Abstract
Oestrogen has been reported to control the invasiveness of endometrial stromal cells in endometriosis. Notch signalling, a master regulator of cell invasion in tumours, is regulated by oestrogen in other diseases and hyperactivated in endometriotic stromal cells. Therefore, we hypothesized that an interaction between Notch signalling and oestrogen may exist in the regulation of endometrial stromal cell invasion, which is essential for the development of endometriosis. Western blot analysis of tissues showed that the expression levels of Notch components (JAG1 and NOTCH1) and Notch activity were markedly higher in ectopic endometria than in their eutopic and normal counterparts. Primary stromal cells obtained from normal endometria cultured with oestrogen presented significant increases in the expression of Notch components and Notch activity, the cytoplasmic and nuclear accumulation of NOTCH1 intracellular domain, the expression of matrix metallopeptidase 9 and vascular endothelial growth factor and cell invasiveness. Knockdown of NOTCH1 markedly alleviated oestrogen-induced matrix metallopeptidase 9 and vascular endothelial growth factor expression and cell invasion. ICI (an oestrogen receptor α antagonist) also blocked these oestrogenic effects. Oestrogen-responsive elements were found in the promoters of NOTCH1 and JAG1. A luciferase reporter analysis revealed that oestrogen regulated the expression of Notch components via oestrogen receptor alpha, which is bound to oestrogen-responsive elements in the JAG1 and NOTCH1 promoters. Collectively, our findings indicate that oestrogen engages in crosstalk with Notch signalling to regulate cell invasion in endometriosis via the activation of oestrogen receptor alpha and the enhancement of Notch activity. Notch signalling blockade may therefore be a novel therapeutic target for endometriosis.
Introduction
Endometriosis (EMS) is a benign but metastatic disease that is characterized by the appearance of endometrial tissue outside of the uterine cavity and causes cyclic pelvic pain and infertility (Giudice & Kao 2004). The most well-accepted hypothesis for the pathogenesis of EMS is the retrograde menstruation/transplantation theory (Sampson 1927). According to this theory, the successful establishment of EMS results from a series of complicated processes, including the adherence, invasion and growth of menstrual endometrial cells which originate from retrograde endomerial fragments. Of these processes, the initial step is the invasion of endometrial stromal cells into the peritoneum is a vital step for EMS formation: a process which is similar to that observed in metastatic neoplasms (Witz et al. 1999). Matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor (VEGF) are genes with well-known roles in cell invasion in metastatic tumours. MMP9 enables invasion mainly by catalysing the degradation of the extracellular matrix (ECM) (Emonard & Grimaud 1990), while VEGF facilitates invasion by interacting with VEGF receptors (VEGFRs) expressed by tumour cells in paracrine or autocrine manner (Lichtenberger et al. 2010). Abundant evidence has shown that MMP9 and VEGF are overexpressed in the ectopic endometrium (Di Carlo et al. 2009, Machado et al. 2010) and their concentrations in peritoneal fluids are elevated in patients with EMS (Bourlev et al. 2006, Liu et al. 2015); these changes may contribute to the invasion of endometrial cells. However, the upstream pathways that regulate the production of these two factors remain to be further investigated.
Oestrogen plays a key role in the pathogenesis of EMS (Kitawaki et al. 2002, Ferrero et al. 2014, da Costa e Silva Rde et al. 2016). It controls not only the maintenance and growth of endometriotic implants (Yland et al. 2018) but also the invasion of endometrial stromal cells. In vitro, oestrogen was found to indirectly regulate MMP9 expression and activity by activating Wnt/b-catenin signalling (Xiong et al. 2015, Zhang et al. 2016a ) or inducing c-Fos (a proto-oncogene) (Pan et al. 2016), thereby contributing to the invasion of endometrial stromal cells. Since Wnt/β-catenin signalling and c-Fos are both essential for tumour invasion (Dong et al. 2015, Paw et al. 2015) and EMS exhibits features of malignant tumours, we hypothesized that additional signalling pathways known to be associated with tumour invasion may also mediate oestrogen-induced invasiveness in endometrial stromal cells.
