Given the increased needs caused by the pandemic, diﬀerent problems with regard to accessing care can be addressed through the organization of eﬃcient, aﬀordable, and human rights protective community-based services as a part of any national COVID-19 recovery plan (16). Obstacles to accessing mental health services should be removed. The stigma and discrimination (20) that seemed to increase during the times of crisis should also be reduced (21). An assessment process with all stakeholders, i.e., service users and providers from diﬀerent professional groups and purchasers (primarily politicians), is needed to provide evidence-based solutions for an eﬀective consensus-based planning (22). User involvement is essential both in decision-making as well as within the implementation (23), through the involvement of peer support (24) and with addressing the needs of diﬀerent vulnerable groups. Mental health services should also be scaled up because the COVID- 19 pandemic produced greater social isolation which increases the risk of domestic violence and poor mental health (25). Educational programs and humanitarian assistance supporting vulnerable groups are needed. Organized support should also be accessible for overburdened frontline workers (26). The majority of the actions mentioned above are also suggested in the United Nations guidelines (16), which deﬁne a set of actions to urgently minimize the mental health consequences of the pandemic. Citation: Makivi ´c I, Švab V and Selak Š (2021) Mental Health Needs Assessment During the COVID-19 Pandemic: Consensus Based on Delphi Study. Front. Public Health 9:732539. doi: 10.3389/fpubh.2021.732539 October 2021 \| Volume 9 \| Article 732539 Frontiers in Public Health \| www.frontiersin.org Mental Health Needs Assessment Makivi ´c et al. care system through a consensus-based process. The present study serves as an evidence-based approach to plan further actions in this context. The results are also a part of a formative evaluation process for further steps in the mental health ﬁeld. Some Slovenian research also examines the health ﬁeld using a Delphi study (27), as do some international studies (28–34), along with addressing the mental health of an individual (35), but none has carried out a mental health needs assessment during the pandemic. Research about mental health services in this context is also an opportunity (36) for service organization improvements and to address mental health needs with the engagement of professional organizations, a political review of mental health plans and policies, and consultation with service users (37). While the COVID-19 pandemic increased the demand for mental health services, the WHO nevertheless reported that mental health services have been disrupted or halted in 93% of countries worldwide, including Slovenia. The WHO identiﬁed mental health as an integral component of COVID-19 responses, and recommended allocating more resources in order to maintain or increase mental health services during the pandemic (13). Moreover, based on the growing trend of mental health problems in many European countries, it is necessary to scale-up mental health services (18), so they can better match the actual mental health needs. In 2018, the National Mental Health Program (NMHP) 2018–2028 was adopted in Slovenia (19) to reform the mental health provisions with the aim of achieving comprehensive, community-oriented, and needs-led services according to the international recommendations. During 2020, the National Institute of Public Health in Slovenia started to coordinate the second Action Plan to implement the NMPH from 2021 to 2023 including a crisis plan regarding the COVID-19 pandemic. Psychiatric services are currently organized in outpatient clinics, psychiatric hospitals, and in the newly developed community- based mental health centers at the primary health care level. During the crisis, improved accessibility was recognized as an essential factor in reducing the mental health burden, and a general recommendation was made to organize services in the community. Frontiers in Public Health \| www.frontiersin.org MATERIALS AND METHODS Speciﬁcally, these call for a whole-society approach to promote, protect, and care for mental health, ensure the widespread availability of emergency mental health and psychosocial support, and support recovery from COVID-19 by building mental health services for the future. The ﬁrst set of priority activities during the COVID-19 pandemic was listed by two interdisciplinary working groups (Figure 1) gathered together to prepare starting points for mental health needs assessment. The ﬁrst group addressed the network of services and the second addressed the mental health education. Together they had a total number of 30 members, including decision-makers, service providers, and service users equally represented. Those groups prepared a list of proposed activities, representing diﬀerent measures in the ﬁeld of systemic regulation and management of mental health. This initial list was sent to 129 participants who were invited to participate in the Delphi study, to make a review, and add additional comments. The purpose of this ﬁrst round of the Delphi study was to obtain comments on the proposed activities and gather proposals for additional activities considered necessary in the system in the long run, and emergency ones that need to be implemented as soon as possible. After the ﬁrst round, all comments and suggestions were reviewed and taken into consideration when designing the ﬁnal set of activities, including proposals for additional measures. All commented and proposed activities were combined in terms of content and meaningfully formed into the ﬁnal set of activities This study aimed to identify and design a set of emergency mental health activities needed in the Slovenian mental health October 2021 \| Volume 9 \| Article 732539 2 Makivi ´c et al. Mental Health Needs Assessment FIGURE 1 \| Workﬂow of the Delphi study. FIGURE 1 \| Workﬂow of the Delphi study. of 41 activities within 16 diﬀerent measures. The ﬁnal set was included in the last Delphi round to be assessed with regard to the level of importance. that was sent to the second and last rounds of the Delphi study to all participants who participated in the ﬁrst round. In the second round, participants rated the importance of the listed activities on a 5-point Likert scale from 1 (not important at all) to 5 (very important). The aim of the last round of the Delphi study was to reach a consensus (at least 70% agreement) on the importance (score 4 or 5) of each activity. MATERIALS AND METHODS For the purpose of the last round, the statistical analysis was used. We calculated arithmetical means and modes for all activities that were assessed by the participants in the second round. Consensus was based on the percentage of those who assessed the activity with the grade of 4 or higher. The study highlighted the most important activities according to the recognized needs in the ﬁeld of mental health during and after the COVID-19 pandemic. These are as follows: (A) Local action groups should be developed or scaled-up to achieve the delivery of food, medicines, and protective equipment to the homes of the people, as well as companionship for those who are lonely, isolated, and excluded because of poverty. The core actions are to be taken through NGO networks and social-care centers. Protocols for connected and continuous outreach are needed, especially for vulnerable groups, such as children, the elderly, and people with preexisting mental disorders. Online and telephone access should be provided to ensure connections among service providers and users. (B) Regular and mandatory consultations should be established among health, social, and educational services, among all sectors with the inclusion of humanitarian and civil society organizations to assess the needs for mental health interventions in the regions, and to enable these to be delivered locally and in good time. (C) A single entry point is necessary to access the psychological support interventions with an accessible telephone number and 24-h information, support, and crisis counseling. (D) Accessible and highly professional psychological assistance for traumatized and overstressed individuals working in the frontlines of treatment and care is needed, as well as specialized care for those infected with COVID-19. The solution chosen as the most eﬀective was to establish educational programs and support for diﬀerent intensity levels of care (from self-help and mutual support to highly qualiﬁed psychotherapy). In addition to lay training about the management of problems related to insecurity, fear, and stress, regular supervision of professionals is necessary, especially during a crisis. More speciﬁcally, continuous training Frontiers in Public Health \| www.frontiersin.org RESULTS (I) To strengthen the network of vocational rehabilitation and supported employment facilities as well as the promotion of social entrepreneurship, because of the deep economic crisis. (J) The need to provide accessible transport for a group of people with mental health problems was also listed among the priorities. (K) Protective equipment must be provided for the related ﬁeld work (for health workers, social workers, including those in accommodation programs, and for relatives or carers of people with mental health problems), as well as equipment for remote access. such as the Delphi technique. The results of our Delphi study are in line with the international guidelines (44) using the existing knowledge about the problem from reports and scientiﬁc databases (45), whereas in our study, the Delphi- based consensus was used, i.e., an open and thorough discussion with stakeholders to highlight the essential activities according to the recognized needs in mental health during and after the COVID-19 pandemic, and to identify various systemic and service measures that need to be addressed. Based on the results, the recognized needs in mental health during and after the COVID-19 pandemic are as follows (Figure 2): (1) Professional level: better management (coordinated work of mental health councils and local health action groups for accessible, needs-based support) and quality (addressing highly professional support through adequate educational programs and supervisions). (2) System and service level: upgrade of the services network (mental health centers organized locally, accommodation facilities, vocational rehabilitation, and day centers), improved access to mental health services (coordinated protocols for diﬀerent stakeholders, transportation, and protective equipment within the COVID-19 pandemic and a single entry point for people with mental health problems), user involvement, and service extension (lay assistance and promoting peer support advocacy as well as addressing the risk of human rights violations bearing in mind all the vulnerable groups, improvement of services working against domestic violence and other forms of abuse). (3) Reducing discrimination and promoting destigmatization. The COVID-19 pandemic revealed many gaps in service provision and put a new emphasis on the community mental health care model (46). A locally accessible service network was seen as one of the highest priorities in this context, together with local and central government and the coordination of services, including protocols for communication among diﬀerent services. RESULTS With regard to the ﬁrst Delphi round, from all the invited participants, more women (40) than men (9) responded to the invitation and participated in the study, and the overall response rate of the ﬁrst round was 37.9%. There were 11 (22.4%) representatives of policy- or decision-makers, 28 (57.2%) representatives of professionals in the ﬁeld of mental health (all professions), and 10 (20.4%) representatives of service users or relatives. All those who participated in the ﬁrst round were asked to assess the importance of each activity. We received 38 responses in the second round, representing a high (77.6%) response rate. There were 7 (18.4%) representatives of decision- makers or politicians, 23 (60.5%) representatives of professionals, and 8 (21.1%) representatives of service users or relatives, with a total of 33 women and 5 men participating in the second round. The initial list contained 53 activities, representing 12 measures covering diﬀerent ﬁelds of service organization supplementing the network of services and education. The list was sent to the ﬁrst Delphi round. The ﬁnal set was composed October 2021 \| Volume 9 \| Article 732539 3 Mental Health Needs Assessment Makivi ´c et al. and supervision are needed for professionals in health, social, and educational services, especially for those working with vulnerable groups (the lonely, elderly, children and adolescents, victims of violence, people with long-term mental health problems, and other vulnerable groups). Peer support has proven to be successful in Slovenia and is also listed among the priority actions. (E) Destigmatization campaigns need to be continuously implemented to improve help-seeking behaviors. (F) Services addressing domestic violence need to be strengthened. Safe points for reporting such violence need to be established and their visibility must be improved. (G) In times of crisis, the risk of human rights violations is high, so there is an important call to strengthen the advocacy and representation services for people with mental health problems and to monitor the protection of their rights, including institutional and hospital settings. (H) To expand the network of accommodation facilities with diﬀerent levels of support for people in need, to increase the number of (smaller) group homes or housing units for people with mental disorders (especially in areas where these facilities do not exist), and to strengthen the network of day-care centers. RESULTS The need to address the needs of the people at their homes was also an important priority because of the obviously inadequate protection of people with mental health disorders in institutions (47). The consensus reached by the Delphi study is in line with this recognition, and is as such, another important call for mental health reform in Slovenia. The need for social protection of service users is included in the Delphi results, including the terms of accommodation, provision of life necessities, and vocational rehabilitation, transport, and employment, as stressed in the last WHO recommendations about community care (47). Human rights protection with adequate direct involvement of service users in the planning and delivery of services was once again stressed, but with little reﬂection in the subsequent political decisions (48). Moreover, before the Delphi study in Slovenia, mental health experts from diﬀerent backgrounds along with volunteers provided a single entry point which was established as a national helpline (49), which is supported and complemented by the local mental health psychological support oﬀered within health promotion centers (50). Furthermore, information and support materials, including guidelines, were developed for professionals and the general public on a large-scale basis, addressing diﬀerent aspects of mental health promotion and prevention during the pandemic, All the given activities were rated on average with a score of 4.3 on the 1-to-5 scale of importance, but most often with a score of 5 (mode of all overall scores). On average, the percentage of agreement or consensus level was 84.3% (from 71.1 to 97.4%). In the vast majority of cases, all participants evaluated all the statements, and only one person did not give a rating for only three statements. On average, all but one of the activities were rated 4 or higher. Only one activity, addressing the work of mental health councils, had an average score of 3.9. Participants, on average, expressed the opinion that all activities are important or very important. Given the high level of consensus reached, the third round was not necessary. Frontiers in Public Health \| www.frontiersin.org DISCUSSION The article describes the Delphi study conducted to identify the needs and gaps in mental health service delivery during the period of the COVID-19 pandemic in Slovenia. The study identiﬁed various systemic and service measures that need to be addressed and/or strengthened. Reports of mental health interventions during the COVID-19 pandemic are multiplying rapidly. Several studies (39–43) have so far suggested diﬀerent activities, measures, and recommendations to address mental health challenges during the pandemic, but to the best of our knowledge, none has used a scientiﬁcally supported method, October 2021 \| Volume 9 \| Article 732539 Frontiers in Public Health \| www.frontiersin.org 4 Makivi ´c et al. Mental Health Needs Assessment FIGURE 2 \| Essential activities during and after the COVID-19 pandemic. FIGURE 2 \| Essential activities during and after the COVID-19 pandemic. some also in the languages of the various national communities in Slovenia (51). representatives of all stakeholder groups and tried to obtain a more balanced group of participants. Another limitation is that the primary-care teams working during the pandemic were not speciﬁcally addressed, with regard to the training and guidelines in the mental health ﬁeld. With the increase in telemedicine services across the globe, there is also a need to address the problems that can arise with these. In the discussion part of this study, the need for better online and telephone access to improve the connections between services and users were mentioned, but the limitations and challenges that can occur with this were not addressed. There is likely to be a need for new guidelines that can help manage this new context for the provision of services. The major strength of this study is the inclusion of policy- makers, service users, and mental health professionals. The Delphi results call for reforming mental health services in line with international recommendations (52, 53) and measures abroad, and we ﬁnd our results comparable to those reported in other countries, as well as applicable to our context. The Delphi consultation achieved the participation of service users, professionals, and providers in the codesign of emergency measures. Particular attention should be focused on additional services, especially for vulnerable groups. The ﬁndings are important to address the knowledge gap in the ﬁeld of mental health needs during pandemics. The results of the current research call for good collaboration among the existing services and sectors, as well as a shift toward a community-based approach. DISCUSSION However, despite the rigorous methodology that was used, the present research has some limitations, as follows. Namely, the Delphi study did not include health- and social-care professionals working in the frontlines with COVID-19 patients at the beginning of the pandemic. It is possible that we addressed more individuals who are already engaged in this ﬁeld but do not necessarily know all the disadvantages of the current system that should be addressed. We also do not know if we included a large group of people with mental health needs to address their diﬀerent views on the focal problems in a suﬃciently comprehensive manner. Nevertheless, we repeatedly searched for Frontiers in Public Health \| www.frontiersin.org REFERENCES Niederberger M, Spranger J. Delphi technique in health sciences: a map. Front Public Heal. (2020) 8:457. doi: 10.3389/fpubh.2020.00457 7. World Health Organization. COVID-19 Disrupting Mental Health Services in Most Countries, WHO Survey. (2020). Available online at: https://www.who. int/news/item/05-10-2020-covid-19-disrupting-mental-health-services-in- most-countries-who-survey (accessed February 10, 2021). 23. Tait L, Lester H. Encouraging user involvement in mental health services. Adv Psychiatr Treat. (2005) 11:168–75. doi: 10.1192/apt.11.3.168 24. Mental Health Foundation. Peer support. Available online at: https://www. mentalhealth.org.uk/a-to-z/p/peer-support (accessed February 10, 2021). 8. Kumar S. Caring for people who care: supporting health workers during the COVID 19 pandemic. EClinicalMedicine. (2020) 28:100667. doi: 10.1016/j.eclinm.2020.100667 25. Usher K, Bhullar N, Durkin J, GyamﬁN, Jackson D. Family violence and COVID-19: Increased vulnerability and reduced options for support. Int J Ment Health Nurs. (2020). 2020, 1–11. doi: 10.1111/inm.12735 9. Pan KY, Kok AAL, Eikelenboom M, Horsfall M, Jörg F, Luteijn RA, et al. The mental health impact of the COVID-19 pandemic on people with and without depressive, anxiety, or obsessive-compulsive disorders: a longitudinal study of three Dutch case-control cohorts. Lancet Psychiat. (2021) 8:121– 9. doi: 10.1016/S2215-0366(20)30491-0 26. El-Hage W, Hingray C, Lemogne C, Yrondi A, Brunault P, Bienvenu T, et al. Health professionals facing the coronavirus disease 2019 (COVID- 19) pandemic: what are the mental health risks? Encephale. (2020) 46:73– 80. doi: 10.1016/j.encep.2020.04.008 10. Liang L, Ren H, Cao R, Hu Y, Qin Z, Li C, Mei S. The eﬀect of COVID-19 on Youth Mental Health. Psychiatr Q. (2020) 21:1– 12. doi: 10.1007/s11126-020-09744-3 27. Susiˇc TP, Švab I, Kolšek M. Community actions against alcohol drinking in Slovenia—a Delphi study. Drug Alcohol Depend. (2006) 83:255– 61. doi: 10.1016/j.drugalcdep.2005.11.021 11. Huang Y, Zhao N. Generalized anxiety disorder, depressive symptoms and sleep quality during COVID-19 outbreak in China: a web-based cross-sectional survey. Psychiatry Res. (2020) 288:112954. doi: 10.1016/j.psychres.2020.112954 28. Howarth E, Vainre M, Humphrey A, Lombardo C, Hanaﬁah AN, Anderson JK, et al. Delphi study to identify key features of community-based child and adolescent mental health services in the East of England. BMJ Open. (2019) 9:e022936. doi: 10.1136/bmjopen-2018-022936 j p y 12. McGinty EE, Presskreischer R, Han H, Barry CL. Psychological distress and loneliness reported by US adults in 2018 and April 2020. JAMA. (2020) 324:94–4. doi: 10.1001/jama.2020.9740 29. Bovopoulos N, Jorm AF, Bond KS, LaMontagne AD, Reavley NJ, Kelly CM, et al. Providing mental health ﬁrst aid in the workplace: A Delphi consensus study. BMC Psychol. (2016) 4:41. REFERENCES coronavirus disease 2019 (COVID-19). JAMA Netw Open. (2020) 3:e2023282. doi: 10.1001/jamanetworkopen.2020.23282 coronavirus disease 2019 (COVID-19). JAMA Netw Open. (2020) 3:e2023282. doi: 10.1001/jamanetworkopen.2020.23282 15. Volkow ND. Collision of the COVID-19 and addiction epidemics. Ann Intern Med. (2020) 2:M20–1212. doi: 10.7326/M20-1212 1. Maunder RG, Peladeau N, Leszcz M, Romano D, Savage D, Rose M, et al. Applying the lessons of SARS to pandemic inﬂuenza an evidence-based approach to mitiga-ting the stress experienced by healthcare workers. Can J Public Health. (2008) 99:486–8. doi: 10.1007/BF03403782 16. United Nations. Policy Brief: COVID-19 and the Need for Action on Mental Health. (2020). Available online at: https://unsdg.un.org/resources/policy- brief-covid-19-and-need-action-mental-health (accessed February 10, 2021). 2. Goodwin R, Haque S, Neto F, Myers LB. Initial psychological responses to Inﬂuenza A, H1N1 (“Swine ﬂu”). BMC Infect Dis. (2009) 9:166. doi: 10.1186/1471-2334-9-166 17. Javed B, Sarwer A, Soto EB, Mashwani ZUR. Impact of SARS-CoV-2 (Coronavirus) Pandemic on Public Mental Health. Front Public Heal. (2020) 8:292. doi: 10.3389/fpubh.2020.00292 3. Khalid I, Khalid TJ, Qabajah MR, Barnard AG, Qushmaq IA. Healthcare workers emotions, perceived stressors and coping strategies during a MERS- CoV outbreak. Clin Med Res. (2016) 14:7–14. doi: 10.3121/cmr.2016.1303 18. Organisation for Economic Co-operation and Development. Tackling the Mental Health Impact of the COVID-19 Crisis: An Integrated, Whole- of-Society Response. (2021). Available online at: https://www.oecd.org/ coronavirus/policy-responses/tackling-the-mental-health-impact-of- the-covid-19-crisis-an-integrated-whole-of-society-response-0ccafa0b/ (accessed February 10, 2021). 4. Mohammed A, Sheikh TL, Gidado S, Poggensee G, Nguku P, Olayinka A, et al. An evaluation of psychological distress and social support of survivors and contacts of Ebola virus disease infection and their relatives in Lagos, Nigeria: a cross sectional study. BMC Public Health. (2015) 15:824. doi: 10.1186/s12889-015-2167-6 19. Resolucija o nacionalnem programu duševnega zdravja 2018–2028. Uradni list RS, št. 24/18; (2018). 5. Brand J, Mckay D, Wheaton MG, Abramowitz JS. The relationship between obsessive compulsive beliefs and symptoms, anxiety and disgust sensitivity, and Swine Flu fears. J Obsessive Compuls Relat Disord. (2013) 2:200– 6. doi: 10.1016/j.jocrd.2013.01.007 20. Mayoclinic. Mental health: Overcoming the Stigma of Mental Illness. (2017). Available online at: https://www.mayoclinic.org/diseases-conditions/mental- illness/in-depth/mental-health/art-20046477 (accessed February 10, 2021). 21. Tehrani H. Mental health stigma related to novel coronavirus disease (COVID-19) in older adults. Geriatr Gerontol Int. (2020) 20:796–7. doi: 10.1111/ggi.13985 6. Pierce M, Hope H, Ford T, Hatch S, Hotopf M, John A, et al. Mental health before and during the COVID-19 pandemic: a longitudinal probability sample survey of the UK population. Lancet Psychiat. (2020) 7:883– 92. doi: 10.1016/S2215-0366(20)30308-4 22. FUNDING Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study. This work was supported by the Slovenian Research Agency, Project No.: Z3-2652. CONCLUSIONS Our main ﬁnding is that a consensus about emergency mental health activities during a pandemic or other natural disaster is needed, and this can be constructed with the Delphi method. The actions already launched to meet the mental health needs of the population in Slovenia during the COVID-19 pandemic correspond well with the results of this study. Nevertheless, continuous consultation among experts, politicians, and service users is needed to make the right decisions at the right time, according to our experience. If the Delphi recommendations were strictly followed in Slovenia, and followed in a timely manner, we could overcome many challenges. October 2021 \| Volume 9 \| Article 732539 5 Mental Health Needs Assessment Makivi ´c et al. DATA AVAILABILITY STATEMENT IM lead methodology and carried out ﬁeld work. IM and VŠ recoded answers and make analysis. ŠS wrote introduction. IM wrote methodology section. VŠ wrote conclusion. All authors contributed to the article and approved the submitted version. The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. REFERENCES doi: 10.1186/s40359- 016-0148-x 13. World Health Organization. The Impact of COVID-19 on Mental, Neurological and Substance Use Services. (2020). Available online at: https://www.who.int/ publications/i/item/978924012455 (accessed February 10, 2021). 30. Lauriks S, de Wit MAS, Buster MCA, Arah OA, Klazinga NS. Composing a core set of performance indicators for public mental health care: a modiﬁed delphi procedure. Adm Policy Ment Heal Ment Heal Serv Res. (2014) 41:625– 35. doi: 10.1007/s10488-013-0506-4 14. Li L, Li F, Fortunati F, Krystal JH. Association of a prior psychiatric diagnosis with mortality among hospitalized patients with October 2021 \| Volume 9 \| Article 732539 Frontiers in Public Health \| www.frontiersin.org 6 Makivi ´c et al. Mental Health Needs Assessment 31. Hemmings CP, Underwood LA, Bouras N. Services in the community for adults with psychosis and intellectual disabilities: a Delphi consultation of professionals’ views. J Intellect Disabil Res. (2009) 53:677–84. doi: 10.1111/j.1365-2788.2008.01146.x 46. Thornicroft G, Tansella M. The balanced care model: the case for both hospital- and community-based mental healthcare. Br J Psychiatry. (2013) 202:246–8. doi: 10.1192/bjp.bp.112.111377 47. World Health Organization. Long-Stay Mental Health Care Institutions and the COVID-19 Crisis: Identifying and Addressing the Challenges for Better Response and Preparedness. (2020). Available online at: https:// www.euro.who.int/en/health-topics/noncommunicable-diseases/mental- health/publications/2020/long-stay-mental-health-care-institutions-and- the-covid-19-crisis-identifying-and-addressing-the-challenges-for-better- response-and-preparedness-2020 (accessed August 30, 2021). 32. Langlands RL, Jorm AF, Kelly CM, Kitchener BA. First aid for depression: a Delphi consensus study with consumers, carers and clinicians. J Aﬀect Disord. (2008) 105:157–65. doi: 10.1016/j.jad.2007.05.004 33. Kirkham EJ, Crompton CJ, Iveson MH, Beange I, McIntosh AM, Fletcher-Watson S. Co-development of a best practice checklist for mental health data science: a Delphi study. Front Psychiatry. (2021) 12:643914. doi: 10.3389/fpsyt.2021.643914 48. Ažman R, Švab V. Mental Health in the Community: the Case of Slovenia. Ljubljana: Šent, Slovensko združenje za duševno zdravje. (2015). 34. Jorm AF. Using the Delphi expert consensus method in mental health research. Aust N Z J Psychiatry. (2015) 49:887– 97. doi: 10.1177/0004867415600891 49. Nacionalni inštitut za javno zdravje. Telefon za Psihološko Podporo ob Epidemiji COVID-19. (2021). Available online at: https://www.nijz.si/sl/ zazivel-nov-telefon-za-psiholosko-podporo-ob-epidemiji-covid-19-080- 51-00 (accessed August 10, 2021). 35. Cleverley K, Bartha C, Strudwick G, Chakraborty R, Srivastava R. The development of a client care needs assessment tool for mental health and addictions settings using a modiﬁed delphi approach. Nurs Leadersh. (2018) 31:52–65. doi: 10.12927/cjnl.2018.25603 50. Vinko M, Roškar S, Šubara TN, Grum AT. Local Mental Health Support in Slovenia During COVID-19: Setting Up Primary Health Care Helplines. EuroHealthNet magazine (2020). REFERENCES Available online at: https://eurohealthnet- magazine.eu/local-mental-health-support-in-slovenia-during-covid- 19-setting-up-primary-health-care-helplines/ (accessed February 10, 2021). 36. Moreno C, Wykes T, Galderisi S, Nordentoft M, Crossley N, Jones N, et al. How mental health care should change as a consequence of the COVID-19 pandemic. Lancet Psychiat. (2020) 7:813–24. doi: 10.1016/S2215-0366(20)30307-2 51. Nacionalni inštitut za javno zdravje. Skrb za duševno zdravje v ˇcasu širjenja novega koronavirusa SARS-CoV-2. (2020). Available online at: https://www. nijz.si/sl/koronavirus-dusevno-zdravje (accessed February 10, 2021). 37. Ghebreyesus TA. Addressing mental health needs: an integral part of COVID- 19 response. World Psychiatry. (2020) 19:129–30. doi: 10.1002/wps.20768 38. Hsu C-C. The Delphi technique: making sense of consensus. Pract Assess Res Eval. (2007) 12:10. doi: 10.7275/pdz9-th90 52. World Health Organization. Substantial Investment Needed to Avert Mental Health Crisis. (2020). Available online at: https://www.who.int/news/item/ 14-05-2020-substantial-investment-needed-to-avert-mental-health-crisis (accessed February 10, 2021). 39. Talevi D, Socci V, Carai M, Carnaghi G, Faleri S, Trebbi E, et al. Mental health outcomes of the covid-19 pandemic. Riv Psichiatr. (2020) 55:137–44. doi: 10.1708/3382.33569 40. Antoine B, Nicolas D, Laurent D. Reshaping community mental health services during the COVID-19 epidemic—report from the 59G21 Service in Lille, France. Health Serv Insights. (2020) 13:1178632920954876. doi: 10.1177/1178632920954876 53. World Health Organization. Looking Back, Looking Forward: Rapid Assessment of the Mental Health System in Slovenia. Report of a virtual mission by the WHO Regional Oﬃce for Europe. Geneva: WHO (2020),1–22. 41. Pelizza L, Pupo S. The COVID-19 pandemic and Italian Public mental health services: experience and future directions. J Patient Exp. (2020) 7:642– 4. doi: 10.1177/2374373520956857 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. 42. Staševi´c-Karliˇci´c I, Dordevi´c V, Staševi´c M, Suboti´c T, Filipovi´c Z, Ignjatovi´c-Risti´c D, et al. Perspectives on mental health services during the COVID-19 epidemic in serbia. Srp Arh Celok Lek. (2020) 148:28– 28. doi: 10.2298/SARH200504028S 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. 43. Kuzman MR, Curkovic M, Wasserman D. Principles of mental health care during the COVID-19 pandemic. Eur Psychiatry. (2020) 63:e45. doi: 10.1192/j.eurpsy.2020.54 44. World Health Organization. Copyright © 2021 Makivi´c, Švab and Selak. 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. Frontiers in Public Health \| www.frontiersin.org REFERENCES Mental Health and COVID-19. (2020). Available online at: https://www.who.int/teams/mental-health-and-substance-use/ covid-19 (accessed February 10, 2021). Copyright © 2021 Makivi´c, Švab and Selak. 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. 45. Almeda N, Garcia-Alonso C, Salvador-Carulla L. Mental health planning at a very early stage of the COVID-19 crisis: a systematic review of online international strategies and recommendations. BMC Psychiatry. (2021) 21:43. doi: 10.1186/s12888-020-03015-y October 2021 \| Volume 9 \| Article 732539 Frontiers in Public Health \| www.frontiersin.org
https://openalex.org/W2093667779	https://www.scielo.br/j/rsp/a/RHQMY74qJfTsRXCxkxy9Jzy/?lang=pt&format=pdf	Portuguese	null	Análise do coeficiente de mortalidade materna no município de Osasco, S. Paulo, Brasil	Revista de saúde pública/Revista de Saúde Pública	1,972	cc-by	3,203	* Da Disciplina de Saúde Materna do Departamento de Prática de Saúde Pública da USP — Av. Dr. Analdo, 715 — São Paulo, S. P., Brasil. ANÁLISE DO COEFICIENTE DE MORTALIDADE MATERNA NO MUNICÍPIO DE OSASCO, S. PAULO, BRASIL* Cyro CIARI Jr. Pedro Augusto Marcondes de ALMEIDA RSPSP-136 1.000, tem sofrido algumas críticas. No entanto, representa ainda a melhor forma de se medir a intensidade do evento, sen- do sistematicamente usado. CIARI JR., C. & ALMEIDA, P.A.M. DE — Aná- lise do coeficiente de mortalidade ma- terna no município de Osasco, S. Pau- lo, Brasil. Rev. Saúde públ., S. Paulo, 6: 237-44, 1972. A ocorrência de óbito materno tem di- minuído sensivelmente em países onde se desenvolvem intensos programas de Saúde Materna. Isto ocorre, por exemplo, na cidade de São Francisco, onde o coe- ficiente é de 1,1/10.000, o mesmo aconte- cendo na cidade de Bristol (1,8/10.000) 4. Porém, em áreas menos desenvolvidas, a freqüência do evento óbito por causas obstétricas ainda se mantém alta, co- mo: Bogotá, 12.6; Lima, 15.6; Santiago, 31.6/10.000 habitantes. RESUMO: Foi analisado o coeficiente de mortalidade materna no município de Osasco e observado que o mesmo ainda é alto comparado ao de S. Paulo (Capital) e ao de cidades localizadas em países mais desenvolvidos. Porém, o comporta- mento deste coeficiente, nos diversos anos, mostra efetiva melhora nos servi- ços de saúde materna, através da redu- ção de 28,99% no período de 1967/1970. Observou-se que, ao ser analisado o coe- ficiente de mortalidade materna deve ser levado em consideração não só a sua magnitude, como também o seu compor- tamento nos diferentes anos. Além disso, deve-se considerar a proporção que apre- senta em relação às demais causas de óbitos no grupo fértil feminino e o rela- cionamento entre si das causas obstétri- cas. Foi salientada a correspondência existente entre causas obstétricas de óbi- to e serviços de saúde materna. Na análise da mortalidade materna, para que seja adequadamente compreen- dida e para que expresse os fatores que a estão determinando, devemos levar em consideração alguns aspectos básicos: 1 — Magnitude do coeficiente compa- rado com outras áreas; UNITERMOS — Saúde materna; Mortali- dade materna; Assistência pré-natal. UNITERMOS — Saúde materna; Mortali- dade materna; Assistência pré-natal. 2 — comportamento do coeficiente nos diversos anos; MATERIAL E MÉTODOS Desta forma, não basta sabermos a magnitude do coeficiente de mortalidade materna. É necessário interpretá-lo em seus aspectos mais intimos. O seu com- portamento durante os anos observados, nos dá a medida da atuação de serviços de maternidade. Além disso, devemos co- nhecer a sua situação em relação ao nú- mero de óbitos que ocorrem no grupo fértil feminino pois, naturalmente, deve sempre estar em último lugar, dado ser a gravidez um processo fisiológico pouco determinante de óbito. Porém, este enfo- que deve ainda ser estendido ao compor- tamento das causas obstétricas entre si, pois, cada uma delas expressa com sua presença, uma deficiência de serviços. De uma maneira geral, os óbitos ocorridos por toxemia, hemorragia ou infecção mostram deficiência, seja na assistência pré-natal, ou seja, na assistência ao par- to. Na avaliação dos serviços de saúde materna ainda é importante conhecer-se a proporção dos óbitos que estão ocor- rendo na gravidez, no parto ou no puer- pério deficiências de serviços. ue mais uma vez evidenciam as Foram levantados no Cartório de Re- gistro Civil de Osasco, os óbitos ocorri- dos nos anos de 1967 a 1970, inclusive. Destes, foram retirados os femininos que ocorreram na idade de 15-50 anos. Do montante destes óbitos tomamos aqueles onde havia a indicação de causa obsté- trica. Nestes procurava-se classificar as causas do óbito e sempre que necessário recorreu-se ao médico assistente ou re- gistro do hospital para esclarecimento, principalmente para identificar a idade da gestação e se o óbito tinha ocorrido na gestação, no parto ou no puerpério. Na classificação de causas adotaram-se os critérios da Classificação Internacio- nal de Doenças, revisão de 1965 2. Quan- do dividimos os óbitos maternos ocorri- dos durante a gravidez, o parto e o puer- pério, adotamos o critério de situá-lo acordo com a causa básica do óbito. As- sim, um caso de infecção puerperal es- tará situado no puerpério, porém, se a infecção iniciou-se no parto (infecção in- tra-parto) foi classificado neste grupo e não mais em puerpério. Pretendemos analisar a situação da mortalidade materna no município de Osasco levando em consideração estes 5 aspectos básicos. Desta forma, podere- mos aquilatar a situação da saúde ma- terna naquele município enfocada apenas pelo aspecto de perdas de gestantes. Não levamos em consideração a evasão de partos que ocorre no município dada sua proximidade de São Paulo. I N T R O D U Ç Ã O 3 — proporção dos óbitos maternos em relação a outras causas de óbito de mulheres em idade fértil; O coeficiente de mortalidade materna, expressado pelo número de óbitos ocor- ridos por causas de gravidez, parto, puer- pério ou aborto sobre nascidos vivos por 4 — proporção das causas de óbito materno entre si; 5 — proporção de óbitos maternos ocorridos durante a gravidez, o parto, o puerpério e por aborto. 5 — proporção de óbitos maternos ocorridos durante a gravidez, o parto, o puerpério e por aborto. 5 — proporção de óbitos maternos ocorridos durante a gravidez, o parto, o puerpério e por aborto. ram incluídos para o cálculo do coefi- ciente de mortalidade materna. MATERIAL E MÉTODOS Acredita- mos que os números obtidos em Osasco exprimem a realidade local, isto é, o nú- mero de partos que são assistidos no município e para os quais são requeridos serviços. O pequeno número de óbitos, como é usual em saúde materna, não nos permi- te, pela análise destes dados, extrapolar para outras áreas ou afirmar um com- portamento uniforme dos eventos. Contu- do, representa subsídio para uma obser- vação mais prolongada da ocorrência que permitirá então uma conclusão segura. Da mesma forma, deixamos de fazer a análise de óbitos maternos por aborto, da- da à escassez de dados. No entanto, fo- A população foi estimada consideran- se os saldos migratório e vegetativo acha- dos mediante coeficientes médios dos censos, extrapolando linearmente. Nesta estimativa, por este método, achamos er- ro de 2,93% para mais em relação ao censo de 1970 1. Para comparar a proporção de óbitos por causas maternas e as demais causas de morte em mulheres na idade fértil, es- colhemos algumas mais significantes e grupamos outras. Ao lado das causas maternas colocamos neoplasias malig- nas, moléstias renais mais hipertensão, cardiopatias e, num único conjunto, as demais. & GRIFFITH 5 preferem 15-45. No entanto, o escolhido é mais freqüentemente usado em nossa literatura. R E S U L T A D O S A magnitude dos coeficientes de morta- lidade materna no município de Osasco e sua comparação com a Capital, interior e o Estado são vistos na Tabela 1. O grupo feminino em idade reprodutiva consideramos como compreendido entre 15-50 anos. Alguns autores como PUFFER Verificamos que nos anos em que é possível a coincidência de dados, Osasco mantinha seu coeficiente mais alto que os demais. No entanto, verifica-se que a tendência de Osasco e Capital é para re- dução progressiva dos coeficientes, ao passo que no Interior permanecem está- veis. nor proporção, seguindo-se cardiopatias, moléstias renais e câncer. O grupo de outras causas naturalmente supera as demais. Durante os anos observados, verifica- se ainda que a proporção de óbitos obs- tétricos em relação aos demais vem se mantendo dentro de, aproximadamente, um décimo do total de mortes do grupo. Interessante ainda é verificar a propor- ção de óbitos por idade dentro do grupo fértil feminino. Para isto somamos as mortes ocorridas nos 4 anos e distribuí- mos por idade e causas situando sua pro- porção. (Tabela 3). Durante os anos observados, verifica- se ainda que a proporção de óbitos obs- tétricos em relação aos demais vem se mantendo dentro de, aproximadamente, um décimo do total de mortes do grupo. A proporção de óbitos maternos rela- cionados com outras causas de óbito de mulheres em idade fértil, no município de Osasco, é vista na Tabela 2. Interessante ainda é verificar a propor- ção de óbitos por idade dentro do grupo fértil feminino. Para isto somamos as mortes ocorridas nos 4 anos e distribuí- mos por idade e causas situando sua pro- porção. (Tabela 3). Pela Tabela 2, nota-se que os óbitos por causa obstétrica são sempre os de me- Verificamos que a distribuição dos óbi- tos se faz de acordo com a idade mais comum de incidência das moléstias. Outro aspecto a destacar é a proporção mantida pelos óbitos de causa obstétrica entre si. (Tabela 4). Finalmente devemos verificar o com- portamento dos óbitos em relação a sua ocorrência na gestação, no parto e no puerpério. Vemos que a toxemia se situa em pri- meiro lugar, seguida de hemorragia e in- fecção, distribuição esta característica do nosso meio. Observa-se uma predominância para eventos ocorridos na gestação e em me- nor porcentagem no puerpério Verificamos que a distribuição dos óbi- tos se faz de acordo com a idade mais comum de incidência das moléstias. Outro aspecto a destacar é a proporção mantida pelos óbitos de causa obstétrica entre si. (Tabela 4). Verificamos que a distribuição dos óbi- tos se faz de acordo com a idade mais comum de incidência das moléstias. Outro aspecto a destacar é a proporção mantida pelos óbitos de causa obstétrica entre si. (Tabela 4). C O M E N T Á R I O S PUFFER & GRIFFITH 5 em seu trabalho sobre o grupo etário feminino de 15-74 anos, no período de 1962 a 1964, observa- ram praticamente o mesmo em relação a várias cidades. Especialmente sobre São Paulo e Ribeirão Preto, encontraram como coeficiente mais elevado de causa d3 óbito, em primeiro lugar, as cardiopa- tias, seguindo-se os tumores malignos. Como já tínhamos assinalado o coefi- ciente de mortalidade materna mede o número de óbitos que estão ocorrendo em relação aos nascidos vivos. Sendo medida quantitativa este valor aparece principalmente quando são comparadas áreas diferentes. Na análise do coeficiente de Osasco, comparando-se com São Paulo, Capital, observa-se considerável diferença que já não é tão acentuada quando comparada com o Estado e é menor em relação ao interior. No entanto, se observarmos o comportamento do coeficiente de Osasco em relação aos demais da Tabela 1, ve- mos que para a Capital, do ano de 65 para o de 68, houve uma redução de 16,09%, ao passo que em Osasco a redu- ção foi de 28,99%, no período de 67/70. Desta forma parece que a tendência dos serviços de Saúde Materna no Muni- cípio de Osasco é no sentido de atuar de forma significativa na redução da mor- talidade materna. Na análise do coeficiente de Osasco, comparando-se com São Paulo, Capital, observa-se considerável diferença que já não é tão acentuada quando comparada com o Estado e é menor em relação ao interior. No entanto, se observarmos o comportamento do coeficiente de Osasco em relação aos demais da Tabela 1, ve- mos que para a Capital, do ano de 65 para o de 68, houve uma redução de 16,09%, ao passo que em Osasco a redu- ção foi de 28,99%, no período de 67/70. Desta forma, o grupo fértil feminino do Município de Osasco está se mantendo, no que diz respeito aos óbitos de causa obstétrica, em nível porcentual semelhan- te ao existente em outras cidades. Dados importantes são colhidos na Ta- bela 3, onde relacionamos as diversas causas de óbito com vários grupos etá- rios. Observamos que no referente ao total de óbitos há uma elevação da por- centagem de forma diretamente propor- cional à elevação da idade dos grupos etários. O mesmo já não podemos cons- tatar em relação aos óbitos maternos, pois estes se mantêm praticamente está- veis em todos os grupos etários, caindo a 0 no grupo etário 45-50 anos. Observa-se uma predominância para eventos ocorridos na gestação e em me- nor porcentagem no puerpério C O M E N T Á R I O S Quanto aos óbitos por câncer, a tabela reflete evidentemente o que já se esperava, isto é, a elevação rápida e efetiva do porcen- tual a partir do grupo etário 25-30, atin- gindo o máximo no grupo de 45-50. O mesmo observamos no que diz respeito aos óbitos por hipertensão arterial e re- nais, sendo que já no grupo etário mínimo, 15-20 anos, são encontrados. Desta forma parece que a tendência dos serviços de Saúde Materna no Muni- cípio de Osasco é no sentido de atuar de forma significativa na redução da mor- talidade materna. A Tabela 2, demonstra que as causas obstétricas situam-se abaixo das demais causas escolhidas para comparação neste trabalho. No grupo fértil feminino esta deve ser a situação normal do item, cau- sas obstétricas. Por outro lado mantém durante anos a proporção de 1/10 das restantes causas. Importante de se considerar para aná- lise dos serviços de assistência à mater- nidade é a situação do porcentual de óbitos por causas obstétricas em relação às demais caudas. Observamos sob este aspecto que o porcentual de óbitos por cardiopatia suplanta em muito o de cau- sas obstétricas, assim como o de renais e hipertensão arterial que se segue em importância ao grupo de cardiopatias. O porcentual devido aos tumores malignos segue-se ao de renais e hipertensão arte- rial, apresentando também índice mais elevado do que o de causas obstétricas. As cardiopatias atuam sobre todos os grupos etários indiferentemente (Tabela 3). As outras causas de óbitos, por reu- nirem um grupo muito heterogêneo de moléstias, atuam diferentemente, atin- gindo porcentual elevado nos grupos etá- rios baixos e diminuindo a medida que se eleva a idade das pacientes. Um dos aspectos importantes da mor- talidade materna é verificar a porcenta- gem com que participam as diferentes causas obstétricas. Na dependência dos Serviços de Assistência Materna ofereci- dos, varia a importância de cada uma delas. A oferta de antibióticos relegou a plano secundário o óbito por infecção, da mesmo forma que o uso adequado do sangue permite controlar o óbito por he- morragia . A toxemia, moléstia própria da gestação, por suas características, é con- trolada pelos serviços de pré-natal. Des- ta forma a maior ou menor incidência de uma destas causas está relacionada ao tipo de atividade desenvolvida no progra- ma de Saúde Materna. C O M E N T Á R I O S Assim, as deficiên- cias de serviços de pré-natal colocam a toxemia em primeiro lugar, bem como as dificuldades na assistênca ao parto incre- mentam os óbitos por hemorragia e in- fecção. durante a gestação, o parto e o puerpé- rio tamb assistência no local. Na Tabela 5, encon- tramos a porcentagem maior durante a gravidez, seguida de parto e puerpério. Estes dados coincidem com os de óbitos, na Tabela 4. Estes resultados da análise dos fatores que interferem com o coeficiente de mor- talidade materna, demonstram uma coe- rência entre os níveis de redução da mor- talidade e o comportamento intrínseco da redução, pois, se bem que devamos considerar o coeficiente como alto, ele está se reduzindo progressivamente e as causas de óbito materno estão ocupando uma proporção correta. Desta forma, um serviço adequado de assistência à maternidade deve determi- nar a seguinte ordem cronológica de im- portância: toxemia, hemorragia e infec- ção. Os óbitos por aborto devem mere- cer análise especial dada as dificuldades que cercam seu registro. Outra variá- vel, não introduzida neste trabalho, por falta de dados precisos, é a dos óbitos por causas associadas à gravidez, como cardiopatias, hemopatias e outras que, na verdade, deveriam ser as mais impor- tantes e que não aparecem pela falta de registros precisos, perdendo-se entre as causas de óbitos gerais. CONCLUSÕES 1— O coeficiente de mortalidade ma- terna no Município de Osasco é alto comparado ao de São Paulo (Capi- tal) e ao de cidades de países mais desenvolvidos. 1— O coeficiente de mortalidade ma- terna no Município de Osasco é alto comparado ao de São Paulo (Capi- tal) e ao de cidades de países mais desenvolvidos. 2 — O importante na análise do coefi- ciente de mortalidade materna é o seu comportamento nos diversos anos, como indicativo de melhoria de serviços de maternidade. 3 — A relação entre os óbitos por cau- sas obstétricas com as demais no grupo fértil feminino, demonstra que as de gravidez, parto e puer- pério tre todas as causas. A Tabela 4, mostra a situação de Osas- co onde vemos a predominância da toxe- mia (46,87) seguida por hemorragia (34,37) e infecção (18,75). No entanto, vemos que em 1967 a importância das três se equi- valia, ao passo que em 1968 a hemorragia predominou e só houve ajustamento a partir de 1969, necessitando maior obser- vação para verificar se estes resultados se mantêm. Isto revela que é muito re- cente a melhoria de serviços da área de Saúde Materna. Nesta nova linha de idéias a situação dos óbitos ocorridos 4 — A proporção entre os óbitos de causa obstétrica presta-se para ava- liar não só a eficiência dos serviços de assistência à maternidade, como também identificar suas deficiên- cias relacionadas às atividades de pré-natal e assistência ao parto. REFERÊNCIAS BIBLIOGRÁFICAS CIARI JR., C. & ALMEIDA, P.A.M. DE — [Ana- lysis of maternal mortality rate in the county of Osasco, S. Paulo State, Brazil]. Rev. Saúde públ., S. Paulo, 6: 237-44, 1972. 1. ALONSO, J. — Estimativas populacionais e construção de tábua de sobrevivência para o sexo feminino no Município de Osasco. Osasco, Secretaria da Saúde. Departamento Técnico Normativo, 1971. SUMMARY: The maternal mortality rate in the county of Osasco (S. Paulo State, Brazil) was analysed and it was observed that this rate is still high comparing with the city of S. Paulo (S. Paulo State, Bra- zil), as well as with towns of more deve- loped countries. Even so, one can notice that this rate shows a real improvement in the maternal health service with a reduction of the rate of 28,99% during the 1967-1970 period. The necessity that the maternal mortality rate be analized in respect to not only its magnitude but also as to its conduct during the different years, was showed. It is important to compare the obstetrical causes of death with others in the feminine fertile group and the relationship between the diffe- rent obstetrical causes. The importance between these and the maternal health services, was showed. 2. MANUAL de classificação estatística inter- nacional de doenças, lesões e causas de óbito; 8.a revisão, 1965. Washington, D. C., Organização Panamericana da Saúde, 1969. (OPAS — Publicação cien- tífica, 190). 3. MONETTI, V. et al. — Situação da saúde materna e infantil no Estado de São Paulo. São Paulo, Instituto de Saúde. Divisão de Saúde Materna e da Crian- ça, 1970. (Publicação, 4 — Série D, n.o 1). 4. NOTICIÁRIO ESTATÍSTICO (Departamen- to de Estatística do Estado de S. Pau- lo) São Paulo, (1):3-10, 1971. 5. PUFFER, R. R. & GRIFFITH, G. W. — Características de la mortalidad urbana. Washington, D. C., Organización Pana- mericana de la Salud, 1968. p. 191-205 (Publiación científica, 151). UNITERMS: Maternal health; Maternal mortality; Prenatal care. UNITERMS: Maternal health; Maternal mortality; Prenatal care. Recebido para publicação em 17-4-1972 Aprovado para publicação em 26-7-1972.
https://openalex.org/W1997807500	https://bmcmolbiol.biomedcentral.com/counter/pdf/10.1186/1471-2199-9-61	English	null	Phosphorylation at Ser473 regulates heterochromatin protein 1 binding and corepressor function of TIF1beta/KAP1	BMC molecular biology	2,008	cc-by	13,481	BioMed Central BioMed Central This article is available from: http://www.biomedcentral.com/1471-2199/9/61 This article is available from: http://www.biomedcentral.com/1471-2199/9/61 © 2008 Chang 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. BMC Molecular Biology Open A Research article Phosphorylation at Ser473 regulates heterochromatin protein 1 binding and corepressor function of TIF1beta/KAP1 Chiung-Wen Chang1, Han-Yi Chou1, Yu-Sheng Lin1, Kuo-Hsiang Huang1, Ching-Jin Chang2, Tsui-Chun Hsu1 and Sheng-Chung Lee1,2,3 Open Access Open Research article Phosphorylation at Ser473 regulates heterochromatin protein 1 binding and corepressor function of TIF1beta/KAP1 Chiung-Wen Chang1, Han-Yi Chou1, Yu-Sheng Lin1, Kuo-Hsiang Huang1, Ching-Jin Chang2, Tsui-Chun Hsu1 and Sheng-Chung Lee1,2,3 Address: 1Institute of Molecular Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan, 2Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan and 3Institute of Clinical Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan Email: Chiung-Wen Chang - ginthrpi@gmail.com; Han-Yi Chou - hyechou@ntu.edu.tw; Yu-Sheng Lin - b90205205@ntu.edu.tw; Kuo- Hsiang Huang - ckhsaturn@hotmail.com; Ching-Jin Chang - chingjin@gate.sinica.edu.tw; Tsui-Chun Hsu - tchsu2204@yahoo.com.tw; Sheng- Chung Lee* - slee@ntu.edu.tw * Corresponding author Received: 9 April 2008 Accepted: 1 July 2008 Received: 9 April 2008 Accepted: 1 July 2008 Published: 1 July 2008 Published: 1 July 2008 BMC Molecular Biology 2008, 9:61 doi:10.1186/1471-2199-9-61 BMC Molecular Biology 2008, 9:61 doi:10.1186/1471-2199-9-61 Background The recruitment of HP1 proteins by the KRAB-TIF1β complex to specific loci within the genome via the formation of heterochromatin-like complexes may thus silence gene activity. Gene-specific repression may be a consequence of the formation of such complexes [13]. Phosphorylation of both human and mouse TIF1β/ Ser473 has been identified by nuclear phosphoprotein analysis of HeLa and WEHI-231 cells [23,24]. Ser473 is located in the HP1-interacting domain of TIF1β – close to the HP1 box (amino acids 486–490, PXVXL). The conser- vation of TIF1β/Ser473 phosphorylation in various cell lines from different species motivated this investigation of the functional significance of this modification. The coil-coiled domain of TIF1β binds to E2F1 and inhib- its its activity [16]. The induction of cyclin A, Cdc2 and Cdc25A genes depends on E2F [25,26]. Cyclin A is a cell cycle-regulating protein that participates in S-phase con- trol and mitosis in mammalian somatic cells. The pro- moter of cyclin A is repressed during the G1 phase of the cell cycle and is activated at S-phase entry [27]. Cdc2 per- mits the transition from G1 through S in conjunction with cyclin E [28]. These E2F downstream genes are important to normal cell cycle progression [29]. This report shows that TIF1β participates in the regulation of cyclin A2, Cdc2 and Cdc25A gene expression. This regu- lation depends on the interaction between TIF1β and HP1β, which is itself regulated by the phosphorylation state of TIF1β/Ser473. The experimental results suggest that the TIF1β/Ser473-unphosphorylated form binds more strongly than the phosphorylated form to the pro- moters of Cyclin A2, Cdc2, and Cdc25A genes. The phos- phorylation/de-phosphorylation of TIF1β/Ser473 may serve as a molecular switch regulating its interaction with HP1β and gene expression. It has been reported that TIF1β directly interacts with the histone methyltransferase SETDB1, which methylates spe- cifically histone H3/Lys9 within euchromatin [14]. Deple- tion of endogenous levels of TIF1β by siRNA significantly inhibited the KRAB-mediated transcriptional repression of a chromatin template and inhibited cell cycle progres- sion. Cell death may occur if TIF1β is severely depleted by siRNA knockdown [unpublished observations, [15-17]]. Similarly, the knock-down of cellular levels of HP1 pro- teins and SETDB1 by siRNA attenuated KRAB-TIF1β repression. The physiological targets and functions of TIF1β remain unclear. Background taining multiprotein complex regulates various protein activities or genes [19]. The activity of TIF1β may be regu- lated by posttranslational modifications. For example, TIF1β is rapidly phosphorylated by members of the phos- phatidylinositol-3 kinase-like family of kinases following DNA damage. Phosphorylated TIF1β colocalizes with numerous damage response factors at DNA lesions [20]. Recent genetic and proteomic studies in mice have shown that TIF1β is a developmental regulatory protein perform- ing cellular function(s) that are critical to early embryonic development, and is a component of the interactive pro- tein network for the pluripotency of embryonic stem cells [21,22]. The binding of the TIF1β corepressor to the retro- virus primer binding site is responsible for the epigenetic silencing of retrovirus transcription [17]. Therefore, the interaction between TIF1β and HP1 or other transcription factors may account for the regulation of gene expression. g Transcriptional intermediary factor TIF1β and heterochro- matin protein 1 profoundly impact the regulation of the structure and function of chromatin [1]. The heterochro- matin protein 1 (HP1) family of proteins (HP1α, HP1β, and HP1γ) participates in gene silencing by forming hete- rochromatic structures [2,3]. HP1 exhibits distinct nuclear localization patterns: HP1α nassociates with centromeres while HP1β and HP1γ are largely localized in distinct nuclear regions. The nuclear arrangement of HP1 proteins and TIF1β is differentiation pathway-specific, and appears to be more important than changes in the levels of these proteins, which are relatively stable during all of the induced differentiation processes [4,5]. HP1 proteins comprise an N-terminal chromodomain, a C-terminal chromoshadow domain and a hinge domain [3,6,7]. The chromodomain functions as a protein inter- action domain, bringing together different proteins in multi-protein complexes and recruiting them to hetero- chromatin. This domain binds to particular proteins which contain an HP1 box (PXVXL) and function at the transcriptional level [2,8-10]. The chromoshadow domain mediates the formation of homodimer. TIF1β interacts with the chromodomain of dimeric HP1β via the HP1 box of TIF1β in the HP1-interaction domain [6]. The TIF1β ntranscriptional repression activity depends on the interaction between TIF1β and HP1 [11]. This interaction is essential in the relocation of TIF1β from euchromatin to heterochromatin that accompanies the differentiation of primitive endoderm-like cells [12]. TIF1β is proposed to function as a universal co-repressor protein for the KRAB zinc finger protein (KRAB-zfp) superfamily of transcrip- tional repressors. Abstract Background: As an epigenetic regulator, the transcriptional intermediary factor 1β (TIF1β)/ KAP1/TRIM28) has been linked to gene expression and chromatin remodeling at specific loci by association with members of the heterochromatin protein 1 (HP1) family and various other chromatin factors. The interaction between TIF1β and HP1 is crucial for heterochromatin formation and maintenance. The HP1-box, PXVXL, of TIF1β is responsible for its interaction with HP1. However, the underlying mechanism of how the interaction is regulated remains poorly understood. Results: This work demonstrates that TIF1β is phosphorylated on Ser473, the alteration of which is dynamically associated with cell cycle progression and functionally linked to transcriptional regulation. Phosphorylation of TIF1β/Ser473 coincides with the induction of cell cycle gene cyclin A2 at the S-phase. Interestingly, chromatin immunoprecipitation demonstrated that the promoter of cyclin A2 gene is occupied by TIF1β and that such occupancy is inversely correlated with Ser473 phosphorylation. Additionally, when HP1β was co-expressed with TIF1β/S473A, but not TIF1β/ S473E, the colocalization of TIF1β/S473A and HP1β to the promoters of Cdc2 and Cdc25A was enhanced. Non-phosphorylated TIF1β/Ser473 allowed greater TIF1β association with the regulatory regions and the consequent repression of these genes. Consistent with possible inhibition of TIF1β's corepressor function, the phosphorylation of the Ser473 residue, which is located near the HP1-interacting PXVXL motif, compromised the formation of TIF1β-HP1 complex. Finally, we found that the phosphorylation of TIF1β/Ser473 is mediated by the PKCδ pathway and is closely linked to cell proliferation. Conclusion: The modulation of HP1β-TIF1β interaction through the phosphorylation/de- phosphorylation of TIF1β/Ser473 may constitute a molecular switch that regulates the expression of particular genes. Higher levels of phosphorylated TIF1β/Ser473 may be associated with the expression of key regulatory genes for cell cycle progression and the proliferation of cells. Page 1 of 16 (page number not for citation purposes) (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 Background Its interactions with chromatin modification factors such as HDAC1, SETDB1 and HP1 correlate with its activities in regulating the chromatin structure and heterochromatin formation, resulting in epigenetic silencing of reporter genes [18]. A TIF1β-con- Characterization of TIF1β and phosphorylated TIF1β/ Ser473 antibodies Western blotting of interphase HeLa cell extract was per- formed to characterize the specificity of monoclonal anti- Page 2 of 16 (page number not for citation purposes) Page 2 of 16 (page number not for citation purposes) BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471 2199/9/ Characterization of monoclonal anti-TIF1β (20A1) and rabb anti-phosphorylated TIF1β/Ser473 (S473) antibodies Figure 1 Characterization of monoclonal anti-TIF1β (20A1) and rabbit anti-phosphorylated TIF1β/Ser473 (S473) antibodies. (A) HeLa c lysate were probed with anti-TIF1β monoclonal antibody, 20A1 (lane 1 Lysate from 293T cells transfected with FLAG-TIF1β were probed with 20A1 antibody (lane 2) or anti-Flag monoclonal antibody, M2 (lane 3). ( Whole cell extracts were prepared from 293T cells and incubated in th absence (lanes 1 and 3) or presence (lanes 2 and 4) of CIP (30 U) for 3 min at 37°C. Western blotting was performed with antibodies indicated (C) M2-beads purified FLAG-TIF1β from 293T cells were treated witho (lane 1) or with (lane 2) CIP (30U) for 30 min. Western blot was probe with rabbit anti-TIF1β/Ser473 antibody (S473), anti-Flag antibody (M2), and anti-TIF1β antibody (20A1). Wild-type FLAG-TIF1β (lane 3), FLAG TIF1β/S473A (lane 4) or FLAG-TIF1β/S473E (lane 5), were immunoprec itated by M2-beads, resolved in SDS-PAGE, transferred onto membrane and probed with antibodies as indicated. body to TIF1β, clone 20A1. A specific band of 100 kDa was detected (Figure 1A). 293T cell-expressed FLAG-TIF1β was recognized either by 20A1 or monoclonal anti-FLAG antibody, M2 (Figure 1A), demonstrating that mono- clonal antibody 20A1 is specific to TIF1β. The specificity of rabbit anti-phospho-Ser473 antibody was characterized by Western blotting of interphase 293T cell extract with rabbit anti-TIF1β/phospho-Ser473 anti- body (S473) or 20A1 monoclonal antibody. The phos- phorylated TIF1β/Ser473 signal was diminished after calf intestine alkaline phosphatase (CIP) treatment (Figure 1B). Ectopically-expressed FLAG-TIF1β was immunopre- cipitated from 293T cells with M2 beads and treated with CIP to examine whether the signal was due to the phos- phorylation of TIF1β/S473. 20A1 and S473 antibodies both recognized FLAG-TIF1β, but the signal recognized by the S473 antibody disappeared upon CIP treatment (Fig- ure 1C). Unlike the FLAG-TIF1β, mutants FLAG-TIF1β/ S473A and FLAG-TIF1β/S473E were not detected by S473 antibody (Figure 1C). These data demonstrate that while 20A1 or S473 antibodies recognize the same protein, TIF1β, the S473 antibody specifically recognizes phos- phorylated TIF1β/S473. Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression Although TIF1β is known to be a general corepressor pro- tein, few studies have suggested that TIF1β has a critical function in cell cycle progression. The phosphorylation of TIF1β/Ser473 was identified in HeLa and WEHI cells at interphase [23,24]. Since Ser473 is located near the HP1- binding motif, an intriguing question arises: what is the functional consequence, if any, of the phosphorylation of TIF1β/Ser473? To elucidate the phosphorylation state of TIF1β/Ser473 during cell cycle progression, Western blot- ting was performed with lysates from synchronized HeLa cells. Cells were synchronized to G1/S transition using double thymidine block. Phosphorylated TIF1β/Ser473 was almost undetectable at the G1/S boundary under the thymidine block (Figure 2A, 0 hr, S473) but the phospho- rylation of TIF1β/Ser473 was maximal 1 hour post-release from the thymidine block, during early S phase (Figure 2A, 1 hr, S473). The level of phosphorylated TIF1β/Ser473 decreased thereafter for 2 hours during the mid to late S- phase cell cycle progression. The level of total TIF1β remained relatively constant (Figure 2A, 20A1). The ele- vated cyclin A and decreased cyclin B evidenced the G1/S to S cell cycle progression of thymidine release (Figure 2A). ( ) p p y ( ) g Characterization of monoclonal anti-TIF1β (20A1) and rabbit anti-phosphorylated TIF1β/Ser473 (S473) antibodies. (A) HeLa cell lysate were probed with anti-TIF1β monoclonal antibody, 20A1 (lane 1). Lysate from 293T cells transfected with FLAG-TIF1β were probed with 20A1 antibody (lane 2) or anti-Flag monoclonal antibody, M2 (lane 3). (B) Whole cell extracts were prepared from 293T cells and incubated in the absence (lanes 1 and 3) or presence (lanes 2 and 4) of CIP (30 U) for 30 min at 37°C. Western blotting was performed with antibodies indicated. (C) M2-beads purified FLAG-TIF1β from 293T cells were treated without (lane 1) or with (lane 2) CIP (30U) for 30 min. Western blot was probed with rabbit anti-TIF1β/Ser473 antibody (S473), anti-Flag antibody (M2), and anti-TIF1β antibody (20A1). Wild-type FLAG-TIF1β (lane 3), FLAG- TIF1β/S473A (lane 4) or FLAG-TIF1β/S473E (lane 5), were immunoprecip- itated by M2-beads, resolved in SDS-PAGE, transferred onto membrane and probed with antibodies as indicated. HeLa cells were synchronized to prometaphase by noco- dazole treatment. The level of phosphorylated TIF1β/ Ser473 was determined from HeLa extracts collected from 0 to 8 hours after nocodazole were removed. The phos- HeLa cells were synchronized to prometaphase by noco- dazole treatment. Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression The level of phosphorylated TIF1β/ Ser473 was determined from HeLa extracts collected from 0 to 8 hours after nocodazole were removed. The phos- Page 3 of 16 (page number not for citation purposes) Page 3 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 uring cell cycle progression ng cell cycle progression. (A) HeLa cells were arrested at G1/S boundary by vels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 so shown. (B) HeLa cells were synchronized by nocodazole (1 μM) treatment for . The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 tone H3 S10 (H3S10), or α-tubulin were also probed as indicated. The relative termination of the images from Western blots (by ImageGauge software, normal- nel). (C) HeLa cells were arrested at G1 phase by serum starvation for three days phorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The pro- ated. HeLa cells at 0 hr (G1) or 2 hours after serum addition (early-S) were immu- ter stained with DAPI (blue). (D) Immunostaining of ectopically expressed FLAG- 2) and wild-type FLAG-TIF1β (column 3) were transiently transfected into HeLa M2 (red) and HP1β (green) antibodies. DNA was counter stained with DAPI ndary antibody alone (data not shown). Bars in C represent 30 μm and bars in D Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression Figure 2 Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression. (A) HeLa cells were arrested at G1/S boundary by double thymidine block and released into S phase for three hours. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The protein levels of cyclin A, cyclin B1, or α-tubulin were also shown. (B) HeLa cells were synchronized by nocodazole (1 μM) treatment for 16 hours. Mitotic shake off cells were collected and released for 8 hours. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The protein levels of cyclin B1, phosphorylated Histone H3 S10 (H3S10), or α-tubulin were also probed as indicated. The relative phosphorylation level of TIF1β/Ser473 was compared by quantitative determination of the images from Western blots (by ImageGauge software, normal- ized to individual TIF1β level) at each time point (shown in the lower panel). Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression (C) HeLa cells were arrested at G1 phase by serum starvation for three days and released to S phase by serum addition. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The pro- tein levels of cyclin A, cyclin B1, and α-tubulin were also probed as indicated. HeLa cells at 0 hr (G1) or 2 hours after serum addition (early-S) were immu- nostained with 20A1 (red) and S473 (green) antibodies, DNA was counter stained with DAPI (blue). (D) Immunostaining of ectopically expressed FLAG- TIF1βs. FLAG-TIF1β/S473A (column 1) or FLAG-TIF1β/S473E (column 2) and wild-type FLAG-TIF1β (column 3) were transiently transfected into HeLa cells. Cells were fixed at 36 hours after transfection and co-stained with M2 (red) and HP1β (green) antibodies. DNA was counter stained with DAPI (blue). Control immunostaining was negative when performed with secondary antibody alone (data not shown). Bars in C represent 30 μm and bars in D show 8 μm. Phosphory Figure 2 Phosphory Figure 2 The M- phase marker cyclin B or phosphorylated histone H3S10 identify specific stages of mitotic release (Figure 2B). date the role of TIF1β in regulating gene expression, we examined whether its interaction with HP1β was affected by the phosphorylation of S473. Immunoprecipitation and Western blotting experiments were performed using 293T cells that were co-transfected with FLAG-TIF1β (TIF1β), FLAG-TIF1β/S473A (S473A) or FLAG-TIF1β/ S473E (S473E) and HA-HP1β. The interaction between FLAG-TIF1β/S473A and HP1β was stronger than that between FLAG-TIF1β/S473E or FLAG-TIF1β and HP1β (Figure 3B). To demonstrate that the reduced phosphor- ylation level of TIF1β/Ser473 was related to its stronger interaction with HP1β, a pull-down assay was performed using purified GST-HP1β from E. coli and FLAG-TIF1β, FLAG-TIF1β/S473A or FLAG-TIF1β/S473E from 293T cells. GST-HP1β interacted with non-phosphorylated- mimetic FLAG-TIF1β/S473A 1.6 times more effectively than with phosphorylated-mimetic FLAG-TIF1β/S473E (Figure 3C). The similarity of the level of interaction between wild-type FLAG-TIF1β and GST-HP1β and that of FLAG-TIF1β/S473A and GST-HP1β (1.39 vs. 1.63) may reflect a lower phosphorylation level of wild-type FLAG- TIF1β here. To verify that TIF1β/Ser473 phosphorylation compromises the interaction between TIF1β and HP1β, M2 or HA antibody pull-down assays were conducted using in vitro translated HA-HP1β and FLAG-TIF1β, FLAG- TIF1β/S473A, or FLAG-TIF1β/S473E. The results of the M2 pull-down and the HA antibody pull-down assay indeed demonstrated that the interaction between HA- HP1β and FLAG-TIF1β/S473A is much stronger than that between FLAG-TIF1β/S473E and FLAG-TIF1β (Figure 3D). Taken together, these data demonstrate that the phosphorylation state of TIF1β/Ser473 influences the interaction between TIF1β and HP1β both in vivo and in vitro. When TIF1β/Ser473 is phosphorylated, its interac- tion with HP1β is compromised. To further evaluate the phosphorylation state of TIF1β/ S473 from G1 to S phase, HeLa cells were serum starved for three days to arrest in G1 phase. The level of phospho- rylated TIF1β/Ser473 was almost non-detectable in G1 phase (Figure 2C, 0 hr, S473), reached a maximum 2 hours after serum was added and declined thereafter (Fig- ure 2C, 2 hr, S473). The gradual increase in the level of cyclin A or decrease in the level of cyclin B demonstrated the specific cell cycle stage from G1 to S (Figure 2C). This dynamic, biphasic (S and M phases) appearance of phos- phorylated TIF1β/Ser473 suggests that TIF1β may be involved in regulating the cell cycle by the switching on and off of Ser473 phosphorylation. Phosphory Figure 2 Various cell cycle regulated proteins may exhibit changed activity when phosphorylation modulates their stability. The total protein level of TIF1β remained approximately constant while the level of phosphorylated TIF1β/Ser473 fluctuated throughout the cell cycle. To further demon- strate this dynamic fluctuation of phosphorylated TIF1β/ Ser473 level, HeLa cells were immunostained using S473 antibody. The level of phosphorylated TIF1β/Ser473 in the cells 2 hr after serum addition markedly exceeded that in the serum-starved cells (Figure 2C, immunofluores- cence staining, compare early S and G1, S473). The pro- tein level and distribution of TIF1β during G1 were the same as those of early S phase (Figure 2C, immunofluo- rescence staining, compare 20A1). Furthermore, in later experiments using over-expressed FLAG-TIF1βs, similar nuclear distribution patterns of over-expressed wild-type FLAG-TIF1β, FLAG-TIF1β/S473A and FLAG-TIF1β/S473E were observed (Figure 2D, compare M2). HP1β staining also showed nuclear localization with over-expressed FLAG-TIF1βs (Figure 2D, compare HP1β in transfected cells). Although the TIF1β co-repressor function is known to be related to HP1β, few studies have addressed the specific gene targets of TIF1β. TIF1β/S473 phosphorylation is up- regulated during the S phase in HeLa cells and the interac- tion between HP1β and TIF1β is compromised when Ser473 is phosphorylated. These observations suggest that the phosphorylation of TIF1β/Ser473 may regulate gene expression by abolishing its interaction with HP1β. As a well known corepressor, TIF1β regulates gene expres- sion through direct binding to HP1 proteins in numerous transcriptional-repressed loci. Thus, the immunostaining data further indicate that TIF1β might control its co-regu- lator function in the nuclear microenvironment via Ser473 phosphorylation/dephosphorylation. Phosphory Figure 2 p y β y y g g y p g g Phosphorylation of TIF1β/Ser473 is dynamically regulated during cell cycle progression. (A) HeLa cells were arrested at G1/S boundary by double thymidine block and released into S phase for three hours. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The protein levels of cyclin A, cyclin B1, or α-tubulin were also shown. (B) HeLa cells were synchronized by nocodazole (1 μM) treatment for 16 hours. Mitotic shake off cells were collected and released for 8 hours. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The protein levels of cyclin B1, phosphorylated Histone H3 S10 (H3S10), or α-tubulin were also probed as indicated. The relative phosphorylation level of TIF1β/Ser473 was compared by quantitative determination of the images from Western blots (by ImageGauge software, normal- ized to individual TIF1β level) at each time point (shown in the lower panel). (C) HeLa cells were arrested at G1 phase by serum starvation for three days and released to S phase by serum addition. The levels of TIF1β and phosphorylated TIF1β/Ser473 were probed with 20A1 and S473 antibodies. The pro- tein levels of cyclin A, cyclin B1, and α-tubulin were also probed as indicated. HeLa cells at 0 hr (G1) or 2 hours after serum addition (early-S) were immu- nostained with 20A1 (red) and S473 (green) antibodies, DNA was counter stained with DAPI (blue). (D) Immunostaining of ectopically expressed FLAG- TIF1βs. FLAG-TIF1β/S473A (column 1) or FLAG-TIF1β/S473E (column 2) and wild-type FLAG-TIF1β (column 3) were transiently transfected into HeLa cells. Cells were fixed at 36 hours after transfection and co-stained with M2 (red) and HP1β (green) antibodies. DNA was counter stained with DAPI (blue). Control immunostaining was negative when performed with secondary antibody alone (data not shown). Bars in C represent 30 μm and bars in D show 8 μm. Page 4 of 16 (page number not for citation purposes) Page 4 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 phorylated TIF1β/Ser473 level was highest during mitosis (Figure 2B, 0–2 hours, S473) and decreased thereafter through G1 phase (Figure 2B). The lower panel of Figure 2B shows the level of phosphorylated TIF1β/Ser473 in relation to that of total TIF1β at each time point. TIF1β regulates key genes expression during G1 to S-phase cell cycle progression Chromatin immunoprecipitation (ChIP) experiments were used to study the possible involvement of Ser473- phosphorylated/dephosphorylated TIF1β in the expres- sion of HeLa cell cycle-specific genes during the transition from G1 phase to S phase. TIF1β/phospho-Ser473 was almost undetectable in the G1 phase, and climbed there- after to reach maximum level in early S phase. As Cyclin A2 is known to be expressed in early S phase but not in G1 Phosphorylation of TIF1β/Ser473 compromises interaction between HP1β and TIF1β The Ser473 is located close to the PXVXL (aa 486–490)- containing HP1 box (aa 483–497) (Figure 3A). To eluci- Page 5 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 Interaction between TIF1β and HP1β is compromised by phosphorylation of TIF1β/Ser473 Figure 3 Interaction between TIF1β and HP1β is compromised by phosphorylation of TIF1β/Ser473. (A) Schematic repre- sentation of domain organization of TIF1β protein. The position of Ser473 relative to the PXVXL motif is shown. (B) Western blot results of co-immunoprecipitation of over-expressed FLAG-TIF1β and HA-HP1β. FLAG-TIF1β/S473A, FLAG-TIF1β/S473E and wild-type FLAG-TIF1β were co-expressed with HA-HP1β into 293T cells. The level of ectopically expressed FLAG-TIF1βs (Input, M2) or M2-beads immunoprecipitated FLAG-TIF1βs (M2-IP, M2) was probed with M2 antibodies (TIF1β, M2). The level of ectopically expressed HA-HP1β (Input, HA) or M2-beads co-immunoprecipitated HA-HP1βs (M2-IP, HA) was probed with monoclonal anti-HA antibodies (HP1β, HA). (C) In vitro GST pull-down assay of TIF1βs by recombinant GST-HP1β. Wild-type and mutant FLAG-TIF1βs were expressed in 293T cells. These FLAG-TIF1βs were immunoprecipitated by M2 beads and eluted with M2 peptide. The levels of FLAG-TIF1βs from 1/10 input for each pull down assay were shown (lanes 8–10). Eluted FLAG-TIF1βs were incubated with GST-beads alone (lanes 1–3) or incubated with recombinant GST-HP1β bound glutathione- beads (lanes 4–6). The levels of GST-HP1β and FLAG-TIF1βs in each pull down assay were shown (lanes 1–6). Quantitation of the Western blots (by ImageGauge software) from each pull down sample (lanes 4–6) showing the relative level of FLAG- TIF1βs (A, TIF1β/S473A; E, TIF1β/S473E; W, wild-type TIF1β) pulled down by GST-HP1β, after normalization to individual GST-HP1β levels (right panel). (D) Interaction between in vitro translated FLAG-TIF1βs and HA-HP1β. Upper panel, In vitro translated [35S] FLAG-TIF1βs and [35S]-HA-HP1β were incubated for 30 min before being immunoprecipitated by M2 beads in the presence of 0.5 M NaCl. The immunoprecipitates were resolved by SDS-PAGE and followed by autoradiography to visual- ize HA-HP1β (upper right panel, M2-pull down). Lower panel, In vitro translated [35S] FLAG-TIF1βs and [35S] HA-HP1β were pulled down by protein-G coupled anti-HA antibody. The immunoprecipitated FLAG-TIF1βs are shown in the lower right panel (TIF1βs, IP). Supernatants after immunoprecipitation are also shown, as indicated (S). Interaction between TIF1β and HP1β is compromised by phosphorylation of TIF1β/Ser473 Figure 3 Interaction between TIF1β and HP1β is compromised by phosphorylation of TIF1β/Ser473. (A) Schematic repre- sentation of domain organization of TIF1β protein. Phosphorylation of TIF1β/Ser473 compromises interaction between HP1β and TIF1β The position of Ser473 relative to the PXVXL motif is shown. (B) Western blot results of co-immunoprecipitation of over-expressed FLAG-TIF1β and HA-HP1β. FLAG-TIF1β/S473A, FLAG-TIF1β/S473E and wild-type FLAG-TIF1β were co-expressed with HA-HP1β into 293T cells. The level of ectopically expressed FLAG-TIF1βs (Input, M2) or M2-beads immunoprecipitated FLAG-TIF1βs (M2-IP, M2) was probed with M2 antibodies (TIF1β, M2). The level of ectopically expressed HA-HP1β (Input, HA) or M2-beads co-immunoprecipitated HA-HP1βs (M2-IP, HA) was probed with monoclonal anti-HA antibodies (HP1β, HA). (C) In vitro GST pull-down assay of TIF1βs by recombinant GST-HP1β. Wild-type and mutant FLAG-TIF1βs were expressed in 293T cells. These FLAG-TIF1βs were immunoprecipitated by M2 beads and eluted with M2 peptide. The levels of FLAG-TIF1βs from 1/10 input for each pull down assay were shown (lanes 8–10). Eluted FLAG-TIF1βs were incubated with GST-beads alone (lanes 1–3) or incubated with recombinant GST-HP1β bound glutathione- beads (lanes 4–6). The levels of GST-HP1β and FLAG-TIF1βs in each pull down assay were shown (lanes 1–6). Quantitation of the Western blots (by ImageGauge software) from each pull down sample (lanes 4–6) showing the relative level of FLAG- TIF1βs (A, TIF1β/S473A; E, TIF1β/S473E; W, wild-type TIF1β) pulled down by GST-HP1β, after normalization to individual GST-HP1β levels (right panel). (D) Interaction between in vitro translated FLAG-TIF1βs and HA-HP1β. Upper panel, In vitro translated [35S] FLAG-TIF1βs and [35S]-HA-HP1β were incubated for 30 min before being immunoprecipitated by M2 beads in the presence of 0.5 M NaCl. The immunoprecipitates were resolved by SDS-PAGE and followed by autoradiography to visual- ize HA-HP1β (upper right panel, M2-pull down). Lower panel, In vitro translated [35S] FLAG-TIF1βs and [35S] HA-HP1β were pulled down by protein-G coupled anti-HA antibody. The immunoprecipitated FLAG-TIF1βs are shown in the lower right panel (TIF1βs, IP). Supernatants after immunoprecipitation are also shown, as indicated (S). Page 6 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 phase, a reasonable question arises: Is the TIF1β/Ser473 phosphorylation correlated with cyclin A2's expression? PCR primers encompassing the E2F site were used to detect DNA fragments corresponding to the promoter region of cyclin A2. A much higher level of TIF1β was asso- ciated with the cyclin A2 promoter at the G1/S boundary than in early S phase (Figure 4A, upper panel). These results are closely correlated with the rise in phosphor- ylated TIF1β/Ser473 in the early S phase, but not at the G1/S boundary (Figure 4A, lower left panel). TIF1β is in Figure 4 TIF1β i i (D) The HP1β complex is preferentially associated with the promoters of Cdc2 and Cdc25A in TIF1β/S473A over-expressing interphase cells of HEK293T. FLAG-TIF1β, FLAG-TIF1β/S473A, FLAG- TIF1β/S473E and HA-HP-1β were ectopically expressed in HEK 293T cells for 36 hours and subjected to ChIP with the indicated antibodies (anti-tubulin, 10D8; anti-TIF1β, 20A1; anti-HA, HA). DNA from the immunoprecipitated chromatin was quantified by real-time PCR. Each result was normalized to input and the relative level to the transfected vector control. Results for the Cdc2 promoter (upper panels) and Cdc25A promoter (lower panels) are shown. TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A Figure 4 TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A. (A) Binding of TIF1β to the promoter of cyclin A2 gene is correlated with its repression. HeLa cells were synchronized to G1/S boundary by double thymidine block and subsequently released for 1 hour to the early S phase. ChIP was performed as described in the Materials and Methods section. Equal loadings of input derived from G1/S and S cells were used for ChIP experiments (upper panels), using 20A1, H3K4diMe, and H3K9diMe antibodies. PCR was performed with input samples (1/50) and ChIP eluates. Relative differences between G1/S and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (comparing the cyclin A2 signal of G1/S and S phase from each probe with the same total area, by ImageGauge soft- ware). The protein level of TIF1β (20A1), α-tubulin and phosphorylated TIF1β/Ser473 (S473) from formaldehyde cross-linked samples are shown in the lower left panel. PCR for cyclin E promoter was used as a control (upper panel). (B) HeLa cells were arrested at G1 phase by serum starvation and released into early S phase by addition of serum to the culture for 2 hours. ChIP was performed as in (A) with 20A1, anti-H3K4diMe and anti-H3K9diMe antibodies (upper panel). Relative differences between G1 and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (by ImageGauge, as in A), while the protein levels are shown in the lower left panel. (C) Flow cytom- etry analysis. Phosphorylation of TIF1β/Ser473 compromises interaction between HP1β and TIF1β Histone H3 p p y y p PCR primers encompassing the E2F site were used to detect DNA fragments corresponding to the promoter region of cyclin A2. A much higher level of TIF1β was asso- y p ( g pp p ) results are closely correlated with the rise in phosphor- ylated TIF1β/Ser473 in the early S phase, but not at the G1/S boundary (Figure 4A, lower left panel). Histone H3 TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A Figure 4 TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A. (A) Binding of TIF1β to the promoter of cyclin A2 gene is correlated with its repression. HeLa cells were synchronized to G1/S boundary by double thymidine block and subsequently released for 1 hour to the early S phase. ChIP was performed as described in the Materials and Methods section. Equal loadings of input derived from G1/S and S cells were used for ChIP experiments (upper panels), using 20A1, H3K4diMe, and H3K9diMe antibodies. PCR was performed with input samples (1/50) and ChIP eluates. Relative differences between G1/S and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (comparing the cyclin A2 signal of G1/S and S phase from each probe with the same total area, by ImageGauge soft- ware). The protein level of TIF1β (20A1), α-tubulin and phosphorylated TIF1β/Ser473 (S473) from formaldehyde cross-linked samples are shown in the lower left panel. PCR for cyclin E promoter was used as a control (upper panel). (B) HeLa cells were arrested at G1 phase by serum starvation and released into early S phase by addition of serum to the culture for 2 hours. ChIP was performed as in (A) with 20A1, anti-H3K4diMe and anti-H3K9diMe antibodies (upper panel). Relative differences between G1 and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (by ImageGauge, as in A), while the protein levels are shown in the lower left panel. (C) Flow cytom- etry analysis. FLAG-TIF1β/S473A, FLAG-TIF1β/S473E and wild-type FLAG-TIF1β were over-expressed in 293T cells, collected 36 hours post-transfection and analyzed by flow cytometry after staining with propidium iodide, and the FCM histograms are shown in the upper panels. Phosphorylation of TIF1β/Ser473 compromises interaction between HP1β and TIF1β Cell cycle phase distributions (%) of TIF1β over-expressing cells were analyzed with the CellQuest software and shown in the lower left panel, according to the gated region from the upper panel (M1, subG1; M2, G1; M3, S; and M4, G2/M, by Cel- lQuest program), and the bar charts represented the cell cycle distribution (in total for 100%) are shown in the lower right panel (by Microsoft Excel). (D) The HP1β complex is preferentially associated with the promoters of Cdc2 and Cdc25A in TIF1β/S473A over-expressing interphase cells of HEK293T. FLAG-TIF1β, FLAG-TIF1β/S473A, FLAG- TIF1β/S473E and HA-HP-1β were ectopically expressed in HEK 293T cells for 36 hours and subjected to ChIP with the indicated antibodies (anti-tubulin, 10D8; anti-TIF1β, 20A1; anti-HA, HA). DNA from the immunoprecipitated chromatin was quantified by real-time PCR. Each result was normalized to input and the relative level to the transfected vector control. Results for the Cdc2 promoter (upper panels) and Cdc25A promoter (lower panels) are shown. TIF1β is in Figure 4 TIF1β i i TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A Figure 4 TIF1β is involved in the regulation of expression of E2F downstream genes, cyclin A2, Cdc2 and Cdc25A. (A) Binding of TIF1β to the promoter of cyclin A2 gene is correlated with its repression. HeLa cells were synchronized to G1/S boundary by double thymidine block and subsequently released for 1 hour to the early S phase. ChIP was performed as described in the Materials and Methods section. Equal loadings of input derived from G1/S and S cells were used for ChIP experiments (upper panels), using 20A1, H3K4diMe, and H3K9diMe antibodies. PCR was performed with input samples (1/50) and ChIP eluates. Relative differences between G1/S and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (comparing the cyclin A2 signal of G1/S and S phase from each probe with the same total area, by ImageGauge soft- ware). The protein level of TIF1β (20A1), α-tubulin and phosphorylated TIF1β/Ser473 (S473) from formaldehyde cross-linked samples are shown in the lower left panel. PCR for cyclin E promoter was used as a control (upper panel). (B) HeLa cells were arrested at G1 phase by serum starvation and released into early S phase by addition of serum to the culture for 2 hours. ChIP was performed as in (A) with 20A1, anti-H3K4diMe and anti-H3K9diMe antibodies (upper panel). Relative differences between G1 and S phase of each indicated probe on cyclin A2 promoter are shown in the lower right panels (by ImageGauge, as in A), while the protein levels are shown in the lower left panel. (C) Flow cytom- etry analysis. FLAG-TIF1β/S473A, FLAG-TIF1β/S473E and wild-type FLAG-TIF1β were over-expressed in 293T cells, collected 36 hours post-transfection and analyzed by flow cytometry after staining with propidium iodide, and the FCM histograms are shown in the upper panels. Cell cycle phase distributions (%) of TIF1β over-expressing cells were analyzed with the CellQuest software and shown in the lower left panel, according to the gated region from the upper panel (M1, subG1; M2, G1; M3, S; and M4, G2/M, by Cel- lQuest program), and the bar charts represented the cell cycle distribution (in total for 100%) are shown in the lower right panel (by Microsoft Excel). Phosphorylation of TIF1β Ser473 in S phase is mediated by PKC pathway p y TIF1β/Ser473 phosphorylation is up-regulated in S phase. To identify which kinase(s) are involved in the phospho- rylation of TIF1β/Ser473 in the S phase, we used Western blotting of cell extracts prepared from cells that had been treated with a panel of kinase inhibitors. Among the inhibitors tested, only the PKC inhibitor (Ro-31-8820) exhibited significant inhibition, while the CK1 inhibitor had a moderate effect on the phosphorylation of Ser473 (Figure 5A). An inhibitory peptide (a substrate-mimetic fragment from aa 466–478 of TIF1β) that contained TIF1β/S471A/S473A also blocked TIF1β/Ser473 phos- phorylation (data not shown). Neither CaMK-II inhibitor nor staurosporine had any effect (Figure 5A). To confirm that the phosphorylation of TIF1β/Ser473 proceeds along the PKC pathway, the effect of 12-O-tetradecanoylphor- bol-13-acetate (TPA) on TIF1β/Ser473 phosphorylation in interphase HeLa cells was tested. TPA treatment induced dramatic phosphorylation of TIF1β/Ser473 within 1 hour (Figure 5B). To further examine whether subtype-PKCδ is involved in the phosphorylation of TIF1β/Ser473, M2-immunoprecipitation and Western blotting with S473 antibody were performed using cell extracts that were prepared from FLAG-TIF1β and HA- PKCδ-cotransfected 293T cells. The phosphorylation level of TIF1β/Ser473 was double that of the control (Figure 5C). Immunostaining also revealed an increase of the TIF1β/Phospho-Ser473 signal in HA-PKCδ-transfected HeLa cells (Figure 5D). Although the majority of over- expressed PKCδ was located in the cytoplasm, a minor portion appeared in the nucleus (Figure 5D, column 3). Collectively, these results indicate that TIF1β/Ser473 phosphorylation in S phase could be mediated by the PKCδ pathway, and that TIF1β/Ser473 phosphorylation correlated with the expression of target genes by weaken- Immunostaining showed that over-expression of FLAG- TIF1β, FLAG-TIF1β/S473A, and FLAG-TIF1β/S473E did not alter their nuclear localization (Figure 2D). A flow cytometric analysis was performed to test further the effects of over-expressed FLAG-TIF1β, FLAG-TIF1β/S473A and FLAG-TIF1β/S473E on cell cycle progression. Over- expression of the FLAG-TIF1β/S473A mutant caused an accumulation of cells stalled at G2/M in 293T cells (Figure 4C), while the profiles of cell cycle progression were sim- ilar in 293T cells over-expressing FLAG-TIF1β and FLAG- TIF1β/S437E, suggesting that the phosphorylation state of TIF1β/Ser473 affects cell cycle progression. Quantitative real-time PCR analysis of 293T cells revealed half the amount of cyclin A2 mRNA in the FLAG-TIF1β/S473A- overexpressing cells compared to the FALG-TIF1β or FALG-TIF1β/S473E-expressing cells (data not shown). TIF1β is in Figure 4 TIF1β i i FLAG-TIF1β/S473A, FLAG-TIF1β/S473E and wild-type FLAG-TIF1β were over-expressed in 293T cells, collected 36 hours post-transfection and analyzed by flow cytometry after staining with propidium iodide, and the FCM histograms are shown in the upper panels. Cell cycle phase distributions (%) of TIF1β over-expressing cells were analyzed with the CellQuest software and shown in