{"paper_id":"5ba761cc-07b2-49aa-acf3-7ccd8606ebe7","body_text":"1Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreports\nNeuropeptides substance p and \nCalcitonin Gene Related peptide \nAccelerate the Development and \nFibrogenesis of endometriosis\nDingmin Yan1, Xishi Liu1,2 & sun-Wei Guo1,2\nendometriotic lesions are known to be hyperinnervated, especially in lesions of deep endometriosis \n(De), which are frequently in close proximity to various nerve plexuses. De lesions typically have \nhigher fibromuscular content than that of ovarian endometriomas (OE) lesions, but the underlying \nreason remains elusive. Aside from their traditional role of pain transduction, however, whether or \nnot sensory nerves play any role in the development of endometriosis is unclear. Here, we show that, \nthorough their respective receptors neurokinin receptor 1 (NK1R), calcitonin receptor like receptor \n(CRLR), and receptor activity modifying protein 1 (RAMP-1), neuropeptides substance P (SP) and \ncalcitonin gene related peptide (CGRP) induce epithelial-mesenchymal transition (EMT), fibroblast-to-\nmyofibroblast transdifferentiation (FMT) and further turn stromal cells into smooth muscle cells (SMCs) \nin endometriotic lesions, resulting ultimately in fibrosis. We show that SP and CGRP , or the rat dorsal \nroot ganglia (DRG) supernatant, through the induction of NK1R and CGRP/CRLR/RAMP-1 signaling \npathways, promoted EMT, FMT and SMM in endometriosis, resulting in increased migratory and \ninvasive propensity, cell contractility, production of collagen, and eventually to fibrosis. Neutralization \nof NK1R and/or CGRP/CRLR/RAMP-1 abrogated these processes. Extended exposure of endometriotic \nstromal cells to SP and/or CGRP or the DRG supernatant induced increased expression of α-SMA, \ndesmin, oxytocin receptor, and smooth muscle myosin heavy-chain. Finally, we show that De lesions \nhad significantly higher nerve fiber density, increased staining levels of α-SMA, NK1R, CRLR, and \nRAMP-1, concomitant with higher lesional fibrotic content than that of OE lesions. The extent of \nlesional fibrosis correlated positively with the staining levels of NK1R, CRLR, and RAMP-1, as well as the \nnerve fiber density in lesions. Thus, this study provides another piece of evidence that sensory nerves \nplay an important role in promoting the development and fibrogenesis of endometriosis. It explains \nas why DE frequently have higher fibromuscular content than that of OE, highlights the importance of \nlesional microenvironment in shaping the lesional fate, gives more credence to the idea that ectopic \nendometrium is fundamentally wounds that go through repeated tissue injury and repair, and should \nshed much needed light into the pathophysiology of endometriosis.\nCharacterized by the ectopic deposition and growth of endometrial-like tissues, endometriosis is an \nestrogen-dependent and inflammatory disorder affecting ~8% of premenopausal women\n1. However, this seem-\ningly innocuous definition camouflages the disease that can manifest a wild variation in size, location, color, \ndepth of infiltration, presence or absence of adhesion, and the proportion of endometriotic epithelial/stromal \ncells, let alone a kaleidoscopic variation in symptomology and severity. It has been widely accepted that there are \nthree major subtypes of endometriosis: ovarian endometriomas (OE), deep endometriosis (DE), and superficial \nperitoneal endometriosis (PE)\n2. Based mainly on their different histology, these subtypes have long been hypoth-\nesized to be three separate disease entities and perhaps have different pathogenesis and pathophysiology2.\nPreviously called deep infiltrating endometriosis3,4 but now redefined as adenomyosis externa or simply deep \nendometriosis5,6, DE is less prevalent than OE7 and is found not only in the rectovaginal septum, but also in all \n1Shanghai OB/GYN Hospital, Fudan University, Shanghai, 200011, China. 2Shanghai Key Laboratory of f emale \nReproductive Endocrine-Related Diseases, Fudan University, Shanghai, China. Dingmin Yan and Xishi Liu contributed \nequally. Correspondence and requests for materials should be addressed to S.-W.G. (email: hoxa10@outlook.com)\nReceived: 10 September 2018\nAccepted: 16 January 2019\nPublished: xx xx xxxx\nopeN\n\n\n2Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nfibromuscular pelvic structures such as the uterosacral and utero-ovarian ligaments and the muscular wall of \npelvic organs6. DE includes rectovaginal lesions as well as infiltrative forms that involve vital structures such as \nbowel, ureters, and bladder8. Though less prevalent than OE, >95% of women with DE complain of severe pain, \nincluding dysmenorrhea, dyspareunia, non-menstrual pelvic pain, and, less commonly, dyschezia and dysuria5,8, \nand DE is the most difficult subtype of endometriosis to manage clinically9–12.\nResearch on DE has been remarkably extensive, yet its pathogenesis and pathophysiology still remain elu-\nsive5,6,8,13,14. One feature that DE stands out from other subtypes of endometriosis is its presence of smooth muscle \nmetaplasia (SMM) and its high degree of fibrotic tissues 4,15–17, which explains the choice of the term adenomy-\nosis externa, presumably because of its enriched fibromuscular content 15 akin to adenomyosis. This is one of \nseveral reasons that prompted a recent proposal to re-define endometriosis to include the pro-fibrotic nature of \nendometriosis\n18.\nDespite all the vast phenotypic variation in different subtypes of endometriosis, however, all subtypes as \nwell as adenomyosis have one defining hallmark in common, namely, they all go through cyclic or repeated \nbleeding similar to eutopic endometrium\n19. Consequently, they are essentially wounds that go through repeated \ntissue injury and repair (ReTIAR) 20,21. As a result of this ReTIAR, the ectopic endometrium actively interacts \nwith various cells in its microenvironment, activates the transforming growth factor (TGF)-β1/Smad3 signaling \nand experiences epithelial-mesenchymal transition (EMT) and fibroblast-to-myofibroblast transdifferentiation \n(FMT), causing increased collagen production and cellular contractility and eventually resulting in fibrosis\n22. \nExtended exposure to TGF-β 1 also results in elevated expression of α -smooth muscle actin (α -SMA) and of \nmarkers of terminally differentiated smooth muscle cells (SMCs) in the stromal component of endometriotic \nlesions, accounting for SMM that is common in endometriotic lesions\n23–26. This essentially depicts the natural \nhistory of endometriotic lesions27.\nIn support for this notion, we recently found that, compared with OE, DE lesions appeared to have gone \nthrough EMT, FMT, and SMM more thoroughly and more extensively, and, accordingly, exhibited significantly \nmore fibromuscular tissues yet reduced cellularity and vascularity\n28. The issue left unresolved is why there are \nsuch differences between OE and DE.\nIn contrast to OE lesions, DE lesions are in uncanny proximity to several nerve plexuses such as inferior hypo-\ngastric, vesical, uterovaginal, and rectal plexus. More strikingly, DE lesions are frequently hyperinnervated29–34. \nThis raises the prospect that sensory nerve-derived neuropeptides such as substance P (SP) may precipitate the \ndevelopment and fibrogenesis of endometriosis. Indeed, we found, via mouse models, that chemical denerva-\ntion of sensory nerves resulted in significantly less lesional fibrosis than denervation of sympathetic nerves\n35. In \naddition, surgical denervation of sensory nerves significantly reduced lesional fibrosis, and the antagonism of \nneurokinin 1 receptor (NK1R), the SP receptor, also achieved the same effect\n35.\nConsequently, we hypothesized that SP and calcitonin gene related peptide (CGRP), another neuropeptide \nsecreted by sensory nerves, induce EMT and FMT in endometriosis, leading to increased collagen production \nand eventually to fibrosis. In addition, extended exposure to SP and CGRP as well as sensory dorsal root ganglia \n(DRG) supernatant further turn endometriotic stromal cells into differentiated smooth muscle cells, yielding \nSMM. This study was undertaken to test these hypotheses.\nResults\nSP and CGRP induce morphological changes consistent with EMT in endometriotic epithelial cells.  \nWe first evaluated the effect of SP and/or CGRP treatment on 11Z cells, an endometriotic epithelial cell line36. We \nfound that 11Z cells treated with SP or CGRP for 72 hours underwent conspicuous morphological changes con-\nsistent with EMT, changing the morphology from round-shaped to spindle-like feature and dispersed (Fig. 1A). \nConsistent with the morphological changes, the protein expression levels of E-cadherin, an epithelial marker, \nwere significantly decreased after 72 hours of SP or CGRP treatment (Fig. 1B), while the expression levels of genes \ninvolved in EMT, such as Snai1 and Slug, and of markers of mesenchymal cells, such as vimentin and N-cadherin, \nas well as the gene involved in fibrosis such as PAI-1 (Serpine1) were significantly elevated in 11Z cells treated \nwith SP and/or CGRP as compared with the controls (all p-values < 0.05; Fig. 1C).\nsp  and CGRp enhance the proliferative, migratory and invasive propensity in endometriotic \nepithelial cells. By CCK-8 cell proliferation and viability assay, we found that 11Z cells treated with SP and/\nor CGRP for 72 hours resulted in significantly increased cellular proliferation as compared with controls (all \np-values < 0.05; Fig. 1D). Using the scratch assay, we found that the migratory capability of 11Z cells treated with \nSP and/or CGRP was significantly increased as compared with untreated ones (all p-values  < 0.001; Fig.  1E). \nMoreover, we found, by invasion assay, that treatment with SP and/or CGRP for 72 hours significantly increased \nthe invasiveness of endometriotic epithelial cells as compared with controls (all p-values < 0.05; Fig. 1F).\nSP and CGRP induce FMT in endometriotic epithelial cells. Consistent with the EMT-like changes in \nendometriotic epithelial cells treated with SP or CGRP , we also found, through laser scanning confocal micros-\ncopy, that the expression of E-cadherin in 11Z cells was substantially reduced when treated with SP or CGRP as \ncompared with controls, especially when treated for an extended period (Fig. 2). In addition, vimentin expression \nbecame progressively elevated (all p-values < 0.01; Fig. 2), indicating that both SP and CGRP induced EMT in \nendometriotic epithelial cells.\nNot surprisingly, endometriotic epithelial cells initially did not show any expression of α -SMA or F-actin \n(Fig.  2). However, their expression became conspicuous when treated with SP or CGRP for 12 days (all \np-values < 0.01; Fig. 2), suggesting that, after a prolonged exposure to SP or CGRP , the endometriotic epithelial \ncells were further transdifferentiated from mesenchymal cells through EMT to myofibroblasts. In contrast, cells \ntreated with vehicle for the same durations were all negative for vimentin, α -SMA and F-actin but positive for \n\n3Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nFigure 1. The effect of SP and CGRP treatment on morphological, molecular and functional changes in \nendometriotic epithelial cells. (A) Representative micrographs of endometriotic epithelial cells (11Z) treated \nwith SP or CGRP for the indicated time. In all experiments, the concentration of both SP and CGRP was 10−7  M \nunless indicated otherwise. Scale bar = 100 μm. (B) Left panel: Detection of protein levels of E-cadherin by \nimmunoblotting of lysates of 11Z cells treated vehicle, SP , CGRP , or both SP and CGRP for 72 hours (n = 3). The \ngrouping of blots from the same protein were not cropped, and all protein blots were from the same gel. Right \npanel: Bar plot summarizing the fold changes in E-cadherin protein expression levels after treatment with SP \nand/or CGRP . (C) Relative fold change in gene expression of Snai1, Slug, vimentin, N-cadherin and PAI-1 in \n11Z cells treated with indicated conditions for 72 hours (n = 3). Values are normalized to GAPDH expression. \nRight panel: Relative fold change of the protein levels of E-cadherin in 11Z cells (n = 3). (D) Proliferation of \n11Z cells treated with SP or CGRP for 72 hours was evaluated by the CCK-8 assay (n = 6). (E) The migratory \ncapacity, as evaluated by the scratch assay, of 11Z cells treated with SP or CGRP was significantly increased as \n\n4Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nE-cadherin as shown in Fig. 2A (only the results at day 0 were shown since the remaining results were identical \nto the cells evaluated at day 0).Thus, we provided another piece of evidence that SP or CGRP secreted by sensory \nnerves promote EMT-like morphological and molecular changes, and extended exposure of endometriotic epi-\nthelial cells to SP or CGRP also promote myofibroblast activation.