Interestingly, Notch signalling was recently found to be activated in endometriotic stromal cells (Gonzalez-Foruria et al. 2017, Brown et al. 2018). Notch signalling regulates the proliferation, apoptosis, migration, invasion and metastasis of tumour cells (Capaccione & Pine 2013). It is classically activated by the binding of receptors to ligands on adjacent cells, which leads to the cleavage of the transmembrane domain and the subsequent generation of a Notch intracellular domain (NICD) (Pancewicz & Nicot 2011, Bray 2016). NICD then translocates into the nucleus to transcriptionally activate its target genes. Overexpression of Notch signalling components and overactivation of Notch are often correlated with increased cancer cell invasiveness (Hafeez et al. 2009, Capaccione & Pine 2013). Knockdown of NOTCH1 decreased the expression of extracellular proteins involved in cell invasion, including MMP9 and VEGF (Wang et al. 2006, Zhou et al. 2012). More importantly, Notch signalling has also been shown to be modulated by oestrogen in the regulation of angiogenesis, cell differentiation and proliferation (Soares et al. 2004, Wei et al. 2012, Fan et al. 2014). Therefore, the aim of the present study was to determine whether Notch signalling mediates oestrogen-induced invasiveness in human endometrial stromal cells in EMS and investigate the underlying molecular mechanism.
Materials and methods
Tissue collection and cell culture
Informed consent was obtained from all patients, and the study was approved by the Institutional Review Board and Research Ethics Committee of Tongji Medical School, Huazhong University of Science and Technology. As previously described (Zhang et al. 2016b ), normal endometrial samples were obtained from 10 patients aged 27–40 years old (mean age: 33.10 years old) with tubal infertility and without endometriosis, as confirmed by laparoscopy. Paired eutopic and ectopic endometria were collected from 10 patients aged 25–44 years (mean age: 30.90 years old) with ovarian endometriosis that was confirmed histologically. According to the criteria of the revised American Society of Reproductive Medicine (R-ASRM) score classification system, all these cases were classified as stage III–IV, and samples were taken during the proliferative phase of the cycle based on the data of the last menstrual period and histology. Endometria in this phase were used to determine whether there are differences in Notch activity and components among the three kinds of endometria due to different local oestrogen concentrations (Huhtinen et al. 2012) while simultaneously avoiding any interference by progesterone. All patients had regular menstrual cycles and received no hormones or GnRH agonist therapy for at least 3 months before surgery. Patients with hyperplasia of the endometrium, pelvic inflammatory diseases or malignant tumours, adenomyosis, dysfunctional uterine bleeding or systemic pathologies were excluded. Both normal and eutopic endometrial samples were only used for protein extraction, while each ectopic endometrial sample was divided into two parts, one of which was used for protein extraction, while the other was used for immunohistochemical studies.
Primary endometrial stromal cells (ESCs) obtained from normal endometria were chosen for the drug treatment experiments. As previously described (Zhang et al. 2016a ), primary ESCs were isolated from each of the five specimens that were obtained from another set of 220 patients aged 25–43 years old (mean age: 31.93 years old) who were without endometriosis and underwent hysteroscopic curettage during the late secretory phase based on previous studies (Yoshida et al. 2004, Li et al. 2018). Endometria in this phase were used as a model for the retrograde endometrial fragments from which endometriotic stromal cells are thought to originate. We chose endometria in this phase because they were readily available and best mimicked the menstrual endometria (i.e. had the most similar tissue composition).
An established human endometrial stromal cell line (ThESC, ATCC, USA) was used in this study. Both ESCs and ThESCs were cultured in DMEM/Ham’s F-12 containing 10% foetal bovine serum (FBS) in a humidified incubator containing 5% CO2 at 37°C. The medium was changed every other day. Before steroid treatment, ESCs and ThESCs were starved for 24 h with phenol red-free and serum-free DMEM/Ham’s F-12 in order to remove endogenous steroids. Then, the medium was changed again to serum-free DMEM/Ham’s F-12 at the time of steroid administration.