the lower left panel, according to the gated region from the upper panel (M1, subG1; M2, G1; M3, S; and M4, G2/M, by Cel- lQuest program), and the bar charts represented the cell cycle distribution (in total for 100%) are shown in the lower right panel (by Microsoft Excel). (D) The HP1β complex is preferentially associated with the promoters of Cdc2 and Cdc25A in TIF1β/S473A over-expressing interphase cells of HEK293T. FLAG-TIF1β, FLAG-TIF1β/S473A, FLAG- TIF1β/S473E and HA-HP-1β were ectopically expressed in HEK 293T cells for 36 hours and subjected to ChIP with the indicated antibodies (anti-tubulin, 10D8; anti-TIF1β, 20A1; anti-HA, HA). DNA from the immunoprecipitated chromatin was quantified by real-time PCR. Each result was normalized to input and the relative level to the transfected vector control. Results for the Cdc2 promoter (upper panels) and Cdc25A promoter (lower panels) are shown. Page 7 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 K9diMe was associated with the silencing of gene expres- sion. The ChIP results demonstrated that the level of H3K9diMe correlated strongly with that of TIF1β. How- ever, histone H3K4diMe is known to associate with actively transcribed genes. Its association with the pro- moter region of cyclin A2 was inversely correlated with that of TIF1β. Quantitative ChIP results comparing the cyclin A2 level in G1/S and S phase are presented under the panel of PCR results obtained using each probe. As a con- trol, TIF1β was not associated with the promoter of cyclin E (Figure 4A, upper panel). TIF1β/phospho-Ser473 was also elevated 2 hours after serum was added (Figure 2C). ChIP was thus used to measure the level of TIF1β bound to the cyclin A2 promoter in G1-phase HeLa cells that had been starved of serum for three days or S-phase HeLa cells that had been released from serum starvation. More TIF1β in the G1 phase cells was associated with the promoter region of cyclin A2 (Figure 4B, upper panel). TIF1β is in Figure 4 TIF1β i i This finding is closely correlated with the elevated level of phosphor- ylated TIF1β/Ser473 in the S phase but not in the G1 phase (Figure 4B, lower left panel). Quantitative ChIP results, comparing cyclin A2 levels in the G1 and S phase, are presented beneath the PCR results obtained using each probe. Taken together, these results indicate that the asso- ciation of TIF1β with the promoter of cyclin A2 is corre- lated with silencing while dissociation is correlated with derepression of the promoter. To further examine the association of unphosphorylated TIF1β/Ser473 with other cell cycle-regulated genes, such as Cdc2 and Cdc25A, quantitative ChIP experiments were conducted with HEK293T cells that had been transfected with HA-HP1β and FLAG-TIF1β, FLAG-TIF1β/Ser473A, and FLAG-TIF1β/Ser473E. When ChIP was performed with HA monoclonal antibody, the association of HP1β with the promoters of Cdc2 or Cdc25A in FLAG-TIF1β/ Ser473A over-expressing cells was stronger than in FLAG- TIF1β/Ser473E-over-expressing cells (Figure 4D). When ChIP was performed using 20A1 (which recognizes the N- terminal of TIF1β), no obvious difference between FLAG- TIF1β/Ser473A and FLAG-TIF1β/Ser473E was observed. Collectively, these results reveal that over-expressed HP1β and TIF1β/Ser473A may form a stronger complex and preferentially associate with the promoter regions of Cdc2 and Cdc25A genes more than over-expressed HP1β and TIF1β/Ser473E in interphase HEK293T cells. Page 8 of 16 (page number not for citation purposes) Phosphorylation of TIF1β Ser473 in S phase is mediated by PKC pathway The phenotype of the G2/M stalls and the reduction of the cyclin A2 mRNA level in TIF1β/S473A over-expressing 293T cells are consistent with observations made in a GFP-cyclin A2 siRNA knockdown experiment published by Kenrick et al. [29]. Taken together, these results suggest that disruption of TIF1β/Ser473 phosphorylation may influence cell cycle-regulated gene expression, and that cyclin A2 may be one of the TIF1β indirect target genes. These data also demonstrate that the Ser473 phosphoryla- tion/dephosphorylation status of TIF1β may regulate cell cycle progression. Page 8 of 16 (page number not for citation purposes) Page 8 of 16 (page number not for citation purposes) BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 PKC is involved in the phosphorylation of TIF1β/Ser473 in early S phase Figure 5 PKC is involved in the phosphorylation of TIF1β/Ser473 in early S phase. (A) HeLa cells were synchronized with double thymidine block at G1/S phase and treated with various kinase inhibitors before being released for 1 hour to early S-phase (experimental procedures are depicted in the bottom scheme). Lanes 1–6, Western blot of whole cell extracts of cells arrested at G1/S boundary. Lanes 7–12, Western blot of early S-phase extracts. The inhibitors used were CaMK-II inhibitor KN93 (KN, 1 μM), PK Discussion This investigation demonstrates that the phosphorylation of TIF1β/Ser473 by kinase(s) (such as PKCδ) may act as a molecular switch in the temporal regulation of gene expression. TIF1β/Ser473 is located close to the HP1 box, the PXVXL motif, which is responsible for the interaction between TIF1β and HP1, and its phosphorylation nega- tively affects this interaction. The dynamic phosphoryla- tion of TIF1β/Ser473 suggests that this event may regulate the functions of TIF1β in cell cycle progression. The cen- tral question then is: How may the phosphorylation of TIF1β/Ser473 regulate TIF1β-mediated gene expression? We addressed this question by investigating the potential regulatory functions of TIF1β on key genes for cell cycle pregression and the effects of over-expressing phosphor- ylation-deficient or phosphomimetic mutants of TIF1β/ Ser473 on cell cycle progression. TIF1β is proposed to function as a universal corepressor protein for the KRAB zinc finger protein (KRAB-zfp) superfamily of transcrip- tional repressors [32]. The recruitment of HP1 proteins by the KRAB-TIF1β complex to specific loci within the genome through the formation of heterochromatin-like complexes may silence gene activity. The level of phosphorylated TIF1β/Ser473 is decreased in TPA-induced differentiated K562 cells Figure 6 The level of phosphorylated TIF1β/Ser473 is decreased in TPA-induced differentiated K562 cells. K562 cells were treated with 10 ng/ml TPA for 4 days. Phase contrast images of control (left panels, TPA-) and differenti- ated (right panels, TPA+) K562 cells are shown. Equal amounts of proteins from these cells were collected (Coomassie Blue stain) and TIF1β (20A1) or phosphorylated TIF1β/S473 (S473) were detected by Western hybridization. The relative TIF1β/S473 level of proliferating or TPA-induced differentiated K562 cells (by ImageGauge software, after nor- malization to individual TIF1β level) is shown. Bars in phase contrast images represent 40 μm (upper panels) and 8 μm (lower panels). PKC is inv Figure 5 p p y β y p g PKC is involved in the phosphorylation of TIF1β/Ser473 in early S phase. (A) HeLa cells were synchronized with double thymidine block at G1/S phase and treated with various kinase inhibitors before being released for 1 hour to early S-phase (experimental procedures are depicted in the bottom scheme). Lanes 1–6, Western blot of whole cell extracts of cells arrested at G1/S boundary. Lanes 7–12, Western blot of early S-phase extracts. The inhibitors used were CaMK-II inhibitor KN93 (KN, 1 μM), PKC inhibitor Ro-31-8220 (Ro, 100 nM), CK1 inhibitor D4476 (CK-I, 10 μM,), staurosporin (St, 20 nM), and DMSO as a control. The levels of TIF1β (20A1), phosphorylated TIF1β/ Ser473 (S473) and α-tubulin are shown. Quantitative results of the Western blots in early S phase (by ImageGauge software) showing the relative TIF1β/Ser473 phosphorylation level in various kinase inhibitor treatment, after normalization to individual TIF1β level. (B) Phosphorylated TIF1β/Ser473 was highly elevated in TPA-treated HeLa cells. Inter- phase HeLa cells were treated with 100 ng/ml or 250 ng/ml TPA for 1 hour, and whole cell extracts were prepared for Western blot. (C) Phosphorylation of TIF1β/Ser473 in 293T cells was highly induced by over-expressed HA-PKCδ. 293T cells were cotransfected with 1 μg of FLAG-TIF1β plasmid and HA-PKCδ plasmids (1 or 2 μg). FLAG-TIF1β was immunoprecipitated by M2-beads 36 hours post-transfection and probed with antibody to phosphorylated TIF1β/Ser473. Quantitative results (by ImageGauge software) show the relative TIF1β/S473 phosphorylation level in the presence of various amounts of transfected PKCδ plasmid, after normalization to individual TIF1β level. (D) Phospho- rylation of TIF1β/Ser473 was specifically increased in HA-PKCδ transfected HeLa cells. HA-PKCδ transfected HeLa cells were fixed 36 hours post-transfection and co-stained with monoclonal anti-HA antibody (red) and TIF1β/S473 antibody (green). DNA was visualized by counter staining with DAPI (blue). Different maginification of images are shown in columns 1–3. The localization of HA-PKCδ or TIF1β in the nucleus is indicated (column 3, arrow). Bars in D represent: column 1, 40 μm; column 2, 20 μm; and column 3, 7 μm. Page 9 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 ing TIF1β/HP1β interaction (and heterochromatin forma- tion). in proliferating cells (Figure 6, CB stain and Western blot), the level of phosphorylated TIF1β/Ser473 was halved in megakaryocytic differentiated K562. Level of phosphorylated TIF1β/Ser473 is reduced in megakaryocytic differentiated K562 cells Level of phosphorylated TIF1β/Ser473 is reduced in megakaryocytic differentiated K562 cells TIF1β/Ser473 phosphorylation fluctuates during cell cycle progression. K562 cells were used as an inducible cell dif- ferentiation system to profile the level of phosphorylated TIF1β/Ser473 in both proliferating and differentiated cells. K562 is an erythroleukemic cell line that can undergo further differentiation in both megakaryocytic and erythroid lineages, depending on the stimulus [30,31]. It is a well-characterized system in which TPA stimulates the megakaryocytic development of K562. Western blotting was performed on extracts prepared from proliferating or TPA-induced differentiated K562 cells. The differentiated K562 cells exhibited megakaryo- cytic morphology (with lobulated nucleus, highly vacu- olated, and histiocytoid), as demonstrated by phase contrast images (Figure 6). Although the total TIF1β pro- tein level in differentiated K562 cells was lower than that PKC is inv Figure 5 Consistent with the observation that TIF1β promotes cell cycle progression by the up-regulation of its Ser473 phosphorylation, the amount of TIF1β/Ser473 phosphorylation may be utilized as a proliferation marker in cancer cells. Page 10 of 16 (page number not for citation purposes) The phosphorylation of TIF1β/S473 compromises its interaction with HP1β in a manner that is related to cell cycle-regulated gene expression The observation that HP1β interacted strongly with un-phosphorylated TIF1β/Ser473 is consistent with the ChIP results concerning the over-expressed recom- binant variants, where over-expressed FLAG-TIF1β/S473A associated better with the promoter region of Cdc2 or Cdc25A than that did FLAG-TIF1β/S473E (Figure 4D). differentiation of primitive endoderm-like cells [12]. TIF1β is known to interact differentially with HP1β and HP1γ in differentiated and non-differentiated cells [33]. In non-differentiated cells, TIF1β/HP1 interaction occurs only in euchromatin and selectively involves HP1β and HP1γ, but not HP1α. In differentiated cells, on the other hand, TIF1β selectively associates with HP1β in hetero- chromatin, while TIF1β and HP1γ interaction occurs only in euchromatin. These conclusions agree with the reduced level of phosphorylated TIF1β/Ser473 seen here in differ- entiated K562 cells (Figure 6). The results herein also revealed that un-phosphorylated TIF1β/Ser473 interacts more strongly with HP1β than its phosphorylated coun- terpart (Figure 4B). These results further suggest that the phosphorylation of TIF1β/Ser473 regulates the differen- tial interaction between TIF1β and HP1β. The treatment of cells with cyclin A2 siRNA led to an accu- mulation of cells in prophase and mitosis to a degree sim- ilar to that observed for cyclin B1, consistent with the requirement of cyclin A for G1/S and G2/M transitions [29]. Interestingly, the ChIP results (Figures 4A and 4B) and the effects of over-expressed FLAG-TIF1β/Ser473A on cell cycle progression (accumulation of G2/M cells, Figure 4C) are consistent with published results for the siRNA knockdown of Cyclin A2 [29]. What may be the functional consequences of the phos- phorylation of TIF1β/Ser473 and its consequent dissocia- tion from HP1s must be addressed. TIF1β interacts with E2F [16], TRIP-Br and CBP/p300, and potentiates the co- activation of E2F-1/DP-1 by TRIP-Br protein [34]. Phos- phorylated TIF1β/Ser473 may thus be involved in this tri- partite functional interaction. This suggestion is supported by our findings that the induction of cyclin A2 is accompanied by an increased level of phosphorylated TIF1β/Ser473 and reduced binding of TIF1β to the pro- moter. A most provocative question that remains to be answered is whether the Ser473-phosphorylated TIF1β may interact preferentially with transcription factors and serve as a coactivator. HP1β recruitment to E2F-binding element of Cdc2 and Cdc25A promoters was affected by the phosphorylation state of TIF1β/Ser473. The level of HP1β recruitment to Cdc2 or Cdc25A promoter was increased when wild-type FLAG-TIF1β or FLAG-TIF1β/S473A were ectopically expressed. PKC mediated TIF1β/S473 phosphorylation may be involved in cell cycle progression Numerous examples have demonstrated that PKCδ is cen- trally involved in cell proliferation and differentiation. The activation of PKCδ uniquely mediates insulin- induced proliferation: PKCδ is activated by insulin and interacts with insulin receptor and IRS [36-38]. Insulin- activated PKCδ interacts with 3-phosphoinositide- dependent protein kinase to regulate Protein Kinase B The phosphorylation of TIF1β/S473 compromises its interaction with HP1β in a manner that is related to cell cycle-regulated gene expression This association was compromised by the phosphomimetic mutant, S473E, which suggests that HP1β recruitment is negatively regulated by phosphoryla- tion of TIF1β/Ser473. Likewise, ectopically expressed HP1β resulted in elevated recruitment of wild-type FLAG- TIF1β or FLAG-TIF1β/S473A. These observations provide a clue that the recruitment of TIF1β and HP1β could work in a positive feedback manner. The above results demonstrate that the dynamic intercon- version between unphosphorylated and phosphorylated forms of TIF1β/Ser473 may be crucial in the regulation of gene expression during cell cycle progression. Despite extensive efforts to perform ChIP with rabbit anti-phos- phorylated TIF1β/Ser473 antibody, no significant result was obtained. A likely explanation of this failure is that phosphorylated TIF1β/Ser473 is not associated with the promoter of cyclin A2 in the G1 phase or, alternatively, the epitope in the multiprotein complex is masked from anti- body access. The results also demonstrated that the level of phosphor- ylated TIF1β/Ser473 peaks at early S- and M-phases. Two phosphorylation sites other than Ser473, Ser752 and Ser757, were also identified in the course of this investiga- tion. Both sites are located in the bromodomain of TIF1β. The phosphorylation-deficient or phosphomimetic mutants of Ser757 did not influence the phosphorylation of Ser473 (Chang, unpublished results), suggesting that these phosphorylations may be independent events dur- ing mitosis. Other phosphorylation sites important for regulating the function of TIF1β have also been identified [35]. The functional consequences of Ser752 and Ser757 phosphorylation remain to be investigated. The way in which TIF1β disassociates from the HP1-con- taining heterochromatin is unclear. However, the findings herein provide a molecular explanation, which involves PKC-mediated phosphorylation at TIF1β/Ser473 (Figure 5). The phosphorylation of TIF1β/S473 compromises its interaction with HP1β in a manner that is related to cell cycle-regulated gene expression The disruption of the interaction between HP1β and TIF1β by the phosphorylation of TIF1β/Ser473 (Figure 3) suggests that this phosphorylation/dephosphorylation of TIF1β/Ser473 may be the means of regulation of TIF1β- and HP1β-mediated gene silencing. The results of ChIP experiments suggest that the majority of TIF1β associated with the promoters of cyclin A2 in G1 phase cells is likely to be unphosphorylated TIF1β/Ser473. This conclusion is supported by two lines of evidence: (1) very low levels of phosphorylated TIF1β/Ser473 were observed in G1 cells, and (2) overexpressed FLAG-TIF1β/S473A bound to the promoters of Cdc2 and Cdc25A better than FLAG-TIF1β/ S473E. When cells were released into the S phase, the association of unphosphorylated TIF1β/Ser473 with these promoters decreased, accompanying an increased level of phosphorylated TIF1β/Ser473. The dynamics of TIF1β/ Ser473 phosphorylation and TIF1β-binding to the cyclin A2 promoter (during the G1 to S phase progression) indi- cate that un-phosphorylated TIF1β/Ser473 is responsible for silencing the cyclin A2 gene in the G1 phase (Figures 4A The level o TPA-induce Figure 6 The level of phosphorylated TIF1β/Ser473 is decreased in TPA-induced differentiated K562 cells. K562 cells were treated with 10 ng/ml TPA for 4 days. Phase contrast images of control (left panels, TPA-) and differenti- ated (right panels, TPA+) K562 cells are shown. Equal amounts of proteins from these cells were collected (Coomassie Blue stain) and TIF1β (20A1) or phosphorylated TIF1β/S473 (S473) were detected by Western hybridization. The relative TIF1β/S473 level of proliferating or TPA-induced differentiated K562 cells (by ImageGauge software, after nor- malization to individual TIF1β level) is shown. Bars in phase contrast images represent 40 μm (upper panels) and 8 μm (lower panels). K562 cells were treated with 10 ng/ml TPA for 4 days. Phase contrast images of control (left panels, TPA-) and differenti- ated (right panels, TPA+) K562 cells are shown. Equal amounts of proteins from these cells were collected (Coomassie Blue stain) and TIF1β (20A1) or phosphorylated TIF1β/S473 (S473) were detected by Western hybridization. The relative TIF1β/S473 level of proliferating or TPA-induced differentiated K562 cells (by ImageGauge software, after nor- malization to individual TIF1β level) is shown. Bars in phase contrast images represent 40 μm (upper panels) and 8 μm (lower panels). Page 10 of 16 (page number not for citation purposes) Page 10 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 and 4B). Page 11 of 16 (page number not for citation purposes) Methods Antibodies Monoclonal antibody against TIF1β was generated by immunizing BALB/c mice with recombinant TIF1β corre- sponding to the N-terminal region (amino acids 1–250). Clone 20A1 was used throughout this investigation. Phos- pho-specific polyclonal antibody (S473) was produced by immunizing rabbits with KLH-conjugated peptide (VKRSRpSGEGEVC). The S473 antibody was purified by peptide-agarose affinity column chromatography. Rabbit anti-H3K9diMe and H3K4diMe antibodies were obtained from Upstate/Millipore, cyclin A and cyclin B antibodies were from BD Science, and M2 monoclonal antibody and M2 beads were from Sigma. Phosphorylated TIF1β/S473 level may serve as a proliferation marker TIF1β transcriptional repression activity depends on the interaction between TIF1β and HP1β [11]. This interac- tion is essential to the relocation of TIF1β from euchroma- tin to heterochromatin that accompanies the Page 11 of 16 (page number not for citation purposes) Page 11 of 16 (page number not for citation purposes) BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 with HP1β. Thus, phosphorylation of TIF1β/Ser473 plays a crucial role in epigenetic regulation of gene expression. with HP1β. Thus, phosphorylation of TIF1β/Ser473 plays a crucial role in epigenetic regulation of gene expression. [39] and is responsible for STAT3 activation and keratino- cyte proliferation [40]. Although the over-expressed PKCδ is mainly located in the cytoplasm (Figure 5D), it has been shown here (Figure 5D) and by Kajimoto et al. to partially localize to the nucleus [41]. The observation herein that the serum-stimulated phosphorylation of TIF1β/Ser473 correlates strongly with G1/S progression and cyclin A2 expression (Figure 4) uncovers a novel mechanism of PKCδ-mediated G1 to S phase cell cycle progression. Conclusion Although the TIF1β co-repressor function is known to be related to HP1β, few studies have addressed the specific gene targets of TIF1β. We have found that cell cycle pro- gression is regulated by TIF1β. The key cell cycle regula- tory genes, Cyclin A2, Cdc2 and Cdc25A are targeted by TIF1β, and phosphorylation of TIF1β/Ser473 is associated with the activation of these genes. TIF1β/S473 phosphor- ylation is up-regulated during the S phase in HeLa cells and the interaction between HP1β and TIF1β is compro- mised when Ser473 is phosphorylated. These observa- tions suggest that the phosphorylation of TIF1β/Ser473 may regulate gene expression by abolishing its interaction Phosphorylation of TIF1β/Ser473, gene activation and stem cell proliferation The dynamic phosphorylation of TIF1β/Ser473 during cell proliferation and differentiation (Figures 2 and 6) suggests that phosphorylation/dephosphorylation is cru- cial for regulating the transcriptional activity of TIF1β. During the differentiation of embryonic carcinoma cells, the intracellular distribution of TIF1β changes from dif- fuse nuclear staining to discrete foci and colocalizes with heterochromatin [42]. The steady-state level of TIF1β is also decreased. The reduced levels of phosphorylated TIF1β/Ser473 as well as the lower steady-state TIF1β levels in differentiated cells suggest that TIF1β may be mainly localized to heterochromatin in differentiated cells. In embryonic stem cells, TIF1β is present in complexes with various pluripotent markers, including Rex-1, Dax-1 and Nanog [22]. Since phosphorylated TIF1β/Ser473 seems to be preferentially associated with cell proliferation, it is important to determine whether it resides in these com- plexes. In fact, the level of phosphorylated TIF1β/Ser473 may serve as a proliferation marker. The epigenetic silenc- ing of retrovirus transcription is caused by the binding of a TIF1β corepressor complex to retrovirus primer binding site [17]. The likely recruitment of TIF1β by a DNA-bind- ing KRAB-box containing zinc finger protein for PBS- mediated silencing should be addressed, and the question of whether the regulation of the phosphorylation of TIF1β/Ser473 or the formation of TIF1β corepressor com- plex is participating in retrovirus transcription should also be investigated. Forward: 5'-GAAACGGTCCCGCGAAGGTGAGGG-3' and Reverse: 5'-CCCTCACCTTCGCGGGACCGTTTC-3'. HA-PKCδ was obtained from Dr. C. K. Chou, HA-HP1β and GST-HP1β plasmids were kindly provided by Dr. Pierre Chambon. PKC inhibitor Ro-31-8820 and calcium/ calmodulin-dependent protein kinase-II (CaMK-II) inhib- itor KN-93 were from BIOMOL. Casein kinase1 (CK1) inhibitor D4476 was from Calbiochem. Staurosporin, 12- O-tetradecanoylphorbol-13-acetate (TPA), thymidine, and nocodazole were from Sigma. Ser473-specific inhibi- tory peptide SGVKRARAGEGEVrrrrrrrrr (r stands for arginine) was obtained from Genemed Synthesis (South San Francisco, CA). Reverse: 5'-CCCTCACCTGCGCGGGACCGTTTC-3'. Reverse: 5'-CCCTCACCTGCGCGGGACCGTTTC-3'. Reverse: 5'-CCCTCACCTGCGCGGGACCGTTTC-3'. Plasmids and chemicals Mouse pCMV-FLAG-TIF1β was cloned as described by Chang et al. [43]. Site-directed mutagenesis was per- formed using pCMV-FLAG-TIF1β to create S473A and S473E mutants with PfuTurbo DNA Polymerase (Strata- gene). The primers, based on nucleotides 1691 to 1721 of NM_005762, were designed as follows. For S473A: Forward: 5'-GAAACGGTCCCGCGCAGGTGAGGG-3' and Forward: 5'-GAAACGGTCCCGCGCAGGTGAGGG-3' and Chromatin immunoprecipitation assays Chromatin immunoprecipitation (ChIP) assays were per- formed as described by Li et al. [44]. Briefly, synchronized G1, G1/S, early S phase, or interphase cells were treated with 1.42% of formaldehyde for 10 min at room temper- ature. Nuclei from 1 × 107 cells were resuspended in ChIP lysis buffer (50 mM Tris-HCl [pH 8.1], 1% SDS, 10 mM EDTA, 1× protease inhibitor cocktail) and used for each immunoprecipitation. After sonication on ice four to six times for 10 seconds followed by centrifugation for 10 min, the chromatin solution was diluted 10-fold with dilution buffer (5 μg/ml of salmon sperm DNA, 5 mg/ml BSA, 20 mM Tris-HCl [pH 8.1], 1% Triton X-100, 2 mM EDTA, 150 mM NaCl, 1× protease inhibitor cocktail). An input control of 100 μl of sonicated solution was saved and processed in parallel with the eluted immunoprecip- itates beginning at the cross-link reversal step. The chro- matin was pre-cleaned with protein G-agarose. Different antibodies (TIF1β monoclonal antibody, 20A1; rabbit anti-H3K9diMe and H3K4diMe antibodies) were bound to protein G-agarose first in dilution buffer containing 5 μg/ml of salmon sperm DNA and 5 mg/ml BSA. Chroma- tin complexes were then incubated with specific antibod- ies-protein G-agarose and rotated for 2–4 hours at 4°C. Immunoprecipitates were sequentially washed for 5 to 10 min in wash buffer I (20 mM Tris-HCl [pH 8.1], 2 mM EDTA, 0.1% SDS, 1% Triton X-100, 150 mM NaCl), wash buffer II (20 mM Tris-HCl [pH 8.1], 2 mM EDTA, 0.1% SDS, 1% Triton X-100, 500 mM NaCl), wash buffer III (0.25 M LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl [pH 8.1]), and then in TE buffer (three times). Washed beads were incubated at 65°C overnight to reverse protein-DNA cross-linking, and then each sam- ple was treated with 10 μg of proteinase K in proteinase K buffer for 6 h at 55°C. The eluted material was combined in one tube, and the DNA was purified with the QIAquick PCR purification kit (Qiagen, Valencia, Calif.28104) and eluted in 50 μl of elution buffer. Extracted total inputs were diluted 1:50 and subjected to PCR analysis with indi- cated primers. Each PCR mixture contained 2 μl of immu- noprecipitate or input, 1 μM each primer, 0.4 mM deoxynucleoside triphosphate mixture, 1× LA-Taq PCR buffer, and 0.25 μl LA-TaqDNA polymerase in a total vol- ume of 50 μl. Cell cultures HeLa, HEK293, and HEK293T cells were cultured in Dul- becco's modified Eagle's medium plus 10% FCS and 100 units/ml penicillin/streptomycin. For synchronization at mitotic phase, HeLa cells were treated with 1 μM nocoda- zole for 16 hours. Cells were collected by shake-off, rinsed with PBS and cultured in complete medium. For synchro- Page 12 of 16 (page number not for citation purposes) Page 12 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 HP1β were prepared by TnT in vitro transcription/transla- tion kit (Promega). nization at G1/S phase, HeLa cells were treated with 2.5 mM thymidine for 19 hours, released for 10 hours, treated for 19 hours again before release at various time points for cell cycle progression. For G1 phase synchronization, HeLa cells were serum-starved for 72 hours and then cul- tured in complete medium for various times. K562 cells were maintained in RPMI-1640 with 10% FCS and 100 units penicillin/streptomycin. K562 cells were induced to differentiate by treating with TPA (10 ng/ml) for 4 days [30,31]. nization at G1/S phase, HeLa cells were treated with 2.5 mM thymidine for 19 hours, released for 10 hours, treated for 19 hours again before release at various time points for cell cycle progression. For G1 phase synchronization, HeLa cells were serum-starved for 72 hours and then cul- tured in complete medium for various times. K562 cells were maintained in RPMI-1640 with 10% FCS and 100 units penicillin/streptomycin. K562 cells were induced to differentiate by treating with TPA (10 ng/ml) for 4 days [30,31]. Chromatin immunoprecipitation assays The E2F responsive primers for each pro- moter were as follows: Whole cell extracts (WCEs) were prepared by treating the cells with lysis buffer [20 mM HEPES, pH 7.4, 200 mM NaCl, 0.5% Triton-X100, 20% glycerol, 1 mM EDTA, 1 mM EGTA, 1 mM orthovanadate, and protease inhibitors (0.1 μg/ml each of aprotinin, leupeptin, pepstatin A, and 10 mM PMSF) and centrifuging at 6,000 × g in a microcen- trifuge for 30 min at 4°C. Nuclear fractions were prepared by lysing the cells in HEPES buffer (pH 7.6) containing 10 mM NaCl, 1.5 mM MgCl2, 20% glycerol, 0.2 mM EDTA, 0.1% Triton X-100 and protease inhibitors for 10 min, centrifuged at 1,250 × g for 5 min and washed once with the same buffer. Nuclear extracts were prepared in nuclear extraction buffer containing 25 mM Tris-HCl, pH 8.0, 0.5 M NaCl, 1 mM EDTA, 10% glycerol, 0.2% NP-40 and pro- tease inhibitors for 30 min at 4°C followed by DNase I treatment for 1 hour. For calf intestine alkaline phos- phatase (CIP) treatment, 293 T cells were lysed with CIP buffer (50 mM Tris-HCl, pH 8.0, 0.1 M NaCl, 10 mM MgCl2, and protease inhibitors). The supernatants were incubated with 30 U of CIP at 37°C for 30 min. Immunoprecipitation, recombinant proteins, in vitro pull- down, and western blotting WCE was pre-cleaned with protein G beads. Antibody was bound to protein G beads and mixed with pre-cleaned WCE for 2 hours at 4°C by gentle rotation. The immuno- complex was washed 3 times with immunoprecipitation buffer before SDS sample buffer was added and proteins were separated by SDS-PAGE. TIF1βs (FLAG-TIF1β/ S473A, FLAG-TIF1β/S473E, and wild type FLAG-TIF1β) were co-transfected with HA-HA-HP1β into 293T cells. Nuclear extracts were used for M2 pull-down assay. GST- HP1β expressed in Escherichia coli DH5α was purified by binding to glutathione beads. FLAG-tagged TIF1β, TIF1β/ S473A and TIF1β/S473E were expressed in 293T cells and purified by FLAG peptide elution of the M2 bead-bound proteins. For pull-down assays, glutathione bead-bound GST-HP1β was re-suspended in nuclear extraction buffer and incubated with the purified FLAG-TIF1β, FLAG- TIF1β/S473A, and FLAG-TIF1β/S473E. For Cyclin A2 promoter: Flow cytometry Cells were collected and fixed with 70% ethanol for 30 minutes at 0°C. Cells were stained with DNA-staining solution (25 μg/ml propidium iodide, 100 μg/ml RNase A, and 0.5% Nonidet P-40 in PBS) at room temperature for 30 min. DNA content was analyzed from 10,000 cells collected with a BD Biosciences flow cytometer in con- junction with the CellQuest software. The distribution of Authors' contributions CWC performed most experiments and participated in the writing of the manuscript. HYC and TCH contributed to pull-down experiments with in vitro translated proteins. YSL performed quantitative ChIP experiments with ectop- ically expressed plasmids. KHH and CJC contributed to the cloning of recombinant plasmids. SCL generated 20A1 and S473 antibodies. SCL supervised and conducted the study and wrote the manuscript. All authors read and approved the final manuscript draft. and for Cyclin E promoter: Forward: 5'-GCCGCCTGTCCATTCATCC-3'; Reverse: 5'-GGTCCTGTGGAGCCTGTAGCC-3'. and for Cyclin E promoter: cell populations in the cell cycle stage was gated with Cel- lQuest analysis program (M1, subG1; M2, G1; M3, S; and M4, G2/M). The cell cycle phase distribution (in total for 100%) in each sample after CellQuest program analysis is shown. The percentage of the cell cycle phase shown in table is also presented with bar charts (by Microsoft Excel). PCR was performed for 26 to 29 cycles with 1 min of denaturing at 94°C, annealing at 62°C, and extension at 68°C. PCR results were analyzed by agarose gel (1.5%) electrophoresis. Immunostaining Cells on cover slips were washed with 1× phosphate-buff- ered saline (PBS), fixed with 2% formaldehyde for 15 min, washed with cold PBS for three times and further permeablized with 1xPBS containing 0.5% Triton-X100 for 5 min. Cells were blocked with 1% BSA for 30 min and incubated with indicated antibodies diluted in 1% bovine serum albumin/PBS, and probed with indicated primary antibodies. Alexa 594-conjugated goat anti-mouse IgG (Molecular Probes, Inc) and Alexa 488-conjugated goat anti-rabbit IgG were used as secondary immunofluores- cent dyes. DAPI was used to visualize DNA. Stained cells were analyzed with a Leica TCS SP2 Confocal Spectral Microscope using a 63X/NA 1.4 oil immersion objective lens. RNA extraction and Real-time PCR The level of gene expression was determined by real-time PCR. Briefly, total RNA was extracted from cells using Blue extract reagent (LTK, Inc., Taiwan) following the proce- dures recommended by the manufacturer. Samples of 5 μg of total RNA were reverse transcribed using M-MLV reverse transcriptase (Promega) and an oligo dT primer. The primers used for real-time PCR were for Cyclin A2: For real time PCR, 5 μg of FLAG-TIF1β and HA-HP-1β expression vectors were transfected into HEK 293T cells in 10-cm dish at about 20% confluence. In each precipita- tion, 300 μg of total cell extract was used, with 50 μg of total cell extract as input. Antibodies for FLAG (M2) or HA were used to probe recombinant TIF1β and HP1β on E2F targets of Cdc2 and Cdc25A promoters. Real-time PCR was conducted on a Roche Light Cycler 480 using the follow- ing primers for Cdc2 promoter: Forward: 5'-GCATGTCACCGTTCCTCCTT-3'; Forward: 5'-GCATGTCACCGTTCCTCCTT-3'; Reverse: 5'-CAGGGCATCTTCACGCTCTAT-3'; Forward: 5'-ACAGTAGGACGACACTC-3'; and and for Actin: Reverse: 5'-GGATTCACCAATCGGGTAG-3'; Forward: 5'-GAGAAAATCTGGCACCACACC-3'; Reverse: 5'-ATACCCCTCGTAGATGGGCAC-3'. Forward: 5'-CTAGCTGCCATTCGGT-3'; and All reactions were carried out using a 7300 Real-Time PCR System (Applied Biosystems) and ABsolute™ QPCR SYBR® Green Mix (ABgene, Epsom, England). The amplification was carried out as follow: initial enzyme activation at 94°C for 15 min, then 40 cycles of 94°C for 15 s and 60°C for 1 min. A total of 50 ng of each diluted reverse transcription product was used for real-time PCR in a final volume of 25 μl containing 160 nM of each specific primer and 1× ABsolute™ QPCR SYBR® Green Mix (ABgene). The relative level of Cyclin A2 gene expression was calculated according to the comparative Ct method using the 2-ΔΔCTCT formula, using the expression of Actin as an endogenous control. Reverse: 5'-CTTCGCTGTTCTCCCA-3'. Forward: 5'-GAGAAAATCTGGCACCACACC-3'; and the following primers for Cdc25A promoter: Reverse: 5'-ATACCCCTCGTAGATGGGCAC-3'. Forward: 5'-CTGCTCAGTTTCCTTTGGTTTACC-3'; Reverse: 5'-CAAAGACGCCCAGAGATGCAG-3'; Forward: 5'-CTGCTCAGTTTCCTTTGGTTTACC-3'; Reverse: 5'-CAAAGACGCCCAGAGATGCAG-3'; Forward: 5'-CTGCTCAGTTTCCTTTGGTTTACC-3'; Reverse: 5'-CAAAGACGCCCAGAGATGCAG-3'; In vitro transcription and translation [35S]Methionine-labeled FLAG-TIF1βs (FLAG-TIF1β, FLAG-TIF1β/S473A, and FLAG-TIF1β/S473E) and HA- Page 13 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 References Brasher SV, Smith BO, Fogh RH, Nietlispach D, Thiru A, Nielsen PR, Broadhurst RW, Ball LJ, Murzina NV, Laue ED: The structure of mouse HP1 suggests a unique mode of single peptide recog- nition by the shadow chromo domain dimer. EMBO J 2000, 19:1587-1597. 26. Vigo E, Müller H, Prosperini E, Hateboer G, Cartwright P, Moroni MC, Helin K: CDC25A phosphatase is a target of E2F and is required for efficient E2F-induced S phase. Mol Cell Biol 1999, 19:6379-6395. 7. Eissenberg JC: Decisive factors: a transcription activator can overcome heterochromatin silencing. Bioessays 2001, 23:767-771. 8. Nielsen AL, Sanchez C, Ichinose H, Cerviño M, Lerouge T, Chambon P, Losson R: Selective interaction between the chromatin- remodeling factor BRG1 and the heterochromatin-associ- ated protein HP1alpha. EMBO J 2002, 21:5797-5806. 27. Henglein B, Chenivesse X, Wang J, Eick D, Bréchot C: Structure and cell cycle-regulated transcription of the human cyclin A gene. Proc Natl Acad Sci USA 1994, 91:5490-5494. g 28. Aleem E, Kiyokawa H, Kaldis P: Cdc2-cyclin E complexes regu- late the G1/S phase transition. Nat Cell Biol 2005, 7:831-836. 29 K k M H k S S bb S Th N GE H l h Th M 9. Thiru A, Nietlispach D, Mott HR, Okuwaki M, Lyon D, Nielsen PR, Hirshberg M, Verreault A, Murzina NV, Laue ED: Structural basis of HP1/PXVXL motif peptide interactions and HP1 localisa- tion to heterochromatin. EMBO J 2004, 23:489-499. 29. Kenrick M, Hancock S, Stubbs S, Thomas N, GE Healthcare, The May- nard Centre, Cardiff, UK: siRNA screening of the cell cycle with two dynamic GFP sensors. Discovery Matters 2005:18-19 [http:w4.gelifesciences.com/APTRIX/ upp00919.nsntenF6C9E078DD4E77C125704500085213?OpenDocu ment&Path=Catalog&Hometitle=Catalog&entry=1&newrel&Link Part=C1256FC4003AED4FADC4C4C717CAEC125706A003C9C32 _RelatedLinksNew]. J , 10. Vassallo MF, Tanese N: Isoform-specific interaction of HP1 with human TAFII130. Proc Natl Acad Sci USA 2002, 99:5919-5924. 11. Nielsen AL, Ortiz JA, You J, Oulad-Abdelghani M, Khechumian R, Gansmuller A, Chambon P, Losson R: Interaction with members of the heterochromatin protein 1 (HP1) family and histone deacetylation are differentially involved in transcriptional silencing by members of the TIF1 family. EMBO J 1999, 18:6385-6395. 30. Belhacène N, Maulon L, Guérin S, Ricci JE, Mari B, Colin Y, Cartron JP, Auberger P: Differential expression of the Kell blood group and CD10 antigens: two related membrane metallopepti- dases during differentiation of K562 cells by phorbol ester and hemin. FASEB J 1998, 12:531-539. 12. References 20. White DE, Negorev D, Peng H, Ivanov AV, Maul GG, Rauscher FJ 3rd: KAP1, a novel substrate for PIKK family members, colocal- izes with numerous damage response factors at DNA lesions. Cancer Res 2006, 66:11594-11599. 1. Le Douarin B, Nielsen AL, Garnier JM, Ichinose H, Jeanmougin F, Los- son R, Chambon P: A possible involvement of TIF1 alpha and TIF1 beta in the epigenetic control of transcription by nuclear receptors. EMBO J 1996, 15:6701-6715. 21. Cammas F, Mark M, Dollé P, Dierich A, Chambon P, Losson R: Mice lacking the transcriptional corepressor TIF1beta are defec- tive in early postimplantation development. Development 2000, 127:2955-2963. p J 2. Eissenberg JC, Elgin SC: The HP1 protein family: getting a grip on chromatin. Curr Opin Genet Dev 2000, 10:204-210. p 3. Wang G, Ma A, Chow CM, Horsley D, Brown NR, Cowell IG, Singh p 3. Wang G, Ma A, Chow CM, Horsley D, Brown NR, Cowell IG, Singh PB: Conservation of heterochromatin protein 1 function. Mol Cell Biol 2000, 20:6970-6983. 22. Wang J, Rao S, Chu J, Shen X, Levasseur DN, Theunissen TW, Orkin SH: A protein interaction network for pluripotency of embry- onic stem cells. Nature 2006, 444:364-368. 4. Grigoryev SA, Nikitina T, Pehrson JR, Singh PB, Woodcock CL: Dynamic relocation of epigenetic chromatin markers reveals an active role of constitutive heterochromatin in the transition from proliferation to quiescence. J Cell Sci 2004, 117:6153-6162. 23. Beausoleil SA, Jedrychowski M, Schwartz D, Elias JE, Villén J, Li J, Cohn MA, Cantley LC, Gygi SP: Large-scale characterization of HeLa cell nuclear phosphoproteins. Proc Natl Acad Sci USA 2004, 101:12130-12135. 24. Shu H, Chen S, Bi Q, Mumby M, Brekken DL: Identification of phosphoproteins and their phosphorylation sites in the WEHI-231 B lymphoma cell line. Mol Cell Proteomics 2004, 3:279-286. 5. Bártová E, Pacherník J, Kozubík A, Kozubek S: Differentiation-spe- cific association of HP1alpha and HP1beta with chromocen- tres is correlated with clustering ofTIF1beta at c these sites. Histochem Cell Biol 2007, 127:375-388. 25. Rayman JB, Takahashi Y, Indjeian VB, Dannenberg JH, Catchpole S, Watson RJ, te Riele H, Dynlacht BD: E2F mediates cell cycle- dependent transcriptional repression in vivo by recruitment of an HDAC1/mSin3B corepressor complex. Genes Dev 2002, 16:933-947. 6. References Cammas F, Oulad-Abdelghani M, Vonesch JL, Huss-Garcia Y, Cham- bon P, Losson R: Cell differentiation induces TIF1beta associa- tion with centromeric heterochromatin via an HP1 interaction. J Cell Sci 2002, 115:3439-3448. J 31. Rosson D, O'Brien TG: Constitutive c-myb expression in K562 cells inhibits induced erythroid differentiation but not tetra- decanoyl phorbol acetate-induced megakaryocytic differen- tiatio. Mol Cell Biol 1995, 15:772-779. J 13. Ryan RF, Schultz DC, Ayyanathan K, Singh PB, Friedman JR, Freder- icks WJ, Rauscher FJ 3rd: KAP1 corepressor protein interacts and colocalizes with heterochromatic and euchromatic HP1 proteins: a potential role for Krüppel-associated box-zinc fin- ger proteins in heterochromatin-mediated gene silencing. Mol Cell Biol 1999, 19:4366-4378. 32. Urrutia R: KRAB-containing zinc-finger repressor proteins. Genome Biol 2003, 4:231. 33. Cammas F, Janoshazi A, Lerouge T, Losson R: Dynamic and selec- tive interactions of the transcriptional corepressor TIF1 beta with the heterochromatin protein HP1 isotypes during cell differentiation. Differentiation 2007, 75:627-637. 14. Schultz DC, Ayyanathan K, Negorev D, Maul GG, Rauscher FJ 3rd: SETDB1: a novel KAP1-associated histone H3, lysine 9-spe- cific methyltransferase that contributes to HP1-mediated silencing of euchromatic genes by KRAB zinc-finger pro- teins. Genes Dev 2002, 16:919-932. 34. Hsu SI, Yang CM, Sim KG, Hentschel DM, O'Leary E, Bonventre JV: TRIP-Br: a novel family of PHD zinc finger- and bromodo- main-interacting proteins that regulate the transcriptional activity of E2F-1/DP-1. EMBO J 2001, 20:2273-2285. 15. Wang C, Ivanov A, Chen L, Fredericks WJ, Seto E, Rauscher FJ 3rd, Chen J: MDM2 interaction with nuclear corepressor KAP1 contributes to p53 inactivation. EMBO J 2005, 24:3279-3290. y J 35. Li X, Lee YK, Jeng JC, Yen Y, Schultz DC, Shih HM, Ann DK: Role for KAP1 serine 824 phosphorylation and sumoylation/des- umoylation switch in regulating KAP1-mediated transcrip- tional repression. J Biol Chem 2007, 282:36177-36189. p J 16. Wang C, Rauscher FJ 3rd, Cress WD, Chen J: Regulation of E2F1 function by the nuclear corepressor KAP1. J Biol Chem 2007, 282:29902-29909. p J 36. Jain N, Zhang T, Kee WH, Li W, Cao X: Protein kinase C delta associates with and phosphorylates Stat3 in an interleukin-6- dependent manner. J Biol Chem 1999, 274:24392-24400. 17. Wolf D, Goff SP: TRIM28 mediates primer binding site-tar- geted silencing of murine leukemia virus in embryonic cells. Cell 2007, 131:46-57. p J 37. Acknowledgements g We thank Drs Pierre Chambon (Institute of Genetics and Cellular and Molecular Biology, Strasbourg, France) for HP1 plasmids and C. K. Chou (Department of Life Science, Chang Gung University, Taiwan) for HA- Page 14 of 16 (page number not for citation purposes) Page 14 of 16 (page number not for citation purposes) http://www.biomedcentral.com/1471-2199/9/61 BMC Molecular Biology 2008, 9:61 PKCδ. This research was supported by a frontier science research grant from National Science Council of Taiwan (NSC96-2321-B002-008) and fund from the Institute of Biological Chemistry, Academia Sinica, Taiwan. PKCδ. This research was supported by a frontier science research grant from National Science Council of Taiwan (NSC96-2321-B002-008) and 19. Sripathy SP, Stevens J, Schultz DC: The KAP1 corepressor func- tions to coordinate the assembly of de novo HP1-demar- cated microenvironments of heterochromatin required for KRAB zinc finger protein-mediated transcriptional repres- sion. Mol Cell Biol 2006, 26:8623-8638. fund from the Institute of Biological Chemistry, Academia Sinica, Taiwan. fund from the Institute of Biological Chemistry, Academia Sinica, Taiwan. 20. Page 15 of 16 (page number not for citation purposes) References Novotny-Diermayr V, Zhang T, Gu L, Cao X: Protein kinase C delta associates with the interleukin-6 receptor subunit glyc- oprotein (gp) 130 via Stat3 and enhances Stat3-gp130 inter- action. J Biol Chem 2002, 277:49134-49142. 18. Li H, Rauch T, Chen ZX, Szabó PE, Riggs AD, Pfeifer GP: The his- tone methyltransferase SETDB1 and the DNA methyltrans- ferase DNMT3A interact directly and localize to promoters silenced in cancer cells. J Biol Chem 2006, 281:19489-19500. Page 15 of 16 (page number not for citation purposes) BMC Molecular Biology 2008, 9:61 http://www.biomedcentral.com/1471-2199/9/61 http://www.biomedcentral.com/1471-2199/9/61 38. Shen S, Alt A, Wertheimer E, Gartsbein M, Kuroki T, Ohba M, Braiman L, Sampson SR, Tennenbaum T: PKCdelta activation: a divergence point in the signaling of insulin and IGF-1-induced proliferation of skin keratinocytes. Diabetes 2001, 50:255-264. 39. Brand C, Cipok M, Attali V, Bak A, Sampson SR: Protein kinase Cδ participates in insulin-induced activation of PKB via PDK1. Biochem Biophys Res Commun 2006, 349:954-962. p y 40. Gartsbein M, Alt A, Hashimoto K, Nakajima K, Kuroki T, Tennen- baum T: The role of protein kinase C delta activation and STAT3 Ser727 phosphorylation in insulin-induced keratino- cyte proliferation. J Cell Sci 2006, 119:470-481. y p J 41. Kajimoto T, Ohmori S, Shirai Y, Sakai N, Saito N: Subtype-specific translocation of the delta subtype of protein kinase C and its activation by tyrosine phosphorylation induced by ceramide in HeLa cells. Mol Cell Biol 2001, 21:1769-1783. 42. Cammas F, Herzog M, Lerouge T, Chambon P, Losson R: Associa- tion of the transcriptional corepressor TIF1beta with hete- rochromatin protein 1 (HP1): an essential role for progression through differentiation. Genes Dev 2004, 18:2147-2160. 43. Chang CJ, Chen YL, Lee SC: Coactivator TIF1beta interacts with transcription factor C/EBPbeta and glucocorticoid receptor to induce alpha1-acid glycoprotein gene expres- sion. Mol Cell Biol 1998, 18:5880-5887. 44. Li X, Wong J, Tsai SY, Tsai MJ, O'Malley BW: Progesterone and glucocorticoid receptors recruit distinct coactivator com- plexes and promote distinct patterns of local chromatin modification. Mol Cell Biol 2003, 23:3763-3773. References Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 16 of 16 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge Page 16 of 16 (page number not for citation purposes)
https://openalex.org/W3159047120	https://link.springer.com/content/pdf/10.1007/s42995-021-00098-8.pdf	English	null	Optimizing the hybridization chain reaction-fluorescence in situ hybridization (HCR-FISH) protocol for detection of microbes in sediments	Marine life science & technology/Marine Life Science & Technology	2,021	cc-by	9,382	Abstract Fluorescence in situ hybridization (FISH) is a canonical tool commonly used in environmental microbiology research to visualize targeted cells. However, the problems of low signal intensity and false-positive signals impede its widespread application. Alternatively, the signal intensity can be amplified by incorporating Hybridization Chain Reaction (HCR) with FISH, while the specificity can be improved through protocol modification and proper counterstaining. Here we optimized the HCR-FISH protocol for studying microbes in environmental samples, particularly marine sediments. Firstly, five sets of HCR initiator/amplifier pairs were tested on the laboratory-cultured bacterium Escherichia coli and the archaeon Methano- coccoides methylutens, and two sets displayed high hybridization efficiency and specificity. Secondly, we tried to find the best combination of sample pretreatment methods and HCR-FISH protocol for environmental sample analysis with the aim of producing less false positive signals. Various detachment methods, extraction methods and formulas of hybridization buffer were tested using sediment samples. Thirdly, an image processing method was developed to enhance the DAPI signal of microbial cells against that of abiotic particles, providing a reliable reference for FISH imaging. In summary, our optimized HCR-FISH protocol showed promise to serve as an addendum to traditional FISH for research on environmental microbes. Fluorescence in situ hybridization · Hybridization chain reaction · HCR-FISH · Microbial detection Keywords Fluorescence in situ hybridization · Hybridization chain reaction · HCR-FISH · Microbial detection · Sediment e in situ hybridization · Hybridization chain reaction · HCR-FISH · Microbial detection · Sediment Marine Life Science & Technology (2021) 3:529–541 https://doi.org/10.1007/s42995-021-00098-8 RESEARCH PAPER Optimizing the hybridization chain reaction‑fluorescence in situ hybridization (HCR‑FISH) protocol for detection of microbes in sediments Zeyu Jia1 · Yijing Dong2 · Heng Xu2,3 · Fengping Wang1,4 Received: 23 September 2020 / Accepted: 17 February 2021 / Published online: 30 April 2021 © The Author(s) 2021 Received: 23 September 2020 / Accepted: 17 February 2021 / Published online: 30 April 2021 © The Author(s) 2021 Introduction Fluorescence in situ hybridization (FISH) is a widely used research tool in studying the environmental microbial com- munity (Yamaguchi and Kubota 2017). By labelling 16S rRNA, FISH can phylogenetically distinguish targeted microbes at the single-cell level. The applications of FISH in environmental research include quantification of specific microbial populations (Baptista et al. 2014; Buongiorno et al. 2017), cell-level exploration of spatial structure of microbial communities (Orcutt and Meile 2008; Wilen et al. 2008), and providing an indication of cell location for other high-resolution imaging techniques such as Nanoscale sec- ondary ion mass spectrometry (nanoSIMS), Bio-Orthogo- nal Non-Canonical Amino acid Tagging (BONCAT), and Raman microscopy (Chen et al. 2014; Hatzenpichler et al. 2016; Huang et al. 2007).l * Fengping Wang fengpingw@sjtu.edu.cn 1 State Key Laboratory of Microbial Metabolism, Joint International Research Laboratory of Metabolic and Developmental Sciences, School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China 2 School of Physics and Astronomy, Shanghai Jiao Tong University, Shanghai 200240, China In order to observe microbes with epi-fluorescence microscopes, the fluorescence signal of the targeted cells must be intensive and specific. High signal inten- sity ensures that targeted cells can be distinguished from 3 Institute of Natural Science, Shanghai Jiao Tong University, Shanghai 200240, China 4 School of Oceanography, Shanghai Jiao Tong University, Shanghai 200240, China (0121 3456789) 3 530 Marine Life Science & Technology (2021) 3:529–541 Hybridization chain reaction (HCR) is another sig- nal amplification method (Dirks and Pierce 2004). It is a strand displacement amplification method, where the nicked double-helix nucleotide strand is produced without the use of enzyme and a change of temperature. For example, in DNA HCR process, the presence of the initiator oligonu- cleotide triggers a repeating hybridization of two species of DNA hairpins. By combining the initiator with different type of molecules, such as aptamer (Bao et al. 2020; Jia et al. 2018; Zhang et al. 2020), antibodies (Choi et al. 2011), DNA probes (Yamaguchi et al. 2015b), nanoparticles (Zeng et al. 2019; Zhang et al. 2012), etc., the long-chain double- stranded DNA (dsDNA) can specifically locate the target molecule. Since the DNA hairpins can incorporate the fluo- rophore (Huang et al. 2011), nanoparticles (Gao et al. 2017; Miao et al. 2015; Wu et al. 2018), electrochemical indicator (Hou et al. Introduction 2015), etc., the amplified signal can be detected by diverse kinds of instruments, such as spectrophotometers, electrodes and transmission electron microscopes (Bi et al. 2017). background signals, and the high specificity ensures that untargeted cells and abiotic particles are not mis-recog- nized as targeted cells. The traditional FISH method usu- ally works well on highly active microbes on both signal intensity and specificity (Wilen et al. 2008). However, microbial cells in sediment are typically less active and/ or smaller in size than E. coli (Amann and Fuchs 2008; Orphan et al. 2009), with inadequate amounts of rRNA for traditional FISH to create signals intense enough to distinguish cells from the background (Fazi et al. 2007). In addition to using fluorophores with higher efficiency, a variety of signal amplification methods have been tried and combined with FISH to explore the microbial com- munities in sediments (Amann et al. 1995). One method is to design multiple probes targeting different loci of the same rRNA, thus strengthening the signal of each rRNA (Morris et al. 2002). Another method is to deploy a multi- labeled polynucleotide probe, e.g., using RNA transcripts from PCR amplicons of 16S and 23S rRNA genes, with multiple fluorescently labeled uridine incorporated dur- ing transcription (Pernthaler et al. 2002). Other methods have been borrowed from sensor technology, of which the catalyzed reporter deposition FISH (CARD-FISH) is the most well-reported (Kubota 2013). This method requires a DNA probe labeled with horseradish peroxidase (HRP). This enzyme can induce the reaction of hydroperoxide, fluorescence-labeled tyramide, and intracellular aromatic compounds, leading to a deposition of tyramide on the target site. Although this method significantly increases the fluorescence signal, some issues have restricted its application on sediment samples (Kubota 2013): the large molecular weight (~ 40 kDa) of HRP prevents the entrance of the probe into the cells, thus extra permeabilization should be undertaken prior to the hybridization; H2O2 is necessary for inactivation of intracellular peroxidase to avoid a false-positive result, however, this may degrade the nucleic acid (Massie et al. 1972). Of the described methods, the combination of HCR with FISH for bio-imaging was first reported in 2010 (Choi et al. 2010). In HCR-FISH, the aforementioned initiators in the HCR system are concatenated to the probes of tra- ditional FISH as initiator probes. Unlike the fluorescently labeled FISH probe, the initiator probes of HCR-FISH are not labeled. Modification of the HCR‑FISH protocol To verify the protocol, HCR-FISH was first tested on the model bacterium E. coli, targeted by the universal bacte- rial probe EUB338, following the procedure proposed by Yamaguchi et al. (2015a, b). In the original protocol, the concentration of initiator probe was 1 μmol/L, similar to that used in the traditional FISH. In these conditions, the signal of HCR-FISH on E. coli was too unclear to identify the exact location and shape of individual cells (Fig. 2a–c). The cell signal became more intensive and clearer when the probe concentration was increased from the original 1 μmol/L to 2.5 or 10 μmol/L (Fig. 2d–i). Other modifi- cations of the fixation procedure and hybridization state, including the temperature, formamide concentration and moisture level, etc., did not show much improvement when the concentration of the initiator probe remained at 1 μmol/L (data not shown). Therefore, the concentration of initiator probe in the hybridization buffer was set to 10 μmol/L for all further experiments. The average fluores- cence signal intensity of HCR-FISH on E. coli (8.85 ± 1.3 arbitrary unit, A.U.) was ~ 5 times that of traditional FISH (1.61 ± 0.58 A.U.) and there were no obvious differences between the cell counting results on E. coli by HCR- FISH ((6.74 ± 2.14) × 108 cells/ml) and by SYBR Green I staining ((7.56 ± 3.25) × 108 cells/ml) (Student’s t-test, n1 = n2 = 3, P > 0.1). The modified protocol also worked well on pure cultures of the archaeon Methanococcoides methylutens (Fig. 2j–l). While several groups have inde- pendently conducted the HCR-FISH protocols of Yama- guchi et al. (2015a, b), they did not arrive at a unified conclusion on the suitability of this protocol (Buongiorno et al. 2017; Francis et al. 2019; Grieb et al. 2020; Mat- subayashi et al. 2017; Royet et al. 2018). Therefore, it is likely that some experimental conditions suggested in the Yamaguchi et al. (2015a, b) protocols are only margin- ally suitable and may need further optimization to achieve robust performance. Here, it is proposed that increasing the concentration of the initiator probe in the hybridization buffer improves the original protocol. In this study, the method was validated by two relatively independent labs. While the signal intensity of FISH can be improved by the above methods, the specificity of FISH can be influenced by additional factors, especially in sediment samples. Modification of the HCR‑FISH protocol Firstly, if the hybridization condition is not stringent enough, the probe may bind to non-targeted RNAs with partially com- plementary sequences. If this occurs, cells without target RNA may also present a probe signal. Secondly, although base-pair matching does not occur, there is still a chance that the probe will be adsorbed by abiotic particles (Amalfitano and Fazi 2008; Daims and Wagner 2007). This would be more likely in HCR-FISH as more DNA probes are involved, increasing the possibility of DNA adsorption by abiotic particles. These kinds of problems prevent the application of HCR-FISH to sediment and soil samples. Indeed, one attempt to apply HCR-FISH to sediments was unsuccess- ful due to strong false-positive signals (Buongiorno et al. 2017). In this study, we first validated and optimized HCR- FISH on pure cultures of the bacterium Escherichia coli and the archaeon Methanococcoides methylutens. We found that the concentration of the initiator probe in the hybridi- zation buffer needed to be increased to 10 μmol/L and two sets of HCR sequences outcompeted others. Then, we tried to reduce the false positive rate of HCR-FISH on sediment samples. Modifications regarding pretreatment methods and hybridization reagents were validated for sediment samples. We also developed an image-processing method that could improve the performance of DAPI counter staining. Combin- ing these methods, we successfully visualized microbes in sediment with HCR-FISH, demonstrating HCR-FISH as a promising method for use in environmental microbial com- munity research Results and discussion of probe B, releasing a sequence identical to the initiator sequence that can also bind to probe A (Fig. 1, process III). Thus, the fluorescence-labeled amplifier probes A and B accumulate around the target sites in a form of elongated double-stranded DNA with a strong fluorescence signal (Fig. 1, process IV). Yamaguchi et al. (2015b) first applied HCR-FISH on environmental microbes, validating it by using it on active sludge samples. Nikolakakis et al. (2015) used HCR-FISH to observe the migration of syntrophic microbes on squids. Recently, Imachi et al. (2020) suc- cessfully enriched strains of Asgard archaea and docu- mented their shapes using HCR-FISH. There are several advantages of using HCR-FISH compared to CARD-FISH. Firstly, the probes in HCR-FISH are much smaller, and so can more easily penetrate the cells (Amann and Fuchs 2008). Secondly, the fluorophores are labeled on amplifier probes rather than initiator probes, thus the initiator probes can easily be changed with demand with only a small addi- tional cost of time and expense. Thirdly, not using H2O2 in the HCR-FISH protocol better preserves the target RNA in the sample (Massie et al. 1972). Fourthly, the HCR-FISH protocol is less time-consuming. Introduction Two amplifier probes, A and B, are also required for HCR-FISH, each of which is fluorescently labeled and has a stable hairpin structure with free tails on their 5′ terminal. At the beginning, initiator probes hybrid- ize with targeted intracellular RNA, leaving their initiator sequence unpaired (Fig. 1, process I). After the removal of excessive initiator probes, amplifier probes A and B can be added. The unpaired initiator sequence binds to the unpaired tail of probe A, linearizing the stem-loop structure (Fig. 1, process II). The released sequence on the stem-loop structure similarly binds to the unpaired tail Fig. 1 Mechanism of HCR-FISH. After the initiator probe hybridized to the target RNA (process I), amplifier probes could bind to the initiator step by step (process II, III, IV) (Dirks and Pierce 2004; Yamaguchi et al. 2015b) Fig. 1 Mechanism of HCR-FISH. After the initiator probe hybridized to the target RNA (process I), amplifier probes could bind to the initiator step by step (process II, III, IV) (Dirks and Pierce 2004; Yamaguchi et al. 2015b) 3 3 Marine Life Science & Technology (2021) 3:529–541 531 Test of HCR probe sets Observing phylogenetically distinct microbes simultane- ously allows more application of HCR-FISH. In order to achieve this goal, at least two sets of HCR sequences are 1 3 3 Marine Life Science & Technology (2021) 3:529–541 532 Fig. 2 HCR-FISH on pure cultured microbes. E. coli was labeled with EUB338 (a–i) and M. methylutens was labeled with ARCH915 (j–l), both with HCR probe set S1. The concentrations of initiator probe in hybridization solution are 1 μmol/L (panel a–c), 2.5 μmol/L (panel d–f), and 10 μmol/L (panel g–l), respectively. The micro- graphs depict DAPI stain (left) and those depict HCR-FISH signals (middle) are overlapped on the right Fig. 2 HCR-FISH on pure cultured microbes. E. coli was labeled with EUB338 (a–i) and M. methylutens was labeled with ARCH915 (j–l), both with HCR probe set S1. The concentrations of initiator probe in hybridization solution are 1 μmol/L (panel a–c), 2.5 μmol/L (panel d–f), and 10 μmol/L (panel g–l), respectively. The micro- graphs depict DAPI stain (left) and those depict HCR-FISH signals (middle) are overlapped on the right Fig. 2 HCR-FISH on pure cultured microbes. E. coli was labeled with EUB338 (a–i) and M. methylutens was labeled with ARCH915 (j–l), both with HCR probe set S1. The concentrations of initiator probe in hybridization solution are 1 μmol/L (panel a–c), 2.5 μmol/L (panel d–f), and 10 μmol/L (panel g–l), respectively. The micro- graphs depict DAPI stain (left) and those depict HCR-FISH signals (middle) are overlapped on the right and unclear signals, similar to the results of the S1 probe using a probe concentration of 1 μmol/L (Supplementary Fig. S1a, b). It is possible that the shorter length gave S1 and S2 a better opportunity to enter the cells and interact with their targeted RNA, while a longer length made L1 and L2 more likely to be influenced by steric factors. Based on these results, experiments with L1 and L2 were discontinued. The S3 probe set resulted in many nonspecific binding signals outside the cells (Supplementary Fig. S1c), consistent with a previous study (Buongiorno et al. 2017). Comparatively, the presence of S1 and S2 signals were well limited in the neigh- borhood of DAPI signal, showing high specificity (Fig. 2g–i, Supplementary Fig. S1d). It’s possible that the sequence of S3 itself tended to trigger nonspecific binding. At last, probe S1 and S2 were selected for further study because of their high signal intensity, clearness, and specificity. Test of HCR probe sets required and they should meet the following criteria: (a) these sequences should be orthogonal, meaning that they won’t hybridize with each other; and (b) using identical pro- tocols, they should give an acceptable performance under the same working conditions, i.e., they will give a strong signal with the targeted microbes and weak signals with the untargeted materials. Based on previous studies (Choi et al. 2014; Yamaguchi et al. 2015b), five orthogonal HCR sequence sets named S1/S2/S3/L1/L2 (Table 1) were chosen and tested on E. coli. Those that worked well under the mod- ified protocol were selected for further experiments. Among them, S1, S2, and S3 were shorter in length, with initiator sequences consisting of 26 nucleotides, while those of L1 and L2 had 36 nucleotides each. Each initiator sequence was fused with bacterial universal probe EUB338. It was revealed that the shorter probe sets, S1 and S2, gave stronger fluorescence signals, while L1 and L2 emitted only vague 1 3 3 Marine Life Science & Technology (2021) 3:529–541 533 Table 1 Probes used in this study Probe name Probe sequence (5′-3′) Reference S1a TCTAGTCGTTGATGCTTTGTATTCGGCGACAGATAACCGAATACAAAGCATC Choi et al. (2010) S1b CCGAATACAAAGCATCAACGACTAGAGATGCTTTGTATTCGGTTATCTGTCG Choi et al. (2010) S2a CATAGGGTTCGGATTCTTAGGGCGTAGCAGCATCAATACGCCCTAAGAATCC Choi et al. (2010) S2b TACGCCCTAAGAATCCGAACCCTATGGGATTCTTAGGGCGTATTGATGCTGC Choi et al. (2010) S3a ATGAAGGACGGACTACTGATAACTGGGACTTCCATACCAGTTATCAGTAGTC Choi et al. (2010) S3b CCAGTTATCAGTAGTCCGTCCTTCATGACTACTGATAACTGGTATGGAAGTC Choi et al. (2010) L1a GAAGCGAATATGGTGAGAGTTGGAGGTAGGTTGAGGCACATTTACAGACCTCAACCT ACCTCCAACTCTCAC Choi et al. (2014) L1b CCTCAACCTACCTCCAACTCTCACCATATTCGCTTCGTGAGAGTTGGAGGTAGGTTG AGGTCTGTAAATGTG Choi et al. (2014) L2a CGGGTTAAAGTTGAGTGGAGATATAGAGGCAGGGACAAAGTCTAATCCGTCCCTGCC TCTATATCTCCACTC Choi et al. (2014) L2b GTCCCTGCCTCTATATCTCCACTCAACTTTAACCCGGAGTGGAGATATAGAGGCAGG GACGGATTAGACTTT Choi et al. (2014) ARCH915-S1 CCGAATACAAAGCATCAACGACTAGAAAAAAGTGCTCCCCCGCCAATTCCT Yamaguchi et al. (2015a, b) ARCH915-S3 CCAGTTATCAGTAGTCCGTCCTTCATTTTTTGTGCTCCCCCGCCAATTCCT This study EUB338-S1 CCGAATACAAAGCATCAACGACTAGAAAAAAGCTGCCTCCCGTAGGAGT Yamaguchi et al. (2015a, b) EUB338-S2 TACGCCCTAAGAATCCGAACCCTATGAAAAAGCTGCCTCCCGTAGGAGT This study EUB338-S3 CCAGTTATCAGTAGTCCGTCCTTCATTTTTTGCTGCCTCCCGTAGGAGT This study EUB338-L1 GTCCCTGCCTCTATATCTCCACTCAACTTTAACCCGAAAAAAGCTGCCTCCCGTAGG AGT This study EUB338-L2 CCTCAACCTACCTCCAACTCTCACCATATTCGCTTCAAAAAAGCTGCCTCCCGTAGG AGT This study ANME-1-350-S1 CCGAATACAAAGCATCAACGACTAGAAAAAAAGTTTTTCGCGCCTGATGC This study SEEP2-658-S2 TACGCCCTAAGAATCCGAACCCTATGAAAAATCCACTTCCCTCTCCGGT This study Decreasing the negative effects of abiotic particles on HCR‑FISH with detergent to wash the cells off the abiotic particles, while method B uses an ultrasonic probe (Kallmeyer et al. 2008). Both methods were evaluated by counting the total cell number in their supernatant after an hour of precipi- tation. The cell counting results obtained using methods A, B and that result of the control group were 5.70 ± 1.89, 3.20 ± 0.52, 2.80 ± 0.25 per nanoliter, respectively. Com- pared with the untreated sample, more cells were obtained in the supernatant through method A (Student’s t-test, n1 = n2 = 5, P < 0.01), whereas method B had no obvious effect (Student’s t-test, n1 = n2 = 5, P > 0.1). Through sorption of probes and fluorophores, abiotic parti- cles in sediment significantly interfere with the performance of traditional FISH and HCR-FISH. Furthermore, large quantities of abiotic particles can cover the microbes during cell embedding (on a filter membrane), causing problems for observation. These negative effects can be alleviated by reducing either the number of abiotic particles or the adsorp- tion of abiotic particles on the probes. To reduce these nega- tive effects several methods for improving the performance of FISH on sediments were tested. f For the separation step, a multi-layer density gradient centrifugation method was tested on samples treated with method A. Multiple density layers were thought to help avoid the co-precipitation of microbes with abiotic parti- cles in the turbulent flow caused by falling of abiotic par- ticles (Morono et al. 2013). Compared with the group that only stood still for one hour as separation, density gradient centrifugation yielded higher cell numbers, and less cells remained in the pellet (Table 2). Thus, multi-layer density gradient centrifugation was adopted in the protocol for cell separation in further experiments. Reducing the sorption of initiator probe by abiotic particles To reduce false-positive signals resulting from the nonspe- cific binding of abiotic particles on probes, three types of hybridization buffer (A, B, C) were tested (Table 3). Buffer A is a widely used formula for FISH on environmental sam- ples (Yamaguchi et al. 2015b). Buffer B is mostly used with eukaryote samples, with several traditional blocking reagents amended (Choi et al. 2014). Buffer C contains a high con- centration of EDTA (Morono et al. 2020). Our experiments demonstrated that Buffer B did not work well on sediment samples, as nonspecific binding remained significant. The performance of Buffer C outcompeted that of Buffer A in decreasing the nonspecific binding. Buffer C was thus selected for further experiments. In the hope that DAPI might have similar proper- ties to SYBR Green, several filter sets were tested. Our results showed that DAPI-DNA complex only emitted a bright signal under light filter UV-2A (excitation filter 355/50 nm, dichroic mirror 400 nm, barrier filter 410 nm), while the DAPI-abiotic-particle complex emitted signals both under UV-2A and BV-2A (excitation filter 420/40 nm, dichroic mirror 455 nm, barrier filter 460 nm). Based on this property, images under UV-2A (Fig. 3a) and BV-2A (Fig. 3b) were acquired, separately, then the latter images were subtracted from the former. This process was able to dramatically decrease the signal of abiotic particles, while still retaining the signal of cells (Fig. 3c). The spectrum shift of the DAPI signal on abiotic particles may be related to the fact that DAPI has two binding modes to DNA, each of which has an independent fluorescence spectrum and fluorescence quantum yield (Manzini et al. 1983). Accord- ing to previous studies, mineral particles in the sediment sample can absorb debris DNA to become DAPI-bindable (Krsek and Wellington 1999). It is possible that DAPI bindings to debris DNA and genomic DNA have different preferences for binding modes. The different ratio of each type of binding modes on debris DNA and genomic DNA may cause the shift of spectrum. Removal of abiotic particles Sediment particles and cells can be easily separated due to their large differences in density, but first, the cells need be detached from the abiotic particles by physical and/ or chemical means to avoid cell loss during separation. Several detachment methods were tested on paraformal- dehyde-treated sediment samples. Method A uses a buffer 1 3 Marine Life Science & Technology (2021) 3:529–541 534 Table 2 Performance of different extraction methods on sediment samples from South China Sea DG density gradient centrifugation method. The number of cells in the supernatant and pallet after centrifugation was counted separately. See Supplementary Table S1 for more information Cell number (105cells/ml) DG Control Supernatant 18.55 ± 2.94 4.30 ± 1.54 Pellet 4.51 ± 1.07 8.74 ± 1.07 to differentiate cells and abiotic particles based solely on their shape and intensity. Similarly, the DNA dye SYBR Green is also absorbed by abiotic particles, although it was found that the emission spectrum of SYBR Green goes through a redshift when it binds to abiotic particles rather than DNA (Sunamura et al. 2003). By using a band-pass filter 490/20 nm (center wavelength/bandwidth) for exci- tation and 528/38 nm and 617/73 nm filters for detection, abiotic particles could easily to be ruled out (Morono et al. 2009). However, SYBR Green produces signals under sev- eral fluorescence channels and thus may not be compatible with FISH. DG density gradient centrifugation method. The number of cells in the supernatant and pallet after centrifugation was counted separately. See Supplementary Table S1 for more information Imaging and image processing to reduce false counterstaining signal In FISH experiments, DNA counterstaining is often applied to locate cells. A widely used counterstaining dye is DAPI, which shows specificity on double-stranded DNA. However, sediment particles, with complex com- positions and structures, inevitably absorb DAPI mol- ecules and emit a background fluorescence signal. In practice, using one band-pass DAPI filter set, it is difficult 1 3 Table 3 Composition of hybridization buffer tested on sediment samples Main ingredients Hybridization buffer A Hybridization buffer B Hybridization buffer C Tris–HCl 20 mmol/L – 20 mmol/L EDTA 0 – 250 mmol/L SDS 0.01% (w/v) – 0.01% (w/v) NaCl 0.9 mol/L – – Dextran sulfate 10% (w/v) 10% (w/v) – 5× Sodium chloride sodium citrate – 5× – Citric acid – 9 mmol/L – 1× Denhardt’ solution – 1× – Tween 20 – 0.1% (v/v) – Heparin - 50 μg/ml – Formamide X% (v/v) X% (v/v) X% (v/v) reference Yamaguchi et al. (2015a) Choi et al. (2014) Morono et al. (2020) Main ingredients Hybridization buffer A Hybridization buffer B Hybridization buffer C Tris–HCl 20 mmol/L – 20 mmol/L EDTA 0 – 250 mmol/L SDS 0.01% (w/v) – 0.01% (w/v) NaCl 0.9 mol/L – – Dextran sulfate 10% (w/v) 10% (w/v) – 5× Sodium chloride sodium citrate – 5× – Citric acid – 9 mmol/L – 1× Denhardt’ solution – 1× – Tween 20 – 0.1% (v/v) – Heparin - 50 μg/ml – Formamide X% (v/v) X% (v/v) X% (v/v) reference Yamaguchi et al. (2015a) Choi et al. (2014) Morono et al. (2020) Main ingredients Hybridization buffer A Hybridization buffer B Hybridization buffer C 1 3 535 Marine Life Science & Technology (2021) 3:529–541 Fig. 3 Image processing excludes false positive signal of DAPI stain. Image taken from UV-2A channel (a) subtracts that taken from BV-2A channel (b) to produce a new image with enhanced signal of cells and decaying signal of abiotic particles (c) Fig. 3 Image processing excludes false positive signal of DAPI stain. Image taken from UV-2A channel (a) subtracts that taken from BV-2A channel (b) to produce a new image with enhanced signal of cells and decaying signal of abiotic particles (c) Potential of HCR‑FISH Fig. 5 Cartoon showing the complete protocol of HCR-FISH on sedi- ment sample. The sediment slurry is fixed by paraformaldehyde and then washed with PBS. Cells are detached through shaking in the detergent buffer and extracted through density gradient centrifuga- tion. 0.22 μm-pore-size polycarbonate membrane is used for captur- ing cells. A piece of membrane is placed in the container like a 6-well plate. After dehydration with series ethanol solution, the air-dry membrane is covered by hybridization buffer and amplification buffer sequentially, each followed a washing step. Then the sample is stained with DAPI and mounted on the slide for microscopy. The image taken under BV-2A channel was subtracted from that under UV-2A channel for cell recognition and counting. The result could be overlaid with FISH probe signals for further analysis In addition to the application to marine sediment samples, our modified protocol may also be directly applied to con- centrated seawater samples, which share the same features of low cellular rRNA content. The 6× signal enhancement ensures a better sensitivity of HCR-FISH than that of tra- ditional FISH. The better penetration performance makes HCR-FISH more competitive on unknown samples than CARD-FISH, whose permeabilization process is empirical and sample dependent (Amann and Fuchs 2008). Mean- while, the DAPI-related image processing method could also be integrated into other fluorescence imaging experiments, including CARD-FISH and auto-fluorescence observation, to help identify cells from false-positive signals. p y p g In addition to detecting microbes, single-cell level detec- tion of mRNA can be more appealing to researchers study- ing unculturable microbes. For example, study of mcrA transcripts in methanogens using a two-step CARD-FISH process (i.e., two-pass Tyramide Signal Amplification FISH) shows the possibility to detect mRNA in microbes by FISH (Kubota et al. 2006). However, this method has not been widely used, probably because of its complexity. Some stud- ies have used single-molecular FISH (smFISH), a method that involves a set of more than 40 probes targeting differ- ent regions on the sequence of mRNA (Sepulveda et al. 2016; Skinner et al. 2013; So et al. 2011; Taniguchi et al. 2010; Yang et al. 2019), to locate mRNA molecules inside E. coli. Furthermore, genes could also be visualized by sig- nal-enhanced FISH like virusFISH (Castillo et al. 2020) or geneFISH (Barrero-Canosa et al. 2017; Moraru et al. 2010), which combines the strategies of dense fluorescent labeling of single probe and multiple probes. Detecting microbial cells in sediment samples clearly visualized by HCR-FISH, both on sediments from South China Sea and the enrichment slurry sample. For the South China Sea sample, traditional FISH was also applied for comparison. The average fluorescence intensity of the signals from bacteria were 10.9 ± 2.1 and 1.79 ± 0.60 A.U. through HCR-FISH and traditional FISH, respectively. There is thus a ~ 6× elevation of sig- nal intensity using the HCR process. It was also notewor- thy that, although multi-layer density gradient centrifu- gation was applied to this sample, many abiotic particles still remained. This indicates that many abiotic particles have a similar density to the cells. During cell embedding (on a filter membrane), the accumulation of these types After improving the performance of HCR-FISH on sediment samples, an optimized protocol was proposed (Supplementary Table S1 and Fig. 5). A sediment sam- ple from the South China Sea and an anaerobic metha- notroph enrichment from the Guaymas Basin sediment (Krukenberg et al. 2018) were further tested using the optimized protocol as shown in Fig. 5. After HCR-FISH, samples were counterstained with DAPI and micro- scopically examined. A post-imaging process was nec- essary to discriminate cells from fluorescent abiotic particles. As shown in Fig. 4, microbial cells could be 1 Fig. 4 HCR-FISH on two environmental samples. a HCR-FISH on sediment sample from South China Sea. Bacteria were labeled by EUB338-S1. Scale bar, 5 μm. b HCR-FISH on an enrichment sample of anaerobic methanotrophic archaea (ANME) consortia. SRB and ANME were labeled by species-specific probe SEEP2-658-S1 and ANME-1-350-S2, respectively Fig. 4 HCR-FISH on two environmental samples. a HCR-FISH on sediment sample from South China Sea. Bacteria were labeled by EUB338-S1. Scale bar, 5 μm. b HCR-FISH on an enrichment sample of anaerobic methanotrophic archaea (ANME) consortia. SRB and ANME were labeled by species-specific probe SEEP2-658-S1 and ANME-1-350-S2, respectively of anaerobic methanotrophic archaea (ANME) consortia. SRB and ANME were labeled by species-specific probe SEEP2-658-S1 and ANME-1-350-S2, respectively Fig. 4 HCR-FISH on two environmental samples. a HCR-FISH on sediment sample from South China Sea. Bacteria were labeled by EUB338-S1. Scale bar, 5 μm. b HCR-FISH on an enrichment sample 1 3 3 Marine Life Science & Technology (2021) 3:529–541 536 Potential of HCR‑FISH Considering the suc- cess of CARD-FISH and smFISH on in situ detection of single molecules in microbes, it would also be expected that HCR-FISH would be an effective single-molecule detection tool based on its efficient signal amplification and relatively easy protocol. Fig. 5 Cartoon showing the complete protocol of HCR-FISH on sedi- ment sample. The sediment slurry is fixed by paraformaldehyde and then washed with PBS. Cells are detached through shaking in the detergent buffer and extracted through density gradient centrifuga- tion. 0.22 μm-pore-size polycarbonate membrane is used for captur- ing cells. A piece of membrane is placed in the container like a 6-well plate. After dehydration with series ethanol solution, the air-dry membrane is covered by hybridization buffer and amplification buffer sequentially, each followed a washing step. Then the sample is stained with DAPI and mounted on the slide for microscopy. The image taken under BV-2A channel was subtracted from that under UV-2A channel for cell recognition and counting. The result could be overlaid with FISH probe signals for further analysis Sample preparation E. coli and M. methylutens were selected as representatives of bacteria and archaea, respectively. E. coli DH5α was cul- tured in Luria–Bertani (LB) broth at 37 ℃ under 200 rpm shaking. M. methylutens DSM 16625 was obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany) and cultured according to DSMZ protocols. Cells were fixed in 4% (v/v) paraform- aldehyde with phosphate-buffered saline (PBS; 136 mmol/L NaCl, 2.6 mmol/L KCl, 8 mmol/L Na2HPO4, and 2 mmol/L KH2PO4 [pH 7.2]) for 6 h at 4 ℃, washed twice by PBS, and preserved in PBS/ethanol 1:1 (v/v) mixture at −20 °C. Conclusions The traditional FISH, with a single probe, was not sensi- tive enough to explore tiny and/or less active microbes in many natural environments, such as marine sediments (Ishii et al. 2004) and open ocean seawater (Morris et al. 2002). Several new technologies have been developed to deal with this problem. In this study, we have demonstrated that HCR- FISH, one of these technologies, can significantly improve the labeling of microbes in sediment samples. The overall performance of the original HCR-FISH on sediment samples could be improved by applying the following modifications: 1) The signal intensity and resolution of HCR-FISH can be of abiotic particles will also cover the cell, making the improvement by density gradient centrifugation less than expected. of abiotic particles will also cover the cell, making the improvement by density gradient centrifugation less than expected. 1 3 Marine Life Science & Technology (2021) 3:529–541 537 optimized through changing the concentration of initiator probes and types of HCR sequences; 2) The negative effects of abiotic particles in HCR-FISH of sediment samples can be reduced by employing sample pretreatment and optimized hybridization buffer; 3) The counterstaining signal of cells could be distinguished from the background fluorescence of abiotic particles by performing an image processing method to emphasize the cellular DAPI signal. Combining these efforts, we developed an optimized HCR-FISH protocol for sediment samples (Supplementary Table S1 and Fig. 5). hour to precipitate the particles, and the supernatant was used for the cell counts. A control group was also set with fixed sediment sample shaken, rested, and counted as for the other groups. Cell counting To capture the cells, pretreated samples were first diluted in 10 ml PBS and then filtered onto 0.22 μm pore size mem- branes (GTBP02500, Isopore, Merck Millipore). Then, the membrane was dipped in 50, 80, and 96% (v/v) ethanol for dehydration, each step was for 5 min (Yamaguchi et al. 2015a). SYBR Green I (Solarbio, China) was diluted in ultrapure water by 100 times to provide the working solution. The working solution was then applied to the membranes containing the samples for 10 min at room temperature. Then, the membranes were placed on the top of the delicate task wipes (Kimtech Science, USA). The remaining working solution would be absorbed by the delicate task wipes. For each filter membrane, 16 fields of view were selected and counted under the microscope. Three filter membranes were analyzed per sample. Sediment samples were collected from the South China Sea and stored at −80 °C before proceeding to HCR-FISH. The same fixation and preservation procedures, as described above and in Supplementary Table S1, were applied on these sediment samples. An enrichment sample of anaerobic methanotroph was transferred from a long-term enrichment incubator at the Max Planck Institute for Marine Microbiology, Germany. The enrichment was transferred and refreshed according to the reference (Krukenberg et al. 2018). The same fixation and preservation procedures, as described above and in Sup- plementary Table S1, were also applied. Cell extraction Cells were extracted from the detergent buffer detached sam- ples with density gradient centrifugation. Samples were cen- trifuged at 1500 g for 30 min through multiple density layers consisting of 30%, 50%, and 80% (w/v) Nycodenz (Axis- Shield, Norway), respectively. All the transparent liquid was carefully transferred to a clean vial to avoid the disturbance on the precipitate. Then, the supernatant was diluted 4 times by PBS and centrifuged at 14,500 g for 15 min to collect the cells. The precipitated cells were resuspended in PBS for further processing. To compare, parts of the detached samples were allowed to stand for an hour to precipitate the abiotic particles and suspend cells. Their supernatants were collected and stored in PBS for further processing. Cell detachment 30 min later, the washing buffer was removed and 40 μl amplifier probes mixed with amplification buffer was added to the membrane. The membrane was then incubated at 35 °C for 20 min under humidified conditions. Sequentially, 10 ml 4 °C PBS was applied to the membrane on ice to remove excessive fluorescent probes. The membrane was washed in ultrapure water and then in 96% (v/v) ethanol each for one minute on ice and air-dried. Imaging and image processing The imaging experiment was carried out by epifluorescence microscope (Nikon ECLIPSE 90i, Tokyo, Japan), coupled with an illuminator (Nikon INTENSILIGHT C-HGFIE, Tokyo, Japan) and a CCD camera (CoolSNAP HQ2, Photo- metrics, USA). Nikon plan-apochromat 100× oil objective lens was used for imaging. The excitation filter, dichroic filter, and barrier filter of light filter cubes are summarized in Table 4. The camera exposure time varied between 50 and 800 ms depending on the type of fluorophore and observed samples. Images of anaerobic methanotrophs were collected with a Zeiss LSM 880 microscope (Carl Zeiss, Germany) equipped with a plan-apochromat 63× oil objective lens and Airyscan super-resolution system for better resolution. 405, 561, and 633 nm lasers were used for excitation of DAPI, Alexa Fluor 555, Alexa Fluor 647, respectively. p y The channel subtraction was performed using the Image Calculator function integrated in Fiji (a "batteries-included" distribution of ImageJ 1.53c) software released on 2017 May 30 (Schindelin et al. 2015). To balance the background dif- ferences between channels, the images were treated under the following rules: define the gray value that most pixels possessed for images from UV-2A and BV-2A channels as IU and IB, respectively. Both IU and IB represent the low- intensity background in the corresponding image channels. Hence, pixels with a grey level of IU in UV-2A channel and those with grey level of IB in BV-2A channel should occupy the same positions in the image. Then, the grey values of these pixels should become zero by channel subtraction to precisely eliminate the background signal. This could be achieved by multiplying the grey value of each pixel on the image from BV-2A by a factor α = IU/IB. Practically, α ≈ 1/8 for our instruments, and thus the goal could also be achieved by adjusting the exposure time or exciting light intensity for an image from BV-2A to 1/8 of that of the image from UV-2A. The factor α may vary with different light sources, but not with samples. For counterstaining, 20 μl of 10 ng/ml DAPI was applied to the samples for at least 10 min. Samples were washed by PBS for 10 min and dehydrated by 80% (v/v) ethanol. Cell detachment The HCR-FISH protocol was modified based on that of Yamaguchi et al. (2015a). Cells were captured on mem- branes as described in “Cell counting” section. The mem- brane was cut into eight pieces, each of which was enough for the following experiment. The unused pieces were stored in −20 °C for future usage. 40 μl of hybridization buffer with 10 μmol/L initiator probes was added to the membrane. The concentration of formamide in the hybridization buffer depends on the sequence of the probe. As tested here, 25% (v/v) formamide was acceptable for simultaneous detection of microbes using probe EUB338, ARCH915, ANME-1-350 and SEEP2-658. The sample was incubated at 46 °C for 2 h Cells were detached from fixed samples using different methods. For method A, eight volumes of sediment were mixed with one volume of detergent buffer (100 mmol/L EDTA, 100 mmol/L sodium pyrophosphate, 1% (v/v) Tween 80) and vortexed for one hour at level 5 (Vortex-Genie 2, Scientific Industries). For method B, samples were placed in an ice-water mixture 2 cm away from an ultrasonic probe (Model 50 Sonic Dismembrator, Thermofisher). A 30-s sonication with 50% maximum power was performed three times, each followed by a resting period of 30 s. To evalu- ate their performance, detached samples were rested for an 1 3 Marine Life Science & Technology (2021) 3:529–541 538 Table 4 Filter information of light filter cubes in microscope for this study a Central wavelength/bandwidth, the same hereinafter b Used for fluorophore Alexa 488 c Used for fluorophore Alexa 594 Excitation filter Dichroic filter Barrier filter UV-2A long-pass filter set 355/50 nma 400 nm 410 nm BV-2A long-pass filter set 420/40 nm 455 nm 460 nm DAPI band-pass filter set 375/28 nm 415 nm 460/60 nm FITC band-pass filter setb 480/30 nm 505 nm 535/45 nm Texas Red band-pass filter setc 560/40 nm 595 nm 630/60 nm Table 4 Filter information of light filter cubes in microscope for this study (v/v) ethanol, each for one minute, on ice and air dried. The counterstaining step remained the same. in humidified conditions. The humidity requirement can be achieved by sealing the sample in a small container and/or placing a wet tissue beside the sample. After incubation, the membrane was washed in 10 ml washing buffer [20 mmol/L Tris–HCl, 0.01% (w/v) SDS, 0.056–0.225 mol/L NaCl] and incubated at 48 ℃ for 30 min to remove excessive ini- tiator probes. Cell detachment The concentration of NaCl in the washing buffer depends on the concentration of formamide in the hybridization buffer. Based on the Pernthaler et al. (2001), 0.056 mol/L NaCl was used for 40% (v/v) formamide, 0.159 mol/L NaCl for 25% (v/v) formamide and 0.225 mol/L NaCl for 20% (v/v) formamide. A complete table of con- centration pairs can be found in Pernthaler et al. (2001). All the following steps should be done in dark to avoid quenching of the fluorophore. During washing, each ampli- fier probe was dissolved in amplification buffer [0.9 mol/L NaCl, 0.05 mol/L Na2HPO4, 0.01% (w/v) SDS] and incu- bated stepwise at 95 °C for 90 s and 25 °C for 30 min, for initialization. Next, these amplifier probes were mixed and the final concentration of each probe was 2.5 μmol/L. 30 min later, the washing buffer was removed and 40 μl amplifier probes mixed with amplification buffer was added to the membrane. The membrane was then incubated at 35 °C for 20 min under humidified conditions. Sequentially, 10 ml 4 °C PBS was applied to the membrane on ice to remove excessive fluorescent probes. The membrane was washed in ultrapure water and then in 96% (v/v) ethanol each for one minute on ice and air-dried. in humidified conditions. The humidity requirement can be achieved by sealing the sample in a small container and/or placing a wet tissue beside the sample. After incubation, the membrane was washed in 10 ml washing buffer [20 mmol/L Tris–HCl, 0.01% (w/v) SDS, 0.056–0.225 mol/L NaCl] and incubated at 48 ℃ for 30 min to remove excessive ini- tiator probes. The concentration of NaCl in the washing buffer depends on the concentration of formamide in the hybridization buffer. Based on the Pernthaler et al. (2001), 0.056 mol/L NaCl was used for 40% (v/v) formamide, 0.159 mol/L NaCl for 25% (v/v) formamide and 0.225 mol/L NaCl for 20% (v/v) formamide. A complete table of con- centration pairs can be found in Pernthaler et al. (2001). All the following steps should be done in dark to avoid quenching of the fluorophore. During washing, each ampli- fier probe was dissolved in amplification buffer [0.9 mol/L NaCl, 0.05 mol/L Na2HPO4, 0.01% (w/v) SDS] and incu- bated stepwise at 95 °C for 90 s and 25 °C for 30 min, for initialization. Next, these amplifier probes were mixed and the final concentration of each probe was 2.5 μmol/L. Traditional FISH Samples underwent the same treatment as those in the HCR- FISH protocol until the hybridization step. 40 μl of hybridi- zation buffer with 10 μmol/L of fluorophore-labeled probes was added to the membrane. The sample was incubated and washed following the same procedure as for HCR-FISH. Then, the membrane was directly washed in water and 96% 1 3 539 Marine Life Science & Technology (2021) 3:529–541 The analysis of cellular fluorescence intensity was also performed on Fiji. To identify the region of positive signal for each channel, two binary images were created from the channel-subtracted DAPI image and (HCR-)FISH image, separately. The thresholds for producing these binary images were decided based on the built-in automatic algorithm “Moments”. The intersection of the two binary images was used as the cellular mask for intensity calculation of true cells. To measure the cellular fluorescence intensity, we first designated the (HCR-)FISH image to be calculated in the “Set Measurements…” window. Then the window of the cel- lular mask image was activated and the “Analyze Particles” function was run. The “target size” parameter of the func- tion was set to be 0.5–100 mm2. With this pipeline, the Fiji software would first mark the cell regions using the cellular mask image, and then calculate the average grey value of each region on the (HCR-)FISH image. For each experiment group, at least 200 cells were taken into account. References Amalfitano S, Fazi S (2008) Recovery and quantification of bacterial cells associated with streambed sediments. J Microbiol Methods 75:237–243i Amann R, Fuchs BM (2008) Single-cell identification in microbial communities by improved fluorescence in situ hybridization tech- niques. Nat Rev Microbiol 6:339–348i Amann RI, Ludwig W, Schleifer K-H (1995) Phylogenetic identifica- tion and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59:143–169i Bao J, Zhang W, Zhou W, Lv M, Lu C, Yu H (2020) An amplifica- tion strategy for detecting HER2 with a quasi-targeted proteom- ics approach coupled with aptamer-triggered hybridization chain reaction. Talanta 215:120918 Baptista JD, Lunn M, Davenport RJ, Swan DL, Read LF, Brown MR, Morais C, Curtis TP (2014) Agreement between amoA gene-spe- cific quantitative PCR and fluorescence in situ hybridization in the measurement of ammonia-oxidizing bacteria in activated sludge. Appl Environ Microbiol 80:5901–5910 Barrero-Canosa J, Moraru C, Zeugner L, Fuchs BM, Amann R (2017) Direct-geneFISH: a simplified protocol for the simultaneous detection and quantification of genes and rRNA in microorgan- isms. Traditional FISH Environ Microbiol 19:70–82 Supplementary Information The online version contains supplemen- tary material available at https://doi.org/10.1007/s42995-021-00098-8. Bi S, Yue SZ, Zhang SS (2017) Hybridization chain reaction: a versa- tile molecular tool for biosensing, bioimaging, and biomedicine. Chem Soc Rev 46:4281–4298 Acknowledgements We thank for the funding: National Key R&D Pro- gram of China (grant numbers 2018YFC0310800, 2018YFC0310803); COMRA Project DY135-B2-12; the National Natural Science Founda- tion of China (grant numbers 41525011, 91751205, 11774225), the Recruitment Program of Global Experts (Program for Young Profes- sionals), and the Natural Science Foundation of Shanghai (grant no. 18ZR1419800). This is also a contribution to the Center for Ocean Mega-Science, Chinese Academy of Sciences, the Senior User Project of RV KEXUE (KEXUE2019GZ06); and the International Center for Deep-Life Investigation (IC-DLI). We thank Gunter Wegener for pro- viding the ANME enrichment sample. Buongiorno J, Turner S, Webster G, Asai M, Shumaker AK, Roy T, Weightman A, Schippers A, Lloyd KG (2017) Interlaboratory quantification of Bacteria and Archaea in deeply buried sediments of the Baltic Sea (IODP Expedition 347). FEMS Microbiol Ecol 93:fix007 i Castillo YM, Sebastian M, Forn I, Grimsley N, Yau S, Moraru C, Vaque D (2020) Visualization of viral infection dynamics in a unicellular eukaryote and quantification of viral production using virus fluorescence in situ hybridization. Front Microbiol 11:1559 l Chen Y, Li YL, Zhou GT, Li H, Lin YT, Xiao X, Wang FP (2014) Biomineralization mediated by anaerobic methane-consuming cell consortia. Sci Rep 4:5696 Author contributions FW and HX conceived and guided the study. ZJ and YD performed laboratory work. ZJ drafted the manuscript, and FW and HX made further revisions. All authors read and approved the final version of the manuscript. Choi HM, Chang JY, le Trinh A, Padilla JE, Fraser SE, Pierce NA (2010) Programmable in situ amplification for multiplexed imag- ing of mRNA expression. Nat Biotechnol 28:1208–1212 Choi J, Love KR, Gong Y, Gierahn TM, Love JC (2011) Immuno- hybridization chain reaction for enhancing detection of individual cytokine-secreting human peripheral mononuclear cells. Anal Chem 83:6890–6895 Declarations ISME J 3:503–511 Hatzenpichler R, Connon SA, Goudeau D, Malmstrom RR, Woyke T, Orphan VJ (2016) Visualizing in situ translational activity for identifying and sorting slow-growing archaeal-bacterial consortia. Proc Natl Acad Sci USA 113:E4069-4078 Morono Y, Terada T, Kallmeyer J, Inagaki F (2013) An improved cell separation technique for marine subsurface sediments: applica- tions for high-throughput analysis using flow cytometry and cell sorting. Environ Microbiol 15:2841–2849 Hou T, Li W, Liu X, Li F (2015) Label-free and enzyme-free homoge- neous electrochemical biosensing strategy based on hybridization chain reaction: a facile, sensitive, and highly specific microRNA assay. Anal Chem 87:11368–11374fi l sorting. Environ Microbiol 15:2841–2849 Morono Y, Kubota K, Tsukagoshi D, Terada T (2020) EDTA-FISH: a simple and effective approach to reduce non-specific adsorption of probes in fluorescence in situ hybridization (FISH) for environ- mental samples. Microbes Environ 35:3 Huang WE, Stoecker K, Griffiths R, Newbold L, Daims H, Whiteley AS, Wagner M (2007) Raman-FISH: combining stable-isotope Raman spectroscopy and fluorescence in situ hybridization for the single cell analysis of identity and function. Environ Microbiol 9:1878–1889 Morris RM, Rappe MS, Connon SA, Vergin KL, Siebold WA, Carlson CA, Giovannoni SJ (2002) SAR11 clade dominates ocean surface bacterioplankton communities. Nature 420:806–810 Huang J, Wu Y, Chen Y, Zhu Z, Yang X, Yang CJ, Wang K, Tan W (2011) Pyrene-excimer probes based on the hybridization chain reaction for the detection of nucleic acids in complex biological fluids. Angew Chem Int Edit 50:401–404 Nikolakakis K, Lehnert E, McFall-Ngai MJ, Ruby EG (2015) Use of hybridization chain reaction-fluorescent in situ hybridization to track gene expression by both partners during initiation of sym- biosis. Appl Environ Microbiol 81:4728–4735 Orcutt B, Meile C (2008) Constraints on mechanisms and rates of anaerobic oxidation of methane by microbial consortia: process- based modeling of ANME-2 archaea and sulfate reducing bacteria interactions. Biogeosciences 5:1587–1599 l Imachi H, Nobu MK, Nakahara N, Morono Y, Ogawara M, Takaki Y, Takano Y, Uematsu K, Ikuta T, Ito M, Matsui Y, Miyazaki M, Murata K, Saito Y, Sakai S, Song C, Tasumi E, Yamanaka Y, Yamaguchi T, Kamagata Y et al (2020) Isolation of an archaeon at the prokaryote-eukaryote interface. Nature 577:519–525 Orphan VJ, Turk KA, Green AM, House CH (2009) Patterns of 15N assimilation and growth of methanotrophic ANME-2 archaea and sulfate-reducing bacteria within structured syntrophic consortia revealed by FISH-SIMS. Declarations Choi HMT, Beck VA, Pierce NA (2014) Next-generation in situ hybrid- ization chain reaction: higher gain, lower cost, greater durability. ACS Nano 8:4284–4294i Conflict of interest The authors declare no conflicts of interests. Animal and human rights statement No animal and human rights are involved in this article. Daims H, Wagner M (2007) Quantification of uncultured microorgan- isms by fluorescence microscopy and digital image analysis. Appl Microbiol Biotechnol 75:237–248i Dirks RM, Pierce NA (2004) Triggered amplification by hybridization chain reaction. Proc Natl Acad Sci USA 101:15275–15278ifi Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Fazi S, Amalfitano S, Pizzetti H, Pernthaler J (2007) Efficiency of fluorescence in situ hybridization for bacterial cell identification in temporary river sediments with contrasting water content. Syst Appl Microbiol 30:463–470 Francis TB, Krüger K, Fuchs BM, Teeling H, Amann RI (2019) Can- didatus Prosiliicoccus vernus, a spring phytoplankton bloom associated member of the Flavobacteriaceae. Syst Appl Micro- biol 42:41–53 Gao Z, Qiu Z, Lu M, Shu J, Tang D (2017) Hybridization chain reac- tion-based colorimetric aptasensor of adenosine 5’-triphosphate 1 3 540 Marine Life Science & Technology (2021) 3:529–541 on unmodified gold nanoparticles and two label-free hairpin probes. Biosens Bioelectron 89:1006–1012 Moraru C, Lam P, Fuchs BM, Kuypers MM, Amann R (2010) Gen- eFISH–an in situ technique for linking gene presence and cell identity in environmental microorganisms. Environ Microbiol 12:3057–3073 Grieb A, Bowers RM, Oggerin M, Goudeau D, Lee J, Malmstrom RR, Woyke T, Fuchs BM (2020) A pipeline for targeted metagenomics of environmental bacteria. Microbiome 8:21 Morono Y, Terada T, Masui N, Inagaki F (2009) Discriminative detec- tion and enumeration of microbial life in marine subsurface sedi- ments. Declarations Environ Microbiol 11:1777–1791 Ishii K, Mussmann M, MacGregor BJ, Amann R (2004) An improved fluorescence in situ hybridization protocol for the identification of bacteria and archaea in marine sediments. FEMS Microbiol Ecol 50:203–213 Pernthaler A, Preston CM, Pernthaler J, DeLong EF, Amann R (2002) Comparison of fluorescently labeled oligonucleotide and polynu- cleotide probes for the detection of pelagic marine bacteria and archaea. Appl Environ Microbiol 68:661–667 Jia LP, Zhao RN, Wang LJ, Ma RN, Zhang W, Shang L, Wang HS (2018) Aptamer based electrochemical assay for protein kinase activity by coupling hybridization chain reaction. Biosens Bio- electron 117:690–695 Royet D, Dempsey NM, Simonet P, Frénéa-Robin M (2018) A new magnetic cell fishing approach based on hybridization chain reac- tion: HCR-MISH. Sensor Actuat B Chem 273:126–132 Kallmeyer J, Smith DC, Spivack AJ, D’Hondt S (2008) New cell extraction procedure applied to deep subsurface sediments. Lim- nol Oceanogr Methods 6:236–245f i tion: HCR-MISH. Sensor Actuat B Chem 273:126–132 Schindelin J, Rueden CT, Hiner MC, Eliceiri KW (2015) The ImageJ ecosystem: an open platform for biomedical image analysis. Mol Reprod Dev 82:518–529 Krsek M, Wellington E (1999) Comparison of different methods for the isolation and purification of total community DNA from soil. J Microbiol Methods 39:1–16 Sepulveda LA, Xu H, Zhang J, Wang MY, Golding I (2016) Measure- ment of gene regulation in individual cells reveals rapid switching between promoter states. Science 351:1218–1222 Krukenberg V, Riedel D, Gruber-Vodicka HR, Buttigieg PL, Teget- meyer HE, Boetius A, Wegener G (2018) Gene expression and ultrastructure of meso- and thermophilic methanotrophic consor- tia. Environ Microbiol 20:1651–1666 Skinner SO, Sepulveda LA, Xu H, Golding I (2013) Measuring mRNA copy number in individual Escherichia coli cells using single-mol- ecule fluorescent in situ hybridization. Nat Protoc 8:1100–1113 Kubota K (2013) CARD-FISH for environmental microorganisms: technical advancement and future applications. Microbes Envi- ron 28:3–12 l So LH, Ghosh A, Zong C, Sepulveda LA, Segev R, Golding I (2011) General properties of transcriptional time series in Escherichia coli. Nat Genet 43:554–560 Kubota K, Ohashi A, Imachi H, Harada H (2006) Visualization of mcr mRNA in a methanogen by fluorescence in situ hybridization with an oligonucleotide probe and two-pass tyramide signal amplifica- tion (two-pass TSA-FISH). J Microbiol Meth 66:521–528 Sunamura M, Maruyama A, Tsuji T, Kurane R (2003) Spectral imaging detection and counting of microbial cells in marine sediment. Declarations J Microbiol Methods 53:57–65 Taniguchi Y, Choi PJ, Li GW, Chen HY, Babu M, Hearn J, Emili A, Xie XS (2010) Quantifying E. coli proteome and transcriptome with single-molecule sensitivity in single cells. Science 329:533–538 Manzini G, Barcellona ML, Avitabile M, Quadrifoglio F (1983) Inter- action of diamidino-2-phenylindole (DAPI) with natural and syn- thetic nucleic acids. Nucleic Acids Res 11:8861–8876 Massie H, Samis H, Baird M (1972) The kinetics of degradation of DNA and RNA by H2O2. Biochem Biophys Acta 272:539–548 Wilen BM, Onuki M, Hermansson M, Lumley D, Mino T (2008) Microbial community structure in activated sludge floc analysed by fluorescence in situ hybridization and its relation to floc stabil- ity. Water Res 42:2300–2308 Matsubayashi M, Shimada Y, Li YY, Harada H, Kubota K (2017) Phy- logenetic diversity and in situ detection of eukaryotes in anaerobic sludge digesters. PLoS ONE 12:e0172888 Wu J, Li N, Yao Y, Tang D, Yang D, Ong’achwa Machuki J, Li J, Yu Y, Gao F (2018) DNA-stabilized silver nanoclusters for label- free fluorescence imaging of cell surface glycans and fluorescence guided photothermal therapy. Anal Chem 90:14368–14375 Miao P, Tang Y, Yin J (2015) MicroRNA detection based on analyte triggered nanoparticle localization on a tetrahedral DNA modified electrode followed by hybridization chain reaction dual amplifica- tion. Chem Commun 51:15629–15632 Yamaguchi T, Kubota K (2017) Visualization of microorganisms in bioprocesses. In: Purohit HJ, Kalia VC, Vaidya AN, Khardenavis 1 3 541 Marine Life Science & Technology (2021) 3:529–541 AA (eds) Optimization and applicability of bioprocesses. Springer Singapore, Singapore, pp 13–26 AA (eds) Optimization and applicability of bioprocesses. Springer Singapore, Singapore, pp 13–26 AA (eds) Optimization and applicability of bioprocesses. Springer Singapore, Singapore, pp 13–26 Zeng R, Zhang L, Su L, Luo Z, Zhou Q, Tang D (2019) Photoelectro- chemical bioanalysis of antibiotics on rGO-Bi2WO6-Au based on branched hybridization chain reaction. Biosens Bioelectron 133:100–106 Yamaguchi T, Fuchs BM, Amann R, Kawakami S, Kubota K, Hatamoto M, Yamaguchi T (2015a) Rapid and sensitive identification of marine bacteria by an improved in situ DNA hybridization chain reaction (quickHCR-FISH). Syst Appl Microbiol 38:400–405 Zhang B, Liu B, Tang D, Niessner R, Chen G, Knopp D (2012) DNA-based hybridization chain reaction for amplified bioelec- tronic signal and ultrasensitive detection of proteins. Declarations Anal Chem 84:5392–5399 Yamaguchi T, Kawakami S, Hatamoto M, Imachi H, Takahashi M, Araki N, Yamaguchi T, Kubota K (2015b) In situ DNA-hybridiza- tion chain reaction (HCR): a facilitated in situ HCR system for the detection of environmental microorganisms. Environ Microbiol 17:2532–2541 Zhang K, Lv S, Zhou Q, Tang D (2020) CoOOH nanosheets-coated g-C3N4/CuInS2 nanohybrids for photoelectrochemical biosensor of carcinoembryonic antigen coupling hybridization chain reac- tion with etching reaction. Sensor Actuat B Chem 307:127631 Yang S, Kim S, Kim D-K, Jeon An H, Bae Son J, Hedén Gynnå A, Ki Lee N (2019) Transcription and translation contribute to gene locus relocation to the nucleoid periphery in E. coli. Nat Com- mun 10:5131 1 3 1 3 3
