Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis

ARTICLE IN PRESS Reproductive BioMedicine Online (2016) ■■, ■■–■■

w w w. s c i e n c e d i r e c t . c o m w w w. r b m o n l i n e . c o m

ARTICLE

Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis Qi Zhang a,1, Xishi Liu a,b,1, Sun-Wei Guo a,b,* a

Shanghai OB/GYN Hospital, Fudan University, Shanghai 200011, China; b Shanghai Key Laboratory of Female Reproductive Endocrine-Related Diseases, Fudan University, Shanghai, China * Corresponding author.

E-mail address: [email protected] (S-W Guo). 1 These two authors contributed equally to this work. Sun-Wei Guo received his PhD from the University of Washington. He was a research scientist at the University of Michigan, Associate Professor at the University of Minnesota and tenured full Professor at the Medical College of Wisconsin. He served a directorship at the Institute of OB/GYN Research, Shanghai Jiao Tong University School of Medicine. Since 2010, he has been Professor at Shanghai OB/GYN Hospital, and the Department of Biochemistry and Molecular Biology, Fudan University Shanghai College of Medicine. He is an Adjunct Professor at the Department of OB/GYN and Reproductive Sciences, Michigan State University College of Human Medicine, Michigan, USA.

We have recently shown that platelets drive smooth muscle metaplasia (SMM) and fibrogenesis in endometriosis through epithelial-mesenchymal transition (EMT) and fibroblast-to-myofibroblast transdifferentiation (FMT). To see whether this is true in vivo, this prospective, randomized, and serially evaluated mouse investigation was conducted. Endometriosis was induced in female Balb/C mice, which were then randomly divided into two groups: Tanshinone IIA (TAN) and control (CTL) groups. TAN mice were treated with TAN but CTL mice received none. Every week until the 6th week after induction, five mice from each group were killed. Lesion weight was measured and lesion samples were subjected to immunohistochemistry and histochemistry analysis of platelet aggregation (CD41), E-cadherin, TGF-β1, phosphorylated Smad3, α-SMA, collagen I, CCN2, LOX, desmin and SM-MHC, and the extent of fibrosis was evaluated by Masson trichrome staining. It was found that endometriotic lesions exhibited progressive cellular changes consistent with the progressive EMT, FMT, SMM, and fibrogenesis. TAN treatment resulted in significant hindrance of EMT, FMT, SMM and fibrogenesis, and reduced lesion weight (all P-values <0.05). These data corroborate the notion that endometriotic lesions undergo progressive EMT and FMT, giving rise to SMM and ultimately fibrosis. This understanding sheds new light onto the natural history of endometriosis.

Abstract

© 2016 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: endometriosis, epithelial-mesenchymal transition, fibroblast-to-myofibroblast transdifferentiation, fibrosis, platelet, smooth muscle metaplasia

http://dx.doi.org/10.1016/j.rbmo.2016.11.006 1472-6483/© 2016 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 2

Q Zhang et al.

Introduction Characterized by the deposition and growth of endometriallike tissues outside the uterine cavity, endometriosis is an oestrogen-dependent disorder and a major contributor to pelvic pain and subfertility affecting 6–10% of women of reproductive age (Giudice and Kao, 2004). Despite exponential growth in the number of publications on endometriosis in the last four decades (Guo, 2014), its pathophysiology is incompletely understood (Giudice and Kao, 2004). As a result, the development of effective, targeted therapy or preventative measures for this debilitating disease, especially the development of non-hormonal drugs, has been painfully slow (Guo, 2014), over which there is a palpable disappointment (Vercellini et al., 2011). One seemingly insurmountable roadblock to the elucidation of the pathophysiology of endometriosis is our rudimentary knowledge of the natural history of endometriosis, even though the consensus is that over time endometriosis is a progressive and dynamic disease in spontaneous and induced endometriosis (D’Hooghe et al., 1996a, 1996b; Harirchian et al., 2012). Despite thousands of published studies documenting various histologic, cellular and molecular aberrations in endometriotic lesions, the lesions appear to refuse to part with their secrets of natural history. Consequently, the currently widely used rAFS/rASRM staging system does not correlate well with either the severity of symptoms, progression or prognosis (Koninckx et al., 2011). It is also difficult to piece together most, if not all, published studies. Taking the cue from the observation that ectopic endometrium experiences cyclic and thus repeated bleeding (Brosens, 1997), a cardinal sign of tissue injury, we have recently shown that platelets play important roles in the development of endometriosis (Ding et al., 2015; Guo et al., 2015a, 2015b, 2015c; Zhang et al., 2015) and proposed that endometriotic lesions are essentially wounds that undergo repeated tissue injury and repair (ReTIAR) (Ding et al., 2015; Guo et al., 2015a, 2015b, 2015c). Our recent investigation indicates that, as the result of this ReTIAR, the endometriotic lesions, stimulated by platelet-derived transforming growth factor-β1 (TGF-β1), activate the TGF-β1/Smad3 signalling pathway and undergo epithelial-mesenchymal transition (EMT) and fibroblast-to-myofibroblast transdifferentiation (FMT), resulting in increased cellular contractility and collagen production, leading ultimately to fibrosis (Zhang et al., 2016a, 2016b). Prolonged exposure to activated platelets also leads to increased expression of α-smooth muscle actin (α-SMA) as well as markers of differentiated smooth muscle cells (SMC) in endometriotic stromal cells, which may be responsible for what is termed smooth muscle metaplasia (SMM) that is nearly universally seen in endometriotic lesions (Itoga et al., 2003; Khare et al., 1996; Matsuzaki and Darcha, 2013; Mechsner et al., 2005). In light of these findings, we postulated that during the progression of endometriosis, we should see indications of the activated TGF-β1/Smad3 signalling pathway, progressive EMT and FMT, as shown by increased vimentin expression but decreased E-cadherin expression in the epithelial component of the endometriotic lesions. In addition, we should see increased expression of α-SMA, a marker for myofibroblasts and SMC (Hasegawa et al., 2003; Hinz et al., 2001), in the stromal

component of the lesions. Concomitant with the increased α-SMA expression and other markers of SMC, we should see progressive SMM as well as increased fibrotic tissue content in endometriotic lesions. Our recent serial evaluation of endometriotic tissue samples harvested from time points after induction of endometriosis in female baboons are consistent with our hypothesis (Zhang et al., 2016a, 2016b). In this study, this hypothesis was further tested in a mouse model of endometriosis. Our goal was two-fold. First, to test this hypothesis using serially harvested endometriotic tissue samples and immunohistochemistry (IHC) and histochemistry analyses. Second, to test the hypothesis that anti-platelet treatment should be effective in impeding the EMT, FMT, SMM and fibrogenesis.