\nSP and CGRP induce FMT and further differentiate endometrial and primary endometriotic \nstromal cells into smooth muscle cells (SMCs). We next investigated the effect of SP and/or CGRP \non endometrial and endometriotic stromal cells in the development of endometriosis. We found that ESCs and \nHESCs treated with SP and/or CGRP underwent noticeable morphological changes reminiscent of myofibroblast \ntransdifferentiation, since the morphology of these cells went from spindle like and became thinner and more \nelongated reminiscent of muscle fibers and further dispersed (Fig. 3A).\nα -SMA is considered as a marker for myofibroblast and SMC\n37. Oxytocin receptor (OTR) is detected only in \nfully differentiated SMCs38, desmin is a marker for differentiated and mature SMC37, and smooth muscle-myosin \nheavy chain (SM-MHC) is considered as a marker restricted for the SMC 39. Since myofibroblasts are the most \nimportant effector cells in fibrogenesis40, we next evaluated the expression of genes known to be involved in FMT \nin ESCs and HESCs. Treatment with SP and/or CGRP resulted in significant upregulation of CCN2 (CTGF), \nFN, LOX, COL1A1, and α-SMA in both ESCs and HESCs as compared with that in controls (all p-values < 0.05; \nFig. 3B). More remarkably, when ESCs and HESCs were exposed to SP and/or CGRP for an extended time (12 \ndays), the protein expression levels of α-SMA and of desmin and OTR, the two markers for differentiated SMCs, \nwere significantly increased as compared with that treated with vehicle (Fig. 3B).\nIn addition, immunofluorescence confirmed the progressively increased expression of α -SMA, F-actin, \ndesmin, OTR and SM-MHC in HESCs treated with SP as the treatment duration increased (Fig. 3C), suggesting \nprolonged exposure of HESCs to SP and/or CGRP results in SMCs-like differentiation (Fig. 3C,D). In contrast, \nHESCs treated with vehicle for the same durations showed no change in the staining levels of either α -SMA, \nF-actin, desmin, OTR or SM-MHC and, as such, the results were identical to the treated cells evaluated at day \n0 and were thus not shown. These results strongly indicate that SP and/or CGRP secreted from sensory nerves \ninduce FMT and further SMM in endometriotic stromal cells.\nsp  and CGRp increase cellular contractility and collagen production in endometriotic and nor-\nmal endometrial stromal cells after FMT. One important step in wound healing is the wound contrac-\ntion, which is accomplished by myofibroblasts41. Collagen gel contraction by fibroblasts has been ascribed to the \ncontraction of actin filaments, which generate the cumulative traction force exerted by fibroblasts on collagen \nfibrils\n42. In vitro collagen gel contraction is considered to simulate the wound contraction process in vivo43.\nWe found that, compared to that treated with vehicle, the contractility of 11Z cells, HESCs, and ESCs was all \nsignificantly increased after treatments with SP and/or CGRP (all p-values < 0.05) just for 3 hours and further \nincreased progressively and then plateaued at 48 hours (similar results for 3 cell types, but only results on HESCs \nare shown in Fig. 4A). In addition, and consistent with FMT and increased contractility, treatment with SP and/\nor CGRP for 72 hours resulted in significantly increased production of soluble collagens in 11Z cells, ESC and \nHESCs as compared with controls (all p-vales < 0.05; Fig. 4B). These results clearly show that treatment with SP \nand/or CGRP led to myofibroblast activation and increased collagen production in endometriotic epithelial and \nstromal cells as well as normal endometrial stromal cells.\nThe DRG supernatant induces fibroblast transdifferentiation to myofibroblasts and further to \nSMC. Considering that SP and/or CGRP , which are secreted mostly from sensory nerves, induced FMT and \nSMM in endometriotic stromal cells as shown above, we next investigated whether co-culture with the DRG \nsupernatant can also induce these cells to undergo differentiation into SMCs as shown above. By immunoflu-\norescence, we found that HESCs treated with the DRG supernatant for 4 days had slightly increased α -SMA \nand F-actin staining but, not surprisingly, negative staining of OTR, desmin, and SM-MHC. However, OTR, \ndesmin and SM-MHC staining was progressively and significantly elevated as the treatment duration increased, \nconcomitant with progressively increased α-SMA and F-actin staining (Fig. 5). In contrast, HESCs treated with \njust medium for the same durations showed no change in the staining levels of either α -SMA, F-actin, desmin, \nOTR or SM-MHC and, as such, the results were identical to the treated cells evaluated at day 0 and were thus not \nshown. Consistently, the gene expression levels of α-SMA, OTR, desmin and SM-MHC were progressively and \nsignificantly elevated as the duration of treatment increased, indicating that sensory nerves induce endometriotic \nstromal cells to undergo further differentiation into fully differentiated SMCs and are thus responsible for SMM \nin endometriotic lesions.\nSP and CGRP neutralization reverses DRG-induced EMT and FMT in endometriotic cells. Given \nthe roles of SP and CGRP in inducing EMT, FMT and SMM in endometriotic cells as shown above, we next \ninvestigated whether SP and CGRP neutralization, respectively, by Aprepitant, a potent and selective NK1R \ncompared with untreated ones. The cells were photographed at 0, 12 and 24 h after being scratched (n = 6). The \ndistance between edges of cells traversed was calculated, in pixel numbers, relative to the initial scratch distance. \nScale bar = 100 μm. (F) The representative photomicrographs of the invaded 11Z cells in transwell assay under \ndifferent treatments for 72 h. The total number of cells invaded to the bottom of the transwell was then counted \n(n = 5). Magnification: × 200. Scale bar = 100 μm. Symbols of statistical significance: *p < 0.05, **p < 0.01, \n***p < 0.001. Data are represented in mean ± SD. C: Control; SP: Substance P; CGRP: calcitonin gene related \npeptide.\n\n5Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nantagonist, and CGRP Fragment 8–37, a selective competitive antagonist of CGRP receptors, would abolish \nsensory nerve-induced EMT, FMT, and SMM in endometriotic cells. We found that neutralization of SP and/\nor CGRP significantly abrogated DRG-induced changes in morphology and the expression of genes/proteins \ninvolved in EMT (with the only exception of Snail when CGRP was neutralized, likely due to the lack of statis-\ntical power) in endometriotic epithelial cells, especially when SP and CGRP were both neutralized (Fig. 6A–C). \nIn addition, neutralization of either SP , CGRP or both significantly and nearly completely abolished sensory \nnerve-induced proliferative, migratory and invasive propensity in endometriotic epithelial cells (Fig. 6D–F).\nSimilarly, with the only exception of α -SMA, CCN2, and OTR, for which CGRP neutralization did not yield \nstatistically significant reduction likely due to lack of sufficient statistical power, SP and/or CGRP neutralization \ncompletely abolished DRG-induced FMT and SMM as manifested by changes in morphology (Fig. 7A) and the \nexpression levels of markers of myofibroblasts and of SMCs in endometriotic stromal cells (Fig. 7B). Consistent \nwith the immunofluorescence results shown above (Fig. 5), the treatment of the DRG supernatant significantly \nFigure 2. SP and CGRP induce transdifferentiation to a myofibroblast phenotype in endometriotic epithelial \ncells. Immunofluorescence evaluation of the expression of E-cadherin, vimentin, α-SMA and F-actin in 11Z \ncells after treatment with SP or CGRP for indicted times. (A) Representative photomicrographs of 11Z cells \nshowing immunofluorescent staining of E-cadherin (in green), vimentin (in red), α-SMA (in red) and F-actin \nfibers (in green) after treatments with SP (10\n−7  M) or CGRP (10−7  M) for 6 and 12 days, respectively. The \nnucleus was stained blue. The control group was also evaluated at day 0 before the treatment, and no change was \nfound treated with vehicle for the same durations. Since the results were identical to the cells evaluated at 0 days, \nthe figures are not shown. Magnification: ×400. Scale bar = 50 μm. (B) Summary of the immunofluorescence \nresults by mean optical density (MOD) (in pixels) on the same exposure condition. **p < 0.01. Data are \nrepresented in mean ± SD.\n\n6Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nFigure 3. The effect of SP and CGRP on morphological and molecular changes in an endometrial stromal \ncell line (ESCs) and primary endometriotic stromal cells (HESCs). (A) Representative morphology of ESCs \nand HESCs treated with SP or CGRP for 72 h. Scale bar = 100 μm. (B) Relative fold change of gene expression \nlevels of CCN2 (CTGF), FN, LOX, Collagen I (COL1A1), α-SMA, desmin and OTR in ESCs and HESCs \ntreated with vehicle, SP , CGRP , or both SP and CGRP for 12 days (triplicates for ESCs and n = 8 for HESCs). \nValues are normalized to the GAPDH expression. (C) Upper panel: Immunofluorescence results showing \nprogressively increased staining of α-SMA, F-actin, desmin, OTR and SM-MHC in HESCs treated with SP as \nthe duration of treatment increases. In contrast, HESCs treated with vehicle for the same durations showed \nno change in the staining levels of either α-SMA, F-actin, desmin, OTR or SM-MHC and, as such, the results \nwere identical to the treated cells evaluated at day 0 and were thus not shown. Lower panel: Summary of the \nimmunofluorescence results for HESCs treated with SP for different durations by MOD (in pixels). (D) Upper \npanel: Immunofluorescence results showing progressively increased staining of α-SMA, OTR and SM-MHC \n\n7Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nincreased the gene expression levels of α-SMA, desmin and OTR (Fig. 7B). Moreover, consistent with the changes \ninduced by SP and/or CGRP treatment, neutralization of SP and/or CGRP also abolished increased cellular con-\ntractility induced by the DRG supernatant in endometriotic stromal cells, and abolished the production of soluble \ncollagens induced by the treatment with the DRG supernatant in endometriotic epithelial and stromal cells as well \nas endometrial stromal cells (Fig. 7C–D).\nThe extent of lesional fibrosis correlates with nerve fiber density and lesional NK1R/RAMP-1/CRLR  \nexpression levels. Having demonstrated the effect of sensory nerve-derived SP and/or CGRP in promoting \nprogression of endometriotic lesions through EMT, FMT, and SMM via in vitro and in vivo studies, we now turn \nto human endometriotic lesions. We predicted that lesional expression levels of NK1R, the SP receptor, as well \nas calcitonin receptor like receptor (CRLR) and receptor activity modifying protein 1 (RAMP-1), the two CGRP \nreceptors, correlate with the extent of lesional fibrosis.\nWe first performed an IHC analysis of α-SMA, NK1R, RAMP-1 and CRLR in both OE and DE tissue samples. \nWe also evaluated the density of CGRP + sensory never fibers within the lesions as well as the extent of lesional \nfibrosis by Masson trichrome staining. Since the expression of RAMP-1 and CRLR, the two receptors for CGRP , \nhas not been reported in endometrium or endometriosis, we also stained the two for normal endometrial tissues.\nWe found that the immunoreactivity against α-SMA was seen mostly in cytoplasm in endometriotic stromal \ncells, as expected, and NK1R staining was seen mostly in cytoplasm and membranes in endometriotic epithelial \ncells (Fig. 8A). The lesional density of CGRP positive nerve fibers correlated with the pain severity in women with \nendometriosis (Spearman’s r = 0.85, p < 2.2 × 10\n− 16). Concomitantly, lesional staining levels of α -SMA, NK1R, \nRAMP-1 (either epithelial or stromal), and CRLR (either epithelial or stromal) correlated with the sensory nerve \nfiber density in the lesions (all r’s > 0.69, p < 8.7 × 10\n− 10). As shown with Masson trichrome staining, the extent \nof fibrosis was also remarkably aggravated in endometriotic lesions as nerve fiber density increased, especially \nin DE lesions (Fig. 8B). The RAMP-1 and CRLR immunostaining could be found in both epithelial and stromal \ncomponents of normal endometrium, but their staining levels were significantly lower than that in either OE or \nDE lesions (Fig. 8A,B).