Estradiol promotes cells invasion by activating β-catenin signaling pathway in endometriosis

Endometriosis is an estrogen-dependent disease that involves the adhesion, invasion, and angiogenesis of endometrial tissues outside of the uterine cavity. We hypothesized that a link exists between estrogen and beta-catenin (β-catenin) signaling in the pathogenesis of endometriosis. Human endometrial stromal cells (HESCs) were separated from eutopic endometrial tissues that were obtained from patients with endometriosis. β-catenin expression and cells invasiveness ability were up-regulated by 17β-estradiol (E2) in an estrogen receptor (ESR)-dependent manner, whereas β-catenin siRNA abrogated this phenomenon. Moreover, co-immunoprecipitation and dual immunofluorescence studies confirmed ESR1, β-catenin, and lymphoid enhancer factor 1/T cell factor 3 co-localization in the nucleus in HESCs after E2 treatment. To determine the role of β-catenin signaling in the implantation of ectopic endometrium, we xenotransplanted eutopic endometrium from endometriosis patients into ovariectomized severe combined immunodeficiency mice. The implantation of the endometrium was suppressed by β-catenin siRNA. Collectively, studies regarding β-catenin signaling are critical for improving our understanding of the pathogenesis of estrogen-induced endometriosis, which can translate into the development of treatments and therapeutic strategies for endometriosis.


Introduction
Endometriosis is a common benign gynecological disease that is defined as the presence of endometrial tissue outside the uterine cavity. This disease affects women of reproductive age and causes pelvic pain and infertility. The incidence rate of endometriosis is w10% (Giudice & Kao 2004). The pathogenesis of endometriosis is multifactorial, and endogenous and exogenous estrogens play key roles in its development and progression (Huhtinen et al. 2012). Studies have shown greater synthesis of 17b-estradiol (E 2 ) in ectopic endometrium than in eutopic endometrium, which is critical for endometrial implantation (Delvoux et al. 2009). Accordingly, the inhibition of systemic estrogen action is the current medical therapy for endometriosis and results in restricted proliferation. Aberrant production of estrogen plays an indispensable role in the pathogenesis of this disease.
Beta-catenin (b-catenin) acts as both a regulator of cell adhesion and migration in its cadherin-bound form and a transcriptional factor in Wnt signaling that plays a key role in the regulation of proliferation, growth, and differentiation in endometrial physiology and disease (van der Horst et al. 2012). Wnts binding to Frizzled results in the activation of receptor kinases, phosphorylation of the cytoplasmic mediator Dishevelled, and inhibition of the multifunctional serine/threonine kinase glycogen synthase kinase 3, ultimately allowing for the accumulation of b-catenin (Mulholland et al. 2005). b-catenin then translocates into the nucleus and associates with transcription factors of the T cell factor/ lymphoid enhancer factor (TCF/LEF) family to induce the expression of its downstream target genes (Akiyama 2000, Peifer & Polakis 2000. A previous study of Matsuzaki et al. (2010) suggested that aberrant b-catenin expression is found in the endometrium of patients with endometriosis. Moreover, endometriotic lesion development was shown to be inhibited by targeting the b-catenin/TCF complex (Matsuzaki & Darcha 2013). Many in vivo physiological studies regarding the cross-talk between estrogen and b-catenin signaling have been conducted in tissues of the brain (Cardona-Gomez et al. 2004) and uterus (Gunin et al. 2004). In neuroblastoma cells, b-catenin is stabilized by estrogen in neurons, exerting a significant effect at the transcriptional level (Varea et al. 2009). The growing number of nuclear receptors interacting with b-catenin causes alterations in cell proliferation and tumorigenesis (Mulholland et al. 2005). Based on the results of the cited studies, we hypothesized that a link exists between estrogen and b-catenin signaling in the pathogenesis of endometriosis.
The aims of the present study were to investigate the molecular mechanism underlying the interaction between estrogen and b-catenin signaling in human endometrial stromal cells (HESCs), illustrate whether the b-catenin signaling pathway plays a critical role in estrogen-facilitated invasion and angiogenesis by HESCs, and determine the role of b-catenin in a nonobese diabetic severe combined immunodeficiency (NOD-SCID) mouse endometriosis model.