Materials and methods Mouse experiment protocol Ninety virgin female Balb/C mice, 6 weeks old and about 16– 18 g in bodyweight, were purchased from Shanghai BiKai Laboratory Animal Centre (Shanghai, China) and used for this study. They were maintained under controlled conditions with a light/ dark cycle of 12/12 h, and had access to chows and water ad libitum. All experiments were performed under the guidelines of the National Research Council’s Guide for the Care and Use of Laboratory Animals (National Research Council, 1996) and approved by the institutional experimental animals review board of Shanghai OB/GYN Hospital, Fudan University on 14 May 2014. Among the 90 mice, 30 were randomly selected as donors that contributed endometrial tissue fragments, and the remaining 60 were recipients who received endometrial tissues from donor mice. After 2 weeks of acclimatization and before the endometriosis-inducing procedure (see below), all recipient mice were administrated a baseline hotplate test as reported previously (Ding et al., 2015). The donor mice were initially treated i.m. with oestradiol benzoate (0.2 μg/g bodyweight, Xinyi Chemistry, Shanghai, China) after 1 week of acclimation. One week later they were killed and their uteri were removed and harvested as in (Bacci et al., 2009). The uterine tissues were seeded in a Petri dish containing warm sterile saline, and split longitudinally with a pair of scissors, as in (Somigliana et al., 1999, 2001). The maximal diameter of the processed fragments was consistently smaller than 1 mm. Approximately 40–50 fragments per mouse were then injected i.p. into recipient mice. The recipient mice were divided randomly into two groups in equal size: Tanshinone IIA (TAN) and control (CTL) group. Mice in the TAN group were injected i.p. with TAN 12.5 μg/g dissolved in 300 μl sterile saline (shielded from light) every other day (i.e. days 2, 4, 6, . . ., 38 and 40), starting from 2 days after the induction of endometriosis (day 0, the day when induction was performed). Mice in the CTL group received no treatment at all. This ‘intact’ mode was chosen mainly because the study aimed to see the natural development history of endometriosis at the price of getting more conservative treatment effect due to the promotional effect of stress resulting from repeated injections (Cuevas et al., 2012; Long et al., 2016). To minimize possible individual variation, endometrial tissue

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis fragments obtained from one mouse were injected i.p. to two mice, one each from CTL and TAN groups. TAN is a pharmacologically active diterpenoid extracted from the roots of Salvia milthiorriza Bunge, a plant used in traditional Chinese medicine as a remedy for ‘blood stasis’, which is a term used in traditional Chinese medicine standing for hypercoagulability. It has been demonstrated that TAN inhibits platelet aggregation (Maione et al., 2014) and suppresses TGF-β1/Smad3 and NF-κB signalling pathways and thus fibrogenesis (Jiang et al., 2015; Tang et al., 2015; Wang et al., 2015). Sodium tanshinone IIA sulfonate injection is widely used in China as a prescription drug for treating cardiovascular disease and is known to have an excellent safety profile. Every week until the 6th week after the induction (i.e. days 7, 14, 21, 28, 35 and 42), five mice each from the two groups were killed. Hotplate test, bodyweight measurement and blood samples collection (about 10 μl) from tail veins were performed to all mice before sacrifice. The blood sample was used for the quantification of the platelet activation rate by flow cytometry (see below). Then the abdominal cavity was immediately opened, examined and evaluated carefully. The endometriotic lesions were harvested and weighed (dry weight). After quantification, the lesion samples were immediately processed, as detailed below, for histological confirmation, and immunohistochemical analysis of CD41 (a marker for platelets), TGF-β1, phosphorylated Smad3 (pSmad3), E-cadherin, α-SMA, collagen I, lysyl oxidase (LOX), CCN2 (also called connective tissue growth factor, or CTGF), desmin (a marker for differentiated and mature SMC (Hasegawa et al., 2003)), and smooth muscle, myosin heavy chain (SM-MHC) (a marker restricted for the SMC (Manabe and Owens, 2001)), along with Masson trichrome staining to quantify the extent of fibrosis in lesions.

The hotplate test and lesion measurement The hotplate test was performed with a commercially available Hot Plate Analgesia Meter (Model BME-480, Institute of Biomedical Engineering, Chinese Academy of Medical Sciences, Tianjin, China) as reported previously (Lu et al., 2010). Since mouse is not vocal about its pain severity, and since women with endometriosis as well as rodents with induced endometriosis exhibit central sensitization (Bajaj et al., 2003; Berkley et al., 2007; He et al., 2010), the hotplate latency can be used as a surrogate measure for the severity of endometriosisrelated generalized hyperalgesia (Lu et al., 2010). The endometriotic lesions were excised and carefully weighed and then fixed for IHC and histochemistry analyses. The lesion dry weight was determined following (Bacci et al., 2009).

Fluorescence-activated cell sorting (FACS) assessment of platelet activation rate in peripheral blood Following Qureshi et al. (Qureshi et al., 2009), the platelet activation rate in the peripheral blood was assessed by FACS. About 10 μl of blood sample was mixed with EDTA-K2. Five micro litres of the whole blood was diluted with 45 μl 0.5%

3 paraformaldehyde (PFA), a fixation agent. The isotype control antibodies (eBioscience, San Diego, CA, USA) were added to the negative control tubes while fluorescein isothiocyanate (FITC) conjugated anti-mouse CD61 antibody (eBioscience) was used to label platelets and phycoerythrin (PE) conjugated antimouse CD62p (P-selectin) antibody (eBioscience) was used to label the activated platelets. Two antibodies were incubated together in test tubes at room temperature for 20 min. Four hundred and fifty micro litres 0.5% PFA was added to each tube, and then analysed by FACS (BD FACS Calibur, BD, Franklin Lakes, NJ, USA). After identifying platelets by gating on FITC positivity (CD61 positivity), P-selectin positivity was determined by analysing phycoerythrin fluorescence. In all cases, the quantification was made under the identical experimental protocol and condition and analysed under the fixed flowcytometry cell sorting voltage. One representative FACS result for assessment of the platelet activation rate is shown in Supplementary Figure S1.