\nAs expected and also consistent with previously reported\n28,44, we found that, compared with OE lesions, DE \nlesions have significantly higher nerve fiber density, higher extent of fibrosis, and higher α-SMA and NK1R stain-\ning levels (Fig. 8). In addition, the staining levels of both RAMP-1 and CRLR in both OE and DE lesions were sig-\nnificantly higher than that of normal endometrium irrespective of cellular component (Fig. 8). Linear regression \nanalysis indicated that, after controlling for age, menstrual phase, parity, presence or absence of adenomyosis, and \npresence or absence of uterine fibroids, the RAMP-1 and CRLR staining levels were still significantly higher in OE \nand DE lesions than that of normal endometrium (all p-values < 0.036, all R2 ≥ 0.34). For endometriotic lesions, \nthe NK1R, RAMP-1, and CRLR staining levels were significantly higher in DE lesions than OE lesions, even after \nthe adjustment for age, menstrual phase, parity, presence or absence of adenomyosis, and presence or absence of \nuterine fibroids (all p-values < 0.008).\nNot surprisingly, the severity of dysmenorrhea correlated positively with the nerve fiber density in lesions \n(Spearman’s r = 0.85, p < 2.2 × 10\n− 16; Supplementary Fig. 1A). It also positively correlated with the extent of \nfibrosis in endometriotic lesions (r = 0.90, p < 2.2 × 10− 16; Supplementary Fig. 1B), and the lesional α -SMA \nand NK1R immunostaining levels (r = 0.81, p = 2.3 × 10− 15, and r = 0.71, p = 1.8 × 10− 10; Supplementary \nFigure 1C,D), as well as with that of RAMP-1 and CRLR (Supplementary Fig. 1E–H). As expected, the extent of \nlesional fibrosis positively correlated with the lesional staining levels of NK1R, RAMP-1, and CRLR (all r’s ≥  0.60, \np < p < 5.2 × 10\n−17 ; Fig. 9).\nDiscussion\nIn this study, we have shown that, as with platelet-derived TGF-β122, sensory nerve-derived neuropeptides SP and \nCGRP facilitate EMT, FMT and differentiation to SMC in endometriosis, yielding increased collagen production, \nelevated cellular contractility, and eventually fibrosis. Neutralization of their respective receptors, such as NK1R, \nRAMP-1 and CRLR, however, abrogates these processes. Extended exposure to sensory DRG supernatant further \nturned endometriotic stromal cells into differentiated SMCs, resulting in SMM. More remarkably, lesional nerve \nfiber density correlated with the lesional expression levels of NK1R, RAMP-1 and CRLR, and ultimately with the \nextent of lesional fibrosis as well as the severity of pain in women with endometriosis. These data, taken together \nwith our in vivo data\n35 in conjunction with the successful establishment of a mouse DE model of by infusion of \nSP and CGRP in addition to i.p. endometrium injection45 provide an additional strong piece of evidence that sen-\nsory nerve fibers play a potent facilitatory role in expediting the development and fibrogenesis of endometriotic \nlesions. This also provides an answer to a long-standing conundrum as why DE lesions have abundant SMC-like \ncells\n17,25,46 and more extensive fibrosis than that of OE lesions2,4,24,28,46.\nConsidering the consensus that DE patients suffer much more serious pain, these results may explain the \ndifferences between OE and DE that sensory nerve fibers and their secreted neuropeptides accelerate the lesional \ndevelopment through facilitating EMT, FMT and SMM and ultimately fibrosis, making the DE lesions that are \ntypically more fibromuscular than that of OE lesions.\nin HESCs treated with CGRP as the treatment duration increases. The untreated cells at the same time points \nshowed no change at all time points and were identical to the treated cells evaluated at day 0, and, as such, the \nresults are not shown. Lower panel: Summary of the immunofluorescence results for HESCs treated with CGRP \nfor different durations by MOD (in pixels). Scale bar = 50 μm. Symbols of statistical significance: *p < 0.05; \n**p < 0.01, ***p < 0.001. Data are represented in mean ± SD. In all experiments, the concentration of both SP \nand CGRP was 10\n−7  M.\n\n8Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nViewed from the lens of the ReTIAR22, our results are consistent with the well documented roles of SP/CGRP \nand their receptors in wound healing and fibrogenesis. Successful repair of injured tissues requires diverse inter-\nactions among different cells, biochemical mediators, and the cellular microenvironment 47–49. It is well docu-\nmented that sensory denervation impairs cutaneous wound healing through increased apoptosis and reduced \nproliferation50,51. Indeed, SP acts as an immune modulator and injury messenger in various peripheral tissues52. \nFurthermore, SP mobilizes mesenchymal stem cells52 and endothelial progenitor cells (EPCs)53 in the bone mar-\nrow, and induces them to migrate into the injured peripheral tissues where they are involved in tissue regenera-\ntion. SP accelerates the normal acute and chronic wound healing processes\n54–56. Subcutaneous administration of \nSP accelerates the normal acute wound healing response via increased angiogenesis, resulting from SP-mediated \nEPC mobilization\n57. Similarly, genetic deletion or blockade of CGRP receptors has been suggested to be detrimen-\ntal to wound healing58–60. SP can indirectly promote hemangiogenesis in tissues by recruiting granulocytes with \nangiogenic potential from the blood circulation61. As both SP and CGRP are vasodilators62, the activation of the \nSP/CGRP and their receptors would result in plasma extravasation63,64 and platelet extravasation, contributing to \nplatelet aggregation in endometriosis as we demonstrated previously65.\nNK1R is expressed in human and rodent uterus66–69, so are CRLR and RAMP170. The expression of NK1R and \nSP-coding gene TAC1 is reported to be upregulated by estrogen68,71 and TNF-α44. This may suggest that increased \nlocal production of estrogen and proinflammatory cytokines may induce NK1R in myometrium, causing uterine \nhyperactivity and, subsequently, pain, as in colon\n72,73. As local production of TNF-α and estrogen is increased in \nendometriotic lesions, TAC1 and NK1R expression would be and in fact has been reported to be elevated44. NK1R \nactivation has been shown to be involved in ERK1/2 protein (MAPK), p38 MAPK, NF-κB, PI3K, Akt, Src, EGFR \nand Rho/Rock signaling pathways in different cell types\n74. Importantly, all these proteins have been implicated in \nthe development of endometriosis.\nFigure 4. SP and CGRP increase cellular contractility and promote collagen production in endometriotic \ncells. (A) Left panel: the representative results of collagen gel contraction assay for primary endometriotic \nstromal cells (HESCs) after treatments with SP (10\n−7  M) or CGRP (10−7  M) for 48 h. Right panel: Summary \nof the contractility experiment for HESCs, in terms of diameter of the gel surface, measured at 0, 3, 24, and \n48 h, respectively (n = 5). (B) The amount of soluble collagen secreted by 11Z cells (n = 3), ESCs (n = 3) and \nHESCs (n = 8) after treatment with SP and/or CGRP for 72 h. The absorbance value was determined at 570 nm \n(optical density, or OD), and the concentration of collagen (μg/mL) was determined by the collagen reference \nstandard curves. Data are represented in mean ± SD. Symbols of statistical significance: *p < 0.05; **p < 0.01, \n***p < 0.001.\n\n9Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nFigure 5. Immunofluorescence results showing the expression of myofibroblast and smooth muscle cell \nmarkers in HESCs treated with DRG supernatant for different durations as indicated. (A) Representative \nphotomicrographs showing that DRG supernatant gradually induces the transdifferentiation of primary \nendometriotic stromal cells to a myofibroblast and then smooth muscle cell like phenotype. After treatment \nwith the DRG supernatant for 12 days, the staining levels of α-SMA, F-actin, desmin, OTR and SM-MHC \nwere significantly elevated. In contrast, HESCs treated with just medium for the same durations showed no \nchange in the staining levels of either α-SMA, F-actin, desmin, OTR or SM-MHC and, as such, the results were \nidentical to the treated cells evaluated at day 0 and were thus not shown. Scale bar = 50 μm. (B) Summary of \nthe immunofluorescence results for HESCs treated with DRG supernatant for different durations by MOD (in \npixels). Data are represented in mean ± SD. **p < 0.01.\n\n10Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nFigure 6. SP and CGRP neutralization reverses DRG-induced EMT and FMT in 11Z cells. (A) Representative \nmorphology of 11Z cells treated with medium, the DRG supernatant with pre-treatment with aprepitant (10− 6 M), \nCGRP fragment 8–37 (10− 6 M), or both aprepitant (10− 6 M) and CGRP 8–37 (10− 6 M), or without for 12 days. Scale \nbar = 100 μm. (B) Left panel: Detection of protein levels of E-cadherin by immunoblotting of lysates of 11Z cells \ntreated with indicated condition for 12 days (n = 3). The grouping of blots from the same protein were not cropped, \nand all protein blots were from the same gel. Right panel: Relative fold change in protein levels of E-cadherin in 11Z \ncells treated with indicated condition for 12 days (n = 3). (C) Relative fold change of gene expression levels of Snai1, \nSlug, vimentin, N-cadherin, and PAI-1 in 11Z cells treated with indicated conditions for 12 days (n = 3). Values are \nnormalized to the GAPDH expression levels. (D) Results of SP and CGRP neutralization on cellular proliferation of \n11Z cells, as measured by CCK-8 assay (n = 8). The 11Z cells were treated with the indicated conditions for 12 days. \n(E) Results of SP and CGRP neutralization on cell migratory capacity, as evaluated by the scratch assay, of 11Z cells \ntreated with indicated conditions for 12 days. The cells were photographed at 0, 12, 24 and 48 hours, respectively, \nafter being scratched. The distance between two edges that cells traversed was calculated relative to the initial \nscratch distance as measured with pixel values. Scale bar = 100 μm. (F) Results of SP and CGRP neutralization on \ninvasiveness. The representative photomicrographs of the invaded 11Z cells in the transwell assay after indicated \ntreatments (Magnification: × 200). Cells, after 12 days’ indicated treatments, were added to the top of transwells \ncoated with Matrigel and treated as indicated. The total number of cells invaded to the bottom of the transwell was \nthen counted. Scale bar = 100 μm. * or \n#p < 0.05, ** or ##p < 0.01, *** or ###p < 0.001; N: not statistically significant \n\n11Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nSP can also directly induce M2 polarization of inflammatory macrophages, which participate in tissue repair75 \nand are involved in promoting lesion growth and fibrogenesis in endometriosis76,77. Functioning through NK1R, \nwhich is widely expressed in immune cells, SP has been reported to be a potent neuroimmunomodulator78. SP has \nbeen shown to inhibit NK cell cytotoxicity through NK1R79.\nSP plays an important role in cardiac fibrosis in response to hypertension 80. Co-culture with rat primary \nsensory neurons from DRG and as well as SP , CGRP and vasoactive intestinal peptide or VIP facilitate fibroblasts \nand keratinocytes proliferation and induce collagen I production 81. Direct contact of fibroblasts with neuronal \nprocesses promotes differentiation to myofibroblasts and their contractility82.\nIn view of the above, the results presented in this and the companion manuscript are consistent with the roles \nof SP/CGRP and their receptors in wound healing and fibrogenesis, and underscores the notion that sensory \nnerves are a notable feature of the lesional microenvironment, especially in DE lesions, and innervation plays an \nimportant role in lesional development and fibrogenesis.\nWe have previously shown that platelet-derived TGF-β1 promotes EMT, FMT, SMM and fibrogenesis in endo-\nmetriotic lesions\n21,22,83. We have shown recently that macrophages also promote lesional fibrogenesis through \nidentical processes 77. We found recently that regulatory T cells also facilitate lesional fibrogenesis through the \nsame processes (Xiao et al ., unpublished data). In this and our in vivo  studies35, we have shown that sensory \nnerve-derived neuropeptides also accelerate lesional fibrogenesis through similar processes. While in each case \nthe identity of the culprit in promoting lesional fibrogenesis is different, all of them nonetheless can induce \nEMT, FMT, and SMM, resulting in ultimately lesional fibrosis. Hence in lesional microenvironment, there are \nseveral accessories to the crime of promoting endometriosis, who may conspire together (e.g., lesion- and/or \nplatelet-derived TXA2 induces neurite outgrowth, and thus lesional innervation\n84) or may act independently, that \ntogether promote lesional fibrogenesis. Thus, progressive EMT, FMT, SMM and fibrogenesis jointly constitute the \nnatural history of endometriotic lesions27.