Tissue collection
All eutopic endometrial tissues were obtained from endometriosis patients by hysterectomy, which was performed at the Department of Obstetrics and Gynecology, Tongji Medical College, Union Hospital, Huazhong University of Science and Technology. The average age was 39 years (range 33-44) at the time of laparoscopy. The patients had not received any hormonal treatment for at least 6 months before the surgical procedure. Informed consent was obtained from each patient using protocols approved by the Human Investigation Committee at Tongji Medical College, Union Hospital, Huazhong University of Science and Technology. All samples were obtained in the proliferative phase of the cycle, which was confirmed histologically according to established criteria.

Cell culture
Isolation and culture of HESCs began with the collection of eutopic tissues under sterile conditions. The tissues were then transported to the laboratory on ice in a 1:1 formula of DMEM/F12 (Gibco) with 10% fetal bovine serum (FBS; Gibco). The minced eutopic endometrium was digested with collagenase type 2 (0.1%; Sigma-Aldrich) for 30 min at 37 8C with constant agitation. The tissue pieces were filtered on a 100 mm wire sieve to remove debris. Following gentle centrifugation, the supernatant was discarded, and the cells were resuspended in DMEM/F12. The HESCs were separated from epithelial cells by passing them over a 400 mm wire sieve. The filtered suspension was layered over Ficoll and centrifuged at 1200 g for 20 min to further remove leukocytes and erythrocytes, and the middle layer was collected and then washed with D-Hanks solution. The HESCs were placed in a culture flask and allowed to adhere for 24 h. The adherent stromal cells were cultured as a monolayer in flasks with DMEM/F12 containing 15% FBS, 20 mmol/l HEPES, 100 IU/ml penicillin, and 100 mg/ml streptomycin and incubated in 5% CO 2 at 37 8C. This method supplied 95% pure HESCs.
Hormone treatment E 2 was purchased from Sigma-Aldrich (E-2758) and dissolved in DMSO. ICI 182 780 (ICI), an estrogen receptor antagonist was purchased from Cayman Chemical (CAS 129453-61-8; Ann Arbor, MI, USA) and dissolved in DMSO. Cells were incubated with E 2 at various doses for various times or with DMSO (control group).

Western blot analysis
After the cells were washed with PBS, cell extracts were prepared in lysis buffer containing 40 mM Tris-HCl, 100 mM NaCl, and 0.1% Nonidet P-40 and supplemented with protease inhibitors. Whole-cell lysates were collected by centrifugation at 16000 g at 24 8C for 10 min after the cells were incubated for 30 min on ice. Cytoplasmic and nuclear proteins were extracted according to the instructions of a cytoplasmic and nuclear protein extraction kit (Sangon Biotech, Shanghai, China). Cells were harvested in ice-cold PBS, scraped from culture dishes on ice using a plastic cell scraper, and collected in 1.5 ml micro-centrifuge tubes. The cytoplasmic lysis buffer contains 1 ml hypotonic buffer, 5 ml phosphatase inhibitor, 1 ml DL-dithiothreitol (DTT), and 10 ml phenylmethanesulfonyl fluoride (PMSF). The material was centrifuged at 3! 1000 r.p.m. for 5 min to sediment the nuclei. The supernatant was then centrifuged at 12!1000 r.p.m. for 10 min to get the non-nuclear supernatant fraction. The nuclear pellet was then washed three times with cytoplasmic lysis buffer and resuspended in nuclear lysis buffer (1 ml lysis buffer, 5 ml phosphatase inhibitor, 1 ml DTT, and 10 ml PMSF). The mixture was sonicated briefly to aid nuclear lysis. The nuclear lysates were collected after centrifugation at 12!1000 r.p.m. for 10 min at 4 8C. The samples were boiled for 10 min and then separated by 10% SDS-PAGE. The proteins were transferred to a PVDF membrane (Millipore, Billerica, MA, USA) in the presence of 20% methanol and 0.1% SDS. Nonspecific signals were blocked by incubating the membrane in 5% nonfat milk in Tris-buffered saline-Tween-20 for 1 h. The membranes were incubated with primary antibodies of interest overnight at 4 8C. The antibodies were directed against the following proteins: b-catenin (#8480, Cell Signaling Technology, Danvers, MA, USA), non-phospho b-catenin (#8814, Cell Signaling Technology), GAPDH (ab181602, Abcam, Cambridge, UK), or histone H1 (ab61177, Abcam). The blots were developed using HRP-conjugated secondary antibodies (Cell Signaling Technology), and the proteins were visualized by an ECL procedure (Millipore) according to the manufacturer's recommendations.