https://openalex.org/W2279368225	https://dro.deakin.edu.au/articles/journal_contribution/A_process_evaluation_of_the_Supermarket_Healthy_Eating_for_Life_SHELf_randomized_controlled_trial/20887123/1/files/37155757.pdf	English	null	A process evaluation of the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial	The international journal of behavioural nutrition and physical activity	2,016	cc-by	12,731	A process evaluation of the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial AUTHOR(S) Dana Olstad, Kylie Ball, Gavin Abbott, Sarah McNaughton, Ha Le, C Ni Mhurchu, C Pollard, David Crawford Dana Olstad, Kylie Ball, Gavin Abbott, Sarah McNaughton, Ha Le, C Ni Mhurchu, C Pollard, David Crawford 10536/DRO/DU:30081935 Downloaded from Deakin University’s Figshare repository Deakin University CRICOS Provider Code: 00113B Olstad, Dana Lee, Ball, Kylie, Abbott, Gavin, McNaughton, Sarah A., Le, Ha N. D., Mhurchu, Cliona Ni, Pollard, Christina and Crawford, David A. 2016, A process evaluation of the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial, International journal of behavioral nutrition and physical activity, vol. 13, no. 1, pp. 1-15. Olstad, Dana Lee, Ball, Kylie, Abbott, Gavin, McNaughton, Sarah A., Le, Ha N. D., Mhurchu, Cliona Ni, Pollard, Christina and Crawford, David A. 2016, A process evaluation of the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial, International journal of behavioral nutrition and physical activity, vol. 13, no. 1, pp. 1-15. Abstract Background: Supermarket Healthy Eating for Life (SHELf) was a randomized controlled trial that operationalized a socioecological approach to population-level dietary behaviour change in a real-world supermarket setting. SHELf tested the impact of individual (skill-building), environmental (20 % price reductions), and combined (skill-building + 20 % price reductions) interventions on women’s purchasing and consumption of fruits, vegetables, low-calorie carbonated beverages and water. This process evaluation investigated the reach, effectiveness, implementation, and maintenance of the SHELf interventions. Methods: RE-AIM provided a conceptual framework to examine the processes underlying the impact of the interventions using data from participant surveys and objective sales data collected at baseline, post-intervention (3 months) and 6-months post-intervention. Fisher’s exact, χ2 and t-tests assessed differences in quantitative survey responses among groups. Adjusted linear regression examined the impact of self-reported intervention dose on food purchasing and consumption outcomes. Thematic analysis identified key themes within qualitative survey responses. Results: Reach of the SHELf interventions to disadvantaged groups, and beyond study participants themselves, was moderate. Just over one-third of intervention participants indicated that the interventions were effective in changing the way they bought, cooked or consumed food (p < 0.001 compared to control), with no differences among intervention groups. Improvements in purchasing and consumption outcomes were greatest among those who received a higher intervention dose. Most notably, participants who said they accessed price reductions on fruits and vegetables purchased (519 g/week) and consumed (0.5 servings/day) more vegetables. The majority of participants said they accessed (82 %) and appreciated discounts on fruits and vegetables, while there was limited use (40 %) and appreciation of discounts on low-calorie carbonated beverages and water. Overall reported satisfaction with, use, and impact of the skill-building resources was moderate. Maintenance of newly acquired behaviours was limited, with less than half of participants making changes or using study-provided resources during the 6-month post-intervention period. Conclusions: SHELf’s reach and perceived effectiveness were moderate. The interventions were more effective among those reporting greater engagement with them (an implementation-related construct). Maintenance of newly acquired behaviours proved challenging. Trial registration: Current controlled trials ISRCTN39432901. Keywords: Process evaluation, Dietary behaviours, Fruits and vegetables, Carbonated beverages, Water, Food consumption, Food purchasing, Women, Supermarkets, RE-AIM © 2016 Olstad et al. A process evaluation of the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial Dana Lee Olstad1, Kylie Ball1, Gavin Abbott1, Sarah A. McNaughton1, Ha N. D. Le2, Cliona Ni Mhurchu3, Christina Pollard4 and David A. Crawford1 Correspondence: kylie.ball@deakin.edu.au 1Centre for Physical Activity and Nutrition Research, Deakin University, 221 Burwood Highway, Burwood, VIC 3125, Australia Full list of author information is available at the end of the article http://hdl.handle.net/10536/DRO/DU:30081935 http://hdl.handle.net/10536/DRO/DU:30081935 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 DOI 10.1186/s12966-016-0352-3 Open Access Background always consistent with expectations. Specifically, in the immediate post-intervention period purchase of FV in- creased in the price reduction group, and purchase of fruit increased in the combined group, while no changes were found within the skill-building group. In addition, small unexpected increases in consumption of high-calorie car- bonated beverages were observed in the skill-building and price reduction groups. Some of these behaviours were maintained at 6-months post-intervention, while others were not. These mixed and sometimes unexpected find- ings suggested a need to investigate the processes under- lying these results. Unhealthy dietary patterns are among the leading public health concerns worldwide given their significant contri- bution to the global burden of disease [1]. Shifting dietary behaviours in health promoting directions is challenging however, given the ubiquity and relatively entrenched na- ture of unhealthy dietary patterns. In Australia, although most individuals ≥2 years of age consume fruits and vege- tables (FV) on a daily basis, fewer consume sufficient quantities of fruits (54 %) and vegetables (7 %) for health benefits [2]. Intake of discretionary foods such as confec- tionary, cakes, sweet biscuits, pastries and soft drinks appear to be displacing intake of healthy foods such as FV within the Australian population, with more than one- third of daily energy intake contributed by these and other discretionary foods [2]. Strategies to promote purchase and consumption of FV and other healthy foods are there- fore the subject of intense investigation. y g Given the increasingly complex nature of public health interventions, and the need to evaluate their impacts in real-world contexts, it is no longer sufficient to confine evaluative activities to the realm of efficacy testing. A focus on program efficacy without attention to process contributes to a reductionist paradigm that oversimpli- fies reality and ignores essential details regarding why an intervention did or did not achieve its intended effects [17]. Process evaluation provides a means to monitor and document the implementation of health promotion programs in order to better understand why a program was or was not successful, and how any effects were achieved [18]. However despite its importance, few inter- ventions within food shopping environments have under- taken process evaluation [19], and RCTs similar to SHELf have only examined selected aspects of reach and imple- mentation [16, 20–23]. Background Limited availability of process- related data constrains efforts to apply lessons from past studies to improve the design and execution of future in- vestigations. The purpose of this study was to conduct a process evaluation to investigate the reach, effective- ness, implementation, and maintenance of the SHELf interventions. Until recently, a paradigm of individual responsibility for health behaviours predominated within the nutri- tion community, such that many past investigations sought to improve dietary behaviours through increas- ing individual-level nutrition-related knowledge, motiv- ation, and self-efficacy for healthy eating [3]. However, although intrapersonal factors are important correlates of food intake patterns [4–8], individual-level behaviour change interventions have generally yielded only mod- est outcomes [9, 10]. Indeed, behaviour change is not easily achieved nor maintained in the absence of struc- tural supports, and therefore recent initiatives have sought to modify the context within which individual behaviours occur by creating environments that sup- port healthy eating. Socioecological approaches inte- grate these diverse perspectives by acknowledging the joint contribution of individual, social and environmental factors to dietary behaviours [11, 12]. y The current study is based on the Supermarket Healthy Eating for Life (SHELf) project, which was a randomized controlled trial (RCT) that operationalized a socioecological approach to population-level dietary be- haviour change in a real-world supermarket setting [13, 14]. SHELf tested the impact of individual (skill-building), environmental (price reductions), and combined (skill- building + price reductions) interventions on women’s food purchasing and consumption behaviours. The study complemented past ecologically-informed supermarket- based RCTs performed in other nations. In a New Zealand study, 12.5 % price discounts on healthier foods led to in- creased purchasing of FV and other healthier foods, while in the Netherlands a 50 % discount on FV with and with- out additional education increased purchasing and con- sumption of FV [15, 16]. Education alone had no impact in either of these studies. Findings from the SHELf project differed somewhat from these earlier studies and were not Abstract Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Page 2 of 15 Page 2 of 15 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Theoretical framework Women who responded to the mail-out (n = 693; 23 % response rate) were invited to participate if they met the following eligibility criteria: aged 18–60 years, the main household food shopper, shopped regularly at Coles super- markets in one of the two catchment areas, held a Coles store loyalty card and were willing to use it when they shopped at Coles during the study period, able to speak, read and write in English, willing to disclose household in- come, and the only woman in their household taking part in the study (n = 674 women met inclusion criteria). The study was approved by the Deakin University Faculty of Health Human Ethics Advisory Group. All women pro- vided written, informed consent prior to participating. RE-AIM provides a conceptual model for understanding the impact of public health interventions [17, 27]. Overall impact is modelled as a function of five factors: reach, ef- fectiveness, adoption, implementation, and maintenance [17]. We used RE-AIM as an organizing framework to investigate the reach, effectiveness, implementation and maintenance of the SHELf interventions (i.e. excluding adoption). This framework was selected because of its comprehensiveness, robustness demonstrated through sev- eral decades of use in a variety of fields, and adaptability to different purposes and contexts [27–29]. Kessler et al. [30] have outlined specific items across RE-AIM dimensions that represent comprehensive application of the full model. Our measures included these items with adaptations, addi- tions and exclusions in some instances to accommodate the study design and availability of data. Factors at the organizational level (i.e. adoption, which refers to the pro- portion and representativeness of settings that adopt a pro- gram, along with maintenance of new programs at the organizational level) were not considered relevant given that study sites were pre-selected and were not expected to maintain the interventions beyond the study period. Participants and setting p g Full methodological details of the SHELf study have been previously described [13, 14] and are briefly sum- marized here. SHELf was a socioecologically informed intervention conducted in two Coles® supermarkets (the second largest grocery chain in Australia) from May, 2011 to November, 2012. The Australian Bureau of Sta- tistics’ Socioeconomic Index for Areas (SEIFA), derived from aggregate measures of social and economic infor- mation from the population census (such as the propor- tion of low-income households and the proportion of adults with a tertiary education) [24], was used to ran- domly select one advantaged and one disadvantaged neighbourhood that were serviced by a Coles supermar- ket within 25 km of Deakin University in Melbourne, Australia. Two target stores from within these areas were then purposively selected, and a random sample of Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 3 of 15 3000 women identified via customer loyalty cards (Fly- Buys) who shopped regularly at these, or any other Coles stores with a 5 km radius, received a mail-out of a study recruitment pack. Women were targeted for recruitment because they tend to assume the majority of household food-related responsibilities [25, 26]. discussion boards and threads that coincided with news- letter content. Women in the combined price reduction + skill-building treatment arm received 20 % price dis- counts and all skill-building resources and supports. Interventions Following baseline measures (n = 32 did not return a baseline survey at all or in time to participate), 642 women were randomized using a computer-generated block randomization sequence into one of four study conditions: 1) Control (n = 161), 2) Price reductions (n = 161), 3) Skill-building (n = 160), or 4) Combined price reductions + skill-building (n = 160). The sample size provided 90 % power to detect an increase in vegetable consumption of at least 0.5 servings/day. Retrospective retrieval of 3 months of electronic purchasing data was followed immediately by a 3-month intervention period, with a subsequent 6-month no-intervention follow-up phase. Women in the control group did not receive any interventions during the study period, and were asked to maintain their usual shopping habits. The price reduc- tion intervention aimed to reduce the cost of targeted healthier foods through 20 % price discounts on the pur- chase of all FV (including all varieties of fresh, canned, dried and frozen FV, excluding fruit juice, frozen hot chips and fried potatoes), low-calorie carbonated bever- ages, and water. Discounts were applied at the checkout counter upon swiping a FlyBuys card and were in addition to existing store promotions and discounts. Par- ticipants received a list of discounted items at the start and mid-way through the intervention. The skill- building treatment arm was developed on the basis of social cognitive theory and addressed theoretical con- structs of nutrition-related self-efficacy, knowledge, and perceived facilitators and barriers for healthy eating through key messages provided in eight mailed newslet- ters and accompanying resources (e.g. recipes, goal- setting and self-monitoring exercises), and access to a dietitian-facilitated web-based forum that contained Data collection f h Items for the process evaluation were assessed via completion of self-report surveys at baseline (T1), post-intervention (T2: 3 months) and 6-months post- intervention (T3: 9 months). We aimed to achieve a broad coverage of major issues relating to the impact of the interventions. It was not our intent to explore the hidden meanings underlying why participants did or did not find the interventions to be impactful, thus a brief survey was most appropriate to our aims. To promote retention, all participants received small gifts, in- cluding a A$20 shopping voucher for each of three survey completions, 1000 FlyBuys rewards points (approximately equivalent to A$15 value), and water bottles, shopping bags, spice packs and tea bags. Questions for the process evaluation were purpose-developed for the SHELf study and tailored according to treatment group assignment and contained a mix of closed- and open-ended questions. Core measures were identical on all surveys (e.g. sociode- mographic details), however questions relating to specific elements of each intervention differed by intervention group (e.g. frequency of use of recipes, receipt of price dis- counts) as described below. The specific closed-ended questions asked and response options provided are pre- sented in Tables 1, 2 and 3. Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 4 of 15 Table 1 Process evaluation of the skill-building intervention at post-intervention Survey questions Skill-building n = 160a % (n) Combined n = 160a % (n) P-value Newsletters In the past 3 months, did you receive newsletters from us about healthy eating? (implementation) 0.493 Yes 98.0 (146) 99.3 (151) No 1 (2) 0 (0) Don’t know 0.7 (1) 0.7 (1) If yes, how useful did you find these newsletters? (implementation) 0.457 5 Extremely useful 14.9 (21) 17.2 (25) 4 34.0 (48) 29.0 (42) 3 30.5 (43) 35.2 (51) 2 18.4 (26) 13.8 (20) 1 Not at all useful 2.1 (3) 4.8 (7) How many of the 8 newsletters did you read? (implementation) 0.919 All 51.8 (75) 51.7 (78) About 75 % 19.3 (28) 22.5 (34) About 50 % 16.6 (24) 13.3 (20) About 25 % 9.7 (14) 9.9 (15) None 2.8 (4) 2.7 (4) On average how long did it take you to read through a newsletter? (implementation) (mean ± SD) 11.9 ± 12.8 mins 11.8 ± 8.8 mins 0.899 Recipes Did you use any of the recipes provided? Data collection f h (implementation) 0.114 Yes 45.6 (67) 53.6 (81) No 53.1 (78) 46.3 (70) Don’t know 1.4 (2) 0 (0) If yes, how many have you used? (implementation) 0.860 All 1.5 (1) 2.4 (2) 13–15 1.5 (1) 1.2 (1) 10–12 0 (0) 1.2 (1) 7–9 1.5 (1) 4.9 (4) 4–6 30.8 (20) 24.4 (20) 1–3 64.6 (42) 64.6 (53) How often have you used the recipes? (implementation) 0.815 Every day 0 0 Every 2–3 days 3.0 (2) 3.7 (3) About once a week 16.7 (11) 15.9 (13) About once every 2 weeks 34.9 (23) 28.1 (23) Table 1 Process evaluation of the skill-building intervention at post-intervention (Continued) About once a month 19.7 (13) 28.1 (23) Less than once a month 25.8 (17) 23.2 (19) Did you find the additional materials useful? (e.g. the Food for Health brochures, Food Cents Eat Smart for 4 booklets, etc.) (implementation) 0.335 Yes 65.1 (97) 72.8 (107) No 18.1 (27) 12.9 (19) Don’t know 16.8 (25) 14.3 (21) Did you share any of the newsletters, additional materials or recipes with friends or family over the past 3 months? (reach) 0.853 Yes 30.4 (45) 33.1 (49) No 68.9 (102) 66.2 (98) Don’t know 0.7 (1) 0.7 (1) What do you think of the amount of materials (newsletters, recipes, and other materials) you received from us? (implementation) 0.969 Too many materials 21.0 (31) 21.9 (32) About the right amount 76.4 (113) 75.3 (110) Did not receive enough 2.7 (4) 2.7 (4) RE-AIM constructs examined are listed in brackets after each question Fisher’s exact tests and χ2 tests (for categorical data) and two-sided unpaired t-tests with unequal variances were conducted to assess differences in survey responses between treatment groups. Unadjusted multinomial logistic regression with robust standard errors was used to assess differences between groups where Fisher’s or χ2 tests revealed overall differences aRepresents highest possible sample size for each question. The sample size for each question differs due to missing responses Table 1 Process evaluation of the skill-building intervention at post-intervention (Continued) Reach Reach refers to the proportion of the target population that participates in an intervention and their representa- tiveness [17]. Participants self-reported sociodemo- graphic characteristics including age, highest educational qualification, household income, country of birth, mari- tal status and number of children living in the home. Demographic characteristics of participants were com- pared to 2011 census data for the Greater Melbourne area [31] to determine the representativeness of partici- pants. We additionally examined the reach of the inter- ventions within participants’ social circles by asking whether they had shared skill-building intervention ma- terials or price discounts with others, and whether par- ticipants thought their partner and/or children (where applicable) were more willing to eat FV as a result of their participation in the study. Skill-building g Participants were asked whether they had received news- letters, how many they had read, how useful they found them, and the average time taken to read one. Partici- pants were also asked to qualitatively describe reasons why they may not have read all of the newsletters, the main messages they recalled from reading them, and the most and least useful parts. With respect to recipes, par- ticipants reported whether they had used recipes, how many they had used, and how often. Those who had not used recipes were asked to describe why this was the case. Finally, participants indicated whether they found the additional skill-building materials (i.e. brochures, booklets) useful, and whether the volume of skill-building materials provided was appropriate. Implementation l Implementation Implementation is a multi-dimensional construct that refers to what a program consists of when it is deliv- ered [32]. Effectiveness Effectiveness refers to the positive and negative impacts of interventions on outcomes [17]. To supplement previously Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 5 of 15 Page 5 of 15 reporte process tive im accesse caused and if n used an they ha as a res scribe a Implem Implem refers t ered [3 Skill-bui Particip letters, them, a pants w why th main m most an ticipant many th used re case. Fi the ad booklet materia Table 2 Process evaluation of the price reduction intervention at post-intervention Survey questions Price reduction n = 161a % (n) Combined n = 160a % (n) P-value Did you receive discounts on fruit and vegetables as part of this study? (implementation) 0.157 Yes 84.8 (134) 78.4 (120) No 5.7 (9) 11.8 (18) Don’t know 9.5 (15) 9.8 (15) If yes, did you buy more fruit and vegetables because of the discount? (effectiveness) 0.474 Yes 65.4 (87) 62.7 (74) No 28.6 (38) 33.9 (40) Don’t know 6.0 (8) 3.4 (4) Did you receive discounts on diet/ low-calorie soft drinks or water as part of this study? (implementation) 0.429 Yes 43.0 (68) 36.8 (56) No 30.4 (48) 36.8 (56) Don’t know 26.6 (42) 26.3 (40) If yes, did you buy more diet/low- calorie soft drinks or water because of the discount (effectiveness) 0.684 Yes 36.8 (25) 29.1 (16) No 58.8 (40) 67.3 (37) Don’t know 4.4 (3) 3.6 (2) If you received discounts do you think that you bought less sugar- sweetened/full calorie soft drinks because of the discount on diet/ low calorie soft drinks? (effectiveness) 0.389 Yes 23.5 (16) 21.8 (12) No 61.8 (42) 70.9 (39) Don’t know 14.7 (10) 7.3 (4) Please indicate how much you liked or disliked the discounts you were offered (implementation) 0.457 5 I liked the discounts very much 68.8 (108) 65.1 (99) 4 10.8 (17) 7.9 (12) 3 7.6 (12) 9.9 (15) 2 1.9 (3) 1.3 (2) 1 I did not like the discounts 3.2 (5) 2.0 (3) Did not receive 7.6 (12) 21 (13.8) If you received discounts, did you use the ‘supermarket items included in our 20 % price di t’ h d t t 0.193 Table 2 at post-i Yes No Don’t k If you rec buy thing discount? Effectiveness Yes No Don’t k RE-AIM co Fisher’s ex t-tests wit responses with robus where Fish aRepresen for each q Table 2 Process evaluation of the price reduction intervention at post-intervention (Continued) Yes 53.1 (77) 45.8 (60) No 33.1 (48) 43.5 (57) Don’t know 13.8 (20) 10.7 (14) If you received discounts did you buy things for others using your discount? (reach) 0.450 Yes 13.8 (20) 11.5 (15) No 83.5 (121) 87.8 (115) Don’t know 2.8 (4) 0.8 (1) RE-AIM constructs examined are listed in brackets after each question Fisher’s exact tests and χ2 tests (for categorical data) and two-sided unpaired t-tests with unequal variances were conducted to assess differences in survey responses between treatment groups. Unadjusted multinomial logistic regression with robust standard errors was used to assess differences between groups where Fisher’s or χ2 tests revealed overall differences aRepresents highest possible sample size for each question. The sample size for each question differs due to missing responses Table 2 Process evaluation of the price reduction intervention at post-intervention (Continued) reported main outcomes from the SHELf study [13], this process evaluation assessed perceived positive and nega- tive impacts of the interventions by asking those who accessed discounts whether the price reductions had caused them to purchase more of the discounted items, and if not, why this was the case, as well as how they had used any saved funds. All participants were asked whether they had changed the way they bought, cooked or ate food as a result of study participation, and to qualitatively de- scribe any changes they had made. Price reductions Participants reported whether they had accessed price discounts on FV and on low-calorie soft drinks and Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 6 of 15 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 6 of 15 Table 3 Process evaluation of the overall study at post-intervention Survey questions Control n = 161a % (n) Price reduction n = 161a % (n) Skill-building n = 160a % (n) Combined n = 160a % (n) P-value Have you changed the way you buy, cook or eat food after taking part in this study? (effectiveness) <0.001 Yes 14.7 (23)b 32.9 (52)c 34.5 (51)c 37.5 (57)c No 77.6 (121) 59.5 (94) 52.0 (77) 54.6 (83) Don’t know 7.7 (12) 7.6 (12) 13.5 (20) 7.9 (12) (If you have children aged 12 years of younger) Do you think your child/ children are more willing to eat fruit and vegetables as a result of you taking part in this study? (reach) 0.227 Yes 15.4 (14) 26.0 (19) 22.1 (15) 29.7 (19) No 61.5 (56) 58.9 (43) 54.4 (37) 45.3 (29) Don’t know 23.1 (21) 15.1 (11) 23.5 (16) 25.0 (16) (If you have a partner) Do you think your partner is more willing to eat fruit and vegetables as a result of you taking part in this study? (reach) 0.017 Yes 18.1 (23)d 34.7 (42)e 30.6 (34)e 37.1 (43)e No 59.1 (75) 45.5 (55) 46.9 (52) 38.8 (45) Don’t know 22.8 (29) 19.8 (24) 22.5 (25) 24.1 (28) RE-AIM constructs examined are listed in brackets after each question Fisher’s exact tests and χ2 tests (for categorical data) and two-sided unpaired t-tests with unequal variances were conducted to assess differences in survey responses between treatment groups. Unadjusted multinomial logistic regression with robust standard errors was used to assess differences between groups where Fisher’s or χ2 tests revealed overall differences aRepresents highest possible sample size for each question. The sample size for each question differs due to missing responses b,cValues for variables without a common letter differ (p < 0.0001) d,eValues for variables without a common letter differ (p < 0.05) Table 3 Process evaluation of the overall study at post-intervention of recipes (yes vs no), and whether price discounts on FV (yes vs no) and beverages (yes vs no) were accessed. Maintenance Maintenance measures the extent to which newly ac- quired behaviours are maintained over time [17], and was assessed at T3 (6-months post-intervention). Price reductions Out- comes for the analysis included purchase (grams/week) and consumption (servings/day) of FV (including all var- ieties of fresh, canned, dried and frozen FV, excluding fruit juice, frozen hot chips and fried potatoes), and purchase (millilitres/week) and consumption (servings/day) of car- bonated beverages and water. As described in a previous publication, consumption data were self-reported via a Food Frequency Questionnaire that used validated ques- tions, while purchasing outcomes were assessed using electronic sales data obtained from customer store loyalty cards [13]. water. A subsequent question asked whether participants had ever consulted the list of discounted items they had been given. Participants were also requested to quantita- tively rate and qualitatively describe the degree to which they liked the price reductions overall. Overall Participants were asked to describe what they liked most and least about the study. Impact of intervention dose Two dose-related variables were selected for each of the skill-building and price reduction interventions to evalu- ate whether the dose of the interventions received influ- enced purchase and consumption of FV and beverages at T2 (immediately post-intervention). Within the skill- building intervention, dose-related variables selected were those where there was sufficient variability in re- sponses (e.g. we did not consider whether participants had received newsletters as nearly 99 % reported receiv- ing them), and where responses were characterized by meaningful cut-points (e.g. number of recipes used was not examined as most participants used 1–3 or 4–6 rec- ipes and the difference between these was not theoretically meaningful). Thus, the dose-related variables analysed were: number of newsletters read (100 % vs < 100 %), use Quantitative findings g A total of 642 of the 3,000 women identified by Coles supermarkets as shopping at least once every 2 weeks within the target stores or any other Coles within a 5 km radius were randomized. Three individuals actively withdrew from the study following randomization, thus 21.3 % of the target population participated. Notably, this proportion might have been higher had we devoted additional efforts to enrolling additional non-responders. Demographic characteristics of participants have been previously presented [13]. Participants were nearly evenly split according to neighbourhood-level socioeconomic sta- tus (SES), with 44 % of participants residing in low, and 55 % in high SES areas [13]. Comparison of participant characteristics with the general population of females liv- ing in the Greater Melbourne region showed that a higher proportion of women enrolled in SHELf were born in Australia (71 % vs 63 %), and compared to the population of females age ≥20 years, a greater proportion of women in SHELf were married (71 % vs 55 %), and had less than a high school education (12 % vs 35 %). Compared to all households in Melbourne, a lower proportion of women in SHELf had a household income < $52,000 AUD/year (24 % vs 34 %). Self-reported consumption of fruit (1.9 servings/day) and vegetables (2.5 servings/day) was higher than the Australian average of 1.0 servings/day and 2.2 servings/day, respectively, among women ≥19 years [2]. Linear regression analyses assessed differences in FV and beverage purchasing and consumption at T2 ac- cording to the dose-related variables selected for the analysis. Individuals in the combined group were pooled with those in the price reduction or skill-building groups for interventions that both groups received (e.g. those who accessed price reductions on FV in the price reduc- tion and combined groups were pooled, those who used recipes in the skill-building and combined groups were pooled, etc.) to assess the impact of all four dose-related variables (i.e. number of newsletters read, use of recipes, whether price discounts on FV and beverages were accessed). All models controlled for baseline values of the outcome and intervention group, and for the follow- ing a priori-determined covariates: age, country of birth, catchment area, marital status, household income, and the number of children living at home. Given the skewed distributions of several outcomes, bootstrapping with 1000 resamples was used to produce more robust stand- ard errors. Results Participants reported whether they had changed the way they bought, cooked or ate food as a result of the study, and whether their partner and/or children (where applic- able) were more willing to eat FV as a result of their in- volvement in the study. We present quantitative findings first, followed by qualita- tive findings which provide context for, and help to ex- plain the reasons underlying the quantitative results. We chose to summarize qualitative responses rather than to present specific quotations given the brief nature of the open-ended responses provided. Full quotations would not have provided additional information in this instance. Skill-building Participants were asked whether they had continued to use any study materials in the post-intervention period, to report the number of materials used, and the fre- quency with which they had used them. Participants also indicated whether they had shared any intervention ma- terials with friends/family in the past 6 months. Page 7 of 15 Page 7 of 15 Page 7 of 15 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Statistical analyses Data from all participants who completed any portion of the T2 (n = 619) and T3 (n = 608) surveys were included, thus the sample size differed for each question. Fisher’s exact tests and χ2 tests (for categorical data) and un- paired t-tests with unequal variances (for continuous data) were conducted to assess differences in survey re- sponses between treatment groups. Unadjusted multi- nomial logistic regression with robust standard errors was used to assess differences between groups where Fisher’s or χ2 tests revealed overall differences, with ‘no’ as the reference category. Qualitative data analysis Qualitative data were coded in an inductive manner by a single investigator using thematic analysis procedures [33]. NVivo software (version 10.0; QSR International, Doncaster, VIC, Australia) was used to organize the data. Participants were asked whether they had changed their purchasing of FV and of soft drinks and water in the past 6 months. Quantitative findings Nearly two-thirds of participants who reported that they had accessed price discounts on FV said they purchased more FV because of the discounts, while the reverse was true among those receiving discounts on low-calorie car- bonated beverages and water, with approximately two- thirds reporting they did not purchase more low-calorie carbonated beverages and water, or fewer high-calorie carbonated beverages because of the discounts (Table 2). Qualitative findings Reported changes made as a result of study involvement were very consistent, as participants overwhelmingly in- dicated that they had increased their purchase and/or consumption of FV, and of vegetables in particular. Many said they had increased the variety of foods they ate, especially of FV, and more often purchased these items fresh. Very few participants reported changing their beverage purchasing or consumption patterns as a result of study involvement. Some participants described changes they had made in more general terms, saying they had increased healthy, and reduced unhealthy eat- ing. Additional changes made primarily by those who re- ceived skill-building materials included planning meals and shopping trips more often, incorporating FV within recipes, and increased awareness of food choices, al- though these types of changes were also reported by in- dividuals in the price reduction and control groups. Individuals who received price discounts also said they had increased purchasing of FV at Coles to take advantage of the price discounts. Only three participants suggested the possibility of unintended negative consequences, in- cluding eating more unhealthy food or feeling guilty about eating unhealthy foods during the intervention period, al- though it was not clear whether they attributed these to the interventions specifically. Quantitative findings g Nearly all participants randomized to receive skill- building materials reported receiving newsletters, with more than half reading all of them (Table 1). On average, participants spent just under 12 min reading each news- letter. On a 5-point scale where 5 indicates ‘extremely useful’ and 1 indicates ‘not at all useful’, 80 % of partici- pants who reported receiving newsletters gave them a score of 3 or higher. Although participants said they ap- preciated the recipes within the newsletters, only ap- proximately half reported using the recipes provided, with most using only 1–3 recipes and doing so on an in- frequent basis (i.e. no more than once every 2 weeks). Overall, three-quarters of participants thought the amount of material provided in the skill-building intervention was ‘about right’. Qualitative findings Those who did not increase their purchasing of FV said they had not done so because they already purchased what they required and additional purchases would have gone to waste. Others said they already purchased their FV elsewhere, often for reasons related to perceived cost and/or quality. Notably, a number indicated that al- though they had not increased their overall FV purchas- ing, they were now buying a greater proportion of their FV at Coles to take advantage of the price discounts they received through the study. Themes that emerged with respect to why so many participants did not increase purchasing of low-calorie carbonated beverages and water in response to price discounts related to the fact that many participants tended not to buy these bever- ages, often because they preferred tap water or wanted to avoid artificial sweeteners. Other participants indi- cated that they did not increase purchasing of dis- counted beverages simply because they did not require more of them. Participants were asked to report how they used the funds they saved from the FV and beverage discounts. Many reported using them to purchase more food over- all, with a large constituency indicating they spent these funds on FV specifically. Very few used the money to purchase additional discounted beverages. Others used their savings to fund other household-related expendi- tures, while still others used the money to purchase ‘lux- ury’ items they would not normally buy, such as more expensive types of FV, restaurant meals or magazines. One participant used the funds to purchase cigarettes. This was the only instance in which it was possible to ascertain that saved funds had clearly been used to pur- chase unhealthy items. Finally, some participants did not spend the saved funds or were unsure how they had used them, often indicating this was because the savings were too small to notice. Quantitative findings Participants who withdrew from the study (n = 3) and who were missing main outcomes from T1 (n = 10) or T2 (n = 30) were excluded. Participants who responded ‘don’t know’ (see numbers in Tables 1 and 2) to individual questions were additionally excluded from these analyses, as the dose received was uncertain in these cases. The study’s reach extended to some of the families and friends of participants. Just under one-third of partici- pants who received skill-building materials shared them with family/friends (Table 1), while 13 % of those who reported they had accessed the price discounts shared them with others (Table 2). Compared to individuals in the control group, participants in the intervention groups reported that their partners were more willing to consume FV as a result of their involvement in the study (p < 0.05), although their children were not (Table 3). All statistical tests were two-sided and the significance level was set at p < 0.05. Stata software (version 13; Sta- taCorp LP, TX) was used for all statistical analyses. Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 8 of 15 Effectiveness (35 %) said they had changed the way they bought, cooked or ate food as a result of study involvement (p < 0.001), with no differences among intervention groups (Table 3). Qualitative findings The vast majority of those who did not read all eight newsletters failed to do so due to time constraints, with many indicating they had put them aside to read later. Others indicated that they had not read all of the news- letters because the information they contained was too simplistic and they already knew much of it. Participants overwhelmingly appreciated the recipes that newsletters provided. Many also liked the information about FV and sections providing practical tips and time-efficient strat- egies for healthy living. Most participants said there was nothing they disliked about the newsletters. The primary messages they recalled from reading newsletters were often very general messages emphasizing the importance of eating healthfully, eating more FV, and healthy lifestyles. When participants were asked what they liked most about the study, increased awareness of current eating behaviours was a strong theme across all groups, includ- ing those in the control group. This was attributed to the chance for self-reflection through completing sur- veys, information provided in the skill-building interven- tion, as well as through the foods to which discounts were applied. Appreciation for the incentives and dis- counts provided through the study emerged as another strong and consistent theme. Many participants also val- ued the opportunity to contribute to research, as they found the study interesting and worthwhile, and enjoyed completing the surveys. Those in the skill-building arms found the newsletters and recipes helpful and liked learning new things. The vast majority of participants in- dicated that they did not dislike anything about the study. The small number of participant-expressed con- cerns related primarily to the surveys, as some disliked the repetitive nature of the questions, the length of the surveys, or that they could not always find their preferred response options. Others disliked the non-monetary in- centives, and felt this money could have been better spent elsewhere. Complaints related to the discounts were rare, whereas concerns were expressed with the newsletters, in terms of their messaging, the frequency with which they were received (i.e. too frequent), and their length (i.e. too long). Finally, environmental concerns were raised regard- ing the amount of packaging required to post non- monetary incentives, and provision of paper-based news- letters and surveys. Quantitative findings More than 80 % of individuals in the price reduction groups reported accessing discounts on FV, with just under 40 % reporting they had accessed the beverage discounts (Table 2). More than two-thirds liked the col- lective discounts ‘very much’. Quantitative findings The majority of participants (85 %) signed up to par- ticipate in the web-based forums. Of those participants who signed up, 25 % logged onto the forums at least Compared to those in the control group (15 %), signifi- cantly more participants in the intervention groups Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 9 of 15 Page 9 of 15 once, although only 10 % actively posted on the forums during the intervention period. small, and that even with the discounts they could still purchase less expensive produce elsewhere. Overall Qualitative findings Qualitative findings The vast majority of those who did not read all eight newsletters failed to do so due to time constraints, with many indicating they had put them aside to read later. Others indicated that they had not read all of the news- letters because the information they contained was too simplistic and they already knew much of it. Participants overwhelmingly appreciated the recipes that newsletters provided. Many also liked the information about FV and sections providing practical tips and time-efficient strat- egies for healthy living. Most participants said there was nothing they disliked about the newsletters. The primary messages they recalled from reading newsletters were often very general messages emphasizing the importance of eating healthfully, eating more FV, and healthy lifestyles. Most of those who did not use the recipes had not done so because they already had a collection of recipes they enjoyed and were accustomed to using, which in some cases were similar to those provided by SHELf. Others indicated they had not had the time to use the recipes, while many said they did not find the recipes appealing, primarily because they were too basic. A number said the recipes did not fit with their current diet (followed for cultural, medical or lifestyle-related reasons). Qualitative findings Most of those who did not use the recipes had not done so because they already had a collection of recipes they enjoyed and were accustomed to using, which in some cases were similar to those provided by SHELf. Others indicated they had not had the time to use the recipes, while many said they did not find the recipes appealing, primarily because they were too basic. A number said the recipes did not fit with their current diet (followed for cultural, medical or lifestyle-related reasons). Qualitative findings Aspects of the discounts that participants most enjoyed were that they allowed them to save money, primarily on FV, with very few mentioning the beverage savings. Many indicated the discounts allowed them to buy more and/or a greater variety of FV, particularly the more ex- pensive types. Participants said that receiving discounts made them feel appreciated and like they were being rewarded for making healthy choices. A number also in- dicated that they shifted their FV purchasing from other stores to Coles to take advantage of the discounts. In adjusted analyses, individuals who said they read all of the newsletters had a higher intake of low-calorie car- bonated beverages and water (0.97 servings/day, 95 % CI 0.34–1.60; p < 0.01) compared to those who said they did not read all of the newsletters immediately post- intervention, while those who reported using the pro- vided recipes had a higher intake of fruits (0.24 servings/ day, 95 % CI 0.03–0.45; p < 0.05) and vegetables (0.27 servings/day, 95 % CI 0.05–0.49); p < 0.05) compared to those who did not use these recipes (Table 4). Those who reported accessing the FV discounts purchased (518.86 g/week, 95 % CI 158.41–879.31; p < 0.01) and consumed (0.48 servings/day, 95 % CI 0.13–0.84; p < 0.01) more vegetables compared to those who said they did not access discounts on FV. Those who said they accessed beverage discounts purchased more low-calorie Participants expressed very few complaints related to the discounts, and were disappointed when they ended. Many wished the discounts had applied to more, or even all supermarket items. Several indicated they did not like the beverage discounts because they were not useful for them. Some indicated the size of the discounts was too Olstad et al. Qualitative findings International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Table 4 Impact of intervention dose on purchase and consumption of fruits, vegetables and beverages at post-intervention Dose-related variables Fruit purchase (g/week) Fruit consumption (servings/day)a Vegetable purchase (g/week) Vegetable consumption (servings/day)a Low-calorie carbonated beverage and water purchase (mL/week) Low-calorie carbonated beverage and water consumption (servings/day)a High-calorie carbonated beverage purchase (mL/week) High-calorie carbonated beverage consumption (servings/day)a β (95 % confidence interval) Number of newsletters read All vs < allb (n = 285) 2.49 (−241.77, 246.75) −0.06 (−0.27, 0.14) 113.82 (−103.66, 331.31) 0.06 (−0.17, 0.30) 188.98 (−502.78, 880.74) 0.97** (0.34, 1.60) 507.94 (−407.14, 1423.02) −0.07 (−0.16, 0.01) Used recipes Yes vs nob (n = 282) 128.72 (−114.44, 371.88) 0.24* (0.03, 0.45) −13.87 (−233.24, 205.50) 0.27* (0.05, 0.49) 248.27 (−317.73, 814.26) 0.38 (−0.24, 1.00) 95.23 (−635.92, 826.38) −0.05 (−0.16, 0.06) Accessed price discounts on fruit and vegetables Yes vs nob (n = 274) 238.93 (−152.89, 630.75) 0.28 (−0.08, 0.63) 518.86 (158.41, 879.31) 0.48 (0.13, 0.84) n/a n/a n/a n/a Accessed price discounts on beverages Yes vs nob (n = 221) n/a n/a n/a n/a 1004.24* (134.44, 1874.03) 0.48 (−0.23, 1.19) 1074.80 (−656.43, 2806.03) −0.02 (−0.13, 0.08) Beta coefficients represent the between-group mean differences in purchasing and consumption of fruits, vegetables and beverages. Linear regression models controlled for baseline values of the outcome, interven- tion group, age, country of birth, catchment area, marital status, household income, and the number of children living at home a1 serving of fruit = 150 g; 1 serving of vegetables = 75 g; 1 serving of beverage = 125 mL bIndicates reference category p < 0.05; *p < 0.01; bolded values highlight significant differences on purchase and consumption of fruits, vegetables and beverages at post-intervention Table 4 Impact of intervention dose on purchase and consumption of fruits, veget Page 11 of 15 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 carbonated beverages and water (1004.24 mL/week, 95 % CI 134.44–1874.03; p < 0.05) compared to those who said they did not access these discounts. framework proved valuable in guiding this analysis, and in collating the most consequential and policy-relevant findings. Overall, the reach of the SHELf interventions to disadvantaged groups, and beyond study participants themselves was moderate. Quantitative findings A minority (27 %) of those who said they had accessed price reductions during the intervention period reported that they changed their purchasing of FV after the price reductions stopped, while even fewer (9 %) reported that they had changed their beverage purchasing patterns in this period. Quantitative findings At the 6-month follow-up, similar proportions of indi- viduals in all groups (31 %) reported that the study had influenced the way they purchased, cooked or ate food. There were no differences among groups in the propor- tion of participants who reported their partners (31 %) and/or children (28 %) were more willing to eat FV as a result of their involvement in the study. Relative to those in the control group, significantly more participants in the intervention groups reported that the study had influenced the way they purchased, cooked or consumed food at post-intervention, with no differences among intervention groups. This suggests that the skill-building intervention may have had modest positive impacts that were not captured in our earlier analysis, either because our measures were not suffi- ciently inclusive or because they lacked sensitivity. The former appears most likely, as those who received skill- building materials indicated that study involvement prompted them to plan shopping trips and meals more often, to incorporate more FV during cooking, to be more aware of their dietary behaviours, and to make other changes we did not previously assess. A focus on purchasing and consumption outcomes may have been responsible for similarly null findings in other educa- tional interventions in supermarkets [15, 16], pointing to the need for broader outcome assessments to capture the potential benefits of skill-building interventions in this context. While health benefits can only be achieved through improved dietary behaviours, behaviour change Qualitative findings With respect to their maintenance of intervention- promoted behaviours, participants described attempts to increase healthy eating, primarily through including more or a greater variety of FV in their diets, and to drink more water. Participants also mentioned attempts to reduce unhealthy eating, such as through drinking fewer sugar-sweetened beverages. Many had become more aware of the importance of healthy eating. Others de- scribed more frequently planning their meals and shop- ping, and using more, or a greater variety of recipes. Qualitative findings Participants indicated limited appreciation for, and use of beverage compared to FV discounts. All of the interventions were at least moder- ately efficacious in changing the way participants bought, cooked or ate food, with the strongest impacts observed among those who received a greater interven- tion dose. Finally, maintenance of newly acquired behav- iours proved challenging. Quantitative findings Nearly 44 % of individuals who received skill-building materials reported using them in the 6-month post- intervention period, with 32 % using newsletters and 47 % using recipes, albeit on an infrequent basis (i.e. no more than once every 2 weeks). Only 23 % reported sharing skill-building materials with family/friends in the post-intervention period. It is important to understand the degree to which a program reaches those most in need of intervention [17]. Socioeconomic disadvantage is a risk factor for poor dietary quality, including low FV consumption [4, 35–39] and therefore we sought to ensure equal repre- sentation of high and low SES groups through targeted recruitment procedures. Although the sample was nearly evenly split according to neighbourhood-level SES, less than a quarter of participants were from lower income groups according to household-level measures. Dispar- ities in study participation have also been reported in other supermarket-based studies despite the use of tar- geted recruitment strategies [16, 20]. SHELf had some success in reaching participants’ friends and family, with nearly one-third of participants sharing skill-building materials, 13 % sharing discounts, and an increased will- ingness of some participants’ partners (34 %) to consume FV due to the study. Moreover, community-wide impacts of SHELf were evident, as rollout of Coles’ fresh produce ‘super-specials’ was informed by study findings [13]. Discussion The challenge in the former in- stance will be to balance intervention intensity with program reach, as participants’ time constraints may limit participation in a more intensive study. Moreover, given the low usage of the web-based forums compared to the mailed resources, we would recommend combin- ing web- and paper-based strategies in future studies, as evidence suggests that the internet is the predominant, and an increasing source of nutrition information for consumers [40], and that messages may be reinforced when delivered through multiple channels [41]. Notably, although less preferred, this process evaluation revealed that the beverage discounts were not entirely without effect, as one-third of those who accessed bever- age discounts reported purchasing more low-calorie car- bonated beverages and water as a consequence, while those who reported accessing beverage discounts pur- chased an additional litre of these beverages per week. Purchase of low-calorie carbonated beverages and water did not displace purchases of high-calorie carbonated bev- erages among these individuals, however, as purchasing of high-calorie carbonated beverages remained stable. Thus the true impact of the beverage discounts is unclear. It is possible that these beverages are not direct alternatives and that the impacts observed were due to individuals who already regularly purchased low-calorie carbonated beverages and water using price discounts to “stock up” on these items. The current findings support our previously reported results showing positive outcomes of FV discounts [13], as nearly two-thirds of participants indicated that they pur- chased more FV due to the discounts. Those who re- ported using the FV discounts, and who therefore received the intended FV “discount dose”, used it primar- ily to purchase and consume more vegetables, rather than fruit. This coincides with qualitative findings, as partici- pants indicated a particular focus on increasing purchase and consumption of vegetables. Price reductions on FV appeared more effective than price reductions on bever- ages. Overall, study findings are therefore consistent with the notion that price elasticities differ across foods [44, 45]. Price discounts were clearly not sufficient to induce all participants to purchase and consume more of the discounted items, as consumers do not select foods solely based on price. Thus, multi-component interventions that simultaneously target multiple determinants of food selection and consumption are essential to im- prove population-level dietary behaviours. Discussion SHELf was a RCT that compared the impact of individ- ual, environmental and a combined approach to promot- ing healthy eating in the supermarket setting, a key environment for influencing food selection and con- sumption [34]. Process evaluation was undertaken with the aim of elucidating factors implicated in the study’s previously reported main outcomes [13]. The RE-AIM Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 12 of 15 Page 12 of 15 Page 12 of 15 is an iterative process. Thus it remains important to quantify changes in attitudes and beliefs arising from this study as they might prompt future dietary change. may also explain similar null findings by others [43]. Comments indicating that some participants who re- ceived price discounts displaced their FV purchasing from other venues into study supermarkets to take ad- vantage of price discounts should encourage supermar- kets to offer FV at affordable prices. Such comments also suggest implications for the validity of food pur- chasing data in supermarket-based pricing interventions (i.e. apparent increases in FV purchasing could be partly related to a displacement effect), and will therefore be examined in a subsequent quantitative study. y y g p p y g In the SHELf study researchers controlled the dose of skill-building materials delivered to participants, how- ever the dose received was under individual control. Thus, participants had to actively engage with the inter- vention by reading newsletters, using recipes, and com- pleting skill-building exercises. While approximately half of participants said they read all eight newsletters and used recipes, they spent an average of just 12 min read- ing each newsletter, and most used between one and three of the recipes provided. Thus, a low level of en- gagement with skill-building materials may partly ex- plain their lack of impact on food purchasing and consumption in the full sample, as among those who re- ceived a higher dose of intervention, positive, albeit modest, benefits were observed. At least two important implications emerge from these findings. First, newslet- ters and recipes may be core aspects of the skill-building intervention that should be retained in future studies, and second, a more intensive, structured and/or more engaging intervention format might be required to en- sure a sufficient dose. Discussion Similarly, positive impacts of the study on the reported willingness of inter- vention participants’ partners to increase their FV con- sumption at T2 disappeared at T3 largely due to reported study-associated increases in FV consumption among part- ners of control group participants in the post-intervention period. There is currently a strong focus on evaluating the im- pact of health promotion interventions, however less emphasis is placed on process evaluation [50]. This study was unique because of its strong design, real- world supermarket setting, grounding in an explicitly socioecological approach, multi-sectoral partnerships (i.e. Heart Foundation, Coles® supermarkets), and collec- tion of quantitative and qualitative process-related data at multiple time points. These data can help to ascertain which aspects of the SHELf interventions were most useful, and provide more complete information for deci- sion makers in deciding whether it is worthwhile to initi- ate or continue to support similar programs. While substituting healthier for less healthy products in the diet is desirable, simply adding more food, even healthy foods such as FV, can lead to weight gain. Stud- ies conducted in real-world supermarket settings have reported no change in overall food expenditures [15, 16], purchase of unhealthy foods [15], or total caloric intake [43] in response to price discounts on FV and healthier beverages. At least three studies in simulated supermar- ket environments have, however, reported counterpro- ductive outcomes of pricing interventions whereby discounting of healthy foods has led to higher calorie purchases [46, 47], purchase of more items overall [47] or of additional unhealthy foods [46, 48]. Qualitative comments indicating that some participants used their savings from price discounts to purchase more food overall, or more FV specifically, suggest the possibility of unintended negative consequences (i.e. increased caloric intake, although recognizing that increased FV intake confers health benefits) that should be probed in future analyses. In addition, the broader economic and health implications of food-related price reductions should be examined, as individuals could theoretically use saved funds on a variety of other more or less healthy items or activities. Our sample was more highly educated and had a higher income than the general population of women living in Melbourne, and therefore the generalizability of study findings is unclear. The process measures were self-reported, and may therefore be influenced by social desirability and recall biases, however these should be relatively balanced across groups. Discussion g p Price reductions on FV proved extremely popular, with > 80 % of participants reporting they had accessed these, compared to < 40 % who accessed beverage dis- counts. Herman et al. [42] also found that FV discounts were highly accessed, with redemption rates for FV vouchers of 90 % among low-income women, whereas Waterlander et al. [16] recorded much lower redemption rates of 15–45 %. Qualitative comments further substan- tiated these findings, as participants overwhelmingly in- dicated that they appreciated and used FV discounts, whereas only a small fraction mentioned liking beverage discounts. Indeed, many deemed the beverage discounts useless because they tended not to buy carbonated bev- erages and water. This coincides with results from the 2011–12 Australian Health Survey, where a majority of females reported consuming fruits (65 %) and vegetables (77 %) on a given day, whereas fewer consumed regular (17 %) and diet (9 %) soft drinks [2]. These findings highlight factors underlying the absence of positive im- pacts of beverage discounts in our previous analysis, and Main outcomes from the SHELf study suggested the possibility of unintended negative consequences, as indi- viduals in some groups increased purchase and/or con- sumption of high-calorie carbonated beverages at T2 and T3 [13]. We previously speculated that participants may have used saved funds to purchase more high- Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 13 of 15 Page 13 of 15 calorie beverages (a substitution effect), or that behav- iour change activities may have unintentionally pro- moted increased intake of these beverages [13]. This process evaluation, however, found no evidence for these suppositions, as participants did not specifically mention using saved funds to purchase high-calorie beverages, al- though they did report using saved funds to purchase additional unspecified ‘foods’, which conceivably could have included high-calorie beverages. In addition, very few participants recalled any beverage-related messaging associated with the skill-building intervention, and those messages that were recalled were positive (e.g. drink more water). Thus, the observed increase in purchase and consumption of high-calorie carbonated beverages appears unrelated to the SHELf interventions. responses suggest that these positive impacts may have been partly related to the opportunity for additional self- reflection on personal eating behaviours occasioned by completion of a second survey at T2. Discussion Survey questions were not validated or examined for their reliability, nor were we able to use multiple methods (e.g. individual interviews) to triangulate findings. The consistency of responses within and across questions, however, suggests that participants correctly understood the nature of the questions. Finally, we did not have purchasing and consumption data for all foods and beverages, and we were therefore limited in our ability to assess substitution effects. Implications and conclusions While many interventions have been successful in pro- ducing short-term behaviour change, benefits often erode over time with the resumption of previous behav- ioural patterns [49]. Maintenance of newly established behaviours proved challenging in the SHELf study, al- though successes were achieved in some areas. Overall, positive perceived impacts of the interventions on the way participants bought, cooked and consumed food were largely maintained in the post-intervention period, but were no longer statistically significant, seemingly due to positive impacts of the study on these behaviours amongst participants in the control group in the post- intervention period (i.e. ~31 % of participants in all groups reported that the study had positively influenced the way they bought, cooked or ate food). Qualitative Given the resources required to implement public health programs, the potential value of new programs should not be judged solely on the basis of their efficacy, but on a thorough and comprehensive analysis of all aspects of a program. SHELf was the first supermarket-based RCT to perform a thorough process evaluation assessing the reach, effectiveness, implementation and maintenance of a skill-building, price discount, and combined interven- tion. Process evaluation proved critical to a correct inter- pretation of study outcomes by enabling a more holistic assessment of intervention impact, and by accounting for factors such as their level of implementation. Based on these findings we have formulated several recom- mendations for future nutrition-related interventions in Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Page 14 of 15 Page 14 of 15 supermarkets. First, given our mixed success in recruit- ing disadvantaged women into the SHELf study (i.e. while 44 % of women lived in disadvantaged neighbour- hoods, participants were more advantaged than average Melbournians according to individual-level SES measures) more specific actions may be required to minimize socio- economic disparities in study participation where examin- ing impact on disadvantaged populations is an explicit goal. Second, given that intervention impact varied ac- cording to the dose received, future studies should exam- ine the impact of multiple doses of intervention. Third, given the importance of ensuring delivery of a sufficient intervention dose without overburdening participants, fu- ture studies could vary intervention intensity to assess feasibility and impact. Fourth, our inability to examine food substitution effects limited our conclusions related to the ultimate health impacts of our findings, and thus fu- ture studies should examine these where possible. References 1. Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012; 380(9859):2224–60. 2. Australian Bureau of Statistics. 2014. Australian Health Survey: Nutrition First Results - Foods and Nutrients, 2011–12. http://www.abs.gov.au/ausstats/ abs@.nsf/Lookup/by%20Subject/4364.0.55.007~2011-12~Main%20Features~ Key%20Findings~1. Accessed January 5, 2015. 2. Australian Bureau of Statistics. 2014. Australian Health Survey: Nutrition First Results - Foods and Nutrients, 2011–12. http://www.abs.gov.au/ausstats/ abs@.nsf/Lookup/by%20Subject/4364.0.55.007~2011-12~Main%20Features~ Key%20Findings~1. Accessed January 5, 2015. 2. Australian Bureau of Statistics. 2014. Australian Health Survey: Nutrition First Results - Foods and Nutrients, 2011–12. http://www.abs.gov.au/ausstats/ abs@.nsf/Lookup/by%20Subject/4364.0.55.007~2011-12~Main%20Features~ Key%20Findings~1. Accessed January 5, 2015. 3. Brownell KD, Kersh R, Ludwig DS, Post RC, Puhl RM, Schwartz MB, et al. Personal responsibility and obesity: a constructive approach to a controversial issue. Health Aff (Millwood). 2010;29(3):379–87. Supermarkets represent a critical, yet so far underuti- lized venue for influencing dietary behaviours at a population-level. This study points to the potential to le- verage the power of supermarkets within food systems to virtuous ends, contributing to accomplishment of key public health objectives that might otherwise be more difficult, and take much longer to achieve. Despite mod- est reported effectiveness, even small changes among some population segments can shift population-level dietary intake distributions in healthier directions. It is hoped that the collective successes and failures of the SHELf intervention described herein will inform devel- opment and implementation of more effective policies and programs to optimize food selection in this setting. 4. Ball K, Crawford D, Mishra G. Socio-economic inequalities in women’s fruit and vegetable intakes: a multilevel study of individual, social and environmental mediators. Public Health Nutr. 2006;9(5):623–30. 5. Shaikh AR, Yaroch AL, Nebeling L, Yeh MC, Resnicow K. Psychosocial predictors of fruit and vegetable consumption in adults a review of the literature. Am J Prev Med. 2008;34(6):535–43. 6. Thornton LE, Lamb KE, Tseng M, Crawford DA, Ball K. Does food store access modify associations between intrapersonal factors and fruit and vegetable consumption? European journal of clinical nutrition. 2015. 6. Thornton LE, Lamb KE, Tseng M, Crawford DA, Ball K. Does food store access modify associations between intrapersonal factors and fruit and vegetable consumption? European journal of clinical nutrition. 2015. 7. Smith KJ, McNaughton SA, Cleland VJ, Crawford D, Ball K. Competing interests Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. 13. Ball K, McNaughton S, Le H, Gold L, Ni Mhurchu C, Abbott G, et al. Influence of price discounts and skill-building strategies on purchase and consumption of healthy food and beverages: outcomes of the Supermarket Healthy Eating for Life randomized controlled trial. Am J Clin Nutr. 2015; 101(5):1055–64. Implications and conclusions Finally, future studies could also measure additional relevant out- comes such as change in food preparation, shopping and awareness of food selection, as these may contribute to additional behaviour change over time. FT100100581. The study funders had no role in the collection, analysis and interpretation of data, or in writing the manuscript. The contents of this paper are the responsibility of the authors and do not reflect the views of NHMRC. FT100100581. The study funders had no role in the collection, analysis and interpretation of data, or in writing the manuscript. The contents of this paper are the responsibility of the authors and do not reflect the views of NHMRC. Authors’ contributions DLO analysed and interpreted the data, wrote the manuscript; KB designed the study, supervised data collection, interpreted the data; GA analysed and interpreted the data; HNDL analysed and interpreted the data; SM, CNM, CP, DAC designed the study, interpreted the data. All authors read, edited and approved the final manuscript. 14. Ball K, McNaughton SA, Mhurchu CN, Andrianopoulos N, Inglis V, McNeilly B, et al. Supermarket Healthy Eating for Life (SHELf): protocol of a randomised controlled trial promoting healthy food and beverage consumption through price reduction and skill-building strategies. BMC Public Health. 2011;11:715. References Health, behavioral, cognitive, and social correlates of breakfast skipping among women living in socioeconomically disadvantaged neighborhoods. J Nutr. 2013;143(11): 1774–84. 7. Smith KJ, McNaughton SA, Cleland VJ, Crawford D, Ball K. Health, behavioral, cognitive, and social correlates of breakfast skipping among women living in socioeconomically disadvantaged neighborhoods. J Nutr. 2013;143(11): 1774–84. 8. Ferranti EP, Narayan KM, Reilly CM, Foster J, McCullough M, Ziegler TR, et al. Dietary self-efficacy predicts AHEI diet quality in women with previous gestational diabetes. Diabetes Educ. 2014;40(5):688–99. 8. Ferranti EP, Narayan KM, Reilly CM, Foster J, McCullough M, Ziegler TR, et al. Dietary self-efficacy predicts AHEI diet quality in women with previous gestational diabetes. Diabetes Educ. 2014;40(5):688–99. 9. Bowen DJ, Beresford SA. Dietary interventions to prevent disease. Annu Rev Public Health. 2002;23:255–86. 9. Bowen DJ, Beresford SA. Dietary interventions to prevent disease. Annu Rev Public Health. 2002;23:255–86. 10. Beresford SA, Curry SJ, Kristal AR, Lazovich D, Feng Z, Wagner EH. A dietary intervention in primary care practice: the Eating Patterns Study. Am J Public Health. 1997;87(4):610–6. 10. Beresford SA, Curry SJ, Kristal AR, Lazovich D, Feng Z, Wagner EH. A dietary intervention in primary care practice: the Eating Patterns Study. Am J Public Health. 1997;87(4):610–6. Abbreviations f d FV: fruits and vegetables; RCT: randomized controlled trial; RE-AIM: reach, effectiveness, adoption, implementation, maintenance; SES: socioeconomic status; SHELf: Supermarket Healthy Eating for Life. 11. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Educ Q. 1988;15(4):351–77. 11. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Educ Q. 1988;15(4):351–77. health promotion programs. Health Educ Q. 1988;15(4):351–77. 12. Stokols D. Establishing and maintaining healthy environments. Toward a social ecology of health promotion. Am Psychol. 1992;47(1):6–22. 12. Stokols D. Establishing and maintaining healthy environments. Toward a social ecology of health promotion. Am Psychol. 1992;47(1):6–22. 12. Stokols D. Establishing and maintaining healthy environments. Toward a social ecology of health promotion. Am Psychol. 1992;47(1):6–22. Author details 1 1Centre for Physical Activity and Nutrition Research, Deakin University, 221 Burwood Highway, Burwood, VIC 3125, Australia. 2Deakin Health Economics, Deakin University, 221 Burwood Highway, Burwood, VIC 3125, Australia. 3National Institute for Health Innovation, School of Population Health, Tamaki Campus, University of Auckland, Auckland 1072, New Zealand. 4School of Public Health, Faculty of Sciences, Curtin University, GPO Box U1987, Perth, WA 6845, Australia. Received: 8 July 2015 Accepted: 16 February 2016 Received: 8 July 2015 Accepted: 16 February 2016 Acknowledgements Ni Mhurchu C, Eyles H, Schilling C, Yang Q, Kaye-Blake W, Genc M, et al. Food prices and consumer demand: differences across income levels and ethnic groups. PLoS ONE. 2013;8(10), e75934. 21. Ni Mhurchu C, Blakely T, Wall J, Rodgers A, Jiang Y, Wilton J. Strategies to promote healthier food purchases: a pilot supermarket intervention study. Public Health Nutr. 2007;10(6):608–15. 45. Powell LM, Chriqui JF, Khan T, Wada R, Chaloupka FJ. Assessing the potential effectiveness of food and beverage taxes and subsidies for improving public health: a systematic review of prices, demand and body weight outcomes. Obes Rev. 2013;14(2):110–28. 22. Phipps EJ, Braitman LE, Stites SD, Singletary SB, Wallace SL, Hunt L, et al. Impact of a rewards-based incentive program on promoting fruit and vegetable purchases. Am J Public Health. 2015;105(1):166–72. 46. Epstein LH, Dearing KK, Roba LG, Finkelstein E. The influence of taxes and subsidies on energy purchased in an experimental purchasing study. Psychol Sci. 2010;21(3):406–14. 23. Blakely T, Ni Mhurchu C, Jiang Y, Matoe L, Funaki-Tahifote M, Eyles HC, et al. Do effects of price discounts and nutrition education on food purchases vary by ethnicity, income and education? Results from a randomised, controlled trial. J Epidemiol Community Health. 2011;65(10):902–8. 47. Waterlander WE, Steenhuis IH, de Boer MR, Schuit AJ, Seidell JC. Introducing taxes, subsidies or both: the effects of various food pricing strategies in a web-based supermarket randomized trial. Prev Med. 2012;54(5):323–30. 24. Australian Bureau of Statistics. 2006. Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia - Data only, 2006. http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/2033.0.55. 0012006?OpenDocument. Accessed April 11, 2015. 48. Giesen JC, Havermans RC, Nederkoorn C, Jansen A. Impulsivity in the supermarket. Responses to calorie taxes and subsidies in healthy weight undergraduates. Appetite. 2012;58(1):6–10. 25. Smith LP, Ng SW, Popkin BM. Trends in US home food preparation and consumption: analysis of national nutrition surveys and time use studies from 1965–1966 to 2007–2008. Nutr J. 2013;12:45. 49. Jeffery RW, Drewnowski A, Epstein LH, Stunkard AJ, Wilson GT, Wing RR, et al. Long-term maintenance of weight loss: current status. Health Psychol. 2000;19(1 Suppl):5–16. 26. Australian Bureau of Statistics. 2009. Australian social trends. http://www.abs. gov.au/socialtrends. Accessed 50. Estabrooks PA, Allen KC. Updating, employing, and adapting: a commentary on What does it mean to “employ” the RE-AIM model. Eval Health Prof. 2013;36(1):67–72. 27. King DK, Glasgow RE, Leeman-Castillo B. Acknowledgements 15. Ni Mhurchu C, Blakely T, Jiang Y, Eyles HC, Rodgers A. Effects of price discounts and tailored nutrition education on supermarket purchases: a randomized controlled trial. Am J Clin Nutr. 2010;91(3):736–47. This study was funded by a National Health & Medical Research Council (NHMRC) Project Grant, ID 594767. Coles® supermarkets and the National Heart Foundation of Australia provided in-kind support for the study. DLO is supported by a Canadian Institutes of Health Research Fellowship and an Endeavour Research Fellowship. KB is supported by a National Health & Medical Research Council Principal Research Fellowship, ID 1042442. SM is supported by an Australian Research Council Future Fellowship, ID 16. Waterlander WE, de Boer MR, Schuit AJ, Seidell JC, Steenhuis IH. Price discounts significantly enhance fruit and vegetable purchases when combined with nutrition education: a randomized controlled supermarket trial. Am J Clin Nutr. 2013;97(4):886–95. Page 15 of 15 Page 15 of 15 Page 15 of 15 Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 17. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health. 1999;89(9):1322–7. 40. Pollard CM, Pulker CE, Meng X, Kerr DA, Scott JA. Who uses the internet as a source of nutrition and dietary information? An Australian population perspective. J Med Internet Res. 2015;17(8), e209. 41. Korda H, Itani Z. Harnessing social media for health promotion and behavior change. Health Promot Pract. 2013;14(1):15–23. 18. Saunders RP, Evans MH, Joshi P. Developing a process-evaluation plan for assessing health promotion program implementation: a how-to guide. Health Promot Pract. 2005;6(2):134–47. 42. Herman DR, Harrison GG, Jenks E. Choices made by low-income women provided with an economic supplement for fresh fruit and vegetable purchase. J Am Diet Assoc. 2006;106(5):740–4. 19. Escaron AL, Meinen AM, Nitzke SA, Martinez-Donate AP. Supermarket and grocery store-based interventions to promote healthful food choices and eating practices: a systematic review. Prev Chronic Dis. 2013;10, E50. 43. Geliebter A, Ang IY, Bernales-Korins M, Hernandez D, Ochner CN, Ungredda T, et al. Supermarket discounts of low-energy density foods: effects on purchasing, food intake, and body weight. Obesity (Silver Spring). 2013; 21(12):E542–8. 20. Mhurchu CN, Blakely T, Funaki-Tahifote M, McKerchar C, Wilton J, Chua S, et al. Inclusion of indigenous and ethnic minority populations in intervention trials: challenges and strategies in a New Zealand supermarket study. J Epidemiol Community Health. 2009;63(10):850–5. 44. Acknowledgements Reaiming RE-AIM: using the model to plan, implement, and evaluate the effects of environmental change approaches to enhancing population health. Am J Public Health. 2010; 100(11):2076–84. 28. Jilcott S, Ammerman A, Sommers J, Glasgow RE. Applying the RE-AIM framework to assess the public health impact of policy change. Ann Behav Med. 2007;34(2):105–14. 29. Gaglio B, Shoup JA, Glasgow RE. The RE-AIM framework: a systematic review of use over time. Am J Public Health. 2013;103(6):e38–46. 30. Kessler RS, Purcell EP, Glasgow RE, Klesges LM, Benkeser RM, Peek CJ. What does it mean to “employ” the RE-AIM model? Eval Health Prof. 2013;36(1): 44–66. 31. Australian Bureau of Statistics. 2011. Census of population and housing. http://www.abs.gov.au/websitedbs/censushome.nsf/home/data. Accessed January 20, 2015. 32. Durlak JA, DuPre EP. Implementation matters: a review of research on the influence of implementation on program outcomes and the factors affecting implementation. Am J Community Psychol. 2008;41(3–4):327–50. 33. Patton MQ. Qualitative Research & Evaluation Methods. Thousand Oaks, CA: Sage Publications, Inc; 2002. 34. Hawkes C. Dietary implications of supermarket development: a global perspective. Develop Policy Rev. 2008;26(6):657–92. 35. Irala-Estevez JD, Groth M, Johansson L, Oltersdorf U, Prattala R, Martinez- Gonzalez MA. A systematic review of socio-economic differences in food habits in Europe: consumption of fruit and vegetables. Eur J Clin Nutr. 2000; 54(9):706–14. Olstad et al. International Journal of Behavioral Nutrition and Physical Activity (2016) 13:27 Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step: 36. Johansson L, Thelle DS, Solvoll K, Bjorneboe GE, Drevon CA. Healthy dietary habits in relation to social determinants and lifestyle factors. Br J Nutr. 1999; 81(3):211–20. • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit and we will help you at every step: • We accept pre-submission inquiries 37. Subar AF, Heimendinger J, Patterson BH, Krebs-Smith SM, Pivonka E, Kessler R. Fruit and vegetable intake in the united states: the baseline survey of the five a Day for better health program. Am J Health Promot. 1995;9(5):352–60. 38. Thornton LE, Bentley RJ, Kavanagh AM. Individual and area-level socioeconomic associations with fast food purchasing. J Epidemiol Community Health. 2011;65(10):873–80. y 39. Rehm CD, Matte TD, Van Wye G, Young C, Frieden TR. Demographic and behavioral factors associated with daily sugar-sweetened soda consumption in New York City adults. J Urban Health. 2008;85(3):375–85. 39. Rehm CD, Matte TD, Van Wye G, Young C, Frieden TR. Demographic and behavioral factors associated with daily sugar-sweetened soda consumption in New York City adults. J Urban Health. 2008;85(3):375–85.