\nDE is quite different from OE or PE2 and its histology resembles that of uterine adenomyosis6. Our findings \nprovide an answer to a long-standing conundrum in endometriosis: Why DE tends to have more fibromuscular \ncontent than other subtypes of endometriosis? The answer simply lies in the locations of the DE lesions: unlike \nother subtypes, these lesions typically are in close proximity to various sensory nerve plexuses and additionally \nhave higher density of sensory nerves than other subtypes\n29,30,85,86. Moreover, the increased density of sensory \nnerves in or around DE lesions also results from neurotrophic factors secreted by endometriotic lesions 84,87–90 \nor platelets84. However, the sensory nerves in or around endometriotic lesions also become accessories to the \ncrime in inflicting pains in women with endometriosis through accelerated lesional fibrogenesis and enhanced \ntransduction of pain mediators to the CNS. Just as the location is vitally important to a real estate, the location \nof an endometriotic lesion, which defines its lesional microenvironment, is also crucially important to deter -\nmine its phenotype and, consequently, symptomology. As a corollary, OE and DE, and likely other subtypes \nof endometriosis and adenomyosis as well\n91,92, have the identical pathophysiology and similar natural history \nas wounds undergoing ReTIAR. What makes them different is their locations which encompass their lesional \nmicroenvironment.\nThat said, some caveats should be noted. In this study, we used an endometriotic epithelial cell line instead of \nprimary endometriotic epithelial cells due mainly to the technical difficulty in culturing the latter. In addition, \nwe used endometrial stromal cell line (i.e. ESCs) and primary endometriotic stromal cells derived from OE (i.e. \nHESCs). None of these cells came from DE lesions. As such, these cells are likely to be phenotypically and func-\ntionally different from those derived from OE. However, deriving epithelial cells from DE lesions may be out of \nthe question because of the technical difficulty and also due to the phenomenon of “stromal endometriosis” , i.e. \nendometriotic lesions of absent glandular epithelium\n93,94 in 12–15% of DE patients95,96. In addition, as DE lesions \nbecome highly fibromuscular17, genuine endometriotic stromal cells may be hard to found. Even one could har-\nvest them and the resultant cells have the look and feel of endometriotic stromal cells, there is still a question as \nhow we can be certain that they have not undergone FMT and are genuinely mesenchymal cells. Equally likely, \nthey could have been recruited to the lesion site from elsewhere. On balance, given somewhat consistent changes \nin endometrial stromal (i.e. ESCs) and endometriotic stromal cells (i.e. HESCs) (see, for example, Fig. 3 ), our \nchoice may be justifiable, suggesting that the phenotypic differences between OE and DE lesions may result from \ntheir respective microenvironments. This, in fact, is what this study is trying to show: sensory nerve-derived \nneuropeptides, which are abundant in the DE microenvironment, can turn non-DE endometriotic or even endo-\nmetrial stromal cells into DE-like cells, resulting in a fibromuscular phenotype that is characteristically DE.\nOur finding that sensory nerve-secreted neuropeptides promote lesional fibrosis has significant implications. \nFirst, by administration of SP and/or CGRP to mice with induced endometriosis, a mouse DE model can be \ndeveloped\n97. Currently, the only animal DE model that has been reported is through grafting uterine specimens, \nespecially in full uterine thickness, to the peritoneal cavity in baboons98. Issues of cost, facility and surgical skill \naside, the DE mouse model can be established in as short as 3 weeks, as compared with 20–24 weeks minimum for \nbaboon models. Obviously, the mouse DE model has many advantages over the baboon model. Moreover, if one \ncan establish the disease model at will, the chance that we can develop novel therapeutics can be greatly increased.\nSecond, our results strongly suggest that NK1R may be a promising drug target for treating endometriosis \nthrough stalling lesional fibrogenesis via suppression of EMT, FMT, and SMM. Aside from delaying fibrogenesis, \nNK1R antagonism can also mitigate alterations in the brain induced by chronic psychological stress\n99–101, which \n(p > 0.05). Data are represented in means ± SDs. Symbols of statistical significance: *, **, and *** indicate different \nsignificant levels when compared with the untreated cells, while #, ##, and ### indicate different significant levels when \ncompared with the cells treated with the DRG supernatant. * or #p < 0.05; ** or ##p < 0.01; *** or ###p < 0.001.\n\n12Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nis associated with visceral hypersensitivity and spinal NK1R up-regulation in female rodents102,103 and possibly in \nmouse with induced endometriosis as well104–106. Moreover, NK1R antagonism has been shown to reduce anxiety \nand emotional arousal circuit response to noxious visceral distension in women with irritable bowel syndrome107, \nwhich is likely attributable, at least in part, to increased colonic hypermotility induced by SP and NK1R activa -\ntion72,73. Given that endometriosis can induce anxiety and depression in mouse 108 and possibly in humans as \nwell109–114, the use of NK1R inhibitors as a therapeutics for endometriosis may have added benefits.\nFigure 7. SP and/or CGRP neutralization completely abolished sensory nerve-induced FMT and further SMM \nin primary endometriotic stromal cells. (A) Representative morphology of HESCs treated with medium, DRG \nsupernatant with pre-treatment with aprepitant (10−6  M), CGRP 8–37 (10−6  M), or both or without for 12 days. \nScale bar = 100 μm. (B) Relative fold change of gene expression of α-SMA, CCN2 (CTGF), FN, LOX, COL1A1, \ndesmin and OTR in HESCs treated with indicated conditions for 12 days (n = 8). (C) Left panel: SP and/or \nCGRP neutralization abolished increased cellular contractility induced by DRG supernatant treatment for 12 \ndays in endometriotic stromal cells (HESCs). Right panel: Summary of the contractility experiment for HESCs, \nin terms of diameter of the gel surface, measured at 0, 3, 24, and 48 h, respectively, after release (n = 5). (D) The \namount of soluble collagen produced by11Z cells (n = 3), ESCs (n = 3) and HESCs (n = 8) after treatment with \nindicated conditions for 12 days. The absorbance value was determined at 570 nm (optical density, or OD) and \nthe concentration of collagen (μg/mL) was determined by the collagen reference standard curves. Data are \nrepresented in mean ± SD. Symbols of statistical significance: * or \n#p < 0.05, ** or ##p < 0.01, *** or ###p < 0.001; \nN: not statistically significant (p > 0.05) Note that *, **, and *** are the statistical significance levels when \ncompared with the untreated cells, while #, ##, and ### indicate statistical significant levels when compared with \nthe cells treated with the DRG supernatant.\n\n13Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nLastly, our study further highlights the importance of the lesional microenvironment in shaping the lesional \ndevelopment route and fate. Besides endometriotic epithelial and stromal cells, which traditionally have been \nthe major research focus, other cells, such as immune cells (including platelets) and now the sensory nerve cells, \nare also critical aiders and abettors to the crime of inflicting pains and misery to women with endometriosis. \nFocusing exclusively on endometriotic cells would be difficult to explain as why OE and DE lesions differ dramati-\ncally in phenotype and symptomology and insufficient to fully understand the pathophysiology of endometriosis. \nIn addition, the identical processes, instituted by platelets, immune cells and sensory nerves, of EMT, FMT, SMM, \nand fibrogenesis experienced by endometriotic cells underscore the dynamic nature in cellular identity, pheno-\ntype, function and behavior of endometriotic cells.\nTo conclude, we have shown that sensory nerve-derived neuropeptides SP and CGRP promote the develop-\nment and fibrogenesis of endometriotic lesions through EMT, FMT and transdifferentiation to SMC. Antagonism \nof their respective receptors, however, stalls these processes. Consequently, sensory nerve fibers facilitate the \ndevelopment and fibrogenesis of endometriotic lesions along with other cells in the lesional microenvironment. \nOur study highlights the importance of lesional microenvironment in lesional development and fibrogenesis and \nalso the dynamic nature of endometriotic cells. Finally, our study suggests that NK1R may be a promising thera-\npeutic target for treating endometriosis.\nMaterials and Methods\nHuman samples. This study strictly adhered to the ethical principles outlined by the Helsinki Declaration \nand was approved by the institutional ethics review board of Shanghai OB/GYN Hospital, Fudan University. \nAfter informed consent, endometriotic tissue samples harvested for isolation and primary culture were obtained \nfrom 8 patients (mean age = 32.9 ± 5.3 years) with laparoscopically and histologically diagnosed OE but no other \ngynecological diseases, who had a regular menstrual cycle (proliferative phase, n = 2 and secretory phase, n = 6).\nFor immunohistochemistry analysis, all human tissues samples were obtained from premenopausal patients \nwith laparoscopically and histologically diagnosed OE (n = 30) or DE (n = 30), admitted to the Shanghai OB/\nGYN Hospital, Fudan University, from February, 2014 to Jun, 2016. Written informed consent was obtained \nfrom all study subjects prior to sample collection. In all cases, the OEs, staged as III-IV by the revised American \nSociety of Reproductive Medicine classification system (rASRM), were removed by stripping the cyst wall from \nthe ovaries, and the DE samples were taken from uterosacral ligaments. For controls, endometrial tissue samples \nFigure 8. Immunohistochemistry evidence of fibrosis and nerve fiber density in endometriotic lesions. (A) \nRepresentative photomicrographs of immunohistochemistry analysis of α-smooth muscle actin (α-SMA), \nneurokinin 1 receptor (NK1R), calcitonin gene related peptide (CGRP) and substance P (SP), along with \nMasson trichrome staining in ovarian endometriotic lesions and in DIE lesions. Magnification: ×400. Scale \nbar = 50 μm. (B) Immunostaining levels of CGRP-positive nerve fibers, Masson staining, α-SMA and NK1R. \nScatter plot shows the relationship between the nerve fibers density via CGRP immunoreactivity staining and \nextent of fibrosis via Masson trichrome staining in both ovarian endometriotic and DIE lesions, with its color \ncorresponding to the pain stage of patients. **p < 0.01, ***p < 0.001. Data are represented in means ± SDs.\n\n14Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nwere harvested from 24 cycling women, matching, in frequency, in age- and menstrual phase with patients with \nOE or DE after written informed consent. These women underwent surgery because of teratoma, cervical intra-\nperitoneal neoplasia III, and tubal ligation, but were found to be free of endometriosis, adenomyosis, and uterine \nabnormalities, such as uterine leiomyomas. Table 1 lists the characteristics of all recruited subjects.\nNone of the patients had received hormonal or anti-platelet treatment for at least 90 days before the surgery. \nIn addition, none of them had any malignant or other inflammatory disease.\nCells and reagents. Established by Strazinski-Powitz et al.115, the endometriotic epithelial cell line (HEES or \n11Z)36 was kindly provided by Professor Jung-Hye Choi of Kyung Hee University, Seoul, Republic of Korea, and \ncultured in RPMI 1640 medium (Gibco Laboratories, Life Technologies, Grand island, NY , USA) supplemented \nwith 5% fetal bovine serum (FBS) (Gibco), 100 IU/mL penicillin G, 100 μg/mL streptomycin and 2.5 μg/mL \nAmphotericin B (Hyclone, Utah, USA). The human endometrial stromal cell line (ESC), as reported by Krikun  \net al.\n116, was kindly provided by Professor Asgi Fazleabas of Michigan State University, Michigan, U.S.A., and cul-\ntured in Dulbecco’s modified Eagle’s medium/Ham’s F-12 medium (DMEM/F-12, Hyclone) supplemented with \n5% FBS, 100 IU/mL penicillin G, 100 μg/mL streptomycin and 2.5ug/mL Amphotericin B.