Reagents
The 17β-oestradiol (E2, Sigma-Aldrich) was dissolved in DMSO, stored at 80°C and diluted in PBS before use. ICI-182780 (ICI, Cayman Chemicals) was dissolved in ethanol, stored at 80°C and diluted in PBS before use.
In vitro treatments with E2 or ICI
As simultaneous administration of E2 and ICI would probably interfere with the patterns of genes expressed upon ICI treatment, cells were incubated in serum-free medium in the presence of 10−5 mol/L ICI during starvation. After 24 h of serum starvation, the ESCs were maintained in serum-free medium with E2 at 10−8 mol/L. The concentrations of E2 and ICI used in this study were based on previous studies (Wei et al. 2012, Zhang et al. 2016a ). PBS + DMSO and PBS + ethanol were used as the vehicle control treatments for E2 and ICI, respectively. Each experiment was carried out in triplicate and repeated at least three times.
Western blot analysis
Equal amounts of proteins extracted from tissue samples (30 µg) and cell samples (90 µg) were loaded and subjected to 8% SDS-PAGE, and the following steps were performed as previously described (Zhang et al. 2016a ). The primary antibodies used in these experiments were JAG1 (ab85763, 1.25 µg/mL, Abcam), NOTCH1 (ab65297, 1.6 µg/mL, Abcam), VEGF (19003-1-AP, 0.73 µg/mL, Proteintech), MMP9 (10375-2-AP, 0.55 µg/mL, Proteintech), N1ICD (AF5307, 1 µg/mL, Affinity, USA) and GAPDH (AF7021, 1 µg/mL, Affinity). Optical densities (ODs) were measured by ImageJ (Rawak Software, Inc. Germany). Expression was calculated as the ratio of the target OD to the GAPDH OD in each case.
Immunohistochemical staining
Formalin-fixed paraffin-embedded tissues were sectioned at 4 µm thickness using routine deparaffinization and rehydration procedures. The sections were stained as previously described (Zhang et al. 2016a ). The primary antibodies used in these experiments were JAG1 (#70109, 0.5 μg/mL, CST, USA), NOTCH1 (#3608, 4.2 μg/mL, CST) and N1ICD (AF5307, 5 μg/mL, Affinity). Sections incubated without the primary antibody served as negative controls, while sections obtained from squamous cell lung cancer, cervical squamous carcinoma and prostatic carcinoma that positively expressed JAG1, NOTCH1 and N1ICD respectively were used as positive controls as shown in Supplementary Fig. 1 (see section on supplementary data given at the end of this article).
Immunofluorescence
Primary stromal cells were seeded on glass coverslips placed in six-well plates. After the cells were treated with E2 for 24 h, they were fixed with 4% paraformaldehyde for 15 min, permeabilized in 0.05% Triton X-100 (Sigma) for 30 min and blocked with 5% BSA containing 0.01% Triton X-100 for 1 h at room temperature. Then, the cells were incubated with anti-N1ICD antibodies (10 µg/mL) at 4°C overnight. The next day, the cells were washed and incubated with CY3-conjugated secondary antibodies (GB21303, 1:50, Google Biological Technology, China) for 1 h at 37°C in the dark. Nuclear counterstaining was applied with DAPI (Sigma) for 5 min at 37°C, and the cells were then washed, mounted with fluorescence-quenching inhibitor and examined to determine the N1ICD distribution under an Olympus FV1000 laser-scanning confocal microscope (Olympus).