https://openalex.org/W2340155273	http://liu.diva-portal.org/smash/get/diva2:904632/FULLTEXT01	English	null	Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model	Scientific reports	2,016	cc-by	11,570	Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model Els Willems, Carlos Guerrero-Bosagna, Eddy Decuypere, Steven Janssens, Johan Buyse, Nadine Buys, Per Jensen and Nadia Everaert Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model received: 13 August 2015 accepted: 07 January 2016 Published: 10 February 2016 Els Willems1,2, Carlos Guerrero-Bosagna2, Eddy Decuypere1, Steven Janssens3, Johan Buyse Nadine Buys3, Per Jensen2 & Nadia Everaert1,4 Previously, long-term effects on body weight and reproductive performance have been demonstrated in the chicken model of prenatal protein undernutrition by albumen removal. Introduction of such persistent alterations in phenotype suggests stable changes in gene expression. Therefore, a genome- wide screening of the hepatic transcriptome by RNA-Seq was performed in adult hens. The albumen- deprived hens were created by partial removal of the albumen from eggs and replacement with saline early during embryonic development. Results were compared to sham-manipulated hens and non-manipulated hens. Grouping of the differentially expressed (DE) genes according to biological functions revealed the involvement of processes such as ‘embryonic and organismal development’ and ‘reproductive system development and function’. Molecular pathways that were altered were ‘amino acid metabolism’, ‘carbohydrate metabolism’ and ‘protein synthesis’. Three key central genes interacting with many DE genes were identified: UBC, NR3C1, and ELAVL1. The DNA methylation of 9 DE genes and 3 key central genes was examined by MeDIP-qPCR. The DNA methylation of a fragment (UBC_3) of the UBC gene was increased in the albumen-deprived hens compared to the non- manipulated hens. In conclusion, these results demonstrated that prenatal protein undernutrition by albumen removal leads to long-term alterations of the hepatic transcriptome in the chicken. In utero growth retardation in humans as inferred from low birth weight has repercussions on postnatal health and performance, as exemplified by the increased risk of adult degenerative diseases such as type 2 diabetes1,2. The maternal low protein rat model is one of the most extensively studied animal models of in utero growth restriction3 and findings similar to those in humans are observed. The chicken can be used as a unique avian model to study prenatal protein undernutrition4–7 by replacing a part of the albumen with saline. As albumen is the main source of protein for the developing embryo8, the net effect is prenatal protein undernutrition. Thus, in the chicken only strictly nutritional effects are involved, in contrast to mammalian models where maternal effects (e.g. hormonal effects) are implicated. Linköping University Post Print N.B.: When citing this work, cite the original article. N.B.: When citing this work, cite the original article. Original Publication: Els Willems, Carlos Guerrero-Bosagna, Eddy Decuypere, Steven Janssens, Johan Buyse, Nadine Buys, Per Jensen and Nadia Everaert, Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model, 2016, Scientific Reports, (6). http://dx.doi.org/10.1038/srep20837 Copyright: Nature Publishing Group: Open Access Journals - Option C / Nature Publishing Group http://www.nature.com/ Postprint available at: Linköping University Electronic Press http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-125286 Original Publication: Els Willems, Carlos Guerrero-Bosagna, Eddy Decuypere, Steven Janssens, Johan Buyse, Nadine Buys, Per Jensen and Nadia Everaert, Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model, 2016, Scientific Reports, (6). http://dx.doi.org/10.1038/srep20837 p g p Copyright: Nature Publishing Group: Open Access Journals - Option C / Nature Publishing Group Postprint available at: Linköping University Electronic Press http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-1252 Postprint available at: Linköping University Electronic Press http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-125286 www.nature.com/scientificreports www.nature.com/scientificreports received: 13 August 2015 accepted: 07 January 2016 Published: 10 February 2016 Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model Indeed, in mammalian models manipulation of the maternal diet influences both the maternal nutritional and hormonal status, thereby exerting both nutritional and hormonal effects on the offspring.f ff p g Previously, the effects of albumen removal in the chicken have been studied by other groups4,9 and our own group5–7,10,11. Although the one-day-old chick weight was not significantly decreased by albumen removal, the body weight was reduced during the juvenile phase5,7,11. At adult age however, the effect of the body weight depended on the posthatch environmental conditions. When kept in battery cages with limited possibility for 1KU Leuven, Department of Biosystems, Laboratory of Livestock Physiology, Kasteelpark Arenberg 30 box 2456, 3001 Leuven, Belgium. 2Linköping University, IFM Biology, AVIAN Behavioural Genomics and Physiology Group, Linköping 581 83, Sweden. 3KU Leuven, Department of Biosystems, Research Group Livestock Genetics, Kasteelpark Arenberg 30 box 2456, 3001 Leuven, Belgium. 4University of Liège, Gembloux Agro-Bio Tech, Precision Livestock and Nutrition Unit, Passage des Déportés 2, 5030 Gembloux, Belgium. Correspondence and requests for materials should be addressed to J.B. (email: Johan.buyse@biw.kuleuven.be) Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 1 www.nature.com/scientificreports/ Figure 1. Correlation between biological replicates. Heat-map of Spearman’s correlation of the normalized counts as expression levels from all samples compared against each other, represented by a colored field ranging from green (0.95) to red (1). Figure 1. Correlation between biological replicates. Heat-map of Spearman’s correlation of the normalized counts as expression levels from all samples compared against each other, represented by a colored field ranging from green (0.95) to red (1). exercise, catch-up growth was observed in the albumen-deprived hens11. However when kept in floor pens–a more competitive environment for feed, space and water–the body weight of the albumen-deprived hens remained lower throughout the entire experimental period (55 weeks of age)7. Irrespective of the body weight in adulthood, the reproductive performance was markedly diminished by embryonic albumen removal as reflected in the reduced number and weight of the eggs7,11. At 10 weeks of age, glucose intolerance was observed in the albumen-deprived hens. This difference however, disappeared in adulthood due to age-related loss in glucose tolerance of the hens7. exercise, catch-up growth was observed in the albumen-deprived hens11. However when kept in floor pens–a more competitive environment for feed, space and water–the body weight of the albumen-deprived hens remained lower throughout the entire experimental period (55 weeks of age)7. Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model Irrespective of the body weight in adulthood, the reproductive performance was markedly diminished by embryonic albumen removal as reflected in the reduced number and weight of the eggs7,11. At 10 weeks of age, glucose intolerance was observed in the albumen-deprived hens. This difference however, disappeared in adulthood due to age-related loss in glucose tolerance of the hens7. Induction of an altered phenotype that persists throughout the lifespan implies stable changes in gene expres- sion which would result in altered activities of metabolic pathways12. In the low-protein-diet rat model, changes in both hepatic gene expression and DNA methylation have been reported. Six days after weaning, the peroxisome proliferator-activated receptor α (PPARα ) expression is 10.5-fold higher and DNA methylation 20.6% lower, whereas expression of the glucocorticoid receptor (GR) is 200% higher and DNA methylation 22.8% lower13. Moreover, these changes persist at least until postnatal day 3414. Gong et al. reported an increase in gene expres- sion in the rat of the Insulin-like growth factor 2 (IGF2) in the liver and an increase in DNA methylation of the regulatory region of IGF2 in the liver of male low-protein diet offspring at day 015.hf g y gf g y Therefore, the objective of the present study was to perform a genome-wide screening for differences in gene expression using RNA-Sequencing (RNA-Seq) in liver samples collected from adult laying hens and differentially expressed genes were grouped according to biological function to discover affected pathways. In addition, it was investigated whether the alterations in gene expression coincided with changes in DNA methylation, in order to examine the possibility of epigenetic mechanisms underlying the observed long-term programming effects. Results h i l To proceed with confirmation and validation, an addi- tional 28 previously identified genes where the albumen-deprived hens differed from the sham-manipulated group were included in a list to select genes to validate the RNA-Seq, making a total of 31 genes (Table 1). Confirmation and validation of DE genes of RNA-Seq via qPCR. Half of these 31 genes were selected (15 genes) for technical confirmation covering a range of expression levels and biological functions (Fig. 3). Relative expression levels of the albumen-deprived hens versus the non-manipulated and the sham-manipulated hens (n = 3 per group) are displayed as obtained from the RNA-Seq and qPCR results. The fold change estimates by qPCR and by RNA-Seq were strongly correlated (Pearson correlation coefficient was 0.85). The genes of which the expression levels did not fully match are the genes with low expression levels, pointing to decreased sensitivity of the RNA-Seq technique at low expression levels. q q p To validate the biological significance of the 15 DE genes, sample size was increased to 8 samples per group (Table 2). As observed from both the RNA-Seq and the qPCR results, 7 genes (46.7%) (TNFSF10, LAPTM4B, TMEM86A, CKS1B, NXPH-2, LRRC3C and BMF) had differentially expression in the liver of the albumen-deprived hens compared to the sham-manipulated hens (P < 0.1) and were thus validated. Eight genes could not be validated. Two genes of these (SEMA6D and H2B-I) had significantly increased expression in the albumen-deprived hens compared to the sham-manipulated hens in the RNA-Seq, but were significantly decreased in the qPCR results. Another 6 genes did not display significant differences in the gene expression measured by qPCR. Grouping of DE genes according to biological function. The cut-off criteria were loosened (P < 0.005 and Fold change > 1.5) to include more DE genes in our analysis to find key metabolic and biological path- ways affected by prenatal protein undernutrition by systems biology analysis using Ingenuity Pathways analysis (IPA) software. Only previously identified genes were included, and all DE genes for which the non-manipulated group differed from the sham-manipulated group were excluded as these do not represent effects of prena- tal protein undernutrition. 116 DE genes were obtained, for 13 genes albumen-deprived differed from both non-manipulated and sham-manipulated; for 31 genes albumen-deprived differed from non-manipulated and for 88 genes albumen-deprived differed from sham-manipulated group. Significantly involved biological pathways are listed in Table 3. Results h i l Physiological results. Detailed physiological results have been published previously7. In brief, body weight of the albumen-deprived hens was reduced throughout the entire experimental period (0–55 weeks). In addition, the abdominal fat weight was also reduced in the albumen-deprived hens as compared to the sham-manipulated hens. No differences in absolute or relative liver weight were observed. The reproductive capacity was diminished in the albumen-deprived hens as reflected in the reduced number of eggs and lower egg weight. The plasma tri- iodothyronine (T3) levels were increased in the albumen-deprived compared to the non-manipulated hens, but not the sham-manipulated hens. An oral glucose tolerance test (OGTT) at 10 weeks of age revealed a decreased glucose tolerance in the albumen-deprived hens. During adulthood, an age-related loss of glucose tolerance was observed in the hens, leading to disappearance of treatment differences in the OGTT. Genome-wide screening for differentially expressed (DE) genes using RNA-Seq. The logarithms of the fold change and associated P-values of treatment differences are depicted in Supplementary Table S3 online. The heat map of Spearman correlation between all samples using the normalized counts as expression values is shown in Fig. 1. The correlation between all samples was very high (> 0.95) and the biological replicates did not cluster well together. When using P < 0.001 and the log2-fold change higher than 1 as cut-off, 156 genes were dif- ferentially expressed between the treatments, only 75 of these were previously identified genes (Fig. 2). A heatmap Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 2 www.nature.com/scientificreports/ Figure 2. Venn-diagram showing 75 significantly differentially expressed (DE) genes. Genes are DE between the non-manipulated, sham-manipulated and albumen-deprived hens, including only previously identified genes. DE genes were filtered with a cut-off of P-value < 0.001 and log2-fold change > 1. Figure 2. Venn-diagram showing 75 significantly differentially expressed (DE) genes. Genes are DE between the non-manipulated, sham-manipulated and albumen-deprived hens, including only previously identified genes. DE genes were filtered with a cut-off of P-value < 0.001 and log2-fold change > 1. enerated based on these differentially expressed genes is depicted in Supplementary Figure 1 and demonstrates good separation of the three treatments (except for sample sham 3).if Only 3 previously identified genes were differentially expressed between the albumen-deprived hens and both the non-manipulated and the sham-manipulated hens. Results h i l Two of these networks had key central genes, interacting with many DE genes. The first was a pathway involved in embryonic development, organ development and organ morphology, including 14 DE genes and had ubiquitin C (UBC) as central gene, whereas the second one was involved in cell cycle and carbohydrate metabolism, including 12 DE genes and had glucocorticoid receptor (NR3C1) and Embryonic lethal, abnormal Vision, Drosophila-like 1 (ELAVL1) as central genes (Fig. 4). None of the three central genes (UBC, NR3C1 and ELAVL1) were differentially expressed in the RNA-Seq dataset. DNA methylation analysis using MeDIP-qPCR. The 9 DE genes of the qPCR (TNFSF10, LAPTM4B, TMEM86A, CKS1B, NXPH-2, LRRC3C, BMF, SEMA6D and H2B-I) were selected for DNA methylation anal- ysis. However, no specific primers for any of the CpG rich fragments of TNFSF10 could be optimised and this gene was therefore excluded from the analysis. In addition, the DNA methylation of the 3 key central regu- latory genes (UBC, NR3C1, ELAVL1) identified by the pathway analysis was also examined. For each gene, several CpG rich fragments were examined in the promoter region or around the transcription start site via MeDIP-qPCR (Table 4). The DNA methylation of most of the examined fragments did not display a treatment effect. Only the DNA methylation of fragment (UBC_3) of the UBC gene was affected by treatment (P = 0.0442). Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 3 www.nature.com/scientificreports/ Gene Description Accession log2-fold change P value C vs. A S vs. A C vs. S C vs. A S vs. A C vs. Results h i l S C≠A; S≠A GRIN2C glutamate receptor ENSGALG00000027415 3.98 3.53 0.45 0.00018 0.00065 NS IDO2 indoleamine 2,3-dioxygenase2 ENSGALG00000024085 3.91 3.22 0.69 0.00001 0.00020 NS SPAG-4 like sperm-associated antigen 4-like ENSGALG00000000443 − 3.03 − 4.28 1.25 0.00021 0.00000 NS S≠A H2B-I histone H2B 1/2/3/4/6 ENSGALG00000027174 0.35 1.04 − 0.69 NS 0.00059 NS uncharacterized protein ENSGALG00000008635 1.01 2.51 − 1.49 NS 0.00001 NS H2A-VII histone H2A-IV ENSGALG00000027113 0.66 1.06 − 0.41 NS 0.00034 NS SEMA6D semaphorin 6D ENSGALG00000004844 0.69 1.07 − 0.38 NS 0.00012 NS HERPUD1 homocysteine-inducible, endoplasmic reticulum stress-inducible, ubiquitin-like domain member 1 ENSGALG00000001220 − 0.34 − 1.21 0.87 NS 0.00003 NS IP6K2 inositol hexakisphosphate kinase 2 ENSGALG00000005701 0.41 1.07 − 0.66 NS 0.00092 NS SLC6A6 sodium- and chloride-dependent taurine transporter ENSGALG00000006425 0.76 1.16 − 0.40 NS 0.00060 NS PROK2 prokineticin 2 ENSGALG00000007785 − 0.91 − 2.93 2.02 NS 0.00083 NS LECT2 myeloid protein 1 precursor ENSGALG00000006323 − 1.17 − 2.03 0.87 NS 0.00038 NS TMEM116 transmembrane protein 116 ENSGALG00000004760 0.40 1.03 − 0.63 NS 0.00052 NS LAPTM4B lysosomal protein transmembrane 4 beta ENSGALG00000028628 − 0.90 − 1.92 1.02 NS 0.00071 NS GATA5 transcription factor GATA-5 ENSGALG00000005352 2.55 3.42 − 0.87 NS 0.00012 NS ENSGALG00000000850 − 0.89 − 1.83 0.94 NS 0.00066 NS CKS1B CDC28 protein kinase regulatory subunit 1B ENSGALG00000028664 − 1.04 − 1.65 0.61 NS 0.00047 NS LRRC3C leucine rich repeat containing 3C ENSGALG00000026789 − 0.83 − 2.39 1.56 NS 0.00073 NS uncharacterized protein ENSGALG00000010993 0.67 1.46 − 0.79 NS 0.00007 NS GAL7 gallinacin-7 ENSGALG00000022817 − 0.92 − 1.87 0.95 NS 0.00071 NS GAL2 gallinacin-2 ENSGALG00000016669 − 1.36 − 2.47 1.10 NS 0.00004 NS GJA1 gap junction alpha-1 protein ENSGALG00000014873 − 0.27 − 1.15 0.87 NS 0.00025 NS SLITRK4 SLIT and NTRK-like family, member 4 ENSGALG00000007242 1.31 4.45 − 3.14 NS 0.00009 NS TMEM86A transmembrane protein 86A ENSGALG00000006358 0.91 1.29 − 0.38 NS 0.00078 NS uncharacterized protein ENSGALG00000009387 − 1.08 − 2.26 1.18 NS 0.00038 NS BMF bcl-2-modifying factor ENSGALG00000014537 0.46 1.04 − 0.58 NS 0.00002 NS TC2N tandem C2 domains, nuclear ENSGALG00000010738 1.04 1.37 − 0.32 NS 0.00043 NS NXPH2 neurexophilin 2 ENSGALG00000029083 0.55 1.54 − 0.99 NS 0.00009 NS GLUL glutamine synthetase ENSGALG00000003678 − 0.08 − 1.04 0.96 NS 0.00061 NS TNFSF10 tumor necrosis factor ligand superfamily member 10 ENSGALG00000009179 0.44 1.23 − 0.80 NS 0.00064 NS HMCN1 hemicentin 1 ENSGALG00000005141 − 0.57 − 1.49 0.92 NS 0.00017 NS Table 1. List of differentially expressed genes from RNA-Seq. Results h i l Cut-off criteria were P < 0.001 and log2-fold change > 1 between the albumen-deprived hens and the non-manipulated and sham-manipulated hens (3 genes) or between the albumen-deprived hens and the sham-manipulated hens (28 genes). Genes in italic were selected for measurements of gene expression via qPCR. The albumen-deprived hens had significantly more DNA methylation in this fragment than the non-manipulated group, whereas the sham-manipulated group had an intermediate value (Fig. 5). The albumen-deprived hens had significantly more DNA methylation in this fragment than the non-manipulated roup, whereas the sham-manipulated group had an intermediate value (Fig. 5). Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 Discussionh The aim of the present study was to identify differences in gene expression using RNA-Seq and find pathways long-term affected by prenatal protein undernutrition by albumen removal in the liver of the chicken. In addi- tion, it was investigated whether alterations in gene expression coincide with changes in DNA methylation (MeDIP-qPCR).i q A first striking observation of the RNA-Seq dataset was the very high correlation between normalized counts as expression levels between all samples (Spearman correlation > 0.95), as demonstrated by the heat map in Fig. 1. This shows that both within as well as between treatments’ differences in expression were very small and all sam- ples were very similar. Indeed, posthatch environmental conditions were the same for all groups and therefore differences should only be due to the nutritional programming performed during embryonic development. A possible explanation for the small differences is that the treatments were applied early during embryonic devel- opment and measurements were performed in adulthood. The time span between treatment and sampling was nearly 58 weeks. Hence, treatment differences might have been much larger if measurements would have been performed earlier in life. However, at hatch, 2-D DIGE results revealed only 8 differential protein spots between the albumen-deprived hens and either the non-manipulated or the sham-manipulated hens or both5. Perhaps, other organs than the liver such as the hypothalamus are the sites of major changes in gene expression causing the Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 4 www.nature.com/scientificreports/ Figure 3. Correlation between RNA-Seq and qPCR results for 15 DE genes. The log2-fold change between the albumen-deprived hens and the non-manipulated (black box) and sham-manipulated hens (grey box) is displayed for both the RNA-Seq results as obtained from edgeR and the qPCR results obtained from the 2−ddCt method. The Pearson’s correlation coefficient between relative expression levels is displayed. Figure 3. Correlation between RNA-Seq and qPCR results for 15 DE genes. The log2-fold change between the albumen-deprived hens and the non-manipulated (black box) and sham-manipulated hens (grey box) is displayed for both the RNA-Seq results as obtained from edgeR and the qPCR results obtained from the 2−ddCt method. The Pearson’s correlation coefficient between relative expression levels is displayed. Figure 3. Correlation between RNA-Seq and qPCR results for 15 DE genes. Discussionh Using P< 0 001 and log fold change >1 as cut off to determine DE genes in RNA Seq analysis 3 genes were phenotypic differences. Indeed, the hypothalamus functions as the center of regulation of energy and feed intake and is therefore a good candidate for future research. U i P 0 001 d l f ld h 1 ff d i DE i RNA S l i 3 f and is therefore a good candidate for future research. Using P < 0.001 and log2-fold change > 1 as cut-off to determine DE genes in RNA-Seq analysis, 3 genes were DE between the albumen-deprived and both the non-manipulated and sham-manipulated group, 39 between the albumen-deprived and the sham-manipulated and 21 between the albumen-deprived and the non-manipulated group (Fig. 2). To our knowledge, no published studies have examined the effect of low protein diet in mam- malian models on the offspring via RNA-Seq. Previously, the effect of low protein diet during gestation on the offspring has been analyzed by microarray in the liver of mouse16 and rat17. In adult rat offspring (day 84), only a small number of genes was affected, 222 genes were upregulated, whereas 89 were downregulated (False discovery rate < 0.05; fold change ≥ 1.5)17, although this number was much larger than in the present study. This could point at a distinction between application of strictly nutritional effects and involvement of secondary maternal effects. In addition, in these mammalian models the reduction in protein content in the diet is frequently accompanied by an increase in carbohydrates (e.g. glucose, sucrose or starch) as the diets need to be isocaloric. Therefore, observed effects might just as well be effects of carbohydrate overload instead of protein restriction18. Technical arguments to explain the low number of DE genes include lack of annotation of certain genes in the chicken. Indeed, from the present dataset 156 genes were differentially expressed, but only 75 of these were annotated in the databank. Although many of these ‘novel genes’ are probably mapping artefacts, some of these might repre- sent true new genes. RNA-Seq datasets will become very important in the near future to improve the annotation of the chicken genome and identify more genes. Discussionh The log2-fold change between the albumen-deprived hens and the non-manipulated (black box) and sham-manipulated hens (grey box) is displayed for both the RNA-Seq results as obtained from edgeR and the qPCR results obtained from the 2−ddCt method. The Pearson’s correlation coefficient between relative expression levels is displayed. Gene qPCR (2−ddCt) P-value Validated? non-manipulated sham-manipulated albumen-deprived C≠A; S≠A SPAG4like 1.00 ± 0.27 0.56 ± 0.23 0.75 ± 0.17 NS No S≠A TNFSF10 1.00 ± 0.13b 0.69 ± 0.09b 1.40 ± 0.07a 0.0006 Yes LAPTM4B 1.00 ± 0.16ab 1.96 ± 0.62b 0.50 ± 0.09a 0.0520 Yes TMEM86A 1.00 ± 0.07ab 0.93 ± 0.15b 1.42 ± 0.16a 0.0489 Yes CKS1B 1.00 ± 0.12ab 1.74 ± 0.39a 0.96 ± 0.08b 0.0777 Yes NXPH2 1.00 ± 0.08a 0.45 ± 0.07b 1.06 ± 0.20a 0.0091 Yes LRRC3C 1.00 ± 0.14ab 1.68 ± 0.45a 0.70 ± 0.10b 0.0776 Yes BMF 1.00 ± 0.10ab 0.74 ± 0.08b 1.32 ± 0.12a 0.0048 Yes SEMA6D 1.00 ± 0.17ab 1.70 ± 0.50a 0.63 ± 0.08b 0.0916 No H2B-I 1.00 ± 0.16ab 2.12 ± 0.68a 0.67 ± 0.08b 0.0681 No GLUL 1.00 ± 0.14 1.20 ± 0.23 1.13 ± 0.12 NS No HERPUD1 1.00 ± 0.10 1.38 ± 0.36 1.17 ± 0.29 NS No SLC6A6 1.00 ± 0.11 1.18 ± 0.21 1.03 ± 0.06 NS No LECT2 1.00 ± 0.29 1.33 ± 0.33 0.65 ± 0.16 NS No GJA1 1.00 ± 0.11 0.94 ± 0.11 0.97 ± 0.12 NS No Table 2. Biological validation of RNA-Seq results. Gene expression of selected DE genes (P < 0.001 and log2-fold change > 1) were measured in 8 samples per treatment via qPCR. a,btreatment means with different superscript are significantly different (P < 0.1). Table 2. Biological validation of RNA-Seq results. Gene expression of selected DE genes (P < 0.001 and log2-fold change > 1) were measured in 8 samples per treatment via qPCR. a,btreatment means with different superscript are significantly different (P < 0.1). Table 2. Biological validation of RNA-Seq results. Gene expression of selected DE genes (P < 0.001 and log2-fold change > 1) were measured in 8 samples per treatment via qPCR. a,btreatment means with different superscript are significantly different (P < 0.1). phenotypic differences. Indeed, the hypothalamus functions as the center of regulation of energy and feed intake and is therefore a good candidate for future research. www.nature.com/scientificreports/ Abbreviations: ABCC9 (ABC transporter C family member 9); ABP1 (Actin binding protein 1); ADAMTS1 and 5 (A disintegrin and metalloproteinase with thrombospondin motifs 1 and 5); ARHGEF28 (Rho guanine nucleotide exchange factor 28); ASNS (Asparagine synthetase); BAIAP2 (Brain-specific angiogenesis inhibitor 1-associated protein 2); BMF (Bcl-2-modifying factor); CD200 (OX-2 membrane glycoprotein); CKS1B (CDC28 protein kinase regulatory subunit 1B): DCT (L-dopachrome tautomerase precursor); DIO2 (Iodothyronine deiodinase); ERRFI1 (ERBB receptor feedback inhibitor 1); FOXP2 (Forkhead box protein P2); FRZB (secreted frizzled-related protein 3 precursor); GATA5 (transcription factor GATA-5); GFPT2 (glutamine-fructose-6-phosphate transaminase 2); GRIN2C (glutamate receptor); HPS5 (Hermansky-Pudlak syndrome 5); IDO2 (indoleamine 2,3-dioxygenase 2); IL12RB1 (interleukin 12 receptor, beta 1); IP6K2 (inositol hexakisphosphate kinase 2); LAMA2 (laminin, alpha 2); LECT2 (myeloid protein 1 precursor); LRAT (Lecithin retinol acyltransferase); MTTP (microsomal triglyceride transfer protein large subunit precursor); NR0B1 (nuclear receptor subfamily 0 group B member 1); PDE11A (phosphodiesterase 11A); PITX2 (pituitary homeobox 2); PLCL1 (phospholipase C-like 1); PRF1 (Perforin-1); PROK2 (prokineticin 2); RACGAP1 (Rac GTPase activating protein 1); SIK1 (serine/threonine-protein kinase); SLC20A2 (sodium-dependent phosphate transporter 2); SLC39A14 (solute carrier family 39 (zinc transporter), member 14); SLC3A1 (Neutral and basic amino acid transport protein); SLC6A6 (sodium- and chloride- dependent taurine transporter); SLC7A10 (Asc-type amino acid transporter 1); SLITRK4 (SLIT and NTRK-like protein 4); TECTA (tectorin alpha); TNFSF10 (tumor necrosis factor ligand superfamily member 10); TP53I3 (tumor protein p53 inducible protein 3); ULK1 (Serine/threonine-protein kinase); WNT11 (Protein Wnt-11). Expression of 15 DE genes of the RNA-Seq dataset was validated using qPCR. Technically (n = 3), a strong correlation between results from both techniques could be seen. The genes of which the expression levels did not fully match are the genes with low expression levels, pointing to decreased sensitivity of the RNA-Seq technique at low expression levels. Biologically (n = 8), however, only half of the DE genes showed the same results using qPCR. This is in agreement however with the level of consistency observed in other studies19. The differences between both techniques may be ascribed to several factors such as interindividual variation as liver samples were collected from different chickens. Many publications of genome-wide expression studies lack validation, which may provide misleading conclusions. y p g DE genes (P < 0.005 and fold change > 1.5) were grouped to find key metabolic and biological pathways affected by prenatal protein undernutrition using systems biology analysis (Ingenuity Pathways analysis (IPA) software). www.nature.com/scientificreports/ www.nature.com/scientificreports/ Molecular and cellular functions Molecules P-value Number of molecules Amino acid metabolism SLC3A1, SLC7A10, ASNS, DIO2, GFPT2, IDO2, SLC6A6 2.91E-05–4.78E-02 7 Molecular transport SLC3A1, SLC7A10, TNFSF10, MTTP, LRAT, SLC6A6, DCT, IP6K2, RACGAP1, SLC20A2, SLC39A14, ABCC9, GRIN2C, SIK1, DIO2, CD200, NR0B1, ULK1, FOXP2, HPS5 2.91E-05–4.78E-02 20 Small molecule biochemistry SLC3A1, SLC7A10, ASNS, DIO2, GFPT2, IDO2, SLC6A6, TP53I3, ABP1, LRAT, MTTP, PDE11A, TNFSF10, DCT, FOXP2, HPS5 2.91E-05–4.78E-02 16 Cell death and survival DCT, PRF1, TNFSF10, PITX2, CD200, LAMA2, BMF, SIK1, LECT2, SLC6A6, PROK2, IP6K2, FRZB 4.11E-04–4.78E-02 13 Cell-to-cell signaling and interaction DCT, PRF1, TNFSF10, FRZB, BAIAP2, LAMA2, CD200, PITX2, HPS5,TECTA, FOXP2, IL12RB1 3.48E-03–4.78E-02 12 Carbohydrate metabolism TNFSF10, GFPT2 5.42E-03–4.26E-02 2 Protein synthesis MTTP, CD200, GRIN2C, NR0B1, ULK1, IL12RB1, PRF1 5.42E-03–2.28E-02 7 Physiological system development and function Nervous system development and function CD200, LAMA2, FOXP2, BAIAP2, PDE11A, ARHGEF28, SLITRK4, ULK1, FRZB, PRF1, TECTA, 4.30E-04–4.78E-02 11 Organ morphology LRAT, NR0B1, ADAMTS1, ERRFI1, GATA5, LAMA2, FRZB, PITX2, SLC39A14, SLC6A6, TECTA, FOXP2, WNT11, DCT, HPS5, ABCC9, PLCL1, CD200, PDE11A, DIO2 4.30E-04–4.86E-02 20 Reproductive system development and function LRAT, NR0B1, ADAMTS1, ERRFI1, GATA5, LAMA2, BMF, WNT11, PROK2, PRF1, TNFSF10, PITX2 4.30E-04–3.74E-02 12 Tissue development CD200, LAMA2, ADAMTS1, WNT11, NR0B1, FOXP2, BMF, DIO2, FRZB, PITX2, ERRFI1, SLC39A14, PROK2, HPS5, ARHGEF28, BAIAP2, SLITRK4, ULK1, LRAT, TNFSF10, ADAMTS5 4.30E-04–4.78E-02 21 Embryonic development WNT11, LAMA2, NR0B1, FOXP2, PITX2, ADAMTS1, FRZB, SLC39A14, BMF, DIO2, CKS1B, PROK2, ERRFI1, HPS5, LRAT 5.42E-03–4.78E-02 15 Organ development WNT11, NR0B1, FOXP2, DIO2, ADAMTS1, PITX2, FRZB, SLC39A14, BMF, PROK2, ERRFI1, HPS5, LECT2, MTTP, PDE11A, PRF1, TNFSF10, LAMA2, LRAT 5.42E-03–4.78E-02 19 Organismal development ADAMTS1, BMF, NR0B1, PITX, DIO2, FRZB, ERRFI1, FOXP2, SLC39A14, GATA5, HPS5, LAMA2, LRAT, WNT11, PROK2, TNFSF10 5.42E-03–4.78E-02 16 Table 3. Identification of relevant biological pathways affected by prenatal protein undernutrition by albumen removal in the chicken. Differential expressed genes were grouped through the use of Ingenuity Pathway Analysis (IPA). IPA-analysis (www.ingenuity.com) was used to identify key biological pathways comprising the differentially identified proteins after prenatal protein undernutrition by albumen removal in chicken. The significance of the canonical pathways was tested by Fisher’s exact test. The following genes are included in the biological pathways. Discussionh Using P < 0.001 and log2-fold change > 1 as cut-off to determine DE genes in RNA-Seq analysis, 3 genes were DE between the albumen-deprived and both the non-manipulated and sham-manipulated group, 39 between the albumen-deprived and the sham-manipulated and 21 between the albumen-deprived and the non-manipulated group (Fig. 2). To our knowledge, no published studies have examined the effect of low protein diet in mam- malian models on the offspring via RNA-Seq. Previously, the effect of low protein diet during gestation on the offspring has been analyzed by microarray in the liver of mouse16 and rat17. In adult rat offspring (day 84), only a small number of genes was affected, 222 genes were upregulated, whereas 89 were downregulated (False discovery rate < 0.05; fold change ≥ 1.5)17, although this number was much larger than in the present study. This could point at a distinction between application of strictly nutritional effects and involvement of secondary maternal effects. In addition, in these mammalian models the reduction in protein content in the diet is frequently accompanied by an increase in carbohydrates (e.g. glucose, sucrose or starch) as the diets need to be isocaloric. Therefore, observed effects might just as well be effects of carbohydrate overload instead of protein restriction18. Technical arguments to explain the low number of DE genes include lack of annotation of certain genes in the chicken. Indeed, from the present dataset 156 genes were differentially expressed, but only 75 of these were annotated in the databank. Although many of these ‘novel genes’ are probably mapping artefacts, some of these might repre- sent true new genes. RNA-Seq datasets will become very important in the near future to improve the annotation of the chicken genome and identify more genes. Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 5 www.nature.com/scientificreports/ Seven important physiological system development and functions were affected by the applied Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 6 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 4. Grouping of DE genes according to biological function. Differentially expressed genes (P < 0.005 and fold change > 1.5) are grouped with Ingenuity Pathway Analysis to identify affected pathways and key central regulatory genes. (A) Pathway involved in Embryonic development, organ development and organ morphology. (B) Pathway involved in Cell cycle and carbohydrate metabolism. Grey genes are identified as DE genes in RNA-Seq: 14 and 12 DE genes in the two pathways, respectively. Figure 4. Grouping of DE genes according to biological function. Differentially expressed genes (P < 0.005 and fold change > 1.5) are grouped with Ingenuity Pathway Analysis to identify affected pathways and key central regulatory genes. (A) Pathway involved in Embryonic development, organ development and organ morphology. (B) Pathway involved in Cell cycle and carbohydrate metabolism. Grey genes are identified as DE genes in RNA-Seq: 14 and 12 DE genes in the two pathways, respectively. treatment. These pathways include ‘embryonic and organismal development’, ‘organ morphology and develop- ment’, ‘tissue development’ and ‘reproductive system development and function’, as expected by applying a treat- ment early during embryonic development. These results are in agreement with our previously published data on both the peri- and postnatal development of the albumen-deprived hens5–7,11. Although no differences were detected in one-day-old chick weight, the proportional carcass weight and the water content of the carcass were increased in the albumen-deprived group5. In addition, on embryonic day 20, the plasma thyroxine (T4) concen- tration was reduced for the albumen-deprived group, indicating a decreased metabolic rate. Body weight and feed intake were reduced during the young to juvenile phase, whereas at adult age the body weight either decreased or increased depending on the posthatch environmental conditions7,11. These results are probably the consequence of differences in embryonic development. Still, it is remarkable that this effect was still apparent at adult age. In agreement, maternal protein restriction in the rat offspring affected the ‘developmental process’ as biological pathway process17. p y p Genes involved in the ‘reproductive system development and function’ pathway were also demonstrated to be affected. Irrespective of posthatch body weight, the reproductive performance of the albumen-deprived hens was seriously diminished as reflected in both a reduced number and weight of the eggs7,11. www.nature.com/scientificreports/ Moreover, these eggs had a different composition (increased proportional yolk and decreased proportional albumen) and were of infe- rior quality as more second grade eggs were laid by the albumen-deprived group11. The reduced reproductive performance is in agreement with the study of Rae et al. linking prenatal undernutrition of ewes to diminished ovulation rate of offspring in adult life20.f f Additionally, several molecular and cellular functions were affected: ‘amino acid metabolism’, ‘molecular transport’, ‘small molecule biochemistry’, ‘cell death and survival’, ‘cell-to-cell signaling and interaction’, ‘carbo- hydrate metabolism’ and ‘protein synthesis’. Lillycrop et al. also found altered molecular functions due to prena- tal protein undernutrition in rats: receptor binding, tetrapyrrole binding, and UDP-glycosyltransferase activity, cation and anion transmembrane transporter activity, growth factor activity and ATPase activity17. Genes with a wide range of functions were demonstrated to be altered in the present study, which is consistent with our previous studies demonstrating a wide range of both physiological and metabolic alterations. By partial albumen replacement with saline, the embryonic protein availability was decreased and therefore changes in the expression of genes related to the amino acid metabolism could be expected. Moreover, changes in amino acid metabolism were also observed by screening for differential protein abundances in the liver of newly hatched chicks. Affected pathways included valine, methionine, glutamate and cysteine degradation5, although the precise genes/proteins affected differed between these two studies. In this respect, changes in protein synthesis were also inferred. A previous study from our group also found indications of an altered protein metabolism in broilers treated by albu- men removal before incubation, suggesting a transient increase in muscle proteolysis10. Although these results may be interpreted as ordinary and expected, it provides evidence that the applied treatment creates strictly nutritional changes.f Another affected pathway was the ‘carbohydrate metabolism’, although few genes were represented here. However, changes in glucose metabolism have been demonstrated before by differential protein abundances in the liver of newly hatched chicks. Affected pathways included the gluconeogenesis, the tricarboxyl acid cycle and pyruvate fermentation to lactate5. Furthermore, the albumen-deprived hens exhibited reduced glucose toler- ance at 10 weeks of age possibly by a decreased insulin production or increased insulin resistance. However this difference disappeared at adult age due to age-related loss of glucose tolerance in the chicken7. www.nature.com/scientificreports/ Also, in mamma- lian maternal low protein models, differences in gene expression involved in the carbohydrate metabolism were Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 7 www.nature.com/scientificreports/ Gene Accession number Primer pair MeDIP-qPCR (normalized Ct values) non-manipulated sham-manipulated albumen-deprived P-value BMF ENSGALG00000014537 BMF_1 1.00 ± 0.02 1.03 ± 0.02 1.03 ± 0.01 NS CKS1B ENSGALG00000028664 CKS1B_1 1.00 ± 0.01 1.03 ± 0.02 1.00 ± 0.00 NS H2B-I ENSGALG00000027174 H2B_1 1.00 ± 0.04 1.00 ± 0.04 0.96 ± 0.05 NS H2B_2 1.00 ± 0.01 1.02 ± 0.00 1.01 ± 0.01 NS H2B_3 1.00 ± 0.02 1.08 ± 0.05 1.01 ± 0.02 NS LAPTM4B ENSGALG00000028628 LAPTM4B_1 1.00 ± 0.02 1.03 ± 0.02 1.00 ± 0.02 NS LAPTM4B_2 1.00 ± 0.03 1.06 ± 0.02 0.99 ± 0.02 NS LAPTM4B_3 1.00 ± 0.02 0.99 ± 0.02 1.00 ± 0.03 NS LRRC3C ENSGALG00000026789 LRRC3C_1 1.00 ± 0.02 0.99 ± 0.02 0.95 ± 0.03 NS LRRC3C_2 1.00 ± 0.02 1.05 ± 0.05 1.02 ± 0.04 NS LRRC3C_3 1.00 ± 0.01 1.02 ± 0.03 1.00 ± 0.02 NS NXPH2 ENSGALG00000029083 NXPH2_1 1.00 ± 0.04 0.99 ± 0.04 0.95 ± 0.03 NS NXPH2_2 1.00 ± 0.01 1.00 ± 0.02 1.00 ± 0.01 NS NXPH2_3 1.00 ± 0.04 0.96 ± 0.03 0.94 ± 0.03 NS SEMA6D ENSGALG00000004844 SEMA6D_1 1.00 ± 0.01 0.99 ± 0.01 0.98 ± 0.01 NS SEMA6D_2 1.00 ± 0.01 0.99 ± 0.01 0.99 ± 0.01 NS TMEM86A ENSGALG00000006358 TMEM86A_1 1.00 ± 0.02 0.97 ± 0.01 1.00 ± 0.02 NS ELAVL1 ENSGALG00000000726 ELAVL1_1 1.00 ± 0.01 1.00 ± 0.01 0.99 ± 0.01 NS ELAVL1_2 1.00 ± 0.01 1.00 ± 0.01 0.99 ± 0.01 NS ELAVL1_3 1.00 ± 0.01 0.99 ± 0.01 0.99 ± 0.01 NS NR3C1 ENSGALG00000007394 NR3C1_1 1.00 ± 0.03 1.00 ± 0.03 0.96 ± 0.02 NS NR3C1_2 1.00 ± 0.02 1.01 ± 0.03 0.98 ± 0.02 NS NR3C1_3 1.00 ± 0.01 1.00 ± 0.01 1.00 ± 0.01 NS NR3C1_4 1.00 ± 0.04 0.99 ± 0.05 0.95 ± 0.04 NS UBC ENSGALG00000004509 UBC_1 1.00 ± 0.03 0.97 ± 0.01 1.00 ± 0.02 NS UBC_2 1.00 ± 0.01 1.00 ± 0.01 0.99 ± 0.01 NS UBC_3 1.00 ± 0.00b 1.02 ± 0.01ab 1.03 ± 0.01a 0.0442 Table 4. DNA methylation analysis of genes of interest. Analysis was performed for 9 DE genes from RNA- Seq and qPCR and 3 key central genes identified by pathway analysis by MeDIP-qPCR. Table 4. DNA methylation analysis of genes of interest. Table 4. DNA methylation analysis of genes of interest. Analysis was performed for 9 DE genes from RNA- Seq and qPCR and 3 key central genes identified by pathway analysis by MeDIP-qPCR. www.nature.com/scientificreports/ Analysis was performed for 9 DE genes from RNA- Seq and qPCR and 3 key central genes identified by pathway analysis by MeDIP-qPCR. Figure 5. Part of the promoter sequence of UBC and the primer sequence of UBC_3 (depicted bold and underlined). The qPCR fragment is located around the transcription start site (CTG, italic underlined), starting from 31 bp upstream of the 5′ -UTR (Untranslated Region) to 155 bp into this region and 553 bp upstream of the translation start codon (ATG, italic underlined). In the PCR fragment of the UBC_3 primers there are 14 CpG sites, of which one or more have differential methylation. Table 4. DNA methylation analysis of genes of interest. Analysis was performed for 9 DE genes from RNA- eq and qPCR and 3 key central genes identified by pathway analysis by MeDIP-qPCR. Figure 5. Part of the promoter sequence of UBC and the primer sequence of UBC_3 (depicted bold and underlined). The qPCR fragment is located around the transcription start site (CTG, italic underlined), starting from 31 bp upstream of the 5′ -UTR (Untranslated Region) to 155 bp into this region and 553 bp upstream of the translation start codon (ATG, italic underlined). In the PCR fragment of the UBC_3 primers there are 14 CpG sites, of which one or more have differential methylation. observed. In rat dams, fed a protein-restricted diet, an increase of the regulatory enzyme of the gluconeogenesis phosphoenolpyruvate carboxykinase PCK1 mRNA and increased activity was detected in liver of the progeny Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 8 www.nature.com/scientificreports/ until 11 months of age, suggesting that programming of the metabolism also extends to the regulation of gene expression21.f p In offspring from low-protein-diet rat models, alterations in DNA methylation were observed in association with changes in gene expression13,14. In the present study, a screening of the DNA methylation of the 9 DE genes and the three key central genes (UBC, NR3C1 and ELAVL1) was performed by MeDIP-qPCR. The data suggest that DNA methylation of fragment UBC_3 of the ubiquitin C (UBC) gene is increased in the albumen-deprived hens compared to the non-manipulated hens. The fact that the gene expression in this region did not differ in the RNA-Seq experiment does not exclude the possibility that these changes in DNA methylation could be related with distal regulation of other genes. Methods hi Ethics statement. All experiments were conducted in strict accordance with the European Communities Council Directive (2010/63/EC) and were approved by the Institutional Ethical Committee of KU Leuven (P002–2012). Chickens. The set-up of this experiment was previously described by7. More information on the technique of albumen removal is available in11. In brief, fertilized Isa Brown layer-type eggs (Vepymo, Poppel, Belgium) were randomly divided over the three treatments and incubated together. After one day of incubation, a hole was drilled in the egg, 3 ml of outer thin albumen was removed using a needle and syringe and replaced by about the same volume of sterile saline, followed by sealing of the hole using a drop of paraffin (albumen-deprived group). A sham-manipulated group was mock-treated similar to the albumen-deprived group, except for the actual albu- men removal and saline injection. A third group received no treatment (non-manipulated group). A discussion of the technique of albumen removal can be found in10,11. After hatch, only the female chicks were reared in floor pens with wood shavings as litter until 55 weeks of age. All floor pens were located in one environmentally con- trolled room. Room temperature was initially set at 34 °C and was gradually decreased until 20 °C at 5 weeks of age. This temperature was maintained until 55 weeks of age. A 23 h light cycle was initially provided and gradually decreased to 10 h at 6 weeks. At 14 weeks, the light cycle was gradually prolonged to 15 h at 19 weeks to stimulate sexual maturation. The hens received soy-wheat-corn based diets ad libitum formulated according to the devel- opmental requirements based on the Isa Brown Management Guide (Research Diet Services, Wijk bij Duurstede, the Netherlands)11. At 55 weeks of age, the laying hens were killed and 8 liver samples per group were randomly collected and snap frozen in liquid nitrogen before storage at − 80 °C. Frozen liver tissue was homogenized by grinding into powder in liquid nitrogen before use. Genome-wide screening for differential gene expression using RNA-Seq. RNA isolation. Total RNA was extracted from approximately 50 mg of crushed liver collected from hens at 55 weeks of age (n = 8 per group) as previously described by5. The RNA concentration and quality (260/280 ratio) was measured using UV-spectroscopy (Implen, Westburg, Leusden, The Netherlands). The integrity of the RNA was electrophoreti- cally verified. www.nature.com/scientificreports/ The fragment is located around the transcription start site, starting from 31 bp upstream of the 5′ -UTR (Untranslated Region) to 155 bp into this region and 553 bp upstream of the translation start codon (CTG) (Fig. 5). 14 CpG’s are contained within this fragment and one or more of these CpGs have increased methylation in the albumen-deprived hens. Ubiquitin C is one of the genes that produces ubiquitin, the precursor of polyubiquitin. Ubiquination is associated with protein degradation, DNA repair, cell cycle regulation, kinase modification, endocytosis, and regulation of other cell signaling pathways. Since the pres- ent study only performed a DNA methylation analysis biased towards specific genes, a genome-wide approach will be needed in order to determine the real extent of changes in DNA methylation that could be generated as result of the experimental treatment used. Moreover, other epigenetic modifiers such as histone modification or microRNA could be examined, as demonstrated in a low protein rat model22,23. Conclusions In conclusion, the present results demonstrate for the first time that prenatal protein undernutrition by albu- men removal leads to long-term alterations of the hepatic transcriptome in the chicken. As expected, pathways involved in embryonic development were affected by this treatment. In addition, changes in amino acid metab- olism and protein synthesis prove the efficacy of the application of strictly nutritional effects. Limited effects of DNA methylation were observed in the regulation of the currently examined DE genes. Methods hi Per sample, GC-content was corrected using full quantile normalization on bins of GC-content with the EDASeq package from Bioconductor29 for within-sample nor- malization. Between-sample normalization was carried out to correct for library size and RNA composition, as it is known that these are sources of large variation, using full quantile normalization with the EDASeq pack- age from Bioconductor30. The statistical modeling assumed the design of the experiment as log (Count) = β 1 × non-manipulated + β 2 × sham-manipulated + β 3 albumen-deprived. For each gene, the coefficients β were esti- mated with the EdgeR package (3.2.3)31 of Bioconductor, by fitting a negative binomial generalized linear model (GLM)28. The following comparisons were tested: non-manipulated vs. albumen-deprived, sham-manipulated vs. albumen-deprived and sham-manipulated vs. non-manipulated. Differentially expressed genes were selected for confirmation by qPCR based on the criteria that the p-value should be < 0.001 combined with a cut-off fold change of 2 or absolute log2-ratio larger than 132. Identification of relevant biological pathways. A list of differentially expressed genes for pathway analysis was created by selecting the differentially expressed transcripts with P < 0.005 and fold change > 1.5. This list of differentially expressed genes between the albumen-deprived hens and both non-manipulated and sham-manipulated hens or only the sham-manipulated hens was imported into the Ingenuity Pathway Analysis (IPA; Ingenuity Systems, Redwood City, CA, USA) to identify biological interactions between the gene products. The biological interaction scores were defined by the IPA statistical algorithm based on its knowledge base, and the P value was calculated according to Fisher’s exact test. Confirmation and validation of RNA-Seq results via qPCR. DNase treatment and reverse transcrip- tion. RNA samples (n = 8 per group) were DNase treated with RQ1 RNase-Free DNase (M6101, Promega, Fitchburg, WI, USA), according to the manufacturer’s instructions. The RNA was transcribed into cDNA using the Reverse Transcription system (A3500, Promega, Madison, WI, USA). Denaturation was performed for 3 min at 80 °C followed by 45 min at 42 °C for the reverse transcription. The reaction was terminated by heating the samples for 5 min at 95 °C. Primer design. The primer design was described previously5. In Supplementary Table S1 online, all primers are listed, both for reference genes and the genes of interest. All measurements were performed in the linear range of amplification with correlation coefficient > 0.99. Methods hi In addition, integrity of RNA samples used for RNA-Seq was checked on the BioAnalyzer (Agilent Technologies, Santa Clara, CA, USA). Sequencing and quality control of the reads. The sequencing was performed in collaboration with the Genomics Core of the UZ Leuven (Belgium). Samples (n = 3 per group, 2 μ g RNA) were prepared by TruSeq library prepa- ration (RS122, Illumina, San Diego, CA, USA) to generate single end unstranded sequencing libraries, according to the manufacturer’s recommendations. Sequencing of all samples was carried out in 1 lane of 1 flow cell on the HiSeq2000 (Illumina), using single end chemistry with read lengths of 50 base pairs. Between 17.3 and 26.5 million reads were sequenced for each sample. The quality of the reads was checked using the FASTQC software (Illumina). All the reads passed the Phred score (≥ 28). Data analysis of the RNA-Seq data. Data analysis was performed in collaboration with the Nucleomics Core of the VIB (Belgium). Reads were mapped to the chicken genome (Galgal4.71) using TopHat (v2.0.8b)24. Quality filtering was performed with SAMtools (0.1.19–44428cd) removing all reads from the alignment that are non-primary mappings or have a mapping quality ≤ 2025. The number of mapped reads varied between 16 and Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 9 www.nature.com/scientificreports/ 24.8 million reads per sample as calculated by Picard (1.92) (BroadInstitute) resulting in a mapping efficiency of approximately 93%. Per sample, the transcripts were identified from the mapping with Cufflinks v2.1.126. With the Cuffmerge tool, all per-sample transcript lists were merged together with the reference annotation into one file. With Cuffcompare, the identified transcripts were matched with known transcripts and genes, based on overlapping coordinates on the reference genome. About 22.1% novel loci were identified, resulting in a total of 22,082 genes. The number of reads in the alignments that overlap with the gene features were counted using HTSeq-count (0.5.4p3)27. The reads that could be attributed to more than one gene (ambiguous, about 1.3 million per sample) or could not be attributed to any gene (no feature, about 1.1 million per sample) were not counted. Exclusion caused by ambiguity excluded about 6.9% and no feature about 6.1% of the mapped reads per sam- ple. The 8,483 genes that had less than 1 counts-per-million were considered ‘absent’ and therefore removed from the dataset28. As such, 13,599 genes were left. Methods hi Final primer concentration was 0.3 μ M for all primer pairs, except for BMF, where 0.15 μ M was used. qRT-PCR. Quantitative real-time PCR (qRT-PCR) measurements were performed as described previously5. For the analysis of the qRT-PCR output, the 2−ΔΔCT method of relative quantification was used33. Expression of genes was normalized to the geometric average of the two references genes: β -actin (ACTB) and peptidylprolyli- somerase D (PPID).Technical confirmation results were based on the same three samples as used in the RNA-Seq, whereas in the biological validation the sample size was increased to 8 samples per group. Methylated DNA immunoprecipitation and quantitative real-time PCR (MeDIP-qPCR). Genomic DNA isolation. One ml of SNET buffer (1% SDS; 0.4M NaCl; 5 mM EDTA; 20 mM Tris.HCl pH 8.0) and 20 μ l of Proteinase K (00120437, Thermo Scientific, Waltham, MA, USA) was added to 100 mg of homog- enized liver tissue and incubated overnight in a water bath at 56 °C. The next day 1 ml of a phenol: chloroform: isoamyl alcohol 25:24:1 mixture (A0889, AppliChem, Darmstadt, Germany) was added and thoroughly vortexed. After centrifugation (14,000 rpm, 10 min, 4 °C), the top layer was transferred to a new tube and the process of adding, vortexing and centrifugating was repeated several times with addition of subsequently 1 ml of the phe- nol: chloroform: isoamyl alcohol 25:24:1 mixture and 1 ml of chloroform (1.02445, Merck Millipore, Readington Township, NJ, USA). Finally, 1 ml of isopropanol (20842, Prolabo, VWR, West Chester, PA, USA) was gently mixed with the top layer, centrifuged (14,000 rpm, 5 min, 4 °C) and the resulting supernatant was discarded. The DNA pellet was washed with 500 μ l 70 ethanol (14,000 rpm, 5 min, 4 °C) and the pellet was dissolved in 150 μ l milliQ water. The DNA concentration and purity was measured on a NanoDrop 1000 Spectrophotometer (Thermo Scientific, Waltham, MA, USA). Shearing of genomic DNA. 6 μ g of genomic DNA was diluted in 30 μ l milliQ and subsequently fragmented by sonication in icecold water in the Biorupter (Diagenode, Liège, Belgium) for 1 to 4 cycles of 15 times 10 sec ON–10 sec OFF (and briefly spinned down after each cycle) until the fragment size was between 100–600 bp, with the majority of the fragments around 250 bp. Fragment size was verified by electrophoresis on a 2% agarose gel (A9539, Sigma-Aldrich, St. Louis, MO, USA). Methods hi 5.5 μ g of frag- mented genomic DNA was diluted to 400 μ l in TE buffer (10 mM Tris.HCl pH 7.5, 1 mM EDTA) and dena- tured for 10 minutes at 95 °C, followed by immediate cooling on ice for 5 minutes. 100 μ l of 5X IP buffer (50 mM Na.H2PO4 pH 7, 0.7M NaCl, 0.4 mM Triton X-100) and 2.5 μ l 5-mC monoclonal antibody (C15200006, Diagenode) were added and incubated overnight on a rotating platform at 4 °C. Protein A/G PLUS-Agarose Immunoprecipitation Reagent (sc-2003, Santa Cruz Biotechnology, Inc, Dallas, TX, USA) was pre-washed with 0.1 mg/ml BSA in 1X PBS and resuspended in 40 μ l IP buffer. Beads were added to the DNA-antibody complex and incubated for 2 h on a rotating platform at 4 °C. Beads bound to the DNA-antibody complex were washed 3 times with 1 ml 1X IP buffer. To release the methylated DNA from the beads, the beads were resuspended in 250 μ l digestion buffer (50 mM Tris.HCl pH 8, 10 mM EDTA, 0.5% SDS), 3.5 μ l proteinase K (00120437, Thermo Scientific) and incubated overnight on a rotating platform at 55 °C. The methylated DNA was purified by first mixing with 250 μ l phenol solution (P4557, Sigma-Aldrich) and centrifugation (14,000 rpm, 5 min, room tem- perature) and repeating this with addition of 250 μ l chloroform:isoamyl alcohol (24:1, C0549, Sigma-Aldrich). Washing was performed by addition of 2 μ l GlycoBlue Coprecipitant (AM9515, Applied Biosystems, Foster City, CA, USA), 20 μ l 5M NaCl and 500 μ l 99.5% icecold ethanol and placement at − 20 °C for at least 1 hour. After centrifugation (14,000 rpm, 15 min, 4 °C), the supernatant was removed and the pellet was washed with 500 μ l of icecold 70% ethanol and centrifuged again. The pellet was resuspended in 30 μ l milliQ. The DNA concentration was measured on the NanoDrop 1000 Spectrophotometer (Thermo Scientific). The yield was about 10–30 ng/μ l. Whole Genome amplification. 30 ng of each MeDIP sample was amplified using the GenomePlex Complete Whole Genome Amplification kit (WGA2, Sigma-Aldrich) according the manufacturer’s instruction, except no fragmentation was performed, as the samples were already sonicated. After amplification, the DNA was cleaned using the GeneJET PCR Purification kit (K0701, Thermo Scientific). Methods hi The DNA concentration was measured on the NanoDrop 1000 Spectrophotometer (Thermo Scientific) and the samples were diluted to 20 ng/μ l. Selection of CpG rich regions around transcription start site of interesting genes and primer design. For each gene of interest, several CpG rich (> 4 CpG’s) fragments (150–200 bp) were selected in the promoter region (5000 bp upstream of transcription start site) and around the transcription start site and the corresponding primers for these fragments were designed (Supplementary Table S2 online) using Primer Designing Tool (NCBI). The prim- ers were purchased from Life Technologies (Carlsbad, CA, USA) and verified as described previously. Quantitative real-time PCR (qPCR). Quantitative real-time PCR (qRT-PCR) measurements were performed in triplicate using the LightCycler480 qPCR machine (Roche Applied Science, Penzberg, Germany). For every reaction, 5 μ l of LightCycler 480 SYBR Green I Master Mix (04 887 352 001, Roche), 0.5–1 μ M of forward and reverse primer, 2.5 μ l of DNA (20 ng/μ l) and milliQ to a final volume of 10 μ l were added together in a 96-well plate (LightCycler 480 96 Multiwell Plate white, 04 729 692 001, Roche) and sealed with LightCycler 480 Sealing Foil (04 729 757 001, Roche). The PCR reaction program began with 5 min heating at 95 °C followed by 45 cycles of 10 sec at 95 °C, 20 s at 60 °C and 30 sec at 72 °C. In addition, a melting curve analysis was performed to check the specificity of the primers (5 seconds at 95 °C, 1 minute at 55–60 °C, temperature gradually increased until 97 °C is reached). Statistical Analysis. qPCR and MeDIP-qPCR data were processed with the statistical software package SAS version 9.3 (SAS Institute Inc., Cary). The gene expression values (2−ddCt) and Ct values of the MeDIP-qPCR were analyzed using one-way ANOVA with treatment (non-manipulated, sham-manipulated and albumen-deprived) as independent variable. When a significant effect of treatment was demonstrated, the means were further com- pared by a posthoc Tukey’s test. All data are shown as mean ± SEM. Methods hi Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 10 www.nature.com/scientificreports/ Methylated DNA immunoprecipitation (MeDIP). This method was previously described by34. 5.5 μ g of frag- mented genomic DNA was diluted to 400 μ l in TE buffer (10 mM Tris.HCl pH 7.5, 1 mM EDTA) and dena- tured for 10 minutes at 95 °C, followed by immediate cooling on ice for 5 minutes. 100 μ l of 5X IP buffer (50 mM Na.H2PO4 pH 7, 0.7M NaCl, 0.4 mM Triton X-100) and 2.5 μ l 5-mC monoclonal antibody (C15200006, Diagenode) were added and incubated overnight on a rotating platform at 4 °C. Protein A/G PLUS-Agarose Immunoprecipitation Reagent (sc-2003, Santa Cruz Biotechnology, Inc, Dallas, TX, USA) was pre-washed with 0.1 mg/ml BSA in 1X PBS and resuspended in 40 μ l IP buffer. Beads were added to the DNA-antibody complex and incubated for 2 h on a rotating platform at 4 °C. Beads bound to the DNA-antibody complex were washed 3 times with 1 ml 1X IP buffer. To release the methylated DNA from the beads, the beads were resuspended in 250 μ l digestion buffer (50 mM Tris.HCl pH 8, 10 mM EDTA, 0.5% SDS), 3.5 μ l proteinase K (00120437, Thermo Scientific) and incubated overnight on a rotating platform at 55 °C. The methylated DNA was purified by first mixing with 250 μ l phenol solution (P4557, Sigma-Aldrich) and centrifugation (14,000 rpm, 5 min, room tem- perature) and repeating this with addition of 250 μ l chloroform:isoamyl alcohol (24:1, C0549, Sigma-Aldrich). Washing was performed by addition of 2 μ l GlycoBlue Coprecipitant (AM9515, Applied Biosystems, Foster City, CA, USA), 20 μ l 5M NaCl and 500 μ l 99.5% icecold ethanol and placement at − 20 °C for at least 1 hour. After centrifugation (14,000 rpm, 15 min, 4 °C), the supernatant was removed and the pellet was washed with 500 μ l of icecold 70% ethanol and centrifuged again. The pellet was resuspended in 30 μ l milliQ. The DNA concentration was measured on the NanoDrop 1000 Spectrophotometer (Thermo Scientific). The yield was about 10–30 ng/μ l. Methylated DNA immunoprecipitation (MeDIP). This method was previously described by34. References e e e ces 1. Ravelli, A. C. J. et al. Glucose tolerance in adults after prenatal exposure to famine. Lancet 351, 173–177 (1998).ht t 2. Hales, C. N. & Barker, D. J. The thrifty phenotype hypothesis. Br. Med. Bull. 60, 5–20 (2001).h t 2. Hales, C. N. & Barker, D. J. The thrifty phenotype hypothesis. Br. Med. Bull. 60, 5–20 (2001).h ht 3. Ozanne, S. E. & Hales, C. N. The long term consequences of intra-uterine protein malnutrition on glucose metabolism. Proc. Nut Soc. 58, 615–619 (1999). ll f l h d l f l h k h b l 4. Hill, W. L. Importance of prenatal nutrition to the development of a precocial chick. Dev. Psychobiol. 26, 237–249 (1993). 5. Willems, E. et al. Embryonic protein undernutrition by albumen removal programs th during the perinatal period in an avian model. PLoS One 9(4), e94902 (2014a). 5. Willems, E. et al. Embryonic protein undernutrition by albumen removal programs the hepatic amino acid and glucose metabolis during the perinatal period in an avian model. PLoS One 9(4), e94902 (2014a). g p p 6. Willems, E. et al. Reduced protein availability by albumen removal during chicken embryogenesis decreases body weight and induces hormonal changes. Exp Physiol, doi: 10.1113/EP085313 (2015). g p y 7. Willems, E. et al. Effects of nutritional programing on growth and metabolism caused by albumen removal in an avian model J. Endocrinol. 225, 89–100 (2015). J. Endocrinol. 225, 89–100 (2015). 8 F B M & Vi M A D l p t Of Ih A i E b b h i l d ph i l i l t d (Ch d H ll 8. Freeman, B. M. & Vince, M. A. Development Of Ihe Avian Embryo: a behavioural and physiological study (Chapman and Hall London, 1974).l 9. Finkler, M. S., Van Orman, J. B. & Sotherland, P. R. Experimental manipulation of egg quality in chickens: influence of albumen and yolk on the size and body composition of near-term embryos in a precocial bird. J. Comp. Physiol. B168, 17–24 (1998). y y p y p p y 10. Everaert, N. et al. The effect of albumen removal before incubation (embryonic protein undernutrition) on the post-hatch performance, regulators of protein translation activation and proteolysis in neonatal broilers. Br. J. Nutr. 110, 265–274 (2013). www.nature.com/scientificreports/ K. Epigenetic transgenerational actions of vinclozolin on promote regions of the sperm epigenome. PLoS One 5(9), e13100 (2010). g p p g 5. Edgar, R., Domrachev, M. & Lash, A. E. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository Nucleic Acids Res. 30(1), 207–210 (2002). Author Contributions N.E., J.B., E.D. and E.W. designed and analysed the chicken experiment. N.E., E.W. and N.B. designed the RNA- Seq. S.J. supported with the data analysis. E.W., C.G.-B. and P.J. designed, performed and analysed the MeDIP- qPCR. E.W. prepared and edited the manuscript prior to submission. All authors read and agree with the final format of the manuscript. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 14. Lillycrop, K. A. et al. Induction of altered epigenetic regulation of the hepatic glucocorticoid receptor in the offspring of rats fed a protein-restricted diet during pregnancy suggests that reduced DNA methyltransferase-1 expression is involved in impaired DNA methylation and changes in histone modifications. Br. J. Nutr. 97, 1064–1073 (2007).f y gi ( ) 5. Gong, L., Pan, Y. X. & Chen, H. Gestational low protein diet in the rat mediates Igf2 gene expression in male offspring via altered hepatic DNA methylation. Epigenetics 5, 619–626 (2010).f p y p g 6. van Straten, E. M. E. et al. Sex-dependent programming of glucose and fatty acid metabolism in mouse offspring by maternal protein restriction. Gender Med. 9, 166–179 (2012). , ( ) 17. Lillycrop, K. A. et al. Maternal protein restriction with or without folic acid supplementation during pregnancy alters the hepatic transcriptome in adult male rats. Br. J. Nutr. 103, 1711–1719 (2010). 18. Armitage, J. A., Khan, I. Y., Taylor, P. D., Nathanielsz, P. W. & Poston, L. Developmental programming of the metabolic syndrome by maternal nutritional imbalance: how strong is the evidence from experimental models in mammals? J. Physiol. 561.2, 355–377 (2004).fi 9. Shi, Y. & He, M. Differential gene expression identified by RNA-Seq and qPCR in two sizes of pearl oyster (Pinctada fucata). Gene 538, 313–322 (2014).hf ( ) 20. Rae, M. T. et al. The effects of undernutrition, in utero, on reproductive function in adult male and female sheep. Anim. Reprod. Sci. 72, 63–71 (2002).f 1. Desai, M. et al. Programming of hepatic insulin-sensitive enzymes in offspring of rat dams fed a protein-restricted diet. Am. J Physiol. 272, G1083–1090 (1997). 22. Zheng, S., Rollet, M. & Pan, Y. X. Maternal protein restriction during pregnancy induces CCAAT/enhancer-binding protein (C/EBPβ ) expression through the regulation of histone modification at its promoter region in female offspring rat skeletal muscle. Epigenetics 6, 161–170 (2011). p g ( ) 23. Sohi, G., Revesz, A., Ramkumar, J. & Hardy, D. B. Higher hepatic miR-29 expression in undernourished male rats durin postnatal period targets the long-term repression of IGF-1. Endocrinology 156, 3069–3076 (2015). ll h l b d l h f ( p p g g p gy 24. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009). p p g g p gy 24. Trapnell, C., Pachter, L. & Salzberg, S. L. www.nature.com/scientificreports/ TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009). 25. Li, H. et al. 1000 Genome Project Data Processing Subgroup: The Sequence alignment/map (SAM) format and SAMtools. Bioinformatics 25, 2078–2079 (2009).i 24. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009). 25 Li H et al 1000 Genome Project Data Processing Subgroup: The Sequence alignment/map (SAM) format and SAMtools 24. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering sp 26. Trapnell, C. et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515 (2010). f 27. Anders, S., Pyl, P. T. & Huber, W. HTSeq: A Python framework to work with high-throughput sequencing data. bioRxiv, doi: 10.1101/002824 (2014).f 28. Robinson, M. D. & Smyth, G. K. Moderated statistical tests for assessing differences in tag abundance. Bioinformatics 23, 2881– (2007). ( ) 29. Risso, D., Schwartz, K., Sherlock, G. & Dudoit, S. GC-content normalization for RNA-Seq data. BMC Bioinformatics 12, 480 (2 29. Risso, D., Schwartz, K., Sherlock, G. & Dudoit, S. GC-content normalization for RNA-Seq data. BMC Bioinformatics 12, 480 (2011). 30 Robinson M D & Oshlack A A scaling normalization method for differential expression analysis of RNA Seq data Genome Biol . Risso, D., Schwartz, K., Sherlock, G. & Dudoit, S. GC-content nor 29. Risso, D., Schwartz, K., Sherlock, G. & Dudoit, S. GC-content normalization for RNA-Seq data. BMC Bioinformatics 12, 480 (2011). 30. Robinson, M. D. & Oshlack, A. A scaling normalization method for differential expression analysis of RNA-Seq data. Genome Biol. 11 R25 (2010) 9. Risso, D., Schwartz, K., Sherlock, G. & Dudoit, S. GC-content normalization for RNA-Seq data. BMC Bioinformatics 12, 480 (2011) 0. Robinson, M. D. & Oshlack, A. A scaling normalization method for differential expression analysis of RNA-Seq data. Genome Biol 11, R25 (2010).f 31. Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital expression data. Bioinformatics 26, 139–140 (2010).h p f 32. MAQC Consortium. The MicroArray quality control (MAQC) project shows inter- and intraplatform reproducibility of expression measurements (2006). Nat. Biotechnol. 24, 1151–1161 (2006). 33. Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCt met Methods 25, 402–408 (2001). 4. Guerrero-Bosagna, C., Settles, M., Lucker, B. & Skinner, M. Acknowledgements g The authors would like to thank the Genomics Core (UZ Leuven) for the sequencing of the reads and the Nucleomics core (VIB, Vlaams Instituut voor Biotechnologie) for the data analysis of the RNA-Seq. This research was supported by the Fund for Scientific Research Flanders (FWO-Vlaanderen) (FWO nr. G.0566.08) and the European Research Council senior researcher project: GENEWELL–Genetics and epigenetics of animal welfare. The authors are members of the COST Action FA1201 Epiconcept: Epigenetics and periconception environment. g The authors would like to thank the Genomics Core (UZ Leuven) for the sequencing of the reads and the Nucleomics core (VIB, Vlaams Instituut voor Biotechnologie) for the data analysis of the RNA-Seq. This research was supported by the Fund for Scientific Research Flanders (FWO Vlaanderen) (FWO nr G 0566 08) and the The authors would like to thank the Genomics Core (UZ Leuven) for the sequencing of the reads and the Nucleomics core (VIB Vlaams Instituut voor Biotechnologie) for the data analysis of the RNA-Seq This research The authors would like to thank the Genomics Core (UZ Leuven) for the sequencing of the reads and the Nucleomics core (VIB, Vlaams Instituut voor Biotechnologie) for the data analysis of the RNA-Seq. This research was supported by the Fund for Scientific Research Flanders (FWO-Vlaanderen) (FWO nr. G.0566.08) and the European Research Council senior researcher project: GENEWELL–Genetics and epigenetics of animal welfare. The authors are members of the COST Action FA1201 Epiconcept: Epigenetics and periconception environment References 11 Willems E et al Partial albumen removal early during embryonic development of layer type chickens has negative consequences 10. Everaert, N. et al. The effect of albumen removal before incubation (embryonic protein undernutrition) on the post-hatch performance, regulators of protein translation activation and proteolysis in neonatal broilers. Br. J. Nutr. 110, 265–274 (2013). 11. Willems, E. et al. Partial albumen removal early during embryonic development of layer-type chickens has negative consequences on laying performance in adult life Poult Sci 92 1905 1915 (2013) on laying performance in adult life. Poult. Sci. 92, 1905–1915 (20f y g p 12. Burdge, G. C., Hanson, M. A., Slater-Jefferies, J. L. & Lillycrop, K. A. Epigenetic regulation of transcription: a mechanism for inducing variations in phenotype (fetal programming) by differences in nutrition during early life? Br. J. Nutr. 97, 1036–1046 (2007). ll h ll k d f d d 12. Burdge, G. C., Hanson, M. A., Slater-Jefferies, J. L. & Lillycrop, K. A. Epigenetic regulation of transcription: a mechanism for inducing variations in phenotype (fetal programming) by differences in nutrition during early life? Br. J. Nutr. 97, 1036–1046 (2007). 13. Lillycrop, K. A., Phillips, E. S., Jackson, A. A., Hanson, M. A. & Burdge, G. C. Dietary protein restriction of pregnant rats induces and folic acid supplementation prevents epigenetic modification of hepatic gene expression in the offspring. J. Nutr. 135, 1382–1386 (2005). f 13. Lillycrop, K. A., Phillips, E. S., Jackson, A. A., Hanson, M. A. & Burdge, G. C. Dietary protein restriction of pregnant rats induces and folic acid supplementation prevents epigenetic modification of hepatic gene expression in the offspring. J. Nutr. 135, 1382–1386 (2005). Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 11 Additional Informationh Data availability: The data sets supporting the results of this article are available in the NCBI’s Gene Expression Omnibus repository35, and are available through GEO Series accession number GSE57938(http://www.ncbi. nlm.nih.gov/geo/query/acc.cgi?acc=GSE57938). Supplementary information accompanies this paper at http://www.nature.com/srep Competing financial interests: The authors declare no competing financial interests. Competing financial interests: The authors declare no competing financial interests. How to cite this article: Willems, E. et al. Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model. Sci. Rep. 6, 20837; doi: 10.1038/ srep20837 (2016). How to cite this article: Willems, E. et al. Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model. Sci. Rep. 6, 20837; doi: 10.1038/ srep20837 (2016). How to cite this article: Willems, E. et al. Differential Expression of Genes and DNA Methylation associated with Prenatal Protein Undernutrition by Albumen Removal in an avian model. Sci. Rep. 6, 20837; doi: 10.1038/ srep20837 (2016). 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/ Scientific Reports \| 6:20837 \| DOI: 10.1038/srep20837 12
https://openalex.org/W2921583638	http://pure-oai.bham.ac.uk/ws/files/59421566/fcell_07_00034.pdf	English	null	How HSCs Colonize and Expand in the Fetal Niche of the Vertebrate Embryo: An Evolutionary Perspective	Frontiers in cell and developmental biology	2,019	cc-by	11,381	How HSCs colonize and expand in the fetal niche of the vertebrate embryo Mahony, Christopher B.; Bertrand, Julien Y. DOI: 10.3389/fcell.2019.00034 10.3389/fcell.2019.00034 License: Creative Commons: Attribution (CC BY) License: Creative Commons: Attribution (CC BY) Document Version Publisher's PDF, also known as Version of record Citation for published version (Harvard): Mahony, CB & Bertrand, JY 2019, 'How HSCs colonize and expand in the fetal niche of the vertebrate embryo: an evolutionary perspective', Frontiers in cell and developmental biology, vol. 7, no. MAR, 34. https://doi.org/10.3389/fcell.2019.00034, https://doi.org/10.3389/fcell.2019.00034 Link to publication on Research at Birmingham portal Publisher Rights Statement: Checked for eligibility: 08/04/2019 Publisher Rights Statement: Checked for eligibility: 08/04/2019 Publisher Rights Statement: Checked for eligibility: 08/04/2019 General rights U l li General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. d l d d/ i f h bli i f h i i •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non commercial research y y y p •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. y •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. here a licence is displayed above, please note the terms and conditions of the licence govern your use of this document Where a licence is displayed above, please note the terms an When citing, please reference the published version. Citation for published version (Harvard): Mahony, CB & Bertrand, JY 2019, 'How HSCs colonize and expand in the fetal niche of the vertebrate embryo: an evolutionary perspective', Frontiers in cell and developmental biology, vol. 7, no. MAR, 34. https://doi.org/10.3389/fcell.2019.00034, https://doi.org/10.3389/fcell.2019.00034 How HSCs Colonize and Expand in the Fetal Niche of the Vertebrate Embryo: An Evolutionary Perspective Christopher B. Mahony and Julien Y. Bertrand* Department of Pathology and Immunology, Faculty of Medicine, CMU, University of Geneva, Geneva, Switzerland Department of Pathology and Immunology, Faculty of Medicine, CMU, University of Geneva, Geneva, Switzerland Rare hematopoietic stem cells (HSCs) can self-renew, establish the entire blood system and represent the basis of regenerative medicine applied to hematological disorders. Clinical use of HSCs is however limited by their inefﬁcient expansion ex vivo, creating a need to further understand HSC expansion in vivo. After embryonic HSCs are born from the hemogenic endothelium, they migrate to the embryonic/fetal niche, where the future adult HSC pool is established by considerable expansion. This takes place at different anatomical sites and is controlled by numerous signals. HSCs then migrate to their adult niche, where they are maintained throughout adulthood. Exactly how HSC expansion is controlled during embryogenesis remains to be characterized and is an important step to improve the therapeutic use of HSCs. We will review the current knowledge of HSC expansion in the different fetal niches across several model organisms and highlight possible clinical applications. Take down policy Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, please contact UBIRA@lists.bham.ac.uk providing details and we will remove access to the work immediately and investigate. Download date: 24. Oct. 2024 REVIEW published: 12 March 2019 doi: 10.3389/fcell.2019.00034 published: 12 March 2019 doi: 10.3389/fcell.2019.00034 Edited by: Edited by: Hatem E. Sabaawy, Rutgers University, The State University of New Jersey, United States Reviewed by: César Nombela Arrieta, University of Zurich, Switzerland Eirini Trompouki, Max-Planck-Institut für Immunbiologie und Epigenetik, Germany Keywords: zebraﬁsh, mammals, CHT, fetal liver, hematopoietic (stem) cells, caudal hematopoietic tissue, microenvironment Eirini Trompouki, Max-Planck-Institut für Immunbiologie und Epigenetik, Germany Keywords: zebraﬁsh, mammals, CHT, fetal liver, hematopoietic (stem) cells, caudal hematopoietic tissue, microenvironment Correspondence: Hematopoiesis is a highly conserved process across many organisms that culminates with the emergence of hematopoietic stem cells (HSCs). In zebraﬁsh and mammals, hematopoiesis initiates with the emergence of primitive myeloid and erythroid cells (Palis et al., 2001; Bertrand et al., 2005; Palis, 2014; McGrath et al., 2015). Similar cells, prohemocytes, are also detected in drosophila larvae that give rise to plasmatocytes (macrophage-like cells) and crystal cells (platelet-like cells) (Lebestky et al., 2000). Primitive myeloid and erythroid cells are also detected in xenopus embryos (Ciau- Uitz et al., 2014). Following this, deﬁnitive hematopoiesis then occurs in two distinct waves in vertebrates. The ﬁrst wave is characterized by the transient erythro-myeloid precursors (EMPs) that arise in the yolk sac in mice and humans (Bertrand et al., 2005; McGrath et al., 2015), the posterior blood island in zebraﬁsh (Bertrand et al., 2007) and the posterior-lateral ventral blood island in xenopus (Ciau-Uitz et al., 2014). The appearance of EMPs in chicken embryos remains to be determined. The second wave consists of HSC speciﬁcation from the aortic hemogenic endothelium by the highly conserved process of endothelial-to-hematopoietic transition (EHT). The formation of the hemogenic endothelium requires the correct balance of extrinsic and intrinsic factors to initiate the expression of speciﬁc transcription factors, such as runx1 and gata2. During mammalian and avian development, HSC speciﬁcation occurs in the aorta-gonads-mesonephros (AGM) region where they form intra-aortic clusters (Jaﬀredo et al., 2000; Bollerot et al., 2005a,b; Zovein et al., 2008; Chen et al., 2009; Boisset et al., 2010) between embryonic (E) day 9.5 and 11.5 in mice, between E26 Specialty section: This article was submitted to Stem Cell Research, a section of the journal Frontiers in Cell and Developmental Biology Received: 23 November 2018 Accepted: 25 February 2019 Published: 12 March 2019 Citation: Mahony CB and Bertrand JY (2019) How HSCs Colonize and Expand in the Fetal Niche of the Vertebrate Embryo: An Evolutionary Perspective. Front. Cell Dev. Biol. 7:34. doi: 10.3389/fcell.2019.00034 Specialty section: This article was submitted to Stem Cell Research, a section of the journal Frontiers in Cell and Developmental Biology Received: 23 November 2018 Accepted: 25 February 2019 Published: 12 March 2019 INTRODUCTION Correspondence: Julien Y. Bertrand julien.bertrand@unige.ch Mammals Mammalian HSCs are produced from the ﬂoor of the embryonic aorta in the AGM region (Zovein et al., 2008; Chen et al., 2009; Ivanovs et al., 2011) but also in the vitelline and umbilical arteries (de Bruijn et al., 2000). Additionally, human and murine studies have detected HSCs in the placenta that arise independently from and in parallel with the HSCs from the AGM (Gekas et al., 2005; Ottersbach and Dzierzak, 2005; Rhodes et al., 2008; Gekas et al., 2010). However it remains unknown what contribution placenta-derived HSCs make to the adult stem cell pool. HSCs then colonize the FL from E11 in mice and E28 in humans, mainly in response to CXCL12 (Chou and Lodish, 2010). CXCL12, released from ECs, stromal cells and mesenchymal progenitors, is well characterized for its role in HSC homing, retention and survival in the niche (Ara et al., 2003; Christensen et al., 2004; Sugiyama et al., 2006; Sawitza et al., 2009; Greenbaum et al., 2013). CXCL12 enhances migration of FL-HSCs in combination with stem cell factor (SCF), when compared to BM-HSCs (Christensen et al., 2004). Furthermore, mice lacking CXCL12 or its receptor (CXCR4) display normal FL hematopoiesis but aberrant spleen and BM colonization, suggesting that speciﬁc and distinct signaling environments attract and maintain HSCs (Nagasawa et al., 1996; Ara et al., 2003). In the FL, HSCs are found closely associated with ECs and stromal cells that promote HSC expansion (Tamplin et al., 2015; Khan et al., 2016). Mouse and zebraﬁsh studies have shown that HSCs physically interact with ECs that promote their proliferation in the fetal niche (Tamplin et al., 2015; Khan et al., 2016). We, and others, have shown that this expansion depends on the expression of several cytokines produced by stromal cells and caudal ECs (cECs) (Tamplin et al., 2015; Mahony et al., 2016, 2018). Additional signals are also important for their expansion, which will be discussed in this review. This embryonic expansion is an essential step in the formation of the adult HSC pool and the correct maturation of HSCs. The self-renewing and multipotent properties of HSCs make these cells an excellent target for regenerative medicine protocols (Cavazzana-Calvo et al., 2000; Walasek et al., 2012; Aiuti et al., 2013). Citation: Mahony CB and Bertrand JY (2019) How HSCs Colonize and Expand in the Fetal Niche of the Vertebrate Embryo: An Evolutionary Perspective. March 2019 \| Volume 7 \| Article 34 1 Frontiers in Cell and Developmental Biology \| www.frontiersin.org The Embryonic HSC Niche Mahony and Bertrand TABLE 1 \| Summary of anatomical sites of hematopoiesis in the mentioned species. and E40 in humans and between E3-4 in chickens. During zebraﬁsh development, HSCs emerge between 32 to 60 hours post fertilization (hpf), from the hemogenic endothelium in the dorsal aortal (Bertrand et al., 2010; Kissa and Herbomel, 2010), a process that requires inﬂammatory cytokines produced by neutrophils (Espin-Palazon et al., 2014) and extracellular matrix (ECM) degradation by macrophages to allow HSCs to enter circulation (Travnickova et al., 2015). HSCs are ﬁrst detected in the ventral blood island and then later in the dorsal lateral plate mesoderm in xenopus (Ciau-Uitz et al., 2000) and transient cells with HSC characteristics are closely associated to the cardiac tube in drosophila (Dey et al., 2016). mentioned species. Species HSC emergence HSC expansion Adult hematopoiesis Human AGM Fetal liver Bone marrow Mouse AGM Fetal liver Bone marrow Chicken AGM PAF/YS Bone marrow Xenopus VBI/DLP Fetal liver Bone marrow Zebraﬁsh AGM CHT Kidney marrow Drosophila Lymph node Lymph node Hematopoietic hubs AGM, aorta, gonads, mesonephros. VBI/DLP, ventral blood island/dorsal lateral plate mesoderm. PAF, para aortic foci. YS, yolk sac. CHT, Caudal hematopoietic tissue. p y In all these organisms, the initial speciﬁcation from endothelial cells (ECs) results in a limited number of HSCs that must mature and expand. This is achieved by migrating through diﬀerent niches, each in distinct anatomical locations that contain speciﬁc microenvironments. The ﬁrst niche that expands HSCs in mouse, humans and xenopus is the fetal liver (FL) (Ema and Nakauchi, 2000; Ciau-Uitz et al., 2014) before they migrate to the bone marrow (BM). In contrast, zebraﬁsh HSCs expand in the caudal hematopoietic tissue (CHT) (Tamplin et al., 2015) and then migrate to the kidney marrow (KM) and chicken HSCs expand in the para-aortic foci (PAF) before seeding the BM (Dunon and Imhof, 2000; Jaﬀredo et al., 2000; Bollerot et al., 2005a,b). This process is diﬀerent in drosophila where an initial wave of HSCs (derived from head mesoderm) arises early during larvae development, followed by a second wave of HSCs found in the lymph gland (Lebestky et al., 2000; Dey et al., 2016). Citation: HSCs are then seeded in hematopoietic clusters in the dorsal abdomen of adult drosophila (Ghosh et al., 2015; Dey et al., 2016). Frontiers in Cell and Developmental Biology \| www.frontiersin.org Mammals Many therapies currently make use of ex vivo expansion of autologous human HSCs, using a cytokine cocktail, but always with limited eﬃciency (Petzer et al., 1996; Cavazzana-Calvo et al., 2000; Schuster et al., 2012). Therefore a better understanding of the diﬀerent combinations of cytokines present in the niche along with the additional mechanisms and signaling pathways that normally expand HSCs is required to improve clinical treatment of a range of hematopoietic diseases. Here, we review the recent literature that describes the extrinsic signals important for HSC homing, expansion and ﬁnally release from the embryonic niche across zebraﬁsh, xenopus, chicken, and mammals. The anatomical sites where hematopoiesis occurs in these organisms are summarized in Table 1. We will then brieﬂy discuss the possible clinical implications of this current knowledge. In addition to cytokine secretion, a direct contact between the diﬀerent cells within the FL and hematopoietic progenitors is also important to maintain and expand HSCs (Nanno et al., 1994; Corlu et al., 1998). HSCs express a number of integrins and adhesion receptors that are critical for the correct traﬃcking of HSCs to the FL and could mediate cell contact. For example, VE- Cadherin (CD144), α2b-integrin (CD41), β1-integrin (CD29), cKIT and CXCR4 are well established traﬃcking molecules expressed by HSCs and play a key role in HSC guidance to the fetal niche (Mazo et al., 2011). Further studies have demonstrated that umbilical HSCs have a higher aﬃnity to adhere to adult BM than embryonic HSCs, which is due to a speciﬁc shift in the expression of speciﬁc integrins by HSCs. This suggests that integrin expression is March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 2 The Embryonic HSC Niche Mahony and Bertrand 32–60 hpf (Bertrand et al., 2010; Kissa and Herbomel, 2010; Mahony et al., 2018). Contrary to mammals, zebraﬁsh HSCs then migrate toward the vein, where they enter circulation to migrate to the CHT (Bertrand et al., 2010). Within the CHT they considerably expand between 3 and 4 days post fertilization (dpf) (Murayama et al., 2006; Tamplin et al., 2015; Mahony et al., 2016; Staal et al., 2016). HSCs migrate to the CHT embryonic niche in response to cxcl12a, expressed from stromal cells (Tamplin et al., 2015). Further zebraﬁsh studies have identiﬁed klf6a as an important transcription factor that directly regulates ccl25b expression in ECs (Xue et al., 2015, 2017). Xue et al. Chicken Deﬁnitive chicken hematopoiesis is initiated at E3-4 by the emergence of intra-aortic clusters of HSCs derived from endothelium (as seen in mammals) (Jaﬀredo et al., 2000; Bollerot et al., 2005a,b; Yvernogeau and Robin, 2017). HSCs then migrate to the neighboring mesenchyme, ventral to the aorta and located in the PAFs, that support the development of CD45+ cells (Cormier, 1993; Geerts et al., 1993), such as myeloerythroid progenitor cells and immature thymic precursors (that have not yet undergone T-cell receptor rearrangements) (Lampisuo et al., 1999; Jaﬀredo et al., 2000; Liippo et al., 2000; Saynajakangas et al., 2009). An additional site of embryonic hematopoiesis includes the yolk sac, which also contributes to the expansion and maturation of erythroid and myeloid cells (Guedes et al., 2014). However, the homing signals to the chicken PAFs remain unidentiﬁed. Although little is known about the microenvironment that would support HSCs in the chicken PAFs, diﬀerential expression of integrins may play an important role in supporting HSCs (Corbel, 2002). Stromal Cells In the mouse embryo, HSCs are closely associated with Nestin+ periportal stromal cells that express many HSC expansion factors, such as angptl2 (Khan et al., 2016). Many diﬀerent supportive stromal cell lines also have been derived from the mouse FL, such as AFT024, that support HSCs in vitro (Nolta et al., 2002), and the KM3 cell line, that supports human embryonic stem cells (Hu et al., 2012). The analysis of the AFT024 cell line revealed an enrichment in secreted factors such as insulin like growth factor, SCF, angiopoietin-3, Wnts and Ephrin2a that support HSCs (Charbord and Moore, 2005). NON-CELL-AUTONOMOUS MEDIATORS OF HSC EXPANSION IN THE EMBRYONIC NICHE The HSC pool ﬁrst undergoes expansion shortly after HSC emergence from the AGM (Taoudi et al., 2008; Rybtsov et al., 2016), before migrating to their fetal niche. The number of HSCs then greatly expands to around 38 times their original number, peaking at around E14 in mice and ceasing around 2–4 days postnatal (Morrison et al., 1995; Ema and Nakauchi, 2000; Baumann et al., 2004; Lessard et al., 2004; Chen et al., 2009; Payushina, 2012). Therefore, fully characterizing the diﬀerent cells and environmental cues that expand HSCs in diﬀerent organisms is required to improve the currently limited regenerative therapies. We will hereafter describe the diﬀerent elements of the microenvironment that contribute to this expansion, across the vertebrate phylum. Xenopus Fate-mapping and grafting experiments showed that bona ﬁde HSCs are generated in the dorsal lateral plate (DLP), the equivalent of the mammalian AGM (Turpen et al., 1981; Maeno et al., 1985; Ciau-Uitz et al., 2000; Clements and Traver, 2013). In larval stages, DLP-derived HSCs reach maturity and seed the FL where they produce erythrocytes that will replace embryonic primitive erythrocytes. The FL is the main site of HSC expansion and diﬀerentiation during embryogenesis, i.e., before metamorphosis (Chen and Turpen, 1995). Classical studies made use of kidney and liver sections from bullfrog tadpoles to reveal hematopoietic microenvironments, supporting red blood cell development (Broyles et al., 1981). After metamorphosis, the majority of the blood cells are DLP-derived (Ciau-Uitz et al., 2014). Mammals (2017) demonstrated that ccl25b is expressed in the CHT at 48hpf and is an important cytokine for HSC chemoattraction to and expansion within the CHT niche. These results were further corroborated by the ex vivo culture of murine HSCs in the presence of Ccl21 (murine ortholog of ccl25b) that enhanced HSC expansion through activation of its receptor, Ccr7 (Xue et al., 2017). Upon arrival in the CHT niche, VCAM+ macrophages are also required to direct HSCs (through binding to α4-integrin expressed by HSCs) toward venous capillaries and retain them in their embryonic niche (Li et al., 2018). required during development to mediate homing to speciﬁc niches (Roy and Verfaillie, 1999). Integrins (mainly α4-integrin) are implicated in mediating HSC interaction with the vascular niche in the BM (Mazo et al., 1998), and their inhibition mobilizes HSCs from the FL (Kim et al., 2016). Human endothelium- derived HSCs also express and use a myeloid adhesion factor, glycosylphosphatidylinositol-anchored surface protein (GPI-80; also known as Vanin-2, or VNN2) (Prashad et al., 2015), to facilitate their migration and expansion in the fetal niche. Several ECM, cell adhesion and cytoskeleton pathways are enriched in the AFT024 murine FL ﬁbroblast-derived stromal cell line [a cell line that supports HSC expansion in vitro (Nolta et al., 2002)] and within HSCs, permitting HSC migration and anchoring to their niches (Charbord and Moore, 2005). Endothelial Cells Both lpl and its cofactor apolipoprotein c2 (apoc2) controlled the release of an essential fatty acid, (Docosahexaenoic acid), which was identiﬁed as a novel HSC expansion factor (Liu et al., 2018). This study highlighted an additional possible pathway important for improving HSC expansion ex vivo. Direct physical contacts between HSCs and ECs have been described in the mouse embryo, similarly to the zebraﬁsh CHT (Tamplin et al., 2015), and may be mediated by E-selectin and VCAM1 (Schweitzer et al., 1996; Wittig et al., 2010). Murine FL-HSCs express the endothelial protein C receptor that can bind the activated protein C. This induces protease-activated receptor 1 signaling which inhibits apoptosis and maintains self- renewal activity (Iwasaki et al., 2010). In contrast to murine HSCs, human HSCs express endothelial protein C receptor in the FL, but lose its expression once they have migrated to the BM (Subramaniam et al., 2018). p The anterior lobe of the drosophila lymph gland consists of a medullary zone (MZ), a cortical zone (CZ) and the posterior signaling center (PSC). Prohemocytes are located in the MZ and give rise to mature hemocytes, plasmatocytes, crystal cells and rare lamellocytes, in response to immune challenge (Evans et al., 2003). These progeny cells will then colonize the CZ. The drosophila PSC is an important signaling niche that controls blood cell production and maturation and has been associated to mammalian stromal cells (Lebestky et al., 2003). The PSC is a source of Hedgehog signal that activates the zinc ﬁnger transcription factor Cubitus interruptus (Ci) (homolog of the vertebrate Gli proteins) in cells located in the MZ to maintain quiescence of blood progenitors and ﬁne tune diﬀerentiation upon immune challenge, independently of the EBF transcription factor Col within blood progenitors (Mandal et al., 2007; Benmimoun et al., 2015a,b; Pennetier et al., 2012). Further studies have suggested that the PSC may interact with nearby cells directly though thin processes that extend into the MZ (Mandal et al., 2007). Additional signals that aﬀect PSC signaling and HSC maintenance include Fetal liver ECs also support hematopoiesis ex vivo and promote HSC diﬀerentiation (Ohneda and Bautch, 1997; Wittig et al., 2010). FL ECs support hematopoiesis through the expression SCF/KitL that is normally membrane-bound. The conversion of this cytokine to its soluble form occurs under the control of MMP9, which transcription is regulated by Ezh2, a transcriptional repressor expressed by FL ECs. Endothelial Cells This process was only described in the context of erythropoiesis (Neo et al., 2018) but might be relevant for HSC expansion. These studies underscore the importance of ECs in expanding HSCs in the zebraﬁsh CHT and the FL in mammals. However, many other cell types support HSC expansion during embryogenesis. Endothelial Cells HSCs arrive in the zebraﬁsh CHT niche and progressively colonize this tissue from 48 hpf until 80 hpf (Tamplin et al., 2015). Although some HSCs are still present at 96 hpf, the majority have proliferated and left the CHT niche (Mahony et al., 2016). To accommodate HSCs within the CHT, the vascular niche is remodeled to improve stem cell seeding, a process that is controlled by cxcr1 (Blaser et al., 2017). Upon arrival to the CHT niche, HSCs trigger a “cuddling” behavior from caudal endothelial cells (cECs). This cuddling allows the cECs to maintain close proximity between an HSC and a stromal cell, which induces HSC proliferation (Tamplin et al., 2015). The cuddled HSC then undergoes cell division where either both daughter cells will leave the niche, one will leave and the one most proximal to the stromal cell will stay or both will stay. Strikingly, pharmacological stimulation of HSC niche engraftment leads to an overall increase in the number of adult stem cells, as shown by lineage tracing (Tamplin et al., 2015). This indicates the contribution of HSC embryonic niche engraftment to increasing the adult stem cell pool size, and highlights the importance of fully understanding the cytokines expressed by ECs that expand HSCs. p g p Recent studies have also focused on zebraﬁsh stromal cells, derived from the somites (Murayama et al., 2015). Isolated CHT zebraﬁsh stromal cells [caudal hematopoietic embryonic stromal tissue (CHEST)] from 3dpf embryos express a range of hematopoietic cytokines, some of which are not present in isolated kidney cells from adult ﬁsh (such as gcsfb, il11a, il11b, and fgf21). Furthermore CHEST cells were able to expand cultured HSCs and stimulate HSC diﬀerentiation in vitro (Wolf et al., 2017). The development of these zebraﬁsh stromal cell lines is an important step and represents a valuable tool to study hematopoiesis in this model. Indeed, by comparing the transcriptome of CHEST, ZKS (zebraﬁsh kidney stromal cells) and ZEST (zebraﬁsh embryonic stromal trunk cells), Berrun and colleagues highlighted the hematopoietic role of isthmin 1 (ism1), a secreted protein required for HSC development as well as erythro-myeloid diﬀerentiation (Berrun et al., 2018). In addition to hematopoietic cytokines, lipid metabolism is also important for HSC expansion and development during embryogenesis. This was recently demonstrated through the study of lipoprotein lipase (lpl), an enzyme expressed by stromal and/or ECs in the CHT, and required for fatty acid metabolism. Zebraﬁsh During zebraﬁsh development, gata2b (the earliest hemogenic endothelium marker) is expressed in ECs in the ﬂoor of the dorsal aorta (Butko et al., 2015). HSCs are then speciﬁed through the expression of runx1 and cmyb, and can be observed undergoing EHT from the ﬂoor of the dorsal aorta between A subtype of stromal cells, stellate cells, are fat storing hepatic sinusoid cells that appear around E10-11 in the mouse embryo and express a number of cytokines, ECM and adhesion molecules (Ramadori and Saile, 2002; Tan et al., 2017). Stellate cells are Desmin-positive and are found in close proximity to March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 3 Mahony and Bertrand The Embryonic HSC Niche HSCs (Kiassov et al., 1995). These cells express many diﬀerent supportive hematopoietic cytokines such as OSM, Csf1, THPO, EPO, Igf1, SCF/KitL and Cxcl12 (Fujio et al., 1994; Kubota et al., 2007; Tan et al., 2017). Stellate cells also express VCAM, ﬁbronectin1, vitronectin1, Lamb1-1 (Laminin-b1-1) and Lamc1 (Laminin-c1) (Kubota et al., 2007; Tan et al., 2017). In vitro, adult hepatic stellate cells can maintain HSCs in a similar manner to BM mesenchymal stem cells (Kordes et al., 2013). Stellate cells may therefore play an important role in maintaining the liver microenvironment and expanding the HSC pool. odorants that stimulate γ-aminobutyric acid release from the brain and drosophila insulin-like growth factor 2 by adipocytes (Yu et al., 2018). Mammalian Hepatocytes In the zebraﬁsh as well, we have shown the role of tfec, a mitf family zinc ﬁnger transcription factor (Lister et al., 2011), in HSC expansion. Tfec is highly expressed in cECs of the CHT and controls the expression of several CHT niche cytokines (such as kitlgb, thpo, csf1a, and csf3b) (Mahony et al., 2016). These cytokines are known to promote HSC proliferation and survival, resulting in their expansion. Consequently, in tfec mutants, HSCs fail to mature, resulting in severe anemia (Mahony et al., 2016). However, this phenotype can be rescued by the overexpression of kitlgb (Mahony et al., 2016). In addition, tfec also controls the expression of oncostatin M (osm) from cECs that can synergistically enhance the expansion of HSCs with kitlgb (Mahony et al., 2018). Tfec is therefore a critical transcription factor required for the proper role of the embryonic hematopoietic niche. This role of tfec has probably been conserved in mammals. Indeed, in rats, the comparison of liver sinusoidal ECs to lung microvascular ECs revealed a speciﬁc combination of transcription factors that controlled liver ECs function and identity (Geraud et al., 2010). Tfec was identiﬁed as a liver speciﬁc transcription factor, together with Gata4, Lmo3, and Maf. Whereas the roles of Tfec, Lmo3, and Maf have not yet been investigated during mouse fetal hematopoiesis, Gata4 seems to be required for ECs to function as an HSC niche (Geraud et al., 2017). Further studies demonstrated that Gata4 is required for correct EC maturation in the embryonic liver and correct HSPC colonization, as Gata4 allows the speciﬁcation of a discontinuous endothelium that allows HSC colonization. Therefore, when Gata4 was speciﬁcally deleted in liver sinusoidal ECs during development, the FL vasculature failed to develop normally and hematopoiesis was not supported by the FL (Geraud et al., 2017). The deletion of another transcription factor, activating transcription factor 4 (ATF4) impaired the ability of both endothelial and stromal cells to support FL HSC development. Further analysis revealed that Atf4 directly regulates the expression of Angptl3 to control the repopulating eﬃciency of HSCs (Zhao et al., 2015). Of note, Atf4 is also required at the cell-autonomous level for proper self-renewal of FL HSCs (Rieger, 2015). Hepatocytes contribute to most of the liver mass and therefore also play a role in the HSC niche. HSC Regulation by Their Hematopoietic Progeny Regulation of HSC expansion by their progeny has been described in the drosophila lymph gland. PDGF and VEGF- related factor- 1 (Pvf1) is produced by the PSC and activates March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 4 The Embryonic HSC Niche Mahony and Bertrand its receptor (PVR) on diﬀerentiated cells in the cortical zone. This activates JAK/STAT signaling and consequently controls the expression and secretion of adenosine deaminase Growth Factor-A (Adgf-A) from diﬀerentiated cells. This enzyme converts adenosine into inosine that promotes progenitor maintenance, through the PKA intracellular pathway (Mondal et al., 2011). Similarly in mammals, the placental microenvironment supports hematopoiesis through PDGF-B expression to overall inhibit HSC diﬀerentiation (Chhabra et al., 2012). Drosophila bip1 (bric-à-brac interacting protein 1) and Nucleoporin 98 (Nup98) are additional genes identiﬁed in controlling blood progenitor proliferation that control PVR expression (Mondal et al., 2014). The negative feedback of diﬀerentiated cells into HSCs has been evidenced in the adult mouse BM, but remains undocumented in the FL. diﬀerentiation in the MZ by regulating E-cadherin levels (Gao et al., 2014). Although ROS levels and hypoxia are well- described cues relevant to the adult BM niche, very little is known about their role during embryonic HSC expansion in any other models. Transcription Factors Controlling the HSC Niche Understanding the genetic network that controls the speciﬁcation and maintenance of the niche will allow building or reproducing these niches in vitro. Transcription factors control many aspects of the niche, from how the niche attracts HSCs to their nurturing and release. As mentioned earlier, klf6a is a transcription factor that controls the expression of ccl25b in cECs from the zebraﬁsh CHT. Therefore, in the absence of klf6a, HSCs cannot colonize the CHT and do not expand (Xue et al., 2017). Mammalian Hepatocytes The analysis of human FL samples has revealed the presence of E-selectin positive bipotent FL hepatoblasts, capable of diﬀerentiating into hepatic or biliary epithelial cells (Terrace et al., 2007). Murine hepatoblasts have been characterized by their expression of Protein delta homolog 1 (DLK-1) and a range of important hematopoietic cytokines, such as Epo, Thpo, Il6, SCF, and Flt3l. Hepatoblasts also express high levels of ECM molecules, including vitronectin, ﬁbronectin and tenascin C that are expressed under the control of TBGβ1 signaling (Sugiyama et al., 2013). The importance of hepatocytes in expanding HSCs was demonstrated by the co- culture of hepatocyte cell lines with FL cells, which resulted in a large increase in the number of HSCs (Aiuti et al., 1998). Their importance was further underscored by the analysis of Map2k4-mutant mouse embryos, lacking hepatoblasts, where hematopoiesis was strongly aﬀected following a decrease in cytokines expression in the FL, such as EPO and SCF. Accordingly, these embryos displayed a strong decrease in the number of HSCs (Sugiyama et al., 2011). Another study showed that DLK1-positive hepatocyte progenitors (or hepatoblasts) were the main contributor in the FL niche, as they were the main source of SCF, IGF-2, Cxcl12 as well as angptl2 and angptl3. Initially, these cells were identiﬁed as T cells as they were stained with the anti-CD3 antibody (Chou and Lodish, 2010). Ex vivo co-culture of these FL hepatoblasts with BM-HSCs leads to enhanced HSC long- term repopulation and ex vivo expansion of HSCs (Zhang and Lodish, 2004; Zhang et al., 2006). It was further suggested that Angptl2 could mediate its eﬀects through the LILRB2 receptor (Deng et al., 2014). Hypoxia and ROS Hypoxia is an important maturation signal in the microenvironment. In drosophila, low oxygen levels are sensed by Sima, the ortholog of hypoxia-inducible factor 1-α (Hif1-α) that induces crystal cell diﬀerentiation. Moderate ROS levels are required for proliferation of early progenitors. Fine-tuning levels of ROS is also required to balance progenitor March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 5 The Embryonic HSC Niche Mahony and Bertrand FIGURE 1 \| Summary of vertebrate HSC expansion in the embryonic niche. Following their derivation from aortic endothelium, HSCs home to their embryonic niche in response to several attractive cytokines, such as cxcl12 and ccl25b/Ccl21. HSCs are directed toward vascular cells by Mϕ (macrophages). The vascular cells in the embryonic niche then remodel to accommodate the arriving HSCs. HSCs then become lodged in the fetal niche and undergo cell division to expand their initial number. This expansion is in response to several cytokines released from many different cell types. Endothelial cells release kitlg and osm, under the control of tfec. Stromal cells release cxcl12, under the control of atf4. Hepatoblasts release Kit-ligand and angiopoietins. After considerable expansion, the ECM is remodeled by Mmp9 released from neutrophils and HSCs leave their fetal niche to migrate to their adult niche. Outlined here is an overview of the main cell types involved in fetal HSC expansion along with some examples of important cytokines/signals that they secrete, although many others exist. FIGURE 1 \| Summary of vertebrate HSC expansion in the embryonic niche. Following their derivation from aortic endothelium, HSCs home to their embryonic niche in response to several attractive cytokines, such as cxcl12 and ccl25b/Ccl21. HSCs are directed toward vascular cells by Mϕ (macrophages). The vascular cells in the embryonic niche then remodel to accommodate the arriving HSCs. HSCs then become lodged in the fetal niche and undergo cell division to expand their initial number. This expansion is in response to several cytokines released from many different cell types. Endothelial cells release kitlg and osm, under the control of tfec. Stromal cells release cxcl12, under the control of atf4. Hepatoblasts release Kit-ligand and angiopoietins. After considerable expansion, the ECM is remodeled by Mmp9 released from neutrophils and HSCs leave their fetal niche to migrate to their adult niche. Hypoxia and ROS Outlined here is an overview of the main cell types involved in fetal HSC expansion along with some examples of important cytokines/signals that they secrete, although many others exist. TABLE 2 \| Summary of the important cells and tissues required to mediate HSC expansion through evolution. TABLE 2 \| Summary of the important cells and tissues required to mediate HSC expansion through evolution. Drosophila (Lymph gland) Zebraﬁsh (CHT) Xenopus (FL) Chicken (PAF +YS) Mammals (FL) HSC Progeny Yes ? ? ? ? Stromal cells Yes Yes Yes Yes Yes Endothelial cells No Yes Yes Yes (YS) Yes Hepatocytes No No ? No Yes Nervous system ? ? ? ? ? CHT, caudal hematopoietic tissue; FL, fetal liver; PAF, para-aortic foci; YS, yolk sac. of the survival signal c-Abl in FL-HSCs, premature death and reduced BM colonization, highlighting a cell-intrinsic pathway that is required for FL-to-BM transition. The HSC migration from FL to the BM is mediated by diﬀerent cytokines, ECM signals and adhesion molecules (e.g., CXCL12, SCF, cadherins, integrins) (Ciriza et al., 2013). HSCs are then maintained in a quiescent state in BM throughout adulthood by many diﬀerent signals and cells (Boulais and Frenette, 2015). The mechanisms driving HSC release from the embryonic niche are similarly elusive in zebraﬁsh. However, live imaging has revealed signiﬁcant changes in the vascular architecture that normally accommodates HSCs. At 4 dpf, the number of HSCs present in the CHT decreases and they are no longer embedded in the vascular network, but they are instead found loosely associated to vascular cells (Tamplin et al., 2015; Mahony et al., 2016). Further studies have indicated that mmp9 expression in macrophages was required to modulate the accessibility of HSCs to cxcl12. Indeed, the inhibition of mmp9 resulted in the accumulation of HSCs in the CHT, which was rescued in cxcl12a- morphants (Theodore et al., 2017). Therefore the modulation of the vascular structure and the ECM by mmp9 is an important feature to control retention and release of HSCs from the embryonic niche. CLINICAL IMPLICATIONS The recent use of drug screens has identiﬁed several promising candidates to improve the clinical use of HSCs in regenerative medicine. For example, StemRegenin-1 (SR1), an antagonist of the aryl hydrocarbon receptor (AHR) enhances the ex vivo expansion of CD34+ cells and improves long-term engraftment in murine models (Boitano et al., 2010). Recently, early phase clinical trials have further highlighted the eﬀectiveness of this compound in improving the expansion of CD34+ cells from umbilical cord blood (Wagner et al., 2016). UM171, another compound, improves human cord blood cell expansion and engraftment, although the exact mechanism remains to be fully characterized (Fares et al., 2014). Zebraﬁsh high throughput screens have also highlighted the role of dmPGE2 that confers a competitive advantage to treated HSCs and has been successful in subsequent clinical trials (North et al., 2007; Cutler et al., 2013; Goessling and North, 2014). Identifying additional compounds and combinations of cytokines will improve HSC expansion for therapeutic use. Hematopoietic stem cell emergence and derivation from ECs is a highly conserved process across chick, zebraﬁsh, mouse and humans (Jaﬀredo et al., 2000; Bollerot et al., 2005a,b; Zovein et al., 2008; Chen et al., 2009; Bertrand et al., 2010; Boisset et al., 2010). Even drosophila blood cells are found in close association to the cardiac tube (Dey et al., 2016). Furthermore, the cell autonomous expression of transcription factors (such as runx1 and gata2) required to achieve EHT is highly conserved. This knowledge gained from the study of diﬀerent animal models has made it possible to generate inducible HSCs from ECs in vitro (Gomes et al., 2018). Across the species discussed in this review there are several conserved processes but also many diﬀerences in how HSCs expand in their fetal niche. Understanding the main similarities and diﬀerences is crucial to fully understanding HSC expansion. Here we will summarize the main conserved elements across drosophila, zebraﬁsh, mouse and human and how these have progressed throughout evolution. The importance in further understanding the niche has been further underscored as it was shown that niche dysfunctions (for example, induced by mesenchymal infections) could lead to genotoxic stress and might sensitize HSCs to leukemia development (Zambetti et al., 2016; Passaro et al., 2017). It remains to be identiﬁed if aberrant cytokine signaling in the fetal niche can sensitize HSCs to diﬀerent disorders. HSC RELEASE FROM THE EMBRYONIC NICHE The cellular processes governing mammalian HSC release from the embryonic niche remain to be fully characterized. However, a recent study has revealed that the FL-to-BM transition required correct formation of the Wiskott–Aldrich syndrome verprolin- homologous (WAVE) protein complex 2, mediated by Hem-1 (Shao et al., 2018). The disruption of this complex leads to a loss March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 6 The Embryonic HSC Niche Mahony and Bertrand FUNDING JYB is endorsed by a Chair in Life Sciences funded by the Gabriella Giorgi-Cavaglieri Foundation and is also funded by the Swiss National Fund (31003_166515). CONCLUSION Understanding the complete genetic and molecular program that controls HSC expansion in the embryonic niche remains an important goal to improve current protocols of regenerative medicine. The use of many animal models across phylogeny will concur to this aim, as most of the mechanisms involved in the control of HSC expansion by the embryonic niche appear to be conserved through evolution (Figure 1 and Table 2). Understanding the complete genetic and molecular program that controls HSC expansion in the embryonic niche remains an important goal to improve current protocols of regenerative medicine. The use of many animal models across phylogeny will concur to this aim, as most of the mechanisms involved in the control of HSC expansion by the embryonic niche appear to be conserved through evolution (Figure 1 and Table 2). From zebraﬁsh to mammals, HSCs expand in an increasingly more complex niche: whereas the CHT in teleosts is a transient vascularized tissue (Tamplin et al., 2015), HSCs will colonize the FL in mammals, a bona ﬁde organ. With this, comes the addition of a new cellular layer to the HSC niche, i.e., the hepatocytes. As stromal and ECs, hepatoblasts will secrete similar survival and proliferative signals, such as SCF and other cytokines (Sugiyama et al., 2011, 2013) (Figure 1 and Table 2). It is interesting to note that even if the structures have changed, the genetic network seems to have been conserved. One such example is the transcription factor Tfec that is speciﬁcally expressed in the zebraﬁsh CHT vasculature, as well as in sinusoidal ECs of the fetal and adult liver in rodents (Geraud et al., 2010; Mahony et al., 2016). CLINICAL IMPLICATIONS The drosophila PSC consists of stromal cells in the lymph gland, and controls blood cell production and maturation (Lebestky et al., 2003). These cells are responsible for the signaling of a number of extrinsic factors to control HSC expansion (Figure 1 and Table 2). Throughout evolution, the structure of the embryonic niche has greatly changed. The vertebrate phylum coincides with the emergence of a closed circulatory system, dependent on vasculature. In the zebraﬁsh, the role of vasculature becomes very important for HSC expansion (Tamplin et al., 2015) and remains so in higher vertebrate species (Table 2). In avian embryos, HSCs expand in the yolk sac, a highly vascularized structure (Guedes et al., 2014), and mouse HSCs are also closely associated to the growing vasculature in the FL that support HSCs (Ohneda and Bautch, 1997; Wittig et al., 2010; Tamplin et al., 2015) (Figure 1 and Table 2). AUTHOR CONTRIBUTIONS CBM wrote the manuscript. JYB edited the manuscript. Aiuti, A., Cicchini, C., Bernardini, S., Fedele, G., Amicone, L., Fantoni, A., et al. (1998). Hematopoietic support and cytokine expression of murine-stable with Wiskott-Aldrich syndrome. Science 341:1233151. doi: 10.1126/science.123 3151 with Wiskott-Aldrich syndrome. Science 341:1233151. doi: 10.1126/science.123 3151 Aiuti, A., Cicchini, C., Bernardini, S., Fedele, G., Amicone, L., Fantoni, A., et al. (1998). Hematopoietic support and cytokine expression of murine-stable REFERENCES doi: 10.1016/j.exphem.2013.01.009 Berrun, A., Harris, E., and Stachura, D. L. (2018). Isthmin 1 (ism1) is required for normal hematopoiesis in developing zebraﬁsh. PLoS One 13:e0196872. doi: 10.1371/journal.pone.0196872 Clements, W. K., and Traver, D. (2013). Signalling pathways that control vertebrate haematopoietic stem cell speciﬁcation. Nat. Rev. Immunol. 