\nThe primary human endometriotic stromal cells (HESCs) were derived as previously reported\n117 and used in \nour previous work22. Briefly, the ectopic endometrial tissues were washed with DMEM/F-12 medium and minced \ninto small pieces of about 1 mm³ in size. After enzymatic digestion of minced tissues with 0.2% collagenase II \nFigure 9. Scatter plot of the extent of lesional fibrosis and lesional nerve fiber density (A), lesional staining \nlevels of NK1R (B), epithelial RAMP-1 (C), stromal RAMP-1 (D), epithelial CRLR (E) and stromal CRLR (F). \nEach dot in the figure represents one data point (one patient) and different colors represent different severity of \ndysmenorrhea. The numbers shown in each figure is the Pearson’s correlation coefficient, and *** indicates that \nthe statistical significance level of the correlation coefficient is less than 0.001.\n\n15Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n(Sigma, St. Louis, MO, USA) in a shaking bed for 1.5 h at 37 °C, they were separated by filtration through a \n76-μm then a 37-μ m nylon mesh. The filtrated cells were centrifuged and suspended in DMEM/F-12 medium \nsupplemented with 10% FBS, 100 IU/mL penicillin, 100 mg/mL streptomycin and 2.5 μg/mL Amphotericin B, \nand seeded into 25-cm2 cell culture flasks and incubated at 37 °C in humidified atmosphere of 5% CO2 in air. The \npurity of endometriotic stromal cells was confirmed by immunocytochemistry using an antibody against vimen-\ntin (Abcam, Cambridge, UK), a specific marker of stromal cells, and an antibody against cytokeratin 7 (CK7) \n(Zhongshan Jinqiao, Beijing, China), a specific marker of epithelial cells. The vimentin staining was positive and \nthe CK7 staining was negative after the third passage. To rule out the possibility of contamination with ovarian \ngranulosa cells, we also stained the primary cells with follicle stimulating hormone receptor (Abcam, Cambridge, \nUK), a specific marker for granulosa cells\n118. The staining was found to be negative, as we previously reported119.\nCells with different treatments were used for quantitative real-time RT-PCR, Western blot, invasion assay, \nscratch test, cell immunofluorescence, cell contractility and collagen assays. SP (Sigma), a neuropeptide and \ninflammatory mediator involved in pain transmission, was administered at the concentration of 10\n− 7 M120. CGRP \n(Sigma), a long-lasting vasodilator, was used at the concentration of 10−7  M following the previous report121. For \ninhibitor experiments, cells were pretreated with vehicle or the potent NK1R antagonist aprepitant (Selleckchem) \n(10\n− 6 M)122 or CGRP Fragment 8–37 (CGRP 8–37) (Sigma) (10 − 6 M)123, a selective competitive antagonist for \nCGRP receptors, for 1 hour at 37 °C. Aside from the references cited above, we note that SP at the concentration of \n10− 7 M is reported to increase cell viability, reduce apoptosis, stimulate proliferation, and inducs proinflammatory \nsignaling in some pathological conditions, such as Crohn’s disease124 and acute intestinal inflammation125. At con-\ncentrations ranged from 10− 12 M to 10− 7 M, CGRP is reported to significantly induce epithelial cell migration and \nproliferation in a dose-dependent manner126. CGRP at the concentration of 10− 7 M induces complete relaxation \nof human coronary arteries, while pre-incubation with CGRP 8–37 at the concentration of 10−6  M causes almost \nperfectly antagonistic effect 127. Moreover, 10− 6 M of aprepitant, or about 534.4 ng/mL, is well within or below \nthe plasma concentration in cancer patients who took aprepitant to prevent chemotherapy-induced nausea and \nvomiting: After taking one 125-mg capsule on day 1 followed by an 80-mg capsule on days 2 and 3\n128, the median \nand interquartile range of plasma concentration of aprepitant 1 day and 3 days were reported to be 768 ng/mL \n(592–949) and 915 ng/mL (563–1203), respectively 129. With these considerations, we believe that our choice of \ndoses are well justified.\nIsolation of DRG-derived neurons.  Thirty virgin female Sprague-Dawley rats (Shanghai Center for \nExperimental Animals, Chinese Academy of Sciences, Shanghai, China), 4–5 weeks old, 100–120  g in body -\nweight, were used for this study following the guidelines of the National Research Council’s Guide for the Care \nand Use of Laboratory Animals\n130 and approval by the Ethics Review Board, Shanghai OB/GYN Hospital.\nVariable\nNormal endometrium \n(n = 24)\nOvarian endometrioma \n(n = 30)\nDeep endometriosis \n(n = 30) p-value\nAge (in years)\n   Mean ± S.D 35.5 ± 6.4 32.8 ± 7.5 38.5 ± 6.1\n0.004\n   Median (Range) 34 (26–48) 31 (23–53) 39 (27–52)\nMenstrual phase\n   Proliferative 18 (75.0%) 13 (43.3%) 11 (36.7%)\n0.084\n   Secretory 6 (25.0%) 17 (56.7%) 19 (63.3%)\nParity\n   0 3 (12.5%) 17 (56.7%) 7 (23.3%)\n0.002   1 18 (75.0%) 11 (36.7%) 22 (73.3%)\n   2 3 (12.5%) 2 (6.7%) 1 (3.3%)\nrASRM stage\n   III\nNA\n14 (46.7%)\nNA NA\n   IV 16 (53.3%)\nSeverity of dysmenorrhea\n   None 24 (100.0%) 14 (46.7%) 0 (0.0%)\n1.3 × 10\n–14   Mild 0 (0.0%) 9 (30.0%) 10 (33.3%)\n   Moderate 0 (0.0%) 4 (13.3%) 5 (16.7%)\n   Severe 0 (0.0%) 3 (10.0%) 15 (50.0%)\nCo-occurrence with adenomyosis\n   No 24 (100.0%) 29 (96.7%) 16 (53.3%)\n1.4 × 10−6\n   Yes 0 (0.0%) 1 (3.3%) 14 (46.7%)\nCo-occurrence with uterine fibroids\n   No 24 (100.0%) 25 (83.3%) 19 (63.3%)\n0.0015\n   Yes 0 (0.0%) 5 (16.7%) 11 (36.7%)\nTable 1. Characteristics of recruited patients who donated their tissue samples for this study. rASRM: Revised \nAmerican Society for Reproductive Medicine classification system for endometriosis.\n\n16Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nAfter sacrifice by decapitation, the lumbar DRG was dissected, which was digested with 1 mg/mL collagenase \ntype 1 A, 0.4 mg/mL trypsin type I, and 0.1 mg/mL DNase I (all from Sigma) in DMEM/F-12 at 37 °C for 30 min, \nand then triturated, as reported previously131. Following filtration through a 100-μm nylon mesh, the dissociated \nDRG neurons were plated on coverslips coated with 1 mg/mL poly-D-lysine at room temperature for 2 hours, \nand then cultured in DMEM/F-12 containing 10% FBS. Six hours later, the culture medium was replaced with \nDMEM/F-12 containing 1% N2 supplement (Life Technologies, Grand Island, NY , USA), and the neurons were \nmaintained at 37 °C in humidified atmosphere of 5% CO\n2 for further experimentation.\nCellular proliferation by CCK-8 assay. The 11Z cells and HESCs were seeded at a density of ∼ 2,000 per \nwell in a 96-well plate and cultured in standard medium, then starved by a serum-free medium for 12 hours. \nThe cells were incubated at 37 °C in a humidified atmosphere of 5% CO2. For stimulation experiment, 11Z cells \nwere treated with vehicle, SP (10− 7 M), CGRP (10− 7 M), or both SP (10− 7 M) and CGRP (10− 7 M); for antagonist \nexperiments, HESCs were treated with medium, DRG supernatant with or without pre-treatment of aprepitant \n(10\n− 6 M), CGRP 8–37 (10− 6 M), or both aprepitant (10− 6 M) and CGRP 8–37 (10− 6 M), for 1 hour. After 48 hours, \n10 μL of CCK-8 solution (Dojindo Co., Ltd. Kumamoto, Japan) was added into each well, followed by incubation \nfor 1–4 h, and the color of the solution was closely monitored. When the difference in the color was significant \nbetween groups, absorbance was measured at 450 nm. The experiment was done in triplicate.\nRNA isolation and real-time Rt-pCR. Total RNA was isolated from 11Z cells, ESCs and HESCs after dif-\nferent treatments for 12 days, using TRIzol (Invitrogen, Carlsbad, CA, USA). The synthesis of cDNA was carried \nout using the reverse transcription kit (Takara, Takara Bio, Inc., Otsu, Shiga, Japan). The mRNA abundance was \nquantitated by real-time PCR using SYBR Premix Ex Taq (Takara). The expression values were normalized to the \ngeometric mean of GAPDH measurements and the quantification of mRNA abundance was performed using the \nmethod as previously described\n132. Table 2 lists the names of genes and their primers used in this study.\nWestern blot analysis. Cells were scraped and their total proteins were extracted in a Radio-  \nImmunoprecipitation Assay (RIPA) buffer (Fermentas, Thermo Fisher Scientific, Pittsburgh, PA, USA). The pro-\ntein concentration was evaluated using a bicinchoninic acid (BCA) protein quantitative analysis kit (P0010S, \nBeyotime, Shanghai, China). Briefly, protein samples were loaded on a 10% SDS-PAGE and transferred to pol-\nyvinyl difluoride (PVDF) membranes (Bio-Rad, Hercules, CA, USA). The membranes were incubated at 4 °C \novernight with the primary antibodies (listed in Table 3). After the membranes were incubated with HRP-labeled \nsecondary antibodies at room temperature for 1 hour, the band images were developed with enhanced \nGene name Sequence\nGAPDH\nforward 5′-GCACCGTCAAGGCTGAGAAC-3′\nreverse 5′-TGGTGAAGACGCCAGTGGA-3′\nVimentin\nforward 5′-GAACGCCAGATGCGTGAAATG-3′\nreverse 5′-CCAGAGGGAGTGAATCCAGATTA-3′\nN-cadherin\nforward 5′-ATCCTACTGGACGGTTCG-3′\nreverse 5′-TTGGCTAATGGCACTTGA-3′\nFibronectin (FN)\nforward 5′-CCATCGCAAACCGCTGCCAT-3′\nreverse 5′-AACACTTCTCAGCTATGGGCTT-3′\nPAI-1 (Serpine1)\nforward 5′-ACCGCAACGTGGTTTTCTCA-3′\nreverse 5′-TTGAATCCCATAGCTGCTTGAAT-3′\nSnail\nforward 5′-TCGGAAGCCTAACTACAGCGA-3′\nreverse 5′-AGATGAGCATTGGCAGCGAG-3′\nSlug\nforward 5′-AAGCATTTCAACGCCTCCAAA-3′\nreverse 5′-GGATCTCTGGTTGTGGTATGACA-3′\nCollagen 1A1 (COL1A1)\nforward 5′-AGGGCCAAGACGAAGACATC-3′\nreverse 5′-GATCACGTCATCGCACAACA-3′\nα-SMA\nforward 5′-GCTTTGCTGGGGACGATGCT-3′\nreverse 5′-GTCACCCACGTAGCTGTCTT-3′\nCCN2 (CTGF)\nforward 5′-GGTCAAGCTGCCCGGGAAAT-3′\nreverse 5′-TGGGTCTGGGCCAAACGTGT-3′\nLOX\nforward 5′-TGCCAGTGGATTGATATTACAGATGT-3′\nreverse 5′-AGCGAATGTCACAGCGTA CAA-3′\nDesmin\nforward 5′-CATCCTCAAGAAGGTGTTGGAG-3′\nreverse 5′-CAAAGAGACGTGGGACGAGT-3′\nOTR\nforward 5′-GTGGTGGCAGTGTTTCAGGT-3′\nreverse 5′-CGTAGAAGCGGAAGGTGATG-3′\nTable 2. List of primers used in the real-time RT-PCR analysis.\n\n17Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nchemiluminescence (ECL) reagents (Pierce, Thermo Scientific, Rockford, IL, USA) and digitized on Image Quant \nLAS 4000 mini (GE Healthcare). Image quantification was performed with Quantity One software (Bio-Rad).\nscratch assay. Scratch assay was used to assess the migratory propensity of 11Z cells and HESCs as described \npreviously22. Briefly, the cells were plated in six-well tissue culture dishes at a concentration of 1 × 105 cells. After \nthe cells reached 80–90% confluence, the tip of a micropipette was used to wound the cells, creating a linear, \ncross-stripe scrape ~2 mm wide. The cells were washed with PBS to remove floating cellular debris and re-fed with \neither serum-free medium (for use as a negative control) or experimental medium (RPMI 1640 or DMEM/F-12 \ncontaining vehicle with a combination of SP , CGRP , or both). Cell migration was photographed (Olympus \nBX53, Olympus, Tokyo, Japan) at 200× magnification and the corresponding distance was evaluated at 0, 12, 24, \n48 hours after the scratch, and recorded with an attached digital camera (Olympus DP73, Olympus). At each time \npoint, 2–3 measurements were carried out for each well and the average distance of each edge of cells traversed \nrelative to the initial scratch distance was calculated in pixel values using Image Pro-Plus software 6.0 (version \n6.0.0.206; Media Cybernetics, Inc, Bethesda, MD, USA). Each assay was done in triplicate. All experiments were \nconducted in the presence of 5 μg/mL of mitomycin-C (Sigma) to suppress cellular proliferation\n133.\nInvasion assay.  To evaluate the effect of SP and/or CGRP or DRG supernatant treatment and of the SP/\nCGRP antagonism on the invasiveness of 11Z cells, Invasion assay with Biocoat 24-well Matrigel (BD Biosciences, \nFranklin Lakes, NJ, USA) invasion chambers (Corning, Tewksbury, MA, USA) was used. Briefly, ~10\n5 11Z cells \nresuspended in 200 μL serum-free culture medium containing different treatments were added into each upper \nchamber. For the stimulation experiment, we added the vehicle, SP (10− 7 M), CGRP (10− 7 M), or both SP (10− 7 M) \nand CGRP (10− 7 M); for the antagonism experiment, we added the medium, DRG supernatant with or without \npre-treatment with aprepitant (10 − 6 M), CGRP 8–37 (10 − 6 M), or both aprepitant (10 − 6 M) and CGRP 8–37 \n(10− 6 M) for 1 hour. The lower chamber was also added with 600 μL culture medium with 20% fetal bovine serum \n(FBS). After incubation at 37 °C for 48 hours in a humidified atmosphere of 5% CO 2 in air, the number of cells \nadhering to the lower surface of the membrane was counted under the microscope. Infiltrated cells were fixed by \n95% alcohol, nuclear stained, and counted under the microscope (Olympus BX53) fitted with a digital camera \n(Olympus DP73). The invasion index was defined to be the average count of the infiltrated cells under 200× mag-\nnification of randomly selected 3–5 fields. All experiments were carried out in triplicates.\nCell immunofluorescence.  