siRNAs and transfection
HESCs were grown in culture medium with 15% FBS before transfection. When the cells were 50-60% confluent, b-catenin siRNA (RiboBio, Guangzhou, China) and Lipofectamine 2000 (Invitrogen) were added to opti-MEM, mixed, incubated for 20 min, and then added to the cells at room temperature (RT) according to the manufacturer's protocol. Scrambled siRNA (RiboBio) was used as a control. After 6 h, the mixture was replaced with phenol red-free DMEM/F12 with 15% FBS in 5% CO 2 at 37 8C.

Plasmids and dual-luciferase assay
TOPflash, a TCF reporter plasmid; FOPflash, a negative control for TOPflash; and pRL-SV40, a Renilla luciferase expression plasmid, were purchased from GeneChem (Shanghai, China). Co-transfection experiments were performed in 24-well plates. In total, 1!10 5 cells were seeded per well in 500 ml medium. Then, 0.8 mg TOPflash or FOPflash and 0.008 mg pRL-SV40 reporter plasmid plus Lipofectamine 2000 transfection reagent were added to the HESCs. After 48 h, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega). Then, firefly luciferase activity was normalized to the corresponding Renilla luciferase activity.

Matrigel invasion assay
Transwell units with 8.0 mm pore-size polycarbonate filters (Corning Costar, Tewksbury, MA, USA) were precoated with 50 ml 1:2 diluted Matrigel (Sigma-Aldrich) and used to investigate cell invasion. A total volume of 100 ml suspension containing w10 5 cells/ml was added to each upper compartment of the precoated units. After the cells were allowed to attach for 30 min, the units were transferred to wells containing 500 ml DMEM/F12 medium with 20% FBS as a chemoattractant and incubation was conducted for 24 h. After removing the cells and Matrigel from the upper surface of the membrane with a cotton bud and staining with 0.1% crystal violet for 15 min, the number of cells on the underside was determined using light microscopy (Olympus, Japan). Five randomly selected fields were counted per insert.

Dual immunofluorescence
HESCs were cultured in 25 mm 2 dishes and treated with E 2 (Sigma-Aldrich) at 10 K8 mol/l for 48 h, with DMSO as a control. Then, the cells were fixed in ice-cold 4% paraformaldehyde for 10 min, washed with 1! PBS, and permeabilized in PBS containing 0.5% Triton X-100 for 3 min at RT. All cells were blocked (5% BSA, 30 min, RT) and incubated overnight (4 8C) with primary antibodies directed against b-catenin (#8814, Cell Signaling Technology), LEF1 (#2286, Cell Signaling Technology), TCF3 (#2883, Cell Signaling Technology), and ESR1 (ab75635, Abcam). After the cells were exposed to the primary antibodies, they were incubated with Alexa Fluor 488-and 568conjugated secondary antibodies and mounted with Prolong Gold antifade medium. Immunofluorescently labeled cell monolayers were visualized under an Olympus FV1000 laser scanning confocal microscope (Olympus).

Xenotransplantation of human eutopic endometrial tissues
Proliferative-phase eutopic endometrial tissues were obtained from ten premenopausal women with endometriosis during hysterectomy at Wuhan Union Hospital. Fresh endometrial tissue samples were fragmented into 2-3 mm 3 sections under sterile conditions. The fragments were cultured in phenol redfree DMEM/F12 (1:1)C10% FBS before transplantation into NOD-SCID mice (Pearson et al. 2008). In total, 50 8-week-old NOD-SCID female mice were used. The guidelines for animal care were approved by the Institutional Animal Care and Use Committee, Tongji Medical College, Huazhong University of Science and Technology. The mice were bilaterally ovariectomized and then left untreated for 14 days. Eight to ten fragments of human endometrial lesions were implanted into the pelvic and peritoneal cavity (Grummer et al. 2001). The mice were divided into six groups at 2 days after implantation.