Immunohistochemistry (IHC) Endometriotic tissue samples from all mice were fixed with 10% formalin (w/v) and paraffin-embedded. Serial 4-μm sections were obtained from each block, with the first resultant slide being stained for haematoxylin and eosin (H and E) to confirm pathologic diagnosis, and the subsequent slides for IHC analysis for CD41 (1:100, Abcam, Cambridge, UK), E-cadherin (1:400, Cell Signaling Technology, Boston, MA, USA), α-SMA (1:100, Abcam), collagen I (1:100, Abcam), CCN2 (CTGF) (1:400, Abcam), LOX (1:100, Abcam), TGF-β1 (1:100, Abcam), p-Smad3 (1:100, Abcam), desmin (1:200, Abcam) and SMMHC (1:100, Abcam). Routine deparaffinization and rehydration procedures were performed. For negative controls, the immunoglobulin G from rabbit serum (Sigma, Darmstadt, Germany) in lieu of the primary antibodies (but in the same concentration as that of primary antibodies) was used. For positive controls, human ovarian endometrioma tissues were used for α-SMA, collagen I, CCN2 (CTGF), LOX, TGF-β1 and p-Smad3 staining, while the deep infiltrating endometriosis (DIE) tissue samples were used for desmin and SM-MHC staining (Supplementary Figure S2). All human tissues samples were obtained after written informed consent from patients with laparoscopically and histologically diagnosed ovarian endometrioma or DIE from premenopausal women with laparoscopically and histologically diagnosed ovarian endometrioma or DIE in our hospital, who received no hormonal or anti-platelet treatment at least 3 months prior to the surgery. Routine deparaffinization and rehydration procedures were performed, as reported previously (Ding et al., 2015). For antigen retrieval, the slides were heated at 98°C in a citrate buffer (pH 6.0) for a total of 30 min for staining for CD41, E-cadherin, p-Smad3, α-SMA, collagen I, CCN2 (CTGF), desmin and SM-MHC or an EDTA buffer (pH 8.0, Shanghai Sun BioTech Company, Shanghai, China) for a total of 20 min for staining for LOX and TGF-β1, and cooled naturally to the room temperature. The slides were incubated with goat blocking serum for 15 min at room temperature and then incubated with the primary antibodies overnight at 4°C. The primary antibodies were diluted by phosphate-buffered saline. After slides were rinsed, the horseradish peroxidase (HRP) labelled secondary antibody Detection Reagent (Sunpoly-HII, BioSun

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 4

Q Zhang et al.

Figure 1 The kinetic changes of platelet activation rate in peripheral blood samples (A), lesion weight (B), and hotplate latency (C) in mice treated with Tanshinone IIA (blue lines) and without (red lines). The data are shown in means and standard deviations. *: P < 0.05; NS = not statistically significant (by Wilcoxon rank test). n = 4–5 at any time point for each group.

Technology Co., Ltd, Shanghai, China) was incubated at room temperature for 30 min. The bound antibody complexes were stained for 3–5 min or until appropriate for microscopic examination with diaminobenzidine and then counterstained with haematoxylin (30 s) and mounted. All lesion samples were evaluated by H and E staining using a H and E staining kit (Sun Biotec, Shanghai, China). Images were obtained with the microscope (Olympus BX53, Olympus, Tokyo, Japan) fitted with a digital camera (Olympus DP73, Olympus). Three to five randomly selected images at 400X magnification of each sample were taken to obtain a mean optional density value by Image Pro-Plus 6.0 (Media Cybernetics, Inc., Bethesda, MD, USA).

Masson trichrome staining Masson trichrome staining was used for the detection of collagen fibres in endometriotic tissue samples. Tissue sections were deparaffinized in xylene and rehydrated in a graded alcohol series, then were immersed in Bouin’s solution at 37°C for 2 h, which was made with saturated picric acid 75 ml, 10% (w/v) formalin solution 25 ml and acetic acid 5 ml. Sections were stained using the Masson’s Trichrome Staining kit (Baso, Wuhan, China) following the manufacturer’s instructions. The areas of the collagen fibre layer stained in blue in proportion to the entire field of the ectopic endometrium were calculated by the Image Pro-Plus 6.0.

Statistical analysis The comparison of distributions of continuous variables between the two groups was made using Wilcoxon’s test, and the paired Wilcoxon test, whenever appropriate, was used when the beforeafter comparison was made for the same group of subjects. Pearson’s or Spearman’s rank correlation coefficient was used when

evaluating correlations between two variables when both variables were continuous or when at least one variable was ordinal. To account for the effect of more than one factor, multiple linear regression analysis was used. To gain more insights into the possible mechanisms underlying the progression of endometriosis, a hierarchical cluster analysis was carried out with scaled data for CTL mice and the Euclidean distance as the similarity metric, with the average linkage being the clustering method. The resulting dendrogram was represented as a heatmap, which is essentially a graphical representation of numerical data by different colourations. A multidimensional scaling (MDS) analysis was performed to discriminate all CTL mice without missing data. P-values of <0.05 were considered statistically significant. All computations were made with R 3.2.4 (R Development Core Team, 2013) (www.r-project.org).

Results No mouse died during the entire course of the experiment, and there was no difference in bodyweight between the two groups (Supplementary Figure S3) and no visible internal haemorrhage, or other adverse effect. However, four mice, one each in the 3-, 5- and 6-week TAN group, and one in the 4-week CTL group, had hotplate latencies over 40 s for unknown reasons. To be conservative, these mice were excluded in all the analyses. As expected, the platelet activation rate in the peripheral blood from mice in the TAN group was consistently lower than the CTL group, but the difference reached statistical significance only at 1 and 5 weeks after induction (P < 0.05; Figure 1A). However, a multiple linear regression analysis incorporating the week at which blood samples were taken and group identity (TAN versus CTL) indicated that the platelet activation rate in the TAN group was significantly lower than