Endometriotic epithelial 11Z cells were seeded into 12-well plates overnight \nand treated with vehicle, SP (10−7  M) and/or CGRP (10−7  M) for 0, 6 or 12 days. HESCs were seeded into 12-well \nplates and treated with either vehicle, SP (10 − 7 M), CGRP (10− 7 M) or DRG supernatant for 0, 4, 8, or 12 days. \nThe cells were then washed with PBS twice, fixed with 95% ethylalcohol for 30 minutes, suspended in 0.3% Triton \nX-100 for 20 minutes, and blocked in 10% normal goat serum followed by incubation with the primary anti-\nbodies. For 11Z cells, E-cadherin, vimentin, F-actin and α -SMA were used to evaluate the occurrence of EMT \nand FMT, respectively; HESCs were incubated at 4 °C overnight with antibodies against α -SMA, F-actin, OTR, \ndesmin, or SM-MHC. The information on these antibodies is listed in Table 3. After washing, cells were incubated \nat 37 °C for 1 hour with Alexa Flour 488-conjugated goat anti-mouse IgG (Abcam) or Alexa Flour (R) 647 goat \nanti-rabbit (Abcam) and then washed with PBS and stained with DAPI. Images of stained cells were obtained by \na laser scanning confocal microscope (Leica TCS SP5 Confocal Microscope, Leica, Solms, Germany) at room \ntemperature. Images were recorded separately with different objective lenses (20x, 40x/1.25-oil and 100x 1.4-oil \nobjective), then exported as a TIFF-format digital file. All experiments were carried out in duplicate.\nCollagen gel contraction assay. The contractility of cells treated with different conditions was evaluated \nby the cellular collagen gel contraction assay kit (CBA-201, Cell Biolabs, San Diego, CA, USA) according to the \nvendor’s instructions. Briefly, HESCs were embedded in the collagen gel and cultured three-dimensionally. They \nwere suspended in the collagen solution (2–5  × 10\n6 cells/mL), and the collagen/cell mixture (0.5  mL/plate) was \nAntibody name\nCatalog \nnumber\nVendor name and \nlocation\nConcentration Immunofluorescence /\nWestern blot\nGAPDH (loading control) 5174 CST, Boston, MA, USA −/1:1000\nE-cadherin 3195 S CST 1:100/1:1000\nVimentin ab8978 Abcam, Cambridge, UK 1:100\nAlpha smooth muscle actin (α-SMA) ab5694 Abcam 1:100\nF-actin ab205 Abcam 1:100\nOxytocin receptor (OTR) ab115664 Abcam 1:200\nDesmin ab6322 Abcam 1:200\nSmooth muscle, myosin heavy chain (SM-MHC) ab81031 Abcam 1:200\nNeurokinin 1 receptor (NK1R) ab183713 Abcam 1:50\nCalcitonin receptor like receptor (CRLR) ab84467 Abcam 1:200\nReceptor activity modifying protein 1 (RAMP-1) ab203282 Abcam 1:100\nCalcitonin gene related peptide (CGRP) ab47027 Abcam 1:200\nTable 3. List of antibodies used in the Western blot analyses, immunofluorescence and immunohistochemistry.\n\n18Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\ndispensed into 24-well plates (Corning) and incubated at 37 °C for 1 hour. Immediately after collagen polymeri-\nzation, 1 mL of culture medium with designated treatment was added to the top of each collagen gel lattice. After \nincubation for 72 hours, the collagen gels were gently released from the side of the culture dishes with a sterile \nspatula, and the gels were photographed and the then the diameter of each gel surface was carefully measured \nafter release with a vernier caliper at 3, 24 and 48 hours, respectively. In case of oval-shaped gel surface, the longest \nand shortest diameters were measured and then averaged. The difference between the diameter of the well and the \ndiameter of the gel surface indicates the extent of cellular contractility.\nImmunohistochemistry. Tissue samples were fixed (10% formalin (w/v)) and then paraffin embedded. \nEach tissue block was serially sectioned (4-μ m), and the first resultant slide was stained with hematoxylin and \neosin to validate pathologic diagnosis, with the subsequent slides being used for IHC analysis of α -smooth muscle \nactin (α -SMA) (1:100, Abcam), NK1R (1:50, Santa Cruz), CRLR (1:200, Abcam), RAMP-1 (1:100, Abcam), and \nCGRP (1:200, Abcam). α -SMA was stained for myofibroblasts and differentiated smooth muscle cells (SMCs), \nand CGRP were used as specific markers for sensory nerve fibers. The number of nerve fibers was determined as \npreviously described\n87. The area with the greatest number of nerves was selected, and after scanning the section \nat low magnification (100x), five randomly selected areas were evaluated and averaged for each lesions. Negative \ncontrol sections were processed by omitting the primary antibody. Dorsal root ganglia were used as positive con-\ntrol for the CGRP staining.\nThe endogenous receptor for SP is NK1R\n44, while for CGRP , its major receptors are calcitonin receptor like \nreceptor (CRLR) and receptor activity modifying protein 1 (RAMP-1). CRLR by itself cannot function as a recep-\ntor for CGRP as it needs to form complexes with accessory proteins from the RAMP family\n134. Routine depa-\nraffinization and rehydration were carried out, as reported previously135.\nTo retrieve antigens, the slides were heated at 98 °C in a citrate buffer (pH6.0) for 30 minutes and then cooled \nto room temperature. After incubation with goat blocking serum for 15 min at room temperature, the processed \nslides were then incubated at 4 °C overnight with the intended primary antibodies (listed in Table 3). After wash-\ning with phosphate-buffered saline, the slides were added with the horse reddish peroxidae-labeled secondary \nantibody Detection Reagent (Sunpoly-HII; BioSun Technology Co, Ltd, Shanghai, China) and incubated at room \ntemperature for 30 minutes. The resultant bound antibody complexes were stained with diaminobenzidine for \n3–5 minutes or until appropriate for microscopic examination, followed by counterstaining with hematoxylin \n(30 seconds) and then mounted. The staining images were procured with an Olympus BX53 microscope fitted \nwith an Olympus DP73 digital camera. For each sample, 3–5 images at 400x magnification, selected at random, \nwere taken to arrive a mean density value using the software Image Pro-Plus 6.0 (Media Cybernetics, Inc).\nMasson trichrome staining. To identify collagen fibers in endometriotic tissue samples, Masson trichrome \nstaining was performed. The sections (4-μm, paraffin embedded) were deparaffinized in xylene and rehydrated \nin a series of graded alcohol, then were soaked in Bouin’s solution at 37 °C for 2 h. The Bouin’s solution was made \nwith 75 mL of saturated picric acid, 25 mL of 10% (w/v) formalin solution and 5 mL of acetic acid. Following ven-\ndor’s instructions, the sections were processed using the Masson Trichrome Staining kit (Baso, Wuhan, China). \nThe extent of lesional fibrosis, defined to be the areas of the collagen fiber layer (stained in blue) relative in pro-\nportion to the entire portion of the endometriotic lesions, was quantitated by the software Image Pro-Plus 6.0.\nCollagen assay.  The cell culture medium was collected after 11Z cells (n  = 3), ESCs (n  = 3) and HESCs \n(n = 8) treated with different conditions for 72 hours and then subjected to Sircol soluble collagen assay (S1000, \nBiocolor, Carrickfergus, UK) following the manufacturer’s instructions. Briefly, the culture medium was collected \nand then centrifuged to discard the particulate materials on the bottom. Since the medium contained serum sup-\nplement, low protein binding microcentrifuge tubes (Eppendorf, Hamburg, Germany) were used. The absorbance \nvalue at 570 nm filter indicates the amount of collagens in the culture medium. The concentration of collagen \nin 1 mL culture medium was determined by the reference standard curves obtained using a microplate reader \n(Biotek, Winooski, VT, USA).\nstatistical analysis. To compare the distributions of continuous variables between two groups, Wilcoxon’s \nrank test was employed. To compare the before-after difference for the same group of subjects, the paired t-test \nwas used for data from experiments using cell lines due to uniformity, and paired Wilcoxon test was used for data \nfrom experiments using primary cells because of possible heterogeneity. Pearson’s correlation coefficient was used \nfor two continuous variables. When one or more variables were ordinal, Spearman’s correlation coefficient was \nused. The result was considered to be statistically significant when the P value was less than 0.05. All calculations \nwere carried out using the software R (version 3.5.0)\n136.\nReferences\n 1. Giudice, L. C. & Kao, L. C. Endometriosis. Lancet 364, 1789–1799, S0140673604174035 (2004).\n 2. Nisolle, M. & Donnez, J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum \nare three different entities. Fertil Steril 68, 585–596, S0015-0282(97)00191-X (1997).\n 3. Koninckx, P . R. & Martin, D. C. Deep endometriosis: a consequence of infiltration or retraction or possibly adenomyosis externa? \nFertil Steril 58, 924-928, S0015-0282(16)55436-3 (1992).\n 4. Cornillie, F . J., Oosterlynck, D., Lauweryns, J. M. & Koninckx, P . R. Deeply infiltrating pelvic endometriosis: histology and clinical \nsignificance. Fertil Steril 53, 978–983, S0015-0282(16)53570-5 (1990).\n 5. Koninckx, P . R., Ussia, A., Adamyan, L., Wattiez, A. & Donnez, J. Deep endometriosis: definition, diagnosis, and treatment. Fertil \nSteril 98, 564–571, https://doi.org/10.1016/j.fertnstert.2012.07.1061 (2012).\n 6. Gordts, S., Koninckx, P . & Brosens, I. Pathogenesis of deep endometriosis. Fertil Steril 108, 872–885 e871, S0015-0282(17)31884-8 \n(2017).\n\n19Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n 7. Koninckx, P . R. Biases in the endometriosis literature. Illustrated by 20 years of endometriosis research in Leuven. Eur J Obstet \nGynecol Reprod Biol 81, 259-271, S0301211598002000 (1998).\n 8. Vercellini, P . et al. Deep endometriosis: definition, pathogenesis, and clinical management. J Am Assoc Gynecol Laparosc 11, \n153–161 (2004).\n 9. Vercellini, P . et al. Medical treatment in the management of deep endometriosis infiltrating the proximal rectum and sigmoid \ncolon: a comprehensive literature review. Acta Obstet Gynecol Scand, https://doi.org/10.1111/aogs.13328 (2018).\n 10. Abrao, M. S., Borrelli, G. M., Clarizia, R., Kho, R. M. & Ceccaroni, M. Strategies for Management of Colorectal Endometriosis. \nSemin Reprod Med 35, 65–71, https://doi.org/10.1055/s-0036-1597307 (2017).\n 11. Abrao, M. S. et al. Deep endometriosis infiltrating the recto-sigmoid: critical factors to consider before management. Hum Reprod \nUpdate 21, 329–339, https://doi.org/10.1093/humupd/dmv003 (2015).\n 12. Donnez, J. & Squifflet, J. Complications, pregnancy and recurrence in a prospective series of 500 patients operated on by the \nshaving technique for deep rectovaginal endometriotic nodules. Hum Reprod  25, 1949–1958, https://doi.org/10.1093/humrep/\ndeq135 (2010).\n 13. Tosti, C., Pinzauti, S., Santulli, P ., Chapron, C. & Petraglia, F . Pathogenetic Mechanisms of Deep Infiltrating Endometriosis. Reprod \nSci, https://doi.org/10.1177/1933719115592713 (2015).\n 14. Zondervan, K. T. et al. Endometriosis. Nature reviews. Disease primers 4, 9, https://doi.org/10.1038/s41572-018-0008-5 (2018).\n 15. Anaf, V ., Simon, P ., Fayt, I. & Noel, J. Smooth muscles are frequent components of endometriotic lesions. Hum Reprod 15, 767–771 \n(2000).\n 16. Donnez, J. et al. Current thinking on the pathogenesis of endometriosis. Gynecol Obstet Invest 54 Suppl 1, 52–58; discussion 59–62, \n66295 (2002).\n 17. van Kaam, K. J., Schouten, J. P ., Nap, A. W ., Dunselman, G. A. & Groothuis, P . G. Fibromuscular differentiation in deeply infiltrating \nendometriosis is a reaction of resident fibroblasts to the presence of ectopic endometrium. Hum Reprod  23, 2692–2700, den153 \n(2008).\n 18. Vigano, P . et al. Time to redefine endometriosis including its pro-fibrotic nature. Hum Reprod , 1–6, https://doi.org/10.1093/\nhumrep/dex354 (2017).\n 19. Brosens, I. A. Endometriosis–a disease because it is characterized by bleeding. Am J Obstet Gynecol  176, 263–267, S0002-\n9378(97)70482-4 (1997).\n 20. Guo, S. W ., Ding, D., Shen, M. & Liu, X. Dating Endometriotic Ovarian Cysts Based on the Content of Cyst Fluid and its Potential \nClinical Implications. Reprod Sci 22, 873–883, https://doi.org/10.1177/1933719115570907 (2015).\n 21. Zhang, Q., Duan, J., Olson, M., Fazleabas, A. & Guo, S. W . Cellular changes consistent with epithelial-mesenchymal transition and \nfibroblast-to-myofibroblast transdifferentiation in the progression of experimental endometriosis in baboons. Reprod Sci 23(10), \n1409–21, https://doi.org/10.1177/1933719116641763. (2016).\n 22. Zhang, Q., Duan, J., Liu, X. & Guo, S. W . Platelets drive smooth muscle metaplasia and fibrogenesis in endometriosis through \nepithelial-mesenchymal transition and fibroblast-to-myofibroblast transdifferentiation. Mol Cell Endocrinol 428, 1–16, S0303-\n7207(16)30063-6 (2016).\n 23. Itoga, T. et al. Fibrosis and smooth muscle metaplasia in rectovaginal endometriosis. Pathol Int 53, 371–375, 1483 (2003).\n 24. Khare, V . K., Martin, D. C. & Eltorky, M. A comparative study of ovarian and pelvic wall-infiltrating endometriosis. J Am Assoc \nGynecol Laparosc 3, 235–239 (1996).