Immunocytochemistry
After 24 h of treatment with E2, the cells were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.1% Triton X-100 for 10 min. Endogenous peroxidases were blocked with 3% H2O2. After the cells were blocked with 10% goat serum, they were incubated with anti-N1ICD antibodies (10 µg/mL) at 4°C overnight. Then, the cells were incubated with a secondary antibody at 37°C for 30 min, stained with 3,30-diaminobenzidine (DAB) for 5 min and counterstained with alcohol-free haematoxylin. After the cells were washed with PBS, they were visualized by light microscopy to determine the distribution of N1ICD labelling.
Migration and invasion assays
Transwell units (24-well with 8.0-mm pore-size polycarbonate filters; Corning Costar, Tewksbury, MA, USA) were used for migration and invasion assays. The approach used for the invasion assay was identical to that used for the migration assay, except that the chambers were precoated with 100 µL of 1:3 diluted Matrigel (Sigma-Aldrich) and pre-hydrated in serum-free medium. After the cells (ESCs or ThESCs) were detached in 10% trypsin and counted, they were diluted to 1 × 105/mL in serum-free medium. A total of 200 µL of cells were plated in each upper chamber with or without E2 while 500 µL of complete medium containing 10% FBS was added to each lower chamber. All assays were performed at 37°C in humidified air with 5% CO2 and 95% O2. After 24 h (migration assay) or 48 h (invasion assay) of incubation, the cells (together with Matrigel for the invasion assay) on the upper surface of the membrane were removed, and the migrated or invaded cells on the underside of the membrane were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Then, the number of these cells was counted in 5 randomly selected fields per well at 200× magnification by light microscopy (Olympus), and the counts were averaged.
In the ICI group, ESCs were pre-treated as previously described. Similarly, to confirm that Notch signalling mediated oestrogen-induced cell invasion, ThESCs were transfected with control shRNA or NOTCH1 shRNA 24 h before the migration and invasion assays. The transfection method is described below.
NOTCH1 knockdown and plasmid transfection
The NOTCH1 shRNA and control shRNA were purchased from GeneChem (China). Four pairs of oligonucleotides encoding shRNAs for NOTCH1 were designed (data not shown), and knockdown efficiency was examined by Western blot analysis. The shRNA with the best knockdown efficiency was used for subsequent experiments. The following were the most effective sequences selected in the pilot experiment: (NOTCH1 shRNA) 5′-CCGGGCCGAACCAATACAACCCTCTCTCGAGAGAGGGTTGTATTGGTTCGGCTTTTTG-3′ (sense) and 5′-AATTCAAAAAGCCGAACCAATACAACCCTCTCTCGAGAGAGGGTTGTATTGGTTCGGC-3′ (antisense).
For plasmid transfection, 1.3 × 106 ThESCs were seeded in each well of a six-well plate. At 40–60% confluence, the cells were transiently transfected with the NOTCH1 shRNA construct or shRNA control vector using Neofect (Neofect Beijing Biotech Co. Ltd, China) according to the manufacturer’s instructions. Transfection efficiency was tracked by detecting the green fluorescent protein (GFP) co-expressed by the GV248 vector. The cells were cultured for 24 h after transfection and subsequently incubated with or without E2 for another 24 h in serum-free medium before collection. Each experiment was carried out in triplicate and repeated three times.
RNA extraction and real-time PCR
Total RNA extraction, cDNA synthesis and quantitative real-time PCR were performed as previously described (Xiong et al. 2016). The primer sequences are presented in Supplementary Table 1. The PCR amplification products were analysed by melting curve and 1.5% agarose gel electrophoresis, followed by sequencing to verify their authenticity. ACTB and GAPDH were used as internal controls and the data were normalized to the geometric mean of ACTB and GAPDH. Quantitation of target genes was performed using the 2−ΔΔt method. All real-time PCRs were performed in triplicate and repeated at least three times.