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis the controls (log-transformed, P < 0.001; R2 = 0.27). Thus, the lack of statistical significance at some time points may have been attributable to the small sample sizes. In CTL mice, the lesion weight increased progressively as they aged, especially 5 and 6 weeks after induction (Figure 1B). While point-wise comparison indicated that TAN treatment resulted in significant reduction in lesion weight only at 5 and 6 weeks after induction (P < 0.05; Figure 1B), a multiple linear regression analysis incorporating the week at which tissue samples were harvested, bodyweight, the group identity and their interaction indicated that the lesion weight increased progressively (P = 0.008), and that mice in the TAN group had significantly smaller lesions than the control group (P < 0.05; R2 = 0.52). The lack of statistical significance at some time points may have been due to the lack of statistical power. Similarly, a multiple linear regression analysis incorporating the bodyweight, baseline latency, week at which the hotplate test was administrated, the group identity and their interaction indicated that the hotplate latency decreased progressively as mice aged (P < 0.001; Figure 1C), that the baseline latency was positively associated with the final latency (P < 0.05) and that mice in the TAN group had significantly longer latency than the control group (P < 0.001; R2 = 0.55). These results are consistent with results previously reported (Ding et al., 2015; Guo et al., 2015a, 2015b, 2015c). Iimmunohistochemistry analysis was also conducted of CD41 (for platelet aggregation), TGF-β1, p-Smad3, E-cadherin, α-SMA, collagen I, CCN2 (CTGF) and LOX in endometriotic tissue samples. It was found that platelets aggregated mostly in the stromal component of endometriotic lesions, and some could be found around epithelial cells. Platelet aggregation was increased as endometriosis progressed but treatment with TAN resulted in significantly decreased platelet aggregation (Figures 2 and 3). E-cadherin staining was seen mostly in cytoplasm and membranes in endometriotic epithelial cells but its immunoreactivity was decreased as endometriosis progressed (Figure 2). Treatment with TAN resulted in significantly reduced platelet aggregation but significantly increased E-cadherin staining (Figures 2 and 4; both P-values <0.001 by regression analysis, R2 > 0.57), suggesting that a progressive EMT occurred as the disease progressed, but the treatment with TAN hindered this progression. TGF-β1 and LOX staining was seen mostly in endometriotic glandular epithelial cells and was localized in the cytoplasm. The p-Smad3 staining was seen in nuclei as well as cytoplasm in endometriotic epithelial cells. The nuclei staining of p-Smad3 appeared to be more prominent as the lesions aged. Collagen I and α-SMA staining was seen primarily in the endometriotic stromal component. In contrast, CCN2 staining was mainly in cytoplasm and extracellular matrix (ECM) in endometriotic glandular epithelial as well as stromal cells (Figures 2 and 3). The immunoreactivity against CD41, TGF-β1, p-Smad3, CCN2, α-SMA, collagen I and LOX was all elevated as endometriosis progressed, consistent with the progressive FMT in the development of endometriosis if unimpeded. Consistent with the gradual but progressive EMT and FMT, the extent of fibrosis increased progressively, starting from 2 weeks after the induction (Figure 4). Remarkably, multiple linear regression analyses indicated that the older the lesion was, the lower E-cadherin staining (P < 0.001) and, with the only exception of CCN2, the

5 higher staining levels of CD41, TGF-β1, p-Smad3, α-SMA, collagen I and LOX were (all P-values <0.01, all R2 > 0.44; Figure 4). In addition, TAN treatment significantly elevated the E-cadherin staining while significantly reduced the immunoreactivity against CD41, TGF-β1, p-Smad3, CCN2, α-SMA, collagen I and LOX, concomitant with reduced fibrosis (all P-values <0.001; Figure 4A–H). Remarkably, the extent of platelet aggregation in ectopic endometrium was correlated with all the IHC measurements (Supplementary Figure S4). IHC analysis was also performed for desmin and SM-MHC, two markers of highly differentiated smooth muscle cells. Desmin and SM-MHC staining was seen mostly in cytoplasm of the stromal component in ectopic implants. Consistent with the gradual but progressive EMT and FMT as shown in invitro studies (Zhang et al., 2016a, 2016b), it was found that the extent of desmin and SM-MHC-positive SMC and therefore the extent of SMM increased progressively, starting from 3 weeks after the induction (Figures 4 and 5). Treatment with TAN resulted in significantly reduced staining of both desmin and SM-MHC (both P-values <0.01 by linear regression, R2 > 0.55; Figures 4I, J and 5), suggesting that the treatment with TAN significantly impeded the progression of SMM. While point-wise comparison indicated that TAN treatment resulted in significant reduction in desmin and SMMHC staining compared with the controls only at 3 and 5 weeks after induction (all P < 0.05; Figures 4I, J and 5), a multiple linear regression analysis incorporating the week at which tissue samples were harvested and the group identity indicated that the both desmin and SM-MHC staining levels increased progressively in the controls (P < 0.01 and P < 0.01, respectively) but decreased significantly in mice treated with TAN (P < 0.001 and P < 0.001, and R2 = 0.55 and R2 = 0.46, respectively). Desmin and SM-MHC staining levels were positively correlated (r = 0.89, P < 0.001). Remarkably, both staining levels were significantly positively correlated with markers of EMT (E-cadherin and vimentin) and FMT (α-SMA, CCN2, LOX and collagen I) and the extent of fibrosis (r ranged from 0.50, P < 0.01, to 0.72, P < 0.001) but significantly negatively correlated with that of E-cadherin (r = −0.72, P < 0.001, and r = −0.67, P < 0.001, respectively). These two markers of SMM (i.e. desmin and SM-MHC) were also correlated with the extent of fibrosis in ectopic endometrium (Supplementary Figure S5). Finally, the extent of fibrosis in endometriotic lesions were evaluated by Masson trichrome staining (Figure 5). By multiple linear regression, it was found that the extent of fibrosis in ectopic endometrium in control mice also increased progressively, starting from week 2 after induction (P < 0.001; Figure 4K; the dependent variable was arc sin transformed). Treatment with TAN, however, significantly attenuated the increase in fibrotic content (P < 0.001, R2 = 0.83). Not surprisingly, the extent of fibrosis was correlated with markers of myofibroblast activation (α-SMA and CCN2), and the extent of ECM deposits (collagen I and LOX) (Supplementary Figure S5). To gain more insights into the possible mechanisms underlying the progression of endometriosis and also the TAN treatment effect, a hierarchical cluster analysis of some select IHC immunostaining measurements (and the platelet activation rate) were performed, all square-root transformed, and the results are presented as a heatmap (Figure 6A). We can see from Figure 6A that all staining measurements were grouped into 4 clusters: cluster 1 consisted of TGF-β1,

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 6

Q Zhang et al.