\n 25. Matsuzaki, S. & Darcha, C. Involvement of the Wnt/beta-catenin signaling pathway in the cellular and molecular mechanisms of \nfibrosis in endometriosis. PLoS One 8, e76808, https://doi.org/10.1371/journal.pone.0076808 (2013).\n 26. Mechsner, S. et al . Oxytocin receptor expression in smooth muscle cells of peritoneal endometriotic lesions and ovarian \nendometriotic cysts. Fertil Steril 83(Suppl 1), 1220–1231, https://doi.org/10.1016/j.fertnstert.2004.11.038 (2005).\n 27. Guo, S.-W . Fibrogenesis resulting from cyclic bleeding: The Holy Grail of the natural history of ectopic endometrium. Hum Reprod \n33(3), 353–356, https://doi.org/10.1093/humrep/dey015. (2018).\n 28. Liu, X., Zhang, Q. & Guo, S. W . Histological and Immunohistochemical Characterization of the Similarity and Difference Between \nOvarian Endometriomas and Deep Infiltrating Endometriosis. Reprod Sci , 1933719117718275, https://doi.org/10.1177/  \n1933719117718275 (2017).\n 29. Wang, G., Tokushige, N., Markham, R. & Fraser, I. S. Rich innervation of deep infiltrating endometriosis. Hum Reprod 24, 827–834, \nden464 (2009).\n 30. Wang, G. et al. Hyperinnervation in intestinal deep infiltrating endometriosis. J Minim Invasive Gynecol 16, 713–719, S1553-\n4650(09)00357-4 (2009).\n 31. Anaf, V . et al. Increased nerve density in deep infiltrating endometriotic nodules. Gynecol Obstet Invest 71, 112–117, https://doi.\norg/10.1159/000320750 (2011).\n 32. Anaf, V . et al. Hyperalgesia, nerve infiltration and nerve growth factor expression in deep adenomyotic nodules, peritoneal and \novarian endometriosis. Hum Reprod 17, 1895–1900 (2002).\n 33. Ferrero, S. et al. Loss of sympathetic nerve fibers in intestinal endometriosis. Fertil Steril 94, 2817–2819, S0015-0282(10)01022-8 \n(2010).\n 34. Arnold, J. et al. Imbalance between sympathetic and sensory innervation in peritoneal endometriosis. Brain Behav Immun  26, \n132–141, S0889-1591(11)00489-2 (2012).\n 35. Liu, X., Y an, D. & Guo, S.-W . Sensory nerve-derived neuropeptides accelerate the development and fibrogenesis of endometriosis. \nHum Reprod In press. (2018).\n 36. Gaetje, R., Kotzian, S., Herrmann, G., Baumann, R. & Starzinski-Powitz, A. Nonmalignant epithelial cells, potentially invasive in \nhuman endometriosis, lack the tumor suppressor molecule E-cadherin. Am J Pathol 150, 461–467 (1997).\n 37. Hasegawa, T., Hasegawa, F ., Hirose, T., Sano, T. & Matsuno, Y . Expression of smooth muscle markers in so called malignant fibrous \nhistiocytomas. J Clin Pathol 56, 666–671 (2003).\n 38. Busnelli, M. et al . Oxytocin-induced cell growth proliferation in human myometrial cells and leiomyomas. Fertil Steril  94, \n1869–1874, https://doi.org/10.1016/j.fertnstert.2009.10.064 (2010).\n 39. Manabe, I. & Owens, G. K. The smooth muscle myosin heavy chain gene exhibits smooth muscle subtype-selective modular \nregulation in vivo. J Biol Chem 276, 39076–39087, https://doi.org/10.1074/jbc.M105402200 (2001).\n 40. Wynn, T. A. Cellular and molecular mechanisms of fibrosis. J Pathol 214, 199–210, https://doi.org/10.1002/path.2277 (2008).\n 41. Shaw, T. J. & Martin, P . Wound repair at a glance. J Cell Sci 122, 3209–3213, https://doi.org/10.1242/jcs.031187 (2009).\n 42. Grinnell, F . Fibroblasts, myofibroblasts, and wound contraction. J Cell Biol 124, 401–404 (1994).\n 43. Bell, E., Ivarsson, B. & Merrill, C. Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of \ndifferent proliferative potential in vitro. Proc Natl Acad Sci USA 76, 1274–1278 (1979).\n 44. McKinnon, B. D., Evers, J., Bersinger, N. A. & Mueller, M. D. Induction of the neurokinin 1 receptor by TNFalpha in endometriotic \ntissue provides the potential for neurogenic control over endometriotic lesion growth. J Clin Endocrinol Metab 98, 2469–2477, \nhttps://doi.org/10.1210/jc.2013-1019 (2013).\n 45. Y an, D., Liu, X. & Guo, S. W . The Establishment of a Mouse Model of Deep Endometriosis. Hum Reprod, https://doi.org/10.1093/\nhumrep/dey361.  (2018).\n 46. Bonte, H. et al. Histologic appearance of endometriosis infiltrating uterosacral ligaments in women with painful symptoms. J Am \nAssoc Gynecol Laparosc 9, 519–524 (2002).\n\n20Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n 47. Barbul, A. Immune aspects of wound repair. Clin Plast Surg 17, 433–442 (1990).\n 48. Cioffi, W . G., Burleson, D. G. & Pruitt, B. A. Jr. Leukocyte responses to injury. Arch Surg 128, 1260–1267 (1993).\n 49. Guirao, X. & Lowry, S. F . Biologic control of injury and inflammation: much more than too little or too late. World J Surg  20, \n437–446 (1996).\n 50. Smith, P . G. & Liu, M. Impaired cutaneous wound healing after sensory denervation in developing rats: effects on cell proliferation \nand apoptosis. Cell Tissue Res 307, 281–291, https://doi.org/10.1007/s00441-001-0477-8 (2002).\n 51. Engin, C., Demirkan, F ., Ayhan, S., Atabay, K. & Baran, N. K. Delayed effect of denervation on wound contraction in rat skin. Plast \nReconstr Surg 98, 1063–1067 (1996).\n 52. Hong, H. S. et al. A new role of substance P as an injury-inducible messenger for mobilization of CD29(+) stromal-like cells. Nat \nMed 15, 425–435, nm.1909 (2009).\n 53. Amadesi, S. et al. Role for substance p-based nociceptive signaling in progenitor cell activation and angiogenesis during ischemia \nin mice and in human subjects. Circulation  125(1774–1786), S1771–1719, https://doi.org/10.1161/CIRCULATIONAHA.111.  \n089763 (2012).\n 54. Kant, V . et al. Topical application of substance P promotes wound healing in streptozotocin-induced diabetic rats. Cytokine  73, \n144–155, https://doi.org/10.1016/j.cyto.2014.12.015 (2015).\n 55. Leal, E. C. et al. Substance P promotes wound healing in diabetes by modulating inflammation and macrophage phenotype. Am J \nPathol 185, 1638–1648, https://doi.org/10.1016/j.ajpath.2015.02.011 (2015).\n 56. Y ang, L. et al. Substance P promotes diabetic corneal epithelial wound healing through molecular mechanisms mediated via the \nneurokinin-1 receptor. Diabetes 63, 4262–4274, https://doi.org/10.2337/db14-0163 (2014).\n 57. Um, J. et al. Substance P enhances EPC mobilization for accelerated wound healing. Wound Repair Regen 24, 402–410, https://doi.\norg/10.1111/wrr.12403 (2016).\n 58. Brain, S. D. Sensory neuropeptides: their role in inflammation and wound healing. Immunopharmacology 37, 133–152 (1997).\n 59. Brain, S. D. & Grant, A. D. Vascular actions of calcitonin gene-related peptide and adrenomedullin. Physiol Rev  84, 903–934, \nhttps://doi.org/10.1152/physrev.00037.2003 (2004).\n 60. Toda, M. et al. Roles of calcitonin gene-related peptide in facilitation of wound healing and angiogenesis. Biomed Pharmacother 62, \n352–359, https://doi.org/10.1016/j.biopha.2008.02.003 (2008).\n 61. Kohara, H., Tajima, S., Y amamoto, M. & Tabata, Y . Angiogenesis induced by controlled release of neuropeptide substance P . \nBiomaterials 31, 8617–8625, https://doi.org/10.1016/j.biomaterials.2010.07.079 (2010).\n 62. Kawasaki, H., Takasaki, K., Saito, A. & Goto, K. Calcitonin gene-related peptide acts as a novel vasodilator neurotransmitter in \nmesenteric resistance vessels of the rat. Nature 335, 164–167, https://doi.org/10.1038/335164a0 (1988).\n 63. Lembeck, F . & Holzer, P . Substance P as neurogenic mediator of antidromic vasodilation and neurogenic plasma extravasation. \nNaunyn Schmiedebergs Arch Pharmacol 310, 175–183 (1979).\n 64. Eglezos, A., Andrews, P . V ., Boyd, R. L. & Helme, R. D. Tachykinin-mediated modulation of the primary antibody response in rats: \nevidence for mediation by an NK-2 receptor. J Neuroimmunol 32, 11–18 (1991).\n 65. Ding, D., Liu, X., Duan, J. & Guo, S. W . Platelets are an unindicted culprit in the development of endometriosis: clinical and \nexperimental evidence. Hum Reprod 30, 812–832, dev025 (2015).\n 66. Patak, E. et al . Tachykinins and tachykinin receptors in human uterus. Br J Pharmacol  139, 523–532, https://doi.org/10.1038/\nsj.bjp.0705279 (2003).\n 67. Patak, E. et al. Functional and molecular characterization of tachykinins and tachykinin receptors in the mouse uterus. Biol Reprod \n72, 1125–1133, https://doi.org/10.1095/biolreprod.104.036814 (2005).\n 68. Pinto, F . M. et al. Tachykinin receptor and neutral endopeptidase gene expression in the rat uterus: characterization and regulation \nin response to ovarian steroid treatment. Endocrinology 140, 2526–2532, https://doi.org/10.1210/endo.140.6.6695 (1999).\n 69. Gonzalez-Santana, A. et al . Altered expression of the tachykinins substance P/neurokinin A/hemokinin-1 and their preferred \nneurokinin 1/neurokinin 2 receptors in uterine leiomyomata. Fertil Steril 106, 1521–1529, https://doi.org/10.1016/j.\nfertnstert.2016.07.007 (2016).\n 70. Nikitenko, L. L. et al. Differential and cell-specific expression of calcitonin receptor-like receptor and receptor activity modifying \nproteins in the human uterus. Mol Hum Reprod 7, 655–664 (2001).\n 71. Pinto, F . M., Pintado, C. O., Pennefather, J. N., Patak, E. & Candenas, L. Ovarian steroids regulate tachykinin and tachykinin \nreceptor gene expression in the mouse uterus. Reprod Biol Endocrinol 7, 77, https://doi.org/10.1186/1477-7827-7-77 (2009).\n 72. Bradesi, S., Eutamene, H., Fioramonti, J. & Bueno, L. Acute restraint stress activates functional NK1 receptor in the colon of female \nrats: involvement of steroids. Gut 50, 349–354 (2002).\n 73. Lu, P . et al. The role of substance P in the maintenance of colonic hypermotility induced by repeated stress in rats. Neuropeptides \n56, 75–82, https://doi.org/10.1016/j.npep.2016.01.006 (2016).\n 74. Garcia-Recio, S. & Gascon, P . Biological and Pharmacological Aspects of the NK1-Receptor. Biomed Res Int 2015, 495704, https://\ndoi.org/10.1155/2015/495704 (2015).\n 75. Lim, J. E., Chung, E. & Son, Y . A neuropeptide, Substance-P , directly induces tissue-repairing M2 like macrophages by activating \nthe PI3K/Akt/mTOR pathway even in the presence of IFNgamma. Sci Rep 7, 9417, https://doi.org/10.1038/s41598-017-09639-7  \n(2017).\n 76. Bacci, M. et al. Macrophages are alternatively activated in patients with endometriosis and required for growth and vascularization \nof lesions in a mouse model of disease. Am J Pathol 175, 547–556, S0002-9440(10)60569-X (2009).\n 77. Duan, J., Liu, X. & Guo, S.-W . The M2a macrophage subset may be critically involved in fibrogenesis of endometriosis in mouse. \nReprod Biomed Online 37, 254–268, https://doi.org/10.1016/j.rbmo.2018.05.017. (2018).\n 78. Mashaghi, A. et al. Neuropeptide substance P and the immune response. Cell Mol Life Sci 73, 4249–4264, https://doi.org/10.1007/\ns00018-016-2293-z.\n 79. Monaco-Shawver, L. et al. Substance P inhibits natural killer cell cytotoxicity through the neurokinin-1 receptor. J Leukoc Biol 89, \n113–125, https://doi.org/10.1189/jlb.0410200 (2011).\n 80. Dehlin, H. M., Manteufel, E. J., Monroe, A. L., Reimer, M. H. Jr. & Levick, S. P . Substance P acting via the neurokinin-1 receptor \nregulates adverse myocardial remodeling in a rat model of hypertension. International journal of cardiology 168, 4643–4651, \nhttps://doi.org/10.1016/j.ijcard.2013.07.190 (2013).\n 81. Cheret, J. et al. Influence of sensory neuropeptides on human cutaneous wound healing process. Journal of dermatological science \n74, 193–203, https://doi.org/10.1016/j.jdermsci.2014.02.001 (2014).\n 82. Fujiwara, T. et al. Direct contact of fibroblasts with neuronal processes promotes differentiation to myofibroblasts and induces \ncontraction of collagen matrix in vitro. Wound Repair Regen 21, 588–594, https://doi.org/10.1111/wrr.12059 (2013).\n 83. Zhang, Q., Liu, X. & Guo, S. W . Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with \ninduced endometriosis. Reprod Biomed Online 34, 124–136, S1472-6483(16)30611-3 (2017).\n 84. Y an, D., Liu, X. & Guo, S. W . Endometriosis-Derived Thromboxane A2 Induces Neurite Outgrowth. Reprod Sci  24, 829–835, \n1933719116670037 (2017).\n 85. Tokushige, N., Markham, R., Russell, P . & Fraser, I. S. Nerve fibres in peritoneal endometriosis. Hum Reprod 21, 3001–3007, del260 \n(2006).\n 86. Tokushige, N. et al . Nerve fibers in ovarian endometriomas. Fertil Steril 94, 1944–1947, https://doi.org/10.1016/j.\nfertnstert.2009.12.074 (2010).\n\n21Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n 87. Barcena de Arellano, M. L. et al. Evidence of neurotrophic events due to peritoneal endometriotic lesions. Cytokine 62, 253–261, \nhttps://doi.org/10.1016/j.cyto.2013.03.003 (2013).\n 88. Barcena de Arellano, M. L. et al. Overexpression of nerve growth factor in peritoneal fluid from women with endometriosis may \npromote neurite outgrowth in endometriotic lesions. Fertil Steril 95, 1123–1126, https://doi.org/10.1016/j.fertnstert.2010.10.023 \n(2011).