Estradiol activates b-catenin signaling
Mice were co-treated with E 2 (0.2 mg/day) plus positive b-catenin siRNA (125 mg/kg per day) or scrambled siRNA (125 mg/kg per day) and saline solution control. All treatments were administrated via peritoneal injection. E 2 was purchased from Sigma-Aldrich (Y0000852). Mice were sacrificed 10 and 21 days after implantation, the peritoneal cavity was examined, and tissue collected. The implanted endometrial lesions were harvested for hematoxylin and eosin staining and for immunohistochemical (IHC) analysis. Western blot analysis of the tissues was performed as described previously.

Statistical analysis
All values are shown as the meanGS.E.M. Unpaired t-tests and one-way ANOVA were used to demonstrate statistical significance. Statistical significance was set at P!0.05. All experiments were performed in triplicate.

E 2 up-regulates b-catenin mRNA and protein expression with the involvement of ERs in HESCs
Endometriosis is an estrogen-dependent disease (Huhtinen et al. 2012). Abnormal b-catenin protein expression has been identified in ovarian endometrioid adenocarcinoma associated with endometriosis (Stewart et al. 2013), and E 2 has been shown to stabilize b-catenin expression through estrogen receptor binding in neuronal cells (Varea et al. 2009). First, we tested the relationship between E 2 and b-catenin, a key transcription factor in Wnt signaling. The qRT-PCR results showed that E 2 was capable of inducing b-catenin expression in a dose-and time-dependent manner. E 2 caused up-regulation of b-catenin mRNA at 10 K10 mol/l (w2.3-fold), followed by maximal up-regulation at 10 K8 mol/l (w2.8-fold) and a decline at 10 K6 mol/l (w1.9-fold) while remaining high compared with the control (Fig. 1A). As shown in Fig. 1B, qRT-PCR analysis demonstrated that E 2 (10 K8 mol/l) treatment caused b-catenin up-regulation (w2.2-fold) at 24 h and that this up-regulation continued to increase until 48 h (w2.9-fold). Next, we explored whether this E 2regulated response was reflected in the protein level of b-catenin. Western blot analysis demonstrated that b-catenin was up-regulated (w4.1-fold) at 10 K10 mol/l E 2 , then increased (w4.5-fold) at 10 K8 mol/l E 2 but slightly declined at 10 K6 mol/l E 2 (Fig. 1C). b-catenin protein expression was also up-regulated (w4.6-fold) at mol/l 10 -12 10 -10 10 -8 10 -6 10 -12 10 -10 10 -8 10 -6 10 -12 10 -10 10 -8 10 -6 12 h 24 h and continued to increase until 48 h (w6.1-fold) at 10 K8 mol/l E 2 (Fig. 1D). E 2 up-regulated b-catenin mRNA and protein expression in a dose-and timedependent manner in HESCs, and the effect was maximal at 10 K8 mol/l E 2 at 48 h (**P!0.01). Furthermore, we wanted to determine whether the estrogenregulated responses are dependent on estrogen receptor (ESR). To address this question, HESCs were treated with ICI (ER inhibitor, 10 K6 mol/l) in conjunction with E 2 at various doses and times. We observed that the E 2 -regulated inductive responses of b-catenin mRNA and protein were indeed antagonized by ICI (Fig. 1A, B, C and D). Collectively, these suggested that estrogen induces b-catenin up-regulation in HESCs via the involvement of ER.
E 2 activates the b-catenin signaling pathway in HESCs Next, we wanted to confirm the effects of E 2 on the b-catenin signaling pathway in HESCs. To examine whether E 2 alters the subcellular distribution of b-catenin, cells were treated with E 2 (10 K8 mol/l) or  Figure 2 (A) The interaction between ESR1 and b-catenin in HESCs was examined. Cells were incubated with DMSO and E 2 (10 K8 mol/l) for 24 h. Then cell lysates were prepared and immunoprecipitated with IgG and ESR1 antibody. The second panel is ESR1 shown as an internal loading control. Immunoblots of input lysate controls (5% of input) are also shown. (B and C) HESCs were incubated with E 2 (10 K8 mol/l) under different time scales (0, 12, 24, and 48 h). (B) Cell lysates were prepared and immunoprecipitated with ESR1 antibody or IgG. The interaction between ESR1 and b-catenin/TCF3/LEF1 was examined. (C) Cell lysates were prepared and immunoprecipitated with b-catenin antibody or IgG. The interaction between b-catenin and ESR1/TCF3/LEF1 was examined. (D and E) Co-localization of ESR1, b-catenin, and LEF1/TCF3 in nucleus were stimulated with E 2 in HESCs. Cells were cultured in phenol red-free DMEM/F12 for 24 h and incubated with DMSO and E 2 (10 K8 mol/l) for 48 h. (D) Representative confocal microscopy images of HESCs immunostained for ERS1 (green, left panels) and its co-localization with the LEF1 and TCF3 (red, middle panels), and the merged images (right panels). (E) Representative confocal microscopy images of HESCs immunostained for b-catenin (green, left panels) and its co-localization with the LEF1 and TCF3 (red, middle panels), and the merged images (right panels). The bars indicate 20 mm. Data presented are from three independent experiments. with control (DMSO) for 48 h. Western blot analysis indicated that E 2 treatment led to an increase in b-catenin protein levels in the nucleus (w2.6-fold) and to a slight decrease in b-catenin protein levels in the cytoplasm compared with the control (w0.7-fold) (Fig. 1E). As shown in Fig. 1F, non-phosphorylated b-catenin (active b-catenin) was up-regulated by E 2 (w3.67-fold) compared with the control. Next, we investigated the effect of E 2 on TCF-induced transcriptional activity in HESCs. As shown in Fig. 1G, TOPflash activity markedly increased (**P!0.01) upon stimulation with E 2 and could be abolished by ICI (**P!0.01). Collectively, these findings demonstrated that E 2 activates b-catenin signaling in HESCs.