Figure 2 Representative immunostaining staining results in ectopic lesions in mice treated with and without Tanshinone IIA (TAN) during the progression of endometriosis (Part I). TAN: mice treated with TAN; control (CTL): mice without treatment of TAN. Different columns represent tissue samples harvested at different time points (after induction of endometriosis). The level of CD41 staining reflected the aggregation of platelets, and mostly seen in cytoplasmic components in stromal cells. E-cadherin staining was seen mainly in cytoplasmic and membranous components in epithelial cells. Transforming growth factor-β1 (TGF-β1) staining was seen primarily in glandular epithelial cells and was localized in the cytoplasm. The phosphorylated Smad3 (p-Smad3) staining was seen in nuclei as well as cytoplasm in endometriotic epithelial cells. The p-Smad3 staining results for week 1 were unavailable since the tissue sample sections were run out. Scale bar = 50 μm.

collagen I and the extent of fibrosis; cluster 2 consisted of α-SMA and the extent of platelet aggregation; cluster 3, LOX and CCN2; and cluster 4, E-cadherin and the platelet activation rate (Figure 6A). Corresponding to the clustering of staining measurements, all mice could be grouped roughly into 2 clusters: cluster I (top 12 rows) is featured by a high fibrotic content and low E-cadherin expression in ectopic endometrium; and cluster II (the remaining rows), a relatively low fibrotic content and high E-cadherin expression. Remarkably, cluster I includes all mice that were killed at 5 and 6 weeks after endometriosis induction and one each that was killed at 3 and 4 weeks. Cluster II includes all tissue samples harvested at 1 as well as 2 weeks after induction and

four and three samples that were harvested at 3 and 4 weeks after induction, respectively. In other words, the clustering essentially grouped all mice into early and late stages of endometriosis. Consequently, this study attempted to classify all endometriotic tissue samples from CTL mice using immunostaining or histochemistry marker, i.e. E-cadherin, α-SMA or the extent of fibrosis, and the results are shown in Figure 6B. We can see that, based on simply the staining levels of E-cadherin, α-SMA or the extent of fibrosis, all tissue samples can be ranked roughly by the ‘age’ of the lesions (Figure 6B). In other words, just a few select lesional immunohistochemistry and histochemistry markers representing

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis

7

Figure 3 Representative immunostaining staining results in ectopic lesions in mice treated with and without Tanshinone IIA (TAN) during the progression of endometriosis (Part II). TAN: mice treated with TAN; control (CTL): mice without treatment of TAN. Different columns represent tissue samples harvested at different time points (after induction of endometriosis). Collagen I and α-smooth muscle actin (α-SMA) staining was seen mostly in the stromal component of the ectopic lesions. Connective tissue growth factor (CCN2 or CTGF) staining was seen in cytoplasmic and extracellular matrix components in both glandular epithelial and stromal cells. Lysyl oxidase (LOX) staining was seen primarily in glandular epithelial cells and was localized in the cytoplasm. Scale bar = 50 μm.

different landmarks in endometriosis development chronicled all lesion samples quite well.

Discussion In this study, it has been shown that, consistent with our hypothesis, endometriotic lesions seemingly undergo progressive platelet aggregation, and progressive EMT, FMT and SMM, leading ultimately to fibrosis in a mouse model of endometriosis. In addition, anti-platelet treatment resulted in decreased platelet activation and TGF-β1/p-Smad3 expression indicative of attenuated TGF-β/Smad activation, and reduced lesion size as well as the extent of EMT, FMT, SMM and of fibrosis. These data provide a strong piece of corroborative evi-

dence that endometriotic lesions and their microenvironment have all the necessary molecular machinery that gives rise to SMM and promotes fibrogenesis (Zhang et al., 2016a, 2016b). The biggest strength of this study is the serial observation of cellular changes in ectopic endometrium. This, coupled with the IHC and histochemistry analyses of carefully selected markers based on our ReTIAR hypothesis (Zhang et al., 2016a, 2016b), helps to illuminate the natural history of endometriosis, which so far has been elusive. This study also has several limitations. Firstly, the sample sizes at each time point are rather moderate, sometimes resulting in inadequate statistical power. This is further compounded by missing data in some measurements due to the lack of enough tissue sections. This problem is partially remedied by the use of linear regression modelling

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 8

Q Zhang et al.

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis

9

Figure 4 Changes in platelet aggregation shown by CD41 staining (A), Transforming growth factor-β1 (TGF-β1) staining (B), phosphorylated Smad3 (p-Smad3) staining (C), E-cadherin staining (D), CCN2 staining (E), α-smooth muscle actin (α-SMA) staining (F), collagen I staining (G), lysyl oxidase (LOX) staining (H), desmin staining (I), and smooth muscle, myosin heavy chain (SM-MHC) staining (J), and the proportion of fibrotic content (K) in ectopic endometrium from treated with Tanshinone IIA (blue line) and without (red line). The data are shown in means and standard deviations. *: P < 0.05; **: P < 0.01; NS = not statistically significant (Wilcoxon rank test). n = 4–5 at any time point for each group.

Figure 5 Representative immunostaining of desmin and SM-MHC and Masson trichrome staining in ectopic lesions in mice treated with and without Tanshinone IIA (TAN) during the progression of endometriosis. TAN: mice treated with TAN; control (CTL): mice without treatment of TAN. Different columns represent tissue samples harvested at different time points (after induction of endometriosis). For desmin and SM-MHC, staining of ectopic endometrium harvested 6 weeks after induction was not performed since all sections were exhausted. Desmin and SM-MHC staining was seen mostly in the stromal component of the ectopic lesions. In Masson trichrome staining, the collagen fibres in ectopic endometrium were stained in blue. Scale bar = 50 μm. SM-MHC = smooth muscle, myosin heavy chain.

incorporating all observations, but larger sample sizes should provide more precise estimates. While the study may be somewhat underpowered to detect difference at some time points between the TAN treated and untreated mice, the difference and the time-dependent changes that it did detect are more likely to be genuine due to our highly structured hypothesis. Secondly, while the use of immunohistochemistry analysis provides a convenient way to ascertain the cell type, location and expression levels of select markers, it cannot determine whether the protein is expressed or secreted by other neighbouring cells. Thirdly, the administration of TAN started just one day after the induction of endometriosis, and lasted for the entire course of the experiment until the mice were

killed. Hence the ‘treatment effect’ of TAN should be viewed with caution, since in actuality, the medical treatment does not start shortly after the ectopic endometrium is taking root, nor does the medical treatment last forever. Therefore, the ‘treatment effect’ of TAN is really meant to demonstrate the effect of anti-platelet intervention shortly after the ectopic endometrium starts to grow after successful penetration into the peritoneum. Fourthly, while the mice in the TAN group received i.p. injection of TAN every other day until they were killed, mice in the CTL group were intact, receiving no treatment whatsoever. Thus, the CTL is not a perfect control group for assessing the treatment efficacy. However, since i.p. injection would inevitably elicit pain sensation, this, in