Luciferase reporter assay
We searched for oestrogen-responsive elements (EREs) using the JASPAR Database in the 3000 bp regions located in the 5′-flanking regions of the JAG1 and NOTCH1 genes. One ERE and three EREs were found in the JAG1 and NOTCH1 genes, respectively: the JAG1 gene contained GCCCAGGGTGAGCACGCCCT, and the NOTCH1 gene contained GGCCAATGTCAGGGCCT (N1a), TCCTTAGCTCACCCTGACAA (N1b) and AGGCACGGGCACCCTGTGCC (N1c). The following series of constructs was generated: the wild-type JAG1 (FLJ1) and NOTCH1 promoters (FLN1), an ERE mutation identified in the JAG1 promoter (MuJ1) and the ERE deletions identified in the NOTCH1 promoter (N1a, N1b, N1c). They were cloned into the pGL3 luciferase reporter plasmid.
ThESCs were plated in 24-well plates 1 day before transfection. At 40–60% confluence, the cells were transfected with the indicated luciferase reporter constructs or the empty vector, as previously described. A pRL-SV40 (Promega) control vector containing the Renilla luciferase gene was also co-transfected at the same time for normalization of luciferase activity. Twenty-four hours after transfection, E2 was added. After another 24 h, cells were collected for the luciferase assay, which was performed using a Dual-Luciferase Assay Kit (GN201-01, YHP-Bio, China), and measurements were made on a GloMax 20/20 luminometer (Promega). These experiments were performed in triplicate and repeated three times.
Statistical analysis
The trial number is indicated in the figure legends. All statistical analyses were performed using GraphPad InStat 7.0 (GraphPad Software, Inc). The unpaired Student’s t test was used to identify significant differences between two groups, and one-way ANOVA with Tukey’s post hoc analysis was used for comparisons between three or more groups. All data sets are shown as the mean ± standard deviation (s.d.) of three independent experiments. A P value <0.05 was considered statistically significant.
Results
Notch components and activity are increased in EMS
JAG1 and NOTCH1 are the ligand and receptor, respectively, of the Notch family. Their binding leads to the activation of Notch signalling, which can be quantified according to N1ICD levels (Bray 2006, Pancewicz & Nicot 2011). Therefore, to assess the role of Notch signalling in EMS, the protein levels of JAG1, NOTCH1 and N1ICD were analysed using Western blot in normal (control), eutopic (Eu) and ectopic (Ec) endometria. The results showed that the expression levels of JAG1 and NOTCH1 and the level of N1ICD were significantly higher in the Ec and Eu groups than in the control group (n = 10, Fig. 1). Additionally, all three were present at higher levels in the Ec group than in the Eu group (n = 10, Fig. 1). These results suggest that Notch signalling is hyperactivated in EMS and might therefore be associated with the development of EMS.
JAG1, NOTCH1 and NOTCH1 intracellular domain (N1ICD) levels are significantly higher in eutopic and ectopic endometria in endometriosis than in normal endometria. (A) Left: Western blot analysis of JAG1, NOTCH1 and N1ICD protein levels in normal endometria obtained from patients (n = 10) with tubal infertility and paired eutopic and ectopic endometria obtained from patients (n = 10) with endometriosis. GAPDH was used as the internal control. Right: Bars indicate relative expression levels after normalization to GAPDH. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA. Normal, normal endometria; Eu, eutopic endometria; Ec, ectopic endometria; N1ICD, NOTCH1 intracellular domain. (B) Representative photomicrograph of immunohistochemical staining for the JAG1 (left), NOTCH1 (middle) and N1ICD (right) proteins in human endometriotic endometrium.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
To further examine the cell-specific expression of Notch components and distribution of N1ICD in EMS, we next performed an immunohistochemical analysis. Representative images of immunostaining for JAG1, NOTCH1 and N1ICD are shown in Fig. 1B. Both NOTCH1 and JAG1 were predominantly expressed in epithelial cells and weakly expressed in stromal cells in endometriotic lesions. In contrast, N1ICD was highly expressed in both epithelial cells and stromal cells. For intracellular localization, both JAG1 and NOTCH1 were localized at the cell membrane, while N1ICD was localized in cell plasma and predominantly in cell nucleus. These results suggest that Notch activation may be associated with the role of stromal cells in EMS.