E 2 stimulates ESR1 and b-catenin co-localization in the nucleus in HESCs
Estrogen induces cellular effects via two molecular mechanisms: a classic genomic mechanism in which E 2 binds ER to stimulate target gene expression and a nongenomic mechanism in which E 2 induces cellular effects independent of ER-related transcriptional activity (Acconcia & Kumar 2006). Therefore, we wanted to clarify which mechanism is responsible for the observed effects in HESCs. Further investigation was conducted using COIP. As shown in Fig. 2A, the interaction between ESR1 and b-catenin was significantly more frequent when cells were treated with E 2 . Moreover, we wanted to examine whether TCF3/LEF1 cross-talk with ESR1. As shown in Fig. 2B, the interaction between ESR1 and b-catenin/TCF3/LEF1 were detected at all time points (0, 12, 24, and 48 h) by E 2 (10 K8 mol/l) treatment, and the effect was maximal at 48 h. E 2 stimulated ESR1 binding to b-catenin/TCF3/LEF1 in a time-dependent manner. Furthermore, b-catenin pulldown analysis detected an ESR1/TCF3/LEF1 complex. In contrast, the effects of ESR1, TCF3, and LEF1 interactions with b-catenin were maximal at 24 h (Fig. 2C). Dual immunofluorescence studies revealed intense nuclear co-localization, with ESR1, b-catenin, and LEF1/TCF3 co-localizing in the nucleus in HESCs after treatment with E 2 for 48 h. The expression of these proteins in the control (DMSO) group also showed a co-localization pattern, but the co-localization was less intense than that in the E 2 group ( Fig. 2D and E). These results suggested that E 2 induced b-catenin activation and led to a molecular association with ESR1 and LEF1/TCF3 in HESCs.

b-catenin deletion represses E 2 -induced expression of MMP9 and VEGF in HESCs
In endometrial implants, VEGF induces neovascularization (Harada et al. 2001), and MMPs promote matrix remodeling (Sillem et al. 2001). A previous study suggested that MMP levels were significantly higher in epithelial and stromal cells in patients with endometriosis compared to patients without endometriosis (Becker et al. 2010). Estrogen has been shown to induce VEGF and MMP9 production in endometriotic lesions, leading to the adhesion and growth of endometrial cells, which can spill into the peritoneal cavity (Wang et al. 2011). These two factors seem to be vital for the establishment of endometriosis. Therefore, we focused on whether VEGF and MMP9 are up-regulated by E 2 interaction with the b-catenin signaling pathway. QRT-PCR analysis indicated that E 2 treatment led to increased levels of MMP9 (w3.5-fold) and VEGF (w4.2-fold) in HESCs, whereas ICI treatment significantly abrogated this phenomenon (*P!0.05; Fig. 3A). HESCs were transiently transfected with b-catenin siRNA. This RNA interference resulted in w90% knockdown of b-catenin protein levels in the HESCs. The expression levels of VEGF and MMP9 in the transfected HESCs decreased significantly when the cells were also treated with E 2 (*P!0.05; Fig. 3B). These results suggested that up-regulation of MMP9 and VEGF by E 2 was dependent on b-catenin.