13, 336–348. doi: 10.1038/nri3443 Bertrand, J. Y., Chi, N. C., Santoso, B., Teng, S., Stainier, D. Y., and Traver, D. (2010). Haematopoietic stem cells derive directly from aortic endothelium during development. Nature 464, 108–111. doi: 10.1038/nature08738 Corbel, C. (2002). Expression of alphaVbeta3 integrin in the chick embryo aortic endothelium. Int. J. Dev. Biol. 46, 827–830. Bertrand, J. Y., Jalil, A., Klaine, M., Jung, S., Cumano, A., and Godin, I. (2005). Three pathways to mature macrophages in the early mouse yolk sac. Blood 106, 3004–3011. doi: 10.1182/blood-2005-02-0461 Corlu, A., Lamy, I., Ilyin, G. P., Fardel, O., Kneip, B., Le Jossic, C., et al. (1998). Hematopoiesis-promoting activity of rat liver biliary epithelial cells: involvement of a cell surface molecule, liver-regulating protein. Exp. Hematol. 26, 382–394. Bertrand, J. Y., Kim, A. D., Violette, E. P., Stachura, D. L., Cisson, J. L., and Traver, D. (2007). Deﬁnitive hematopoiesis initiates through a committed erythromyeloid progenitor in the zebraﬁsh embryo. Development 134, 4147– 4156. doi: 10.1242/dev.012385 Cormier, F. (1993). Avian pluripotent haemopoietic progenitor cells: detection and enrichment from the para-aortic region of the early embryo. J. Cell Sci. 105( Pt 3), 661–666. Blaser, B. W., Moore, J. L., Hagedorn, E. J., Li, B., Riquelme, R., Lichtig, A., et al. (2017). CXCR1 remodels the vascular niche to promote hematopoietic stem and progenitor cell engraftment. J. Exp. Med. 214, 1011–1027. doi: 10.1084/jem. 20161616 Cutler, C., Multani, P., Robbins, D., Kim, H. T., Le, T., Hoggatt, J., et al. (2013). Prostaglandin-modulated umbilical cord blood hematopoietic stem cell transplantation. Blood 122, 3074–3081. doi: 10.1182/blood-2013-05-503177 Boisset, J. C., van Cappellen, W., Andrieu-Soler, C., Galjart, N., Dzierzak, E., and Robin, C. (2010). In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium. Nature 464, 116–120. doi: 10.1038/nature 08764 de Bruijn, M. F., Speck, N. A., Peeters, M. C., and Dzierzak, E. (2000). Deﬁnitive hematopoietic stem cells ﬁrst develop within the major arterial regions of the mouse embryo. EMBO J. 19, 2465–2474. doi: 10.1093/emboj/19.11.2465 Deng, M., Lu, Z., Zheng, J., Wan, X., Chen, X., Hirayasu, K., et al. (2014). A motif in LILRB2 critical for Angptl2 binding and activation. Blood 124, 924–935. REFERENCES with Wiskott-Aldrich syndrome. Science 341:1233151. doi: 10.1126/science.123 3151 Aiuti, A., Biasco, L., Scaramuzza, S., Ferrua, F., Cicalese, M. P., Baricordi, C., et al. (2013). Lentiviral hematopoietic stem cell gene therapy in patients Aiuti, A., Cicchini, C., Bernardini, S., Fedele, G., Amicone, L., Fantoni, A., et al. (1998). Hematopoietic support and cytokine expression of murine-stable March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org Mahony and Bertrand The Embryonic HSC Niche hepatocyte cell lines (MMH). Hepatology 28, 1645–1654. doi: 10.1002/hep. 510280626 Chhabra, A., Lechner, A. J., Ueno, M., Acharya, A., Van Handel, B., Wang, Y., et al. (2012). Trophoblasts regulate the placental hematopoietic niche through PDGF-B signaling. Dev. Cell 22, 651–659. doi: 10.1016/j.devcel.2011.12.022 hepatocyte cell lines (MMH). Hepatology 28, 1645–1654. doi: 10.1002/hep. 510280626 Ara, T., Tokoyoda, K., Sugiyama, T., Egawa, T., Kawabata, K., and Nagasawa, T. (2003). Long-term hematopoietic stem cells require stromal cell-derived factor- 1 for colonizing bone marrow during ontogeny. Immunity 19, 257–267. doi: 10.1016/S1074-7613(03)00201-2 Chou, S., and Lodish, H. F. (2010). Fetal liver hepatic progenitors are supportive stromal cells for hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 107, 7799–7804. doi: 10.1073/pnas.1003586107 Christensen, J. L., Wright, D. E., Wagers, A. J., and Weissman, I. L. (2004). Circulation and chemotaxis of fetal hematopoietic stem cells. PLoS Biol. 2:E75. doi: 10.1371/journal.pbio.0020075 Baumann, C. I., Bailey, A. S., Li, W., Ferkowicz, M. J., Yoder, M. C., and Fleming, W. H. (2004). PECAM-1 is expressed on hematopoietic stem cells throughout ontogeny and identiﬁes a population of erythroid progenitors. Blood 104, 1010–1016. doi: 10.1182/blood-2004-03-0989 Ciau-Uitz, A., Monteiro, R., Kirmizitas, A., and Patient, R. (2014). Developmental hematopoiesis: ontogeny, genetic programming and conservation. Exp. Hematol. 42, 669–683. doi: 10.1016/j.exphem.2014.06.001 Benmimoun, B., Haenlin, M., and Waltzer, L. (2015a). Haematopoietic progenitor maintenance by EBF/Collier: beyond the Niche. Cell Cycle 14, 3517–3518. doi: 10.1080/15384101.2015.1093449 Ciau-Uitz, A., Walmsley, M., and Patient, R. (2000). Distinct origins of adult and embryonic blood in Xenopus. Cell 102, 787–796. doi: 10.1016/S0092-8674(00) 00067-2 Benmimoun, B., Polesello, C., Haenlin, M., and Waltzer, L. (2015b). The EBF transcription factor Collier directly promotes Drosophila blood cell progenitor maintenance independently of the niche. Proc. Natl. Acad. Sci. U.S.A. 112, 9052–9057. doi: 10.1073/pnas.1423967112 Ciriza, J., Thompson, H., Petrosian, R., Manilay, J. O., and Garcia-Ojeda, M. E. (2013). The migration of hematopoietic progenitors from the fetal liver to the fetal bone marrow: lessons learned and possible clinical applications. Exp. Hematol. 41, 411–423. REFERENCES doi: 10.1182/blood-2014-01-549162 Boitano, A. E., Wang, J., Romeo, R., Bouchez, L. C., Parker, A. E., Sutton, S. E., et al. (2010). Aryl hydrocarbon receptor antagonists promote the expansion of human hematopoietic stem cells. Science 329, 1345–1348. doi: 10.1126/science. 1191536 Dey, N. S., Ramesh, P., Chugh, M., Mandal, S., and Mandal, L. (2016). Dpp dependent Hematopoietic stem cells give rise to Hh dependent blood progenitors in larval lymph gland of Drosophila. Elife 5:e18295. doi: 10.7554/ eLife.18295 Bollerot, K., Pouget, C., and Jaﬀredo, T. (2005a). The embryonic origins of hematopoietic stem cells: a tale of hemangioblast and hemogenic endothelium. APMIS 113, 790–803. Dunon, D., and Imhof, B. A. (2000). The role of cell traﬃc in the emergence of the T lymphoid system. Semin. Immunol. 12, 429–433. doi: 10.1006/smim.2000.0266 Bollerot, K., Romero, S., Dunon, D., and Jaﬀredo, T. (2005b). Core binding factor in the early avian embryo: cloning of Cbfbeta and combinatorial expression patterns with Runx1. Gene Expr. Patterns 6, 29–39. Ema, H., and Nakauchi, H. (2000). Expansion of hematopoietic stem cells in the developing liver of a mouse embryo. Blood 95, 2284–2288. Boulais, P. E., and Frenette, P. S. (2015). Making sense of hematopoietic stem cell niches. Blood 125, 2621–2629. doi: 10.1182/blood-2014-09-570192 Espin-Palazon, R., Stachura, D. L., Campbell, C. A., Garcia-Moreno, D., Del Cid, N., Kim, A. D., et al. (2014). Proinﬂammatory signaling regulates hematopoietic stem cell emergence. Cell 159, 1070–1085. doi: 10.1016/j.cell. 2014.10.031 Broyles, R. H., Johnson, G. M., Maples, P. B., and Kindell, G. R. (1981). Two erythropietic microenvironments and two larval red cell lines in bullfrog tadpoles. Dev. Biol. 81, 299–314. doi: 10.1016/0012-1606(81)90293-1 Evans, C. J., Hartenstein, V., and Banerjee, U. (2003). Thicker than blood: conserved mechanisms in Drosophila and vertebrate hematopoiesis. Dev. Cell 5, 673–690. doi: 10.1016/S1534-5807(03)00335-6 Butko, E., Distel, M., Pouget, C., Weijts, B., Kobayashi, I., Ng, K., et al. (2015). Gata2b is a restricted early regulator of hemogenic endothelium in the zebraﬁsh embryo. Development 142, 1050–1061. doi: 10.1242/dev.119180 Fares, I., Chagraoui, J., Gareau, Y., Gingras, S., Ruel, R., Mayotte, N., et al. (2014). Cord blood expansion. Pyrimidoindole derivatives are agonists of human hematopoietic stem cell self-renewal. Science 345, 1509–1512. doi: 10.1126/ science.1256337 Cavazzana-Calvo, M., Hacein-Bey, S., de Saint Basile, G., Gross, F., Yvon, E., Nusbaum, P., et al. (2000). Gene therapy of human severe combined immunodeﬁciency (SCID)-X1 disease. Science 288, 669–672. doi: 10.1126/ science.288.5466.669 Fujio, K., Evarts, R. P., Hu, Z., Marsden, E. REFERENCES R., and Thorgeirsson, S. S. (1994). Expression of stem cell factor and its receptor, c-kit, during liver regeneration from putative stem cells in adult rat. Lab. Invest. 70, 511–516. Charbord, P., and Moore, K. (2005). Gene expression in stem cell-supporting stromal cell lines. Ann. N. Y. Acad. Sci. 1044, 159–167. doi: 10.1196/annals. 1349.020 Gao, H., Wu, X., Simon, L., and Fossett, N. (2014). Antioxidants maintain E-cadherin levels to limit Drosophila prohemocyte diﬀerentiation. PLoS One 9:e107768. doi: 10.1371/journal.pone.0107768 Chen, M. J., Yokomizo, T., Zeigler, B. M., Dzierzak, E., and Speck, N. A. (2009). Runx1 is required for the endothelial to haematopoietic cell transition but not thereafter. Nature 457, 887–891. doi: 10.1038/nature07619 Geerts, W. J., Lamers, W. H., and Moorman, A. F. (1993). Diﬀerences in erythropoiesis in normal chicken and quail embryos. Histochem. J. 25, 280–290. doi: 10.1007/BF00159119 Chen, X. D., and Turpen, J. B. (1995). Intraembryonic origin of hepatic hematopoiesis in Xenopus laevis. J. Immunol. 154, 2557–2567. March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 8 Mahony and Bertrand The Embryonic HSC Niche Gekas, C., Dieterlen-Lievre, F., Orkin, S. H., and Mikkola, H. K. (2005). The placenta is a niche for hematopoietic stem cells. Dev. Cell 8, 365–375. doi: 10.1016/j.devcel.2004.12.016 Lampisuo, M., Liippo, J., Vainio, O., McNagny, K. M., Kulmala, J., and Lassila, O. (1999). Characterization of prethymic progenitors within the chicken embryo. Int. Immunol. 11, 63–69. doi: 10.1093/intimm/11.1.63 Gekas, C., Rhodes, K. E., Van Handel, B., Chhabra, A., Ueno, M., and Mikkola, H. K. (2010). Hematopoietic stem cell development in the placenta. Int. J. Dev. Biol. 54, 1089–1098. doi: 10.1387/ijdb.103070cg Lebestky, T., Chang, T., Hartenstein, V., and Banerjee, U. (2000). Speciﬁcation of Drosophila hematopoietic lineage by conserved transcription factors. Science 288, 146–149. doi: 10.1126/science.288.5463.146 Lebestky, T., Jung, S. H., and Banerjee, U. (2003). A Serrate-expressing signaling center controls Drosophila hematopoiesis. Genes Dev. 17, 348–353. doi: 10. 1101/gad.1052803 Geraud, C., Koch, P. S., Zierow, J., Klapproth, K., Busch, K., Olsavszky, V., et al. (2017). GATA4-dependent organ-speciﬁc endothelial diﬀerentiation controls liver development and embryonic hematopoiesis. J. Clin. Invest. 127, 1099– 1114. doi: 10.1172/JCI90086 Lessard, J., Faubert, A., and Sauvageau, G. (2004). Genetic programs regulating HSC speciﬁcation, maintenance and expansion. Oncogene 23, 7199–7209. doi: 10.1038/sj.onc.1207940 Geraud, C., Schledzewski, K., Demory, A., Klein, D., Kaus, M., Peyre, F., et al. (2010). REFERENCES tfec controls the hematopoietic stem cell vascular niche during zebraﬁsh embryogenesis. Blood 128, 1336–1345. doi: 10.1182/blood-2016-04-710137 Mahony, C. B., Pasche, C., and Bertrand, J. Y. (2018). Oncostatin M and kit-ligand control hematopoietic stem cell fate during zebraﬁsh embryogenesis. Stem Cell Rep. 10, 1920–1934. doi: 10.1016/j.stemcr.2018.04.016 Hu, J., Hu, S., Ma, Q., Wang, X., Zhou, Z., Zhang, W., et al. (2012). Immortalized mouse fetal liver stromal cells support growth and maintenance of human embryonic stem cells. Oncol. Rep. 28, 1385–1391. doi: 10.3892/or.2012.1909 Mandal, L., Martinez-Agosto, J. A., Evans, C. J., Hartenstein, V., and Banerjee, U. (2007). A Hedgehog- and Antennapedia-dependent niche maintains Drosophila haematopoietic precursors. Nature 446, 320–324. doi: 10.1038/nature05585 Ivanovs, A., Rybtsov, S., Welch, L., Anderson, R. A., Turner, M. L., and Medvinsky, A. (2011). Highly potent human hematopoietic stem cells ﬁrst emerge in the intraembryonic aorta-gonad-mesonephros region. J. Exp. Med. 208, 2417–2427. doi: 10.1084/jem.20111688 Mazo, I. B., Gutierrez-Ramos, J. C., Frenette, P. S., Hynes, R. O., Wagner, D. D., and von Andrian, U. H. (1998). Hematopoietic progenitor cell rolling in bone marrow microvessels: parallel contributions by endothelial selectins and vascular cell adhesion molecule 1. J. Exp. Med. 188, 465–474. doi: 10.1084/jem. 188.3.465 Iwasaki, H., Arai, F., Kubota, Y., Dahl, M., and Suda, T. (2010). Endothelial protein C receptor-expressing hematopoietic stem cells reside in the perisinusoidal niche in fetal liver. Blood 116, 544–553. doi: 10.1182/blood-2009-08-240903 Jaﬀredo, T., Gautier, R., Brajeul, V., and Dieterlen-Lievre, F. (2000). Tracing the progeny of the aortic hemangioblast in the avian embryo. Dev. Biol. 224, 204–214. doi: 10.1006/dbio.2000.9799 Mazo, I. B., Massberg, S., and von Andrian, U. H. (2011). Hematopoietic stem and progenitor cell traﬃcking. Trends Immunol. 32, 493–503. doi: 10.1016/j.it.2011. 06.011 Khan, J. A., Mendelson, A., Kunisaki, Y., Birbrair, A., Kou, Y., Arnal-Estape, A., et al. (2016). Fetal liver hematopoietic stem cell niches associate with portal vessels. Science 351, 176–180. doi: 10.1126/science.aad0084 McGrath, K. E., Frame, J. M., Fegan, K. H., Bowen, J. R., Conway, S. J., Catherman, S. C., et al. (2015). Distinct sources of hematopoietic progenitors emerge before HSCs and provide functional blood cells in the mammalian embryo. Cell Rep 11, 1892–1904. doi: 10.1016/j.celrep.2015.05.036 Kiassov, A. P., Van Eyken, P., van Pelt, J. F., Depla, E., Fevery, J., Desmet, V. J., et al. (1995). Desmin expressing nonhematopoietic liver cells during rat liver development: an immunohistochemical and morphometric study. Diﬀerentiation 59, 253–258. doi: 10.1046/j.1432-0436.1995.5940253.x Mondal, B. C., Mukherjee, T., Mandal, L., Evans, C. REFERENCES Liver sinusoidal endothelium: a microenvironment-dependent diﬀerentiation program in rat including the novel junctional protein liver endothelial diﬀerentiation-associated protein-1. Hepatology 52, 313–326. doi: 10.1002/hep.23618 Li, D., Xue, W., Li, M., Dong, M., Wang, J., Wang, X., et al. (2018). VCAM- 1(+) macrophages guide the homing of HSPCs to a vascular niche. Nature 564, 119–124. doi: 10.1038/s41586-018-0709-7 Liippo, J., Koskela, K., and Lassila, O. (2000). Prethymic progenitors from the avian para-aortic mesoderm express GATA-3 and distinct chTcf isoforms but still lack T-cell receptor-gamma rearrangements. Scand. J. Immunol. 52, 502–509. doi: 10.1046/j.1365-3083.2000.00807.x Ghosh, S., Singh, A., Mandal, S., and Mandal, L. (2015). Active hematopoietic hubs in Drosophila adults generate hemocytes and contribute to immune response. Dev. Cell 33, 478–488. doi: 10.1016/j.devcel.2015.03.014 j Goessling, W., and North, T. E. (2014). Repairing quite swimmingly: advances in regenerative medicine using zebraﬁsh. Dis. Model. Mech. 7, 769–776. doi: 10.1242/dmm.016352 Lister, J. A., Lane, B. M., Nguyen, A., and Lunney, K. (2011). Embryonic expression of zebraﬁsh MiT family genes tfe3b, tfeb, and tfec. Dev. Dyn. 240, 2529–2538. doi: 10.1002/dvdy.22743 Gomes, A. M., Kurochkin, I., Chang, B., Daniel, M., Law, K., Satija, N., et al. (2018). Cooperative transcription factor induction mediates hemogenic reprogramming. Cell Rep. 25, 2821–2835.e7. doi: 10.1016/j.celrep.2018.11.032 Liu, C., Han, T., Stachura, D. L., Wang, H., Vaisman, B. L., Kim, J., et al. (2018). Lipoprotein lipase regulates hematopoietic stem progenitor cell maintenance Liu, C., Han, T., Stachura, D. L., Wang, H., Vaisman, B. L., Kim, J., et al. (2018). Lipoprotein lipase regulates hematopoietic stem progenitor cell maintenance through DHA supply. Nat. Commun. 9:1310. doi: 10.1038/s41467-018-03775-y p p p g p p g through DHA supply. Nat. Commun. 9:1310. doi: 10.1038/s41467-018-03775-y Greenbaum, A., Hsu, Y. M., Day, R. B., Schuettpelz, L. G., Christopher, M. J., Borgerding, J. N., et al. (2013). CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature 495, 227–230. doi: 10.1038/nature11926 Maeno, M., Tochinai, S., and Katagiri, C. (1985). Diﬀerential participation of ventral and dorsolateral mesoderms in the hemopoiesis of Xenopus, as revealed in diploid-triploid or interspeciﬁc chimeras. Dev. Biol. 110, 503–508. doi: 10. 1016/0012-1606(85)90108-3 Guedes, P. T., de Oliveira, B. C., Manso, P. P., Caputo, L. F., Cotta-Pereira, G., and Pelajo-Machado, M. (2014). Histological analyses demonstrate the temporary contribution of yolk sac, liver, and bone marrow to hematopoiesis during chicken development. PLoS One 9:e90975. doi: 10.1371/journal.pone.009 0975 Mahony, C. B., Fish, R. J., Pasche, C., and Bertrand, J. Y. (2016). REFERENCES J., Sinenko, S. A., Martinez- Agosto, J. A., et al. (2011). Interaction between diﬀerentiating cell- and niche- derived signals in hematopoietic progenitor maintenance. Cell 147, 1589–1600. doi: 10.1016/j.cell.2011.11.041 Kim, A. G., Vrecenak, J. D., Boelig, M. M., Eissenberg, L., Rettig, M. P., Riley, J. S., et al. (2016). Enhanced in utero allogeneic engraftment in mice after mobilizing fetal HSCs by alpha4beta1/7 inhibition. Blood 128, 2457–2461. doi: 10.1182/blood-2016-06-723981 Mondal, B. C., Shim, J., Evans, C. J., and Banerjee, U. (2014). Pvr expression regulators in equilibrium signal control and maintenance of Drosophila blood progenitors. Elife 3:e03626. doi: 10.7554/eLife.03626 Kissa, K., and Herbomel, P. (2010). Blood stem cells emerge from aortic endothelium by a novel type of cell transition. Nature 464, 112–115. doi: 10. 1038/nature08761 Morrison, S. J., Hemmati, H. D., Wandycz, A. M., and Weissman, I. L. (1995). The puriﬁcation and characterization of fetal liver hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 92, 10302–10306. doi: 10.1073/pnas.92.22.10302 Kordes, C., Sawitza, I., Gotze, S., and Haussinger, D. (2013). Hepatic stellate cells support hematopoiesis and are liver-resident mesenchymal stem cells. Cell. Physiol. Biochem. 31, 290–304. doi: 10.1159/000343368 Murayama, E., Kissa, K., Zapata, A., Mordelet, E., Briolat, V., Lin, H. F., et al. (2006). Tracing hematopoietic precursor migration to successive hematopoietic organs during zebraﬁsh development. Immunity 25, 963–975. doi: 10.1016/j. immuni.2006.10.015 Kubota, H., Yao, H. L., and Reid, L. M. (2007). Identiﬁcation and characterization of vitamin A-storing cells in fetal liver: implications for functional importance of hepatic stellate cells in liver development and hematopoiesis. Stem Cells 25, 2339–2349. doi: 10.1634/stemcells.2006-0316 Murayama, E., Sarris, M., Redd, M., Le Guyader, D., Vivier, C., Horsley, W., et al. (2015). NACA deﬁciency reveals the crucial role of somite-derived stromal March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 9 Mahony and Bertrand The Embryonic HSC Niche cells in haematopoietic niche formation. Nat. Commun. 6:8375. doi: 10.1038/ ncomms9375 cells in haematopoietic niche formation. Nat. Commun. 6:8375. doi: 10.1038/ ncomms9375 Sawitza, I., Kordes, C., Reister, S., and Haussinger, D. (2009). The niche of stellate cells within rat liver. Hepatology 50, 1617–1624. doi: 10.1002/hep.23184 p gy p Saynajakangas, R., Uchida, T., and Vainio, O. (2009). Diﬀerential gene expression Saynajakangas, R., Uchida, T., and Vainio, O. (2009). Diﬀerential gene expression in CD45 cells at para-aortic foci stage of chicken haematopoiesis. Scand. J. Immunol. 70, 288–294. REFERENCES doi: 10.1111/j.1365-3083.2009.02304.x Nagasawa, T., Hirota, S., Tachibana, K., Takakura, N., Nishikawa, S., Kitamura, Y., et al. (1996). Defects of B-cell lymphopoiesis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSF/SDF-1. Nature 382, 635–638. doi: 10.1038/382635a0 in CD45 cells at para-aortic foci stage of chicken haematopoiesis. Scand. J. Immunol. 70, 288–294. doi: 10.1111/j.1365-3083.2009.02304.x Schuster, J. A., Stupnikov, M. R., Ma, G., Liao, W., Lai, R., Ma, Y., et al. (2012). Expansion of hematopoietic stem cells for transplantation: current perspectives. Exp. Hematol. Oncol. 1:12. doi: 10.1186/2162-3619-1-12 Nanno, M., Hata, M., Doi, H., Satomi, S., Yagi, H., Sakata, T., et al. (1994). Stimulation of in vitro hematopoiesis by a murine fetal hepatocyte clone through cell-cell contact. J. Cell. Physiol. 160, 445–454. doi: 10.1002/jcp. 1041600307 Schweitzer, K. M., Drager, A. M., van der Valk, P., Thijsen, S. F., Zevenbergen, A., Theijsmeijer, A. P., et al. (1996). Constitutive expression of E-selectin and vascular cell adhesion molecule-1 on endothelial cells of hematopoietic tissues. Am. J. Pathol. 148, 165–175. Neo, W. H., Booth, C. A. G., Azzoni, E., Chi, L., Delgado-Olguin, P., de Bruijn, M., et al. (2018). Cell-extrinsic hematopoietic impact of Ezh2 inactivation in fetal liver endothelial cells. Blood 131, 2223–2234. doi: 10.1182/blood-2017-10- 811455 Shao, L., Chang, J., Feng, W., Wang, X., Williamson, E. A., Li, Y., et al. (2018). The Wave2 scaﬀold Hem-1 is required for transition of fetal liver hematopoiesis to bone marrow. Nat Commun 9, 2377. doi: 10.1038/s41467-018-04716-5 Nolta, J. A., Thiemann, F. T., Arakawa-Hoyt, J., Dao, M. A., Barsky, L. W., Moore, K. A., et al. (2002). The AFT024 stromal cell line supports long- term ex vivo maintenance of engrafting multipotent human hematopoietic progenitors. Leukemia 16, 352–361. doi: 10.1038/sj.leu.2402371 Staal, F. J., Spaink, H. P., and Fibbe, W. E. (2016). Visualizing human hematopoietic stem cell traﬃcking in vivo using a zebraﬁsh xenograft model. Stem Cells Dev. 25, 360–365. doi: 10.1089/scd.2015.0195 Subramaniam, A., Safaee Talkhoncheh, M., Magnusson, M., and Larsson, J. (2018). Endothelial protein C receptor (EPCR) expression marks human fetal liver hematopoietic stem cells. Haematologica 104, e47–e50. doi: 10.3324/haematol. 2018.198515 North, T. E., Goessling, W., Walkley, C. R., Lengerke, C., Kopani, K. R., Lord, A. M., et al. (2007). Prostaglandin E2 regulates vertebrate haematopoietic stem cell homeostasis. Nature 447, 1007–1011. doi: 10.1038/nature05883 Ohneda, O., and Bautch, V. L. (1997). Murine endothelial cells support fetal liver erythropoiesis and myelopoiesis via distinct interactions. Br. J. Haematol. 98, 798–808. REFERENCES doi: 10.1046/j.1365-2141.1997.3163133.x Sugiyama, D., Kulkeaw, K., and Mizuochi, C. (2013). TGF-beta-1 up-regulates extra-cellular matrix production in mouse hepatoblasts. Mech. Dev. 130, 195– 206. doi: 10.1016/j.mod.2012.09.003 Ottersbach, K., and Dzierzak, E. (2005). The murine placenta contains hematopoietic stem cells within the vascular labyrinth region. Dev. Cell 8, 377–387. doi: 10.1016/j.devcel.2005.02.001 Sugiyama, D., Kulkeaw, K., Mizuochi, C., Horio, Y., and Okayama, S. (2011). Hepatoblasts comprise a niche for fetal liver erythropoiesis through cytokine production. Biochem. Biophys. Res. Commun. 410, 301–306. doi: 10.1016/j.bbrc. 2011.05.137 Palis, J. (2014). Primitive and deﬁnitive erythropoiesis in mammals. Front. Physiol. 5:3. doi: 10.3389/fphys.2014.00003 Sugiyama, T., Kohara, H., Noda, M., and Nagasawa, T. (2006). Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity 25, 977–988. doi: 10.1016/j.immuni. 2006.10.016 Palis, J., Chan, R. J., Koniski, A., Patel, R., Starr, M., and Yoder, M. C. (2001). Spatial and temporal emergence of high proliferative potential hematopoietic precursors during murine embryogenesis. Proc. Natl. Acad. Sci. U.S.A. 98, 4528–4533. doi: 10.1073/pnas.071002398 Tamplin, O. J., Durand, E. M., Carr, L. A., Childs, S. J., Hagedorn, E. J., Li, P., et al. (2015). Hematopoietic stem cell arrival triggers dynamic remodeling of the perivascular niche. Cell 160, 241–252. doi: 10.1016/j.cell.2014.12.032 Passaro, D., Di Tullio, A., Abarrategi, A., Rouault-Pierre, K., Foster, K., Ariza- McNaughton, L., et al. (2017). Increased vascular permeability in the bone marrow microenvironment contributes to disease progression and drug response in acute myeloid leukemia. Cancer Cell 32, 324–341.e6. doi: 10.1016/j. ccell.2017.08.001 Tan, K. S., Kulkeaw, K., Nakanishi, Y., and Sugiyama, D. (2017). Expression of cytokine and extracellular matrix mRNAs in fetal hepatic stellate cells. Genes Cells 22, 836–844. doi: 10.1111/gtc.12517 Payushina, O. V. (2012). Hematopoietic microenvironment in the fetal liver: roles of diﬀerent cell populations. ISRN Cell Biol. 2012:979480. doi: 10.5402/2012/ 979480 Taoudi, S., Gonneau, C., Moore, K., Sheridan, J. M., Blackburn, C. C., Taylor, E., et al. (2008). Extensive hematopoietic stem cell generation in the AGM region via maturation of VE-cadherin+CD45+ pre-deﬁnitive HSCs. Cell Stem Cell 3, 99–108. doi: 10.1016/j.stem.2008.06.004 Pennetier, D., Oyallon, J., Morin-Poulard, I., Dejean, S., Vincent, A., and Crozatier, M. (2012). Size control of the Drosophila hematopoietic niche by bone morphogenetic protein signaling reveals parallels with mammals. Proc. Natl. Acad. Sci. U.S.A. 109, 3389–3394. doi: 10.1073/pnas.1109407109 Terrace, J. D., Currie, I. S., Hay, D. C., Masson, N. M., Anderson, R. A., Forbes, S. J., et al. (2007). REFERENCES Progenitor cell characterization and location in the developing human liver. Stem Cells Dev. 16, 771–778. doi: 10.1089/scd.2007.0016 Petzer, A. L., Zandstra, P. W., Piret, J. M., and Eaves, C. J. (1996). Diﬀerential cytokine eﬀects on primitive (CD34+CD38-) human hematopoietic cells: novel responses to Flt3-ligand and thrombopoietin. J. Exp. Med. 183, 2551–2558. doi: 10.1084/jem.183.6.2551 Theodore, L. N., Hagedorn, E. J., Cortes, M., Natsuhara, K., Liu, S. Y., Perlin, J. R., et al. (2017). Distinct roles for matrix metalloproteinases 2 and 9 in embryonic hematopoietic stem cell emergence, migration, and niche colonization. Stem Cell Rep. 8, 1226–1241. doi: 10.1016/j.stemcr.2017.03.016 Prashad, S. L., Calvanese, V., Yao, C. Y., Kaiser, J., Wang, Y., Sasidharan, R., et al. (2015). GPI-80 deﬁnes self-renewal ability in hematopoietic stem cells during human development. Cell Stem Cell 16, 80–87. doi: 10.1016/j.stem.2014.10.020 Travnickova, J., Tran Chau, V., Julien, E., Mateos-Langerak, J., Gonzalez, C., Lelievre, E., et al. (2015). Primitive macrophages control HSPC mobilization and deﬁnitive haematopoiesis. Nat. Commun. 6:6227. doi: 10.1038/ncomms7227 Ramadori, G., and Saile, B. (2002). Mesenchymal cells in the liver–one cell type or two? Liver 22, 283–294. Turpen, J. B., Knudson, C. M., and Hoefen, P. S. (1981). The early ontogeny of hematopoietic cells studied by grafting cytogenetically labeled tissue anlagen: localization of a prospective stem cell compartment. Dev. Biol. 85, 99–112. doi: 10.1016/0012-1606(81)90239-6 Rhodes, K. E., Gekas, C., Wang, Y., Lux, C. T., Francis, C. S., Chan, D. N., et al. (2008). The emergence of hematopoietic stem cells is initiated in the placental vasculature in the absence of circulation. Cell Stem Cell 2, 252–263. doi: 10.1016/j.stem.2008.01.001 Rieger, M. A. (2015). ATF4, a new player in fetal HSC expansion. Blood 126, 2351–2352. doi: 10.1182/blood-2015-10-671370 Wagner, JE Jr, Brunstein, C. G., Boitano, A. E., DeFor, T. E., McKenna, D., Sumstad, D., et al. (2016). Phase I/II trial of stemregenin-1 expanded umbilical cord blood hematopoietic stem cells supports testing as a stand-alone graft. Cell Stem Cell 18, 144–155. doi: 10.1016/j.stem.2015.10.004 Roy, V., and Verfaillie, C. M. (1999). Expression and function of cell adhesion molecules on fetal liver, cord blood and bone marrow hematopoietic progenitors: implications for anatomical localization and developmental stage speciﬁc regulation of hematopoiesis. Exp. Hematol. 27, 302–312. doi: 10.1016/ S0301-472X(98)00031-9 Walasek, M. A., van Os, R., and de Haan, G. (2012). Hematopoietic stem cell expansion: challenges and opportunities. Ann. N. Y. Acad. Sci. 1266, 138–150. doi: 10.1111/j.1749-6632.2012.06549.x Wittig, O., Paez-Cortez, J., and Cardier, J. E. (2010). REFERENCES Liver sinusoidal endothelial cells promote B lymphopoiesis from primitive hematopoietic cells. Stem Cells Dev. 19, 341–350. doi: 10.1089/scd.2009.0300 Rybtsov, S., Ivanovs, A., Zhao, S., and Medvinsky, A. (2016). Concealed expansion of immature precursors underpins acute burst of adult HSC activity in foetal liver. Development 143, 1284–1289. doi: 10.1242/dev.131193 March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 10 Mahony and Bertrand The Embryonic HSC Niche Zhang, C. C., and Lodish, H. F. (2004). Insulin-like growth factor 2 expressed in a novel fetal liver cell population is a growth factor for hematopoietic stem cells. Blood 103, 2513–2521. doi: 10.1182/blood-2003-08-2955 Wolf, A., Aggio, J., Campbell, C., Wright, F., Marquez, G., Traver, D., et al. (2017). Zebraﬁsh Caudal Haematopoietic Embryonic Stromal Tissue (CHEST) Cells Support Haematopoiesis. Sci. Rep. 7:44644. doi: 10.1038/srep44644 Zhao, Y., Zhou, J., Liu, D., Dong, F., Cheng, H., Wang, W., et al. (2015). ATF4 plays a pivotal role in the development of functional hematopoietic stem cells in mouse fetal liver. Blood 126, 2383–2391. doi: 10.1182/blood-2015-03-63 3354 Xue, Y., Gao, S., and Liu, F. (2015). Genome-wide analysis of the zebraﬁsh Klf family identiﬁes two genes important for erythroid maturation. Dev. Biol. 403, 115–127. doi: 10.1016/j.ydbio.2015.05.015 Xue, Y., Lv, J., Zhang, C., Wang, L., Ma, D., and Liu, F. (2017). The vascular niche regulates hematopoietic stem and progenitor cell lodgment and expansion via klf6a-ccl25b. Dev. Cell 42:e344. doi: 10.1016/j.devcel.2017.07.012 Zovein, A. C., Hofmann, J. J., Lynch, M., French, W. J., Turlo, K. A., Yang, Y., et al. (2008). Fate tracing reveals the endothelial origin of hematopoietic stem cells. Cell Stem Cell 3, 625–636. doi: 10.1016/j.stem.2008. 09.018 Yu, S., Luo, F., and Jin, L. H. (2018). The Drosophila lymph gland is an ideal model Yu, S., Luo, F., and Jin, L. H. (2018). The Drosophila lymph gland is an ideal model for studying hematopoiesis. Dev. Comp. Immunol. 83, 60–69. doi: 10.1016/j.dci. 2017.11.017 for studying hematopoiesis. Dev. Comp. Immunol. 83, 60–69. doi: 10.1016/j.dci. 2017.11.017 Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Yvernogeau, L., and Robin, C. (2017). Restricted intra-embryonic origin of bona ﬁde hematopoietic stem cells in the chicken. Development 144, 2352–2363. doi: 10.1242/dev.151613 Zambetti, N. A., Ping, Z., Chen, S., Kenswil, K. J. G., Mylona, M. A., Sanders, M. A., et al. Frontiers in Cell and Developmental Biology \| www.frontiersin.org REFERENCES (2016). Mesenchymal inﬂammation drives genotoxic stress in hematopoietic stem cells and predicts disease evolution in human pre-leukemia. Cell Stem Cell 19, 613–627. doi: 10.1016/j.stem.2016.08.021 Copyright © 2019 Mahony and Bertrand. 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. Copyright © 2019 Mahony and Bertrand. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Zhang, C. C., Kaba, M., Ge, G., Xie, K., Tong, W., Hug, C., et al. (2006). Angiopoietin-like proteins stimulate ex vivo expansion of hematopoietic stem cells. Nat. Med. 12, 240–245. doi: 10.1038/nm1342 March 2019 \| Volume 7 \| Article 34 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 11
https://openalex.org/W4302418533	https://iris.unimore.it/bitstream/11380/1104226/1/jhep1605.pdf	English	null	Mixed symmetry tensors in the worldline formalism	arXiv (Cornell University)	2,016	cc-by	15,371	Published for SISSA by Springer Received: April 7, 2016 Accepted: May 4, 2016 Published: May 10, 2016 Received: April 7, 2016 Accepted: May 4, 2016 Published: May 10, 2016 Keywords: Gauge Symmetry, Wilson, ’t Hooft and Polyakov loops, Scattering Ampli- tudes Open Access, c⃝The Authors. Article funded by SCOAP3. Mixed symmetry tensors in the worldline formalism doi:10.1007/JHEP05(2016)056 doi:10.1007/JHEP05(2016)056 Contents 1 Introduction 1 2 Worldline theory and anti-commuting colour ﬁelds 3 3 Extension of the worldline action 6 3.1 Gauging the U(F) symmetry 7 4 Path integral quantisation 10 4.1 Evaluation of the functional integrals 12 5 Bosonic colour ﬁelds 15 6 The generalised worldline theory 17 7 Path integral quantisation 19 7.1 Evaluation of the path integral 21 8 Conclusion 23 A U(F ) gauge ﬁxing 25 B Functional determinants 26 JHEP05(2016)056 A U(F ) gauge ﬁxing B Functional determinants Mixed symmetry tensors in the worldline formalism JHEP05(2016)056 Olindo Corradinia,b and James P. Edwardsc aDipartimento di Scienze Fisiche, Informatiche e Matematiche, Universit`a degli Studi di Modena e Reggio Emilia, via Campi 213/A, I-41125 Modena, Italy bINFN — Sezione di Bologna, via Irnerio 46, I-40126 Bologna, Italy cDepartment of Mathematical Sciences, University of Bath, Claverton Down, Bath BA2 7AY, U.K. E-mail: olindo.corradini@unimore.it, jpe28@bath.ac.uk Olindo Corradinia,b and James P. Edwardsc aDipartimento di Scienze Fisiche, Informatiche e Matematiche, Universit`a degli Studi di Modena e Reggio Emilia, via Campi 213/A, I-41125 Modena, Italy bINFN — Sezione di Bologna, via Irnerio 46, I-40126 Bologna, Italy cDepartment of Mathematical Sciences, University of Bath, Claverton Down, Bath BA2 7AY, U.K. E-mail: olindo.corradini@unimore.it, jpe28@bath.ac.uk E-mail: olindo.corradini@unimore.it, jpe28@bath.ac.uk Abstract: We consider the ﬁrst quantised approach to quantum ﬁeld theory coupled to a non-Abelian gauge ﬁeld. Representing the colour degrees of freedom with a single family of auxiliary variables the matter ﬁeld transforms in a reducible representation of the gauge group which — by adding a suitable Chern-Simons term to the particle action — can be projected onto a chosen fully (anti-)symmetric representation. By considering F families of auxiliary variables, we describe how to extend the model to arbitrary tensor products of F reducible representations, which realises a U(F) “ﬂavour” symmetry on the world- line particle model. Gauging this symmetry allows the introduction of constraints on the Hilbert space of the colour ﬁelds which can be used to project onto an arbitrary irreducible representation, speciﬁed by a certain Young tableau. In particular the occupation num- bers of the wavefunction — i.e. the lengths of the columns (rows) of the Young tableau — are ﬁxed through the introduction of Chern-Simons terms. We verify this projection by calculating the number of colour degrees of freedom associated to the matter ﬁeld. We sug- gest that, using the worldline approach to quantum ﬁeld theory, this mechanism will allow the calculation of one-loop scattering amplitudes with the virtual particle in an arbitrary representation of the gauge group. Keywords: Gauge Symmetry, Wilson, ’t Hooft and Polyakov loops, Scattering Ampli- tudes ArXiv ePrint: 1603.07929 ArXiv ePrint: 1603.07929 Open Access, c⃝The Authors. Article funded by SCOAP3. 1 Introduction The worldline formalism [1, 2] is proving to be a valuable alternative to the more tra- ditional approaches to quantum ﬁeld theory. In particular it provides computationally eﬃcient methods of obtaining scattering amplitudes, having been originally applied to re-derive the Bern-Kosower master formulae [3, 4] without recourse to string theory. It has now been applied to many problems, such as multi-loop amplitudes [5], the analysis of eﬀective actions [6], ﬁeld theory on curved space time [7] and photon-graviton mix- ing [8], higher spin ﬁelds [9, 10] and tree-level scattering [11, 12], amongst many others. There are also phenomenological applications such as in photon / gluon scattering [13], bound states computations [14], and the description of standard model matter [15] and its uniﬁcation [16]. These ﬁrst quantised theories involve actions which realise a worldline supersymmetry, with the super-partners to the worldline coordinates providing the ﬁeld’s spin degrees of freedom. The interaction between the matter ﬁeld and the gauge ﬁeld is expressed through the gauge invariant Wilson-loop coupling. More recently the worldline approach to non-Abelian ﬁeld theory has been im- proved [17, 18], and the diﬃculty related to the path ordering of the Wilson loop (necessary – 1 – to guarantee the gauge invariance) has been lifted, thus making the perturbative expan- sion more manageable. This was achieved by representing colour degrees of freedom (and the coupling to the Lie algebra valued gauge ﬁeld) with additional auxiliary ﬁelds [19– 21]. These ﬁelds enlarge the Hilbert space of the quantum theory but lead to the matter ﬁeld transforming in a reducible representation of the gauge group. The Hilbert space may be ﬁnite-dimensional or inﬁnite dimensional, depending on how the colour degrees of freedom have been included in the worldline theory. Either way, projection onto a ﬁnite- dimensional irreducible representation has been achieved by coupling the colour variables to a U(1) gauge ﬁeld, including an additional Chern-Simons term whose level is ﬁxed so as to select the required representation [22, 23]. The worldline gauge ﬁeld does not modify the theory in any other way than to impose the constraint enforcing irreducibility, since in one dimension the U(1) theory has trivial dynamics. By tuning the Chern-Simons coupling constant it is possible to pick out each representation at will. 1 Introduction JHEP05(2016)056 One of the limitations that previous approaches have encountered is that it has only been possible to describe fully anti-symmetric or fully symmetric representations of the gauge group. This has followed from the use of additional ﬁelds which are anti-commuting (Grassmann variables) or commuting respectively. This is certainly suﬃcient to describe standard model fermions — which transform in (conjugate-)fundamental or trivial repre- sentations [15] — and the multiplets used in SU(5) uniﬁcation — which are fully anti- symmetric [16] — but more exotic matter and a worldline description of gluons require a generalisation of these ideas. It is also interesting to uncover the additional structure the worldline theory requires in order to accommodate the description of more complex matter. In this article we intend to explain how this generalisation can be achieved by introducing several “families” of the auxiliary ﬁelds and partially gauging a unitary symmetry that rotates between them. The Hilbert space that arises will be shown to be the tensor product of the spaces associated to each set of ﬁelds which provides a much richer structure. For this reason a more general procedure of projecting onto an irreducible representation is required which we will construct. In the present article we ﬁrst deal with the case that the colour ﬁelds are fermionic, before repeating our analysis in the event that they are bosonic so as to provide a complete picture of our approach. For both scenarios we will study the phase-space theory that is familiar from the worldline approach and describe the introduction of the colour ﬁelds and the projection onto an arbitrary irreducible representation of the gauge group. To demonstrate that our construction is correct we will functionally quantise the resulting theories on the circle, using the path integral to compute the number of degrees of freedom associated to the matter ﬁeld. This article begins by revising the quantum mechanical description of spinning point particles and the incorporation of a coupling to the gauge ﬁeld. In section 3 we then discuss additional anti-commuting worldline ﬁelds that carry the particle’s colour information, and the extra worldline global symmetry that arises. We achieve irreducibility by partially gauging this symmetry. In section 4 we carry out the functional quantisation of the colour ﬁelds to count the number of degrees of freedom they associate to the matter ﬁeld. 1 Introduction Having achieved this for fermionic colour ﬁelds, in sections 5 and 6 we brieﬂy provide an account of – 2 – the worldline theory for bosonic auxiliary ﬁelds, again quantising the theory on the circle to count the number of degrees of freedom in section 7. 1We take {xµ, pν}P B = δµν and {ψµ, ψν}P B = −iδµν which generate gauge transformations through δz = z, ξ(τ)H + iη(τ)Q P B. 2 Worldline theory and anti-commuting colour ﬁelds First quantised approaches to quantum ﬁeld theory begin with a quantum mechanical description of the dynamical and spin degrees of freedom of the ﬁeld. For example, the phase-space description of a classical spin 1/2 point particle in Minkowski space (which is relevant for a worldline description of the Dirac ﬁeld) takes the following form [24, 25] JHEP05(2016)056 S [ω, p, ψ, e, χ] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ −eH −iχQ , (2.1) (2.1) where the ωµ represent the embedding of the particle’s worldline in space and the ψµ describe its spin degrees of freedom. We have also introduced the functions H = 1 2p2 (Hamiltonian) and Q = p·ψ (super-charge) which are constrained to vanish by the equations of motion of the einbein, e(τ), and the gravitino, χ(τ), respectively. Through Poisson (Dirac) brackets1 these functions generate the local super-symmetry transformations of the matter ﬁelds. δωµ = ξ(τ)pµ + iη(τ)ψµ(τ); δpµ = 0; δψµ = −η(τ)pµ (2.2) (2.2) where ξ(τ) is the generator of reparameterisations and η(τ) is the (Grassmann) generator of local supersymmetry transformations. Invariance of the action is achieved through the accompanying transformations of the super-gravity multiplet: δe = ˙ξ(τ) + 2iχη(τ); δχ = ˙η(τ), (2.3) (2.3) whilst the constraint functions satisfy the N = 1 supersymmetry algebra {Q, Q}P B = −2iH. (2.4) (2.4) In the canonical quantisation scheme, the phase space variables are promoted to linear operators with the fundamental relations [ˆxµ, ˆpµ] = iδµν and { ˆψµ, ˆψν} = δµν and the constraints must be imposed as operator equations which deﬁne the physical state space: ˆp2 \|phys⟩= 0 and ˆψ · ˆp \|phys⟩= 0. The anti-commutation algebra can be solved by taking ψµ = 1 √ 2γµ, which shows that the purpose of the spin degrees of freedom is to generate the γ-matrices. The constraints are then seen to ensure that the physical states are on shell and satisfy the Dirac equation. To couple this particle (and consequently the underlying ﬁeld which it is representing) to an Abelian gauge potential requires the modiﬁcation of H and Q. Minimal coupling is achieved by replacing the conjugate momentum, pµ, by its covariant version, πµ = pµ −Aµ, where we have absorbed the coupling strength into the gauge ﬁeld Aµ. 2 Worldline theory and anti-commuting colour ﬁelds The eﬀect of this – 3 – is to the re-deﬁne the susy generator as Q ≡ψ · π = ψ · (p −A). Now the Hamiltonian, H, is determined by the supersymmetry algebra (2.4) where a further change is encountered: {Q, Q}P B = −2iH ⇒H = 1 2π2 + i 2ψµFµνψν, (2.5) (2.5) where Fµν is the ﬁeld strength tensor built out of Aµ as Fµν = ∂µAν −∂νAµ. This idea can be extended to the non-Abelian case, where the vector potential is Lie algebra valued. We will take the generators of this algebra, {T a}, to be Hermitian and choose them in the fundamental representation of the symmetry group (we limit attention to the gauge group SU(N) for physical reasons), so Aµ = AaµT a. These additional details can be incorporated into the worldline action at a classical level by introducing additional Grassmann variables which carry the colour degrees of freedom that upon quantisation will create the associated Hilbert space [19, 20]. To do this we follow [17] and [15, 16], deﬁning N pairs of Grassmann ﬁelds, ¯cr and cr, which transform in the (anti-)fundamental representation of SU(N), giving them the following Poisson brackets: JHEP05(2016)056 {¯cr, cs}P B = −iδr s; {¯cr, ¯cs}P B = 0 = {cr, cs}P B. (2.6) (2.6) By taking [T a, T b] = ifabcT c, it is easy to check that the expressions By taking [T a, T b] = ifabcT c, it is easy to check that the expressions Ra ≡¯cr(T a)rscs , (2.7) (2.7) which can be used to absorb the gauge group indices of the generators, supply us with a (classical) representation of the Lie algebra, i.e. n Ra, Rbo P B = fabcRc . (2.8) (2.8) To correctly produce the above Poisson brackets between the auxiliary ﬁelds, their dynamics is speciﬁed as S[¯c, c] = R dτ i¯cr ˙cr. In the path integral formulation this ﬁrst order action is responsible for producing the path-ordering prescription that is required to maintain gauge invariance for the coupling to a non-Abelian ﬁeld. 2 Worldline theory and anti-commuting colour ﬁelds The full particle action thus reads S [ω, p, ψ, e, χ, ¯c, c] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr ˙cr −e eH −iχ eQ , (2.9) (2.9) where eH = eπ2 + i 2ψµF a µνψν¯cr(T a)rscs; eQ = ψ · eπ; eπµ = pµ −Aaµ¯cr(T a)rscs (2.10) (2.10) and F a µν = ∂µAa ν −∂νAa µ + ifabcAb µAc ν has been completed to the full (non-Abelian) ﬁeld strength tensor. The above charges provide the modiﬁed constraints which impose the new mass-shell condition and the covariant Dirac equation γ · D \|phys⟩= 0 in the presence of the gauge ﬁeld. Due to (2.8), the terms in eH and eQ which couple the gauge ﬁeld to the particle worldline retain the correct group structure, while keeping the particle action gauge-invariant. This is easy to verify. By calling U a generic SU(N) gauge transformation, the coloured ﬁelds transform as (colour indices are left implied when unnecessary) Aµ →U Aµ + i∂µ U†, c →Uc, ¯c →¯c U† (2.11) (2.11) – 4 – and the momentum ˜πµ can be made gauge-invariant by imposing the gauge transformation (2.12) pµ →pµ −i¯c U†∂µUc (2.12) that in turn makes the symplectic form R dτ(p · ˙ω + i¯c˙c) gauge-invariant. So the additional ﬁelds simply ensure the desired interactions are maintained without the need to rely on algebra valued potentials which would require path ordering upon functional quantisation. that in turn makes the symplectic form R dτ(p · ˙ω + i¯c˙c) gauge-invariant. So the additional ﬁelds simply ensure the desired interactions are maintained without the need to rely on algebra valued potentials which would require path ordering upon functional quantisation. It is easy to see that there is an additional global U(1) symmetry in (2.9) which acts on the auxiliary Grassmann ﬁelds as cr →e−iϑcr and ¯cr →¯creiϑ. The conserved current associated to this symmetry is L = ¯crcr and it is useful to gauge this symmetry by modifying the action to JHEP05(2016)056 S [ω, p, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr ˙cr −e eH −iχ eQ −a(L −s) , (2.13) (2.13) where the new gauge ﬁeld a(τ) transforms under the U(1) symmetry as δa = ˙ϑ. 2 Worldline theory and anti-commuting colour ﬁelds We have also introduced a constant, s = n −N 2 , to build a gauge invariant Chern-Simons term for the gauge ﬁeld SCS = R dτ a(τ)s, which modiﬁes the constraint produced by the equation of motion for a(τ) to L + N 2 = n. To see why these quantities should be introduced we must examine the Hilbert space of the additional Grassmann ﬁelds. The anti-commutation relations associated to the canonical quantization of (2.6) can be realised by promoting ¯cr and cr to creation and annihilation operators, ˆc† r and ˆcr, acting on coherent states ⟨¯u\| = ⟨0\|e¯urˆcr; ⟨¯u\| ˆc†r = ¯ur ⟨¯u\| ; ⟨¯u\| ˆcr = ∂¯ur ⟨¯u\| , (2.14) (2.14) whereby the operators naturally act by multiplication or derivation with respect to the Grassmann variables ¯ur. Wave functions are then built out of components which transform as anti-symmetric tensor products of the original representation of the gauge ﬁeld — for a state \|Ψ⟩its associated wave function Ψ(x, ¯u) = ⟨¯u, x\|Ψ⟩has a ﬁnite expansion Ψ(x, ¯u) = ψ(x) + ψr1(x)¯ur1 + ψr1r2(x)¯ur1 ¯ur2 + . . . + ψr1r2...rN (x)¯ur1 ¯ur2 · · · ¯urN , (2.15) (2.15) with totally antisymmetric tensors ψr1r2...rl. The wave function is consequently not de- scribed by an irreducible representation of the gauge group. However, the constraint func- tion, L, becomes the number operator ˆL = ˆc† rcr whose eigenvalues indicate the occupation of each state. Acting on the Fock space we solve an operator ordering ambiguity by choos- ing the anti-symmetric combination ˆL = 1 2 (¯ur∂¯ur −∂¯ur ¯ur) so that the constraint imposed by the gauge ﬁeld becomes ˆL + N 2 \|Ψ⟩= n \|Ψ⟩−→ ¯ur ∂ ∂¯ur −n Ψ(x, ¯u) = 0. (2.16) (2.16) This constraint selects from (2.15) the component that transforms with n fully anti- symmetrised indices, ψr1...rn, by enforcing all other components to vanish, therefore acting to project the wavefunction onto a single irreducible representation. This method has been used in the past to construct worldline theories describing higher spin ﬁelds [22, 26], diﬀer- ential forms [23, 27] and one-loop gluon amplitudes in the presence of bosonic and spinor matter [17, 18]. This constraint selects from (2.15) the component that transforms with n fully anti- symmetrised indices, ψr1...rn, by enforcing all other components to vanish, therefore acting to project the wavefunction onto a single irreducible representation. 2 Worldline theory and anti-commuting colour ﬁelds This method has been used in the past to construct worldline theories describing higher spin ﬁelds [22, 26], diﬀer- ential forms [23, 27] and one-loop gluon amplitudes in the presence of bosonic and spinor matter [17, 18]. – 5 – There is an obvious limitation, however, in using the worldline action (2.13) because the representations in which the matter ﬁeld can transform are restricted to those built out of fully antisymmetric tensor products of the original representation of the gauge group. In this work we generalise the worldline action to overcome this inadequacy and provide the means to project onto any, arbitrarily chosen representation of the symmetry group. This provides a complete framework in which the worldline approach can be applied to any form of fermionic matter ﬁeld by merely introducing some additional degrees of freedom to take part in quantisation whose rˆole is to ensure that the contribution from undesired wavefunction components are excluded from physical results. JHEP05(2016)056 3 Extension of the worldline action To achieve a richer Hilbert space associated to the additional Grassmann ﬁelds the simple generalisation we propose is to include F families of these “colour” ﬁelds. We will denote the families by an additional subscript (continuing to prescribe to the summation convention) and deﬁne the generalised Poisson brackets {¯cr f, cgs}P B = −iδfgδr s; {¯cr f, ¯cs g}P B = 0 = {cfr, cgs}P B. (3.1) (3.1) The worldline action (2.9) is modiﬁed to The worldline action (2.9) is modiﬁed to S [ω, p, ψ, e, χ, ¯c, c] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr f ˙crf −e eH −iχ eQ , (3.2) (3.2) where in the tilded quantities the conjugate momentum now involves a sum over the F families: eπµ = pµ −Aa µ¯cr f(T a)rscfs. This action remains invariant under local supersymme- try transformations generated by eQ and local reparameterisations generated by eH, but the global U(1) symmetry that was previously encountered has been enlarged to a non-Abelian “ﬂavour group” U(F). Speciﬁcally, if Λ = e−iλ is an element of the global symmetry group, the action (3.2) is invariant under cfr →Λfgcgr and ¯cr f →¯cr gΛ† gf. f gf We ﬁnd the conserved currents associated to these transformations by using the Noether trick. The result is a generalisation of the occupation number2 Lfg = ¯cr fcfg. These currents satisfy the following U(F) algebra Lfg, Lf′g′ P B = iδfg′Lf′g −iδf′gLfg′ (3.3) (3.3) so they can be used to build an inﬁnitesimal U(F) transformation by taking Poisson brack- ets with λ = λfgLfg: δλcfr = {cfr, λ} P B = −iλfgcgr; δλ¯cr f = ¯cr f, λ P B = i¯cr gλgf. (3.4) δλcfr = {cfr, λ} P B = −iλfgcgr; δλ¯cr f = ¯cr f, λ P B = i¯cr gλgf. (3.4) (3.4) The Fock space associated to the Grassmann ﬁelds is also more complicated, because one can now act on the vacuum with creation operators associated to diﬀerent families: Ψ(x, ¯u) = X {n1,n2,...nF } ψr1...rn1,s1...sn2,...,t1...tnF (x) ¯ut1 F . . . ¯u tnF F · · · ¯us1 2 . . . ¯u sn2 2 ¯ur1 1 . . . ¯u rn1 1 (3.5) (3.5) 2Note that setting g = f, then Lff coincides with the familiar current related to the occupation number of family f. 3 Extension of the worldline action 2Note that setting g = f, then Lff coincides with the familiar current related to the occupation number of family f. – 6 – – 6 – where each component is completely antisymmetric in the blocks of indices associated to the individual families (but has no special symmetry between indices from separate families). So wavefunctions are in general described by a reducible representation of the symmetry group, with the components in (2.15) transforming as tensor products of the representation created by each family — aﬀecting a Young tableaux notation we may write Ψ(x, ¯u) ∼ X {n1,n2,...nF } .. \|{z} nF ⊗. . . ⊗ .. \|{z} n2 ⊗ .. \|{z} n1 (3.6) Ψ(x, ¯u) ∼ X {n1,n2,...nF } .. \|{z} nF ⊗. . . ⊗ .. \|{z} n2 ⊗ .. \|{z} n1 (3.6) (3.6) JHEP05(2016)056 n2 nF to highlight how the components transform (the fth term in the product has nf boxes, representing nf fully anti-symmetrised indices). We have made a positive step in terms of enlarging the Hilbert space to include components transforming in more general represen- tations, but it is now necessary to generalise the procedure which projects out undesired contributions to physical phenomena in order to select a chosen component that transforms irreducibly from (3.6). 3Note that {Lf, Lf′}P B = 0 but {Lf, Lgg′}P B = iδfgLfg′−iδfg′Lgf does not vanish in general. Summing over f, however, is suﬃcient to produce an invariant U(1) generator L. 3.1 Gauging the U(F ) symmetry To attain the desired projection we follow the ideas of section 2 by gauging the U(F) symmetry (3.4) through the introduction of gauge ﬁelds afg(τ). Under a local U(F) trans- formation afg(τ) is taken to transform as δλafg(τ) = ˙λfg(τ) + i [a(τ), λ(τ)]fg (3.7) δλafg(τ) = ˙λfg(τ) + i [a(τ), λ(τ)]fg (3.7) which ensures the invariance of which ensures the invariance of S [ω, p, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr f ˙crf −e eH −iχ eQ −afgLfg , (3.8) (3.8) under local U(F) transformations. At this juncture, we would also like to introduce an appropriate Chern-Simons term for each of the diagonal entries Lf ≡Lff (no sum implied), so as to project onto the subspace of components which have a given number of indices associated with each family. This means that it is not possible to gauge the entire U(F) group, since in that case the only Abelian invariant3 is L = P f Lf. This would only allow the addition of a single Chern-Simons term s R dτ P f aff and we would be unable to control the occupation number of individual families separately. Instead, we begin by gauging the U(1)F ⊂U(F) subgroup of the full symmetry group by introducing the diagonal ﬁelds af ≡aff and diagonal generators Lf only. In this case the F diagonal gauge ﬁelds af have an Abelian transformation. We can thus introduce F Chern-Simons terms so as to impose the constraints on the Fock space ˆLf + N 2 \|Ψ⟩= nf \|Ψ⟩−→ ¯ur f ∂ ∂¯ur f −nf ! Ψ(x, ¯u) = 0, (3.9) (3.9) 3Note that {Lf, Lf′}P B = 0 but {Lf, Lgg′}P B = iδfgLfg′−iδfg′Lgf does not vanish in general. Summing over f, however, is suﬃcient to produce an invariant U(1) generator L. – 7 – which is suﬃcient to project onto the wavefunction component with nf antisymmetric indices associated to each family, ψr1...rn1,s1...sn2,...,t1...tnF (x). Gauging only U(1)F is not suﬃcient to achieve irreducibility, since the selected wave- function component itself transforms as a tensor product of the antisymmetric representa- tion created by each family (see (3.6) for illustration). To then pick out a single irreducible representation we must gauge a larger part of the U(F) symmetry whilst keeping this sub- group invariant. 3.1 Gauging the U(F ) symmetry Consider the partial gauging of those Lfg with 1 ⩽g ⩽f ⩽F, for which the following ﬁrst class sub-algebra holds Lfg, Lf′f P B = −iδf′gLf + iLf′g g′ = f Lfg, Lf′g′ P B = −iδf′gLfg′ g′ < f Lfg, Lf′g′ P B = 0 f < g′. (3.10) JHEP05(2016)056 (3.10) So the subalgebra is ﬁrst class. Later we will refer to the group associated to these genera- tors as the “auxiliary gauge group.” With this choice it is easy to check that the diagonal elements of the gauge ﬁeld still transform in an Abelian manner, δλafg(τ) = ˙λfg(τ) + i X g⩽k⩽f (afk(τ)λkg(τ) −λfk(τ)akg(τ)) (3.11) ⇒δλaff(τ) = ˙λff(τ), (3.12) (3.11) (3.12) ⇒δλaff(τ) = ˙λff(τ), which means that F independent Chern-Simons terms remain admissible and the con- straints (3.9) continue to be imposed. We therefore arrive at the complete expression for the worldline action in phase-space S′ [ω, p, ψ, e, χ, ¯c, c, a]= Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr f ˙crf −e eH −iχ eQ − F X f=1 af (Lf −sf) − X g<f afgLfg , (3.13) (3.13) where we have used the notation af ≡aff. This is the worldline theory that we will use for the rest of this paper. First we examine the eﬀect that the larger gauging has on the Fock space of the Grassmann ﬁelds and in the next section we will quantise the worldline theory in the path integral formulation. In canonical quantisation the U(F) generators Lfg become (for g < f) ˆLfg = ¯ur f∂¯urg when acting on coherent states and the equation of motion for afg imposes the constraint ˆLfg \|Ψ⟩= 0 −→¯ur f ∂ ∂¯urg Ψ(x, ¯u) = 0. (3.14) (3.14) To understand how this additional constraint acts on the wavefunction selected by the Lf, it suﬃces to recall that ψr1...rn1,s1...sn2,...,t1...tnF (x) can be written as a sum over compo- nents transforming in irreducible representations with various symmetries between indices belonging to the diﬀerent families. 3.1 Gauging the U(F ) symmetry For example, taking F = 2, and n2 = 2, n1 = 1, the – 8 – surviving part — satisfying (3.9) — of Ψ(x, ¯u) can be written (exploiting the SF symmetry allowing the permutation of the nf) as Ψ(x, ¯u) = ψr1r2,s1(x) ¯ur1 2 ¯ur2 2 ¯us1 1 = ψ[r1r2,s1](x) + 1 3 2ψr1r2,s1(x) −ψs1r1,r2(x) −ψr2s1,r1(x) ¯ur1 2 ¯ur2 2 ¯us1 1 , (3.15) (3.15) which represents the tensor product decomposition ⊗ = ⊕ . The constraints (3.14) imply that only the components in the kernel of all the ˆLfg survive. Each ˆLfg acts as an “anti-symmetriser,” i.e. it replaces a ¯ur g with a ¯ur f and anti-symmetrises over the indices involved. For the current example, the only non-diagonal number operator that exists is ˆL21, which acts as JHEP05(2016)056 ˆL21 ¯ur1 2 ¯ur2 2 ¯us1 1 = ¯ut 2 ∂ ∂¯ut 1 ¯ur1 2 ¯ur2 2 ¯us1 1 = ¯us1 2 ¯ur1 2 ¯ur2 2 (3.16) (3.16) which is now anti-symmetric under interchange of the indices r1, r2 and s1. It is easy to verify that the result of contracting the indices of the term in rounded brackets of (3.15) with the above product identically vanishes (this component transforms in the representation which implies symmetrisations over the r1, s1 and r2, s1 pairs). This means that the constraint acts as 0 = ˆL21Ψ(x, ¯u) = ψ[r1r2,s1](x)¯ur1 2 ¯ur2 2 ¯us1 2 =⇒ψ[r1r2,s1](x) = 0 (3.17) (3.17) so that the only surviving component of the wavefunction is that which transforms in the representation. This is described by a Young tableau with n2 rows in the ﬁrst column and n1 rows in the second column. In general, the eﬀect of the constraints can be summarised as follows. The diagonal operators ˆLf act to select the occupation number of each family of Grassmann ﬁelds to be equal to nf, picking out a single wavefunction component (i.e. with ﬁxed nf) from the tensor product (3.6). The oﬀ-diagonal operators ˆLfg then select from this reducible sum a single component with compatible symmetries. To lie in the kernel of all of the ˆLfg requires the maximum inter-family symmetry so that the single surviving component that the constraints project onto transforms in the representation with Young tableau Ψ(x, ¯u) ∼ nF ... .. ... . . . ... ...n1 .. \| {z } F columns (3.18) Ψ(x, ¯u) ∼ nF ... .. 3.1 Gauging the U(F ) symmetry ... . . . ... ...n1 .. \| {z } F columns (3.18) Ψ(x, ¯u) ∼ nF ... .. ... . . . ... ...n1 .. \| {z } F columns (3.18) (3.18) {z F columns which has nF rows in the ﬁrst column, nF−1 rows in the second and so forth, where N ⩾nF ⩾nF−1 ⩾. . . ⩾n2 ⩾n1. The symmetrisation between indices associated to – 9 – diﬀerent families implied by (3.18) means that this is the only component that lies in the kernel of all of the ˆLfg. This ensures that the wavefunction transforms in an irreducible representation of SU(N). The conclusion is that with the current approach it is possible to project onto an arbitrary representation, without being restricted to those with single- column tableaux. We provide a much more versatile method for the worldline description of SU(N)-charged matter ﬁelds forming any multiplet. In the next section we verify this analysis by carrying out functional quantisation for a particle traversing a closed path. This has application in the worldline formalism where it is related to the partition function and one-loop eﬀective action of the quantum ﬁeld theory describing matter interacting with a gauge ﬁeld. We will do so by counting the number of degrees of freedom (DoF) associated to the matter ﬁeld for comparison with the dimension of the representation denoted by (3.18). JHEP05(2016)056 4 Path integral quantisation To facilitate the path integral quantisation of the particle described by (3.13) we ﬁrst rewrite the action in conﬁguration space by solving the equation of motion for pµ and putting it on-shell. This leads to a (Euclidean) path integral Z DeDχDωDψD¯cDcDa Vol(Gauge) exp (−S [ω, ψ, e, χ, ¯c, c, a]) (4.1) (4.1) where where S [ω, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ " e−1 ˙ω2 2 + 1 2ψ · ˙ψ −χ e ˙ω · ψ + ¯cr f ˙cfr −i¯cr fAa(T a)rscfs + F X f=1 iaf ¯cr fcfr−sf + X g<f iafg¯cr fcgr # . (4.2) (4.2) Above, the division by the volume of the gauge group (i.e. diﬀeomorphisms, supersymmetry and auxiliary gauge group) takes care of summing only over gauge-inequivalent conﬁgu- rations. In equation (4.2) we have denoted A = ˙ω · A −e 2ψµFµνψν which makes up the Wilson loop exponent as W = Pei R 1 0 A(τ)dτ and we denote the F independent Chern- Simons levels by sf. The path ordering prescription is now taken care of by the inclusion of the additional Grassmann ﬁelds. The ends of the interval are identiﬁed to form S1 — the boundary conditions on ω, e and a are periodic, whereas the Grassmann ﬁelds ψ, χ, ¯c and c are taken to be anti-periodic. This form of the action would be crucial in the context of the worldline formalism where it would form the basis for the calculation of multi-loop scattering amplitudes. Indeed, the quantisation of the N = 1 point particle leads to the functional integrals involving the embedding coordinates, ω, its spin degree of freedom, ψ and the worldline supergravity multiplet e and χ. For simplicity, however, we will focus here only on the dimension associated to the colour degrees of freedom. This allows us to focus on the restricted path integral relating to the auxiliary Grassmann ﬁelds and to – 10 – temporarily drop the coupling to the external gauge ﬁeld (so we set A = 0). 4 Path integral quantisation The restricted path integral which counts the particle degrees of freedom is then 1 Vol(Gauge) Z Y f,r D¯cr fDcfr Y h<g Dagh Y f Daf × exp  − Z 1 0 dτ   F X f=1 ¯cr fDfcfr −isfaf + X h<g iagh¯cr fcgr    , (4.3) (4.3) we we have used the diagonal gauge ﬁelds to construct the covariant derivative Df ≡ d dτ + iaf , and “Vol(Gauge)” now only refers to the auxiliary gauge group. The calculation of this functional integral will be shown to yield the correct number of degrees of freedom associated to the colour space for any choice of s = (s1, s2, . . . , sF ). JHEP05(2016)056 The action in (4.3) enjoys the (local) auxiliary gauge group symmetry that was dis- cussed in the previous section and so to compute the path integral we must ﬁrst employ a gauge ﬁxing procedure. On the circle, it is not possible to entirely gauge away the ﬁelds afg; at best one may set the akj to be constant (see appendix A). The most general form of akj would be a constant upper-triangular matrix, however as discussed in appendix A the functional integral over the colour ﬁelds is independent of the oﬀ-diagonal entries so it suﬃces to take   ˆafg =      θ1 0 · 0 0 θ2 · 0 · · · · 0 0 · θF     . (4.4) (4.4) The {θf} then remain as moduli to be integrated over and can be identiﬁed as angles in [0, 2π] by examining large U(1)F transformations. This gauge ﬁxing can be compensated for by introducing Faddeev-Popov ghosts, γfg and ¯γfg, which ensure that gauge invariance is maintained. The precise form of the ghost action depends on how the global U(F) symmetry was gauged. With the partial gauging discussed in the previous section, the gauge transformation of the ˆafg is given by (3.11): The {θf} then remain as moduli to be integrated over and can be identiﬁed as angles in [0, 2π] by examining large U(1)F transformations. This gauge ﬁxing can be compensated for by introducing Faddeev-Popov ghosts, γfg and ¯γfg, which ensure that gauge invariance is maintained. The precise form of the ghost action depends on how the global U(F) symmetry was gauged. 4 Path integral quantisation With the partial gauging discussed in the previous section, the gauge transformation of the ˆafg is given by (3.11): δˆagh = ˙λgh + i X h⩽k⩽g (θgδgkλkh −θhλgkδkh) = d dτ + i (θg −θh) λgh (4.5) (4.5) which leads to a ghost action Sg [¯γ, γ, {θ}] = Z 1 0 dτ X h⩽g ¯γgh d dt + i (θg −θh) γgh. (4.6) (4.6) The result of these calculations is the following gauge ﬁxed path integral F Y f=1 Z 2π 0 dθf 2π Z D(¯c, c)D(¯γ, γ) exp − Z 1 0 dτ " F X f=1 ¯cr fDfcfr −iθfsf + X h⩽g ¯γgh d dt +i (θg−θh) γgh #! (4.7) F Y f=1 Z 2π 0 dθf 2π Z D(¯c, c)D(¯γ, γ) exp − Z 1 0 dτ " F X f=1 ¯cr fDfcfr −iθfsf d (4.7) – 11 – – 11 – where on the chosen gauge slice Df → d dτ + iθf and the sf are determined by our choice of population numbers for each family of colour ﬁelds. The ghosts are taken to be periodic around the circle. The functional integration leads to a product of various determinants which we now calculate. 4.1 Evaluation of the functional integrals The integration over the colour degrees of freedom factorises due to our choice of gauge ﬁxing the ﬁelds of the auxiliary gauge group to ˆafg and provides a product of functional determinants F Y f=1 Det APB d dτ + iθf N (4.8) JHEP05(2016)056 (4.8) which we deﬁne as the product of their eigenvalues for functions on S1 with anti-periodic boundary conditions (APB). In appendix B we show that this yields which we deﬁne as the product of their eigenvalues for functions on S1 with anti-periodic boundary conditions (APB). In appendix B we show that this yields F Y f=1 2 cos θf 2 N . (4.9) (4.9) Similarly, integrating over the ghosts leads to a product of functional determinants on functions with periodic boundary conditions (PB), deﬁned as the product of their non-zero eigenvalues, which eﬀectively provides a measure for the U(F) moduli: Similarly, integrating over the ghosts leads to a product of functional determinants on functions with periodic boundary conditions (PB), deﬁned as the product of their non-zero eigenvalues, which eﬀectively provides a measure for the U(F) moduli: µ ({θk}) = F Y f=1 Det PB d dτ × Y h⩽g Det PB d dτ + i (θg −θh) (4.10) (4.10) which evaluates to (see appendix B) which evaluates to (see appendix B) µ ({θk}) = Y h<g 2i sin θg −θh 2 . (4.11) (4.11) Putting this together we ﬁnd that the number of colour degrees of freedom is represented by a simple multiple integral KF F Y f=1 Z 2π 0 dθf 2π eiθfsf µ ({θk}) 2 cos θf 2 N (4.12) (4.12) which represents the main result of this article. In this formula, KF is a normalisation constant whose inverse is equal to the number of fundamental domains included in the integration over the U(F) moduli. We illustrate the application of this formula in a few special cases, before rewriting it in terms of more convenient variables. We ﬁrst verify that for F = 1 formula (4.12) reduces to the previously calculated result for totally anti-symmetric representations. Then we demonstrate the case F = 2 to prove that the correct counting of DoF is reached for the example discussed in (3.15). 4.1 Evaluation of the functional integrals So starting with only one family of additional Grassmann – 12 – ﬁelds we take s1 = n1 −N 2 and note that the measure, µ (θ1) = 1, is trivial and K1 = 1. Then (4.12) becomes Z 2π 0 dθ1 2π eiθ1n1e−iθ1 N 2 ei θ1 2 + e−i θ1 2 N = Z 2π 0 dθ1 2π eiθ1n1 1 + e−iθ1N . (4.13) (4.13) At this stage it is convenient to eﬀect a change of variable by noting that z1 = eiθ1 parametrises the U(1) Wilson-loop on the circle. This leads to At this stage it is convenient to eﬀect a change of variable by noting that z1 = eiθ1 parametrises the U(1) Wilson-loop on the circle. This leads to I dz1 2πizn1−1 1 1 + 1 z1 N = I dz1 2πi (z1 + 1)N zN+1−n1 1 , (4.14) JHEP05(2016)056 (4.14) where the contour of integration is taken to be the circle \|z1\| = 1. For n1 ⩽N there is a pole at z = 0 and we can arrive at the elementary result I dz1 2πi (z1 + 1)N zN+1−n1 =        N n1 ! 0 ⩽n1 ⩽N 0 Otherwise . (4.15) (4.15) This shows that for F = 1, the main formula (4.12) correctly counts the colour degrees of freedom for a matter ﬁeld transforming in the representation with n1 fully anti-symmetrised indices (this corresponds to the dimension of the single-column Young tableau with n1 rows). This result is in agreement with [17, 18], where scattering amplitudes were calculated using worldline techniques for matter ﬁelds transforming in representations built out of antisymmetric tensor products of the fundamental representation of SU(N). A less trivial example is that corresponding to (3.15). It is useful to consider this calculation in order to see how the modular measure provides the required projection onto an irreducible representation. To describe such a wavefunction we need F = 2 families of the additional Grassmann ﬁelds and for generality will take occupation numbers n = (n1, n2) with 0 ⩽n1 ⩽n2 ⩽N. To achieve this requires the Chern-Simons levels to be chosen as N 1 N 1 = n1 −N 2 −1 2 and s2 = n2 −N 2 + 1 2. 4.1 Evaluation of the functional integrals Then application of (4.12) provides (K2 = 1) Z 2π 0 dθ1 2π Z 2π 0 dθ2 2π eiθ1(n1−1 2)eiθ2(n2+ 1 2) 1+e−iθ1N 1+e−iθ2N ei θ1 2 e−i θ2 2 −e−i θ1 2 ei θ2 2 (4.1 Z 2π 0 dθ1 2π Z 2π 0 dθ2 2π eiθ1(n1−1 2)eiθ2(n2+ 1 2) 1+e−iθ1N 1+e−iθ2N ei θ1 2 e−i θ2 2 −e−i θ1 2 ei θ2 2 (4.16) (4.16) ( ) where we have already cancelled the terms e−iθ1 N 2 and e−iθ2 N 2 by factorising their inverses from the ﬁrst two terms in rounded brackets. It is again convenient to make use of the Wilson loop variables z1 = eiθ1 and z2 = eiθ2 to rewrite this expression as an integration on the complex plane: I I dz1 2πi dz2 2πi (z1 + 1)N zN+1−n1 1 (z2 + 1)N zN+1−n2 2 − (z1 + 1)N zN+1−(n1−1) 1 (z2 + 1)N zN+1−(n2+1) 2 ! . (4.17) (4.17) – 13 – – 13 – Using (4.15) this is easily determined to be Using (4.15) this is easily determined to be Using (4.15) this is easily determined to be N n1 N n2 − N n1 −1 N n2 + 1 = N! (N −n2)! (N + 1)! (N −(n1 −1))! n2 + 1 −n1 n1!(n2 + 1)! = N n2 N +1 n1 n2 + 1 −n1 n2 + 1 . (4.18) N n1 N n2 − N n1 −1 N n2 + 1 = N! (N −n2)! (N + 1)! (N −(n1 −1))! n2 + 1 −n1 n1!(n2 + 1)! 1 = N n2 N +1 n1 n2 + 1 −n1 n2 + 1 . (4.18) (4.18) It is straightforward to conﬁrm that this is the dimension of the representation with Young tableau n2 n1 JHEP05(2016)056 n2 n1 .. (4.19) (4.19) having n2 rows in the ﬁrst column and n1 rows in the second.4 In particular, setting n2 = 2 and n1 = 1 projects onto the representation as in (3.17). 4This diagram has “factors” N! (N−n2)! (N+1)! (N−(n1−1))! and “hooks” n1!(n2 −n1)! (n2+1)! (n2+1−n1)! which divide to give its dimension in agreement with (4.18) — see [28]. 4.1 Evaluation of the functional integrals Returning now to the general case it is useful to incorporate the approach taken in these two examples by making a complete change of variables in (4.12) by introducing F complex variables zf = eiθf and to present the main formula in terms of these Wilson loop variables. In general, to project onto the SU(N) representation speciﬁed by occupation numbers n = (n1, n2, . . . nF ) requires Chern-Simons levels sf = nf −N 2 −F−(2f−1) 2 to impose the correct constraints on the physical state space. Overall, following some elementary manipulations the general formula can be written rather concisely in a similar manner to the examples above as KF F Y f=1 I dzf 2πi Y h<g 1 −zh zg (zf + 1)N zN+1−nf f , (4.20) (4.20) which represents the colour degrees of freedom for the matter ﬁeld in a ﬁrst quantised approach. The complex integrals simply pick out the poles at zf = 0 for each term in the expansion of the product in rounded brackets. We have checked that this provides the correct counting of degrees of freedom for arbitrary representations of SU(N) given by Young tableaux with nf rows in the fth column. The ﬁnal formula (4.20) is applied by ﬁxing the vector n and expanding the expression into a sum over products of complex integrals. For illustration, setting F = 6 and n = (1, 3, 5, 6, 6, 8), assuming N ⩾8, the integration evaluates to dim (4.21) g dim (4.21) 4This diagram has “factors” N! (N−n2)! (N+1)! (N−(n1−1))! and “hooks” n1!(n2 −n1)! (n2+1)! (n2+1−n1)! which divide to give its dimension in agreement with (4.18) — see [28]. (4.21) 4This diagram has “factors” N! (N−n2)! (N+1)! (N−(n1−1))! and “hooks” n1!(n2 −n1)! (n2+1)! (n2+1−n1)! which divide to give its dimension in agreement with (4.18) — see [28]. 4This diagram has “factors” N! (N−n2)! (N+1)! (N−(n1−1))! and “hooks” n1!(n2 −n1)! (n2+1)! (n2+1−n1)! which divide to give its dimension in agreement with (4.18) — see [28]. 4This diagram has “factors” N! (N−n2)! (N+1)! (N−(n1−1))! and “hooks” n1!(n2 −n1)! (n2+1)! (n2+1−n1)! which divide to give its dimension in agreement with (4.18) — see [28]. – 14 – which shows that the partial gauging of the U(F) symmetry group correctly projects onto the desired sector in the Hilbert space of the colour ﬁelds. 4.1 Evaluation of the functional integrals Returning to (4.1), the remaining functional integration over the worldline matter and super-gravity ﬁelds generates the partition function for the dynamics and spin degrees of freedom of the particle. In the simple case we have considered here (without turning on the gauge ﬁeld) we just reproduce the well known functional quantisation of a free point particle multiplet. We leave it to future work to demonstrate that, when the gauge ﬁeld is coupled to the particle as in (4.1), the colour degrees of freedom produce the correct Wilson-loop interaction for the chosen representation of SU(N). In that context, one of the physically signiﬁcant representations that could be used is the adjoint of SU(N). This requires the use of F = 2 families and the choice n = (1, N −1), whereby (4.18) evaluates to N2 −1 as required. JHEP05(2016)056 5 Bosonic colour ﬁelds This approach has been used successfully in worldline applications in the past where the same ﬁrst order action was used to enforce the path ordering inside functional integrals [18]. As before, this action is appended to the phase-space action (2.1) along with the associated change to the conjugate momentum: πµ is replaced by eπµ = pµ −Aaµ¯cr(T a)rscs, which modiﬁes the super-charge and Hamiltonian to eQ = ψ · eπ; eH = eπ2 + i 2ψµF a µνψν¯cr(T a)rscs (5.4) (5.4) JHEP05(2016)056 with F a µν the components of the full (non-Abelian) ﬁeld strength tensor with respect to the gauge group generators. To brieﬂy examine the Hilbert space associated to the additional ﬁelds we promote ¯c and c to creation and annihilation operators ˆc† and ˆc and consider their action on bosonic coherent states ⟨¯u\| = ⟨¯0\| e¯urˆcr; ⟨¯u\| ˆc†r = ¯ur ⟨¯u\| ; ⟨¯u\| ˆcr = ∂¯ur ⟨¯u\| , (5.5) (5.5) where the ¯ur are the complex (commuting) eigenvalues of the ˆc†r. The commutation relations are solved in this basis by taking ˆc†r ∼¯ur and ˆcr ∼∂¯ur. The Fock space is built by acting with creation operators on the vacuum and (in contrast to when the additional ﬁelds are Grassmann valued) is inﬁnite dimensional. Indeed, wavefunctions Φ(x, ¯u) now have a non-terminating expansion Φ(x, ¯u) = φ(x) + φr1(x)¯ur1 + φr1r2 ¯ur1 ¯ur2(x) + . . . + φr1r2...rp ¯ur1 ¯ur2. . . ¯urp + . . . (5.6) (5.6) where the components transform in representations constructed out of p fully symmetric tensor products. The wavefunction thus transforms as a reducible sum. We resolve this as before, constructing a projector by gauging the U(1) symmetry present in the worldline theory, cr →e−iϑcr, ¯cr →¯creiϑ. This leads to a new action S [ω, p, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ p · ˙ω + i 2ψ · ˙ψ + i¯cr ˙cr −e eH −iχ eQ −a(L −s) , (5.7) (5.7) where a(τ) is the gauge ﬁeld which behaves under the U(1) symmetry as δa = ˙ϑ and L = ¯crcr is the number operator for the colour ﬁelds which generates their U(1) transformations. The Chern-Simons term, SCS = R dτ a(τ)s, is familiar by now and involves the constant, s = n + N 2 . 5 Bosonic colour ﬁelds In the previous sections the anti-commuting nature of the additional colour ﬁelds meant that their Hilbert space was spanned by states transforming in representations built out of totally anti-symmetric tensor products. In turns out, as has been shown in [18], that bosonic colour ﬁelds can also provide the correct coupling to the gauge ﬁeld. In the pro- ceeding sections we will show that this choice of commuting ﬁelds instead supplies a basis of fully symmetric representations of SU(N) which will form the building blocks of the complete colour information associated to the matter ﬁeld. We will be able to arrive at a similar formula to (4.20) based on this alternative approach to generating the Wilson loop interaction, which may be useful in instances where it is more natural — or less awkward — to use bosonic variables. Before we introduce this new approach it should be pointed out that it is already possible to generate fully symmetric representations using our pre- vious results. Following the notation above, this can be done by using F = p families of anti-commuting colour ﬁelds and projecting onto the sector of the Hilbert space with occupation numbers given by n = (1, 1, . . . , 1). The total number of degrees of freedom is easily calculated using the main formula (4.20) and turns out to be N+p−1 p , coinciding with the dimension of the representation with p fully symmetric indices. We prefer, however, to develop here a formalism where the fully symmetric represen- tations are the fundamental constituents. To this end we deﬁne N bosonic ﬁelds ¯cr and cr which transform in the (anti-)fundamental representation of SU(N). These ﬁelds have the following Poisson brackets {¯cr, cs}P B = iδr s ; {¯cr, ¯cs}P B = 0 = {cr, cs}P B , (5.1) (5.1) that can be seen to follow from the particle action S[¯c, c] = R 1 0 dτ¯cr ˙cr, and we use them to incorporate the colour indices of the gauge group generators since, deﬁning Sa = ¯cr(T a)rscs , (5.2) (5.2) one gets {Sa, Sb}P B = fabcSc . (5.3) (5.3) – 15 – – 15 – We can therefore also use these commuting variables to represent the coupling between the particle and the gauge ﬁeld and so take care of the non-commuting nature of the gauge group generators in a fully classical worldline action. 6 The generalised worldline theory To describe representations other than those corresponding to fully symmetric tensor prod- ucts requires us to generalise these ideas to include multiple families of the additional com- muting ﬁelds. Following section 3 an additional index is appended to the colour ﬁelds to produce F copies of these colour degrees of freedom, ¯cr f and cfr. The Poisson brackets are generalised to {¯cr f, cgs}P B = iδr sδfg; {¯cr f, ¯cs g}P B = 0 = {cfr, cgs}P B (6.1) (6.1) JHEP05(2016)056 and a sum over these families of ﬁelds is included in each of the functions in the action (for example, eπµ →pµ −Aaµ¯cr f(T a)rscfs). This again enlarges the U(1) symmetry group to the U(F) that rotates between the new ﬁelds. The U(F) rotations are given by cfr →Λfgcgr and ¯cr f →¯cr gΛ† gf where Λ := e−iλ and λ is a constant Lie-algebra valued generator. We again introduce the generalised occupation numbers which are the conserved currents of this global symmetry, now built out of the bosonic colour ﬁelds as Lfg = ¯cr fcfg. At the classical level, these currents obey the U(F) algebra Lff′, Lgg′ P B = −iδf′gLfg′ + iδfg′Lgf′, (6.2) (6.2) nerate inﬁnitesimal U(F) transformations through Poisson brackets with G = so they generate inﬁnitesimal U(F) transformations through Poisson brackets with G = λfgLfg: δλcfr = {cfr, G} P B = −iλfgcgr; δλ¯cr f = ¯cr f, G P B = i¯cr gλgf. (6.3) (6.3) The Hilbert space is now substantially enlarged, consisting of components that trans- form in tensor products of the representation associated to each family and to achieve irreducibility it will be necessary to partially gauge this new symmetry. The richer Fock space can be expressed as a non-terminating sum over components of the form Φ(x, ¯u) = X {n1,n2,...nF } φr1...rn1,s1...sn2,...,t1...tnF (x)¯ut1 F . . . ¯u tnF F . . . ¯us1 2 . . . ¯u sn2 2 . . . ¯ur1 1 . . . ¯u rn1 1 , (6.4) (6.4) where the indices are arranged into blocks of length nf which are fully symmetric. The form of the wavefunctions in (6.4) is nicely explained diagrammatically by using Young tableaux notation: Φ(x, ¯u) ∼ X {n1,n2,...nF } ·· \| {z } nF ⊗. . . 5 Bosonic colour ﬁelds The equation of motion for a(τ) imposes the constraint L = s which acts on the wavefunctions as ˆL −N 2 \|Φ⟩= n \|Φ⟩→ ¯ur ∂ ∂¯ur −n Φ(x, ¯u) = 0, (5.8) (5.8) where this time we resolve the operator ordering ambiguity via symmetrisation so that ˆL = 1 2 (¯ur∂¯ur + ∂¯ur ¯ur). With Φ given by (5.6), this constraint picks out the component that transforms with exactly n fully symmetrised indices, φr1r2...rn. This brief analysis can be compared to the discussion in section 2, where many similarities can be found. – 16 – 6 The generalised worldline theory ⊗ ·· \| {z } n2 ⊗ ·· \| {z } n1 , (6.5) (6.5) so that in general the wavefunction does not transform in an irreducible representation. In order to have irreducibility, following the success of earlier sections, we gauge only a part of this U(F) symmetry by introducing gauge ﬁelds afg for 1 ⩽g ⩽f ⩽F correspond- ing to a subset of the Lfg. We showed above that this subset satisﬁes a ﬁrst class subalgebra and that it led to the desired projection onto a speciﬁed irreducible representation. To this – 17 – end we modify the action to S′ [ω, p, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ " p · ˙ω + i 2ψ · ˙ψ + i¯cr f ˙crf −e eH −iχ eQ − F X f=1 af (Lf −sf) − X g<f afgLfg # , (6.6) − F X f=1 af (Lf −sf) − X g<f afgLfg # , (6.6) (6.6) which will form the basis for the remainder of this article. In equation (6.6) we have split the U(F) gauge ﬁelds into two types. As before, the diagonal generators Lf ≡Lff are gauged by af ≡aff from which the oﬀ-diagonal elements have been separated and gauged by afg. To the former we have also introduced an independent Chern-Simons term associated to each family whose level is related to the desired occupation number of that sector of the Hilbert space. Invariance under local U(F) transformations ﬁxes the variation of the gauge ﬁelds to be JHEP05(2016)056 δλafg(τ) = ˙λfg(τ) + i X g⩽k⩽f (afk(τ)λkg(τ) −λfk(τ)akg(τ)) (6.7) (6.7) as is already familiar. as is already familiar. We very brieﬂy review how this achieves irreducibility, as the argument is essentially the same as in the case that the colour ﬁelds are anti-commuting. It suﬃces to consider the equations of motion implied by the gauge ﬁelds. These impose constraints on the physical state space We very brieﬂy review how this achieves irreducibility, as the argument is essentially the same as in the case that the colour ﬁelds are anti-commuting. It suﬃces to consider the equations of motion implied by the gauge ﬁelds. These impose constraints on the physical state space ˆLf −N 2 \|Φ⟩= nf \|Φ⟩−→ ¯ur f ∂ ∂¯ur f −nf ! Φ(x, ¯u) = 0, (6.8) (6.8) where this time we have resolved an operator ordering ambiguity by symmetrising. 6 The generalised worldline theory This picks out from (6.4) the wavefunction component with nf symmetric indices associated to each family, φr1...rn1,s1...sn2,...,t1...tnF (x). The remaining oﬀ-diagonal constraints then se- lect only one irreducible representation from the tensor product denoted in (6.5). These constraints act on the Fock space as ˆLfg \|Φ⟩= 0 −→¯ur f ∂ ∂¯urg Φ(x, ¯u) = 0. (6.9) (6.9) To illustrate the eﬀect of this requirement we consider the same illustrative example as before, namely when F = 2 and n2 = 2, n1 = 1. In this case the constraints implied by L1 and L2 pick out the wavefunction component φr1r2,s1(x) so that the surviving part of Φ(x, ¯u) can be written as a reducible sum Φ(x, ¯u) = φr1r2,s1(x)¯ur1 2 ¯ur2 2 ¯us1 1 = φ(r1r2,s1)(x) + 1 3 (2φr1r2,s1(x) −φs1r1,r2(x) −φs1r2,r1(x)) ¯ur1 2 ¯ur2 2 ¯us1 1 (6.10) (6.10) which reﬂects the tensor product decomposition ⊗ = ⊕ (we have made use of the symmetries of the wavefunction to carry out some relabelling of indices to arrive at – 18 – this). The surviving components must lie in the kernel of the Lfg, which act to remove a ¯ur g, to introduce a ¯ur f and to symmetrise between indices. Indeed, acting with ˆL21 on the creation operators above yields this). The surviving components must lie in the kernel of the Lfg, which act to remove a ¯ur g, to introduce a ¯ur f and to symmetrise between indices. Indeed, acting with ˆL21 on the creation operators above yields ˆL21 ¯ur1 2 ¯ur2 2 ¯us1 1 = ¯ut 2 ∂ ∂¯ut 1 ¯ur1 2 ¯ur2 2 ¯us1 1 = ¯us1 2 ¯ur1 2 ¯ur2 2 (6.11) (6.11) whose result is symmetric under interchange of the r1, r2 and s1. Now the component in rounded brackets transforms under the representation with Young tableau which implies an antisymmetrisation between the r1, s1 and r2, s2 indices. For this reason, the only term that survives the action of ˆL21 is φ(r1r2,s1) which the constraint therefore forces to vanish and the only surviving component is that which transforms in the representation with n2 columns in the ﬁrst row and n1 columns in the second. JHEP05(2016)056 The general case is similar to this example. 6 The generalised worldline theory The Lf constraints pick out the wavefunc- tion component with nf fully symmetrised indices associated to each family, which further breaks down into components which transform in the tensor product decomposition illus- trated in the summand of (6.5). The Lfg constraints further restrict attention to a single component which lies in their kernel. Given the form of the Lfg this can only be achieved with a maximum amount of anti-symmetrisation between the indices of diﬀerent families. Then the Young tableau denoting the representation of the surviving component is Φ(x, ¯u) ∼ nf.. ..n1 .. ... . . . ... ..              F rows (6.12) (6.12) which contains nf columns in the fth row with nF ⩾nF−1 ⩾. . . ⩾n2 ⩾n1. Due to the anti-symmetrisation between the rows of each column implied by the shape of (6.12), only this component lies in the kernel of all of the Lfg. In this way, the partial gauging of the U(F) symmetry provides a projection onto a single irreducible representation of SU(N). This complements our earlier work where the tableau was made up by instead specifying the number of rows in each of F columns and provides another approach which projects onto an arbitrarily chosen representation without being limited to those with single-row tableaux. which contains nf columns in the fth row with nF ⩾nF−1 ⩾. . . ⩾n2 ⩾n1. Due to the anti-symmetrisation between the rows of each column implied by the shape of (6.12), only this component lies in the kernel of all of the Lfg. In this way, the partial gauging of the U(F) symmetry provides a projection onto a single irreducible representation of SU(N). This complements our earlier work where the tableau was made up by instead specifying the number of rows in each of F columns and provides another approach which projects onto an arbitrarily chosen representation without being limited to those with single-row tableaux. In the following section we will carry out the path integral quantisation of this theory for a particle whose path is closed. As discussed above this is relevant for calculations in the worldline approach to quantum ﬁeld theory where it is related to the partition function of the quantum ﬁeld theory describing a matter ﬁeld interacting with a gauge ﬁeld. 6 The generalised worldline theory We will calculate the degrees of freedom associated to the colour space of the matter ﬁeld and compare the result to the dimension of the representation in (6.12). 7 Path integral quantisation The situation is the same as in section 4: we consider a point particle that traverses a closed path by identifying its coordinates, ωµ, at either end of the domain which therefore – 19 – becomes S1 as before. We proceed by integrating the momenta pµ out so as to arrive (rotating to Euclidean signature) at a conﬁguration space action S [ω, ψ, e, χ, ¯c, c, a] = Z 1 0 dτ " e−1 ˙ω2/2 + 1 2ψ · ˙ψ −χ e ˙ω · ψ + ¯cr f ˙cfr −i¯cr fAa(T a)rscfs + F X f=1 iaf ¯cr fcfr−sf + X g<f iafg¯cr fcgr # , (7.1) where sf are the Chern-Simons level and we recall the previously deﬁned worldline function A = ˙ω·A−e 2ψµFµνψν which parameterises the phase of the Wilson loop, W = Pei R 1 0 A(τ)dτ . The object in question is the functional integral over conﬁgurations JHEP05(2016)056 Z DeDχDωDψD¯cDcDa Vol(Gauge) exp (−S [ω, ψ, e, χ, ¯c, c, a]). (7.2) (7.2) The introduction of the additional bosonic ﬁelds now replaces the path ordering prescrip- tion. We take periodic boundary conditions on ω, e and a, but now also on on ¯c, c. The remaining ﬁelds, ψ and χ, have anti-periodic boundary conditions as usual. We follow the analysis employed in section 4 by restricting attention to the colour degrees of freedom, considering the reduced path integral 1 Vol(Gauge) Z Y f,r D¯cr fDcfr Y h<g Dagh Y f Daf × exp  − Z 1 0 dτ   F X f=1 ¯cr fDfcfr −isfaf + X h<g iafg¯cr fcgr    , (7.3) (7.3) where the diagonal gauge ﬁelds are absorbed into deﬁning a covariant derivative Df ≡ d dτ + iaf and we have dropped the coupling to the external gauge ﬁeld. We will show that this functional integral evaluates to the correct number of degrees of freedom associated to the colour space for any choice of s = (s1, s2, . . . , sF ). In the worldline formalism we would also have to include the functional integrals involving the embedding coordinates, ω, the spin degree of freedom, ψ and the worldline supergravity multiplet e and χ, which would allow us to calculate scattering amplitudes and other physical observables. 7 Path integral quantisation For the present article we will continue to deal only with the colour degrees of freedom, leaving to future work the inclusion of the interaction with the gauge ﬁeld. The action of the colour ﬁelds in (7.3) is of course still invariant under the auxiliary symmetry group discussed above. As before, we gauge ﬁx this symmetry by ﬁxing the ﬁelds afg to be the constant matrix ˆafg =      θ1 0 · 0 0 θ2 · 0 · · · · 0 0 · θF     . (7.4) (7.4) where the parameters are interpreted as angles in [0, 2π] and are the moduli to be inte- grated over. Indeed, integrating over these variables will provide the projection onto an where the parameters are interpreted as angles in [0, 2π] and are the moduli to be inte- grated over. Indeed, integrating over these variables will provide the projection onto an – 20 – irreducible representation of SU(N). We have already found the Faddeev-Popov determi- nant associated to this gauge ﬁxing which led to the measure on the the moduli, (4.11), which takes the form irreducible representation of SU(N). We have already found the Faddeev-Popov determi- nant associated to this gauge ﬁxing which led to the measure on the the moduli, (4.11), which takes the form θ θ µ ({θk}) = Y h<g 2i sin θg −θh 2 (7.5) (7.5) and is the correct factor ensuring gauge invariance of functional integrals on our chosen gauge slice. This measure is also the factor that is required to ensure a projection onto a single irreducible representation of the symmetry group. 