\n 89. Greaves, E. et al. Estrogen receptor (ER) agonists differentially regulate neuroangiogenesis in peritoneal endometriosis via the \nrepellent factor SLIT3. Endocrinology 155, 4015–4026, https://doi.org/10.1210/en.2014-1086 (2014).\n 90. Greaves, E. et al . Estradiol Is a Critical Mediator of Macrophage-Nerve Cross Talk in Peritoneal Endometriosis. Am J Pathol, \nhttps://doi.org/10.1016/j.ajpath.2015.04.012 (2015).\n 91. Liu, X., Shen, M., Qi, Q., Zhang, H. & Guo, S. W . Corroborating evidence for platelet-induced epithelial-mesenchymal transition \nand fibroblast-to-myofibroblast transdifferentiation in the development of adenomyosis. Hum Reprod 31, 734–749, dew018 (2016).\n 92. Shen, M., Liu, X., Zhang, H. & Guo, S. W . Transforming growth factor beta1 signaling coincides with epithelial-mesenchymal \ntransition and fibroblast-to-myofibroblast transdifferentiation in the development of adenomyosis in mice. Hum Reprod 31, \n355–369, dev314 (2016).\n 93. Clement, P . B. The pathology of endometriosis: a survey of the many faces of a common disease emphasizing diagnostic pitfalls and \nunusual and newly appreciated aspects. Adv Anat Pathol 14, 241–260, https://doi.org/10.1097/PAP .0b013e3180ca7d7b (2007).\n 94. Mai, K. T., Y azdi, H. M., Perkins, D. G. & Parks, W . Pathogenetic role of the stromal cells in endometriosis and adenomyosis. \nHistopathology 30, 430–442 (1997).\n 95. Abrao, M. S., Neme, R. M., Carvalho, F . M., Aldrighi, J. M. & Pinotti, J. A. Histological classification of endometriosis as a predictor \nof response to treatment. Int J Gynaecol Obstet 82, 31–40, S0020729203000791 (2003).\n 96. Kamergorodsky, G. et al . Histologic classification of specimens from women affected by superficial endometriosis, deeply \ninfiltrating endometriosis, and ovarian endometriomas. Fertil Steril 92, 2074–2077, S0015-0282(09)01224-2 (2009).\n 97. Liu, X., Y an, D. & Guo, S. W . Establishment of a mouse model of deep infiltrating endometriosis [Abstract]. Reprod Sci 25, 61A \n(2018).\n 98. Donnez, O. et al. Induction of endometriotic nodules in an experimental baboon model mimicking human deep nodular lesions. \nFertil Steril 99, 783–789 e783, S0015-0282(12)02344-8 (2013).\n 99. Czeh, B. et al . Examining SLV-323, a novel NK1 receptor antagonist, in a chronic psychosocial stress model for depression. \nPsychopharmacology 180, 548–557, https://doi.org/10.1007/s00213-005-2184-8 (2005).\n 100. Czeh, B. et al. Chronic stress decreases the number of parvalbumin-immunoreactive interneurons in the hippocampus: prevention \nby treatment with a substance P receptor (NK1) antagonist. Neuropsychopharmacology  30, 67–79, https://doi.org/10.1038/sj.\nnpp.1300581 (2005).\n 101. van der Hart, M. G. et al . Substance P receptor antagonist and clomipramine prevent stress-induced alterations in cerebral \nmetabolites, cytogenesis in the dentate gyrus and hippocampal volume. Molecular psychiatry 7, 933–941, https://doi.org/10.1038/\nsj.mp.4001130 (2002).\n 102. Bradesi, S., Eutamene, H., Garcia-Villar, R., Fioramonti, J. & Bueno, L. Stress-induced visceral hypersensitivity in female rats is \nestrogen-dependent and involves tachykinin NK1 receptors. Pain 102, 227–234 (2003).\n 103. Bradesi, S. et al. Role of spinal microglia in visceral hyperalgesia and NK1R up-regulation in a rat model of chronic stress. \nGastroenterology 136, 1339–1348, e1331–1332, https://doi.org/10.1053/j.gastro.2008.12.044 (2009).\n 104. Berkley, K. J., Cason, A., Jacobs, H., Bradshaw, H. & Wood, E. Vaginal hyperalgesia in a rat model of endometriosis. Neurosci Lett \n306, 185–188, S0304394001019061 (2001).\n 105. Berkley, K. J., McAllister, S. L., Accius, B. E. & Winnard, K. P . Endometriosis-induced vaginal hyperalgesia in the rat: effect of \nestropause, ovariectomy, and estradiol replacement. Pain 132 Suppl 1, S150–159, S0304-3959(07)00552-0 (2007).\n 106. McAllister, S. L., McGinty, K. A., Resuehr, D. & Berkley, K. J. Endometriosis-induced vaginal hyperalgesia in the rat: role of the \nectopic growths and their innervation. Pain 147, 255–264, S0304-3959(09)00558-2 (2009).\n 107. Tillisch, K. et al. Neurokinin-1-receptor antagonism decreases anxiety and emotional arousal circuit response to noxious visceral \ndistension in women with irritable bowel syndrome: a pilot study. Alimentary pharmacology & therapeutics 35, 360–367, https://\ndoi.org/10.1111/j.1365-2036.2011.04958.x (2012).\n 108. Li, T. et al. Endometriosis alters brain electro-physiology, gene expression and increased pain sensitization, anxiety, and depression \nin female mice. Biol Reprod 4841840 (2018)\n 109. Low, W . Y ., Edelmann, R. J. & Sutton, C. A psychological profile of endometriosis patients in comparison to patients with pelvic \npain of other origins. J Psychosom Res 37, 111–116 (1993).\n 110. Barnack, J. L. & Chrisler, J. C. The experience of chronic illness in women: a comparison between women with endometriosis and \nwomen with chronic migraine headaches. Women Health 46, 115–133, https://doi.org/10.1300/J013v46n01_08 (2007).\n 111. Petrelluzzi, K. F ., Garcia, M. C., Petta, C. A., Grassi-Kassisse, D. M. & Spadari-Bratfisch, R. C. Salivary cortisol concentrations, \nstress and quality of life in women with endometriosis and chronic pelvic pain. Stress 11, 390–397, 902547461 (2008).\n 112. Eriksen, H. L. et al. Psychological aspects of endometriosis: differences between patients with or without pain on four psychological \nvariables. Eur J Obstet Gynecol Reprod Biol 139, 100–105, S0301-2115(07)00436-8 (2008).\n 113. Sepulcri Rde, P . & do Amaral, V . F . Depressive symptoms, anxiety, and quality of life in women with pelvic endometriosis. Eur J \nObstet Gynecol Reprod Biol 142, 53–56, S0301-2115(08)00335-7 (2009).\n 114. Quinones, M., Urrutia, R., Torres-Reveron, A., Vincent, K. & Flores, I. Anxiety, coping skills and hypothalamus-pituitary-adrenal \n(HPA) axis in patients with endometriosis. J Reprod Biol Health 3, https://doi.org/10.7243/2054-0841-3-2 (2015).\n 115. Starzinski-Powitz, A. et al. Tracing cellular and molecular mechanisms involved in endometriosis. Hum Reprod Update 4, 724–729 \n(1998).\n 116. Krikun, G. et al. A novel immortalized human endometrial stromal cell line with normal progestational response. Endocrinology \n145, 2291–2296, https://doi.org/10.1210/en.2003-1606 (2004).\n 117. Ryan, I. P ., Schriock, E. D. & Taylor, R. N. Isolation, characterization, and comparison of human endometrial and endometriosis \ncells in vitro. J Clin Endocrinol Metab 78, 642–649, https://doi.org/10.1210/jcem.78.3.8126136 (1994).\n 118. Lan, C. W ., Chen, M. J., Jan, P . S., Chen, H. F . & Ho, H. N. Differentiation of human embryonic stem cells into functional ovarian \ngranulosa-like cells. J Clin Endocrinol Metab 98, 3713–3723, https://doi.org/10.1210/jc.2012-4302 (2013).\n 119. Zhang, Q., Ding, D., Liu, X. & Guo, S. W . Activated Platelets Induce Estrogen Receptor beta Expression in Endometriotic Stromal \nCells. Gynecol Obstet Invest 80, 187–192, https://doi.org/10.1159/000377629 (2015).\n 120. Nilsson, J., von Euler, A. M. & Dalsgaard, C. J. Stimulation of connective tissue cell growth by substance P and substance K. Nature \n315, 61–63 (1985).\n 121. Liang, W ., Zhuo, X., Tang, Z., Wei, X. & Li, B. Calcitonin gene-related peptide stimulates proliferation and osteogenic differentiation \nof osteoporotic rat-derived bone mesenchymal stem cells. Mol Cell Biochem 402, 101–110, https://doi.org/10.1007/s11010-014-\n2318-6 (2015).\n 122. Chernova, I. et al. Substance P (SP) enhances CCL5-induced chemotaxis and intracellular signaling in human monocytes, which \nexpress the truncated neurokinin-1 receptor (NK1R). Journal of leukocyte biology 85, 154–164, https://doi.org/10.1189/jlb.0408260 \n(2009).\n 123. Li, X. W . et al. Inhibitory effect of calcitonin gene-related peptide on hypoxia-induced rat pulmonary artery smooth muscle cells \nproliferation: role of ERK1/2 and p27. Eur J Pharmacol 679, 117–126, https://doi.org/10.1016/j.ejphar.2012.01.015 (2012).\n\n22Scientific  RepoRts  |          (2019) 9:2698  | https://doi.org/10.1038/s41598-019-39170-w\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n 124. Gross, K. et al. Substance P promotes expansion of human mesenteric preadipocytes through proliferative and antiapoptotic \npathways. Am J Physiol Gastrointest Liver Physiol 296, G1012–1019, https://doi.org/10.1152/ajpgi.90351.2008 (2009).\n 125. Koon, H. W . et al. Substance P-stimulated interleukin-8 expression in human colonic epithelial cells involves protein kinase Cdelta \nactivation. J Pharmacol Exp Ther 314, 1393–1400, https://doi.org/10.1124/jpet.105.088013 (2005).\n 126. Bulut, K. et al. Sensory neuropeptides and epithelial cell restitution: the relevance of SP- and CGRP-stimulated mast cells. Int J \nColorectal Dis 23, 535–541, https://doi.org/10.1007/s00384-008-0447-7 (2008).\n 127. Hasbak, P . et al. Investigation of CGRP receptors and peptide pharmacology in human coronary arteries. Characterization with a \nnonpeptide antagonist. J Pharmacol Exp Ther 304, 326–333, https://doi.org/10.1124/jpet.102.037754 (2003).\n 128. Majumdar, A. K. et al. Pharmacokinetics of aprepitant after single and multiple oral doses in healthy volunteers. J Clin Pharmacol \n46, 291–300, https://doi.org/10.1177/0091270005283467 (2006).\n 129. Motohashi, S. et al. Interindividual variations in aprepitant plasma pharmacokinetics in cancer patients receiving cisplatin-based \nchemotherapy for the first time. Biol Pharm Bull 36, 676–681 (2013).\n 130. In Guide for the Care and Use of Laboratory Animals The National Academies Collection: Reports funded by National Institutes of \nHealth (2011).\n 131. Han, Q. J. et al. IPP5 inhibits neurite growth in primary sensory neurons by maintaining TGF-beta/Smad signaling. J Cell Sci 126, \n542–553, https://doi.org/10.1242/jcs.114280 (2013).\n 132. Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta \nC(T)) Method. Methods 25, 402–408, https://doi.org/10.1006/meth.2001.1262 (2001).\n 133. Lee, J. G. & Kay, E. P . FGF-2-induced wound healing in corneal endothelial cells requires Cdc42 activation and Rho inactivation \nthrough the phosphatidylinositol 3-kinase pathway. Invest Ophthalmol Vis Sci 47, 1376–1386, https://doi.org/10.1167/iovs.05-1223 \n(2006).\n 134. McLatchie, L. M. et al. RAMPs regulate the transport and ligand specificity of the calcitonin-receptor-like receptor. Nature 393, \n333–339, https://doi.org/10.1038/30666 (1998).\n 135. Zhang, Q., Liu, X. & Guo, S. W . Progressive development of endometriosis and its hindrance by anti-platelet treatment in mice with \ninduced endometriosis. Reproductive biomedicine online, https://doi.org/10.1016/j.rbmo.2016.11.006 (2016).\n 136. R: A Language and Environment for Statistical Computing. (R Foundation for Statistical Computing, Vienna, Austria, 2016).\nAcknowledgements\nThe authors would like to thank Drs. Jung-Hye Choi, Asgi Fazleabas, Graciela Krikun and Anna Strazinski-Powitz \nfor providing the cell lines used this study. This research was supported in part by grants from the National Science \nFoundation of China (81471434, 81530040, and 81771553 to SWG; 81671436 and 81871144 to XSL), and an \nExcellence in Centers of Clinical Medicine grant (2017ZZ01016) from the Science and Technology Commission \nof Shanghai Municipality. National Science Foundation of China (81471434, 81530040 and 81771553 to S.W .G.; \n81671436 and 81871144 to X.S.L.) and an Excellence in Centers of Clinical Medicine grant (2017ZZ01016) from \nthe Science and Technology Commission of Shanghai Municipality.\nAuthor Contributions\nS.W .G. conceptualized the study and carried out its design, analyzed and interpreted data, and drafted the \nmanuscript, and prepared Fig. 9 and part of Fig. 8. D.Y . performed all the experiments and carried out data \nanalysis together with S.W .G., and prepared Figs 1–8, and X.S.L. recruited patients and secured specimens. All \nparticipated in the writing and revision and approved the final version of the manuscript.\nAdditional Information\nSupplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-39170-w.\nCompeting Interests: The authors declare no competing interests.\nPublisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and \ninstitutional affiliations.\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International \nLicense, which permits use, sharing, adaptation, distribution and reproduction in any medium or \nformat, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-\native Commons license, and indicate if changes were made. The images or other third party material in this \narticle are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons license and your intended use is not per-\nmitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the \ncopyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.\n \n© The Author(s) 2019","source_license":"CC0","license_restricted":false}