Oestrogen enhances Notch signalling and the invasiveness of ESCs
Since oestrogen concentration is elevated in endometriotic lesions (Bulun et al. 2000) and oestrogen plays a role in regulating Notch signalling in other diseases, we next examined the changes in Notch components and activity in response to exogenous E2 in normal primary ESCs. After treatment with E2, the protein levels of JAG1, NOTCH1 and N1ICD were all strongly upregulated in a time-dependent manner (Fig. 2A), indicating that increased levels of Notch components and activity in endometriotic lesions maybe associated with the local high concentration of E2. Together with the previous observation that oestrogen promoted JAG1-induced Notch signalling activity (Soares et al. 2004), our findings indicate that E2 may enhance Notch activity by increasing the levels of Notch components and NOTCH1 cleavage.
Oestrogen increases Notch activity (assessed as the N1ICD level), the protein levels of JAG1, NOTCH1, MMP9 and VEGF and the intracellular distribution of N1ICD. (A) The protein levels of JAG1, NOTCH1, N1ICD, MMP9 and VEGF by time course analysis after treatment with 17β-oestradiol (E2) as determined by Western blot. The bars indicate relative expression levels after normalization to GAPDH. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA. n.s., not significant. (B) Immunofluorescence showing the subcellular localization of N1ICD in the control (PBS + DMSO-treated) and E2-treated groups. (C) Immunohistochemistry showing the subcellular localization of N1ICD in the control and E2-treated groups. The scale bars represent 10, 5 and 2.5 µm, as indicated.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
To further examine the effect of oestrogen on Notch activity, we next performed immunofluorescence and immunochemistry to morphologically define the cellular distribution of N1ICD. The immunofluorescence results showed that N1ICD was detected in a punctate pattern that was distributed in the cytoplasm and nucleus (Fig. 2B). In addition, exposure to E2 induced a striking increase in the signal intensity in both locations, especially in the nucleus. Immunochemical analysis revealed a similar pattern, and within the cytoplasm, the increase in the accumulation of N1ICD was observed mainly around the nucleus (Fig. 2C), suggesting an increase in the translocation of N1ICD into the nucleus after E2 treatment.
To explore the role of oestrogen-induced Notch signalling activation in the regulation of ESC invasiveness, we simultaneously examined MMP9 and VEGF expression and cell invasiveness following exposure to oestrogen. Consistent with our previous findings (Zhang et al. 2016a ,b ), Western blot analysis showed that the protein levels of MMP9 and VEGF were significantly higher after E2 treatment (Fig. 2A). Migration and invasion assays also showed that more cells migrated and invaded when they were incubated with E2 (Fig. 3). These data suggest that oestrogen-induced Notch activity may be correlated with oestrogen-induced ESC invasiveness.
E2 enhances the migratory capacity and invasiveness of endometrial stromal cells (ESCs). Left: Representative photographs of ESC migration and invasion in the control (PBS + DMSO-treated) and E2-treated groups. Original magnification, 200×. Right: Quantified results of the migration and invasion assays. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by unpaired Student’s t-test.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
Knockdown of NOTCH1 alleviates the E2-mediated migration and invasion of ThESCs
To confirm our hypothesis that Notch signalling may mediate the effects of oestrogen on the migration and invasion of ESCs, shRNA-mediated RNA interference was utilized to suppress NOTCH1 expression in ThESCs. ThESCs were pre-transfected with NOTCH1 shRNA or control shRNA before treatment with E2. The results of Western blot analysis showed that knockdown of NOTCH1 significantly lowered the E2-induced expression of MMP9 and VEGF proteins (Fig. 4A) and resulted in low-level migration and invasion in cells incubated with E2 (Fig. 4B and C). This observation suggests that Notch activity is indeed involved in the regulation of E2-induced ThESC invasiveness. Notably, even in the absence of E2, NOTCH1 knockout still reduced MMP9 and VEGF expression and cell invasiveness, consistent with the results reported in the literature (Wang et al. 2006, Zhou et al. 2012) and implying that Notch signalling itself has a controlling effect on endometrial cell invasion in the absence of E2.