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 10

conjunction with the extra handling due to drug administration, could result in stress of some sort, which, as in surgical stress, could facilitate the development of endometriosis that is already in existence (Long et al., 2016). In other words, while the endometriosis in CTL mice represents the natural development of endometriosis in mouse, the treatment effect, as compared with the CTL group, may well be conservative. Lastly, the mouse model that was used may not fully repre-

Q Zhang et al.

sent human endometriosis, and as such, extrapolation to humans may exercise caution. However, the results are not only consistent with the in-vitro data (Zhang et al., 2016a, 2016b) but also essentially identical to that for baboon models of endometriosis (Zhang et al., 2016a, 2016b). Consequently, we believe that these results are credible. Consistent with the in-vitro finding that prolonged exposure to activated platelets results in activation of

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis

11

Figure 6 (A) Hierarchical clustering heatmap of immunoreactivity measurements and mice in the control group. The heatmap was organized by clustering both mice (by rows) and immunoreacivity measurements (by columns). The red colour represents the minimal values while the bright yellow colour represents the maximal values. The white colour strips represent missing values. The numeric on the right-hand side represents which time points (weeks after the induction of endometriosis) when the tissue samples were harvested. The labels in the bottom are the names of the immunoreactivity measurements. The colour strip on the left: The pink colour of increasing darkness indicates tissue samples taken from mice with more advanced endometriosis. The numeric on the rightmost panel represents which time points (weeks after the induction of endometriosis, i.e. age of lesions) when the tissue samples were harvested. PLT Ag: The extent of platelet aggregation; % aPLT: The platelet activation rate in the peripheral blood sample. (B) Results of multidimensional scaling analysis using the E-cadherin and α-smooth muscle actin (α-SMA) immunostaining levels and the extent of fibrosis in ectopic endometrium. Each ellipse indicates one group. The numbers within the ellipses indicate which time points (i.e. weeks after the induction of endometriosis) when the tissue samples were harvested, that is, the age of the lesion tissues. Each number represents a tissue sample from one mouse. Different colours of the tissue samples are used to highlight the age of the lesions, which are consistent with the colours used in panel (A).

SMC-specific markers in endometriotic stromal cells (Zhang et al., 2016a, 2016b), the expression of desmin and SM-MHC in the endometriotic stromal compartment became prominent starting from 4 weeks after the induction of endometriosis (Figure 4), and their expression levels were positively correlated with the extent of fibrosis, as were the FMT markers α-SMA, CCN2, collagen I and LOX (Supplementary Figure S5). These results provide in-vivo evidence that platelets are the driver for SMM and fibrogenesis in endometriosis. As an immediate corollary, it is evident that endometriotic lesions interact closely with their microenvironment and change their morphology and function in a dynamic fashion. It could be argued that this view may succinctly capture the essence of the natural history of endometriosis. It should be noted that while non-human primate models of endometriosis have shown that over time endometriosis is a progressive and dynamic disease in spontaneous and induced endometriosis (D’Hooghe et al., 1996a, 1996b; Harirchian et al., 2012), the evidence was based on repeated serial laparoscopic examinations. However, repeated laparoscopy itself may promote the development of endometriosis (D’Hooghe et al., 1996a, 1996b), thus confounding the observation on the changes in endometriotic lesions. The mouse model used in this study rules out the possibility of possible promotional effect of repeated surgery (Long et al., 2016). While female mice do not menstruate as in humans and thus there is no apparent cyclic bleeding and then subsequent tissue repair, endometriotic stromal cells nonetheless secrete platelet-activating factors such as thrombin and thromboxane A2 (Guo et al., 2016). This, coupled with increased angiogenesis in endometriosis due to increased expression of genes such as VEGF (Donnez et al., 1998) and thus increased vascular permeability, would result in platelet extravasation and aggregation in endometriosis. Results from this study have several potentially important clinical as well as biological implications. First, the data clearly show that anti-platelet therapy may be promising in treating endometriosis, which has been shown previously using other means of intervention (Ding et al., 2015; Guo et al., 2015a, 2015b, 2015c). Second, our data clearly show gradual but progressive development of endometriotic lesions, not only their sizes become larger but also the extent of fibrosis is increased. Many preclinical studies, including our own, give treatment to rodents with induced endometriosis when endometriosis is fully established, say, 2 weeks after induction (Guo et al., 2015a, 2015b, 2015c). However, as seen from Figure 1B, the lesions

continue to grow after week 2 and reach a highly fibrotic state at week 6 (Figure 4E, F, J). This suggests that the chosen time interval between induction and the start of the treatment may be too short, and different time points for starting intervention may yield entirely different conclusions. Given the welldocumented diagnostic delay of about 4–12 years from the onset of first endometriosis-related symptoms to the definitive diagnosis (Arruda et al., 2003; Dmowski et al., 1997; Hadfield et al., 1996), it is likely that the endometriotic lesions in these patients may be already highly fibrotic and thus highly resistant to hormonal or non-hormonal treatment due to decreased vasculature and increased extracellular matrix deposits in interstitial spaces in lesions. In light of this, it is important to justify the time point at which the drug intervention is started in preclinical studies. Lastly, the data show that the natural history of endometriosis can be fully silhouetted or sketched using just a few select landmark markers. Given the inherent variation in IHC and histochemistry analyses as well as individual variation, such an achievement is remarkable. This suggests that endometriotic lesions contain inherent markers that can serve essentially as a ‘fossil record’ of the lesion, and may be used to date the lesions. In view of fact that the currently widely used rAFS/rASRM staging system (American Society for Reproductive Medicine, 1997) does not correlate well with either the severity of symptoms, progression or prognosis (Koninckx et al., 2011), results from this study justify a histological staging of endometriotic lesions based on a few, carefully chosen biomarkers. The results, in conjunction with the finding that a similar developmental process appears to occur in adenomyosis (Shen et al., 2016), suggest that such a histological classification of endometriosis progression is feasible. In fact, this has been validated in principle in ovarian endometriomas (Guo et al., 2015a, 2015b, 2015c). In particular, in view of the increased extent of fibrosis concomitant with decreased PR-B expression in ectopic endometrium (Shen et al., 2016), we believe that such a classification system may be potentially useful in predicting response to medical treatment or even recurrence. That said, future research is needed to precisely define the relationship, if any, between endometriosis progression and the symptomology of the disease and the recurrence risk. In conclusion, this study provides in-vivo evidence in support for the notion that, driven by activated platelets, endometriotic lesions undergo progressive EMT, FMT and SMM, leading ultimately to fibrosis. Anti-platelet treatment impedes this progression, resulting in reduced lesion size and the extent