7.1 Evaluation of the path integral JHEP05(2016)056 d path integral takes the form The gauge ﬁxed path integral takes the form The gauge ﬁxed path integral takes the form KF F Y f=1 Z 2π 0 dθf 2π Z D(¯c, c) µ ({θk}) exp  − Z 1 0 dτ F X f=1 ¯cr fDfcfr −iθfsf   (7.6) (7.6) where the covariant derivative takes the simple form Df = d dτ + iθf and KF is the normalisation constant which also appeared for fermionic colour ﬁelds. The integration over ¯c and c leads to a functional determinant which is deﬁned as the product of its non- zero eigenvalues: F Y f=1 Det PB d dτ + iθf −N = F Y f=1 2i sin θf 2 −N , (7.7) (7.7) where we have made use of the results of appendix B. This gives a formula for the number of degrees of freedom associated to the colour space, which is the main result of this section, KF F Y f=1 Z 2π 0 dθf 2π eiθfsf µ ({θk}) 2i sin θf 2 −N . (7.8) (7.8) We will now give some example calculations making use of this formula to demonstrate that it provides the expected results. The simplest example to consider is for F = 1, whereby the formula is familiar from previous work and determines the dimension of fully symmetric representations. Taking s1 = n1 + N 2 and using µ(θ1) = 1, K1 = 1, (7.8) becomes Z 2π 0 dθ1 2π eiθ1n1eiθ1 N 2 ei θ1 2 −e−i θ1 2 −N = Z 2π 0 dθ1 2π eiθ1n1 1 −e−iθ1N . (7.9) (7.9) It so far has proven convenient to make a change of variable to write our expressions in terms of the U(1) Wilson-loop on the circle, z1 = eiθ1. Doing so we ﬁnd I dz1 2πiz1 zn1 1 (1 −1 z1 )N = I dz1 2πi zN−1+n1 1 (z1 −1)N . (7.10) (7.10) The contour of integration is the circle \|z1\| = 1, deformed to enclose the pole on the real axis at z1 = 1. This follows from the regularisation procedure outlined in appendix B, – 21 – – 21 – which requires a shift θ1 →θ1 −iϵ, sending z1 →z1eϵ for a small positive parameter ϵ. 7.1 Evaluation of the path integral The integral is easily calculated to be I dz1 2πi zN+n1 1 (z1 −1)N = N +n1−1 n1 , (7.11) (7.11) which agrees with the dimension of the SU(N) representation with n1 fully symmetrised indices. This is in agreement with the results of [18], where scattering amplitudes were calculated in the worldline formalism involving matter ﬁelds which transformed in fully symmetric representations of SU(N). We also return to (6.10) where the calculation is more interesting. This requires the use of F = 2 families of ﬁelds and we take the general case of occupation numbers n = (n1, n2) where n1 ⩽n2. Similarly to section 4, this is enforced by Chern-Simons terms s1 = n1 + N 2 −1 2 and s2 = n2 + N 2 + 1 2, in which case (7.8) gives (recall that K2 = 1) JHEP05(2016)056 Z 2π 0 dθ1 2π Z 2π 0 dθ2 2π eiθ1(n1−1 2)eiθ2(n2+ 1 2) × × 1 −e−iθ1−N 1 −e−iθ2−N ei θ1 2 e−i θ2 2 −e−i θ1 2 ei θ2 2 . (7.12) In this expression we have extracted factors of e−iθ1 N 2 and e−iθ2 N 2 from the ﬁrst two terms in rounded brackets to cancel them oﬀagainst their counterparts for simpliﬁcation. We will again make use of the Wilson-loop variables by deﬁning z1 = eiθ1 and z2 = eiθ2 which allows us to re-express this formula as a multiple integral on the complex plane: I I dz1 2πi dz2 2πi zN−1+n1 1 (z1 −1)N zN−1+n2 2 (z2 −1)N −zN−1+(n1−1) 1 (z1 −1)N zN−1+(n2+1) 2 (z2 −1)N ! . (7.13) (7.13) Making use of (7.11) this evaluates to Making use of (7.11) this evaluates to N +n1−1 n1 N +n2−1 n2 − N +n1−2 n1−1 N +n2 n2+1 = (N + n2 −1)! n2!(N −1)! (N + n1 −2)! (N −2)!n1! n2 + 1 −n1 n2 + 1 = N +n2−1 n2 N +n1−2 n1 n2 + 1 −n1 n2 + 1 . (7.14) (7.14) It is simple to verify that this corresponds to the dimension of the representation of SU(N) with Young tableau n2 n1 ·· (7.15) (7.15) having n2 columns in the ﬁrst row and n1 columns in the second.5 Specifying n2 = 2 and n1 = 1 projects on the representation as in (6.10). 5Following [28] this tableaux has “factors” (N+n2−1)! (N−1)! (N+n1−2)! (N−2)! and “hooks” n1!(n2+1)! (n2+1−n1)! whose quotient gives the dimension in (7.14). 7.1 Evaluation of the path integral In future work we intend to also include the coupling to the gauge ﬁeld in order to show that the additional colour ﬁelds produce the expected interaction — this would be the Wilson-loop coupling for the chosen representation of SU(N). This shows that the modular measure µ({θk}) correctly projects onto the wavefunction component transforming in the desired irreducible representation of the gauge group. This measure follows from the partial gauging of the U(F) symmetry that existed in the worldline theory. In the worldline approach it would be necessary to also complete the remaining integrals over the matter and supergravity ﬁelds in (7.1). This procedure is well known and gives the dynamical and spin degrees of freedom of the particle. In future work we intend to also include the coupling to the gauge ﬁeld in order to show that the additional colour ﬁelds produce the expected interaction — this would be the Wilson-loop coupling for the chosen representation of SU(N). As we have already mentioned one of the most physically signiﬁcant representations that one could project onto is the adjoint of SU(N). It remains possible to achieve this in the current setting: we would need F = N −1 families of ﬁelds and should take the vector n = (2, 1, . . . 1). We have veriﬁed that formula (7.16) then evaluates to N2 −1 as desired. Finally, before introducing the bosonic colour ﬁelds we commented that fully symmetric representations can be constructed out of a theory based on fully anti-symmetric representations. The converse is also true; choosing 0 ⩽F = p ⩽N and n = (1, 1, . . . 1), our main formula gives the correct number of degrees of freedom of an SU(N) tensor with p fully antisymmetric indices, Np , and vanishes if F > N. 7.1 Evaluation of the path integral having n2 columns in the ﬁrst row and n1 columns in the second.5 Specifying n2 = 2 and n1 = 1 projects on the representation as in (6.10). As before it is possible to write our main formula more compactly by representing all of the U(F) moduli by Wilson-loop variables, deﬁning zf = eiθf for f ∈[1, . . . F]. Assuming 5Following [28] this tableaux has “factors” (N+n2−1)! (N−1)! (N+n1−2)! (N−2)! and “hooks” n1!(n2+1)! (n2+1−n1)! whose quotient ves the dimension in (7.14). – 22 – occupation numbers n = (n1, n2, . . . , nF ) we take Chern-Simons levels sf = nf + N+1 2 −f. Then following similar lines to before the number of degrees of freedom carried by the matter ﬁeld in the worldline approach can be cast into the form occupation numbers n = (n1, n2, . . . , nF ) we take Chern-Simons levels sf = nf + N+1 2 −f. Then following similar lines to before the number of degrees of freedom carried by the matter ﬁeld in the worldline approach can be cast into the form F Y f=1 I dzf 2πi Y h<g 1 −zh zg zN−1+nf f (1 −zf)N , (7.16) (7.16) where each integration contour encloses the pole at z = 1. We have veriﬁed that this produces the correct number of degrees of freedom for any representation of SU(N) whose Young tableau has nf columns in the fth row. The formula (7.16) is used by ﬁxing the numbers in n and expanding the product into a sum of multiple complex integrals. For ex- ample, with F = 6 and n = (1, 2, 2, 4, 6, 7) (assuming that N ⩾6), the integration provides JHEP05(2016)056 dim . (7.17) (7.17) This shows that the modular measure µ({θk}) correctly projects onto the wavefunction component transforming in the desired irreducible representation of the gauge group. This measure follows from the partial gauging of the U(F) symmetry that existed in the worldline theory. In the worldline approach it would be necessary to also complete the remaining integrals over the matter and supergravity ﬁelds in (7.1). This procedure is well known and gives the dynamical and spin degrees of freedom of the particle. 8 Conclusion We have presented the construction of two related worldline theories which describe a spin 1/2 matter ﬁeld transforming in an arbitrary representations of SU(N). We have shown how coupling the matter ﬁeld to families of either commuting or anti-commuting auxiliary ﬁelds replaces the awkward path ordering prescription and allows the description of richer matter content, including tensors with mixed symmetry. We examined the Hilbert space of the auxiliary worldline ﬁelds which carry the colour information of the matter ﬁeld and – 23 – described how to isolate wavefunction components transforming in a desired irreducible representation. In the case that the auxiliary ﬁelds are anti-commuting, they provide a Fock space whose states transform in tensor products of F anti-symmetric representations of SU(N), whereas their commuting counterparts span a Hilbert space with states that are tensor products of F symmetric representations. The projection on an irreducible representation was achieved by partially gauging a U(F) symmetry which rotates between the F families of additional ﬁelds, and the introduction of Chern-Simons terms to constrain the occupation numbers of the particle wavefunction. Our main result arose from path integral quantisation of the worldline theory which provided a compact formula for the colour degrees of freedom of the matter ﬁeld. JHEP05(2016)056 In future work we will include the coupling to the gauge ﬁeld in order to complete a worldline description of arbitrary matter multiplets coupled to an external non-Abelian ﬁeld. This will be done by reinstating the Wilson-loop coupling between the colour ﬁelds and the gauge ﬁelds and computing the resulting path integral: this will provide a complete framework for the computation of gluon amplitudes in the presence of any form of matter (note that chiral fermions can also be described in a ﬁrst quantised setting by following the ideas presented in [15, 16]). The utility of this approach is clear from the existing eﬃciency of the worldline formalism, which ought to be preserved by the introduction of the colour ﬁelds. In general, the worldline formalism leads to Bern-Kosower rules that simplify the calculation of scattering amplitudes. In the non-Abelian case, these rules state that in order to achieve the full gauge invariance of the on-shell gluon amplitudes, one must append the one-particle irreducible expressions, that naturally come from the worldline representation, with “tree replacement rules” that allow the inclusion of gluon self-interactions. Acknowledgments The authors would like to express their deepest thanks to the warmth and hospitality of INFN Bologna where this work began life. The progress of this research has beneﬁted greatly from the insight of Fiorenzo Bastianelli and Christian Schubert with whom the au- thors have enjoyed many helpful discussions which have had direct bearing on the material presented in this article. JPE would thank INFN Bologna and University of Modena and Reggio Emilia for additional ﬁnancial support to make this manuscript possible. 8 Conclusion We wish to use our formalism to generalise Bern-Kosower rules for gluon scattering in the presence of an arbitrary matter ﬁeld. Although we have limited our attention to closed worldlines for the sake of simplicity, these methods are also well-suited for application to tree-level amplitudes, where the par- ticle endpoints are ﬁxed [11, 12]. This has application to dressed propagators in quantum ﬁeld theory and the analysis of bound states [14]. Finally, the old link between the world- line approach and string theory has recently been re-examined in [29, 30], where a contact interaction was introduced on the worldsheets of tensionless spinning strings. Identifying the endpoints of the strings with particle worldlines reproduced the Wilson-loop interaction for spinor QED. Generalising this theory to non-Abelian symmetry groups would require the incorporation of the colour degrees of freedom onto the string worldsheet and would lead to an alternative description of particle interactions in terms of interacting coloured strings. Furthermore, it would be worthwhile examining the relative merits of using the bosonic colour carrying ﬁelds compared to the Grassmann auxiliary ﬁelds as it would be useful to know which approach tends to oﬀer the most simplicity and versatility for the purpose of calculations in the worldline approach. Much work has also been carried out using worldline techniques for ﬁelds that are cou- pled to the gravitational ﬁeld. The techniques we have presented here can be incorporated into that context with minor modiﬁcations and may provide a powerful approach to the calculation of amplitudes involving gluons and/or gravitons. – 24 – A U(F ) gauge ﬁxing JHEP05(2016)056 Here we present the ﬁxing of the non-Abelian U(F) symmetry discussed in the main text. The generator of the symmetry, λ, is in the Lie algebra of U(F) and produces the following ﬁnite transformation on the gauge ﬁeld: a →U −1aU −iU −1 ˙U; U(τ) = eiλ(τ). (A.1) (A.1) It is tempting to try to use such a transforming to ﬁx the gauge ﬁeld to vanish identically. Simple algebra shows this is equivalent to solving d dτ U(τ) = −ia(τ)U(τ) which can be done by making use of the Path ordering prescription It is tempting to try to use such a transforming to ﬁx the gauge ﬁeld to vanish identically. Simple algebra shows this is equivalent to solving d dτ U(τ) = −ia(τ)U(τ) which can be done by making use of the Path ordering prescription U(τ) = Pe−i R τ 0 a(τ ′)dτ ′. (A.2) (A.2) Since the domain is S1 we require the transformation to be periodic, so this solution is not admissible. Instead, we can deﬁne a constant group valued matrix e−iΘ = Pe−i R 1 0 a(τ)dτ and construct a gauge transformation which takes afg(τ) onto the constant Θ in the Lie algebra of the symmetry group: Since the domain is S1 we require the transformation to be periodic, so this solution is not admissible. Instead, we can deﬁne a constant group valued matrix e−iΘ = Pe−i R 1 0 a(τ)dτ and construct a gauge transformation which takes afg(τ) onto the constant Θ in the Lie algebra of the symmetry group: ˜U(τ) = U(τ)eiΘ ⇒akj(τ) →Θkj. (A.3) (A.3) With the partial gauging we use in this paper, the Θ will be upper-triangular. However, when we return to the functional integral over the colour ﬁelds, the resulting functional determinant is easily shown to be independent of the oﬀdiagonal terms. These then factor out of the path integral, leading to an overall constant which is cancelled upon normalisation. So it suﬃces to keep only the diagonal elements of the Cartan subalgebra which leaves ˆakj in (4.4). These remaining moduli can be identiﬁed as angles by considering large U(1)F ⊂U(F) transformations and requiring periodicity. With the partial gauging we use in this paper, the Θ will be upper-triangular. However, when we return to the functional integral over the colour ﬁelds, the resulting functional determinant is easily shown to be independent of the oﬀdiagonal terms. A U(F ) gauge ﬁxing These then factor out of the path integral, leading to an overall constant which is cancelled upon normalisation. So it suﬃces to keep only the diagonal elements of the Cartan subalgebra which leaves ˆakj in (4.4). These remaining moduli can be identiﬁed as angles by considering large U(1)F ⊂U(F) transformations and requiring periodicity. The ﬁnal ingredient is the integration measure on the moduli, which we derive using the Faddeev-Popov formalism. To avoid overcounting one factorises the integration over orbits of the gauge group out of the functional integration. This is achieved by integrating only over the moduli θk, compensating for this gauge ﬁxing by introducing the ghosts in (4.6) That is, Z Da → Z DU Z Da Z D(¯γ, γ) e−Sg[¯γ,γ,{θ}]δ aU −ˆa (A.4) (A.4) – 25 – where ˆa is the gauge ﬁxed form of a and aU represents the gauged transformed value of a. In the main text we used the results of appendix B to integrate over the ghost degrees of freedom which, cancelling the functional integration over U with the volume of the gauge group, provided gauge ﬁxed integrals of the form where ˆa is the gauge ﬁxed form of a and aU represents the gauged transformed value of a. In the main text we used the results of appendix B to integrate over the ghost degrees of freedom which, cancelling the functional integration over U with the volume of the gauge group, provided gauge ﬁxed integrals of the form Z Da Ω[a] → F Y k=1 Z dθk 2π µ({θk}) Ω[ˆa({θk})] (A.5) (A.5) where Ωis any functional of the gauge ﬁelds and µ denotes the Faddeev-Popov measure maintaining gauge invariance. where Ωis any functional of the gauge ﬁelds and µ denotes the Faddeev-Popov measure maintaining gauge invariance. JHEP05(2016)056 B Functional determinants In this appendix we calculate the various functional determinants which arose in the main text. The main diﬀerence between the situations that are encountered is in the boundary conditions imposed on the ﬁelds in the functional integration. We deﬁne the functional determinant Det d dτ + iθ (B.1) (B.1) as the product of non-zero eigenvalues of the operator in brackets. On the space of periodic functions, the eigenvalue equation as the product of non-zero eigenvalues of the operator in brackets. On the space of periodic functions, the eigenvalue equation d dτ + iθ f(τ) = µf(τ) (B.2) (B.2) is solved by eigenfunctions f(τ) = e−iθτeiµτ, requiring µ = θ + 2nπ for n ∈Z. We arrange the product over these eigenvalues by pairing up the positive and negative integers (taken as part of our regularisation of the inﬁnite product) as θ ∞ Y n=1 1 − θ2 (2nπ)2 ∞ Y n=1 (2nπ)2 . (B.3) (B.3) The latter product can be ζ-function regularised and evaluates to a constant, independent of θ (see [11]). The ﬁrst product is well-known from the expansion of the sin function so we arrive at Det PB d dτ + θ = 2i sin θ 2 . (B.4) (B.4) For anti-periodic boundary conditions the eigenvalues are modiﬁed to µ = θ + (2n + 1)π, so their product becomes For anti-periodic boundary conditions the eigenvalues are modiﬁed to µ = θ + (2n + 1)π, so their product becomes θ ∞ Y n=0 1 − θ2 ((2n + 1) π)2 ∞ Y n=0 ((2n + 1) π)2 . (B.5) (B.5) The ﬁnal product gives an overall constant and the ﬁrst is related to the inﬁnite product expansion of the cos function, so we ﬁnd The ﬁnal product gives an overall constant and the ﬁrst is related to the inﬁnite product expansion of the cos function, so we ﬁnd Det APB d dτ + θ = 2 cos θ 2 . (B.6) (B.6) These results are used extensively in the main text. These results are used extensively in the main text. These results are used extensively in the main text. B Functional determinants which does not require such regularisation. which does not require such regularisation. Open Access. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited. B Functional determinants – 26 – We also wish to return to the calculation of the determinant on the space of periodic functions as the regularisation is important for the discussion presented in the main text. For this reason we re-derive the result using techniques derived from canonical quantisation. The determinant arising out of the following integration can be expressed as a trace over states in the Hilbert space: Det PB d dτ + θ −1 = I PB D(¯c, c) e− R 1 0 dτ ¯c( d dτ +θ)c = tr e−i θ 2(ˆc†ˆc+ˆcˆc†). (B.7) (B.7) JHEP05(2016)056 The trace is most easily evaluated in the occupation number basis, where the states are eigenvalues of ˆn = ˆc†ˆc. This leads to The trace is most easily evaluated in the occupation number basis, where the states are eigenvalues of ˆn = ˆc†ˆc. This leads to Det PB d dτ + θ −1 = ∞ X n=0 ⟨n\| e−i θ 2(ˆn+ 1 2) \|n⟩ = ∞ X n=0 e−i θ 2(n+ 1 2). (B.8) (B.8) We are obliged to regularise the geometric series by taking θ →θ −iϵ and arrive at Det PB d dτ + θ −1 = e−i θ 2 1 −e−iθ = 2i sin θ 2 −1 . (B.9) (B.9) The necessary regularisation of θ aﬀects the Wilson-loop variable z = eiθ →z = zeϵ, which is important in determining the contour in the complex plane along which this parameter is integrated. This changes the original circle of radius \|z\| = 1 to a circle that is slightly larger, therefore encompassing any poles that lie at unit length from the origin. On the other hand, when the variables ¯c and c are Grassmann valued then the trace becomes The necessary regularisation of θ aﬀects the Wilson-loop variable z = eiθ →z = zeϵ, which is important in determining the contour in the complex plane along which this parameter is integrated. This changes the original circle of radius \|z\| = 1 to a circle that is slightly larger, therefore encompassing any poles that lie at unit length from the origin. On the other hand, when the variables ¯c and c are Grassmann valued then the trace becomes Det APB d dτ + θ = X n=0,1 e−iθ(n−1 2) = 2 cos θ 2 (B.10) (B.10) which does not require such regularisation. References [1] M.J. Strassler, Field theory without Feynman diagrams: one loop eﬀective actions, Nucl. Phys. B 385 (1992) 145 [hep-ph/9205205] [INSPIRE]. [2] C. Schubert, Perturbative quantum ﬁeld theory in the string inspired formalism, Phys. Rept. 355 (2001) 73 [hep-th/0101036] [INSPIRE]. [3] Z. Bern and D.A. Kosower, Eﬃcient calculation of one loop QCD amplitudes, Phys. Rev. Lett. 66 (1991) 1669 [INSPIRE]. [3] Z. Bern and D.A. Kosower, Eﬃcient calculation of one loop QCD amplitudes, Phys. Rev. Lett. 66 (1991) 1669 [INSPIRE]. – 27 – [4] H.-T. Sato and M.G. Schmidt, Worldline approach to the Bern-Kosower formalism in two loop Yang-Mills theory, Nucl. Phys. B 560 (1999) 551 [hep-th/9812229] [INSPIRE]. [5] G.V. Dunne and C. Schubert, Multiloop information from the QED eﬀective Lagrangian, J. Phys. Conf. Ser. 37 (2006) 59 [hep-th/0409021] [INSPIRE]. [6] M.G. Schmidt and C. Schubert, On the calculation of eﬀective actions by string methods, Phys. Lett. B 318 (1993) 438 [hep-th/9309055] [INSPIRE]. 7] F. Bastianelli and A. Zirotti, Worldline formalism in a gravitational background, Nucl. Phys. B 642 (2002) 372 [hep-th/0205182] [INSPIRE]. [8] F. Bastianelli, J.M. Davila and C. Schubert, Gravitational corrections to the Euler-Heisenberg Lagrangian, JHEP 03 (2009) 086 [arXiv:0812.4849] [INSPIRE]. JHEP05(2016)056 [9] F. Bastianelli, O. Corradini and E. Latini, Higher spin ﬁelds from a worldline perspective, JHEP 02 (2007) 072 [hep-th/0701055] [INSPIRE]. [10] F. Bastianelli, O. Corradini and E. Latini, Spinning particles and higher spin ﬁelds on (A)dS backgrounds, JHEP 11 (2008) 054 [arXiv:0810.0188] [INSPIRE]. [11] J.P. Edwards, Contact interactions between particle worldlines, JHEP 01 (2016) 033 [arXiv:1506.0813]. [12] N. Ahmadiniaz, F. Bastianelli and O. Corradini, Dressed scalar propagator in a non-abelian background from the worldline formalism, Phys. Rev. D 93 (2016) 025035 [arXiv:1508.0514]. [13] N. Ahmadiniaz, C. Schubert and V.M. Villanueva, String-inspired representations of photon/gluon amplitudes, JHEP 01 (2013) 132 [arXiv:1211.1821] [INSPIRE]. [14] F. Bastianelli, A. Huet, C. Schubert, R. Thakur and A. Weber, Integral representations combining ladders and crossed-ladders, JHEP 07 (2014) 066 [arXiv:1405.7770] [INSPIRE]. [15] P. Mansﬁeld, The fermion content of the Standard Model from a simple world-line theory, Phys. Lett. B 743 (2015) 353 [arXiv:1410.7298] [INSPIRE]. [16] J.P. Edwards, Uniﬁed theory in the worldline approach, Phys. Lett. B 750 (2015) 312 [arXiv:1411.6540] [INSPIRE]. [17] F. Bastianelli, R. Bonezzi, O. Corradini and E. Latini, Particles with non abelian charges, JHEP 10 (2013) 098 [arXiv:1309.1608] [INSPIRE]. [18] F. Bastianelli, R. Bonezzi, O. Corradini, E. Latini and K.H. Ould-Lahoucine, A worldline approach to colored particles, arXiv:1504.0361. [30] J.P. Edwards and P. Mansﬁeld, QED as the tensionless limit of the spinning string with contact interaction, Phys. Lett. B 746 (2015) 335 [arXiv:1409.4948] [INSPIRE]. References [19] A.P. Balachandran, P. Salomonson, B.-S. Skagerstam and J.-O. Winnberg, Classical description of particle interacting with nonabelian gauge ﬁeld, Phys. Rev. D 15 (1977) 2308 [INSPIRE]. [20] A. Barducci, R. Casalbuoni and L. Lusanna, Classical scalar and spinning particles interacting with external Yang-Mills ﬁelds, Nucl. Phys. B 124 (1977) 93 [INSPIRE]. [21] E. D’Hoker and D.G. Gagne, Worldline path integrals for fermions with general couplings, Nucl. Phys. B 467 (1996) 297 [hep-th/9512080] [INSPIRE]. [22] F. Bastianelli, P. Benincasa and S. Giombi, Worldline approach to vector and antisymmetric tensor ﬁelds, JHEP 04 (2005) 010 [hep-th/0503155] [INSPIRE]. – 28 – [23] F. Bastianelli and R. Bonezzi, Quantum theory of massless (p, 0)-forms, JHEP 09 (2011) 018 [arXiv:1107.3661] [INSPIRE]. [24] L. Brink, P. Di Vecchia and P.S. Howe, A Lagrangian formulation of the classical and quantum dynamics of spinning particles, Nucl. Phys. B 118 (1977) 76 [INSPIRE]. [25] L. Brink, S. Deser, B. Zumino, P. Di Vecchia and P.S. Howe, Local supersymmetry for spinning particles, Phys. Lett. B 64 (1976) 435 [Erratum ibid. B 68 (1977) 488] [INSPIRE]. [26] F. Bastianelli, O. Corradini and A. Waldron, Detours and paths: BRST complexes and worldline formalism, JHEP 05 (2009) 017 [arXiv:0902.0530] [INSPIRE]. [27] P. Howe, S. Penati, M. Pernici and P.K. Townsend, A particle mechanics description of antisymmetric tensor ﬁelds, Class. Quant. Grav. 6 (1989) 1125. JHEP05(2016)056 [28] H. Georgi, Lie algebras in particle physics, Frontiers in Physics, Perseus Books, Reading U.S.A. (1999). [29] J.P. Edwards and P. Mansﬁeld, Delta-function interactions for the bosonic and spinning strings and the generation of abelian gauge theory, JHEP 01 (2015) 127 [arXiv:1410.3288] [INSPIRE]. [30] J.P. Edwards and P. Mansﬁeld, QED as the tensionless limit of the spinning string with contact interaction, Phys. Lett. B 746 (2015) 335 [arXiv:1409.4948] [INSPIRE]. – 29 –
https://openalex.org/W3201486201	https://europepmc.org/articles/pmc8469791?pdf=render	English	null	Children Learn, Children Do! Results of the “Planning Health in School”, a Behavioural Change Programme	International journal of environmental research and public health/International journal of environmental research and public health	2,021	cc-by	14,330	Citation: Vieira, M.; Carvalho, G.S. Children Learn, Children Do! Results of the “Planning Health in School”, a Behavioural Change Programme. Int. J. Environ. Res. Public Health 2021, 18, 9872. https://doi.org/10.3390/ ijerph18189872 Keywords: health promotion; health education; Transtheoretical model; eating behaviour; obesity prevention; children’s health Academic Editors: Immacolata Cristina Nettore, Raffaele Teperino and Paolo Emidio Macchia International Journal of Environmental Research and Public Health International Journal of Environmental Research and Public Health Margarida Vieira * and Graça S. Carvalho Margarida Vieira * Margarida Vieira * and Graça S. Carvalho Research Centre on Child Studies, University of Minho, 4710-057 Braga, Portugal; graca@ie.uminho.pt * Correspondence: m.margarida.vieira@gmail.com; Tel.: +351-253601212 Abstract: The ‘Planning Health in School’ programme (PHS-pro) is a behavioural change intervention to assess and improve the eating habits of children, particularly the intake of fruit and vegetables, and to guide them towards healthy choices. The programme and its educational components are based on the Transtheoretical Model of stages of change to integrate nutritional literacy and build up problem- solving and decision-making skills. Children (n = 240, ages 10–12) of one large suburban school in Porto’s metropolitan area (Portugal) were evaluated throughout PHS-pro implementation during one school year in a repeated time–series design. Children’s outcome evaluations were conducted through seven 3-day food records for nine eating behaviour, documented after each learning module and through participatory activities which analysed attitudes, preferences and expectations. Changes were observed in children’s eating behaviour, supported by changes in motivation as perceived in their attitudes and expectations. Signiﬁcant changes were found in a higher consumption of vegetable soup (p = 0.003), milk products (p = 0.024), and fruit (p = 0.008), while the consumption of high-energy dense food (p = 0.048) and soft drinks (p = 0.042) signiﬁcantly decreased. No positive effects on fried food, water, vegetables and bread consumption were found. The PHS-pro intervention proved to be effective in developing healthy eating behaviour in young people. Keywords: health promotion; health education; Transtheoretical model; eating behaviour; obesity prevention; children’s health 1. Introduction A healthy diet is one of the key elements to ensure proper growth during childhood [1,2]. However, at present, it is abundantly evident that most children have unhealthy dietary habits, mainly because they do not eat fruit and vegetables (F&V) in sufﬁcient quantities, while consuming excessively high-fat snacks and high-sugar foods and beverages [3–6]. Received: 7 August 2021 Accepted: 17 September 2021 Published: 19 September 2021 g y g g g g The scientiﬁc literature warns of the negative impact of such eating patterns, particu- larly low F&V intakes, which are associated with obesity and chronic severe diseases, such as diabetes, cardiovascular disease and cancer [7–9]. Additionally, eating sweet candy and cake, as well as drinking soft drinks regularly, are identiﬁed as risk factors for obesity and early childhood caries, two related health problems affecting children worldwide [10]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. A primary consideration is to encourage children to eat healthily to ensure the ad- equate dietary intake of central nutrients and micronutrients, based on a wide diversity of food sources, wherein F&V have a large share [11]. A healthy diet can provide all the conditions for optimum growth during childhood and adolescence and prevent all forms of poor nutrition to move away from the main risk factors of the obesity epidemic and related chronic diseases [12]. For this reason, helping children in the changing process for better eating behaviour is now a much-needed measure, and school-based health promo- tion programmes are clearly identiﬁed as the most effective strategy to promote healthy behaviour among young people [13,14]. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Schools have been identiﬁed as the best settings to implement interventions to educate young people and promote healthy habits [15]. The strong argument for health interven- tions in schools is that children can be reached there with relatively less effort and ensured Int. J. Environ. Res. Public Health 2021, 18, 9872. https://doi.org/10.3390/ijerph18189872 https://www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2021, 18, 9872 2 of 19 global coverage. 1. Introduction There are, however, more advantages: (i) children spend the larger part of the day at school, and for about nine months per year, reﬂecting the importance of school in children’s lives; (ii) children usually eat at least one meal and two snacks in the ﬁve weekdays, which can inﬂuence the eating patterns and the way of living; (iii) the school allows for the combining education and health, and these two components can interact and be cooperative for learning and promoting health, but also have effective results such as health knowledge and skills improvement; and (iv) the school can provide the opportunity for children to be physically active and eat healthy foods [16–18]. A systematic review [19] assessing 55 international studies reporting health promotion interventions for preventing obesity in children found strong evidence to support beneﬁcial effects on children’s health, particularly programmes targeted at children aged 6 to 12 years. Of the 55 examined studies, positive changes were found in children’s eating behaviour in 20 studies regarding nutrition knowledge, eating practices, F&V consumption, energy- dense snack foods, sweetened/carbonated drinks, sweet foods and other indicators of better diets [19]. More recently, another systematic review [20] restricted to randomised controlled trials (RCTs) included 153 RCTs of the programmes which aimed to prevent obesity in children aged from 0 to 18 years. Of the 153 RCTs, 45 programmes targeted children aged 6 to 12 years, showing some evidence that diet combined with physical activity interventions may be effective in obesity prevention and reinforcing that a healthy diet and a physically active lifestyle have many health beneﬁts beyond the promotion of a healthy weight. y g Bearing this in mind, a school-based prevention obesity intervention called the ‘Plan- ning Health in School’ programme (PHS-pro) was designed, implemented and evaluated on children attending school grade 6, aged 10 to 12 years old. The PHS-pro aimed to improve children’s eating behaviour and active living with the implementation of a nu- tritional education programme shaped in eight learning modules and delivered monthly over a full school year. y The PHS-pro provided easy-to-use food and nutrition knowledge for healthy choices, increasing skills for better decision-making and nutrition literacy, and was shown to be effective in improving children’s nutritional statuses by several anthropometric mea- sures [21]. This study aims to determine the impact of the eight learning modules on children’s eating behaviour and lifestyle over a 1-year school intervention. 1. Introduction Additionally, it provides detailed information about how children’s behaviour and attitudes improved and under which circumstances. The effectiveness of the programme was evaluated according to the selected key behaviour goals of the learning modules, and children’s eating behaviour changes were assessed using a 3-day food record following each learning module. 2. Materials and Methods 2.1. Study Design The PHS-pro learning modules were implemented three weeks a baseline data collection and evaluations were conducted at monthly intervals e two distinct periods: in December, the Christmas month, to avoid the effects of the season which can cause changes in the usual diet; and in May after the last l The Scientiﬁc Council of the Institute of Education of the University of Minho ap- proved this study, and ethical permission was obtained from the Pedagogical School Board. Furthermore, teachers’ class coordinators and the Health Promoter Ofﬁce (HPO) of the intervention school made themselves available to offer assistance in developing the research activities. season which can cause changes in the usual diet; and in May, after the last l module since the evaluation period overlapped with the last days of school, m impossible to control the return of the usual data collection. The Scientific Council of the Institute of Education of the University of approved this study, and ethical permission was obtained from the Pedagogica Board. Furthermore, teachers’ class coordinators and the Health Promoter Office of the intervention school made themselves available to offer assistance in develop research activities. To initiate the PHS-pro, several arrangements were carried out to evaluate th setting environment: the food infrastructures of the school with respect to the ba vending machines, cafeteria and the food menus; nutrition and health-related curriculum; physical education classes; and health prevention initiatives. Aside f celebration of World Food Day, the school did not have any initiatives or aw campaigns for healthy behaviour. Thus, a partnership was established with the To initiate the PHS-pro, several arrangements were carried out to evaluate the school setting environment: the food infrastructures of the school with respect to the bar/buffet, vending machines, cafeteria and the food menus; nutrition and health-related school curriculum; physical education classes; and health prevention initiatives. Aside from the celebration of World Food Day, the school did not have any initiatives or awareness campaigns for healthy behaviour. Thus, a partnership was established with the HPO of the intervention school, enabling the inclusion of the PHS-pro in the Yearly Activity Plan of the school. 2. Materials and Methods 2.1. Study Design In order to achieve the objectives of this longitudinal study, the research design was based on a repeated time–series design. The time–series strategy consisted of taking a series of initial observations and introducing a variable or a new dynamic into the research ﬁeld, after which another series of observations were made and in which only one group was available [22,23]. Hence, a baseline data collection was conducted before the intervention, followed by a series of repeated measures to determine the inﬂuence of the independent variable (the PHS-pro) to provide evidence about its effects on children’s eating behaviour (see Figure 1). g The study period extended from September 2011 to May 2012, and the data of the present study corresponded to the implementation of the programme content over one school year. The PHS-pro learning modules were implemented three weeks after the baseline data collection and evaluations were conducted at monthly intervals except in two distinct periods: in December, the Christmas month, to avoid the effects of the festive season which can cause changes in the usual diet; and in May, after the last learning module Int. J. Environ. Res. Public Health 2021, 18, 9872 3 of 19 3 of 19 since the evaluation period overlapped with the last days of school, making it impossible to control the return of the usual data collection. on. Res. Public Health 2021, 18, since the evaluation period overlapped with the last days of school, making it impossible to control the return of the usual data collection. n. Res. Public Health 2021, 18, since the evaluation period overlapped with the last days of school, making it impossible to control the return of the usual data collection. n. Res. Public Health 2021, 18, since the evaluation period overlapped with the last days of school, making it impossible to control the return of the usual data collection. o e u i ea , , Figure 1. Research design and analysed variables according the learning modules’ goals. Figure 1. Research design and analysed variables according the learning modules’ goals. Figure 1. Research design and analysed variables according the learning modules’ goals. Figure 1. Research design and analysed variables according the learning modules’ goals. The study period extended from September 2011 to May 2012, and the dat present study corresponded to the implementation of the programme content o school year. 2. Materials and Methods 2.1. Study Design In addition, the HPO coordinator conducted all the arrangements required for implementing this intervention in the school setting, connecting the PHS-pro with several stakeholders: class coordinator teachers, grade 6 teachers, school principals, the school parents’ association and the school canteen, in order to guarantee the availability of F&V at lunch with an extra salad buffet. ca paig s o ea y be a iou us, a pa e s ip as es ab is e i e the intervention school, enabling the inclusion of the PHS-pro in the Yearly Activ of the school. In addition, the HPO coordinator conducted all the arrangements r for implementing this intervention in the school setting, connecting the PHS-p several stakeholders: class coordinator teachers, grade 6 teachers, school princip school parents’ association and the school canteen, in order to guarantee the ava of F&V at lunch with an extra salad buffet. Parents were invited to participate in the longitudinal study and were fully in of its content and objecti es All parents signed the informed consent form A Parents were invited to participate in the longitudinal study and were fully informed of its content and objectives. All parents signed the informed consent form. A similar procedure was delivered to children in the classroom on the ﬁrst day of school. Detailed information was ﬁrst conveyed orally, then through an individual distribution of a paper package that included participants’ information sheets to explain the research purpose and the child’s informed consent to enable free and active participation, beyond the formal request previously addressed to parents. 2.3. Intervention The implementation of the programme was designed upon the Transtheoretical Model (TTM) to encourage eating behaviour changes and engage children to participate actively in this process of change [24]. The TTM provides an appropriate framework to build up the educational content of PHS-pro, expecting that children would progress through the ﬁve stages identiﬁed by Prochaska and his collaborators [24,25]: (i) precontemplation—no intention to change be- haviour or take action in the near future, unawareness of their problems; (ii) contemplation— aware of unhealthy habits, interest and intention to change in the next 6 months; (iii) preparation—intending to take action in the next 30 days; (iv) action—making changes and visible modiﬁcations on speciﬁc behaviour targets along 6 months; (v) maintenance— working to continue and consolidate healthy habits. In addition, the core constructs of TTM (processes of change, decision balance, self-efﬁcacy) were applied for supporting progress change across stages and facilitate behaviour’ change over the PHS-pro [26,27]. Based on this, the intervention consisted of eight learning modules, which followed the ﬁve-stages of TTM readiness, for moving children from inaction to action or to maintenance, in order to a successful behaviour change. The process of behaviour change was assessed through activities developed over the learning modules, and the progress on dietary behaviour was monitored through the application of a 3-day food record after each module, which provided children’s feedback. The intervention consisted of eight learning modules focused on the four main goals of the programme, which were based on the international guidelines of the WHO [28]: adequate consumption of ﬁve servings F&V/day; decreasing high-sugar food and beverage intake (to 10% of free sugar of total daily energy); decreasing high-fat and energy-dense food consumption; 1 h of physical activity and no more than 2 h spent watch TV. p p y y p The children chose eight topics about food, eating and active living out of a list of 16 to be used for the learning modules, following the participatory methodology principles [29,30]. Each learning module (LM) emphasized a different topic with a speciﬁc behaviour change, and was designed to last 45 min with one scheduled per month as part of the Natural Science classes over the academic year. 2.2. Participants information was first conveyed orally, then through an individual distribution of package that included participants’ information sheets to explain the research All grade 6 children (n = 240, ages 10–12) of one large suburban school in Porto’s metropolitan area (Portugal) participated in this study. Of these, 11 (4.6%) were excluded from the study: one refused to participate, two moved to another school and eight were special educational needs cases. Consequently, the sample size at the onset of this study Int. J. Environ. Res. Public Health 2021, 18, 9872 4 of 19 was 229 children. However, at the end of the study, the sample was reduced to 157 children (with 87 girls and 70 boys) due to failure to meet the following criteria: (i) having complete baseline data evaluation (corresponding to the food record of zero); (ii) having returned two or more complete food records over the intervention; (iii) having parents’ informed consent and their informed consent for participation. 2.3. Intervention The learning modules sequence was as follows: LM1 (“10 steps to be healthier”); LM2 (“Water & milk help you to grow up”); LM3 (“Training every day to be healthier”); LM4 (“3 fruits a day, how much good it does?”); LM5 (“F&V are essential to life”); LM6 (“Start on moving!”); LM7 (“The best snacks”); LM8 (“Final game: who has learned about everything?”—a programme overview). The learning modules goals were to provide nutritional education, identify obstacles and training problem-solving skills, to support children to ﬁnd solutions and strategies, and motivate them to set a personal behaviour change goal and implement it in daily life. All the modules were conducted to follow the TTM stages of change and the processes of change. Additionally, in each LM, a participatory activity was included for supporting children to enhance skills and make decisions to encourage healthier behaviour. This kind of activity enabled children to participate, as well as to receive feedback about speciﬁc behaviour change goals. Table 1 shows the eight modules, describing the scope and the sequence of activities (column 1), the goals (column 2), the behavioural change goals and the expected outcomes in the food records (column 3). Int. J. Environ. Res. Public Health 2021, 18, 9872 5 of 19 Table 1. Learning modules of the intervention and related educational and behaviour change goals. Educational Components Educational Goals Behaviour Change Goals and Expected Outcomes in the Food Record ule-one (LM1): “10 steps to be healthier” nciples of healthy eating presentation in a simple 10 stepwise format p discussion to explore obstacles, solutions/options to meet the ps A1): Choosing the step, changing for better e most difﬁcult step between the 10 wise steps and one step to ehaviour change goal. To reinforce healthy eating and active living; To recognize short and long-term beneﬁts; To enhance motivation for changing behaviour; To identify main obstacles for a healthy eating; To encourage daily attitudes for positive changes. Choosing a personal behaviour change goal Setting one practical measure to implement on daily behaviour Children’s commitment to accomplish the goal in the next days of the learning module Expected outcomes in Food Record 1 (FR1) - To ﬁnd behaviour changes regarding the children’s selection steps: step 2 (starting lunch and dinner by eating vegetable soup), and step 7 (avoiding fried foods). Setting a speciﬁc daily fruit serving goal to be accomplished before the next module. Expected outcomes in Food Record 4 (FR4): Increasing fruit consumption. 2.3. Intervention ule-two (LM2): Water & milk help you to grow up ms were identiﬁed (water, milk and yogurt) and their beneﬁts for a growth were described the pros and cons of beverages usually consumed: nutrients, prices, es and comparative nutritional contents A2): Favourite beverage of the month tion to be achieved in the group Calling for consumption of healthy choices: water, milk and yogurt, which contribute to a proper growth; Identiﬁcation of beneﬁts and adequate portions; Recommendations to decrease/moderate consumption of sugar-sweetened soft drinks; Children resolution: adopting water or milk instead of soft drinks Expected outcomes in Food Record 2 (FR2): - Increasing water consumption and milk, and decreasing of sugar-sweetened soft drinks. ule-three (LM3): Training everyday to be healthier ing the concept of health: what it is to be healthy and its relationship beneﬁts of health promotion and the inﬂuence of nutrition and oices. scussion around concepts of health (A3): Planning for a healthier family trategy for changes at home to a healthier family. A card was quest each child to pay attention to their family’s lifestyle during the days. Raising awareness of being healthy by giving priority to health choices and behaviour; Identifying unhealthy behaviour in the family’s lifestyle; Finding practical measures to implement a healthy environment at home. Children were invited to design what would they consider to change their families’ behaviour to have a healthier life. No Food Record due to the Christmas season. ule-four (LM4): 3 fruits a day, how much good it does? uit intake beneﬁts mendation intake: 3 portions of fruit per day tration of ways of eating fruit the fruits’ top list A4): My Fruit planning: take a step forward into the future e fruit portion intakes from the day before, the present day and hildren prefer to eat on the day after or even if they would prefer to e. Focusing on two key-points: the great variety of fruits available, and to respect each other’s preferences; Understanding differences and characteristics of different fruits; Identifying obstacles to fruit consumption, opportunities to increase the daily intake of fruit, and ways to add fruit to the diet; To raise awareness for daily fruit recommendations: compare the usual consumptions to increase children perception on their consumption. Setting a speciﬁc daily fruit serving goal to be accomplished before the next module. Expected outcomes in Food Record 4 (FR4): Increasing fruit consumption. Educational Goals Calling for the consumption of fruits and vegetables in the daily diet; Recognizing the existing plethora of vegetables (legumes, roots and tubers, pulses); Showing different ways of eating vegetables (soups, salads, and ways of preparation); Encouragment for choosing the favourite ones; Making children understand that they can have preferences for vegetables in the big vegetables family and can eat their favourite vegetables to reach a healthy goal. Calling for the consumption of fruits and vegetables in the daily diet; Recognizing the existing plethora of vegetables (legumes, roots and tubers, pulses); Showing different ways of eating vegetables (soups, salads, and ways of preparation); Choosing a personal behaviour change goal. Setting one practical measure to implement on daily behaviour. Children’s commitment to accomplish the goal before the next learning module. g ( p y p p ) Encouragment for choosing the favourite ones; Making children understand that they can have preferences for vegetables in the big vegetables family and can eat their favourite vegetables to reach a healthy goal. Expected outcomes in Food Record 5 (FR5): Increasing intake of vegetable soup, vegetables as legumes or salads at meals, and increasing fruit consumption. Behaviour Change Goals and Expected Outcomes in the Food Record Activity-seven (A7): Small meals big impact - Recording which is the best snack (card 1) and after the LM7, what a snack must have to be healthy (card 2). - Understanding the difference between which is considered the best snack and the best choice for a healthy snack. Calling for the consumption of healthy foods during snack-breaks; Reinforcing skills that could help children become aware of their autonomy in food choices and eating behaviour (individual decision is exclusively the option of each child); Guiding how to choose a healthy snack in various aspects: nutrients, preferences, prices and new options; Encouraging one speciﬁc eating habit to change during breaks and to improve nutritional snacks: breaks are an opportunity to add fruit. Choosing a personal behaviour change goal fo alternatives to snack breaks. Implementing a snack rich in essential nutrien Children’s commitment to accomplish the goa next module. Expected outcomes in Food Record 7 (FR7): Ev nutritional quality of food choices of snacks: in bread, yogurt and milk consumptions, and de drinks, high-energy dense food (croissants, pa cakes, pastries, cookies and biscuits). Learning module-eight (LM8): The ﬁnal game: who has learned about everything? (i) Closure of the intervention (ii) Children were invited to play a little game. This activity allowed children’s self-evaluation towards their behaviour changes to their eating habits over the programme. Activity-eight (A8): The ﬁnal game: who has learned about everything? Game—‘let’s play on: who has learned about everything?’ Recalling and re-examining concepts and self-assessment regarding changes over the intervention; Reinforcing skills for healthy eating and active living; To enhance motivation for changing behaviour; To encourage daily attitudes towards positive changes. No application of a food record. Application of a small questionnaire to ask ch they could identify any changes that occurred and lifestyle habits over the programme. Educational Components 2.3. Intervention modules of the intervention and related educational and behaviour change goals. Behaviour Change Goals and Expected Outcomes in the Food Record Children were invited to design what would they consider to change their families’ behaviour to have a healthier life. No Food Record due to the Christmas season. Children were invited to design what would they consider to change their families’ behaviour to have a healthier life. No Food Record due to the Christmas season. Raising awareness of being healthy by giving priority to health choices and behaviour; Identifying unhealthy behaviour in the family’s lifestyle; Finding practical measures to implement a healthy environment at home. Focusing on two key-points: the great variety of fruits available, and to respect each other’s preferences; Understanding differences and characteristics of different fruits; Identifying obstacles to fruit consumption, opportunities to increase the daily intake of fruit, and ways to add fruit to the diet; To raise awareness for daily fruit recommendations: compare the usual consumptions to increase children perception on their consumption. Int. J. Environ. Res. Public Health 2021, 18, 9872 6 of 19 Table 1. Cont. Behaviour Change Goals and Expected Outcomes in the Food Record Behaviour Change Goals and Expected Outcomes in the Food Record Educational Components Educational Goals Behaviour Change Goals and Expected Outcomes in the Food R Learning module-ﬁve (LM5): Fruits & vegetables are essential to life (i) The nutritional role of fruits and vegetables in the daily diet: beneﬁts of nutrients, nutritional value, and richness in ﬁbre. Activity-ﬁve (A5): The most popular ones: soup, salad and side dishes - Recording the favourite vegetables and ranking the most popular soup, salad and side dishes. The intention was to make children understand that they can have preferences for vegetables in the big vegetables family and can eat their favourite vegetables to reach a healthy goal. Calling for the consumption of fruits and vegetables in the daily diet; Recognizing the existing plethora of vegetables (legumes, roots and tubers, pulses); Showing different ways of eating vegetables (soups, salads, and ways of preparation); Encouragment for choosing the favourite ones; Making children understand that they can have preferences for vegetables in the big vegetables family and can eat their favourite vegetables to reach a healthy goal. Choosing a personal behaviour change goal. Setting one practical measure to implement on Children’s commitment to accomplish the goa learning module. Expected outcomes in Food Record 5 (FR5): In vegetable soup, vegetables as legumes or salad increasing fruit consumption. Learning module-six (LM6): Start on moving! (i) The importance of the regular practice of physical activity (PA) (ii) In-group discussion to explore obstacles which blocked the daily habit of practicing PA (iii) Explore solutions/options to try new sports Activity-six (A6): What sort of things do you play on? Recording the sports or playful activities usually practised by children and activities they would like to try. Enhancing the health beneﬁts of PA for self-image; Raising awareness to try new sports; Reinforcing physical activity classes of extracurricular programs at school. Choosing a personal behaviour change goal to of a sport or other forms of PA. No Food Record evaluation. Learning module-seven (LM7): The best snacks (i) Food items that might be included as a healthy snack (ii) Demonstration of ways to change the quality of snacks, and how to conﬁrm this using label readings (iii) In-group discussion to identify the most consumed foods during snack-breaks and arguments of what a snack must have to be healthy, that it must be rich in essential nutrients. Learning module-six (LM6): Start on moving! Enhancing the health beneﬁts of PA for self-image; Raising awareness to try new sports; Reinforcing physical activity classes of extracurricular programs at school. Enhancing the health beneﬁts of PA for self-image; Raising awareness to try new sports; Reinforcing physical activity classes of extracurricular programs at school. Choosing a personal behaviour change goal to begin the practice of a sport or other forms of PA. No Food Record evaluation. Choosing a personal behaviour change goal for healthy alternatives to snack breaks. Choosing a personal behaviour change goal for healthy alternatives to snack breaks. Implementing a snack rich in essential nutrients. Children’s commitment to accomplish the goal before the next module. Calling for the consumption of healthy foods during snack-breaks; Reinforcing skills that could help children become aware of their autonomy in food choices and eating behaviour (individual decision is exclusively the option of each child); Guiding how to choose a healthy snack in various aspects: nutrients, preferences, prices and new options; Encouraging one speciﬁc eating habit to change during breaks and to improve nutritional snacks: breaks are an opportunity to add fruit. Implementing a snack rich in essential nutrients. Children’s commitment to accomplish the goal before the next module. Expected outcomes in Food Record 7 (FR7): Evolution in the nutritional quality of food choices of snacks: increasing fruit, bread, yogurt and milk consumptions, and decreasing soft drinks, high-energy dense food (croissants, panikes, donuts, cakes, pastries, cookies and biscuits). No application of a food record. Application of a small questionnaire to ask children whether they could identify any changes that occurred in their eating and lifestyle habits over the programme. Recalling and re-examining concepts and self-assessment regarding changes over the intervention; Reinforcing skills for healthy eating and active living; To enhance motivation for changing behaviour; To encourage daily attitudes towards positive changes. 7 of 19 Int. J. Environ. Res. Public Health 2021, 18, 9872 2.4. Outcome Measures To measure the eating behaviour of children over the implementation of the ed- ucational components and to evaluate behaviour changes after each LM and over the intervention, a 3-day food record was selected to assess all the food and beverages that children consumed over 3 days. The 3-day food record was considered the most accurate method, both in qualitative and quantitative terms, to describe the food consumed [31,32]. Nevertheless, measuring food intake among free-living individuals continued to be a challenging task [33]. Seven 3-day food records were applied to assess the intervention efﬁcacy. The ﬁrst collected data, for the baseline, were obtained before the intervention and were designated food record zero (FR0), followed by six more 3-day food records applied subsequently to each learning module (FR1, FR2, FR4, FR5, FR6 and FR7). In other words, after each learning module children had to record within the following days the respective 3-day food record. The exceptions for returning the food record were for the LM3, corresponding to the Christmas season, as there were always changes in eating habits during Christmas, with typical food consumptions which varied considerably compared to the usual pattern, and for the LM8 in the last days of the academic year. Instead, a small questionnaire was applied to children to identify any changes which occurred in their eating and lifestyle behaviour over the intervention. Likewise, the LM6 topic concerned sports practice and active living and was not a food topic. Accordingly, the food record (FR6) was applied, yet was also used as a possible follow-up evaluation. The 3-day food record form was introduced to children in the classroom with detailed instructions on how to ﬁll it out correctly and the delivered rules were deﬁned. The in- structions of the food record form provided representative pictures of household measures (cups, spoons or standard packaging) to elucidate the deﬁnitions of portion sizes and to serve as an example of the completed records. This helped children to record all foods and beverages consumed during three consecutive and subsequent days after each LM. We encouraged the children to record events at the time of eating and to use their best judgment in the reports. In the current study, only children who returned the completed FR0 (baseline data) plus at least two or more completed food records were reviewed and validated by the nutritionist and considered for analysis. 2.4.1. Measuring Eating Behaviour Changes 2.4.1. Measuring Eating Behaviour Changes The LM5 (“Fruits & vegetables are essential to life”, see Table 1) was also dedicated to the vegetable topic, in which vegetable soup was considered an important preparation for increasing vegetable intake. Consequently, this variable was assessed by comparing FR0 (baseline data) with FR1 and FR5. (2) Fried food/High-fat food servings: this included all fried foods and products, such as French fries, chips, rissoles, pasties, croquettes, ﬁsh ﬁngers, and nuggets. In LM1 all children chose to take action and go up to step 7, which corresponded to avoiding eating fried foods. This variable was assessed by comparing FR0 with FR1. (2) Fried food/High-fat food servings: this included all fried foods and products, such as French fries, chips, rissoles, pasties, croquettes, ﬁsh ﬁngers, and nuggets. In LM1 all children chose to take action and go up to step 7, which corresponded to avoiding eating fried foods. This variable was assessed by comparing FR0 with FR1. (3) Water servings: this included tap water and bottled water. Improving water consumption as the ﬁrst choice of beverage was developed in LM2 (“Water & milk help you to grow up”, see Table 1). This variable was assessed by comparing FR0 with FR2. y g p y p g (4) Milk products servings: this included plain milk, milk mixed with coffee or barley, and yoghurt, and excluded chocolate milk. LM2 was dedicated to the healthy and adequate consumption of milk products. Additionally, LM7 (“The best snacks”, see Table 1) focused on milk products as healthy choices for smaller snacks. This variable was assessed by comparing FR0 with FR2 and FR7. p g (5) Soft drinks servings: this included soda, cola, iced tea, and added-sugar squash juices. In LM2 the soft drinks’ topic was developed to promote a decrease in children’s consumption of soft drinks. Again, the LM7 strengthened the same idea of avoiding and limiting these beverages on snacks. This variable was assessed by comparing FR0 with FR2 and FR7. (6) Fruit servings: this included all fruits and fresh fruit juices, except fruit in syrup. LM4 was exclusively dedicated to the fruit topic (see Table 1). Then, it was further strength- ened in LM5, together with vegetables, and in LM7 as one of the best choices for children’s snacks. Therefore, the fruit variable was assessed by comparing FR0 with FR4, FR5 and FR7. 2.4.1. Measuring Eating Behaviour Changes The 3-day food records data were coded following the behavioural coding proto- col [34]. Given that people consume foods, not nutrients, the PHS-pro had a behavioural focus instead of a knowledge-based focus. Therefore, to guide children towards desirable nutrition-oriented behaviour changes and healthy food choices, rather than nutrients [35], the ﬁrst step was to categorise all food items reported by the children into a behavioural perspective to assess the eating behaviour changes outcomes. All food items reported in the seven 3-day food records were coded by meal and place of consumption, food category, cooking method, and the number of servings (one serving was the default serving and the usual amount of food recorded by children). Each food item was categorised in terms of providing central nutrients (proteins, vitamins, minerals, dietary ﬁbre, complex carbohydrates, fats, free sugars) based on the six food categories of the New Portuguese Food Guide [36]: (i) fruits; (ii) vegetables; (iii) potato, cereal, and cereal products; (iv) pulses, milk and dairy products; (v) meat, ﬁsh, seafood, and eggs; (vi) fats and oils. Subcategories were created within each of these groups, except for vegetables and fruits, in a total of 21 items. Food items were identiﬁed within each subcategory as low-, medium- and high-fat options or low-, medium- and high-sugar options. Two examples of these were: French fries and potato chips, which were credited as high–fat options, compared to boiled potatoes that were credited as low fat, and donuts or cakes, which were credited as high-sugar foods compared to fresh bread. Int. J. Environ. Res. Public Health 2021, 18, 9872 8 of 19 Additionally, typical Portuguese mixed dishes had to provide at least one cup of cooked vegetables or legumes to be credited as containing one serving of vegetables, for example: ‘chicken and vegetable stew’. g After these procedures, to assess the servings of speciﬁc foods and beverages con- sumed, and to measure behaviour changes, smaller eating sub-behaviour categories were created according to the learning modules goals, and nine study variables were organized for analysis: y (1) Vegetable soup servings: this included all traditional Portuguese preparations of vegetable soups. In LM1 (“10 steps to be healthier”, see Table 1), all children chose to take action and go up to step 2, which corresponded to an increase in their vegetable soup consumption at meal times and an increased intake of vegetables. 2.4.1. Measuring Eating Behaviour Changes (7) Vegetable servings: this included all vegetables, legumes, salads, greenery, and mixed dishes rich in legumes or vegetables, and excluded potatoes. LM5 was entirely dedicated to the vegetable topic. This variable was assessed by comparing FR0 with FR5. (8) Bread servings: this included all varieties of fresh and toasted bread. LM7 explored the best nutritional options for children’s snacks, in which bread had a healthy place. This variable was assessed by comparing FR0 with FR7. (9) High-energy dense food servings: this included croissants, panikes, donuts, cakes, pastries, cookies and biscuits. LM7 explored the best nutritional options for children’s snacks compared with other food choices loaded with fat and sugar. This variable was assessed by comparing FR0 with FR7. y g The variables were limited to the behaviour of these nine foods, which were particu- larly relevant to the learning modules goals (see Table 1) and to the overall objectives of the PHS-pro. Figure 1 provides a summary of the analysed variables according to the learning modules goals. 2.4.2. Measuring Attitudes, Preferences and Expectations Participatory activities were implemented to support peer-led activities and children’s decision-making over the learning modules. Such activities were also used to collect Int. J. Environ. Res. Public Health 2021, 18, 9872 9 of 19 9 of 19 data related to children’s attitudes, preferences and expectations. Different activities were developed for each of the eight learning modules to support children in changing their attitudes and behaviour towards the learning modules goals (see Table 1). The feedback returned by the children in the classroom was then subjected to con- tent analysis. y The eight participatory activities are described below. Activity one (A1): Choosing the step, changing for the better. Children were asked which of the 10 wise steps presented in LM1 was the most difﬁcult. As mentioned above, two steps were equally chosen: step 2, increasing vegetable soup servings on meals; and step 7, related to avoiding fried products. The children were also asked to elect the step that they found most prepared them for immediate behaviour change. Children’s commitments were based on the improvements of their daily behaviour for the same two steps, 2 and 7. The children’s responses were given orally and recorded on the feedback activity sheet. Children’s attitudes towards change behaviour were assessed through this activity, and feedback was evaluated in connection with the FR1. g y Activity two (A2): Favourite beverage of the month. g y Activity two (A2): Favourite beverage of the month. From the discussion about the pros and cons of beverages usually consumed by children, it was proposed that children chose the favourite beverage of the month, and the result of this was: adopting water or milk instead of soft drinks. The children’s responses were recorded on the feedback activity sheet. Children’s attitudes towards change behaviour were assessed throughout this activity, and feedback was evaluated in connection with the FR2. Activity three (A3): Planning for a healthier family. Activity three (A3): Planning for a healthier family. Children were invited to analyse their families’ lifestyles and design what they con- sidered to be of the highest importance for changing their home environment to develop a healthier family. A card was individually delivered to each child at the end of this activity, which asked them to return it after the Christmas holidays. This activity allowed analysing whether children could identify the risky eating behaviour of their own families and develop problem-solving skills. 3. Results Of the 229 grade 6 children, 157 (68.6%) completed the baseline 3-day food record (FR0) and two or more subsequent food records. Therefore, only data for these 157 children were used in the analysis to minimise the unreliability of the dietary outcome measures. The participation rates in the data collection of the other six 3-day food records applied during the intervention were 91.1% in FR1, 92.4% in FR2, 87.9% in FR4, 82.2% in FR5, 59.2% in FR6, and 80.9% in FR7. Considering that the LM6 was dedicated to the sports topic and active living, and since the FR6 had a relatively low returning with the worst children’s participation rate, the FR6 data was withdrawn from the analysis. 2.4.2. Measuring Attitudes, Preferences and Expectations p p g Activity four (A4): My fruit planning: take a step forward into the future. Children were asked to identify and record the fruit servings that they consumed the day before, during the present day, which fruits they would like to eat the next day, or even whether they would like to taste a new fruit. Three cards were delivered to each child to registering their feedback. This activity allowed for the analysis of fruit servings consumed and assessed children’s expectations of fruit serving consumption for the next day. Activity ﬁve (A5): The most popular ones: soup, salad and side dishes. Children were asked to report their food preferences regarding the choice between vegetable soups, salads or vegetable side dishes. During this activity, three cards were delivered to each child to note down their three favourite options. The preferences of vegetable options were assessed. g p Activity six (A6): What sort of sports do you play? Children were asked which sports or playful activities they usually practised outside school, in which environments they practised sport, and whether this was within a family context. Additionally, children were asked which activities they would most like to have the chance of trying. A card was individually delivered to collect children’s feedback regarding different sporting environments. The attitudes towards sports participation and playful activities were assessed, as well as children’s expectations of which sports they would like to try. y Activity seven (A7): Small meals, big impact. Snack choices that children considered to be a healthy snack were assessed with a two-step approach: before implementing the LM7 (“The best snacks”, see Table 1) and immediately after it. To each child two cards were applied for the two moments. The attitudes towards change behaviour were assessed through this activity. Activity eight (A8): The ﬁnal game: who has learned about everything? Int. J. Environ. Res. Public Health 2021, 18, 9872 10 of 19 Children were invited to play a little game called ‘let’s play on: who has learned about everything?’. The children’s responses were given orally and recorded on the feedback activity sheet by the researcher with the assistance of the teacher. This activity allowed children’s self-evaluation of their attitude and behaviour changes which occurred in their eating habits over the PHS-pro. 2.5. Statistical Analysis Descriptive statistics were calculated to describe participants’ characteristics at base- line using the mean and standard deviation for continuous variables and frequencies for categorical variables. g In this study, only children who delivered baseline data (FR0), and at least two post- intervention food records (FR1, FR2, FR4, FR5, and/or FR7) were considered for analysis. To assess the effects of the intervention of the outcome variables representing different food behaviour, an analysis of the Wilcoxon signed-rank test was performed to examine differences between pre-intervention (baseline: FR0) and post-interventions (FR1, FR2, FR4, p ( ) p ( FR5 or FR7). Results were expressed as the median and interquartile range (IQR). ) p q g ( Q ) Initial analyses were performed for the symmetrical distribution of the outcome variables by analysing the histogram. A p-value < 0.05 was accepted as statistically signiﬁcant for all analyses, and the statistical software SPSS (package version 21) was used to perform them. Children’s feedback of the activities, which developed during the learning modules, were treated as categorical variables and analysed in a descriptive way, and no statistical testing strategy was used. Furthermore, content analysis was applied to the qualitative responses of children to the open-ended questions [37]. 3.2. Changes in Food Behaviour Table 3 shows the changes in the behaviour of nine foods by comparing children’s food intakes between baseline (one month before the intervention) and the corresponding 3-day food record. Regarding the two goals of LM1: increasing vegetable soup and decreasing fried food servings, children reported in their FR1 a signiﬁcant change in vegetable soup servings by doubling the median value from 1 to 2 servings/3-day period (p = 0.003) between baseline and the FR1. For the fried food variable, there was no signiﬁcant change. Following the LM2 goals, an increase in water and milk consumption and a decrease in sugar-sweetened soft drinks in the FR2 was expected, but only milk consumption had a statistically signiﬁcant change (p = 0.024). Though the median value was equal at baseline and FR2, higher milk consumption was reported in 75 (51.7%) out of the 145 children analysed in the FR2 compared to baseline, and in 25 children (17.2%), milk quantities were equally consumed (data not shown). Soft drinks did not show signiﬁcant changes (p = 0.340) between the two moments, although there was a reduction in the ﬁrst quartile (Q1) from 1 to 0 servings over 3 days, indicating that 25% of children decreased the intake of soft drinks. A signiﬁcant increase in fruit consumption was observed (p = 0.008) following LM4. Indeed, more children ate fruit, indicated by a median value of 2 fruit servings and showing an increase in the third quartile. In fact, of the 138 children cases analysed, 67 (48.6%) reported a higher number of fruit servings at FR4 compared to baseline, and 35 children (25.4%) reported equal fruit servings. p q g Vegetable soup, fruit and vegetable servings were examined between baseline and FR5, as increasing the consumption of these foods was the main goal of the LM5. There were no signiﬁcant changes in these consumptions. However, vegetable soup preserved the positive tendency as, in the 129 cases analysed, 51 children (39.5%) reported eating more vegetable soup in the FR4 than in the baseline, and 38 (29.5%) retained an equal consumption. Likewise, children did not report changes in fruit consumption; nevertheless, the third quartile maintained the number of fruit servings observed in the previous evaluation at FR4. The vegetables topic was again the main focus of LM5, and increasing consumption was the most important goal, yet no positive changes were found in vegetable servings. 3.1. Baseline Characteristics of Participants The mean age of the children was 10.9 years (SD = 0.6), 55.4% of whom were female with no statistical differences detected between genders (χ2 = 1.84; p = 0.175). Table 2 summarises the baseline eating behaviour of children according to the nine food groups. The most common signiﬁcant dietary concern was the F&V consumption. Children reported an intake of F&V well below the recommended levels over a 3-day period. At baseline, more than one third of the children reported not eating fruit (31.2%) or vegetable soup (36.9%) and more than half did not eat any sort of vegetables (51.0%) over 3 days. Milk products had a regular intake, except for 1.9% of children who reported no consumption and 3.2% who consumed one portion over 3 days. Bread intake was also common for snacks but not for 4.5% of children who reported not eating it. High-energy dense food intake was not reported by 17.8% of children. However, more children ate at least one or two servings of these foods over the 3 days. Finally, about 39.5% of the sample reported eating fried foods once in 3 days. Int. J. Environ. Res. Public Health 2021, 18, 9872 11 of 19 Table 2. Baseline eating behaviour of children in a 3-day period (n = 157). Reported Servings 0 1 2 3 4 5 or More % of Children Fruit 31.2 17.2 17.8 12.7 7.0 14.7 Vegetable soup 36.9 26.8 12.7 10.8 5.1 7.7 Vegetables (in salad or cooked) 51.0 28.0 10.8 6.4 2.5 1.3 Milk products 1.9 1.3 6.4 13.4 15.9 61.0 Bread 4.5 12.7 14.0 24.8 19.7 24.2 High-energy dense food 17.8 25.5 27.4 19.1 4.5 5.5 Fried food 38.2 39.5 14.6 5.7 1.9 0.0 Water 14.6 21.0 18.5 14.6 12.7 18.4 Soft drinks 49.0 22.3 11.5 7.0 3.8 6.2 Table 2. Baseline eating behaviour of children in a 3-day period (n = 157). Daily water consumption was not a regular behaviour for 14.6% of children, who reported not drinking water over 3 days. However, soft drinks consumption was already established, as a current habit among young people, as one can see in more than half of children (51.0%, i.e., counting the frequencies of children who reported daily servings), reporting the intake of one or more servings over the 3 days that were analysed. 3.2. Changes in Food Behaviour On the contrary, vegetables (in a salad or cooked as a side dish) recorded relatively low consumption, where half of the children did not eat any kind of vegetables over a 3-day period at both baseline and FR5 evaluations, except for the vegetable soup that maintained the usual intake. Int. J. Environ. Res. Public Health 2021, 18, 9872 12 of 19 12 of 19 Table 3. Comparison of outcomes between baseline and post-intervention 3-day food records (FR1, FR2, FR4, FR5 and FR7). Food behaviour evaluation between baseline and FR1 Outcome Sample size Baseline 3-day FR1 Consumption (servings) Med (IQR) Med (IQR) p-value Vegetable Soup n = 143 1 (0–2) 2 (0–3) 0.003 * Fried food 1 (0–1) 1 (0–2) 0.673 Food behaviour evaluation between baseline and FR2 Outcome Sample size Baseline 3-day FR2 Consumption (servings) Med (IQR) Med (IQR) p-value Milk products n = 145 5 (4–7) 5 (4–7) 0.024 * Water 2 (1–4) 2 (0–4) 0.435 Soft drinks 1 (0–2) 0 (0–2) 0.340 Food behaviour evaluation between baseline and FR4 Outcome Sample size Baseline 3-day FR4 Consumption (servings) Med (IQR) Med (IQR) p-value Fruit n = 138 2 (0–3) 2 (0–4) 0.008 * Food behaviour evaluation between baseline and FR5 Outcome Sample size Baseline 3-day FR5 Consumption (servings) Med (IQR) Med (IQR) p-value Vegetables (in salad or cooked) n = 129 0 (0–1) 0 (0–1) 0.400 Vegetable Soup 1 (0–3) 1 (0–3) 0.093 Fruit 2 (0–3) 2 (0–4) 0.903 Food behaviour evaluation between baseline and FR7 Outcome Sample size Baseline 3-day FR7 Consumption (servings) Med (IQR) Med (IQR) p-value Bread n = 127 3 (2–5) 3 (2–4) 0.455 High-energy dense food 2 (1–3) 1 (0–2) 0.048 * Milk products 5 (4–7) 5 (3–6) 0.078 Soft drinks 1 (0–2) 0 (0–1) 0.042 * Fruit 2 (0–3) 1 (0–3) 0.176 Med (IQR): median (interquartile range). Wilcoxon Rank Test *: p < 0.050 Table 3. Comparison of outcomes between baseline and post-intervention 3-day food records (FR1, FR2, FR4, FR5 and FR7). Table 3. Comparison of outcomes between baseline and post-intervention 3-day food records (FR1, FR2, FR4, FR5 and FR7). Regarding the ﬁve food variables evaluated following the LM7 (bread, high-energy dense food, milk products, soft drinks and fruit), consumption changes were found sig- niﬁcantly different between baseline (FR = 0) and FR7 for high-energy dense food and soft drinks. 3.2. Changes in Food Behaviour After the intervention with LM7, children signiﬁcantly decreased high-energy dense food from a median of 2 to 1 servings/3-day period (p = 0.048). For soft drinks, children reduced servings of this kind of beverage, with half of the children consuming no soft drinks in a 3-day period. Bread and milk products had no changes and maintained the usual consumption. Similarly, fruit consumption had no signiﬁcant change, yet the con- sumption of fruit decreased among 58 children (45.7%) of the 127 children cases observed (data not shown). In short, regarding the nine studied variables, children improved their nutritional practices in ﬁve variables (vegetable soup, milk products, fruit, high energy-dense food and soft drinks) over the PHS-pro intervention. Int. J. Environ. Res. Public Health 2021, 18, 9872 13 of 19 13 of 19 3.3. Changes in Attitudes, Preferences and Intentions towards Healthy Food Choices 3. Changes in Attitudes, Preferences and Intentions towards Healthy Food Choices The participatory activities were introduced during the learning modules implemen- tation to promote both children’s in-group discussion and active participation. p g p p p The ﬁrst two activities (A1 and A2) allowed children to show their attitudes toward behaviour change. For example, in A1, children decided to improve two steps (instead of one as initially proposed) that were equally chosen: step 2, advocating the increase in vegetable soup intake; and step 7, advocating the avoidance of fried foods. This attitude was followed by a signiﬁcant change in vegetable soup servings found in FR1, already referred to above. However, regarding step 7, there was no change, although children had initially indicated their positive attitude towards changing. y p g g In A2, children’s results showed an increase in the intake of water consumption and milk and a decrease in the intake of sugar-sweetened soft drinks. This attitude was followed by a signiﬁcant change in servings of milk products observed in FR2, but not in water or soft drinks consumption. In the third activity (A3), children were asked which key strategy would be an effective plan for promoting a healthy family. Of the 157 children participating, 118 (75.2%) returned the card, in which each child, individually, identiﬁed the risky eating behaviour that their own family should avoid. 3.2. Changes in Food Behaviour Four children (3.4%) reported parents needing to quit smoking as the primary strategy to promote a healthier family, 97 (82.2%) children reported healthy eating at home, and 12 (10.2%) expressed physical activity as the most important for a healthy family. Five children (4.2%) did not clearly specify a strategy and merely stated, “My family needs to change habits”. Clearly, children were able to determine the key changes for a healthier family environment, and most of them expressed the need to take speciﬁc measures to change eating habits at home. p g g In the fourth activity (A4), 142 children (90.4%) returned the feedback card of fruit servings consumed and their fruit eating intentions. Table 4 shows the results of the detailed analysis. On the day before this assessment, more than 35.9% of participants reported the intake of two fruits, and almost one quarter (23.9%) the intake of three fruits; 11 children (7.7%) still did not eat fruit. For the assessment day, more than one quarter (25.4%) of children recorded one serving fruit intake; nevertheless, the time of assessments ranged over the school day, between morning and afternoon, which meant that it could not be conﬁrmed whether the same fruit intake would follow the proﬁle of the previous day. Table 4. Fruit servings consumed in two moments (previous day and assessment day), and fruit eating intentions (A4). Fruit Servings 0 1 2 3 4 % of Children Previous day 7.7 23.9 35.9 23.9 5.6 Assessment day 67.6 25.4 6.3 0.7 0 Next day intentions 0 10.6 24.6 49.3 11.3 Table 4. Fruit servings consumed in two moments (previous day and assessment day), and fruit eating intentions (A4). Table 4. Fruit servings consumed in two moments (previous day and assessment day), and fruit eating intentions (A4). Fruit Servings 0 1 2 3 4 % of Children Previous day 7.7 23.9 35.9 23.9 5.6 Assessment day 67.6 25.4 6.3 0.7 0 Next day intentions 0 10.6 24.6 49.3 11.3 Children’s intentions to eat fruit for the next day showed highly positive results, as nearly half of the children (49.3%) stated that they intended to eat three servings of fruit the next day. According to this, a great number of children (25.4%) showed the intention of improving fruit intake, and all children had a positive attitude to improve fruit consumption. 3.2. Changes in Food Behaviour In the ﬁfth activity (A5), children reported preferences of vegetables (Table 5) with the 156 children (99.4%) providing feedback, from the total sample of 157. A proportion of 78.2% of children reported having two favourite vegetable soups, green cabbage soup and carrot soup, the two most indicated by children. For fresh salads, the majority (66.7%) had at least two preferences (lettuce and tomato salads, either separated or mixed); however, ﬁve children (3.2%) expressed no appreciation for any kind of salad. Vegetable servings as a side dish received a similar acceptance, with a large percentage of children (68.6%) reporting Children’s intentions to eat fruit for the next day showed highly positive results, as nearly half of the children (49.3%) stated that they intended to eat three servings of fruit the next day. According to this, a great number of children (25.4%) showed the intention of improving fruit intake, and all children had a positive attitude to improve fruit consumption. p In the ﬁfth activity (A5), children reported preferences of vegetables (Table 5) with the 156 children (99.4%) providing feedback, from the total sample of 157. A proportion of 78.2% of children reported having two favourite vegetable soups, green cabbage soup and carrot soup, the two most indicated by children. For fresh salads, the majority (66.7%) had at least two preferences (lettuce and tomato salads, either separated or mixed); however, ﬁve children (3.2%) expressed no appreciation for any kind of salad. Vegetable servings as a side dish received a similar acceptance, with a large percentage of children (68.6%) reporting Int. J. Environ. Res. Public Health 2021, 18, 9872 14 of 19 14 of 19 two preferences: broccolis and peas. In addition, more children indicated having three, four or ﬁve preferences for this kind of vegetable preparation compared with vegetable soups or salads. Table 5. Vegetable preferences on the basis of vegetable soup, salad and vegetable side dishes (A5). Number of Preferences 0 1 2 3 4 % of Children Vegetable soups 0 17.3 78.2 3.2 0.6 Fresh salads 3.2 23.1 66.7 5.1 1.9 Vegetable side dishes 1.9 14.1 68.6 8.3 3.8 Table 5. Vegetable preferences on the basis of vegetable soup, salad and vegetable side dishes (A5). Activity six (A6) applied during the LM6 focused on active living such as physical and leisure activities. Table 6 summarises the 152 children’s reports (96.8%) from the total of 157 participants. ble 6. 3.2. Changes in Food Behaviour Frequency distribution of children practising sports or playful activities (A6). Table 6. Frequency distribution of children practising sports or playful activities (A6). Table 6. Frequency distribution of children practising sports or playful activities (A6). Practice Daily Physical Activities % of Children Nothing 56.6 Sports 43.4 Physically active at school (playground) Playing games 51.3 Walking 25.0 Football 21.1 Dancing 2.6 Physically active after school Cycling 23.7 Football 21.1 Walking 14.5 Other sports 13.2 Playing games 11.8 Nothing 11.8 Dancing 3.9 Physically active at weekend (in the family) Going for a walk 44.1 Cycling 28.9 Nothing 25.7 Swimming 1.3 Expectations of trying sports or playful activities Sports 65.8 Playful activities 31.6 Nothing 2.6 The great majority of children were not engaged in sports practices (56.6%). When asked which activities they usually dedicated their time to in the playground, playing games was the most mentioned activity (51.3%). After school, 18 children (11.8%) indicated not completing any kind of physical activity, but the rest used the time after school to play different leisure activities. Regarding weekends with the family, more than one quarter of children (25.7%) reported not completing any kind of leisure activity or sports, while 44.1% indicated “going for a walk” as the most frequent family activity. Finally, when children were asked about activities they would like to have the chance of trying, the majority showed an interest in practising some sports (65.8%) or playful activities (31.6%); between 34 activities mentioned by children, surﬁng was the most popular (with 16 occurrences), followed by tennis (12), horse-riding (11), and skating (11). These preferences showed Int. J. Environ. Res. Public Health 2021, 18, 9872 15 of 19 15 of 19 that the majority of children were motivated to engage with sports and other activities compared with the great percentage of children (56.6%) who reported not practising any kind of physical activity. In the seventh activity (A7), children were assessed regarding foods that could be part of a healthy snack. Table 7 shows the differences between the two assessment moments of A7. Of 157 children, 155 delivered the two cards applied (98.7%), and among these, a great number of children correctly identiﬁed healthy foods for snacks, with 127 (81.9%) making snack choices such as bread and milk products instead of high-energy dense foods and soft drinks. In contrast, 28 children (18.1%) selected unhealthy choices for their snacks (high-energy dense foods, soft drinks, and fried foods). 3.2. Changes in Food Behaviour Nevertheless, of the 127 children that made healthy choices, 55 (35.5%) did not include fruit as a healthy snack on the ﬁrst card assessment (A7-card 1). After the LM7, a change in children’s attitude was found in 148 children (95.5%) reported in all healthy foods, including fruit servings, against seven children (4.5%) who continued with the previous snack choices which did not include fruit. Accordingly, reported healthy snack choices improved after A7. Table 7. Frequency distribution of children’s healthy food choices for snacks (A7). Food Choices for Snacks A7-Card-One A7-Card-Two % of Children Healthy foods 81.9 95.5 Unhealthy foods 18.1 4.5 Table 7. Frequency distribution of children’s healthy food choices for snacks (A7). The last activity (A8) gathered children’s feedback about the concepts of food topics developed over the seven learning modules. Of the 157 children distributed by nine classes, four classes had difﬁculty recalling the concepts of the topic developed during the third learning module (LM3): training every day to be healthier. However, all the concepts of the other learning modules were 100% recorded, suggesting an increase in the healthy-eating knowledge with the step-by-step learning modules approach. 4. Discussion Therefore, these results make the intervention programmes for promoting vegetables appear even more important, and their continuity must be ensured as long as children are growing because these programmes might be the only effective strategy to face this perplexing trend [13]. In general, the results found in this study support the ﬁndings of similar studies for children with the same age range, as reviewed by Water et al. [19]. High levels of F&V intake were found in ﬁve intervention studies. Reductions in high energy-dense foods were reported in one study, as well as sweet foods in two studies. A decrease in soft drinks intake was reported in two studies [19]. These ﬁndings were in line with the results of this study. However, a detailed comparison with these studies is difﬁcult due to different designs and outcome measures. Additionally, no comparable Portuguese data among similar age groups are available as far as we know [41,42]. Children’s attitudes changed positively immediately following the participative ac- tivity of each learning module. In all participatory activities, children’s motivation for healthy changes was observed, which was encouraging. It was also noteworthy that there was a great agreement between the behaviour changes observed in the current study and children’s attitudes expressed in participation of the activities. Preferences and expecta- tions can be helpful information to both deepen knowledge about children’s trends and create opportunities for the continuous improvement for more effective intervention. For example, the low level in reported vegetable preferences can leave the impression that a wide variety of these foods was never tasted. Furthermore, the high level of willingness to try sports is encouraging, providing a clear sign that implementing various dynamics in the active living ﬁeld into studies is well received by children. The learning modules and the participatory activities, in which children set goals and made commitments to change a speciﬁc behaviour or engage in new behaviour, resulted in signiﬁcant eating changes. Finding out children’s needs, focusing on speciﬁc aspects of their behaviour that they were already prepared to change, and implementing practical recommendations appeared to be a successful strategy. Moreover, participatory methods provided children’s positive feedback; that they could change their daily behaviour and were responsible for their decisions. 4. Discussion This study aimed to determine the impact of eight learning modules on children’s eating behaviour regarding nine foods (fruit, vegetable soup, vegetables in a salad or cooked, milk products, bread, high-energy dense food, fried food, water and soft drinks) and their responses to one learning module dedicated to active living over the PHS- pro intervention. Results showed substantial consumption changes with a statistically signiﬁcant in- crease in vegetable soup, milk products and fruit. At the same time, high-energy dense food and soft drinks signiﬁcantly decreased between the interval of evaluation (from baseline to post-intervention learning modules). In contrast, fried food, vegetables, water and bread consumption had no signiﬁcant change. p g g These ﬁndings indicated that PHS-pro implemented over the school year successfully met some of the selected eating behaviour goals. However, it cannot be expected that, if continued over time, the programme could positively inﬂuence the other children’s eating behaviour, which might help prevent and reduce the development of obesity. The most disappointing results of the study were the lack of signiﬁcant changes in fried food and vegetable servings (in salads or cooked). This can be explained by the fact that these foods are usually served at meals not prepared by children, and such factors either facilitate or halt such behaviour since children do not have absolute control in such circumstances [38]. In contrast, fruit serving consumption signiﬁcantly increased after the PHS-pro intervention. Reinaerts [38] explained that fruit consumption is more under children’s control, and if the child chooses to eat fruit, availability is the only important factor for fruit consumption. Int. J. Environ. Res. Public Health 2021, 18, 9872 16 of 19 As mentioned above, at baseline, children recorded a relatively low vegetable con- sumption, whereas half of the children did not eat any kind of vegetables over a 3-day period. This is a matter of serious concern regarding children’s health. Given that adoles- cence is an accelerated growth period and vegetables are the most major sources of vitamins, minerals and dietary ﬁbre, an adequate intake of vegetables containing these nutrients, which are crucial for healthy growth, should be encouraged to sustain healthy habits that are of greater long-term signiﬁcance for preventing obesity and other chronic diseases. Several studies provided evidence in support of these important mechanisms [11,39,40]. 4. Discussion p In designing the intervention, three elements of the programme seemed essential to the achievement of signiﬁcant results: behaviour commitment based on the TTM of stages change, the participatory activities included in the learning modules, and children involved in their decision-making goals and feedback. Children at these ages made numerous de- cisions about their diet; nevertheless, there was a general difﬁculty of taking the right decisions for everyday healthy eating even with adults. Therefore, the learning modules were designed to convert healthy eating principles into practical and speciﬁc behaviour, but regular guidance was required to support children for change behaviour. Further- more, implementing the programme with this design allowed for several possibilities for improving the programme to further tailor it to the target population. p g p g g p p This study has several strengths. It is one of the ﬁrst Portuguese studies to evaluate the impact of a behavioural change programme over an entire school year. No Portuguese studies, to our knowledge, have reported any intervention approach based on TTM of stage change for the eating behaviour among children in early adolescence. The controlled design with baseline and post-intervention measurements, based on the repeated time– series design, allowed for the observations over the time period of one school year with Int. J. Environ. Res. Public Health 2021, 18, 9872 17 of 19 17 of 19 the inclusion of eight different learning modules that would not have been possible using the single pre- and post-intervention design. Another strength of this study is that data analyses were based on a balanced sample by age and gender parameters. An additional strength was that the intervention was conducted in a real-world school setting and showed it could be included in the school organisation, implemented in different school subjects with different stakeholders. Although the sample was ample, a larger sample may have resulted in more statis- tically signiﬁcant results. Indeed, it was expected that all the available children could be included, but some did not deliver the baseline evaluation, (FR0) or some subsequent food records (inclusion criteria of this study), which led to the withdrawal of several tens of children records. Several factors may have inﬂuenced these results, as this study was not without limitations. First, randomisation was not possible, and the study was conducted under the constraints of school-based research without a concurrent control group submitted to the 3-day food records. 4. Discussion Second, the study was restricted to only one school, and results may not have been generalised to other populations. Third, results were not free of selection and attrition bias, considering that children who participated in the study may have been the most motivated to change their behaviour. Finally, dietary intakes were self-reported and underreporting was a widespread problem in this research ﬁeld. Indeed, measuring eating behaviour was relatively labour-intensive and not free of expectancy bias, although the 3-day food record was considered the “gold standard” for monitoring eating behaviour [31]. Despite the aforementioned limitations, the current published research on developing an effective and sustainable intervention to improve children’s eating behaviour in Por- tugal is scarce (e.g., [41,43,44]). It is expected that the present results can have important implications for improving the design of other school programmes that are intended to promote healthy eating and active living in young people. In addition, participatory activities allowed for the documentation of children’s at- titudes and behaviour changes, which were of great value for this research. Concerning gender differences, this factor was not an important enough component in this analysis to draw strong conclusions about gender differences. Indeed, boys and girls had different food preferences and perceptions; therefore, aspects concerning gender differences should be the subject of careful research. j This study supports preliminary evidence that PHS-pro may be effective in promot- ing behaviour changes to improve children’s eating patterns for preventing overweight and obese children. The failure to obtain some positive results in children’s vegetables consumption indicates that PHS-pro strategies need to be improved effectively in a future PHS-pro implementation. Nevertheless, this study raises several interesting ideas for future research to extend the programme to other ages and promote healthy eating over adolescence, coordinating an ongoing intervention tailored to real population needs. g g g p p Finally, this pilot research did not involve changes in children’s family environment, particularly because of the time and conditions allocated to the research, but it is one important issue to be taken into account in further research. 5. Conclusions 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. References 1. WHO. The European Health Report 2005: Public Health Action for Healthier Children and Populations; World Health Organization: Geneva, Switzerland, 2005. 2. Gidding, S.S.; Dennison, B.A.; Birch, L.L.; Daniels, S.R.; Gilman, M.W.; Lichtenstein, A.H.; Rattay, K.T.; Steinberger, J.; Stettler, N.; Van Horn, L. Dietary recommendations for children and adolescents: A guide for practitioners. Pediatrics 2006, 117, 544–559. [CrossRef] 3. Ambrosini, G.L.; Emmett, P.M.; Northstone, K.; Howe, L.D.; Tilling, K.; Jebb, S.A. Identiﬁcation of a dietary pattern prospectively associated with increased adiposity during childhood and adolescence. Int. J. Obes. 2012, 36, 1299–1305. [CrossRef] 4. WHO. Global Health Risks: Mortality and Burden of Disease Attributable to Selected Major Risks; World Health Organization: Geneva, Switzerland, 2009. 5. Ebbeling, C.B.; Pawlak, D.B.; Ludwig, D.S. A randomized trial of sugar-sweetened beverages and adolescent body weight. N. Engl. J. Med. 2012, 367, 1407–1416. [CrossRef] 6. Diethelm, K.; Jankovic, N.; Moreno, L.A.; Huybrechts, I.; De Henauw, S.; De Vriendt, T.; González-Gross, M.; Leclercq, C.; Gottrand, F.; Gilbert, C.C. Food intake of European adolescents in the light of different food-based dietary guidelines: Results of the HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence) Study. Public Health Nutr. 2012, 15, 386–398. [CrossRef] 6 e e , ; J o c, N ; o e o, ; y ec s, ; e e , S ; e e , ; Go e G oss, ; ec e cq, C ; Gottrand, F.; Gilbert, C.C. Food intake of European adolescents in the light of different food-based dietary guidelines: Results of the HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence) Study. Public Health Nutr. 2012, 15, 386–398. [CrossRef] 7. Reilly, J.J.; Kelly, J. Long-term impact of overweight and obesity in childhood and adolescence on morbidity and premature mortality in adulthood: Systematic review. Int. J. Obes. 2011, 35, 891–898. [CrossRef] [PubMed] ( y y p y ) y , , [ ] 7. Reilly, J.J.; Kelly, J. Long-term impact of overweight and obesity in childhood and adolescence on morbidity and premature mortality in adulthood: Systematic review. Int. J. Obes. 2011, 35, 891–898. [CrossRef] [PubMed] 7. Reilly, J.J.; Kelly, J. Long-term impact of overweight and obesity in childhood and adolescence o mortality in adulthood: Systematic review. Int. J. Obes. 2011, 35, 891–898. [CrossRef] [PubMed] y y 8. Alberga, A.S.; Sigal, R.J.; Goldﬁeld, G.; Prud’homme, D.; Kenny, G.P. Overweight and obese teenagers: Why is adolescence a critical period? Pediatr. Obes. 2012, 7, 261–273. References [CrossRef] [PubMed] p 9. Bunc, V. Obesity-causes and remedies. Phys. Act. Rev. 2016, 7, 50–56. [CrossRef] 10. Hung, H.V.; Ngoc, V.T.N.; Vu, T.H.; Chu, D.T. Early Childhood Caries in Obese Children: The Status and Associated Factors in the Suburban Areas in Hanoi, Vietnam. Int. J. Environ. Res. Public Health 2021, 18, 8844. [CrossRef] [PubMed] 10. Hung, H.V.; Ngoc, V.T.N.; Vu, T.H.; Chu, D.T. Early Childhood Caries in Obese Children: The Status and Associated Factors in the Suburban Areas in Hanoi, Vietnam. Int. J. Environ. Res. Public Health 2021, 18, 8844. [CrossRef] [PubMed] 11 Sl i J L Ll d B H lth b ﬁt f f it d t bl Adv Nutr 2012 3 506 516 [C R f] [P bM d] , J ; y , g , , [ ] [ ] 12. Lobstein, T.; Jackson-Leach, R.; Moodie, M.L.; Hall, K.D.; Gortmaker, S.L.; Swinburn, B.A.; James, W.P.T.; Wang, Y.; McPherson, K. Child and adolescent obesity: Part of a bigger picture. Lancet 2015, 385, 2510–2520. [CrossRef] 13. Khambalia, A.Z.; Dickinson, S.; Hardy, L.L.; Gill, T.; Baur, L.A. A synthesis of existing systematic reviews and meta-analyses of school-based behavioural interventions for controlling and preventing obesity. Obes. Rev. 2012, 13, 214–233. [CrossRef] [PubMed] 14. Kropski, J.A.; Keckley, P.H.; Jensen, G.L. School-based obesity prevention programs: An evidence-based review. Obesity 2008, 6, 1009–1018. [CrossRef] 13. Khambalia, A.Z.; Dickinson, S.; Hardy, L.L.; Gill, T.; Baur, L.A. A synthesis of existing systematic reviews and meta-analyses of school-based behavioural interventions for controlling and preventing obesity. Obes. Rev. 2012, 13, 214–233. [CrossRef] [PubMed] school-based behavioural interventions for controlling and preventing obesity. Obes. Rev. 2012, 13, 214–233. [CrossRef] [PubMed] 14. Kropski, J.A.; Keckley, P.H.; Jensen, G.L. School-based obesity prevention programs: An evidence-based review. Obesity 2008, 6, 1009–1018. [CrossRef] g p g y 14. Kropski, J.A.; Keckley, P.H.; Jensen, G.L. School-based obesity prevention programs: An evidence-based review. Obesity 2008, 6, 1009–1018. [CrossRef] 15. Katz, D.L. School-based interventions for health promotion and weight control: Not just waiting on th Rev. Public Health 2009, 30, 253–272. [CrossRef] 16. Foster, G.D.; Sherman, S.; Borradaile, K.E.; Grundy, K.M.; Vander Veur, S.S.; Nachmani, J.; Karpyn, A.; Kumanyika, S.; Shults, J. A policy-based school intervention to prevent overweight and obesity. Pediatrics 2008, 121, 794–802. [CrossRef] [PubMed] 16. Foster, G.D.; Sherman, S.; Borradaile, K.E.; Grundy, K.M.; Vander Veur, S.S.; Nachmani, J.; Karpyn, A.; Kumanyika, S.; Shults, J. A policy-based school intervention to prevent overweight and obesity. 18. WHO. Promoting Health through Schools: Report of a WHO Expert Committee on Comprehensive School Healt World Health Organization: Geneva, Switzerland, 1997. 5. Conclusions The learning modules and participatory activities designed for the PHS-pro implemen- tation, in which children set goals and made commitments to change a speciﬁc behaviour or engage in a new behaviour, resulted in signiﬁcant eating behaviour changes. Finding out children’s needs, focused on speciﬁc aspects of their behaviour that they were already prepared to change and the implementation of practical recommendations appeared to be a successful strategy. Thus, we conclude that applying these three elements: behaviour commitment based on the TTM of stages change, participatory activities, and children involved in decision-making goals and feedback, seems essential to achieve signiﬁcant results and a promising intervention strategy. 18 of 19 Int. J. Environ. Res. Public Health 2021, 18, 9872 18 of 19 Author Contributions: Conceptualization, M.V. and G.S.C.; methodology, M.V.; formal analysis, M.V. and G.S.C.; data curation, M.V. and G.S.C.; writing—original draft preparation, M.V.; writing—review and editing, M.V. and G.S.C.; funding acquisition, M.V. and G.S.C. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by the FOUNDATION FOR SCIENCE AND TECHNOLOGY (FCT), grant numbers SFRH/BD/79512/2011 and RESEARCH CENTRE ON CHILD STUDIES (R&D Unit 317 of FCT; projects UIDB/00317/2020 and UIDP/00317/2020). Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Scientiﬁc Council of the Institute of Education of the UNIVERSITY OF MINHO. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. 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. The data are not publicly available due to data privacy of the participants. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to data privacy of the participants. Acknowledgments: The authors thank the children, the teachers, the school personnel, and the parents for their participation. Acknowledgments: The authors thank the children, the teachers, the school personnel, and the parents for their participation. 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. References Pediatrics 2008, 121, 794–802. [CrossRef] [PubMed] 16. Foster, G.D.; Sherman, S.; Borradaile, K.E.; Grundy, K.M.; Vander Veur, S.S.; Nachmani, J.; Karpyn, A.; Kumanyika, S.; Shults, J. A policy-based school intervention to prevent overweight and obesity. Pediatrics 2008, 121, 794–802. [CrossRef] [PubMed] 17. St Leger, L. What’s the place of schools in promoting health? Are we too optimistic? Health Promot. Int. 2004, 19, 405–408. [CrossRef] policy-based school intervention to prevent overweight and obesity. Pediatrics 2008, 121, 794–802. [CrossRef] [PubMed] 17. St Leger, L. What’s the place of schools in promoting health? Are we too optimistic? Health Promot. Int. 2004, 19, 405–408. [CrossRef] p y p g y 17. St Leger, L. What’s the place of schools in promoting health? Are we too optimistic? Health Promot. Int. 2004, 19, 405–408. [CrossRef] 17. St Leger, L. What’s the place of schools in promoting health? Are we too optimistic? Health Promot. Int. 2004, 19, 405–408. [CrossRef] 18. WHO. Promoting Health through Schools: Report of a WHO Expert Committee on Comprehensive School Health Education and Promotion; World Health Organization: Geneva, Switzerland, 1997. 18. WHO. Promoting Health through Schools: Report of a WHO Expert Committee on Comprehensive School Health Education and Promotion; World Health Organization: Geneva, Switzerland, 1997. 19 of 19 19 of 19 Int. J. Environ. Res. Public Health 2021, 18, 9872 19. Waters, E.; de Silva-Sanigorski, A.; Burford, B.J.; Brown, T.; Campbell, K.J.; Gao, Y.; Armstrong, R.; Prosser, L.; Summerbell, C.D. Interventions for preventing obesity in children. Cochrane Libr. 2011, 12. [CrossRef] p g y 20. Brown, T.; Moore, T.H.M.; Hooper, L.; Gao, Y.; Zayegh, A.; Ijaz, S.; Elwenspoek, M.; Foxen, S.C.; Magee, L.; O’Malley, C. Interventions for preventing obesity in children. Cochrane Database Syst. Rev. 2019, 7, CD001871. [CrossRef] T.; Moore, T.H.M.; Hooper, L.; Gao, Y.; Zayegh, A.; Ijaz, S.; Elwenspoek, M.; Foxen, S.C.; Magee, L.; O ions for preventing obesity in children. Cochrane Database Syst. Rev. 2019, 7, CD001871. [CrossRef] p g y y 21. Vieira, M.; Carvalho, G.S. Costs and beneﬁts of a school-based health intervention in Portugal. Health Promot. Int. 2019, 34, 1141–1148. [CrossRef] 22. Leedy, P.D. The Experimental Study. In Practical Research: Planning and Design, 6th ed.; Leedy, P.D., Ertmer, P.A., Timothy, J.N., Eds.; Pearson Merrill Prentice Hall: Columbus, OH, USA, 1996; pp. 229–242. 23. Biglan, A.; Ary, D.; Wagenaar, A.C. The value of interrupted time-series experiments for community intervention research. Prev. Sci. 2000, 1, 31–49. [CrossRef] 24. References Prochaska, J.O.; Redding, C.A.; Evers, K.E. The Transtheoretical Model and Stages Change. In Health Behavior and Health Education: Theory, Research, and Practice; Glanz, K., Rimer, B.K., Viswanath, K., Eds.; John Wiley & Sons: San Francisco, CA, USA, 2008; pp. 97–121. pp 25. Prochaska, J.O.; Velicer, W.F.; Redding, C.; Rossi, J.S.; Goldstein, M.; DePue, J.; Greene, G.W.; Rossi, S.R.; Sun, X.; Fava, J.L.; et al. Stage-based expert systems to guide a population of primary care patients to quit smoking, eat healthier, prevent skin cancer, and receive regular mammograms. Prev. Med. 2005, 41, 406–416. [CrossRef] [PubMed] g g 26. Prochaska, J.O.; Norcross, J.C.; DiClemente, C.C. Applying the stages of change. Psychother. Aust. 2013 Norcross, J.C.; DiClemente, C.C. Applying the stages of change. Psychother. Aust. 2013, 19, 10–15. 27. Prochaska, J.O.; Velicer, W.F. The transtheoretical model of health behavior change. Am. J. Health Promot. 1997, 12, 38–48. [CrossRef] 28. WHO. Promoting Fruit and Vegetable Consumption around the World; World Health Organization: Genevan, Switzerland, 2003. Available online: http://www.who.int/dietphysicalactivity/fruit/en/ (accessed on 7 May 2010). Sarmento, M.J.; Almeida, C. Research on childhood and children as researchers: Participatory methodol 29. Soares, N.F.; Sarmento, M.J.; Almeida, C. Research on childhood and children as researchers: Participatory methodologies of social worlds of the children. In Proceedings of the Sixth International Conference on Social Methodology Recent Developments and Applications in Social Research Methodology, Amsterdam, The Netherlands, 16–20 August 2004; pp. 1–10. pp gy g pp 30. Sinclair, R. Participation in practice: Making it meaningful, effective and sustainable. Child. Soc. 2004, 18, 106–118. [CrossRef] 30. Sinclair, R. Participation in practice: Making it meaningful, effective and sustainable. Child. Soc. 2004, 18, 106–118. [CrossRef] 31. Thompson, F.E.; Byers, T. Dietary assessment resource manual. J. Nutr. 1994, 124, 2245S–2317S. 30. Sinclair, R. Participation in practice: Making it meaningful, effective and sustainable. Child. Soc. 20 31. Thompson, F.E.; Byers, T. Dietary assessment resource manual. J. Nutr. 1994, 124, 2245S–2317S. p y y J 32. Crawford, P.B.; Obarzanek, E.; Morrison, J.; Sabry, Z.I. Comparative advantage of 3-day food records over 24-hour recall and 5-day food frequency validated by observation of 9-and 10-year-old girls. J. Am. Diet. Assoc. 1994, 94, 626–630. [CrossRef] 33. Kirkpatrick, S.I.; Reedy, J.; Butler, E.N.; Dodd, K.W.; Subar, A.F.; Thompson, F.E.; McKinnon, R.A. Dietary assessment in food environment research: A systematic review. Am. J. Prev. Med. 2014, 46, 94–102. [CrossRef] 34. Cullen, K.W.; Baranowski, T.; Baranowski, J.; Hebert, D. References Behavioral or epidemiologic coding of fruit and vegetable consumption from 24-hour dietary recalls: Research question guides choice. J. Am. Diet. Assoc. 1999, 99, 849–851. [CrossRef] y q g 35. Hoelscher, D.M.; Day, R.S.; Lee, E.S.; Frankowski, R.F.; Kelder, S.H.; Ward, J.L.; Scheurer, M.E. Desi interventions for adolescents. J. Am. Diet. Assoc. 2002, 102, S52–S63. [CrossRef] 36. Rodrigues, S.S.P.; Franchini, B.; Graça, P.; De Almeida, M.D.V. A new food guide for the Portuguese population: Development and technical considerations. J. Nutr. Educ. Behav. 2006, 38, 189–195. [CrossRef] 37. Pope, C.; Ziebland, S.; Mays, N. Qualitative research in health care: Analysing qualitative data. BMJ 2000, 320, 114–116. [CrossRef] 37. Pope, C.; Ziebland, S.; Mays, N. Qualitative research in health care: Analysing qualitative data. BMJ 2000, 320, 114–116. [CrossRef] 38. Reinaerts, E.; de Nooijer, J.; Candel, M.; de Vries, N. Explaining school children’s fruit and vegetable consumption: The contributions of availability, accessibility, exposure, parental consumption and habit in addition to psychosocial factors. Appetite 37. Pope, C.; Ziebland, S.; Mays, N. Qualitative research in health care: Analysing qualitative data. BMJ 2000, 320, 114–116. [CrossRef] 38. Reinaerts, E.; de Nooijer, J.; Candel, M.; de Vries, N. Explaining school children’s fruit and vegetable consumption: The contributions of availability, accessibility, exposure, parental consumption and habit in addition to psychosocial factors. Appetite 2007, 48, 248–258. [CrossRef] [PubMed] 39. Delisle, H. Nutrition in Adolescence-Issues and Challenges for the Health Sector. Issues in Adolescent Health and Development; World Health Organization: Geneva, Switzerland, 2005; pp. 1–115. 40. Kaganov, B.; Caroli, M.; Mazur, A.; Singhal, A.; Vania, A. Suboptimal micronutrient intake among c 2015, 7, 3524–3535. [CrossRef] [PubMed] M.; Mazur, A.; Singhal, A.; Vania, A. Suboptimal micronutrient intake among children in Europe. Nutrients CrossRef] [PubMed] , , [ ] [ ] 41. Gaspar, T.; Matos, M.G.; Santos, T.C.C.F.; Albergaria, F.S. O projecto europeu TEMPEST (auto-regulação para a prevenção da obesidade em crianças e adolescentes): Análise de esquemas de incentivo em Portugal. Rev. Psicol. Criança Adolesc. 2012, 3, 169–185. [CrossRef] 42. Bemelmans, W.J.E.; Verschuuren, M.; Van Dale, D.; Savelkoul, M.; Wendel-Vos, G.C.W.; Van Raaij, J. An EU-Wide Overview of Community-Based Initiatives to Reduce Childhood Obesity; Dutch Institute for Public Health and the Environment (RIVM): Bilthoven, The Netherlands, 2011. Available online: https://sante.public.lu/fr/publications/e/eu-childhood-obesity/eu-childhood-obesity. pdf (accessed on 5 August 2016). p g 43. Pereira, M.; Padez, C.; Nogueira, H. Municipal health promotion programs: Is childhood obesity a concern at local level in Portugal? Health Promot. Int. 2021. References [CrossRef] 44. Ferreira, R.J.; Nogueira, T.; da Silva, V.D.; Pinto, M.L.; Sousa, J.; Pereira, A.M.; Nogueira, P.J.; Borrego, R.; Raposo, A.; Martins, J. A school-based intervention for a better future: Study protocol of Sintra Grows Healthy. BMC Public Health 2020, 20, 1615. [CrossRef]