Knockdown of NOTCH1 by shRNA transfection alleviates oestrogen-mediated MMP9, VEGF expression and ThESC migration and invasion. (A) ThESCs were pre-transfected with scramble shRNA or NOTCH1 shRNA. NOTCH1 knockdown inhibited E2-induced MMP9 and VEGF expression determined by Western blot analysis. Bars indicate relative expression levels after normalization to GAPDH. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA. (B) Representative photographs of ThESC migration and invasion after transfection with scrambled shRNA or NOTCH1 shRNA in the presence or absence of E2. (C) Bars correspond to the quantified results of the migration and invasion assays. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
Oestrogen-induced Notch activity and cell invasiveness are mediated by oestrogen receptor alpha (ERα)
Oestrogen activity is primarily mediated by oestrogen nuclear receptors (Dutertre & Smith 2000, Thomas & Gustafsson 2011), with ERα playing a pivotal role in the pathogenesis of EMS (Burns et al. 2012). Therefore, the ERα antagonist ICI was used to characterize the molecular mechanism of oestrogen in the regulation of Notch signalling and cell invasiveness. Both PCR and Western blot analyses showed that ICI inhibited the expression of JAG1 and NOTCH1 and reduced the levels of N1ICD that were induced by E2 (Fig. 5A and B). Similarly, the E2-mediated increases in MMP9 and VEGF expression and cell migration and invasion were also blocked by ICI (Fig. 5A, B and C). These data suggest that nuclear ERα is essential for E2 actions in terms of Notch signalling and endometrial stromal cell invasiveness.
ERα mediates the oestrogen-induced activation of Notch signalling, the upregulation of JAG1, NOTCH1, MMP9 and VEGF, and the enhancement of migration and invasion of ESCs. (A and B) JAG1, NOTCH1, MMP9 and VEGF mRNA (A) and protein (B) levels together with N1ICD protein levels (B) were analysed after E2 and ICI treatment by real time-PCR (A) and Western blot (B) analysis. Bars indicate relative expression levels after normalization to GAPDH. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA. (C) Representative photographs of migration and invasion of ESCs treated with or without ICI (left). Bars show the quantified results of migration and invasion assays (right). Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
E2 upregulates Notch signalling components by activating oestrogen receptor elements (EREs)
Nuclear ERs usually act as transcription factors by binding to EREs in downstream target genes to regulate their expression (Dutertre & Smith 2000, Thomas & Gustafsson 2011). In this study, RT-PCR analysis indicated that the effect of E2 on Notch components was mediated by transcriptional regulation (Fig. 5A). Whether oestrogen activates the transcription of Notch signalling components by binding to EREs is unknown. Therefore, to further investigate the potential molecular link between E2 and Notch signalling, luciferase assays were conducted. One and three EREs were found in the 5′-flanking regions of the JAG1 and NOTCH1 genes, respectively (Fig. 6A). E2 stimulated an increase in luciferase activity from FLJ1 (full-length JAG1 promoter from −2970 to −2951 bp), which contains a putative ERE, but not from MuJ1 (full-length JAG1 promoter with mutated ERE) (Fig. 6B), suggesting that the region from −2970 to −2951 containing the putative ERE in the JAG1 promoter is oestrogen responsive. Additionally, E2 also induced luciferase activity from the FLN1 construct (full-length NOTCH1 promoter from −3000 to 0 bp) (Fig. 6A and C), which contains three putative EREs (from −70 to −51, −89 to −70, and −1870 to −1851 bp). When any of the three putative EREs was deleted, luciferase activity was no longer stimulated by E2, indicating that all three putative EREs in the NOTCH1 promoter were needed for oestrogen responsiveness. Altogether, these results indicate that E2 upregulates the NOTCH1 and JAG1 genes via nuclear ERα interactions with JAG1 and NOTCH1 promoter regions that contain putative EREs.