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS 12 of fibrosis and improved generalized hyperalgesia. Through progressive EMT, FMT, SMM and ultimately increased fibrosis, the silhouette, if not the full view, of the natural history of endometriosis is now in plain sight. This understanding should greatly improve our understanding of the pathophysiology of endometriosis, and help design better therapeutics and biomarkers for endometriosis.

Acknowledgements This work was supported in part by grants 81270676 (SWG), 81471434 (SWG), 81530040 (SWG) and 81370695 (XSL) from the National Natural Science Foundation of China.

Appendix: Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.rbmo.2016.11.006.

References American Society for Reproductive Medicine, 1997. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil. Steril. 67, 817–821. Arruda, M.S., Petta, C.A., Abrao, M.S., Benetti-Pinto, C.L., 2003. Time elapsed from onset of symptoms to diagnosis of endometriosis in a cohort study of Brazilian women. Hum. Reprod. 18, 756–759. Bacci, M., Capobianco, A., Monno, A., Cottone, L., Di Puppo, F., Camisa, B., Mariani, M., Brignole, C., Ponzoni, M., Ferrari, S., Panina-Bordignon, P., Manfredi, A.A., Rovere-Querini, P., 2009. Macrophages are alternatively activated in patients with endometriosis and required for growth and vascularization of lesions in a mouse model of disease. Am. J. Pathol. 175, 547–556. Bajaj, P., Bajaj, P., Madsen, H., Arendt-Nielsen, L., 2003. Endometriosis is associated with central sensitization: a psychophysical controlled study. J. Pain 4, 372–380. Berkley, K.J., McAllister, S.L., Accius, B.E., Winnard, K.P., 2007. Endometriosis-induced vaginal hyperalgesia in the rat: effect of estropause, ovariectomy, and estradiol replacement. Pain 132 (Suppl. 1), S150–S159. Brosens, I.A., 1997. Endometriosis–a disease because it is characterized by bleeding. Am. J. Obstet. Gynecol. 176, 263–267. Cuevas, M., Flores, I., Thompson, K.J., Ramos-Ortolaza, D.L., TorresReveron, A., Appleyard, C.B., 2012. Stress exacerbates endometriosis manifestations and inflammatory parameters in an animal model. Reprod. Sci. 19, 851–862. D’Hooghe, T.M., Bambra, C.S., Raeymaekers, B.M., Koninckx, P.R., 1996a. Development of spontaneous endometriosis in baboons. Obstet. Gynecol. 88, 462–466. D’Hooghe, T.M., Bambra, C.S., Raeymaekers, B.M., Koninckx, P.R., 1996b. Serial laparoscopies over 30 months show that endometriosis in captive baboons (Papio anubis, Papio cynocephalus) is a progressive disease. Fertil. Steril. 65, 645–649. Ding, D., Liu, X., Duan, J., Guo, S.W., 2015. Platelets are an unindicted culprit in the development of endometriosis: clinical and experimental evidence. Hum. Reprod. 30, 812–832. Dmowski, W.P., Lesniewicz, R., Rana, N., Pepping, P., Noursalehi, M., 1997. Changing trends in the diagnosis of endometriosis: a comparative study of women with pelvic endometriosis presenting with chronic pelvic pain or infertility. Fertil. Steril. 67, 238–243. Donnez, J., Smoes, P., Gillerot, S., Casanas-Roux, F., Nisolle, M., 1998. Vascular endothelial growth factor (VEGF) in endometriosis. Hum. Reprod. 13, 1686–1690.