b-catenin deletion abrogates the E 2 -induced increase of HESCs invasiveness
After HESCs were treated with E 2 for 48 h, the Transwell assay revealed that these cells were more invasive than untreated cells or cells treated with E 2 plus ICI (*P!0.05; Fig. 3C). As shown in Fig. 3D, the Transwell assay revealed that the cells treated with b-catenin siRNA were less invasive than the untreated cells (vehicle) and the negative controls (treated with scrambled siRNA) (*P!0.05). All experimental groups were treated with E 2 (10 K8 mol/l) after transfection with siRNA. These data indicated that the E 2 -induced increase in the invasiveness of HESCs was dependent on b-catenin.
Silencing of the b-catenin gene by siRNA inhibits the adhesiveness and invasiveness of eutopic endometrium induced by E 2 in a NOD-SCID mouse endometriosis model To determine whether E 2 increases the adhesiveness and invasiveness of endometrial lesions through b-catenin signaling in vivo, we established an animal model that would mimic ectopic implantation of the endometrium. To generate this model, human eutopic endometrial were implanted into the pelvic and peritoneal cavities of ovariectomized NOD-SCID female mice. Ten days after transplantation, endometriotic lesions were detected in 25% (two of eight) animals treated with E 2 and scrambled siRNA and 75% (six of eight) of animals treated with E 2 and b-catenin siRNA (Fig. 4A). Endometriotic lesions were detected in 63.6% (seven of 11 mice) animals treated with E 2 and scrambled siRNA 21 days Estradiol activates b-catenin signaling after transplantation (Fig. 4A). The discovery rate of mice treated with E 2 plus scrambled siRNA that developed viable endometriotic lesions was 63.6% (seven of 11 mice) after a 21-day incubation. No visible endometrial fragments were observed in mice treated with E 2 plus positive b-catenin siRNA after a 21-day incubation or in the control groups after both 10-and 21-day incubations. The lesions were highly vascularized as observed on gross morphological examination ( Fig. 4B  and C). The murine lesions demonstrated the presence of human endometrial glands and stroma along with epithelial cells lining the lumen by histology (Fig. 4D). To adjust for biological variations, western blot and IHC analyses were performed in samples harvested from the mice during laparotomy and compared with the results for pre-transplantation eutopic endometrium. As shown in Fig. 4E, an increased expression of dephosphorylated b-catenin was demonstrated in the implanted tissues from scrambled siRNA-treated mice compared with their pre-transplantation tissues and with implanted tissues from b-catenin siRNA-treated mice. Implanted tissues from both E 2 plus positive b-catenin siRNA-treated mice and scrambled siRNA-treated mice displayed up-regulated ESR1 expression compared with their pre-transplantation tissues. Because no visible endometrial fragments were observed in mice treated with E 2 plus positive b-catenin siRNA after a 21-day incubation, we only used the implanted tissues from scrambled siRNAtreated mice compared with their pre-transplantation tissues. As shown in Fig. 4F, the expression levels of dephosphorylated b-catenin and ESR1 were increased in the implanted tissues from mice treated with E 2 plus scrambled siRNA. We further performed IHC staining of VEGF and MMP9 in endometriotic lesions. In the 10-day incubation group, increased VEGF (w2.6-fold) and MMP9 (w2.4-fold) expression was observed in the endometriotic lesions from mice treated with E 2 plus scrambled siRNA compared with their pre-transplantation tissues (Fig. 5A). In contrast, the expression of VEGF and MMP9 proteins was attenuated in endometriotic lesions from b-catenin siRNA-treated mice compared to scrambled siRNA-treated mice (Fig. 5A). In the 21-day incubation group, increased MMP9 (w2.1-fold) and VEGF (w2.0-fold) expression was demonstrated in the implanted tissues compared with their pre-transplantation counterparts (Fig. 5B). These results confirmed that b-catenin plays a critical role in the E 2 -induced adhesion of implanted endometrial fragments to the peritoneum of mice.