https://openalex.org/W4384115948	https://zenodo.org/records/8140802/files/FIVE%20MITOCHONDRIAL%20GENOMES%20OF%20THE%20LIPOPHILIC%20FUNGUS%20MALASSEZIA%20(Malasseziales%20Exobasidiomycetes).pdf	English	null	FIVE MITOCHONDRIAL GENOMES OF THE LIPOPHILIC FUNGUS MALASSEZIA (MALASSEZIALES EXOBASIDIOMYCETES)	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	3,406	FIVE MITOCHONDRIAL GENOMES OF THE LIPOPHILIC FUNGUS MALASSEZIA (MALASSEZIALES EXOBASIDIOMYCETES) Gabriel Barbosa Huszcz , Gledison Eric Teixeira , Igor Henrique Rodrigues-Oliveira , Renan Rodrigues Rocha , Rafaela de Campos Oliveira , Paulo Sallarola Takao , David Aciole Barbosa , Daniela Leite Jabes , Karine Frehner Kavalco , Rubens Pasa , Fabiano Bezerra Menegidio 1,2 1 3 2,3 2 2 1 1 3 3 1,2 Abstract: Fungi of the Malassezia genus are the most prevalent eukaryotic organisms in the natural skin mycobiome of animals and humans. The genus is found in various parts of the body, causing common skin diseases such as pityriasis versicolor and seborrheic dermatitis, and its high adaptability to the human body and prevalence among other eukaryotes makes it unique. However, even with eighteen known species and genomes available, studies on their phylogeny and basic genomic information are still scarce. To this day, only nine species have their mitochondrial genome described and assembled. In this study, we describe for the mtDNA of Malassezia vespertilionis, Malassezia dermatis, Malassezia nana, Malassezia equina, Malassezia caprae and six new mitogenomes of Malassezia pachydermatis. In addition, to better understand the phylogenetic relationships, we mined 35 mitogenomes belonging to 9 species of Malassezia, along with the mitogenomes gathered in this study, and reconstructed the phylogeny based on mtDNA. The results of this study provide genomic variation information and enhance the understanding of the Malassezia genus. They could later deliver highly valuable new insight into data for phylogenetic analysis and population genetics. Malassezia, along with the mitogenomes gathered in this study, and reconstructed the phylogeny based on mtDNA. The results of this study provide genomic variation information and enhance the understanding of the Malassezia genus. They could later deliver highly valuable new insight into data for phylogenetic analysis and population genetics. Keywords: Mitochondrial genome, Malassezia vespertilionis, Malassezia dermatis, Malassezia nana, Malassezia equina, Malassezia caprae, Malassezia pachydermatis, Phylogeny. Introduction Yeasts of the Malassezia genus are present in the microbiota of many animals, including humans, living as commensal organisms on the skin and scalp (Findley et al., 2020). However, in some cases, they can become pathogenic under the influence of predisposing factors that involve both the conditions of the skin microenvironment and changes in the host’s immune system (Theelen et al., 2018; Aykut et al., 2019; Limon et al., 2019; Saunte et al., 2020; Spatz and Richard 2020). The genus groups basidiomycetic, lipophilic (M. pachydermatis) and/or lipodependent yeasts that show asexual reproduction by unipolar or sympodial budding (M. sympodialis) from the parent cell. Materials & Methods Materials & Methods Another important feature is the possibility of forming pseudohyphae. Another important feature is the possibility of forming pseudohyphae. Differentiation of species in the traditional way is done through lipid and fatty acid assimilation tests (biochemical methods), also considering the characteristics of micromorphology. Currently, molecular biology methodologies have been considered of great importance for the taxonomic grouping of the Malassezia genus. The support of molecular tools is very important, since the morphological characteristics between species are very subtle, which can generate doubts and errors. Study of the D1/D2 regions of the 28S ribosomal RNA gene (rRNA28S) and the ITS (Internal Transcribed Spacer) regions have helped in the differentiation of Malassezia species (Batra et al., 2005). In 1996, based on studies based on morphology, physiology and molecular biology, there was a change in the classification of the genus Malassezia in seven species. As of 2002, another seven new species of Malassezia were cited in the literature. Currently, 18 species are recognized for the genus, with only 17 having available genomes (M. arunalokei, M. caprae, M. cuniculi, M. dermatis, M. equina, M. furfur, M. globosa, M. japonica, M. nana, M. obtusa, M. pachydermatis, M. restricta, M. slooffiae, Malassezia sp., M. sympodialis, M. vespertilionis and M. yamatoensis) and 9 having mitochondrial genomes (M. furfur, M. globosa, M. japonica, M. obtusa, M. pachydermatis, M. restricta, M. slooffiae, M. sympodialis and M. yamatoensis). another seven new species of Malassezia were cited in the literature. Currently, 18 species are recognized for the genus, with only 17 having available genomes (M. arunalokei, M. caprae, M. cuniculi, M. dermatis, M. equina, M. furfur, M. globosa, M. japonica, M. nana, M. obtusa, M. pachydermatis, M. restricta, M. slooffiae, Malassezia sp., M. sympodialis, M. vespertilionis and M. yamatoensis) and 9 having mitochondrial genomes (M. furfur, M. globosa, M. japonica, M. obtusa, M. pachydermatis, M. restricta, M. slooffiae, M. sympodialis and M. yamatoensis). In this study, we describe the mitochondrial genome of five species of Malasseziales: Malassezia vespertilionis, Malassezia dermatis, Malassezia nana, Malassezia equina and Malassezia caprae, also six new mitogenomes of Malassezia pachydermatis. In addition, to better understand the phylogenetic relationships, we mined 35 mitogenomes belonging to 9 species of Malassezia, along with the mitogenomes gathered in this study, and reconstructed the phylogeny based on mtDNA. Mitochondrial phylogenetic analysis To perform the phylogenetic analyses, we manually extracted the sequences of the 15 proteincoding genes (PCGs) of the 11 mitogenomes assembled in this work and added to the dataset the PCGs of the 34 mitogenomes of other Malassezia species available at the NCBI, addition to the Rhodotorula mucilaginous (MF694646.1) mitogenome as an outgroup. We aligned the sequences with MAFFT v7 (Katoh & Standley, 2013), available on the Galaxy Europe online platform (Afgan et al., 2018), and concatenated the alignments in SequenceMatrix v1.8 (Vaidya et al., 2011). We constructed the phylogeny by the Maximum Likelihood (ML) method in the IqTree v2.1.2 software (Nguyen et al., 2015) using 1000 ultra-fast bootstrap repeats and the evolutionary model GTR+F+I+G4 estimated by the IqTree ModelFinder (Kalyaanamoorthy et al., 2017). Mitogenome assembly and annotation We obtained Whole Genome Sequencing (WGS) raw data of Malassezia vespertilionis (SRR6206152; Lorch et al., 2018), Malassezia dermatis (DRR043255; Sugita et al., 2002), Malassezia nana (DRR043258; Hirai et al., 2004), Malassezia equina (SRR2136627; Wu et al.,2015) and Malassezia caprae (SRR2132347; Wu et al., 2015) from the Sequence Reads Archive (SRA) NCBI. The reads were submitted to a workflow previously used and validated in fish mitogenome assemblies by Rocha-Reis et al. (2020), Resende et al. (2020) and Pasa et al. (2021), with minor modifications to adapt to the assembly of fungal mitogenomes. We imported the raw data into the Galaxy Europe platform (https://usegalaxy.eu/) and used NOVOplasty v4.3.1 (Dierckxsens et al.,2017) to assemble the mitochondrial genomes. As a seed, we used the complete mitogenome of M. japonica (KY911090.1). We annotated the sequences obtained on Mitos2 (http://mitos2.bioinf.uni-leipzig.de/; Arab et al., 2017) using Gene Code 3 – Yeast and RefSeq 89 Fungi. To validate our workflow, we also performed the assembly and annotation of six new mitogenomes from Malassezia pachydermatis WGS libraries (SRR12046913, SRR12005311, SRR12005312, SRR12005313, SRR12005314 and SRR2135029). In addition, the 34 mitogenomes of the 9 species of the Malassezia genus available at the NCBI (M. furfur, M. globosa, M. japonica, M. obtusa, M. pachydermatis, M. restricta, M. slooffiae, M. sympodialis and M. yamatoensis) were annotated according to the procedure described above. The circular map of the new mitogenomes was drawn using the online software GenomeVx (Conant and Wolfe, 2008). We used the Blast Ring Image Generator (BRIG) (Alikhan et al., 2011) to perform a comparative Blast analysis of available representative Malassezia mitogenomes against our mitogenome pools. Data Description Data Description Characterization of mitochondrial genomes of Malassezia species The mitogenome of the five Malassezia species are circular DNA molecules with a total length ranging from 27,284 bp to 40,970 bp. Mitogenome sizes varied widely among the five newly assembled mitochondrial genomes, with M. vespertilionis having the largest among the new mtDNA and M. caprae being the smallest mitogenome ever described for the genus. Furthermore, M. equina becomes the second smallest mitogenome described for the genus, followed by the already described M. japonica. However, M. dermatis and M. nana presented a mitogenome with a size close to the average of other species of the genus. Likewise, the new M. pachydermatis mitogenomes were between 35,817 bp and 35,829 bp in size, being slightly larger than the M. pachydermatis strain CBS1879 (KY911092.1) mitogenome previously described. A comparison between the size of the new mitogenomes described in this work and those available in the NCBI database can be seen in Supplementary Tables 1. The GC concentration of the mitogenome differs among species, which may be affected by mutation bias, selection, and biases of reconstitution-related DNA repair (Li et al., 2018). The low G + C content observed is similar to other fungal mitogenomes in Malassezia (Supplementary Table 1). The GC content in the five mitogenomes ranged from 23.4% to 33.7%, among which the GC content in the genome of M. caprae was the highest. The base composition of the mitochondrial genome of M. caprae was estimated as 33.8% A, 17.2%C, 16.5% G, and 32.5% T, thus presenting a bias towards AT (66.3%) and the GC content was 33.7%. M. equina presented the composition of 33.4% A, 33.7% T, 16.0% C and 16.1% G, with a bias towards AT (67.1%) and GC content of 32.9%. M. nana presented the composition of 33.6% A, 34.5% T, 15.8% C and 16.1% G, with a bias towards AT (68.1%) and GC content of 31.9%.M. vespertilionis presented the composition of 36.8% A, 39.8% T, 11.5% C and 11.8% G, with a bias towards AT (76.6%) and GC content of 23.4%. Finally, M. dermatis presented a composition of 34.1% A, 34.3% T, 15.6% C and 16.0% G, with a bias towards AT (68.4%) and GC content of 31.6%. In general, mitogenomic coding is associated with the mitochondrial translation apparatus, electron transport, and oxidative phosphorylation. Characterization of mitochondrial genomes of Malassezia species The potential coding genes (PCGs) of Malassezia species included 14 PCGs related to oxidative phosphorylation [NAD dehydrogenase (nad1-6 and nad4L), cytochrome oxidases (cox1, cox2 and cox3), cytochrome b (cob), and 3 ATP synthases (atp6, atp8 and atp9), ribosomal protein S3 (rps3)] and 2 ribosomal RNA genes (rns and rnl). Only M. equina, M. caprae and M. vespertilionis presented 14 PCGs, due to the absence of the atp9 gene. This distribution is consistent with that observed in typical fungal mitogenomes, such as those belonging to Ascomycota and Basidiomycota (Dikarya Subkingdom). In total, the coding regions containing the 15 PCGs described occupied between 13,725 bp and 14,103 bp in the two mitogenomes presented (M. nana and M. dermatis, respectively). Likewise, the 15 PCGs occupied from 14,037 bp to 14,208 bp among the 6 new mitogenomes of M. pachydermatis described here, in addition to 12,735 bp and 14,784 bp among the mitogenomes available in the NCBI database. In mitogenomes that presented 14 PCGs, the coding regions occupied between 13,437 bp and 14,079 bp (M. equina and M. caprae, respectively). Other existing regions in mitogenomes are tRNAs, rRNA genes and intergenic regions. More details on these regions contained in mitogenomes are available in Supplementary Table 1. Except for M. vespertilionis, all new mitogenomes showed nad4 and atp6 in the light chain. Likewise, a duplication of the atp9 gene in M. dermatis, M. nana and in the new M. pachydermatis mitogenomes was also observed in the light chain. In M. vespertilionis, only the cob gene was observed in the light chain. Apart from those already described, all other genes were observed in the heavy chain for all new mitogenomes. More details on the direction of PCGs can be found in Supplementary Tables 1-2. Besides M. caprae and M. equina, where 18 tRNA genes were identified in the mitogenomes, the other species characterized in this work presented between 20 and 25 tRNAs, encoding the 20 standard amino acids. The length of tRNA genes in the five mitogenomes ranged from 71 bp to 86 bp. Each of the five Malassezia mitogenomes involved 2 rRNA genes: the small subunit ribosomal RNA gene (rns) and the large subunit ribosomal RNA gene (rnl). No significant difference was identified in the number of rRNA genes in the five species, and the total length of the genes. Phylogenetic Analysis We obtained a robust evolutionary tree topology (Figure 1) with all recovered clades well supported. Our phylogeny is mostly accordant with a prior study based on sequences of six mitochondrial and nuclear genes (Wang et al., 2014). The tree has shown two main clades comprising a distinct group of species. As in the literature, our data reached a clade encompassing M. caprae, M. dermatis, M. sympodialis, M. equina, and M. nana. This clade is near related to M. pachydermatis clade. Another well-marked and related clade comprises M. globosa and M. restricta lineages. The other main clade assembles M. sloffiae, M. vespertilionis, M. japonica, M. yamatoensis, M. obtusa, and a species complex formed by samples identified as M. furfur. Our analysis shows two clades of M. furfur with 100% bootstrap confidence and a single strain diverging early. The bootstrap confidence of this strain with the two largest clades reached 100% of confidence. The main difference between our analysis and Wang et al. (2014) study was the position of M. restricta, M. globosa, and M. pachydermatis. In our three (Figure 1), these species are closest to the first mentioned clade, though previous analyzes grouped them with the second clade (Wang et al., 2014). In the phylogenetic analysis, the novel Malassezia pachydermatis mitogenome had clustered consistently with M. pachydermatis strain CBS1879 (KY911092.1). The same had happened with other species with multiple mitogenomes available. 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. Acknowledgments This study was financed in part by scholarship grants from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil – CAPES (www.capes.gov.br/) and Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq (www.gov.br/cnpq/pt-br). Conclusion The methodology used to assembly the mitochondrial genomes of Malassezia vespertilionis, Malassezia dermatis, Malassezia nana, Malassezia equina and Malassezia caprae proved to be satisfactory and allowed the expanded the phylogeny of Malasseziales through mtDNA. We were also able to validate our workflow by assembling 6 new Malassezia pachydermatis mitogenomes that demonstrated high agreement with the mitogenome already available. The methodology used to assembly the mitochondrial genomes of Malassezia vespertilionis, Malassezia dermatis, Malassezia nana, Malassezia equina and Malassezia caprae proved to be satisfactory and allowed the expanded the phylogeny of Malasseziales through mtDNA. We were also able to validate our workflow by assembling 6 new Malassezia pachydermatis mitogenomes that demonstrated high agreement with the mitogenome already available. Furthermore, the study of the complete mitochondrial genome proves to be a tool with potential to solve taxonomic problems and help in understanding the evolutionary and ecological relationships of this group. Furthermore, the study of the complete mitochondrial genome proves to be a tool with potential to solve taxonomic problems and help in understanding the evolutionary and ecological relationships of this group. Supplementary Material The Supplementary Material for this article can be found online at: https://doi.org/10.6084/m9.figshare.23666289.v1 Author Contributions RCO, IHRO, GH, GET, PST, RRR, DAB, DLJ, KFK, RP and FBM contributed to the conceptualization, investigation and design of the study. RCO, IHRO, GH, GET, PST, RRR, DAB, DLJ, KFK, RP and FBM performed the bioinformatics/statistics analysis and were responsible for data validation. FBM and DLJ provided resources to carry out the work. FBM, DLJ, KFK and RP were responsible for the administration and supervision of the research project. FBM, DAB, RCO, IHRO, PST and RRR wrote the first draft of the manuscript. FBM, DLJ, KFK and RP wrote sections of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version. Reference Afgan E, Baker D, Batut B, Van Den Beek M, Bouvier D, Čech M et al. The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2018 update. Nucleic acids research. 2018;46(W1):W537-W544. Al Arab M, Zu Siederdissen CH, Tout K, Sahyoun AH, Stadler PF, Bernt M. Accurate annotation of protein-coding genes in mitochondrial genomes. Molecular phylogenetics and evolution. 2017;106: 209-216. Alikhan, NF., Petty, N.K., Ben Zakour, N.L. et al. BLAST Ring Image Generator (BRIG): simple prokaryote genome comparisons. BMC Genomics 12, 402 (2011). https://doi.org/10.1186/1471-2164-12-402 Aykut B, Pushalkar S, Chen R et al. The fungal mycobiome promotes pancreatic oncogenesis via activation of MBL. Nature. 2019;574:264–7. Dierckxsens N, Mardulyn P, Smits G. NOVO Plasty: assembly of organelle genomes from whole genome data. Nucleic acids research. 2017;45(4):e18-e18. Findley K, Oh J, Yang J et al. Topographic diversity of fungal and bacterial communities in human skin. Nature. 2020;498:367–70. Findley K, Oh J, Yang J et al. Topographic diversity of fungal and bacterial communities in human skin. Nature. 2020;498:367–70. Hirai A, Kano R, Makimura K, Duarte ER, Hamdan JS, Lachance MA, Yamaguchi H, HasegawaA. Malassezia nana sp. nov., a novel lipid-dependent yeast species isolated from animals. Int J Syst Evol Microbiol. 2004 Mar;54(Pt 2):623-627. doi: 10.1099/ijs.0.02776-0. PMID: 15023986. Kalyaanamoorthy S, Minh BQ, Wong TK, Von Haeseler A, Jermiin LS. Model Finder: fast model selection for accurate phylogenetic estimates. Nature Methods 2017;4(6):587-589. Kalyaanamoorthy S, Minh BQ, Wong TK, Von Haeseler A, Jermiin LS. Model Finder: fast model selection for accurate phylogenetic estimates. Nature Methods 2017;4(6):587-589. Katoh K, Standley DM. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution. 2013;30(4):772– 780. Katoh K, Standley DM. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution. 2013;30(4):772– 780. Limon JJ, Tang J, Li D et al. Malassezia is associated with Crohn’s disease and exacerbates colitis in mouse models. Cell Host Microbe. 2019;25:377–88. Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, Von Haeseler A, et al. IQTREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Molecular Biology and Evolution 2020;37(5):1530- 1534. inference in the genomic era. Molecular Biology and Evolution 2020;37(5):1530- 1534. Saunte DML, Gaitanis G, Hay RJ. Malassezia-Associated skin diseases, the use of diagnostics and treatment. Front Cell Infect Microbiol. 2020;10:112. Saunte DML, Gaitanis G, Hay RJ. Malassezia-Associated skin diseases, the use of diagnostics and treatment. Front Cell Infect Microbiol. 2020;10:112. Spatz M, Richard ML. Overview of the potential role of Malassezia in gut health and disease. Front Cell Infect Microbiol. 2020;10:201. Spatz M, Richard ML. Overview of the potential role of Malassezia in gut health and disease. Front Cell Infect Microbiol. 2020;10:201. Theelen B, Cafarchia C, Gaitanis G et al. Malassezia ecology, pathophysiology, and treatment [published correction appears in Med Mycol. 2019 Apr 1;57(3):e2]. Med Mycol. 2018;56:S10– 25. Theelen B, Cafarchia C, Gaitanis G et al. Malassezia ecology, pathophysiology, and treatment [published correction appears in Med Mycol. 2019 Apr 1;57(3):e2]. Med Mycol. 2018;56:S10– 25. Vaidya G, Lohman DJ, Meier R. Sequence Matrix: concatenation software for the fast assembly of multi‐gene datasets with character set and codon information. Cladistics 2011;27(2):171-180. Vaidya G, Lohman DJ, Meier R. Sequence Matrix: concatenation software for the fast assembly of multi‐gene datasets with character set and codon information. Cladistics 2011;27(2):171-180. Wang QM, Theelen B, Groenewald M, Bai FY, Boekhout T. Moniliellomycetes and Malasseziomycetes, two new classes in Ustilaginomycotina. Persoonia- Molecular Phylogeny and Evolution of Fungi. 2014;33(1):41-47. Wu, G., Zhao, H., Li, C., Rajapakse, M. P., Wong, W. C. et al. Genus-wide comparative genomics of Malassezia delineates its phylogeny, physiology, and niche adaptation on human skin. PLoS Genetics 11, e1005614 (2015) FIGURE Integrated Biotechnology Center, University of Mogi das Cruzes, Mogi das Cruzes, SP, Brazil 1 Technological Research Center, University of Mogi das Cruzes, Mogi das Cruzes, SP, Brazil 2 Technological Research Center, University of Mogi das Cruzes, Mogi das Cruzes, SP, Brazil 2 Post seguinte → ← Post anterior ← Post anterior RevistaFT A RevistaFT é uma Revista Científica Eletrônica Multidisciplinar Indexada de Alto Impacto e Qualis “B2” em 2023. Periodicidade mensal e de acesso livre. Leia gratuitamente todos os artigos e publique o seu também clicando aqui. Contato Queremos te ouvir. WhatsApp: 21 98159-7352 e-Mail: contato@revistaft.com.br ISSN: 1678-0817 CNPJ: 48.728.404/0001-22 CAPES – Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), fundação do Ministério da Educação (MEC), desempenha papel fundamental na expansão e consolidação da pós-graduação stricto sensu (mestrado e doutorado) em todos os estados da Federação. Conselho Editorial Editores Fundadores: Editores Fundadores: Dr. Oston de Lacerda Mendes. Dr. Oston de Lacerda Mendes. Dr. João Marcelo Gigliotti. Editor Científico: Dr. Oston de Lacerda Mendes Orientadoras: Dra. Hevellyn Andrade Monteiro Dra. Chimene Kuhn Nobre Revisores: Lista atualizada periodicamente em revistaft.com.br/expediente Venha fazer parte de nosso time de revisores também! Copyright © Editora Oston Ltda. 1996 - 2023 Rua José Linhares, 134 - Leblon \| Rio de Janeiro-RJ \| Brasil Dr. João Marcelo Gigliotti. Editor Científico: Dr. Oston de Lacerda Mendes Orientadoras: Dra. Hevellyn Andrade Monteiro Dra. Chimene Kuhn Nobre
https://openalex.org/W2588827109	https://www.biodiversitylibrary.org/partpdf/319799	English	null	THE DEVELOPMENT OF THE SEED-COAT OF CARICA PAPAYA1	Annals of botany	1,910	public-domain	1,788	NOTES. THE DEVELOPMENT OF THE SEED-COAT OF CARICA PAPAYA.'— The seed-coat of many species of Carica shows a complicated structure, the highest degree of differentiation being reached in the seeds of C. microcarpa , C. hastaefolia , and C. Papaya (the Paupauw). Klebs 2 has described the constitution of the mature seed-coat in the two former species, but has not followed up its development from the unfertilized ovule — a development which in the case of C. Papaya shows points of interest. Material of the fruit and seeds of this species, at various stages before and after fertilization, was obtained in Angola during the recent Percy Sladen Memorial Expedition by Dr. H. H. W. Pearson, who handed it over to me for examination. The fruit of C. Papaya is a one-celled berry, with seeds lining the wall. These seeds, when mature, are oval in form and about 8 mm. long. Their coat consists of two separable layers — a hard reddish-brown endotesta, covered by a soft white sarco- testa.s The endotesta rises into irregular ridge-like outgrowths, which run down the length of the seed (Fig. i, a) and appear in transverse section as pyramidal projec¬ tions (Fig. i, b). The sarcotesta fills up the hollows between these ridges, so that the surface of the seed, while still enclosed within the fruit, is smooth and shiny ; but when it is set free, the sarcotesta quickly loses its watery contents and shrivels down into the hollows. In the young ovule, the two integuments develop as usual, and at the time of pollination consist each of four or five layers of ordinary parenchyma (Fig. 2). Immediately after pollination, these layers begin to show the following series of changes : — 1 Percy Sladen Memorial Expedition in S.W. Africa, 1908-1909; Report No. 3. 2 Klebs : Beitrage zur Morphologie und Biologie der Keimung. 3 This sarcotesta was formerly described as an arillus (Baillon : Natural History of Plants, iv, 293, London, 1880). 3 This sarcotesta was formerly described as an arillus (Baillon : Natural History of Plants, iv, 293, London, 1880). 1 Percy Sladen Memorial Expedition in S.W. Africa, 1908-1909; Report No. 3. [Annals of Botany, Vol. XXIV. No. XCV. July, 1910.] y p 2 Klebs : Beitrage zur Morphologie und Biologie der Keimung. A. The inner integmnent. The cells of the inner epidermis (a) are the only ones in this integument which retain their original form. As the seed matures, they enlarge, and a cuticular layer is formed on the outer side where they are in contact with the nucellus. At the same time, tannin is deposited in their cytoplasm, so that the contents of each cell become converted into a tanniferous block (Fig. 4). All the other cells in the integument begin to show sliding growth immediately after pollination (Fig. 3), the cells of the outer epidermis (h) elongating in a direction parallel to the long axis of the seed, while those of the underlying layers lengthen tangentially. Their walls become thickened [Annals of Botany, Vol. XXIV. No. XCV. July, 1910.] N otes. 6o8 and deeply pitted, and they are gradually transformed into characteristic sclerenchy- matous fibres (Fig. 4). The epidermal fibres are elliptical in transverse section ; their lumina are larger than in the underlying layers, and their walls much thicker. B. The outer integument. Immediately after pollination, this integument begins to increase in thickness, until it attains to the relatively considerable size shown in Figs. 1 and 4. This increase is due to the appearance of bands of meristem, running down the integument Figs. 1-4. Sections through the seed and ovule of C. papaya, to show the structure and develop¬ ment of the seed-coat. Lettering in each figure : ( g g T T . , ( a = inner epidermis Inner Integument \ , . 1 ( 0 = outer L = endotesta. Outer Integument j C,~ *mler ” ) ^ ( a = outer „ = sarcotesta. ( „ e = ridges of the endotesta. A A B IlG. i, A and B. Median longitudinal and transverse sections of the mature seed, x 3. B tudinal and transverse ature seed, x 3. Fig. 2. Transverse section through in¬ teguments and nucellus of ovule before pollination. X120. A B B IlG. i, A and B. Median longitudinal and transverse sections of the mature seed, x 3. IlG. i, A and B. Median longitudinal and transverse sections of the mature seed, x 3. Fig. 3. The same, a short time after pollination, showing a developing ridge of the endotesta on each side of the section, x 120. Fig. 3. The same, a short time after pollination, showing a developing ridge of the endotesta on each side of the section, x 120. in the layer immediately underlying the outer epidermis, and marking the future position of the ridges of the endotesta. By the divisions of these meristems, a ridge gradually arises along each band. These ridges, like the meristems, generally run roughly parallel in a longitudinal direction, but are often quite irregular. Two of these developing ridges are shown in transverse section in Fig. 3, and it will be noted that at the base of the depression between them, six or seven adjacent cells of the hypodermal layer have remained undivided, merely increasing in size. A similar row of about six hypodermal cells is found in transverse section at the base of each depression between the ridges, and these cells never divide, but continue to increase Notes. 609 greatly in size and elongate vertically, so that in the mature seed they form a pave¬ ment, six to ten cells broad, running along the bottom of these depressions (Figs. 4 and 1, b). B. The outer integument. As these ridges develop, the cells composing them increase greatly in size, while those of the original integument remain small. The mature integument (excluding Fig. 4. Transverse section through mature seed-coat, showing further development of the ridges of Fig. 3. x 60. Fig. 4. Transverse section through mature seed-coat, showing further development of the ridges of Fig. 3. x 60. the epidermal layers) thus consists of a region of cells with small lumina passing at the ridges into cells with much larger lumina (Fig. 4). As the seed-coat matures, these cells all begin to lose their contents, and their walls become thicker and rather hard (but not lignified) and take on a yellowish colour. The response of these walls to the ordinary tests for cellulose, both staining and microchemical, grows fainter, until in the mature seed only the outermost still unaltered layers give any reaction with these tests. With a solution of iodine in potassium iodide and with Schulze’s the epidermal layers) thus consists of a region of cells with small lumina passing at the ridges into cells with much larger lumina (Fig. 4). As the seed-coat matures, these cells all begin to lose their contents, and their walls become thicker and rather hard (but not lignified) and take on a yellowish colour. The response of these walls to the ordinary tests for cellulose, both staining and microchemical, grows fainter, until in the mature seed only the outermost still unaltered layers give any reaction with these tests. With a solution of iodine in potassium iodide and with Schulze’s Notes . 6io solution their yellow tinge deepens, and after treatment with iodine and sulphuric acid they become brown ; this coloration in each case passing in the outermost layers into the ordinary bluish tinge of a cellulose reaction. From these facts, it seems probable that they are composed of a hardened horny variety of cellulose, like the ‘ reserve cellulose ’ described by Gardiner and Hill 1 in the endosperm of Tamils, and by Griiss 2 in that of Phoenix and other plants, which is also unaffected by ordinary cellulose tests, and reacts similarly with iodine in potassium iodide and with iodine and sulphuric acid. 1 Gardiner, W., and Hill, A. W. : The Histology of the Endosperm during Germination in Tamus communis and Galium tricorne (Proc. Camb. Phil. Soc., xi, 445-57, PI. V, 1902). 2 Griiss, J. : Studien iiber Reserve-Cellulose (Bot. Centralbl., lxx, 242-61, 1897). B. The outer integument. As in Tamus , moreover, the middle lamella is not evident in most of the walls when mature, though in the outermost layers, and fairly often among the other large cells in the ridges, it can still be recognized as a fine line. Otherwise, the layers of the walls of adjoining cells seem to coalesce to form a homogeneous intercellular matrix. A similar transformation takes place in the outer wall of the inner epidermis (c), which also becomes thickened and loses its cellulose reactions. Otherwise, this epidermal layer remains unaltered, except that the cells increase in size and a crystal of calcium oxalate appears in each. Meanwhile, the outer epidermis (d) is developing into the sarcotesta, which is formed entirely from this layer. Over the ridges it remains single, the cells merely increasing in size, but between the ridges the cells elongate perpendicularly, and divide several times by transverse walls. By continued elongation as the ridges of the endotesta rise up, they fill up the hollows between them, so that the surface of the seed remains smooth. In the latest stages they become separated from the endo¬ testa towards the bottom of the hollows (Fig. 4). As the epidermis develops, its outer wall becomes much thickened and striated. As the water in the seed-coat evaporates, the sarcotesta shrivels up, and sinks into the hollows between the ridges of the endotesta, whose thick-walled cells retain their shape. When moistened, both layers rapidly absorb water. The cells of the sarcotesta expand to fill up the hollows once more, and the outer epidermal wall swells up considerably, so that the surface of the seed becomes smooth and shiny. The cell-walls of the endotesta also expand slightly, and water replaces the air in their lumina, so that the whole seed-coat becomes a spongy reservoir for the developing embryo to draw upon. E. L. STEPHENS. Stephens, Edith Layard. 1910. "The development of the seed-coat of Carica papaya." Annals of botany 24, 607–610. https://doi.org/10.1093/oxfordjournals.aob.a089291. View This Item Online: https://www.biodiversitylibrary.org/item/262605 DOI: https://doi.org/10.1093/oxfordjournals.aob.a089291 Permalink: https://www.biodiversitylibrary.org/partpdf/319799 Botany School, Cambridge. Botany School, Cambridge. Holding Institution New York Botanical Garden, LuEsther T. Mertz Library This file was generated 30 March 2024 at 18:47 UTC Copyright & Reuse Copyright & Reuse Copyright Status: Public domain. The BHL considers that this work is no longer under copyright protection. This document was created from content at the Biodiversity Heritage Library, the world's largest open access digital library for biodiversity literature and archives. Visit BHL at https://www.biodiversitylibrary.org. This file was generated 30 March 2024 at 18:47 UTC
https://openalex.org/W4283779608	http://publikationen.ub.uni-frankfurt.de/files/69477/s41598-022-15287-3.pdf	English	null	Characteristics and mortality of 561,379 hospitalized COVID-19 patients in Germany until December 2021 based on real-life data	Scientific reports	2,022	cc-by	8,607	www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports Scientific Reports \| (2022) 12:11116 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Table 1. DRG and OPS codes. Allocation of diagnosis related groups (DRG) and operation and procedure (OPS) codes as they are coded for reimbursement purposes in Germany. Comorbitiy OPS ICD-10 Arterial hypertension I10.x, I11.x–I13.x, I15.x Chronic pulmonary disease I27.8, I27.9, J40.x–J47.x, J60.x–J67.x, J68.4, J70.1, J70.3 Congestive heart failure I09.9, I11.0, I13.0, I13.2, I25.5, I42.0, I42.5 I42.9, I43.x, I50.x, P29.0 Diabetes E10.0, E10.1, E10.9, E11.0, E11.1, E11.9, E12.0, E12.1, E12.9, E13.0, E13.1, E13.9, E14.0, E14.1, E14.9, E10.2–E10.8, E11.2–E11.8, E12.2E12.8, E13.2–E13.8, E14.2–E14.8 Obesity E66.x Complication Intracranial hemorrhage I60.x, I61.x, I62.x Cardiac arrhytthmias I44.1–I44.3, I45.6, I45.9, I47.x–I49.x, R00.0, R00.1, R00.8, T82.1, Z45.0, Z95.0 Cardiopulmonary resuscitation 8–771, 8–772, 8–779 Embolism/thrombosis I74.x Myocardial infarction I21., I22., I24 Pulmonary embolism I26.x Renal failure I12.0, I13.1, N18.x, N19.x, N25.0, Z49.0Z49.2, Z94.0, Z99.2 Renal replacement therapy 8–853., 8–854., 8–855 Stroke I63, I64 Table 1. DRG and OPS codes. Allocation of diagnosis related groups (DRG) and operation and procedure (OPS) codes as they are coded for reimbursement purposes in Germany. The ongoing COVID-19 pandemic is affecting people worldwide since the first reported case. Up to the end of October, more than 246 million cases and up to 5 million deaths have been reported1,2. The number of unreported cases is probably much higher. This makes the COVID-19 pandemic one of the deadliest in history. h Infection with the SARS-CoV-2 virus presents with a wide variety of symptoms—From none to life threaten- ing. This complicates the detection and containment of the virus. With the worldwide spread of the SARS-CoV-2 virus, a burden has been placed on the health care systems worldwide. Especially the intensive care units (ICU) as a limited resource were occupied with the care of COVID-19 patients. In some countries, the ICU capacities reached their limits3,4, resulting in catastrophic outcomes for the patients. Over the pandemic, admission rates to ICU and mortality rate varied strongly. In the early phase of the pandemic, first reports and characterisations based on smaller populations mainly in China suggested only mild symptoms in 80% of the cases5–8. 20% needed hospital treatment and 20% of the hospitalized patients required ICU treatment6,7,9. Those numbers vary within the European Union10. Estimates suggest that between 10 and 20% of SARS-CoV-2 patients become so severely ill, that hospital treatment is required. www.nature.com/scientificreports/ In addition, the proportion of infected people who require intensive care also varies between 5 and 32%10,11.h Therefore, it is important to be able to adequately calculate the resources of the health care system to avoid a collapse of the system in the event of another wave. The present study therefore provides up-to-date data based on which calculations could be made. The aim of this observational study is to describe the dynamic of the pan- demic in Germany and identify the underlying characteristics of hospitalized patients from January 2020 until the end of December 2021, based on data from the German Institute for Hospital Remuneration System (InEK). Jan Andreas Kloka1,2, Lea Valeska Blum1,2, Oliver Old1, Kai Zacharowski1 & Benjamin Friedrichson1* Jan Andreas Kloka1,2, Lea Valeska Blum1,2, Oliver Old1, Kai Zacharowski1 & Benjamin Friedrichson1* The ongoing SARS-CoV-2 pandemic is characterized by poor outcome and a high mortality especially in the older patient cohort. Up to this point there is a lack of data characterising COVID-19 patients in Germany admitted to intensive care (ICU) vs. non-ICU patients. German Reimbursement inpatient data covering the period in Germany from January 1st, 2020 to December 31th, 2021 were analyzed. 561,379 patients were hospitalized with COVID-19. 24.54% (n = 137,750) were admitted to ICU. Overall hospital mortality was 16.69% (n = 93,668) and 33.36% (n = 45,947) in the ICU group. 28.66% (n = 160,881) of all patients suffer from Cardiac arrhythmia and 17.98% (n = 100,926) developed renal failure. Obesity showed an odds-ratio ranging from 0.83 (0.79–0.87) for WHO grade I to 1.13 (1.08–1.19) for grade III. Mortality-rates peaked in April 2020 and January 2021 being 21.23% (n = 4539) and 22.99% (n = 15,724). A third peak was observed November and December 2021 (16.82%, n = 7173 and 16.54%, n = 9416). Hospitalized COVID-19 patient mortality in Germany is lower than previously shown in other studies. 24.54% of all patients had to be treated in the ICU with a mortality rate of 33.36%. Congestive heart failure was associated with a higher risk of death whereas low grade obesity might have a protective effect on patient survival. High admission numbers are accompanied by a higher mortality rate. Abbreviations CDC Centers for Disease Control and Prevention CI Confidence interval COVID-19 Coronavirus disease 2019 CPR Cardiopulmonary resuscitation DRG Diagnosis-related group ICD International classification of diseases ICU Intensive care unit IHCA In-hospital-cardiac-arrest InEK German Institute for Hospital Remuneration System KHG Krankenhausfinanzierungsgesetz (Hospital Financing Act) LOS Length of in-hospital stay Non-ICU Non-intensive care unit (= general ward) OPS Operation and procedure code RT-PCR Real-time-polymerase chain reaction 1Department of Anaesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Goethe University, Theodor‑Stern Kai 7, 60590 Frankfurt, Germany. 2These authors contributed equally: Jan Andreas Kloka and Lea Valeska Blum. *email: Benjamin.Friedrichson@kgu.de \| https://doi.org/10.1038/s41598-022-15287-3 www.nature.com/scientificreports/ Materials and methods Inclusion criteria. All hospitalized patients in Germany with proven SARS-CoV-2 infection between Janu- ary 1st, 2020 and December 31th, 2021 were included. Definitions and data acquisition. We divided patients into subgroups in dependence of admission to the ICU or to the general ward (non-ICU) and distinguished in both subgroups between survivors and non- survivors. The collected data included age, comorbidities (congestive heart failure, arterial hypertension, chronic pulmonary disease, diabetes and obesity), complications (acute renal failure, dialysis, cardiac arrhythmias, car- diopulmonary resuscitation (CPR), embolism, thrombosis, myocardial infarction, pulmonary embolism, intrac- ranial hemorrhage and stroke), length of hospital stay and mortality. Due to our findings associated with obesity, we used and subdivided obesity according to the WHO-definition into grade I, II and III. Diagnoses were coded according to the tenth revision of the International Classification of Diseases (ICD) and procedures were coded according to the International Classification of Procedures in Medicine in the version of 2020 (Table 1). The InEK only allows anonymized queries; there is no possibility to track a case back to a patient and therefore no further analysis is possible at the individual case level. In this study, we included only patients with a confirmed SARS-CoV-2 infection by Real-Time-(RT)-PCR (ICD U07.1) and admitted to hospital between January 1st, 2020 and December 31th, 2021. Existing comorbidities (e.g. arterial hypertension, ICD I10.x, I11.x–I13.x, I15.x) and complications were defined by their respective ICD codes (Table 1). Statistical analysis. The data were descriptively analyzed. Categorical variables are expressed as absolute numbers and percentages. Due to the lack of median and quartiles in the data source, continuous variables were https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ Table 2. Demographic data. Demographic data of inpatient in Germany from January 1st 2020 to December 31th 2021 with a positive SARS-CoV-2 PCR Test. Admission to general ward or admission to Intensive Care Unit (ICU). Consent for publication. As this are anonymised register data, no consensus of the patients can be col- lected. Consent for publication. As this are anonymised register data, no consensus of the patients can be col- lected. Materials and methods Demographic Total in-patient Admission to general ward Admission to ICU n % n % n % Total 561,379 100 423,611 75.46 137,750 24.54 Male 293,833 52.34 207,715 49.03 86,106 62.51 Female 267,516 47.65 215,874 50.96 51,636 37.49 Mortality 93,668 16.69 47,721 11.27 45,947 33.36 Age groups < 28 days 1105 0.20 901 0.21 204 0.15 28 days–1 year 3225 0.57 3050 0.72 175 0.13 1–2a 1704 0.30 1559 0.37 145 0.11 3–5a 1130 0.20 1021 0.24 109 0.08 6–9a 1352 0.24 1218 0.29 134 0.1 10–15a 2870 0.51 2613 0.62 257 0.19 16–17a 1781 0.32 1610 0.38 171 0.12 18–29a 22,400 3.99 19,887 4.69 2513 1.82 30–39a 34,052 6.07 28,928 6.83 5123 3.72 40–49a 43,916 7.82 34,096 8.05 9818 7.13 50–54a 33,878 6.03 24,920 5.88 8955 6.50 55–59a 42,169 7.51 29,751 7.02 12,417 9.01 60–64a 45,342 8.08 30,308 7.15 15,031 10.91 65–74a 97,435 17.36 64,845 15.31 32,588 23.66 75–79a 60,241 10.73 42,634 10.06 17,607 12.78 ≥ 80a 168,779 30.07 136,270 32.17 32,503 23.60 Table 2. Demographic data. Demographic data of inpatient in 31th 2021 with a positive SARS CoV 2 PCR Test Admission t Demographic Total in-patient Admission to general ward Admission to ICU n % n % n % Total 561,379 100 423,611 75.46 137,750 24.54 Male 293,833 52.34 207,715 49.03 86,106 62.51 Female 267,516 47.65 215,874 50.96 51,636 37.49 Mortality 93,668 16.69 47,721 11.27 45,947 33.36 Age groups < 28 days 1105 0.20 901 0.21 204 0.15 28 days–1 year 3225 0.57 3050 0.72 175 0.13 1–2a 1704 0.30 1559 0.37 145 0.11 3–5a 1130 0.20 1021 0.24 109 0.08 6–9a 1352 0.24 1218 0.29 134 0.1 10–15a 2870 0.51 2613 0.62 257 0.19 16–17a 1781 0.32 1610 0.38 171 0.12 18–29a 22,400 3.99 19,887 4.69 2513 1.82 30–39a 34,052 6.07 28,928 6.83 5123 3.72 40–49a 43,916 7.82 34,096 8.05 9818 7.13 50–54a 33,878 6.03 24,920 5.88 8955 6.50 55–59a 42,169 7.51 29,751 7.02 12,417 9.01 60–64a 45,342 8.08 30,308 7.15 15,031 10.91 65–74a 97,435 17.36 64,845 15.31 32,588 23.66 75–79a 60,241 10.73 42,634 10.06 17,607 12.78 ≥ 80a 168,779 30.07 136,270 32.17 32,503 23.60 Table 2. Demographic data. Demographic data of inpatient in Germany from January 1st 2020 to December 31th 2021 with a positive SARS-CoV-2 PCR Test. Admission to general ward or admission to Intensive Care Unit (ICU). Table 2. Demographic data. Materials and methods Demographic data of inpatient in Germany from January 1st 2020 to December 31th 2021 with a positive SARS-CoV-2 PCR Test. Admission to general ward or admission to Intensive Care Unit (ICU). calculated using mean and standard deviation (SD). Due to the aggregated data, only group comparisons of categorical variables were possible. The OR were calculated by means of 2 × 2 frequency tables. For this purpose, the Pearson Chi square test with an assumed significance level of 0.05 was utilized and the OR was determined with the 95% confidence interval (CI). Excel 2019 (Microsoft Corp., Seattle, WA, USA) and Python with SciPy and the statsmodels (sm) package were used for the analyses. calculated using mean and standard deviation (SD). Due to the aggregated data, only group comparisons of categorical variables were possible. The OR were calculated by means of 2 × 2 frequency tables. For this purpose, the Pearson Chi square test with an assumed significance level of 0.05 was utilized and the OR was determined with the 95% confidence interval (CI). Excel 2019 (Microsoft Corp., Seattle, WA, USA) and Python with SciPy and the statsmodels (sm) package were used for the analyses. Ethics approval and consent to participate. Due to the institutional anonymization, no conclusions can be drawn about individual patients. According to §21KHEntgG the reimbursement data is free for scien- tific use. The Ethics Committee of the University Hospital Frankfurt waived the need for an Ethical Committee approval for this study (Chair: Prof. Dr. Harder, Ref: 2022-766). All data processing was performed in accord- ance with the Declaration of Helsinki. Results Results A total of 561,379 patients were hospitalized with a confirmed SARS-CoV-2 infection in Germany from January 1st, 2020 to December 31th, 2021 and analyzed in this study. A total of 561,379 patients were hospitalized with a confirmed SARS-CoV-2 infection in Germany from January st, 2020 to December 31th, 2021 and analyzed in this study. Patients’ characteristics. The proportion of female hospitalized patients was 47.65% (n = 267,516) overall. In total, 75.46% (n = 423,611) non-ICU and 24.54% (n = 137,750) ICU patients (Table 2). The biggest group was aged 80 years or older, being 30.07% (n = 168,779) overall. Divided in groups ICU and non-ICU and subgroups survivors and non-survivors: Patients aged 80 years or older had the highest conditional (relative) frequencies in non-ICU overall (32.17% (n = 136,270)), non-ICU-survivors (26.72% (n = 100,430)), non-ICU-non-survivors (75.10% (n = 35,840)), ICU overall (32.17% (n = 136,270)) and ICU-non- survivors (37.04% (n = 17,020)). The biggest age group in ICU-survivors was 65–74 years (22.63% (n = 20,772). Length of in‑hospital stay (LOS) comorbidities, complications and mortality rate. In all patients the LOS was 11.2 (SD = 12.2) days in 2020 and 11.7 (SD = 12.4) days in 2021. The most common comor- bidity in all hospitalized patients was arterial hypertension (51.94% n = 291,577), followed by congestive heart failure (24.32% n = 136,505). In the non-ICU-group arterial hypertension was the most common comorbidity in survivors (47.82% (n = 179,738)) and non-survivors (61.79% (n = 29,489)), followed by congestive heart failure https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ Table 3. Comorbidities and complications. Comorbidities and complications among SARS-CoV-2 positive patients in Germany from Jan 2020 to end of December 2021. The percentages refer column wise to the corresponding group. OR odds ratio, CI confidence interval. Results Comorbidities and complications. Comorbidities and complications among SARS-CoV-2 positive patients in Germany from Jan 2020 to end of December 2021. The percentages refer column wise to the corresponding group. OR odds ratio, CI confidence interval. Table 3. Comorbidities and complications. Comorbidities and complications among SARS-CoV-2 positive patients in Germany from Jan 2020 to end of December 2021. The percentages refer column wise to the corresponding group. OR odds ratio, CI confidence interval. (17.25% (n = 179,738) in survivors and 61.79% (n = 29,489) in non-survivors, respectively. All other comorbidi- ties and their frequencies in subgroups are displayed in Table 3.h (17.25% (n = 179,738) in survivors and 61.79% (n = 29,489) in non-survivors, respectively. All other comorbidi- ties and their frequencies in subgroups are displayed in Table 3.h The most common complication was cardiac arrhythmia overall, coded in 28.66% (n = 160,881) of the patients overall and in every subgroup, as well. Especially in non-survivor group in ICU patients 58.04% (n = 26,667) and non-ICU patients 45.99% (n = 21,946), respectively. In each non-survivor group the rates were significantly higher, compared to the survivor group (Table 3).h Overall, 17.98% (n = 100,926) patients developed renal failure. The highest conditional (relative) frequency of patients needing dialysis was seen in non-survivors non-ICU-patients being 35.46% (n = 16,920). Cardiopul- monary resuscitation (CPR) was performed in 1.73% (n = 9728) of all COVID-19 inpatients. In non-survivor ICU-patients 14.23% (n = 6536) were treated with CPR. Pulmonary embolism occurred in 2.27% (n = 12,730) of all patients. p Our data shows an overall mortality rate of 16.69% (n = 93,668), 11.27% (n = 47,721) in non-ICU patients, and 33.36% (n = 45,947) in ICU-patients, respectively. Comparison survivors vs. non‑survivors. The highest OR for comorbidity was seen in congestive heart failure (Overall: OR: 3.81 (3.78–3.89) non-ICU: OR: 4.23 (4.15–4.31) ICU: OR: 2.42 (2.37–2.48)). Obesity WHO grade I and II were the only comorbidities in which an OR < 1 was seen. In particular in obesity WHO grade I (Overall: OR: 0.83 (0.79–0.87), non-ICU: OR: 0.55 (0.50–0.60), ICU: OR: 0.77 (0.72–0.82)) (Table 3).hif There is a significant difference between the groups of non-survivor/survivor within all complications (p < 0.001). The highest OR for the non-survivor groups were seen in CPR (Overall: OR: 20.83 (19.82–21.89) non-ICU: OR: 85.36 (70.31–103.64) ICU: OR: 7.86 (7.46–8.29)). Results Overall Non-ICU ICU Survivors Non-survivors p value OR (95% CI) Survivors Non-survivors p value OR (95% CI) Survivors Non-survivors p value OR (95% CI) Total 561,379 467,693 93,668 375,888 47,721 91,799 45,947 n % n % n % n % n % n % n % Comorbitiy Arterial hyperten- sion 291,577 51.94 233,782 49.99 57,795 61.7 < 0.001 1.61 (1.58–1.64) 179,738 47.82 29,489 61.79 < 0.001 1.76 (1.73–1.80) 54,012 58.84 28,295 61.58 < 0.001 1.12 (1.10–1.15) Chronic pulmonary disease 59,627 10.62 46,728 9.99 12,899 13.77 < 0.001 1.44 (1.41–1.47) 34,728 9.24 5327 11.16 < 0.001 1.23 (1.20–1.27) 11,975 13.04 7548 16.43 < 0.001 1.31 (1.27–1.35) Congestive heart failure 136,505 24.32 91,366 19.54 45,139 48.19 < 0.001 3.81 (3.78–3.89) 64,853 17.25 22,360 46.86 < 0.001 4.23 (4.15–4.31) 26,496 28.86 22,772 49.56 < 0.001 2.42 (2.37–2.48) Diabetes 40,830 7.27 29,807 6.37 11,023 11.77 < 0.001 1.96 (1.92–2.01) 21,978 5.85 5330 11.17 < 0.001 2.03 (1.96–2.09) 7751 8.44 5631 12.26 < 0.001 1.51 (1.46–1.57) Obesity (total) 36,580 6.52 30,501 6.52 6079 6.49 0.726 1.00 (0.97–1.02) 19,330 5.14 1387 2.91 < 0.001 0.55 (0.52–0.58) 11,142 12.14 4686 10.2 < 0.001 0.81 (0.79–0.85) WHO grade I 13,751 2.45 11,736 2.51 2015 2.15 < 0.001 0.83 (0.79–0.87) 8112 2.16 589 1.23 < 0.001 0.55 (0.50–0.60) 3619 3.94 1426 3.1 < 0.001 0.77 (0.72–0.82) WHO grade II 8585 1.53 7221 1.54 1364 1.46 0.047 0.92 (0.86–0.97) 4530 1.21 298 0.62 < 0.001 0.50 (0.44–0.56) 2685 2.92 1063 2.31 < 0.001 0.78 (0.72–0.83) WHO grade III 10,239 1.82 8309 1.78 1930 2.06 < 0.001 1.13 (1.08–1.19) 4689 1.25 309 0.65 < 0.001 0.50 (0.44–0.56) 3613 3.94 1621 3.53 < 0.001 0.88 (0.83–0.94) Complication Intracranial hemorrhage 2958 0.53 1434 0.31 1524 1.63 < 0.001 5.38 (5.00–5.78) 363 0.1 127 0.27 < 0.001 2.76 (2.25–3.38) 1042 1.14 1365 2.97 < 0.001 2.67 (2.46–2.89) Cardiac arrhyt- thmias 160,881 28.66 112,247 24 48,634 51.92 < 0.001 3.42 (3.37–3.47) 80,036 21.29 21,946 45.99 < 0.001 3.14 (3.09–3.21) 32,203 35.08 26,667 58.04 < 0.001 2.56 (2.50–2.62) Cardiopul- monary resuscitation 9728 1.73 2008 0.43 7720 8.24 < 0.001 20.83 (19.82- 21.89) 112 0.03 1184 2.48 < 0.001 85.36 (70.31– 103.64) 1896 2.07 6536 14.23 < 0.001 7.86 (7.46–8.29) Embolism/ thrombosis 2453 0.44 1525 0.33 928 0.99 < 0.001 3.06 (2.82–3.32) 529 0.14 184 0.39 < 0.001 2.75 (2.32 -3.25) 993 1.08 740 1.61 < 0.001 1.50 (1.36–1.65) Myocardial infarction 6726 1.20 3987 0.85 2739 2.92 < 0.001 3.50 (3.34–3.68) 1706 0.45 913 1.91 < 0.001 4.28 (3.95–4.64) 2266 2.47 1814 3.95 < 0.001 1.62 (1.53–1.73) Pulmonary embolism 12,730 2.27 8974 1.92 3756 4.01 < 0.001 2.14 (2.05–2.22) 4454 1.18 757 1.59 < 0.001 1.34 (1.24–1.45) 4520 4.92 2999 6.53 < 0.001 1.35 (1.28–1.41) Renal failure 100,926 17.98 71,277 15.24 29,649 31.65 < 0.001 2.58 (2.54–2.62) 56,744 15.1 16,920 35.46 < 0.001 3.09 (3.03–3.15) 14,527 15.82 12,721 27.69 < 0.001 2.03 (1.98–2.09) Renal replacement therapy 31,847 5.67 14,346 3.07 17,501 18.68 < 0.001 7.26 (7.09–7.43) 4398 1.17 884 1.85 < 0.001 1.59 (1.48–1.72) 9888 10.77 16,557 36.03 < 0.001 4.67 (4.54–4.80) Stroke 5004 0.89 3286 0.7 1718 1.83 < 0.001 2.64 (2.49–2.80) 1036 0.28 412 0.86 < 0.001 3.15 (2.81–3.53) 2250 2.45 1301 2.83 < 0.001 1.16 (1.08–1.24) Table 3. www.nature.com/scientificreports/ Mortality, vaccination and corona cases over time in Germany. Mortality of in hospital patients compared to COVID-19 vaccination and number of COVID-19 cases in Germany over time. Corona cases in Germany [n]. Fraction of mortality for ≥ 60 years [%]. Vaccinated percentage of age group ≥ 60 years [%] and total mortality [%] over time. Figure 1. Mortality, vaccination and corona cases over time in Germany. Mortality of in hospital patients compared to COVID-19 vaccination and number of COVID-19 cases in Germany over time. Corona cases i Germany [n]. Fraction of mortality for ≥ 60 years [%]. Vaccinated percentage of age group ≥ 60 years [%] an total mortality [%] over time. Table 4. Mortality. Mortality [%] over time by age subgroups from January 2020 to end of December 2021. www.nature.com/scientificreports/ Mortality Mar 20 Apr 20 May 20 Jun 20 Jul 20 Aug 20 Sep 20 Oct 20 Nov 20 Dec 20 Jan 21 Feb 21 Mar 21 Apr 21 May 21 Jun 21 Jul 21 Aug 21 Sep 21 Oct 21 Nov 21 Dec 21 16.58 21.23 14.35 8.61 6.82 6.32 6.75 9.65 15.70 21.78 22.99 18.69 13.93 13.40 13.29 10.71 8.05 5.53 9.68 13.84 16.82 16.54 Age groups <28 days 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 28 d–1 year 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.61 0.00 0.00 0.00 0.61 0.00 2.70 0.00 0.00 0.00 0.00 0.21 1–2 0.00 0.00 0.00 0.00 8.33 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.56 1.83 0.00 0.00 0.00 0.00 0.00 0.50 0.94 3–5 0.00 6.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.41 0.00 2.78 0.00 0.00 0.00 0.00 1.47 0.76 1.17 6–9 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.27 1.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.59 0.68 0.00 10–15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.27 0.44 0.00 0.00 0.00 0.00 0.00 1.14 0.00 0.00 0.00 0.56 1.29 0.00 0.44 16–17 0.00 3.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.58 1.00 18–29 0.79 0.37 1.20 0.00 0.00 1.44 0.00 0.23 0.48 0.26 0.55 0.55 0.59 0.42 0.91 0.23 0.29 0.16 0.92 0.57 0.82 0.82 30–39 0.24 0.95 1.79 0.93 0.00 0.44 0.75 0.56 0.48 1.00 1.10 1.49 1.46 0.97 1.56 2.14 1.40 0.55 1.42 2.01 1.28 1.63 40–49 1.05 2.42 3.87 1.70 2.39 0.58 0.52 1.06 1.56 2.27 3.11 2.97 2.10 2.24 3.60 4.11 3.43 1.52 2.80 3.36 3.14 3.90 50–54 3.09 4.66 4.74 3.55 2.14 2.40 2.02 1.34 2.70 4.14 6.03 5.96 3.68 4.33 5.22 6.72 7.98 1.60 4.63 4.72 5.80 7.43 55–59 4.28 6.87 7.79 5.11 3.41 3.46 2.75 2.36 4.18 6.70 8.91 7.43 5.58 6.23 8.27 9.35 9.17 4.46 5.72 8.31 8.04 10.28 60–64 7.25 11.48 8.18 8.84 8.75 6.13 4.15 3.70 7.12 11.33 12.02 10.93 8.72 9.30 10.11 11.42 10.39 7.41 9.40 11.20 11.69 12.98 65–74 12.97 18.97 15.01 10.93 8.25 7.76 6.77 8.56 12.28 17.75 20.14 16.75 14.01 16.52 17.07 14.53 13.41 12.49 15.92 17.09 17.13 19.29 75–79 26.04 27.86 18.27 12.04 15.57 14.38 18.01 14.39 20.95 25.01 24.56 22.23 19.57 22.63 22.78 15.41 14.61 17.56 23.85 20.00 22.71 21.24 ≥ 80 49.69 38.33 21.43 14.30 15.14 19.21 20.93 28.50 33.22 35.72 34.20 27.78 26.40 29.72 27.59 17.08 13.30 24.15 28.26 30.38 31.91 28.52 Sex Male 18.48 23.41 16.59 9.20 7.99 7.32 7.80 10.64 17.27 24.41 25.85 21.26 15.46 15.06 15.04 12.38 8.70 6.17 10.64 15.38 18.77 18.84 Female 13.92 18.59 11.76 7.86 5.26 5.12 5.54 8.53 13.98 19.09 20.06 15.96 12.22 11.47 11.16 8.65 7.27 4.86 8.62 12.17 14.73 13.98 Figure 1. www.nature.com/scientificreports/ Mortality Mar 20 Apr 20 May 20 Jun 20 Jul 20 Aug 20 Sep 20 Oct 20 Nov 20 Dec 20 Jan 21 Feb 21 Mar 21 Apr 21 May 21 Jun 21 Jul 21 Aug 21 Sep 21 Oct 21 Nov 21 Dec 21 16.58 21.23 14.35 8.61 6.82 6.32 6.75 9.65 15.70 21.78 22.99 18.69 13.93 13.40 13.29 10.71 8.05 5.53 9.68 13.84 16.82 16.54 Age groups <28 days 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 28 d–1 year 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.61 0.00 0.00 0.00 0.61 0.00 2.70 0.00 0.00 0.00 0.00 0.21 1–2 0.00 0.00 0.00 0.00 8.33 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.56 1.83 0.00 0.00 0.00 0.00 0.00 0.50 0.94 3–5 0.00 6.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.41 0.00 2.78 0.00 0.00 0.00 0.00 1.47 0.76 1.17 6–9 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.27 1.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.59 0.68 0.00 10–15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.27 0.44 0.00 0.00 0.00 0.00 0.00 1.14 0.00 0.00 0.00 0.56 1.29 0.00 0.44 16–17 0.00 3.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.58 1.00 18–29 0.79 0.37 1.20 0.00 0.00 1.44 0.00 0.23 0.48 0.26 0.55 0.55 0.59 0.42 0.91 0.23 0.29 0.16 0.92 0.57 0.82 0.82 30–39 0.24 0.95 1.79 0.93 0.00 0.44 0.75 0.56 0.48 1.00 1.10 1.49 1.46 0.97 1.56 2.14 1.40 0.55 1.42 2.01 1.28 1.63 40–49 1.05 2.42 3.87 1.70 2.39 0.58 0.52 1.06 1.56 2.27 3.11 2.97 2.10 2.24 3.60 4.11 3.43 1.52 2.80 3.36 3.14 3.90 50–54 3.09 4.66 4.74 3.55 2.14 2.40 2.02 1.34 2.70 4.14 6.03 5.96 3.68 4.33 5.22 6.72 7.98 1.60 4.63 4.72 5.80 7.43 55–59 4.28 6.87 7.79 5.11 3.41 3.46 2.75 2.36 4.18 6.70 8.91 7.43 5.58 6.23 8.27 9.35 9.17 4.46 5.72 8.31 8.04 10.28 60–64 7.25 11.48 8.18 8.84 8.75 6.13 4.15 3.70 7.12 11.33 12.02 10.93 8.72 9.30 10.11 11.42 10.39 7.41 9.40 11.20 11.69 12.98 65–74 12.97 18.97 15.01 10.93 8.25 7.76 6.77 8.56 12.28 17.75 20.14 16.75 14.01 16.52 17.07 14.53 13.41 12.49 15.92 17.09 17.13 19.29 75–79 26.04 27.86 18.27 12.04 15.57 14.38 18.01 14.39 20.95 25.01 24.56 22.23 19.57 22.63 22.78 15.41 14.61 17.56 23.85 20.00 22.71 21.24 ≥ 80 49.69 38.33 21.43 14.30 15.14 19.21 20.93 28.50 33.22 35.72 34.20 27.78 26.40 29.72 27.59 17.08 13.30 24.15 28.26 30.38 31.91 28.52 Sex Male 18.48 23.41 16.59 9.20 7.99 7.32 7.80 10.64 17.27 24.41 25.85 21.26 15.46 15.06 15.04 12.38 8.70 6.17 10.64 15.38 18.77 18.84 Female 13.92 18.59 11.76 7.86 5.26 5.12 5.54 8.53 13.98 19.09 20.06 15.96 12.22 11.47 11.16 8.65 7.27 4.86 8.62 12.17 14.73 13.98 Table 4. Results Followed by renal replacement therapy in patients overall (OR: 7.26 (7.09–7.43)), renal failure in non-ICU patients (OR: 3.09 (3.03–3.15)) and embolism or thrombosis in non-ICU patients (OR: 2.75 (2.32–3.25)). (Table 3). Analysis in the context of time course. The observed in-hospital mortality-rates fluctuated during the analyzed and showed three peaks: April 2020 and January 2021 being 21.23% (n = 4539) and 22.99% (n = 15,724) and in November, December 2021 (16.82%, n = 7173 and 16.54%, n = 9416) (Fig. 1, Table 4). https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ Within the whole-time patients aged 60 years or older had the highest mortality in all hospitalized patients. Analyzed by the time trend of the data divided into months from January 2021 until the end of December 2021, in every month the highest mortality was observed in the age group 80–85 years with particularly high mortality Figure 1. Mortality, vaccination and corona cases over time in Germany. Mortality of in hospital patients compared to COVID-19 vaccination and number of COVID-19 cases in Germany over time. Corona cases in Germany [n]. Fraction of mortality for ≥ 60 years [%]. Vaccinated percentage of age group ≥ 60 years [%] and total mortality [%] over time. Table 4. Mortality. Mortality [%] over time by age subgroups from January 2020 to end of December 2021. Discussion Th Compared to the German population (17%, in 2020) and the share of SARS-CoV-2 infected persons (29%), the group of children (< 18 years) is significantly underrepresented among all hospitalised patients (2.35%, Table 2)18,19. Furthermore, 9% (< 18 years) against 26% (> 60 years) conditional on the respective group were admitted to ICU. There a various explanation for this finding. Children have less comorbidities such as obesity, diabetes, cancer and other chronic diseases. Children and adults show a different immune response to a viral infection, protecting children from severe COVID-1920. Our findings can also be explained with a difference in the expression of ACE2-receptor, the primary receptor of SARS-CoV-221. Children express lower levels of ACE221 and therefore have fewer symptoms and a better prognosis22. Comorbidities. Arterial hypertension was the most common comorbidity in our study (51.94% (n = 291,577) overall), this result is in line with other studies describing hypertension as the most common comorbidity with rates of 42%23 up to 56%16.h Obesity is known to be a risk factor for COVID-1924 and is associated with higher mortality11. The incidence of 6.52% for obesity is lower than rates described in the UK (10.2%)11 or France (47.6%)25. Surprisingly our data showed a reduced odd for death in patients with low grade obesity, especially WHO grade I (non-ICU: 0.55 (0.50–0.60) and ICU: 0.77 (0.72–0.82), respectively). Dana et al. found a lower risk of death for critically ill COVID-19 patients among patients with moderate obesity26. This finding is not in line with other studies27. However, the Centers for Disease Control and Prevention (CDC) states that patients at the “threshold” between healthy weight and overweight are at lower risk for hospitalization, ICU admission, and death while the risk for mechanically ventilation raised continuously with rising body mass index (BMI)28. Our findings support the statement of the CDC. Patients with a low grade overweight (WHO grade I) have a higher chance of survival (With increasing obesity, the effect is lost and the overweight becomes a risk factor for the patient (WHO grade III, OR: 1.13 (1.08–1.19)). However, overweight and moderate obesity are known to have a protective effect on hospitalized non-COVID-19 ICU-Patients29. This effect is not yet fully understood and often controversially referred to as “obesity paradox”30. With a lower rate of high grade obese patients admitted to the hospital, more patients are at the “threshold” described by the CDC. Foo et al. Discussion Th This retrospective study includes a cohort of 561,379 hospitalized patients from January 1st, 2020 to December 31th, 2021 with a positive SARS-CoV-2 PCR test. Our main findings are a mortality rate of 16.69% overall. Between months of the ongoing pandemic, the mortality fluctuates. Especially in the “third wave”, a noticeable decrease in the proportion of elderly patients was seen.hi The COVID-19 pandemic in Germany in the analyzed period can be described in four waves: the first one from March until May 202012, the second one from October 2020 until January 202112 and a third wave from March until April 202113. In the end of 2021 Germany went through the fourth wave. The cause of the undulating course of infection numbers is multifactorial, main reasons are higher temperatures in spring/summer14, social distancing and shutdown measures by the government13. Mortality rate. The mortality rate of 16.69% overall is lower than previously described mortality-rates. Richardson et al. described 5700 hospitalized patients diagnosed with COVID-19 in New York from March until April 2020 and found a mortality rate of 21%, overall but a particularly lower rate of ICU-patients (12.2%)15. Karagiannidis et al. described over 10,000 patients suffering from COVID-19 in Germany from February until April 2020 in Germany and found a mortality rate of 22% overall16. This might be explained by the fact that pre- vious studies describe a shorter time period. In December of 2020 the vaccine against COVID-19 was licensed in Germany17. Over time, more and more treatment options are being explored, and guidelines for the treatment of COVID-19 infection continue to evolve on a regular basis. Due to prioritization rules the vaccines were only available for patients over 75 years of age, employees in high-risk facilities and patients with defined comorbidities were vaccicinated17. The prioritization was lifted in May 2021. The first “vaccination effect” may occur in January 2021, which could explain the decreasing propor- tion of patients aged 60 years or older from there on. p g y In contrast to that we found higher mortality rates in ICU patients, being 33.36%. A possible explanation might the higher capacity of ICU-beds in germany16 and therefore the possibility to admit patients with more comorbidities or at higher age to the ICU, while in some countries ICUs reached their capacity limit3. g g p y Our data shows a distinct mortality between minors (< 18 years) and old adults (> 60 years). www.nature.com/scientificreports/ Mortality. Mortality [%] over time by age subgroups from January 2020 to end of December 2021. Within the whole-time patients aged 60 years or older had the highest mortality in all hospitalized patients. Analyzed by the time trend of the data divided into months from January 2021 until the end of December 2021, in every month the highest mortality was observed in the age group 80–85 years, with particularly high mortality in months when hospitalization rates were the highest (Fig. 1, Table 4). https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ Conclusionh The overall mortality rate of 16.69% in COVID-19 patients in Germany is lower than previously shown in other studies. 24.54% of all patients had to be treated on the ICU with a mortality rate of 33.36%, which was high in comparison to the literature. Congestive heart failure was associated with a significantly higher risk of death. In non-ICU patients suffering from congestive heart failure the OR was especially high, being 4.23 (4.15–4.31). The most common comorbidity in all COVID-19 patients was arterial hypertension. The most common complication were arrhythmias, which were diagnosed significantly more often in non-survivors (p < 0.001). In COVID-19 patients CPR is associated with a high chance of death, especially on normal wards (OR: 85.36 (70.31–103.64)). With an OR: 0.83 (0.79–0.87) WHO grade I obesity might have a protective effect of the patient’s survival.f f Pre-existing cardiac conditions appear to carry a particularly high risk in patients suffering from COVID-19. Due to the limited data, no further research could be done, so it is extremely important to make this data available to the scientific community in order to gain a wider insight and better understand possible causal relationships. Limitations. Due to the provision of data by the InEK during the year, only highly aggregated data are avail- able and no further detailed queries are possible. Patients’ age is only available in the provided subgroups, so no further analysis like mean or median age can be made. As the data (e.g. comorbidities) were anonymized and cannot be tracked back to a single patient further analysis such as linear regression are not possible. Laboratory findings or medication are not coded for reimbursement and are therefore not available for analysis. www.nature.com/scientificreports/ A possible cause for the high differ- ence between non-ICU and ICU patients might be the longer timespan between the circulatory arrest and the actual start of CPR due to the lack of monitoring48.hi g This study is the first in Germany to describe the course of the pandemic from the beginning to the end of December 2021, covering 561 379 hospitalized COVID-19 Patients. g p Further studies should be conducted to identify risk factors for an adverse course of the disease. On one hand, this could help to identify patients with a special risk profile at an early stage (for example Simonnet et al. investigated obesity as a risk factor25) or, on the other hand, serve as criteria in possible triaging. www.nature.com/scientificreports/ One underlying reason may be the mismatch between oxygen supply and oxygen demand resulting in cardiac injury. Therefore, the high OR for death is not surprising.hth j yh g p g Thrombotic events are often seen in COVID-19 patients. Those events are associated with a more severe disease and increased mortality38,39. The molecular explanation range from COVID-19-associated coagulopathy to genetic predisposition and is still subject to research40,41. Our findings are in line with literature, we showed a pulmonary embolism rate of 2.27% (n = 12,730) for all hospitalized COVID-19 patient. Patients suffering from pulmonary embolism have a higher chance of dying (OR: 2.14 (2.05–2.22)). OR on ICU is lower (OR: 1.35 (1.28–1.41). One explanation for this could be that more radiological examinations are performed on ICU and therefore associated with a higher number of incidental findings without clinical relevance42. A second explana- tion is that patients on ICU are more likely to be sufficiently anticoagulated, as recommended in guidelines43,44. Unfortunately, medication application such as anticoagulation is not provided in the reimbursement data.fh y pp g p Renal failure is a common complication in patients suffering from COVID-1915,45. The observed rate of patients with acute kidney failure (17.98% n = 100,926) is in line with other studies, describing rates between 20 and 40%15,45. The higher proportion of patients with renal replacement therapy (RRT) in the ICU-group might be explained by the fact that needing dialysis is a common reason to be admitted to the ICU. Patients who had been diagnosed with acute kidney failure and a following admission to the ICU are displayed in the ICU-group only. In our study 5.67% (n = 31,847) patients needed RRT, this result seems to be higher than in other studies, for example Richardson et al. described a rate of 3.2%.f p COVID-19 patients are more likely to suffer from in-hospital-cadiac-arrest (IHCA)46. In our study 1.73% (n = 9728) of all patients received CPR. The OR of dying was remarkably higher on the normal ward compared to the ICU. This is in accordance to the literature. Acharya et al. described a IHCA for non-ICU patients in 2.2% (in our study: 2.48%) and for ICU patients in 15.4% (in our study 14.2%)47. Data availabilityh y The German Institute for Hospital Remuneration System (InEK) supports hospitals and health insurance funds as well as their associations in the introduction and continuous further development of the German-DRG system in accordance with the Krankenhausfinanzierungsgesetz (KHG; Hospital Financing Act). Since 2020, access to these data has been possible during the year. For this observational study, we used publicly accessible performance data provided by InEK. Since the register data were anonymized, the Ethics Committee of the University Hospital Frankfurt waived the need for an Ethical approval (Chair: Prof. Dr. Harder, Ref: 2022-766). Received: 15 February 2022; Accepted: 22 June 2022 Discussion Th described for every 1% increase in obesity preva- lence, the COVID-19 mortality rate was increased by 8.3%31. This might explain our finding.iiff hi We could find a significant difference in the frequency of patients suffering from congestive heart failure between survivors and non-survivors, with a bigger frequency of heart failure in non-survivors. One possible explanation is that patients suffering from heart failure have a higher overall frailty and are at higher risk for acute cardiac injury. Several studies investigated cardiovascular manifestations and mechanisms in patients suffering from COVID-19 and showed a high prevalence of cardiovascular comorbidities32 and a higher risk for mortality in patients suffering from cardiovascular comorbidities or cardiac injury32,33. Jabri et al. described a significant increase in stress cardiomyopathy during the pandemic34, furthermore a frequent occurrence of cor pulmonale in COVID-19 patients has been described35. All this factors explain the high proportion of cardiac comorbidities in hospitalized patients and the increase of those diagnoses in the non-survivor groups. Complications. We found arrhythmias to be the most common complication in our study. We showed a rate of 42.74% (n = 58,870) for ICU patients. This result is comparable to a study by Duo et al. who described a rate of 44.4% among ICU patients in a study investigating 138 patients overall32.tf g p y g g p Electrocardiographic abnormalities are often described in patients suffering from COVID-1936,37. https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ www.nature.com/scientificreports/ 4. Remuzzi, A. & Remuzzi, G. COVID-19 and Italy: What next? Lancet 395, 1225–1228. https://doi.org/10.1016/s0140-6736(20) 30627-9 (2020). 4. Remuzzi, A. & Remuzzi, G. COVID-19 and Italy: What next? Lancet 395, 1225–1228. https://doi.org/10.1016/s0140-6736(20) 30627-9 (2020). 5 W Z & M G J M Ch i i f d i l f h i di 2019 (COVID 19) b k i ( ) 5. Wu, Z. & McGoogan, J. M. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: Summary of a report of 72314 cases from the Chinese Center for Disease Control and Prevention. JAMA 323, 1239–1242 https://doi.org/10.1001/jama.2020.2648 (2020). g J p ( ) China: Summary of a report of 72314 cases from the Chinese Center for Disease Control and Prevention. JAMA 323, 1239–1242. https://doi.org/10.1001/jama.2020.2648 (2020). 6. Guan, W. J. et al. Clinical characteristics of coronavirus disease 2019 in China. N. Engl. J. Med. 382, 1708–1720. https://doi.org/ y p J https://doi.org/10.1001/jama.2020.2648 (2020). 6. Guan, W. J. et al. Clinical characteristics of coronavirus disease 2019 in China. N. Engl. J. Med. 382, 1708–1720. https://doi.org/ 10 1056/NEJMoa2002032 (2020) p g j 6. Guan, W. J. et al. Clinical characteristics of coronavirus disease 2019 in China. N. Engl. J. Med. 382, 1708–1720. https://doi.org/ 10.1056/NEJMoa2002032 (2020). 7. Huang, C. et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 395, 497–506. https:// doi.org/10.1016/S0140-6736(20)30183-5 (2020).h g ( ) ( ) 8. Du, R. H. et al. Hospitalization and critical care of 109 decedents with COVID-19 pneumonia in Wuhan, China. Ann. Am. Thorac Soc. 17, 839–846. https://doi.org/10.1513/AnnalsATS.202003-225OC (2020). p g 9. Wu, Z. & McGoogan, J. M. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: Summary of a report of 72,314 cases from the Chinese Center for Disease Control and Prevention. JAMA 323, 1239–1242. https://doi.org/10.1001/jama.2020.2648 (2020). p g j 10. European Centre for Disease Prevention and Control, A. a. o. t. E. U. Data on Hospital and ICU Admission Rates and Current Occupancy for COVID-19 https://www.ecdc.europa.eu/en/publications-data/download-data-hospital-and-icu-admission-rates- and-current-occupancy-covid-19. Accessed 3 May 2022. p y y 11. Docherty, A. B. et al. Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO clinical characterisation protocol: Prospective observational cohort study. BMJ 369, m1985. https://doi.org/10.1136/bmj.m1985 (2020). h ll l h d h d d hl d d k l 12. Schilling, J. et al. www.nature.com/scientificreports/ Die verschiedenen Phasen der COVID-19-Pandemie in Deutschland: Eine deskriptive analyse von Januar 2020 bis Februar 2021. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 64, 1093–1106. https://doi.org/10.1007/s00103- 021-03394-x (2021).f 13. Schuppert, A., Polotzek, K., Karschau, J. & Karagiannidis, C. Effectiveness of extended shutdown measures during the ‘Bundesnot- bremse’ introduced in the third SARS-CoV-2 wave in Germany. Infection 49, 1331–1335. https://doi.org/10.1007/s15010-021- 01713-7 (2021). 14. Fan, G. et al. Decreased case fatality rate of COVID-19 in the second wave: A study in 53 countries or regions. Transbound. Emerg. Dis. 68, 213–215. https://doi.org/10.1111/tbed.13819 (2021). p g 5. Richardson, S. et al. Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York city area. JAMA 323, 2052–2059. https://doi.org/10.1001/jama.2020.6775 (2020). y g j 6. Karagiannidis, C. et al. Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 Ger- man hospitals: An observational study. Lancet Respir. Med. 8, 853–862. https://doi.org/10.1016/S2213-2600(20)30316-7 (2020).tk 16. Karagiannidis, C. et al. Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 Ger- man hospitals: An observational study. Lancet Respir. Med. 8, 853–862. https://doi.org/10.1016/S2213-2600(20)30316-7 (2020). 17. Vygen-Bonnet, S. et al. Beschluss und Wissenschaftliche Begründung der Ständigen Impfkommission (STIKO) für die COVID- p y p p g 17. Vygen-Bonnet, S. et al. Beschluss und Wissenschaftliche Begründung der Ständigen Impfkommission (STIKO) für die COVID- 19-Impfempfehlung, 3–63. https://doi.org/10.25646/7755.2 (2020). pf pf g p g ( ) 8. Robert-Koch-Institut. COVID-19-Fälle nach Altersgruppe und Meldewoche https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_ Coronavirus/Daten/Altersverteilung.html (2022). Accessed 24 May 2022. g y 19. Bundesamt, S. (2022). 19. Bundesamt, S. (2022). 0. Yoshida, M. et al. Local and systemic responses to SARS-CoV-2 infection in children and adults. Nature 602, 321–327. https://doi org/10.1038/s41586-021-04345-x (2022). 0. Yoshida, M. et al. Local and systemic responses to SARS-CoV-2 infection in children and adults. Nature 602, 321–327. https://doi org/10.1038/s41586-021-04345-x (2022). 1 V d h M l R i i i ld i hibi i i i h C id 19 N E l J M d 382 1653 y p p org/10.1038/s41586-021-04345-x (2022). 21 V d th M t l R i i t i ld t t i hibit i ti t ith C id 19 N E l J M d 382 1 g 1. Vaduganathan, M. et al. Renin-angiotensin-aldosterone system inhibitors in patients with Covid-19. N. Engl. J. Med. 382, 1653– 1659. https://doi.org/10.1056/NEJMsr2005760 (2020). 22. Ludvigsson, J. F. References References 1. COVID-19 Dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU). https://gisan ddata.maps.arcgis.com/apps/dashboards/bda7594740fd40299423467b48e9ecf6. Accessed 25 May 2022. p g pp y 2. Dong, E., Du, H. & Gardner, L. An interactive web-based dashboard to track COVID-19 in real time. Lancet. Infect. Dis 20, 533–534 https://doi.org/10.1016/S1473-3099(20)30120-1 (2020).h p g 3. Immovilli, P. et al. COVID-19 mortality and ICU admission: The Italian experience. Crit. Care 24, 228. https://doi.org/10.1186/ s13054-020-02957-9 (2020). https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| www.nature.com/scientificreports/ Systematic review of COVID-19 in children shows milder cases and a better prognosis than adults. Acta Paediatr. 109, 1088–1095. https://doi.org/10.1111/apa.15270 (2020). 109, 1088–1095. https://doi.org/10.1111/apa.15270 (2020). p g p 3. Grasselli, G. et al. Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy, Italy JAMA Intern. Med. 180, 1345–1355. https://doi.org/10.1001/jamainternmed.2020.3539 (2020). 23. Grasselli, G. et al. Risk factors associated with mortality among patients with COVID-19 in intensiv JAMA Intern. Med. 180, 1345–1355. https://doi.org/10.1001/jamainternmed.2020.3539 (2020). 23. Grasselli, G. et al. Risk factors associated with mortality amon p g j 4. Kwok, S. et al. Obesity: A critical risk factor in the COVID-19 pandemic. Clin. Obes. 10, e12403. https://doi.org/10.1111/cob.12403 (2020). 5. Simonnet, A. et al. High prevalence of obesity in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) requiring invasive mechanical ventilation. Obesity (Silver Spring) 28, 1195–1199. https://doi.org/10.1002/oby.22831 (2020). y p g g y 6. Dana, R. et al. Obesity and mortality in critically ill COVID-19 patients with respiratory failure. Int. J. Obes. 45, 2028–2037. https:// doi.org/10.1038/s41366-021-00872-9 (2021). g 27. Poly, T. N. et al. Obesity and mortality among patients diagnosed with COVID-19: A systematic review and meta-analysis. Front. Med. https://doi.org/10.3389/fmed.2021.620044 (2021). 28. Kompaniyets, L. et al. Body mass index and risk for COVID-19-related hospitalization, intensive care unit admission, invasive mechanical ventilation, and death-United States, March-December 2020. Morb. Mortal Wkly. Rep. 70, 355–361. https://doi.org/ 10.15585/mmwr.mm7010e4 (2021). ( ) 9. Schetz, M. et al. Obesity in the critically ill: A narrative review. Intens. Care Med 45, 757–769. https://doi.org/10.1007/s00134-019- 05594-1 (2019). 30. Donini, L. M., Pinto, A., Giusti, A. M., Lenzi, A. & Poggiogalle, E. Obesity or BMI paradox? Beneath the tip of the Iceberg. Front. Nutr. 7, 53. https://doi.org/10.3389/fnut.2020.00053 (2020). p g 1. Foo, O., Hiu, S., Teare, D., Syed, A. A. & Razvi, S. A global country-level analysis of the relationship between obesity and COVID- 19 cases and mortality. Diabetes Obes. Metab. 23, 2697–2706. https://doi.org/10.1111/dom.14523 (2021). y , p g ( ) 32. Dou, Q., Wei, X., Zhou, K., Yang, S. & Jia, P. Cardiovascular manifestations and mechanisms in patients with COVID-19. T d l b h //d / / ( ) y p g 32. Dou, Q., Wei, X., Zhou, K., Yang, S. & Jia, P. Cardiovascular manifestations and mechanisms in patients with COVID-19. T Endocrinol. Metab. 31, 893–904. https://doi.org/10.1016/j.tem.2020.10.001 (2020). 32. Dou, Q., Wei, X., Zhou, K., Yang, S. & Jia, P. Cardiovascular manifestations and mechan Endocrinol. Competing interests h p g The authors declare no competing interests. Author contributions L.V.B. and J.K. wrote the manuscript and were in charge of planning the study in close consultation with B.F. and K.Z. B.F. conceived the study and was in charge of overall direction and planning. Statistical analysis and proofreading was done by O.O. All authors contributed to the final version of the manuscript. www.nature.com/scientificreports/ Location of in-hospital cardiac arrest in the United States—Variability in event rate and outcomes. J. Am. Hear Assoc. 5, e003638. https://doi.org/10.1161/JAHA.116.003638 (2016). www.nature.com/scientificreports/ Metab. 31, 893–904. https://doi.org/10.1016/j.tem.2020.10.001 (2020). p g j 33. Shi, S. et al. Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China. JAMA Cardiol. 5, 802–810. https://doi.org/10.1001/jamacardio.2020.0950 (2020). p g j 4. Jabri, A. et al. Incidence of stress cardiomyopathy during the coronavirus disease 2019 pandemic. JAMA Netw. Open 3, e2014780 https://doi.org/10.1001/jamanetworkopen.2020.14780 (2020). p g j p 5. Creel-Bulos, C. et al. Acute cor pulmonale in critically ill patients with covid-19. N. Engl. J. Med. 382, e70. https://doi.org/10.1056/ NEJMc2010459 (2020). 36. He, J. et al. Characteristic electrocardiographic manifestations in patients with COVID-19. Can. J. Cardiol. 36(966), e961–e966. https://doi.org/10.1016/j.cjca.2020.03.028 (2020). 37. Angeli, F. et al. Electrocardiographic features of patients with COVID-19 pneumonia. Eur. J. Intern. Med. 78, 101–106. https://doi. org/10.1016/j.ejim.2020.06.015 (2020).h g j j 8. Klok, F. A. et al. Incidence of thrombotic complications in critically ill ICU patients with COVID-19. Thromb. Res. 191, 145–147 https://doi.org/10.1016/j.thromres.2020.04.013 (2020).h p g j 9. Middeldorp, S. et al. Incidence of venous thromboembolism in hospitalized patients with COVID-19. J. Thromb. Haemost. 18 1995–2002. https://doi.org/10.1111/jth.14888 (2020). 40. Loo, J., Spittle, D. A. & Newnham, M. COVID-19, immunothrombosis and venous thromboembolism: Biological mechanisms. Thorax 76, 412–420. https://doi.org/10.1136/thoraxjnl-2020-216243 (2021). https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \| Funding g Open Access funding enabled and organized by Projekt DEAL. This study was supported by internal institutional research funds from the Department of Anaesthesiology, Intensive Care and Pain Therapy, University Hospital Frankfurt, Goethe University, Frankfurt Germany. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 41. Polak, S. B., Van Gool, I. C., Cohen, D., von der Thusen, J. H. & van Paassen, J. A systematic review of pathological findings in COVID-19: A pathophysiological timeline and possible mechanisms of disease progression. Mod. Pathol. 33, 2128–2138. https:// doi.org/10.1038/s41379-020-0603-3 (2020).i 41. Polak, S. B., Van Gool, I. C., Cohen, D., von der Thusen, J. H. & van Paassen, J. A systematic review of pathological findings in COVID-19: A pathophysiological timeline and possible mechanisms of disease progression. Mod. Pathol. 33, 2128–2138. https:// doi.org/10.1038/s41379-020-0603-3 (2020).i g ( ) 2. Lumbreras, B., Donat, L. & Hernandez-Aguado, I. Incidental findings in imaging diagnostic tests: A systematic review. Br. J. Radiol 83, 276–289. https://doi.org/10.1259/bjr/98067945 (2010). p g j 3. Kluge, S. et al. German recommendations for treatment of critically ill patients with COVID-19-version 3: S1-guideline. Anaesthesis 69, 653–664. https://doi.org/10.1007/s00101-020-00833-3 (2020). p g ( ) 4. Kluge, S. et al. Clinical practice guideline: Recommendations on the In-hospital treatment of patients with COVID-19. Dtsch Arztebl. Int. 118, 865–871. https://doi.org/10.3238/arztebl.m2021.0374 (2021). p g 44. Kluge, S. et al. Clinical practice guideline: Recommendations on the In-hospital treatment of patients with COVID-19. Dtsch. Arztebl. Int. 118, 865–871. https://doi.org/10.3238/arztebl.m2021.0374 (2021). p g 5. Ronco, C., Reis, T. & Husain-Syed, F. Management of acute kidney injury in patients with COVID-19. Lancet Respir. Med. 8 738–742. https://doi.org/10.1016/s2213-2600(20)30229-0 (2020).t p g ( ) ( ) 6. Ippolito, M. et al. Mortality after in-hospital cardiac arrest in patients with COVID-19: A systematic review and meta-analysis Resuscitation 164, 122–129. https://doi.org/10.1016/j.resuscitation.2021.04.025 (2021). p g 46. Ippolito, M. et al. Mortality after in-hospital cardiac arrest in patients with COVID-19: A systematic review and meta-analysis. Resuscitation 164, 122–129. https://doi.org/10.1016/j.resuscitation.2021.04.025 (2021). p g j 47. Acharya, P. et al. Incidence, predictors, and outcomes of in-hospital cardiac arrest in COVID-19 patients admitted to intensive and non-intensive care units: Insights from the AHA COVID-19 CVD registry. J. Am. Heart Assoc. 10, e021204. https://doi.org/ 10.1161/JAHA.120.021204 (2021). p g j 7. Acharya, P. et al. Incidence, predictors, and outcomes of in-hospital cardiac arrest in COVID-19 patients admitted to intensive and non-intensive care units: Insights from the AHA COVID-19 CVD registry. J. Am. Heart Assoc. 10, e021204. https://doi.org/ 10.1161/JAHA.120.021204 (2021). J ( ) 48. Perman, S. M. et al. Location of in-hospital cardiac arrest in the United States—Variability in event rate and outcomes. J. Am. Heart Assoc. 5, e003638. https://doi.org/10.1161/JAHA.116.003638 (2016). 8. Perman, S. M. et al. © The Author(s) 2022 Additional information Correspondence and requests for materials should be addressed to B.F. Correspondence and requests for materials should be addressed to B.F. Reprints and permissions information is available at www.nature.com/reprints. Reprints and permissions information is available at www.nature.com/reprints. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and nstitutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2022 https://doi.org/10.1038/s41598-022-15287-3 Scientific Reports \| (2022) 12:11116 \|
https://openalex.org/W2888405766	https://figshare.com/ndownloader/files/13109270	English	null	Expression of a recombinant, 4’-Phosphopantetheinylated, active <i>M. tuberculosis</i> Fatty acid Synthase I in <i>E. coli</i>	bioRxiv (Cold Spring Harbor Laboratory)	2,018	cc-by	412	S1 method: Purification of Mtb FAS I from M. smegmatis mc2 2700 strain All purification steps described were held at 4°C. The relevant fractions containing FAS I following each purification step were identified by analysis on SDS -PAGE gel. M. smegmatis mc2 2700 bacilli were grown at 37°C to O.D 600=1.5-2, in 7H9 broth supplemented with glycerol at 0.5% v/v and Tween 80 at 0.05%. The bacilli were spun down by continuous flow centrifuge apparatus (CEPA®) at 30, 000 RPM for 30 minutes. The pellet was washed twice with PBS, spun at 3000 RPM for 10 minutes and resuspended in Buffer A (KPB 50 mM pH 7.2, KCl 250 mM, glycerol 5% v/v, EDTA 1mM, DTT 1mM) with the addition of PMSF 0.5mM at a ratio of 4ml/gr of cells. Lysis was conducted by sonication on ice at an amplitude of 65% with 10/20 second on/off cycles for 6 minutes. The soluble fraction was taken after centrifugation at 18,000 RPM for 30 min using an SS34 rotor followed by centrifugation at 40,000 RPM with a 45Ti rotor for 60 min. The supernatant was loaded onto a DEAE cellulose column that was equilibrated with buffer A. FAS I came out in the flow- through of the column with the removal of most of the nucleic acids. The flow through of DEAE was subjected to saturated ammonium sulphate (1.4M pH adjusted) for 30 min with stirring followed by centrifugation at 12,000 RPM for 30 min using an SS34 rotor. The precipitated protein was dissolved in 20 ml of buffer B (KPB 50 mM, KCl 100 mM, glycerol 5% v/v, EDTA 1mM and DTT 1mM) and loaded onto a two-layer sucrose cushion, where the bottom layer contained 5 ml (50% sucrose/50mM KPB) and the upper layer contained 20 ml (30% sucrose/50mM KPB). The cushion was centrifuged for 17 hrs at 40,000 RPM using a 45Ti rotor. Five ml fractions were collected using a blunted syringe. The fractions containing FAS I (at about the interface between 30%-50% sucrose layers) were diluted x3 with buffer A and loaded onto a Q15 ion exchange column equilibrated with buffer B. The protein was eluted with buffer B containing 1M KCl using a gradient up to 35% over 35min, at a rate of 1 ml/min. Fractions containing FAS I were loaded onto a superose 6 10/300 column (GE Healthcare) equilibrated with buffer A (S2A Fig). The peaks were collected, concentrated and analyzed on 6%//15% SDS –PAGE as shown (S2B Fig).