Q Zhang et al. Giudice, L.C., Kao, L.C., 2004. Endometriosis. Lancet 364, 1789– 1799. Guo, S.W., 2014. An overview of the current status of clinical trials on endometriosis: issues and concerns. Fertil. Steril. 101, 183– 190, e184. Guo, S.W., Ding, D., Geng, J.G., Wang, L., Liu, X., 2015a. P-selectin as a potential therapeutic target for endometriosis. Fertil. Steril. 103, 990–1000, e1008. Guo, S.W., Ding, D., Geng, J.G., Wang, L., Liu, X., 2015b. P-selectin as a therapeutic target for endometriosis. Fertil. Steril. doi:10.1016/j.fertnstert.2015.01.001. pii: S0015-0282(15)000308, Epub ahead of print. Guo, S.W., Ding, D., Shen, M., Liu, X., 2015c. Dating endometriotic ovarian cysts based on the content of cyst fluid and its potential clinical implications. Reprod. Sci. 22, 873–883. Guo, S.W., Du, Y., Liu, X., 2016. Endometriosis-derived stromal cells secrete thrombin and thromboxane A2, inducing platelet activation. Reprod. Sci. 23, 1044–1052. Hadfield, R., Mardon, H., Barlow, D., Kennedy, S., 1996. Delay in the diagnosis of endometriosis: a survey of women from the USA and the UK. Hum. Reprod. 11, 878–880. Harirchian, P., Gashaw, I., Lipskind, S.T., Braundmeier, A.G., Hastings, J.M., Olson, M.R., Fazleabas, A.T., 2012. Lesion kinetics in a non-human primate model of endometriosis. Hum. Reprod. 27, 2341–2351. Hasegawa, T., Hasegawa, F., Hirose, T., Sano, T., Matsuno, Y., 2003. Expression of smooth muscle markers in so called malignant fibrous histiocytomas. J. Clin. Pathol. 56, 666–671. He, W., Liu, X., Zhang, Y., Guo, S.W., 2010. Generalized hyperalgesia in women with endometriosis and its resolution following a successful surgery. Reprod. Sci. 17, 1099–1111. Hinz, B., Celetta, G., Tomasek, J.J., Gabbiani, G., Chaponnier, C., 2001. Alpha-smooth muscle actin expression upregulates fibroblast contractile activity. Mol. Biol. Cell 12, 2730–2741. Itoga, T., Matsumoto, T., Takeuchi, H., Yamasaki, S., Sasahara, N., Hoshi, T., Kinoshita, K., 2003. Fibrosis and smooth muscle metaplasia in rectovaginal endometriosis. Pathol. Int. 53, 371–375. Jiang, X., Chen, Y., Zhu, H., Wang, B., Qu, P., Chen, R., Sun, X., 2015. Sodium tanshinone IIA sulfonate ameliorates bladder fibrosis in a rat model of partial bladder outlet obstruction by inhibiting the TGF-beta/smad pathway activation. PLoS ONE 10, e0129655. Khare, V.K., Martin, D.C., Eltorky, M., 1996. A comparative study of ovarian and pelvic wall-infiltrating endometriosis. J. Am. Assoc. Gynecol. Laparosc. 3, 235–239. Koninckx, P.R., Ussia, A., Adamyan, L., Wattiez, A., 2011. An endometriosis classification, designed to be validated. Gynecol. Surg. 8, 1–6. Long, Q., Liu, X., Guo, S.W., 2016. Surgery accelerates the development of endometriosis in mouse. Am. J. Obstet. Gynecol. 215, 320.e1–320.e15. Lu, Y., Nie, J., Liu, X., Zheng, Y., Guo, S.W., 2010. Trichostatin A, a histone deacetylase inhibitor, reduces lesion growth and hyperalgesia in experimentally induced endometriosis in mice. Hum. Reprod. 25, 1014–1025. Maione, F., De Feo, V., Caiazzo, E., De Martino, L., Cicala, C., Mascolo, N., 2014. Tanshinone IIA, a major component of Salvia milthorriza Bunge, inhibits platelet activation via Erk-2 signaling pathway. J. Ethnopharmacol. 155, 1236–1242. Manabe, I., Owens, G.K., 2001. The smooth muscle myosin heavy chain gene exhibits smooth muscle subtype-selective modular regulation in vivo. J. Biol. Chem. 276, 39076–39087. Matsuzaki, S., Darcha, C., 2013. Involvement of the Wnt/betacatenin signaling pathway in the cellular and molecular mechanisms of fibrosis in endometriosis. PLoS ONE 8, e76808. Mechsner, S., Bartley, J., Loddenkemper, C., Salomon, D.S., StarzinskiPowitz, A., Ebert, A.D., 2005. Oxytocin receptor expression in smooth muscle cells of peritoneal endometriotic lesions and ovarian endometriotic cysts. Fertil. Steril. 83 (Suppl. 1), 1220–1231.

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006

ARTICLE IN PRESS Progressive development of endometriosis National Research Council. 1996. Guide for the Care and Use of Laboratory Animals. National Academies Press., Washington, DC. Qureshi, A.H., Chaoji, V., Maiguel, D., Faridi, M.H., Barth, C.J., Salem, S.M., Singhal, M., Stoub, D., Krastins, B., Ogihara, M., Zaki, M.J., Gupta, V., 2009. Proteomic and phospho-proteomic profile of human platelets in basal, resting state: insights into integrin signaling. PLoS ONE 4, e7627. R Development Core Team, 2013. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Shen, M., Liu, X., Zhang, H., Guo, S.W., 2016. Transforming growth factor β1 signaling coincides with -mediated epithelialmesenchymal transition and fibroblast-to-myofibroblast transdifferentiation in drive the development of adenomyosis in mice. Hum. Reprod. 31, 355–369. In press. Somigliana, E., Vigano, P., Rossi, G., Carinelli, S., Vignali, M., PaninaBordignon, P., 1999. Endometrial ability to implant in ectopic sites can be prevented by interleukin-12 in a murine model of endometriosis. Hum. Reprod. 14, 2944–2950. Somigliana, E., Vigano, P., Filardo, P., Candiani, M., Vignali, M., Panina-Bordignon, P., 2001. Use of knockout transgenic mice in the study of endometriosis: insights from mice lacking beta(2)microglobulin and interleukin-12p40. Fertil. Steril. 75, 203– 206. Tang, H., He, H., Ji, H., Gao, L., Mao, J., Liu, J., Lin, H., Wu, T., 2015. Tanshinone IIA ameliorates bleomycin-induced pulmonary fibrosis and inhibits transforming growth factor-beta-betadependent epithelial to mesenchymal transition. J. Surg. Res. 197, 167–175.

13 Vercellini, P., Crosignani, P., Somigliana, E., Vigano, P., Frattaruolo, M.P., Fedele, L., 2011. ‘Waiting for Godot’: a commonsense approach to the medical treatment of endometriosis. Hum. Reprod. 26, 3–13. Wang, D.T., Huang, R.H., Cheng, X., Zhang, Z.H., Yang, Y.J., Lin, X., 2015. Tanshinone IIA attenuates renal fibrosis and inflammation via altering expression of TGF-beta/Smad and NF-kappaB signaling pathway in 5/6 nephrectomized rats. Int. Immunopharmacol. 26, 4–12. Zhang, Q., Ding, D., Liu, X., Guo, S.W., 2015. Activated platelets induce estrogen receptor beta expression in endometriotic stromal cells. Gynecol. Obstet. Invest. 80, 187–192. Zhang, Q., Duan, J., Liu, X., Guo, S.W., 2016a. Platelets drive smooth muscle metaplasia and fibrogenesis in endometriosis through epithelial-mesenchymal transition and fibroblast-tomyofibroblast transdifferentiation. Mol. Cell. Endocrinol. 428, 1–16. In press. Zhang, Q., Duan, J., Olson, M., Fazleabas, A., Guo, S.W., 2016b. Cellular changes consistent with epithelial-mesenchymal transition and fibroblast-to-myofibroblast transdifferentiation in the progression of experimental endometriosis in baboons. Reprod. Sci. 1409–1421. In press.

Declaration: The authors report no financial or commercial conflicts of interest.

Received 1 April 2016; refereed 7 October 2016; accepted 15 November 2016.

Please cite this article in press as: Qi Zhang, Xishi Liu, Sun-Wei Guo, Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with induced endometriosis, Reproductive BioMedicine Online (2016), doi: 10.1016/j.rbmo.2016.11.006