Discussion
E 2 is the most potent estrogen found in humans and plays a critical role in the development and progression of endometriosis (Huhtinen et al. 2012). The activated b-catenin pathway may interact with multiple nuclear signal transducers to coordinate tissue-or cell-specific functions (Yang et al. 2002, Hou et al. 2004, Kouzmenko et al. 2004, Chandar et al. 2005, Armstrong et al. 2007). ESR1 and b-catenin have been found to precipitate within the same immune complexes in colon cancer and breast cancer (Kouzmenko et al. 2004). The b-catenin signaling pathway also plays a significant role in the estrogen-regulated normal physiological processes of the uterus via a specific protein-protein interaction between b-catenin and ESR1 (Hou et al. 2004). However, the cross-talk between E 2 and b-catenin signaling in the pathogenesis of endometriosis is unknown. b-catenin is a major component of the canonical Wnt signaling pathway. In the absence of the Wnt ligand, this pathway is inactive due to continual degradation of nonjunctional b-catenin. In the present study, we found that E 2 increased b-catenin expression at the mRNA and protein levels in a time-and dose-dependent fashion in HESCs. Furthermore, E 2 stimulated ESR1 binding to b-catenin. Taken together, our results showed that the b-catenin destruction complex failed to bind to the ESR1/b-catenin conjugation, leading to b-catenin stabilization and eventual translocation to the nucleus. The ESR1/b-catenin conjugation then associated with the transcription factors TCF3 and LEF1. When treated with ICI to antagonize the function of ESR1, b-catenin expression and stabilization decreased.
Adhesion, invasion, and angiogenesis are the steps in the fundamental pathological process of endometriosis. For ectopic implantation and growth to occur, endometrial tissue must first attach itself to the host tissue and then invade the host tissue and obtain its own blood supply from the local vasculature (Giudice et al. 1998). VEGF induces neovascularization in endometrial implants (Harada et al. 2001). The expression of MMPs in the eutopic endometrium in endometriosis patients differs from that in normal females, and the expression of the MMP9 protein in the endometrium of endometriosis patients is much higher than that in normal females (Collette et al. 2006). MMP9 activity is known to be involved in cell invasion (Moon et al. 2010, Kang et al. 2011. Certain studies have shown that MMP9 and VEGFA are downstream target genes of the Wnt/ b-catenin signaling pathway (Wu et al. 2007, Qu et al. 2014. The present study observed that E 2 induced an increase in HESCs invasiveness and that ICI blocked this induced increase. b-catenin siRNA was then transfected into HESCs, which demonstrated that invasion was significantly decreased when b-catenin expression was inhibited. Similarly, VEGF and MMP9 gene expression were also inhibited. In the NOD-SCID mouse model, b-catenin siRNA abrogated the implantation of xenotransplanted endometrium mediated by E 2 , suggesting a novel and crucial mechanism in which E 2 facilitates eutopic endometrial implantation through the b-catenin signaling pathway. However, the mechanism by which ESR1, b-catenin, and the LEF1/TCF3 complex regulate the expression of VEGF and MMP9 is unclear and requires further exploration.
In the present study, we demonstrated that the molecular and functional activation of b-catenin signaling by E 2 in HESCs leads to the promotion of the nuclear translocation of b-catenin. We further found that ESR1, b-catenin, and LEF1/TCF3 interact in nuclei. All of these results suggested that E 2 treatment leads to the accumulation of b-catenin in the nucleus, where b-catenin associates with the TCF/LEF family to activate the transcription of downstream target genes, such as the MMP and VEGF genes. These results confirm the importance of the b-catenin signaling pathway in endometriosis under conditions of abnormal estrogen levels.
In summary, the present study demonstrated that E 2 enhanced b-catenin expression and, as a consequence, b-catenin associated with the LEF/TCF family, increasing HESCs invasion and angiogenesis. In light of the multiple roles of the b-catenin signaling pathway in promoting the development of endometriosis, E 2 may accelerate disease progression by up-regulating b-catenin expression. Therefore, these findings may provide a potential therapeutic target for the treatment of endometriosis.