W4389263069.txt	https://www.frontiersin.org/articles/10.3389/fphar.2023.1330291/pdf?isPublishedV2=False	en		Corrigendum: Tilianin extracted from Dracocephalum moldavica L. induces intrinsic apoptosis and drives inflammatory microenvironment response on pharyngeal squamous carcinoma cells via regulating TLR4 signaling pathways	Frontiers in pharmacology	2,023	cc-by	674	Correction 01 December 2023 10.3389/fphar.2023.1330291 TYPE PUBLISHED DOI OPEN ACCESS EDITED AND REVIEWED BY Xiaoqin Wu, University of Texas Health Science Center at Houston, United States CORRESPONDENCE Zhuorong Li, lizhuorong@imb.pumc.edu.cn Rui Liu, liurui@imb.pumc.edu.cn † These authors have contributed equally to this work 30 October 2023 10 November 2023 PUBLISHED 01 December 2023 RECEIVED ACCEPTED Corrigendum: Tilianin extracted from Dracocephalum moldavica L. induces intrinsic apoptosis and drives inﬂammatory microenvironment response on pharyngeal squamous carcinoma cells via regulating TLR4 signaling pathways CITATION Jiang H, Zeng L, Dong X, Guo S, Xing J, Li Z and Liu R (2023), Corrigendum: Tilianin extracted from Dracocephalum moldavica L. induces intrinsic apoptosis and drives inﬂammatory microenvironment response on pharyngeal squamous carcinoma cells via regulating TLR4 signaling pathways. Front. Pharmacol. 14:1330291. doi: 10.3389/fphar.2023.1330291 Hailun Jiang 1†, Li Zeng 1†, Xueqi Dong 2, Shuilong Guo 3, Jianguo Xing 4, Zhuorong Li 1 and Rui Liu 1* 1 Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 2Department of Cardiology, Fuwai Hospital, National Center for Cardiovascular Disease, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 3 Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 4 Xinjiang Institute of Materia Medica, Ürümqi, China COPYRIGHT © 2023 Jiang, Zeng, Dong, Guo, Xing, Li 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. KEYWORDS dendritic cells, human pharyngeal squamous cell carcinoma, intrinsic apoptosis, nuclear factor-κappa B, tilianin, toll-like receptor, tumor immunity A Corrigendum on Tilianin extracted from Dracocephalum moldavica L. induces intrinsic apoptosis and drives inﬂammatory microenvironment response on pharyngeal squamous carcinoma cells via regulating TLR4 signaling pathways by Jiang H, Zeng L, Dong X, Guo S, Xing J, Li Z and Liu R (2020). Front. Pharmacol. 11:205. doi: 10. 3389/fphar.2020.00205 In the published article, there was an error in Figure 5 as published. In the originally published version of this article, in Figure 5C, the image for the 10 μM tilianin transfected with NC siRNA group in the cell colony formation assay was incorrect. It inadvertently used the same image as the 100 μM tilianin transfected with p65 siRNA group. The image for the 10 μM tilianin transfected with NC siRNA group in Figure 5C has been corrected with the actual image. The corrected Figure 5 and its caption appear below. The authors apologize for this error and state that this change has no impact on the results and conclusions of the article. The original article has been updated. Frontiers in Pharmacology 01 frontiersin.org Jiang et al. 10.3389/fphar.2023.1330291 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 Frontiers in Pharmacology 02 frontiersin.org Jiang et al. 10.3389/fphar.2023.1330291 FIGURE 5 TLR4 and p65 contribute to the cytotoxic effects of tilianin on FaDu cells. (A) Tilianin treatment does not decrease cell viability of FaDu cells after silencing of TLR4 and p65 by siRNA. (B) Colony numbers calculated by image J. software. (C) Tilianin treatment does not inhibit cell colony formation in the presence of TLR4 siRNA and p65 siRNA. (D) The percentage of apoptosis analyzed by BD FACSCanto II. (E) Tilianin treatment does not induce cell apoptosis after treatment of FaDu cells with TLR4 siRNA and p65 siRNA. Results are expressed as the mean ± SD, n = 6. p < 0.05, **p < 0.001 vs. control. #p < 0.05, ##p < 0.01, ###p < 0.001 vs. tilianin. Frontiers in Pharmacology 03 frontiersin.org
https://openalex.org/W4389299938	https://www.scielo.br/j/ress/a/pmqkdGt3wTZQY7RCWWkkyPr/?lang=pt&format=pdf	Portuguese	null	20 anos da Secretaria de Vigilância em Saúde e Ambiente: análise de duas décadas e perspectivas	Epidemiologia e Serviços de Saúde	2,023	cc-by	1,594	20 anos da Secretaria de Vigilância em Saúde e Ambiente: análise de duas décadas e perspectivas Quando Silva Junior publicou, há duas décadas, o artigo A nova face da vigilância epidemiológica,1 apresentando a nova estrutura do Ministério da Saúde, com a criação da Secretaria de Vigilância em Saúde (SVS), eu era uma estudante de doutorado em epidemiologia, e me lembro bem do contentamento coletivo daquele anúncio e dos novos tempos que sopravam no início do governo que se iniciava em 2003. A ideia de congregar, sob uma mesma secretaria, programas e controle de doenças transmissíveis, que antes estavam dispersos, e as doenças crônicas e os agravos não transmissíveis, garantindo seu monitoramento de forma mais efetiva, animaram todo o campo da saúde coletiva. Em duas décadas, ocorreram muitas mudanças significativas nas ações da SVS. Entre elas, destaca-se o papel de coordenação da implantação da Política Nacional de Vigilância em Saúde (PNVS),2 instituída em 2018, e da Política Nacional de Saúde do Trabalhador e da Trabalhadora,3 criada em 2012, que envolve a Rede Nacional de Atenção Integral à Saúde do Trabalhador (Renast), compreendendo 215 centros de referência em saúde do trabalhador (Cerests), em diversos estados e municípios do território brasileiro. É também um marco o reconhecimento, em 2023, dos agentes de combate às endemias (ACEs) como profissionais de saúde que desempenham atividades de vigilância, prevenção e controle de doenças e promoção da saúde.4 Por meio de seu Departamento de Articulação Estratégica em Vigilância em Saúde (DAEVS), a Secretaria de Vigilância em Saúde e Ambiente (SVSA) repassa mensalmente aos estados, ao Distrito Federal e ao municípios brasileiros recursos para apoiar as ações de vigilância em saúde e ambiente, por meio do Incentivo Financeiro (IF) para os ACEs e do Piso Fixo de Vigilância em Saúde (PFVS), entre outros instrumentos. Através do Departamento de Doenças Transmissíveis (DEDT), do Departamento de Análise Epidemiológica e Vigilância de Doenças Não Transmissíveis (DAENT) e do Departamento de HIV/AIDS, Tuberculose, Hepatites Virais e Infecções Sexualmente Transmissíveis (DATHI), a SVSA promove o aprimoramento e capilarização dos programas e ações de vigilância, controle e eliminação de doenças transmissíveis e de vigilância e monitoramento das doenças não transmissíveis, e dos acidentes e violências, em todo o território nacional. NOTA EDITORIAL doi 10.1590/S2237-962220230004000016.pt NOTA EDITORIAL doi 10.1590/S2237-962220230004000016.pt 20 anos da Secretaria de Vigilância em Saúde e Ambiente: análise de duas décadas e perspectivas Destacam-se também o papel do Departamento de Emergências em Saúde Pública (DEMSP), na coordenação da Rede Nacional de Alerta e Resposta às Emergências em Saúde Pública, com 190 unidades do Centro de Informações Estratégicas em Vigilância em Saúde (CIEVS); da Rede Nacional de Vigilância Epidemiológica Hospitalar (Renaveh); do Programa Nacional de Vigilância em Saúde dos Riscos Associados aos Desastres (Vigidesastres); e do Programa de Treinamento em Epidemiologia de Campo Aplicado aos Serviços do Sistema Único de Saúde (Episus). Agrega- -se, a essas estruturas da Secretaria, a coordenação do Sistema Nacional de Laboratórios de Saúde Pública no Brasil e do Programa Nacional de Imunizações, responsável por uma das políticas públicas mais bem-sucedidas do Brasil, que este ano completou 50 anos. Além disso, a criação da SVS incorporou a visão da necessidade de investimento em um periódico científico que tivesse a missão de difundir o conhecimento epidemiológico aplicável às ações de vigilância, de prevenção e de controle de doenças e agravos de interesse da saúde pública no Epidemiologia e Serviços de Saúde, Brasília, 32(4):e2023373, 2023 NOTA EDITORIAL NOTA EDITORIAL âmbito do SUS. O então Informe Epidemiológico do SUS, criado em 1992, foi reestruturado para atender a essas necessidades, e nasceu a revista Epidemiologia e Serviços de Saúde: revista do SUS (RESS), que, para além de sua missão primordial, contribuiu também para a consolidação de muitos programas de pós-graduação na área de conhecimento da saúde coletiva, principalmente os de perfil profissional. Hoje, a nossa revista está classificada no estrato A3 da Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes) e sua excelência é reconhecida nacional e internacionalmente. Nessas duas décadas, a revista se modernizou e se integrou aos principais indexadores nacionais (SciELO) e internacionais (Scopus, Medline, Web of Science e Pubmed Central), passando a publicar regularmente artigos nas versões em português e inglês, e tendo no seu corpo editorial epidemiologistas e sanitaristas renomados no campo da saúde coletiva. Neste ano, quando a antiga SVS completa 20 anos, novos ventos também sopram. Pela primeira vez, temos uma mulher no comando do Ministério da Saúde, a pesquisadora Nísia Trindade Lima.5 Hoje, a SVS ganha um novo escopo com a inclusão da palavra “ambiente”,6 reconhecendo assim a indissociabilidade das relações entre a saúde humana, a saúde animal e o ambiente, que é determinante e determinado pela saúde. CONFLITOS DE INTERESSE Ethel Leonor Noia Maciel1 REFERÊNCIAS 1. Silva Junior JB. A nova face da vigilância epidemiológica. Epidemiol Serv Saude. 2003;12(1):5-6. doi: 10.5123/S1679-49742003000100001. 2. Brasil. Ministério da Saúde. Conselho Nacional de Saúde. Resolução nº 588, de 12 de julho de 2018. Fica instituída a Política Nacional de Vigilância em Saúde (PNVS), aprovada por meio desta resolução. Diário Oficial da União, Brasília (DF), 2018 ago 13; Seção 1:87. 3. Brasil. Ministério da Saúde. Portaria nº 1.823, de 23 de agosto de 2012. Institui a Política Nacional de Saúde do Trabalhador e da Trabalhadora. Diário Oficial da União, Brasília (DF), 2012 ago 24; Seção 1:46. 4. Brasil. Presidência da República. Lei nº 14.536, de 20 de janeiro de 2023. Altera a Lei nº 11.350, de 5 de outubro de 2006, a fim de considerar os Agentes Comunitários de Saúde e os Agentes de Combate às Endemias como profissionais de saúde, com profissões regulamentadas, para a finalidade que especifica. Diário Oficial da União, Brasília (DF), 2023 jan 20; Seção 1:1. 4. Brasil. Presidência da República. Lei nº 14.536, de 20 de janeiro de 2023. Altera a Lei nº 11.350, de 5 de outubro de 2006, a fim de considerar os Agentes Comunitários de Saúde e os Agentes de Combate às Endemias como profissionais de saúde, com profissões regulamentadas, para a finalidade que especifica. Diário Oficial da União, Brasília (DF), 2023 jan 20; Seção 1:1. 5. Kirby T. Nísia Trindade: Brazil’s Federal Minister of Health. Lancet. 2023;402(10410):1317. doi: 10.1016/ S0140-6736(23)02231-6. 5. Kirby T. Nísia Trindade: Brazil’s Federal Minister of Health. Lancet. 2023;402(10410):1317. doi: 10.1016/ S0140-6736(23)02231-6. 6. Brasil. Presidência da República. Decreto nº 11.358, de 1º janeiro de 2023. Aprova a Estrutura Regimental e o Quadro Demonstrativo dos Cargos em Comissão e das Funções de Confiança do Ministério da Saúde e remaneja cargos em comissão e funções de confiança. Diário Oficial da União, Brasília (DF), 2023 jan 1; Seção 1:262. 6. Brasil. Presidência da República. Decreto nº 11.358, de 1º janeiro de 2023. Aprova a Estrutura Regimental e o Quadro Demonstrativo dos Cargos em Comissão e das Funções de Confiança do Ministério da Saúde e remaneja cargos em comissão e funções de confiança. Diário Oficial da União, Brasília (DF), 2023 jan 1; Seção 1:262. 7. Mora C, McKenzie T, Gaw IM, Dean JM, von Hammerstein H, Knudson TA, et al. Over half of known human pathogenic diseases can be aggravated by climate change. Nat Clim Chang. 2022;12(9):869-75. doi: 10.1038/s41558-022-01426-1 7. 20 anos da Secretaria de Vigilância em Saúde e Ambiente: análise de duas décadas e perspectivas A nova SVSA reconhece, portanto, o papel central das mudanças climáticas7 no surgimento de doenças e a importância da saúde ambiental nesse contexto. Aliada à saúde do trabalhador, coadunam o eixo norteador de populações saudáveis, tendo como estrutura organizacional catalisadora das ações o Departamento de Vigilância em Saúde Ambiental e Saúde do Trabalhador (DSAST). A SVSA reconhece também a singularidade com que acidentes e violências impactam na saúde da população e cria, em 2023, a coordenação de acidentes e violências no DAENT, deixando de nomeá-los com o genérico termo de “agravos à saúde”. A criação do Centro Nacional de Inteligência Epidemiológica (CNIE), programada para 2024, é um pilar para se garantir maior capacidade de utilização de dados na geração de informação oportuna e qualificada. Em conjunto com a transformação do PNI em um departamento (DPNI), com arranjos administrativos que visam qualificar a gestão do programa, essas iniciativas são exemplos deste novo tempo. A elevação da vigilância em saúde de base laboratorial a prioridade nacional no novo Plano de Aceleração do Crescimento (PAC) demonstra, de forma inequívoca, a importância do diagnóstico laboratorial na preparação para futuras emergências em saúde. A agregação do Instituto Evandro Chagas e do Centro Nacional de Primatas à SVSA incrementa a vigilância em saúde de base laboratorial na preparação e resposta às emergências em saúde pública e posiciona a região Norte como protagonista das ações de saúde no e para o Brasil. Após um período de pandemia, em que a negação da ciência foi a conduta normativa da gestão, voltar a posicionar a ciência no centro da tomada de decisão é não apenas simbólico, mas também essencial. A SVS, agora SVSA, cresceu, floresceu, sofreu ameaças de sufocamento, mas resistiu e hoje, sob a liderança de uma ministra que conhece e defende o SUS, se coloca pronta para os desafios das próximas décadas. 2 Epidemiologia e Serviços de Saúde, Brasília, 32(4):e2023373, 2023 2 NOTA EDITORIAL CONFLITOS DE INTERESSE REFERÊNCIAS Mora C, McKenzie T, Gaw IM, Dean JM, von Hammerstein H, Knudson TA, et al. Over half of known human pathogenic diseases can be aggravated by climate change. Nat Clim Chang. 2022;12(9):869-75. doi: 10.1038/s41558-022-01426-1 7. Mora C, McKenzie T, Gaw IM, Dean JM, von Hammerstein H, Knudson TA, et al. Over half of known human pathogenic diseases can be aggravated by climate change. Nat Clim Chang. 2022;12(9):869-75. doi: 10.1038/s41558-022-01426-1 Epidemiologia e Serviços de Saúde, Brasília, 32(4):e2023373, 2023 3 3
https://openalex.org/W3210242604	https://www.researchsquare.com/article/rs-959449/latest.pdf	English	null	Robust Bias-Correction of Precipitation Extremes Using A Novel Hybrid Empirical Quantile Mapping Method: Advantages of A Linear Correction For Extremes	Research Square (Research Square)	2,021	cc-by	13,726	Robust Bias-Correction of Precipitation Extremes Using A Novel Hybrid Empirical Quantile Mapping Method: Advantages of A Linear Correction For Extremes Maike Holthuijzen (  maike.holthuijzen@uvm.edu ) University of Vermont https://orcid.org/0000-0002-6870-3314 Maike Holthuijzen (  maike.holthuijzen@uvm.edu ) University of Vermont https://orcid.org/0000-0002-6870-3314 Brian Beckage University of Vermont Patrick J Clemins University of Vermont Dave Higdon Virginia Tech Jonathan M Winter Dartmouth University Brian Beckage University of Vermont Patrick J Clemins University of Vermont Dave Higdon Virginia Tech Jonathan M Winter Dartmouth University Research Article Keywords: bias-correction, extremes, precipitation Posted Date: October 20th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-959449/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Noname manuscript No. (will be inserted by the editor) Keywords bias-correction · extremes · precipitation Keywords bias-correction · extremes · precipitation Advantages of a linear correction for extremes Maike Holthuijzen · Brian Beckage · Patrick J. Clemins · Dave Higdon · Jonathan M. Winter · Maike Holthuijzen · Brian Beckage · Patrick J. Clemins · Dave Higdon · Jonathan M. Winter · Received: date / Accepted: date Received: date / Accepted: date Abstract High-resolution, daily precipitation climate products that realisti- cally represent extremes are critical for evaluating local-scale climate impacts. A popular bias-correction method, empirical quantile mapping (EQM), can generally correct distributional discrepancies between simulated climate vari- ables and observed data but can be highly sensitive to the choice of calibration period and is prone to overﬁtting. In this study, we propose a hybrid bias- correction method for precipitation, EQM-LIN, which combines the eﬃcacy of EQM for correcting lower quantiles, with a robust linear correction for upper quantiles. We apply both EQM and EQM-LIN to historical daily precipitation data simulated by a regional climate model over a region in the northeastern United States. We validate our results using a ﬁve-fold cross-validation and quantify performance of EQM and EQM-LIN using skill score metrics and several climatological indices. As part of a high-resolution downscaling and Maike Holthuijzen (corresponding author) University of Vermont Tel.: 208-830-0270 E-mail: maike.holthuijzen@uvm.edu Brian Beckage University of Vermont E-mail: brian.beckage@uvm.edu Patrick J. Clemins University of Vermont Tel.: 802-448-2188 E-mail: patrick.clemins@uvm.edu Dave Higdon Virginia Tech Tel.:571-858-3120 E-mail: dhigdon@vt.edu Jonathan M. Winter Dartmouth University Tel.: 603-646-6456 E-mail: jonathan.m.winter@dartmouth.edu 2 Maike Holthuijzen et al. bias-correction workﬂow, EQM-LIN signiﬁcantly outperforms EQM in reduc- ing mean, and especially extreme, daily distributional biases present in raw model output. EQM-LIN performed as good or better than EQM in terms of bias-correcting standard climatological indices (e.g., total annual rainfall, frequency of wet days, total annual extreme rainfall). In addition, our study shows that EQM-LIN is particularly resistant to overﬁtting at extreme tails and is much less sensitive to calibration data, both of which can reduce the uncertainty of bias-correction at extremes. bias-correction workﬂow, EQM-LIN signiﬁcantly outperforms EQM in reduc- ing mean, and especially extreme, daily distributional biases present in raw model output. EQM-LIN performed as good or better than EQM in terms of bias-correcting standard climatological indices (e.g., total annual rainfall, frequency of wet days, total annual extreme rainfall). In addition, our study shows that EQM-LIN is particularly resistant to overﬁtting at extreme tails and is much less sensitive to calibration data, both of which can reduce the uncertainty of bias-correction at extremes. Advantages of a linear correction for extremes Keywords bias-correction · extremes · precipitation 1 Introduction Climate data products with ﬁne spatial resolutions,which are important for studying localized changes in extreme climate events, can be generated by combining statistical and dynamic downscaling, [22]. In this study, we also combine statistical and dynamical downscaling to produce pre- cipitation data products with a ﬁne spatial resolution. p p p Downscaling is complemented by bias-correction, a procedure in which cli- mate model output is adjusted such that its statistical properties (e.g. mean, variance, and potentially higher moments) resemble those of observations in a common climatological period [45], [9]. We note that the terms “downscal- ing” and “bias-correction” are sometimes used to refer to equivalent processes. However, in this study, downscaling only refers to the process in which coarse, gridded climate data is interpolated to a ﬁner spatial resolution, and bias- correction refers speciﬁcally to applying transformations to climate model out- put such that distributional biases are reduced. Most bias-correction methods assume stationarity of model errors over time [70], which can be problematic for bias-correcting future climate model output over multi-decadal time spans [10] [20]. In addition, suﬃcient observational data is necessary to derive robust transfer functions [20]. Bias-correction methods for precipitation range from simple approaches such as the “delta change” or “delta factor” method [78] to more ﬂexible and eﬀective quantile-mapping based methods [78] [10] [86]. In quantile-mapping (QM) based methods, a transfer function (TF) maps quan- tiles of climate model output to those of observed data. QM methods can be parametric [63], non-parametric [45], or a combination of both [77]. Distribu- tion mapping (DM) is a parametric QM method in which known, parametric distributions are ﬁt to observed and model data. The Gamma distribution is often used to model wet-day precipitation (e.g. [45], [28] [50]) but is generally not adequate for modeling extreme precipitation tails [32], [29]. Hybrid DM approaches in which the Gamma distribution is ﬁt to lower quantiles and a heavy-tailed distribution is ﬁt to tail quantiles can improve bias-correction of extreme precipitation [29] [81]. A non-parametric counterpart to DM, empiri- cal quantile mapping (EQM), is a ﬂexible, non-parametric method in which no distributional assumptions are made. In EQM, the TF represents a mapping from empirical model quantiles to observed quantiles and typically outper- forms DM [41] [40]. 1 Introduction Climate data is often necessary for social, ecological, and hydrological mod- els and is routinely used in climate impact models and assessments. Model reliability is largely dependent on the quality and resolution of climate data products [18], [35], [21] [17]. The representation of extremes, in particular, can have a disproportionately large eﬀect on such models [46]. Increases in the frequency, variability, and magnitude of extreme precipitation over the last several decades, especially in the northeastern United States, are well- documented [31] [39]. To study the future impacts of changing extremes at local scales, climate data products must represent extreme events accurately and be available at ﬁne spatial and temporal resolutions [46]. General circula- tion models (GCMs) provide important information about historical and fu- ture larger-scale climate trends, but their resolution is too coarse to investigate localized eﬀects of changes in extreme climate events [13], [45]. Additionally, raw GCM output is characterized by a non-trivial degree of bias [45], and the ability of GCMs to reproduce extreme tails of climate variables is limited [47]. Therefore, prior to its use in hydrological [64] [73], agricultural [34], or ecologi- cal models, GCM output is downscaled to a ﬁner resolution and bias-corrected with respect to observed data [89]. These post-processing techniques result in climate data that is more realistic at ﬁner spatial scales. Here, we propose a bias-correction method that more accurately captures precipitation extremes. We incorporate it into a high-resolution downscaling and bias-correction work- ﬂow for constructing daily, high- resolution data products for use in modeling eﬀorts. In the process of downscaling, model output is converted from a coarse to ﬁner resolution. In dynamical downscaling, a regional climate model (RCM) is forced with a GCM, resulting in ﬁner-scale output in which regional cli- mate processes, topography, and orography are incorporated [16]. In statisti- cal downscaling, statistical relationships between coarse-scale climate variables and local, observed data are established, and the eﬀects of ﬁne-scale predictors are integrated into downscaled data [54]. Dynamical downscaling is computa- tionally intensive and can introduce additional biases [8] [48], but, localized cli- mate processes, including extremes [23], are generally better reproduced than in GCMs [52]. However, RCMs are do not perform well in capturing the most Title Suppressed Due to Excessive Length 3 extreme events [2] [47]. Statistical downscaling is eﬃcient, can be applied to a variety of climate variables [56], and is especially eﬀective in topographically complex terrain [30]. 1 Introduction There is a need for a hybrid EQM method in which bias-correction of extremes can be performed without the risk of overﬁtting and the introduction of outliers. We propose and demonstrate a simple, hybrid EQM method for bias- correction that, when used in conjunction with downscaling, results in high- resolution (1km) daily precipitation data in which precipitation extremes are accurately represented. The proposed method, EQM-LIN, combines the eﬀec- tiveness of EQM for correcting the bulk of the distribution with a robust, linear correction for extremes. As part of a high-resolution, downscaling and bias- correction workﬂow, we use EQM-LIN to bias-correct historical (1976-2005), daily precipitation data that were dynamically downscaled by a regional cli- mate model (RCM). We also compare the eﬀectiveness of EQM-LIN to EQM for bias-correction, with an emphasis on the ability of the two methods to ac- curately capture extremes. Because EQM-LIN is computationally cheap, easy to apply, and corrects both mean and extreme bias for precipitation variables, it is an important methodological addition to the body of bias-correction lit- erature. 1 Introduction EQM is eﬀective in correcting precipitation variables [41], [15] [41], [57] [14] and is attractive as a bias-correction method as it corrects the mean, standard deviation, and higher-order distributional moments [28]. We note that the terms downscaling and bias-correction are sometimes used to refer to equivalent processes. However, in this study, downscaling only refers to the process in which raw, gridded climate model output is interpolated to a ﬁner spatial resolution, and bias-correction refers to methods applied to climate model output that result in a reduction of distributional biases. A disadvantage of QM methods and EQM in particular, is their propensity to overﬁt on calibration data, especially at precipitation extremes where data is scarce and highly variable [37] [45], [26], [37] [63] [51]. In EQM, TFs are Maike Holthuijzen et al. 4 interpolated using linear interpolation, splines, or other smoothing techniques [27]. Flexible methods such as EQM can result in TFs that can correct model data nearly perfectly (overﬁtting) but may not generalize to out-of-sample or future model data. Overﬁtting is problematic because it can lead to instability of the TF at higher quantiles [24] [26] [33]. When applied to future projections, EQM has been shown to signiﬁcantly distort future climate change signals [25] [53] and exaggerate or deﬂate extreme trends, introducing additional uncer- tainty into bias-corrected data [10] [77]. Hybrid EQM approaches that combine parametric and non-parametric modeling can reduce the degree of overﬁtting of the TF at extreme tails [77]. In a hybrid approach, bias-correction below a speciﬁed threshold is achieved via a non-parametric TF (EQM), while bias- correction above the threshold is with DM, based on an extreme distribution, such as the Generalized Pareto distribution [77]. Hybrid EQM methods com- bine the ﬂexbility of EQM for correcting lower to middle quantiles with the robustness of parametric distributions for correcting upper quantiles. In par- ticular, the use of extreme or heavy-tailed distributions for modeling extremes can improve bias-correction of tail quantiles [44] [51] [88] [29] [43] [88] [77]. However, the risk of overﬁtting the TF at distributional tails still exists, as poor ﬁts to heavy-tailed distributions can introduce outliers [50] [72]. In addi- tion, selection of the threshold is diﬃcult, as the amount of data beyond the threshold must be suﬃciently large to allow for distribution ﬁtting and must approximate a known heavy-tailed distribution [5] [29]. 2 Methods 2.1 Data The study area, the Lake Champlain Basin, consists of parts of Vermont, New Hampshire, eastern New York, United States and southern Quebec, Canada Title Suppressed Due to Excessive Length 5 (Figure 1). Four watersheds drain into Lake Champlain, and the Green Moun- tains, Adirondack Mountains, and White Mountains span portions of Vermont, New York, and New Hampshire, respectively [85]. Elevations in the study area range from 30 to 1500 m above mean sea level (MSL). The study area is char- acterized by a subhumid continental climate with cold and snowy winters. At high elevations, total annual precipitation can reach 152cm, but locally in- tense precipitation in the form of thunderstorms is more likely during summer months [76]. Fig. 1 GHCND stations (black) within the study area (red). Fig. 1 GHCND stations (black) within the study area (red). Simulated historical (1976-2005) precipitation (PRCP) data were gener- ated by the Advanced Weather and Research Forecasting model (WRF) ver- sion 3.9.1, an RCM [74]. WRF output was generated at a daily temporal resolution. WRF is a widely used numerical weather prediction system for both research and applied forecasting purposes [74]. Historical simulations (1976-2005) were forced by bias-corrected Community Earth System Model 1 (CESM1), a GCM [58]. CESM1 historical simulations were dynamically down- scaled with WRF to a 4km resolution using three one-way nests (36 km, 12 km, 4km). The 4km resolution WRF data were used for this study. Additional WRF model details are included in the Supplementary Materials, and a full description and evaluation of simulations can be found in [38]. Historical daily climate station data was obtained from the Global His- torical Climate Network (GHCND) (https://www.ncdc.noaa.gov/cdo-web/ search?datasetid=GHCND). GHCND data records are adjusted to account for changes in instrumentation and other anomalies [59,62]. We retained only those stations with at least 70% complete records over the historical time period 1976-2005 (85 stations). We chose to use station data, rather than gridded data products (e.g. Livneh, [49]; Daymet, [79]; and PRISM, [11]), be- 6 Maike Holthuijzen et al. cause interpolation algorithms used to create gridded climate products can introduce bias [4] and additional uncertainty when used for bias-correcting climate model output [83]. Gridded products can misrepresent extreme tails [3], and [87] showed that Daymet, Livneh, and PRISM precipitation products varied widely in their representation of wet-day occurrences, length of wet and dry periods, and precipitation intensity in the South-Central United States. 2 Methods Station data represent direct climatological measurements and are available throughout the Northeast [62] [12]. In the proposed workﬂow, historical WRF simulations (model output) were downscaled to a 1km grid prior to bias-correction using topographic downscal- ing, a variation of inverse distance weighting (IDW) that incorporates eleva- tional lapse rates [85]. Elevation estimates at each 1km grid cell were derived by interpolating elevation values from a 30m digital elevation model (DEM) [82] via IDW. The 1-km grid cell size was chosen based on resolution require- ments for climate impacts modeling eﬀorts over the Lake Champlain Basin [84], [85]. Prior to bias-correction, historical simulations were also interpolated to GHCND station locations using topographical downscaling for the purpose of constructing TFs. Model data interpolated to GHCND station locations are denoted as MODinterp,station. Generally raw WRF model data exhibited a wet bias that was most pronounced during summer months (Figure 2). 7 7 Title Suppressed Due to Excessive Length 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Fig. 2 Mean daily precipitation (mm/day) for raw model (Mod) and observed data (Obs) averaged over 85 GHCND stations; smoothed daily precipitation (using loess smoothers) (a) and mean monthly precipitation (mm/month) for raw model (Mod) and observed data (Obs) (b) over 1976-2005. Fig. 2 Mean daily precipitation (mm/day) for raw model (Mod) and observed data (Obs) averaged over 85 GHCND stations; smoothed daily precipitation (using loess smoothers) (a) and mean monthly precipitation (mm/month) for raw model (Mod) and observed data (Obs) (b) over 1976-2005. 2.2 Bias-correction methods The proposed approach, empirical quantile mapping with a linear correc- tion for extremes, EQM-LIN, was compared to empirical quantile mapping (EQM), which is one of the most frequently-used and eﬀective methods for bias-correction. In addition, we compared EQM-LIN to DM with the Gamma distribution (DM-GAMMA), a hybrid EQM approach in which lower quantiles were corrected using EQM, and upper quantiles were ﬁt to Generalized Pareto Distributions (GPDs) (EQM-GPD) [77], as well as a trend-preserving method, quantile delta mapping (QDM) [10]. The results are presented in the Supple- mentary Material but not evaluated in the main manuscript, since none of the additional methods performed as well as or signiﬁcantly better than EQM or EQM-LIN. For both bias-correction methods EQM-LIN and EQM, TFs were con- structed by spatially pooling observed GHCND station and MODinterp,station Maike Holthuijzen et al. 8 data. The same TF was applied to all model values, regardless of spatial lo- cation. We chose to spatially pool data because 1) ﬁne-scale elevation, which does vary spatially, was already incorporated into the downscaling step, and 2) additional interpolation necessary to construct separate TFs based on spa- tial location would have added uncertainty to bias-corrected data. Spatially explicit bias-correction in general can be a diﬃcult task; it involves estimating the TF at every location at which bias-corrected data is desired [37], which is contrary to our desire to develop a bias-correction approach that is simple, eﬃcient, and easily implemented. For both bias-correction methods, twelve TFs were constructed, one for each month of the year [41] [63]. Daily raw model values interpolated to the 1km grid were corrected with the corresponding monthly TF. Because GHCND station gauges are accurate to 0.1mm [59], we deﬁned wet-day pre- cipitation days as days in which daily precipitation was greater than or equal to 0.1mm. Prior to construction of TFs and bias-correction, values of daily MODinterp,station and model interpolations to the 1km grid below 0.1mm were set to 0. All analyses were conducted in R Statistical Language [66]. Empirical quantile mapping: EQM The TF used in EQM is expressed by the empirical cumulative distribution function (ecdf) and its inverse (ecdf−1). 2.2 Bias-correction methods Monthly TFs are of the form: xcorr,t = ecdf−1 obs(ecdfmod(xmod,t)), (1) (1) xcorr,t = ecdf−1 obs(ecdfmod(xmod,t)), where, xcorr,t is the corrected model precipitation value on day t, ecdf−1 obs is the inverse ecdf of observed data, ecdfmod is the ecdf of model data, and xmod,t is the raw model precipitation value on day t. Monthly TFs were con- structed using 10,000 estimated quantiles and interpolation of the TF was accomplished with monotone Hermite splines using the qmap package [27] in R. Values exceeding the range of the TF were corrected using the method of constant extrapolation [7]. The approximate shape of the TF can be exam- ined by plotting estimated quantiles of model and observed data against one another to form a “quantile-quantile-” or “qq-” map (Figure 3). The shape of the quantile-quantile map can provide insight into the type and magnitude of model bias. For instance, if the TF falls below (rises above) the 1:1 line, model quantiles are too high (low) relative to observed quantiles. 9 Title Suppressed Due to Excessive Length 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 Fig. 3 A quantile-quantile map for August constructed with 10,000 quantiles of model and observed data during the calibration period. The red solid line denotes the 1:1 line. Here, raw model data exhibits a low bias, especially at upper quantiles, as the qq-map lies above the 1:1 line. Fig. 3 A quantile-quantile map for August constructed with 10,000 quantiles of model and observed data during the calibration period. The red solid line denotes the 1:1 line. Here, raw model data exhibits a low bias, especially at upper quantiles, as the qq-map lies above the 1:1 line. Empirical quantile mapping with a linear correction for extremes: EQM-LIN In EQM-LIN, the majority of model data are bias-corrected via EQM using (1), while model data beyond a speciﬁed threshold are adjusted with a constant correction via a linear TF (2). All bias-correction by EQM was done with the qmap package [27] in R, and custom code was used to construct the linear TF. The following steps describe the EQM-LIN procedure: 1. 2.2 Bias-correction methods Title Suppressed Due to Excessive Length 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Fig. 4 The quantile-quantile map and corresponding EQM and EQM-LIN TFs for daily observed and model data during the month of August over the calibration period 1976-2005. a) quantile-quantile map, constructed using 10,000 quantiles evenly spaced between 0 and 1; b) EQM TF ; c) EQM-LIN TF, with the blue line denoting the non-parametric (EQM) portion of the TF and the red line indicating the linear portion; d) enlarged section of EQM- LIN TF in c) (grey box) to illustrate the transition from EQM portion to the linear portion of the TF. In c) and d), the threshold (dashed line), indicates the 79th quantile of model data (6.88mm). Fig. 4 The quantile-quantile map and corresponding EQM and EQM-LIN TFs for daily observed and model data during the month of August over the calibration period 1976-2005. a) quantile-quantile map, constructed using 10,000 quantiles evenly spaced between 0 and 1; b) EQM TF ; c) EQM-LIN TF, with the blue line denoting the non-parametric (EQM) portion of the TF and the red line indicating the linear portion; d) enlarged section of EQM- LIN TF in c) (grey box) to illustrate the transition from EQM portion to the linear portion of the TF. In c) and d), the threshold (dashed line), indicates the 79th quantile of model data (6.88mm). 2.2 Bias-correction methods Calibration data is divided into two datasets in which model data is less than (CAL-LOW ) and greater than a speciﬁed threshold (CAL-HIGH). The threshold, T is a function of the inverse ecdf of model data and is expressed as T = ecdf −1 mod(τLIN), where 0 < τLIN < 1. Thus, T is a precip- itation value in mm that indicates where both model and observed datasets are divided. The procedures for estimating T and τLIN are thoroughly out- lined in the Appendix. 2. Next, the intercept for the linear TF, δ is obtained (details are discussed in the Appendix). The intercept represents the constant correction that will be applied to extreme model values (all model values in CAL-HIGH). The linear TF is expressed as xcorr,t = δ+xmod,t and is applied to model values in CAL-HIGH (eq. 2). Model values in CAL-LOW are corrected via EQM. The TF for EQM-LIN is expressed as: 10 Maike Holthuijzen et al. xcorr,t = ! ecdf−1 obs(ecdfmod(xmod,t)), xmod,t < T xmod,t + δ, xmod,t ≥T, (2) where xcorr,t and xmod,t are as deﬁned in (eq. 1). Thus, the linear portion of the TF (xcorr,t = δ + xmod,t) always has a slope of 1 and intercept δ. In this study, we only consider linear TFs with a slope of 1 and intercept of δ. Optimizing the slope as well as the threshold would increase the overall complexity of EQM-LIN and could introduce the potential for overﬁtting on out-of-sample data. We chose τLIN to be 0.79, based on a grid search over a range of values in a ﬁve fold cross-validation approach (details are discussed in the Appendix). We chose the value of τLIN that resulted in the minimization of the mean absolute error of observed and model ecdfs above the 95th percentile (MAE95), [68] (section 3). MAE95 quantiﬁes the distributional similarity between observed and model data at extremes. Since the focus of this study was on accurately representing distributional extremes, we chose the minimization of MAE95 rather than another metric. However, we found that minimization of MAE95 resulted in improvements in all performance metrics and indices. The shape of the EQM-LIN TF is identical to that of EQM below T, while above the threshold the TF is linear. Figure 4 shows a quantile-quantile map for model and observed data for the month of August and the associated EQM and EQM-LIN TFs. 3 Validation Performance evaluation of EQM and EQM-LIN was accomplished with a ﬁve- fold cross-validation procedure using observed and model data during the cali- bration period (1976-2005). Cross-validation is commonly used to evaluate the eﬃcacy of bias-correction methods, as out-of-sample data can be considered proxies for future projections [77] [28] [41]. Test datasets always consisted of consecutive years (for example, if training data consisted of years 1976-2000, test data would contain years 2001-2005). We chose performance metrics and indices that quantiﬁed 1) model skill and 2) the eﬀectiveness of bias-correction methods in capturing overall clima- tology with an emphasis on extreme tails. Model skill, distributional similarity between model and observed data, was quantiﬁed with the mean absolute error 12 Maike Holthuijzen et al. (MAE). We chose MAE, rather than other skill metrics, such as the Perkins Skill Score [60], because it is more sensitive to outliers. MAE was calculated be- tween distributions of daily observed and raw model data as well as between distributions of daily observed and bias-corrected data [28]. Observed, raw model raw model (MODinterp,stations). MAE95 was used to quantify model skill at extreme tails. MAE95 is computed similarly to MAE, but only the up- per 5% of observed and model distributions and of daily precipitation data are utilized [68]. The number of quantiles estimated in the calculation of MAE95 was equal to the maximum number of 95th quantile values in observed or model distributions. Generally, the number of values greater than the 95th quantile in each data type (raw model, observed, and corrected model) did not diﬀer appreciably. MAE and MAE95 values were calculated by month for each of the ﬁve cross-validated data folds for raw and bias-corrected data, and results are reported as the average metric over the ﬁve folds. Since MAE and MAE95 quantify distributional error between model and observed data, lower values are indicative of better model skill, with an ideal mean absolute error of 0 (no error). We used a subset of ETCCDI indices [61] to assess how well bias-corrected data captured overall climate characteristics of observed data. ETCCDI indices are standard indices that allow for the comparison of results over varying time periods, geographical regions, and source data, and are recommended by the World Research Climate Program (WRCP) [42]. ETTCDI ndices were computed with pooled data. Unlike MAE and MAE95, ETCCDI indices are calculated annually, using spatially-averaged data. 3 Validation Daily precipitation values at individual locations are averaged over each day in the 30-year calibration. Thirty annual values of each ETCCDI index were calculated for observed, raw model, EQM-, and EQM-LIN-corrected model data. The choice of indices was based on the preference of stakeholders. ‘D’ indices (D90, D95, and D99) are deﬁned as the annual number of days in which mean daily precipitation exceeded the 90th, 95th, or 99th quantiles. ‘S’ metrics (S90, S95, and S99) are deﬁned as the annual sum of mean daily precipitation (mm) for days in which mean daily precipitation exceeded the 90th, 95th, or 99th quantiles. TotalP is the annual sum of mean daily precip- itation (mm) on wet days (days for which mean daily precipitation 0.1 mm), WetDays is the annual count of wet days, and the simple precipitation index (SPI) is calculated as TotalP/WetDays (mm/day). SPI is a measure of precip- itation intensity. The nine indices characterize the extreme tails of the 30-year climatology of precipitation, as well as general characteristics. An overview of MAE metrics and ETCCDI indices is given in Table 1. Since we applied the same TF to all raw model data, evaluation metrics were also calculated using pooled data. Performance evaluated by ETCCDI indices or MAE metrics cannot be directly compared, since each provides as- sessments on diﬀerent temporal scales. MAE metrics quantify distributional errors of the entire distribution of daily model data compared to observed data. ETCCDI indices quantify how well model data capture 30-year clima- tology at a temporally coarser (annual) scale using spatially averaged data. In Title Suppressed Due to Excessive Length 13 combination, both evaluation metrics give insight in the overall adequacy of the bias-correction method at both aggregated and ﬁner temporal scales. combination, both evaluation metrics give insight in the overall adequacy of the bias-correction method at both aggregated and ﬁner temporal scales. Table 1 Metric and index deﬁnitions Metric/Index Deﬁnition Reference MAE Mean absolute error of quantiles of observed and raw model or bias-corrected model distributions. MAE is calculated using daily data (not spatially averaged) for the entire historical period using 10,000 estimated quantiles [28] MAE95 Mean absolute error of upper 5% of quantiles of observed and raw model or bias-corrected model distributions. MAE95 is calculated using daily data (not spatially averaged) for the entire historical period. 3 Validation [69] D90, D95, D99 Annual count of days for which mean daily precipitation exceeded the 90th 95th or 99th percentile [1] S90, S95, S99 Annual sum (mm) of mean daily precipitation on days in which mean daily precipitation exceeded the 90th , 95th, or 99th percentile. [1] TotalP Annual sum (mm) of mean daily precipitation on days in which mean daily precipitation ≥0.1mm [1] WetDays Annual count of days in which mean daily precipitation ≥0.1mm [1] SPI Simple precipitation index (mm / day) calculated as TotalP / WetDays [1] 3.1 Analyses Bayesian one-way Analysis of Variance (ANOVA) models were used to deter- mine if mean MAE and MAE95 diﬀered signiﬁcantly among raw model, EQM and EQM-LIN data. Separate ANOVA tests were conducted for MAE and MAE95. ANOVA tests were conducted with data from all ﬁve cross-validated folds, as MAE and MAE95 values within folds can be considered subsamples. All analyses were conducted with the RJags package [65] in R. The response variables, MAE or MAE95 values, were log-transformed prior to analysis to ensure homogeneity of variances, an assumption of ANOVA models. The pre- dictor variable for both ANOVA models was data type, a variable with three levels: raw model (Mod), EQM-LIN, and EQM. Credible intervals in the form of 95% highest posterior density (HPD) intervals were used to determine if the diﬀerence in posterior distributions was signiﬁcantly diﬀerent from 0. Credible intervals were constructed for all pairwise diﬀerences of posterior distributions of EQM-LIN, EQM, and raw model data. Credible intervals can be interpreted as the following: there is a 95% chance that the true pairwise diﬀerence in pos- terior distributions is contained within the interval, given the data. Therefore, if 0 is contained within the interval, the diﬀerence is not signiﬁcant at the 95% conﬁdence level. Full details on these analyses are provided in the Supplemen- tary Materials. Distributions of all nine ETCCDI indices calculated from EQM-, EQM- LIN-corrected, and raw model data were compared to those of observed data. Performance of bias-corrected and raw data relative to observed data was formally assessed using Kolmogorov-Smirnov (KS) tests [75]. The two-sample KS test is a non-parametric test that is used to assess the equality of two empirical distributions. It is sensitive to diﬀerences in both location and shape Maike Holthuijzen et al. 14 of the two ecdfs being compared and is often used in climatological studies [10] [71] [80]. Here, we applied the KS test three times for each ETCCDI index to determine the similarity of ecdfs between observed and EQM- and EQM- LIN-corrected data and between observed and raw model data. All tests were conducted with the two-sided null hypothesis that the samples being compared belonged to a common distribution. The signiﬁcance level, α, was set to 0.05; p-values below 0.05 indicate there is evidence that the two samples do not come from a common distribution. 4 Results Overall, data bias-corrected with either EQM or EQM-LIN exhibited substan- tial improvements in both MAE and MAE95 compared to raw model data (Mod), but improvements were more pronounced for EQM-LIN. Both bias- correction methods generally improved ETCCDI indices compared to Mod, and EQM-LIN performed as well as or slightly better than EQM for all in- dices. 3.1 Analyses We acknowledge that the KS test has low power for small sample sizes (30 values or less) [67]. All KS tests in this study are performed on pairs of distributions composed of 30 annual values; thus, we use KS tests, along with visual inspection of boxplots, to guide our interpretation of results. In addition, to control for multiple comparisons, α was adjusted using the Holm-Bonferroni method [36] (details are shown in the Appendix). 4.1 MAE and MAE95 Mean MAEs of EQM- and EQM-LIN-corrected datasets were 0.704mm and 0.655mm, respectively, while mean MAE of Mod was 1.06mm (Figure 5 a). Mean MAEs of both bias-corrected datasets were signiﬁcantly lower that mean MAE of Mod. Monthly mean MAE values for EQM-LIN were overall slightly lower than those of EQM. The credible interval for the diﬀerence in mean MAE between EQM and EQM-LIN contained 0, indicating that although mean MAE of EQM-LIN was lower than that of EQM, the diﬀerence was not signiﬁcant at the 95% conﬁdence level. Mean MAE95 values of EQM- and EQM-LIN-corrected and Mod were 3.45mm, 2.36mm, and 7.11mm, respectively. For EQM-LIN corrected data, MAE95 varied little among months; however, both raw model and EQM- corrected data exhibited substantial increases in MAE95 between months 8 and 11 (Figure 5 b). Similar to results for MAE, mean MAE95 values of both bias-corrected datasets were signiﬁcantly lower than mean MAE95 of Mod. In contrast to results for MAE, 95% credible intervals for the diﬀerence in mean MAE95 of EQM and EQM-LIN indicated that mean MAE95 of EQM-LIN was signiﬁcantly lower than mean MAE95 of EQM at the 95% conﬁdence level (see Supplementary Materials for full details of ANOVA analysis). Title Suppressed Due to Excessive Length 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Fig. 5 Monthly mean MAE (mm) (a) and MAE95 (mm) (b) for raw model (Mod), EQM- and EQM-LIN-corrected data. Please note the diﬀerence in y-axes limits for plots a and b. Fig. 5 Monthly mean MAE (mm) (a) and MAE95 (mm) (b) for raw model (Mod), EQM- and EQM-LIN-corrected data. Please note the diﬀerence in y-axes limits for plots a and b. 4.2 ETCCDI indices Distributions of ETCCDI indices for both bias-corrected datasets more closely resembled those of observed data compared to Mod, with EQM-LIN perform- ing as good as or slightly better than EQM. Generally, mean and extreme total annual precipitation were overestimated in Mod, and distributions of nearly all ETCCDI indices calculated using Mod were signiﬁcantly diﬀerent than those calculated from observations. While bias-correction resulted in the distributions of most ETCCDI indices becoming more similar to those of ob- served data, it also resulted in an underestimation of wet-day frequency. (See Appendix, Table 3 for selected summary statistics of ETTCDI index distribu- tions for Mod, EQM, and EQM-LIN). D and S indices Less extreme ‘S’ indices (S90 and S95) were substantially overestimated in Mod, and distributions of S90 and S95 calculated from Mod were signiﬁcantly diﬀerent from observed data (Figure 6 a; Table 2). Distribu- tions of S90 and S95 were not signiﬁcantly diﬀerent between EQM-LIN and observed data (Figure 6 a; Table 2). However, although the distribution of S90 did not diﬀer between EQM and observed data, the diﬀerence was signif- icant for S95 (Figure 6 a; Table 2). While both bias-correction methods were able to reduce the substantial overestimation of total extreme annual rainfall exhibited in Mod, EQM-LIN slightly outperformed EQM. Distributions of less extreme ‘D’ indices (D90 and D95) were nearly iden- tical for Mod, bias-corrected data, and observed data (Figure 6 b). P-values Maike Holthuijzen et al. 16 of KS tests for both D90 and D95 conﬁrmed that distributions of Mod and bias-corrected data were not signiﬁcantly diﬀerent from observed data (Table 2). These results show that the frequency of less extreme precipitation days, D90 and D95, are adequately represented in Mod and that bias-correction via either method does not adversely aﬀect the representation of D90 and D95 indices. The more extreme S99 and D99 indices were signiﬁcantly overestimated in Mod (p < 0.0001) (Table 2). S99 and D99 distributions calculated from EQM- and EQM-LIN-corrected data were not signiﬁcantly diﬀerent from ob- served data (Figure 6 a and b). However, the distribution of S99 derived from EQM-LIN was more similar to observed data compared to EQM (EQM-LIN p-value = 0.39, EQM p-value = 0.07) (Table 2). of KS tests for both D90 and D95 conﬁrmed that distributions of Mod and bias-corrected data were not signiﬁcantly diﬀerent from observed data (Table 2). These results show that the frequency of less extreme precipitation days, D90 and D95, are adequately represented in Mod and that bias-correction via either method does not adversely aﬀect the representation of D90 and D95 indices. 4.2 ETCCDI indices Although, as stated in the last paragraph, the frequency of less extreme precipitation days (D90 and D95) is represented well in Mod, the frequency of very extreme precipitation days and intensity of very extreme events (as quantiﬁed by D99 and S99) are ex- tensively overestimated in Mod. Both bias-correction methods substantially reduce overestimation of both the frequency and intensity of very extreme precipitation. Title Suppressed Due to Excessive Length 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 Fig. 6 Boxplots of a) D90, D95, and D99 and b) S90, S95, and S99 for observed (Obs), model (Mod), EQM-, and EQM-LIN-corrected data. Boxplots reﬂect 30 annual values for each data type and ETCCDI index. Signiﬁcance of KS-tests of distributional similarity of Mod, EQM, or EQM-LIN compared to Obs at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (). Fig. 6 Boxplots of a) D90, D95, and D99 and b) S90, S95, and S99 for observed (Obs), model (Mod), EQM-, and EQM-LIN-corrected data. Boxplots reﬂect 30 annual values for each data type and ETCCDI index. Signiﬁcance of KS-tests of distributional similarity of Mod, EQM, or EQM-LIN compared to Obs at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (). TotalP, WetDays, and SPI TotalP was signiﬁcantly overestimated in Mod (p < 0.0001), but distributions of TotalP calculated using either bias-corrected dataset were not signiﬁcantly diﬀerent from observed data (p = 0.81) (Figure 7; TotalP, WetDays, and SPI TotalP was signiﬁcantly overestimated in Mod (p < 0.0001), but distributions of TotalP calculated using either bias-corrected dataset were not signiﬁcantly diﬀerent from observed data (p = 0.81) (Figure 7; 18 Maike Holthuijzen et al. Table 2). Thus, both bias-correction methods were highly eﬀective in correcting total annual precipitation. Table 2). 4.2 ETCCDI indices Table 2 Two-sample Kolmogorov-Smirnov (KS) test results for raw model (Mod), EQM- and EQM-LIN-corrected distributions of ETCCDI indices compared to observed distribu- tions of ETCCDI indices. D is the KS test statistic. P-values refer to a two-sided null hypothesis; p-values < 0.05 indicate that the distribution of a particular ETCCDI index for either Mod, EQM-LIN or EQM is signiﬁcantly diﬀerent from that of observed data at the 5% signiﬁcance level. All ETCCDI index distributions consisted of 30 annual values. Signif- icance of KS-tests at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (). S90 D p Mod 0.73 <0.0001 EQM-LIN 0.33 0.07 EQM 0.40 0.02 S95 D p Mod 0.43 0.007* EQM-LIN 0.33 0.007* EQM 0.40 0.02 S99 D p Mod 0.93 <0.0001* EQM-LIN 0.23 0.39 EQM 0.33 0.07 D90 D p Mod 0.17 0.80 EQM-LIN 0.13 0.95 EQM 0.17 0.80 D95 D p Mod 0.17 0.80 EQM-LIN 0.13 0.95 EQM 0.17 0.80 D99 D p Mod 0.97 <0.0001* EQM-LIN 0.10 1 EQM 0.10 1 TotalP D p Mod 0.73 <0.0001* EQM-LIN 0.17 0.808 EQM 0.17 0.808 WetDays D p Mod 0.30 0.13 EQM-LIN 0.97 <0.0001* EQM 0.90 <0.0001* SPI D p Mod 0.73 <0.0001* EQM-LIN 0.53 0.0003* EQM 0.60 <0.0001* 4.2 ETCCDI indices Thus, both bias-correction methods were highly eﬀective in correcting total annual precipitation. The distribution of WetDays derived from Mod did not diﬀer signiﬁcantly from observed data (p = 0.13) (Table 2). However, WetDay distributions cal- culated from EQM- and EQM-LIN-corrected data were signiﬁcantly underes- timated relative to observed data (p < 0.0001) (Figure 7; Table 2). SPI was overestimated by Mod, due to the large overestimation of Total P; SPI was overestimated to a lesser degree, by EQM- and EQM-LIN-corrected data due to the underestimation of WetDays (Figure 7). Distributions of SPI calcu- lated from EQM, EQM-LIN, and Mod all diﬀered signiﬁcantly from observed data (Table 2). Although bias-correction via either EQM-LIN or EQM results in underestimating WetDays, annual precipitation totals (TotalP) are eﬀec- tively corrected. Moreover, while the distribution of WetDays is adequately represented in Mod, Mod contains an excessive number of low-precipitation occurrences relative to observed data (see Supplementary Materials, section 4). However, despite the underestimation of wet day frequency following bias- correction, precipitation intensity (SPI) is slightly improved compared to raw model data. Fig. 7 Boxplots of TotalP, WetDays, and SPI for observed (Obs), model (Mod), EQM-, and EQM-LIN-corrected data. Boxplots reﬂect 30 annual values for each data type and ETCCDI index. Signiﬁcance of KS-tests of distributional similarity of Mod, EQM, or EQM- LIN compared to Obs at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (). Fig. 7 Boxplots of TotalP, WetDays, and SPI for observed (Obs), model (Mod), EQM-, and EQM-LIN-corrected data. Boxplots reﬂect 30 annual values for each data type and ETCCDI index. Signiﬁcance of KS-tests of distributional similarity of Mod, EQM, or EQM- LIN compared to Obs at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (). Title Suppressed Due to Excessive Length 19 Table 2 Two-sample Kolmogorov-Smirnov (KS) test results for raw model (Mod), EQM- and EQM-LIN-corrected distributions of ETCCDI indices compared to observed distribu- tions of ETCCDI indices. D is the KS test statistic. P-values refer to a two-sided null hypothesis; p-values < 0.05 indicate that the distribution of a particular ETCCDI index for either Mod, EQM-LIN or EQM is signiﬁcantly diﬀerent from that of observed data at the 5% signiﬁcance level. All ETCCDI index distributions consisted of 30 annual values. Signif- icance of KS-tests at α = 0.05, adjusted with the Holm-Bonferroni method, are indicated with (*). 5 Discussion Local-scale modeling eﬀorts in hydrology, ecology, agriculture, and economics, as well as climate impact assessments, require high-resolution climate prod- ucts. Since climate extremes exert a large inﬂuence on humans and the en- 20 Maike Holthuijzen et al. vironment, it is crucial that extremes are accurately represented in climate products. An eﬀective way to obtain high-resolution climate products is to statistically downscale and bias-correct dynamically downscaled output from an RCM. Bias-correction of precipitation extremes, in particular, is a diﬃcult task. In this study, we developed a hybrid bias-correction method, EQM-LIN, that combines the eﬃcacy of EQM for bias-correcting the bulk of raw model data, with a robust linear adjustment for correcting distributional tails. We found that EQM-LIN results in the accurate representation of mean and ex- treme precipitation. EQM-LIN outperformed EQM in terms of model skill (MAE and MAE95) and performed at least as well or better than EQM with respect to most ETCCDI climatological indices. Furthermore, our study in- dicates that a linear correction, as implemented in EQM-LIN, is resistant to overﬁtting and results in a more robust TF at higher quantiles, both of which can decrease uncertainty bias-corrected data. The substantial diﬀerence in performance between EQM-LIN and EQM with respect to model skill is due to the diﬀerent ways in which TFs are constructed at extreme tails. In EQM, distributional tails are corrected with a ﬂexible TF that closely interpolates the quantile-quantile map of raw and observed data. However, since data at extreme tails is, by deﬁnition, scarce and variable, the TF produced by EQM may be unstable and can result in a faulty correction on out-of-sample model data [10] [6]. In our study, MAE95 values of EQM increased markedly between months 8 and 11, reaching a maximum in month 9, while those of EQM-LIN remained near 2.5mm (Figure 5 b). An inspection of training and testing datasets used during cross-validation reveals that often, the association between raw model and observed quantiles (the quantile-quantile map) was quite diﬀerent between training and corresponding testing datasets. In such cases, EQM tended to overﬁt on training data, and consequently, the correction applied to testing data was unsuitable. Figure 9 depicts such a scenario for month 9, when the diﬀerence in MAE95 between the two bias-correction methods was large. In Figure 9, the EQM TF constructed with training data (black dots) extends non-linearly above the one- to-one line and then increases sharply. 5 Discussion The shape of the training TF indicates that, generally, raw model quantiles are too low relative to those of observed data. When the training TF is applied to test data, raw model values in the tails, especially, are increased. For instance, a raw model value of 58.6mm would be corrected to 81.8mm (Figure 9). However, the relationship between raw model and observed quantiles in the test data (blue dots), indicates that raw model quantiles are only slightly too high compared to observed quantiles (Figure 9). When raw model data in the test set are bias-corrected with the training TF, raw model values are increased too much relative to observed values (Figure 9). The quantile-quantile map of corrected model quantiles and observed quantiles (which should lie near or on the one-to-one line if the cor- rection was good) is shifted far to the right of one-to-one line, indicating that corrected model values, especially in the tails, are too high. This example shows that the ﬂexibility of EQM is also what makes it susceptible to over- ﬁtting on calibration data and supports other studies showing that EQM is Title Suppressed Due to Excessive Length 21 sensitive to the choice of, and can overﬁt on calibration data [69] [6] [37] [63] [45]. For the same scenario, EQM-LIN produces a linear TF at extreme tails with a slope of 1 and an intercept of δ (the constant correction factor) (Figure 10). Raw model values are adjusted by a constant, δ. Though this approach is less ﬂexible than that of EQM, it produces more stable TFs and is less sensitive to training data. In Figure 10, the training TF for EQM-LIN (black dots) is linear and does not exhibit the ﬂuctuations apparent in the training TF of EQM (Figure 9). The intercept (δ) of the TF in Figure 10 is slightly less than zero, which means that raw model values will be decreased by δ. The TF for EQM-LIN represents an appropriate correction, as model quantiles in the test dataset are, in fact, too high relative to observed quantiles (Figure 10, blue dots). For instance, the TF of EQM-LIN corrects a raw model value of 58.6mm to 58.1mm (Figure 10). Accordingly, the quantile-quantile map of corrected model quantiles and observed quantiles is close to the one-to-one line, indicating a satisfactory correction. 5 Discussion Figures 9 and 10 are representative of scenarios in which the relationship between raw model and observed quantiles diﬀer between training and test- ing data and highlight the diﬀerences in EQM and EQM-LIN. In our study area, such scenarios are common in months when precipitation is variable and when extreme precipitation events are more likely (months 6-9). The diﬀer- ence in bias-correction between EQM-LIN and EQM can also be inspected in downscaled, bias-corrected data over the study region. Figure 8 depicts raw, downscaled, and corrected and downscaled precipitation data for two days in months 8 (Figure 8 a) and 9 (Figure 8 b). Note that for the two days in Fig- ure 8 EQM results in the increase of high precipitation values (bright yellow regions). Maike Holthuijzen et al. 22 Fig. 8 Raw model data (4km grid), downscaled raw model data (1km grid), downscaled and bias-corrected data via EQM and EQM-LIN for one day in a) August 1987 and b) September 1987. Fig. 8 Raw model data (4km grid), downscaled raw model data (1km grid), downscaled and bias-corrected data via EQM and EQM-LIN for one day in a) August 1987 and b) September 1987. Title Suppressed Due to Excessive Length 23 Though EQM-LIN signiﬁcantly outperformed EQM in terms of model skill (MAE and MAE95), results were not as dramatic for climatological (ETCCDI) indices. ETCCDI indices are calculated using spatially averaged, daily data, which reduces variation and may explain the similarity in performance of EQM and EQM-LIN for ETCCDI indices. Bias correction via both EQM and EQM- LIN resulted in improvements over raw data for most indices. Though both bias-correction methods improved the overestimation of total annual mean precipitation (TotalP) as well as total extreme annual precipitation (Sum90, Sum95) exhibited in raw model data, EQM-LIN performed slightly better than EQM for moderate extremes (Sum95). The diﬀerence between bias-corrected and raw model data was most dramatic for very extreme precipitation totals (S99 and D99), conﬁrming that RCMs are generally unable to model very extreme events [2] [47]. Despite the substantial overestimation of the most extreme events in raw model data, both bias-correction methods were highly eﬀective in correcting S99 and D99. Interestingly, the distribution of raw model wet day frequency (WetDays) was similar to that of observed data, while bias-correction via either method resulted in considerable underestimation of wet day frequency. 5 Discussion The negative impact of bias-correction on wet day frequency is most likely due to the ex- cessive number of low-precipitation occurrences (“drizzle eﬀect”) [2] [47] in raw model data. EQM, which is used to correct low-valued quantiles in both bias-correction methods, results in the majority of excessive low-precipitation days being set to zero. The underestimation of wet day frequency after bias- correction via EQM is not unusual; similar results were found by [19] and [55]. Moreover, although wet-day frequency appears to be adequately represented in raw model data, it comes at the expense of substantial overestimation of total annual precipitation (TotalP) and precipitation intensity (SPI). After bias- correction via either method, precipitation intensity is better represented, and the distribution of total annual precipitation is very close to that of observed data. Thus, for most climatological indices, bias-correction via either method provides critical improvements to raw model data, especially with respect to extremes. Maike Holthuijzen et al. 5.1 Conclusion In this study, we show that a hybrid EQM approach for bias-correction (EQM- LIN), in which the majority of model data is corrected via EQM and extreme tails are corrected by a linear TF, resists overﬁtting on calibration data, in- creases overall and model skill, especially at extreme tails, and results in a better representation of climatological indices compared to conventional EQM. Our method is simple, intuitive, and easy to implement, making it a suitable alternative to EQM for bias-correcting historical and future climate simula- tions. Though we apply the method to precipitation data, we expect it could be applied to other climate variables as well. Future work might include ad- Maike Holthuijzen et al. Maike Holthuijzen et al. 24 justing the slope of the linear correction or using another function to construct the TF at extreme tails. Title Suppressed Due to Excessive Length 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Fig. 9 Construction of the EQM TF in a train-test scenario; data for this plot reﬂect one particular train-test fold used during cross-validation for month 9 (September). The TF obtained from training data is shown in black. The quantile-quantile map of model and observed data in the test set is shown in blue. The corrected quantile-quantile map (quantiles of corrected model data versus quantiles of observed data) in the test set are shown in red. 5.1 Conclusion xmod,t and xcorr,t denote model and corrected model values, respectively, for day t. Gray arrows indicate how model data in the test set is corrected, based on the TF from training data. 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 Fig. 9 Construction of the EQM TF in a train-test scenario; data for this plot reﬂect one particular train-test fold used during cross-validation for month 9 (September). The TF obtained from training data is shown in black. The quantile-quantile map of model and observed data in the test set is shown in blue. The corrected quantile-quantile map (quantiles of corrected model data versus quantiles of observed data) in the test set are shown in red. xmod,t and xcorr,t denote model and corrected model values, respectively, for day t. Gray arrows indicate how model data in the test set is corrected, based on the TF from training data. Maike Holthuijzen et al. 26 26 Maike Holthuijzen et al. Fig. 10 Construction of the EQM-LIN TF in a train-test scenario; data for this plot reﬂect one particular train-test fold used during cross-validation for month 9 (September). The TF obtained from training data is shown in black. The quantile-quantile map of model and observed data in the test set is shown in blue. The corrected quantile-quantile map (quantiles of corrected model data versus quantiles of observed data) in the test set are shown in red. xmod,t and xcorr,t denote raw model and corrected model values, respectively, for day t. Gray arrows indicate how raw model data in the test set is corrected, based on the training-set TF. The threshold (dashed line), indicates the 79th quantile of model data (6.88mm). 5.1 Conclusion For ease of viewing, plot a) (gray box) shows the scenario at selected lower (0-10 mm) precipitation quantiles, and plot b) (gray dotted box) shows the scenario at selected extreme (50-80mm) l 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 0 Fig. 10 Construction of the EQM-LIN TF in a train-test scenario; data for this plot reﬂect one particular train-test fold used during cross-validation for month 9 (September). The TF obtained from training data is shown in black. The quantile-quantile map of model and observed data in the test set is shown in blue. A.1 Estimating the threshold, T The ﬁrst step for obtaining the threshold T is to estimate τLIN from the data. We chose τLIN to be 0.79, based on a grid search over a range of values in a ﬁve fold cross-validation approach. We chose the value of τLIN that resulted in the minimization of the mean absolute error of observed and model ecdfs above the 95th percentile (MAE95), [68] (section 3). It is crucial that τLIN be estimated using cross-validation; our result of τLIN = 0.79 may not generalize to all data. To obtain T, we must assume a ﬁxed value of τLIN. The next steps involves the construc- tion of ecdfs for observed and model data in the calibration period. Ecdfs are constructed using 10,000 quantiles evenly spaced between 0 and 1. Next, the threshold, T is computed as ecdf−1 mod(τLIN). Note that, T is the model precipitation value in mm corresponding to the quantile τLIN (whereas 0 ≤τLIN ≤1). 5.1 Conclusion The corrected quantile-quantile map (quantiles of corrected model data versus quantiles of observed data) in the test set are shown in red. xmod,t and xcorr,t denote raw model and corrected model values, respectively, for day t. Gray arrows indicate how raw model data in the test set is corrected, based on the training-set TF. The threshold (dashed line), indicates the 79th quantile of model data (6.88mm). For ease of viewing, plot a) (gray box) shows the scenario at selected lower (0-10 mm) precipitation quantiles, and plot b) (gray dotted box) shows the scenario at selected extreme (50-80mm) precipitation quantiles. Title Suppressed Due to Excessive Length 27 A.2 Estimating the intercept, δ To obtain δ, we assume that T has been calculated. Ecdfs of observed and model data are constructed using 10,000 quantiles evenly spaced between 0 and 1. Values in the ecdfs of model and observed data are sorted in increasing order. Note the rank of T within the sorted precipitation values of the model ecdf; the rank value will be denoted as RT . For example, suppose T = 12mm and the rank of T within the ecdf of model data is 5,000, then RT = 5000. Next, select the precipitation value from sorted, observed ecdf at rank RT and denote this value as Tobs. The intercept of the linear TF, δ which represents the constant correction, is calculated as the diﬀerence Tobs−T. Continuing with the example, suppose Tobs = 9.1mm; then δ = 9.1 −12 = −2.9. This means model extremes (all values ≥T) will be decreased by 2.9mm. The constant correction at extremes, δ, is similar to the constant extrapolation correction used by [7]. However, here, the constant correction is the diﬀerence T −Tobs, whereas in [7], it is ecdf−1 obs(1.00) −ecdf−1 mod(1.00) as in [7]. The KS test statistic, D is computed as (3) (Dn = sup x \|Fn(x) −Gn(x)\|). (3) (Dn = sup x \|Fn(x) −Gn(x)\|). In (3), Fn and Gn are the two ecdfs being compared, n denotes the number of indepen- dent and identically distributed ordered values used to obtain Fn and Gn, and sup x is the supremum of the collection of n distances. supremum of the collection of n distances. B KS Test The KS test statistic, D is computed as A Estimating the threshold T and intercept δ A.1 Estimating the threshold, T C Holm-Bonferroni method for multiple comparisons When multiple statistical comparisons are made, it is often necessary to adjust the Type I error rate (commonly referred to as the signiﬁcance level or α). The Type I error rate is the probability of falsely rejecting the null hypotheses when it is, in fact, true (a false positive). In the context of multiple hypothesis testing, it is often desirable to adjust the family-wise error rate (FWER), the probability of rejecting one null hypothesis in m hypothesis tests. 28 Maike Holthuijzen et al. The Holm-Bonferroni method is suitable when a less conservative adjustment of the FWER is preferred. Suppose m hypothesis tests have been conducted, and m p-values have been calculated. The Holm-Bonferroni adjustment for the FWER involves two steps: 1. Order p-values from least to greatest and assign each p-value a rank from 1 to k, k = 1 . . . m 2. Find the smallest p-value such that pk < α m+1−k . If the condition in step 2 is true, the p-value is signiﬁcant; if the condition in step 2 if false, the p-value is not signiﬁcant. Author’s Contribution All authors contributed to the study conception and design. Data was prepared by Jonathan M. Winter. Model development, writing, and analyses were performed by Maike F. Holthui- jzen. All authors edited, commented on, and reviewed the manuscript. All authors read and approved the ﬁnal manuscript. Ethics Approval No animals were involved in this manuscript. Consent to Participate Consent to Participate All authors consent to participating in the review process for this manuscript. Consent for Publication All authors consent to the publication of this manuscript. Data Availability Data will be available upon request. Code availability Code will be available in a public Github repository. Funding Statement The development of this manuscript was supported by the National Science Foundation under VT EPSCoR Grant No. NSF OIA 1556770. D Summary results for ETCCDI indices D Summary results for ETCCDI indices Table 3 shows the 25th, 50th, and 75th quantiles for each data type (Mod, EQM, and EQM-LIN) and ETCCDI index. Title Suppressed Due to Excessive Length 29 Table 3 25th (Q25), 50th (Q50), and 75th (Q75) quantiles of ETCCDI indices for observed data (Obs), raw model data (Mod), and EQM-, and EQM-LIN-corrected data during the calibration period (1976-2005). Each ETCCDI index was calculated using 30 annual values. 6-2005). Each ETCCDI index was calculated u Data type Q25 Q50 Q75 Sum90 Obs 401.63 456.20 531.18 Mod 533.62 595.78 668.31 EQM 501.05 574.38 645.14 EQM-LIN 463.71 542.41 605.23 Sum95 Obs 239.03 287.21 338.08 Mod 311.85 377.51 428.63 EQM 307.08 378.51 431.53 EQM-LIN 287.94 337.40 400.09 Sum99 Obs 60.29 83.17 114.33 Mod 311.85 377.51 428.63 EQM 55.09 105.16 154.46 EQM-LIN 55.10 101.02 138.54 D90 Obs 32.25 35.50 42.00 Mod 33.00 36.50 40.00 EQM 32.25 37.00 39.75 EQM-LIN 32.25 37.00 39.75 D95 Obs 14.25 18.00 21.75 Mod 15.25 17.50 20.75 EQM 16.00 19.00 20.75 EQM-LIN 16.00 18.00 21.00 D99 Obs 2.25 3.50 4.00 Mod 15.25 17.50 20.75 EQM 2.00 3.50 5.00 EQM-LIN 2.00 3.50 5.00 TotalP Obs 961.92 1032.81 1112.80 Mod 1242.30 1296.84 1367.50 EQM 991.24 1077.74 1132.50 EQM-LIN 951.56 1022.26 1076.33 WetDays Obs 283.00 289.50 293.00 Mod 289.00 294.00 299.75 EQM 249.25 258.00 265.75 EQM-LIN 240.50 251.50 259.25 SPI Obs 3.39 3.55 3.78 Mod 4.19 4.47 4.61 EQM 3.74 4.13 4.34 EQM-LIN 3.77 4.07 4.29 Maike Holthuijzen et al. 30 References 1. Alexander, L., Donat, M., Takayama, Y., Yang, H.: The climdex project: creation of long-term global gridded products for the analysis of temperature and precipitation extremes. In: WCRP Open Science conference, Denver (2011) 2. Baigorria, G.A., Jones, J.W., Shin, D.W., Mishra, A., OBrien, J.J.: Assessing uncer- tainties in crop model simulations using daily bias-corrected regional circulation model outputs. Climate Research 34(3), 211–222 (2007) 3. Bannister, D., Orr, A., Jain, S.K., Holman, I.P., Momblanch, A., Phillips, T., Adeloye, A.J., Snapir, B., Waine, T.W., Hosking, J.S., et al.: Bias correction of high-resolution regional climate model precipitation output gives the best estimates of precipitation in himalayan catchments. Journal of Geophysical Research: Atmospheres 124(24), 14220– 14239 (2019) ( ) 4. Behnke, R., Vavrus, S., Allstadt, A., Albright, T., Thogmartin, W.E., Radeloﬀ, V.C.: Evaluation of downscaled, gridded climate data for the conterminous united states. Ecological applications 26(5), 1338–1351 (2016) g ( ) ( ) 5. Beirlant, J., Goegebeur, Y., Segers, J., Teugels, J.L.: Statistics of extremes: theory and applications. John Wiley & Sons (2006) ( ) 6. Berg, P., Feldmann, H., Panitz, H.J.: Bias correction of high resolution regional climate model data. Journal of Hydrology 448, 80–92 (2012) 7. Boé, J., Terray, L., Habets, F., Martin, E.: Statistical and dynamical downscaling of the seine basin climate for hydro-meteorological studies. International Journal of Climatol- ogy: A Journal of the Royal Meteorological Society 27(12), 1643–1655 (2007) 8. Caldwell, P., Chin, H.N.S., Bader, D.C., Bala, G.: Evaluation of a WRF dynamical downscaling simulation over California. Climatic change 95(3-4), 499–521 (2009) 9. Cannon, A.J., Piani, C., Sippel, S.: Bias correction of climate model output for impact models. In: Climate Extremes and Their Implications for Impact and Risk Assessment, pp. 77–104. Elsevier (2020) 10. Cannon, A.J., Sobie, S.R., Murdock, T.Q.: Bias correction of gcm precipitation by quantile mapping: How well do methods preserve changes in quantiles and extremes? Journal of Climate 28(17), 6938–6959 (2015) ( ) ( ) 11. Daly, C., Taylor, G., Gibson, W., Parzybok, T., Johnson, G., Pasteris, P.: High-quality spatial climate data sets for the united states and beyond. Transactions of the ASAE 43(6), 1957 (2000) ( ), ( ) 12. Durre, I., Menne, M.J., Gleason, B.E., Houston, T.G., Vose, R.S.: Comprehensive au- tomated quality assurance of daily surface observations. Journal of Applied Meterology and Climatology 49(8), 1615–1633 (2010) ( ) ( ) 13. Conﬂicts of interest The authors have no conﬂicts of interest to declare that are relevant to the content of this article. Title Suppressed Due to Excessive Length 31 10 12 9 11 13 References Ekström, M., Grose, M.R., Whetton, P.H.: An appraisal of downscaling methods used in climate change research. Wiley Interdisciplinary Reviews: Climate Change 6(3), 301–319 (2015) 14. Enayati, M., Bozorg-Haddad, O., Bazrafshan, J., Hejabi, S., Chu, X.: Bias correction capabilities of quantile mapping methods for rainfall and temperature variables. Journal of Water and Climate Change 12(2), 401–419 (2021) 15. Fang, G., Yang, J., Chen, Y., Zammit, C.: Comparing bias correction methods in down- scaling meteorological variables for a hydrologic impact study in an arid area in china. Hydrology and Earth System Sciences 19(6), 2547–2559 (2015) 16. Feser, F., Rockel, B., von Storch, H., Winterfeldt, J., Zahn, M.: Regional climate models add value to global model data: a review and selected examples. Bulletin of the American Meteorological Society 92(9), 1181–1192 (2011) 17. Field, C.B., Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D., Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., et al.: Contribution of working group ii to the ﬁfth assessment report of the intergovernmental panel on climate change. Climate change (2014) ( ) 18. Flint, L.E., Flint, A.L.: Downscaling future climate scenarios to ﬁne scales for hydrologic and ecological modeling and analysis. Ecological Processes 1(1), 2 (2012) 19. Fowler, H., Ekström, M., Blenkinsop, S., Smith, A.: Estimating change in extreme eu- ropean precipitation using a multimodel ensemble. Journal of Geophysical Research: Atmospheres 112(D18) (2007) 32 Maike Holthuijzen et al. 20. Fowler, H.J., Blenkinsop, S., Tebaldi, C.: Linking climate change modelling to impacts studies: recent advances in downscaling techniques for hydrological modelling. Interna- tional journal of climatology 27(12), 1547–1578 (2007) 21. Franklin, J., Davis, F.W., Ikegami, M., Syphard, A.D., Flint, L.E., Flint, A.L., Hannah, L.: Modeling plant species distributions under future climates: how ﬁne scale do climate projections need to be? Global change biology 19(2), 473–483 (2013) 22. Friederichs, P., Hense, A.: Statistical downscaling of extreme precipitation events using censored quantile regression. Monthly Weather Review 135(6), 2365–2378 (2007) 23. Gao, X., Pal, J.S., Giorgi, F.: Projected changes in mean and extreme precipitation over the mediterranean region from a high resolution double nested rcm simulation. Geophysical Research Letters 33(3) (2006) 24. Gobiet, A., Suklitsch, M., Heinrich, G.: The eﬀect of empirical-statistical correction of intensity-dependent model errors on the temperature climate change signal. Hydrology and Earth System Sciences 19(10), 4055–4066 (2015) 25. References Grillakis, M.G., Koutroulis, A.G., Daliakopoulos, I.N., Tsanis, I.K.: A method to pre- serve trends in quantile mapping bias correction of climate modeled temperature. Earth System Dynamics 8(3), 889 (2017) 26. Grillakis, M.G., Koutroulis, A.G., Tsanis, I.K.: Multisegment statistical bias correc- tion of daily gcm precipitation output. Journal of Geophysical Research: Atmospheres 118(8), 3150–3162 (2013) 27. Gudmundsson, L.: qmap: Statistical transformations for post-processing climate model output (2016). R package version 1.0-4 28. Gudmundsson, L., Bremnes, J., Haugen, J., Engen-Skaugen, T.: Downscaling rcm pre- cipitation to the station scale using statistical transformations–a comparison of methods. Hydrology and Earth System Sciences 16(9), 3383–3390 (2012) 29. Gutjahr, O., Heinemann, G.: Comparing precipitation bias correction methods for high- resolution regional climate simulations using cosmo-clm. Theoretical and applied cli- matology 114(3), 511–529 (2013) 30. Hanssen-Bauer, I., Achberger, C., Benestad, R., Chen, D., Førland, E.: Statistical down- scaling of climate scenarios over Scandinavia. Climate Research 29(3), 255–268 (2005) 31. Hayhoe, K., Wake, C.P., Huntington, T.G., Luo, L., Schwartz, M.D., Sheﬃeld, J., Wood, E., Anderson, B., Bradbury, J., DeGaetano, A., et al.: Past and future changes in climate and hydrological indicators in the us northeast. Climate Dynamics 28(4), 381–407 (2007) 32. Heo, J.H., Ahn, H., Shin, J.Y., Kjeldsen, T.R., Jeong, C.: Probability distributions for a quantile mapping technique for a bias correction of precipitation data: A case study to precipitation data under climate change. Water 11(7), 1475 (2019) 33. Hnilica, J., Hanel, M., Puš, V.: Multisite bias correction of precipitation data from regional climate models. International Journal of Climatology 37(6), 2934–2946 (2017) 33. Hnilica, J., Hanel, M., Puš, V.: Multisite bias correction of precipitation data from regional climate models. International Journal of Climatology 37(6), 2934–2946 (2017) 34. Hoﬀmann, H., Rath, T.: Meteorologically consistent bias correction of climate time series for agricultural models. Theoretical and Applied Climatology 110(1), 129–141 (2012) ( ) ( ) 34. Hoﬀmann, H., Rath, T.: Meteorologically consistent bias correction of climate time series for agricultural models. Theoretical and Applied Climatology 110(1), 129–141 (2012) 35. Holden, Z.A., Abatzoglou, J.T., Luce, C.H., Baggett, L.S.: Empirical downscaling of daily minimum air temperature at very ﬁne resolutions in complex terrain. Agricultural and Forest Meteorology 151(8), 1066–1073 (2011) 36. Holm, S.: A simple sequentially rejective multiple test procedure. Scandinavian journal of statistics pp. 65–70 (1979) 37. Holthuijzen, M.F., Beckage, B., Clemins, P.J., Higdon, D., Winter, J.M.: Constructing high-resolution, bias-corrected climate products: A comparison of methods. References Journal of Applied Meteorology and Climatology 60(4), 455–475 (2021) 38. Huang, H., Winter, J.M., Osterberg, E.C., Hanrahan, J., Bruyère, C.L., Clemins, P., Beckage, B.: Simulating precipitation and temperature in the lake champlain basin using a regional climate model: limitations and uncertainties. Climate Dynamics 54(1-2), 69– 84 (2020) 39. Huang, H., Winter, J.M., Osterberg, E.C., Horton, R.M., Beckage, B.: Total and extreme precipitation changes over the northeastern united states. Journal of Hydrometeorology 18(6), 1783–1798 (2017) Title Suppressed Due to Excessive Length 33 40. Ivanov, M.A., Kotlarski, S.: Assessing distribution-based climate model bias correction methods over an alpine domain: added value and limitations. International Journal of Climatology 37(5), 2633–2653 (2017) 40. Ivanov, M.A., Kotlarski, S.: Assessing distribution-based climate model bias correction methods over an alpine domain: added value and limitations. International Journal of Climatology 37(5), 2633–2653 (2017) 41. Jakob Themeßl, M., Gobiet, A., Leuprecht, A.: Empirical-statistical downscaling and er- ror correction of daily precipitation from regional climate models. International Journal of Climatology 31(10), 1530–1544 (2011) 42. Karl, T.R., Nicholls, N., Ghazi, A.: Clivar/gcos/wmo workshop on indices and indicators for climate extremes workshop summary. In: Weather and climate extremes, pp. 3–7. Springer (1999) 43. Kim, D.I., Kwon, H.H., Han, D.: Exploring the long-term reanalysis of precipitation and the contribution of bias correction to the reduction of uncertainty over south korea: A composite gamma-pareto distribution approach to the bias correction. Hydrology and Earth System Sciences Discussions pp. 1–53 (2018) Earth System Sciences Discussions pp. 1–53 (2018) 44. Laﬂamme, E.M., Linder, E., Pan, Y.: Statistical downscaling of regional climate model output to achieve projections of precipitation extremes. Weather and climate extremes 12, 15–23 (2016) , ( ) 45. Lafon, T., Dadson, S., Buys, G., Prudhomme, C.: Bias correction of daily precipitation simulated by a regional climate model: a comparison of methods. International Journal of Climatology 33(6), 1367–1381 (2013) 46. Lanzante, J.R., Dixon, K.W., Adams-Smith, D., Nath, M.J., Whitlock, C.E.: Evaluation of some distributional downscaling methods as applied to daily precipitation with an eye towards extremes. International Journal of Climatology 41(5), 3186–3202 (2021) 47. Leander, R., Buishand, T.A.: Resampling of regional climate model output for the simulation of extreme river ﬂows. Journal of Hydrology 332(3-4), 487–496 (2007) 48. Leung, L.R., Mearns, L.O., Giorgi, F., Wilby, R.L.: Regional climate research: needs and opportunities. Bulletin of the American Meteorological Society 84(1), 89–95 (2003) ( ) ( ) 49. References Perkins, S., Pitman, A., Holbrook, N., McAneney, J.: Evaluation of the AR4 climate models simulated daily maximum temperature, minimum temperature, and precipi- tation over Australia using probability density functions. Journal of climate 20(17), 4356–4376 (2007) ( ) 61. Peterson, T.: Climate change indices. WMO bulletin 54(2), 83–86 (2005) ) : Climate change indices. WMO bulletin 54(2), 83–86 (2 Peterson, T.: Climate change indices. WMO bulletin 54(2 62. Peterson, T.C., Vose, R.S.: An overview of the global historical climatology network temperature database. Bulletin of the American Meteorological Society 78(12), 2837– 2850 (1997) ( ) 63. Piani, C., Haerter, J., Coppola, E.: Statistical bias correction for daily precipitation in regional climate models over europe. Theoretical and Applied Climatology 99(1-2), 187–192 (2010) ( ) 64. Pierce, D.W., Cayan, D.R., Maurer, E.P., Abatzoglou, J.T., Hegewisch, K.C.: Improved bias correction techniques for hydrological simulations of climate change. Journal of Hydrometeorology 16(6), 2421–2442 (2015) y gy ( ) ( ) 65. Plummer, M., Stukalov, A., Denwood, M., Plummer, M.M.: Package rjags. Vienna, Austria (2016) ( ) 66. R Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2018). URL https://www.R-project.org/ f 67. Razali, N.M., Wah, Y.B., et al.: Power comparisons of shapiro-wilk, kolmogorov- smirnov, lilliefors and anderson-darling tests. Journal of statistical modeling and ana- lytics 2(1), 21–33 (2011) 68. Reiter, P., Gutjahr, O., Schefczyk, L., Heinemann, G., Casper, M.: Bias correction of ensembles precipitation data with focus on the eﬀect of the length of the calibration period. Meteorologische Zeitschrift pp. 85–96 (2016) ( ) 69. Reiter, P., Gutjahr, O., Schefczyk, L., Heinemann, G., Casper, M.: Does applying quan- tile mapping to subsamples improve the bias correction of daily precipitation? Interna- tional Journal of Climatology 38(4), 1623–1633 (2018) ( ) ( ) 70. Roberts, D.R., Wood, W.H., Marshall, S.J.: Assessments of downscaled climate data with a high-resolution weather station network reveal consistent but predictable bias. International Journal of Climatology 39(6), 3091–3103 (2019) 71. Rosenberg, E.A., Keys, P.W., Booth, D.B., Hartley, D., Burkey, J., Steinemann, A.C., Lettenmaier, D.P.: Precipitation extremes and the impacts of climate change on stormwater infrastructure in washington state. Climatic change 102(1), 319–349 (2010) 72. Shin, J.Y., Lee, T., Park, T., Kim, S.: Bias correction of rcm outputs using mixture distributions under multiple extreme weather inﬂuences. Theoretical and Applied Cli- matology 137(1), 201–216 (2019) ( ) ( ) 73. References Livneh, B., Bohn, T.J., Pierce, D.W., Munoz-Arriola, F., Nijssen, B., Vose, R., Cayan, D.R., Brekke, L.: A spatially comprehensive, hydrometeorological data set for mexico, the us, and southern canada 1950–2013. Scientiﬁc data 2(1), 1–12 (2015) 50. Luo, M., Liu, T., Meng, F., Duan, Y., Frankl, A., Bao, A., De Maeyer, P.: Comparing bias correction methods used in downscaling precipitation and temperature from re- gional climate models: a case study from the kaidu river basin in western china. Water 10(8), 1046 (2018) ( ) ( ) 51. Mamalakis, A., Langousis, A., Deidda, R., Marrocu, M.: A parametric approach for simultaneous bias correction and high-resolution downscaling of climate model rainfall. Water Resources Research 53(3), 2149–2170 (2017) ( ) ( ) 52. Maraun, D.: Bias correcting climate change simulations-a critical review. Current Cli- mate Change Reports 2(4), 211–220 (2016) 53. Maraun, D., Shepherd, T.G., Widmann, M., Zappa, G., Walton, D., Gutierrez, J.M., Hagemann, S., Richter, I., Soares, P.M., Hall, A., et al.: Towards process-informed bias correction of climate change simulations. Nature Climate Change 7(11), 764 (2017) 54. Maraun, D., Wetterhall, F., Ireson, A., Chandler, R., Kendon, E., Widmann, M., Brienen, S., Rust, H., Sauter, T., Themeßl, M., et al.: Precipitation downscaling un- der climate change: Recent developments to bridge the gap between dynamical models and the end user. Reviews of Geophysics 48(3) (2010) ( ) ( ) 55. Martins, J., Fraga, H., Fonseca, A., Santos, J.A.: Climate projections for precipitation and temperature indicators in the douro wine region: The importance of bias correction. Agronomy 11(5), 990 (2021) 56. Mearns, L., Giorgi, F., Whetton, P., Pabon, D., Hulme, M., Lal, M.: Guidelines for use of climate scenarios developed from regional climate model experiments. Data Distribution Centre of the Intergovernmental Panel on Climate Change (2003) 57. Miao, C., Su, L., Sun, Q., Duan, Q.: A nonstationary bias-correction technique to remove bias in gcm simulations. Journal of Geophysical Research: Atmospheres 121(10), 5718– 5735 (2016) 58. Monaghan, A., Steinhoﬀ, D., Bruyere, C., Yates, D.: Ncar cesm global bias-corrected cmip5 output to support wrf/mpas research. Research Data Archive National Center Atmospheric Research Computational Information Systems Laboratory, Boulder. DOI 10, D6DJ5CN4 (2014) Maike Holthuijzen et al. 34 59. Oceanic, N., Administration, A.: Climate data online search (2018). URL https://www. ncdc.noaa.gov/cdo-web/search?datasetid=GHCND. Accessed: 2017-09-30 60. 12 10 9 References Shrestha, M., Acharya, S.C., Shrestha, P.K.: Bias correction of climate models for hydro- logical modelling–are simple methods still useful? Meteorological Applications 24(3), 531–539 (2017) ( ) 74. Skamarock, W.C., Klemp, J.B., Dudhia, J., Gill, D.O., Liu, Z., Berner, J., Huang, X.: A description of the advanced research wrf model (2019). URL https://opensky.ucar. edu/islandora/object/opensky:2898. Accessed: 2019-03-04 75. Smirnov, N., et al.: Table for estimating the goodness of ﬁt of empirical distributions. Annals of mathematical statistics 19(2), 279–281 (1948) 76. Stager, C., Thill, M.: Climate change in the champlain basin: What natural resource managers can expect and do, the nature conservancy adirondack ch. and vt ch. Rep., Keene Valley, NY (2010) 77. Tani, S., Gobiet, A.: Quantile mapping for improving precipitation extremes from re- gional climate models. Journal of Agrometeorology 21(4), 434–443 (2019) 78. Teutschbein, C., Seibert, J.: Bias correction of regional climate model simulations for hydrological climate-change impact studies: Review and evaluation of diﬀerent methods. Journal of Hydrology 456, 12–29 (2012) 79. Thornton, P.E., Thornton, M.M., Mayer, B.W., Wilhelmi, N., Wei, Y., Devarakonda, R., Cook, R.: Daymet: Daily surface weather on a 1 km grid for north america, 1980- 2008. Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center for Biogeochemical Dynamics (DAAC) (2012) Title Suppressed Due to Excessive Length 35 80. Tschöke, G.V., Kruk, N.S., de Queiroz, P.I.B., Chou, S.C., de Sousa Junior, W.C.: Comparison of two bias correction methods for precipitation simulated with a regional climate model. Theoretical and applied climatology 127(3), 841–852 (2017) 81. Um, M.J., Kim, H., Heo, J.H.: Hybrid approach in statistical bias correction of projected precipitation for the frequency analysis of extreme events. Advances in Water Resources 94, 278–290 (2016) , ( ) 82. USGS: The national map (2018). URL https://viewer.nationalmap.gov/basic/ ( ) USGS: The national map (2018). URL https://viewer.n 83. Walton, D., Hall, A.: An assessment of high-resolution gridded temperature datasets over california. Journal of Climate 31(10), 3789–3810 (2018) ( ) ( ) 84. Wang, T., Hamann, A., Spittlehouse, D.L., Murdock, T.Q.: Climatewna:high-resolution spatial climate data for western north america. Journal of Applied Meteorology and Climatology 51(1), 16–29 (2012) 85. Winter, J.M., Beckage, B., Bucini, G., Horton, R.M., Clemins, P.J.: Development and evaluation of high-resolution climate simulations over the mountainous northeastern united states. Journal of Hydrometeorology 17(3), 881–896 (2016) 86. Wood, A.W., Maurer, E.P., Kumar, A., Lettenmaier, D.P.: Long-range experimental hydrologic forecasting for the eastern united states. 12 13 References Journal of Geophysical Research: Atmospheres 107(D20), ACL–6 (2002) 87. Wootten, A.M., Dixon, K.W., Adams-Smith, D.J., McPherson, R.A.: Statistically down- scaled precipitation sensitivity to gridded observation data and downscaling technique. International Journal of Climatology 41(2), 980–1001 (2021) 88. Yang, W., Andréasson, J., Phil Graham, L., Olsson, J., Rosberg, J., Wetterhall, F.: Distribution-based scaling to improve usability of regional climate model projections for hydrological climate change impacts studies. Hydrology Research 41(3-4), 211–229 (2010) ( ) 89. Zia, A., Bomblies, A., Schroth, A.W., Koliba, C., Isles, P.D., Tsai, Y., Mohammed, I.N., Bucini, G., Clemins, P.J., Turnbull, S., et al.: Coupled impacts of climate and land use change across a river–lake continuum: insights from an integrated assessment model of lake champlains missisquoi basin, 2000–2040. Environmental Research Letters 11(11), 114026 (2016) PrecipManuscriptSupplementaryMaterial.pdf Supplementary Files This is a list of supplementary ¦les associated with this preprint. Click to download. PrecipManuscriptSupplementaryMaterial.pdf
https://openalex.org/W2462427603	https://zenodo.org/records/6137609/files/253-254.pdf	English	null	Non-alkaloidal Components of Tylophora hirsuta	Zenodo (CERN European Organization for Nuclear Research)	1,991	cc-by	1,025	NOTES NOTES Non.alkaloidal Components of Tylopkora hiuuta M.ALI Faculty of Pharmacy, ]amia Hamdard (Hamdard University), Hamdard Nagar, New Delhi-110 062 Manuscript received 13 August 1990, revisetZ 5 April7997, acceptetZ 24 April 7997 RECENTLY, the isolation of phenanthroindolizidine alkaloids from the dried aerial parts of Tylophora hirsuta has been reported by us 1 • Some of these alkaloids possess anticandial and antiamoebic properties2 • The isolation and characterisation of non-alkaloidal constituents ofT. hirsuta are reported here. RE The dried and powdered aerial parts (750 g) of T. hirsuta were exhaustively extracted with chloro- form. The dried extract was treated with 2 N HCl to remove alkaloids and the undissolved portion was subjected to column chromatography over silica gel, eluted with petrol, chloroform and methanol in order of increasing polarities. The constituents have been characterised mainly from their spectral data. Octacos-15,20-dien-11-ol (1) i The residue from the petrol eluate on crystallisation from petrol yielded colourless crystals (65 mg), m.p. 87-89°; l.lmaa (KBr) 3 450, 2 920, 2 840, 2 340, 1 475, 1 465. 1 185 and 1 175 cm- 1 ; " (60 MHz ; CDCI 3 ) 0.88 (6H, br s, 2xCH 8), 1.24 (36H, br s, 18 xCH 2), 3.25 (1H, m, CROH), 4.93 (4H, m, 4xCH=C); m/z (rei. int.) 406 [M+] (C28HHO) (3.1), 388 (3.2), 349 (9.5), 335 (0.3), 321 (0.4), 307 (3.1, ion a), 292 (4.2), 278 (1.5), 283 (3.3, ion c), 264 (4.3, ion /), 240 (2.1, ion e), 235 (3.8, ion j), 220 (3.9), 213 (1.6, ion g), 206 (3.2), 192 (3.1, ion k), 138 (4.0), 124 (0.3, ion d), 98 (3.1, ion b), 84 (6.1), 70 (8.5), 57 (4.7), 43 (33.2), 29 (23.2), 18 (100) and 15 (20.5). Acetylation (Ac20-Py) of 1 afforded a monoacety1 derivative, m.p. 71-73°; llmaz (KBr) 1 725 cm-1 • Trlacont-15,19,23-trien-13-ol (2) : The residue from petrol-CHC13 (9 : 1) eluate on crystallisation from CHC18 - MeOH (1 : 1) afforded white crystals (80 mg), m.p. 92-93•; l.lmu (KBr) 3 580, 2 900, 253 J. INDiAN CHBM, Soc., VOL, 68, APRIL 1991 {3} 2xCH8), 1.10 (3H, s, CH8), 1.16 (3H, s, CH8) an 3.26 (1H, dd, J 3.5 and 4.5 Hz, 3-fJH) ; m/z (re int.) 426 [M+] {C80H 11o0) (72.7), 411 (7.8), 408 (5.2 218 (58.9), 207 (38.1), 205 (49.3), 203 (53.5), 18 (58.4) and 55 (100). Acetylation (ActO-Py) of gave a monoacetyl derivative, m.p. 214-26° ; 11 (K.Br) 1 725 cm- 1 • References 1. K. K. BBU'l'ANI, M. AI.l and C. Ko A'tAX., Pllyl J. INDiAN CHBM, Soc., VOL, 68, APRIL 1991 J. Non.alkaloidal Components of Tylopkora hiuuta M.ALI Faculty of Pharmacy, ]amia Hamdard (Hamdard University), Hamdard Nagar, New Delhi-110 062 Manuscript received 13 August 1990, revisetZ 5 April7997, acceptetZ 24 April 7997 INDiAN CHBM, Soc., VOL, 68, APRIL 1991 {3} 2xCH8), 1.10 (3H, s, CH8), 1.16 (3H, s, CH8) and 3.26 (1H, dd, J 3.5 and 4.5 Hz, 3-fJH) ; m/z (rei. int.) 426 [M+] {C80H 11o0) (72.7), 411 (7.8), 408 (5.2), 218 (58.9), 207 (38.1), 205 (49.3), 203 (53.5), 189 (58.4) and 55 (100). Acetylation (ActO-Py) of3 gave a monoacetyl derivative, m.p. 214-26° ; 11,... (K.Br) 1 725 cm- 1 • References 3. A. SARKAR and S. N. GANGUI.Y, Indian J, Ch6tll·• Sect. B, 1978. 16, 742. 1. K. K. BBU'l'ANI, M. AI.l and C. Ko A'tAX., Pllylo• chemistf'y, 1984, 23. 1765 ; 1985, 24, 2778; M. ALl and K. K. BBU'l'ANI, Phytochemistf'y, 1987, 26, 2089. 2. "Annual Report'', Regional Research LaboratorY (C. S. I. R.), Jammu Tawi. 19S7; K. K. BHU'rA:Nl• G. I,. SHARMA and M. Ar..x, Planta Med .• 1987,53. 532. 3. A. SARKAR and S. N. GANGUI.Y, Indian J, Ch6tll·• S t B 1978 16 742 , y y, , , 2. "Annual Report'', Regional Research LaboratorY (C. S. I. R.), Jammu Tawi. 19S7; K. K. BHU'rA:Nl• G. I,. SHARMA and M. Ar..x, Planta Med .• 1987,53. 532. J References {3} 1. K. K. BBU'l'ANI, M. AI.l and C. Ko A'tAX., Pllylo• chemistf'y, 1984, 23. 1765 ; 1985, 24, 2778; M. ALl and K. K. BBU'l'ANI, Phytochemistf'y, 1987, 26, 2089. 1. K. K. BBU'l'ANI, M. AI.l and C. Ko A'tAX., Pllylo• chemistf'y, 1984, 23. 1765 ; 1985, 24, 2778; M. ALl and K. K. BBU'l'ANI, Phytochemistf'y, 1987, 26, 2089. 2 820, 1 615, 1 355 and 785 em-• ; a (60 MHz, CDC18) 0.84 (6H, s, 2 x CH 8 ), 1.26 (40H, br s, 20 xCH2), 3.10 (lH, m, CHOH) and 4.50 (6H, m, CH=CH) ; m/z (ret. int.) 432 [M+) (C80H 1160) (4.0), 404 (3.1}, 389 (2.8), 347 (0.1, ion a), 331 (10.6, ion n), 307 (0.1, ion c), 303 {10.3), 288 (9.9), 279 (0.6, ion e), 274 (3.1), 260 (5.5), 253 (0.5, ion g), 246 (2.9), 239 (0.5, ion i), 232 (4.6), 218 (3.4, ion 1), 213 (0.6, ion k), 192 (3.1, ion j), 178 {2.3), 164 (8.2, ion h), 138 {1.2, ion f), 124 (3.2), 110 (8.2, ion d), 101 (1.1, ion m), 85 (1.3, ion b), 71 {1.7), 57 (3.1), 43 (3.9) and 29 (100). Acetylation (Ac 20-Py) of 2 yielded a monoacetyl derivative, m.p. 78-79°; Vmoz (KBr) 1 720 em-•. Tritriacontane : Petrol-CHCla (3 : 1) eluates gave the compound (120 mg), m.p 69 -71°; vm.,. {KBr) 3 000, 2 910, 1 445, 1 355 and 1 135 em-• ; 8 (60 MHz; CDC18) 0.83 (6H, s, 2xCH8) and 1.33 (62H, br s, 31 xCH2) ; mfz (rei. int.) 464 [M+] (Ca aH6e) (0.3). Tritracont-1-olt Petrol-CHC18 {1 : 1) eluates afforded the compound (150 mg), m.p. 98-100°; Vmas: (KBr) 3 590, 2 900, 1 440, 1 345 and 1 120 cm- 1 ; 4 (60 MHz; CDC18) 0.84 (3H, s, CH3), 1.36 {62H, br s, 31 xCH2), 3.20 (2H, br s, CH90H), m/z (rat. int.) 480 [M+] (CaaH680) (1.3). Acetylation (AcsO-Py) of 4 afforded a monoacetyl derivative, m.p. 81-82° ; Vmas (KBr) 1 725 cm-1 • Gymnorhizol (3): Fractions 63-65, eluted with C-!fCla gave the compound (65 mg), m.p. 208-10° {bt3 • m.p. 208-09°); Vmaz(KBr) 3560,2910, 1595, 1 465 and 1 360 em·• ; 4 (60 MHz : CDCI8 ) 0.80 (6H, s, 2xCH8), 0.96 (3H, s, CH1), 1.03 (6H, s,

Subsets and Splits

No community queries yet

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