The role of oestrogen in the activation of EREs in the promoters of JAG1 and NOTCH1 was assessed by luciferase assays. (A) Constructs: Full-length JAG1 promoter containing the putative ERE (FLJ1) and mutated ERE (MuJ1), full-length NOTCH1 promoter (FLN1) containing three putative EREs, and FLN1 with EREs deletion (N1a, N1b and N1c, respectively). (B and C) Luciferase activity after transfection of different constructs. Mean values were obtained from n = 3 independent experiments; *P < 0.05 by one-way ANOVA.
Citation: Reproduction 157, 4; 10.1530/REP-18-0326
Discussion
Oestrogen and Notch signalling engage in crosstalk to regulate a wide range of biological functions, such as tumour angiogenesis (Soares et al. 2004) and neuronal development (Ruiz-Palmero et al. 2011). However, whether the two signals are correlated in terms of cell invasiveness is unknown. Notch signalling has been recently demonstrated to be associated with fibrosis and angiogenesis in EMS. Whether it has a relationship with the invasion and metastasis of EMS is also unknown. In the current study, we found that Notch signalling was involved in the oestrogen-induced migration and invasion of ESCs, and these processes might facilitate the establishment and metastasis of EMS. In normal ESCs, oestrogen activates Notch signalling, upregulates its components (JAG1 and NOTCH1) in an ERα-dependent manner and transcriptionally regulates JAG1 and NOTCH1 expression by binding EREs in the promoters of these two genes. Notch signalling additionally mediates oestrogen-induced cell invasion by modulating the expression of MMP9 and VEGF.
JAG1 and NOTCH1 overexpression and high Notch activity are usually associated with increased cell invasiveness (Mitsuhashi et al. 2012, Bednarz-Knoll et al. 2016). In EMS, the migration and invasion capacities of ectopic ESCs are stronger than those of eutopic and normal ESCs (Guan et al. 2016). Our findings that JAG1 and NOTCH1 expression and Notch activity were increased in endometriotic lesions (Fig. 1A) and that N1ICD was highly expressed in ectopic ESCs (Fig. 1B) suggest that Notch signalling may be associated with the enhanced invasiveness of ectopic ESCs. Moreover, our study provides laboratory evidence showing that oestrogen can upregulate the expression of JAG1, NOTCH1 and activate Notch signalling. Therefore, these data imply that the hyperactivation of Notch signalling in ectopic ESCs may be attributed to the locally high oestrogen concentrations observed in EMS (Huhtinen et al. 2012). Additionally, Notch activation may result from the combined effects of the upregulation of NOTCH1 and JAG1 expression and NOTCH1 cleavage since the binding of NOTCH1 to JAG1 leads to the activation of Notch signalling and NOTCH1 cleavage.
The role of oestrogen in the regulation of Notch signalling and cell invasion depends on its receptors. The predominant oestrogen receptors in endometriotic lesions are ERα and ERβ. There is some controversy in the literature regarding the role of ERα in the regulation of Notch signalling. Soares and his colleagues found that oestrogen activated Notch signalling through ERα (Soares et al. 2004), while Rizzo et al. found that oestrogen inhibited Notch signalling via ERα by blocking the cleavage of NOTCH1 (Rizzo et al. 2008). Our data support the report by Soares. We are the first to provide evidence showing that ERα plays a facilitative role in modulating the invasiveness of ESCs by activating Notch signalling. We also directly measured N1ICD protein levels instead of NOTCH1 and thereby avoided the measurement limitations associated with reporter assays or antibodies for NOTCH1 in Soares’ and Rizzol’s studies, respectively. In those reports, it is possible that Notch activation might have been masked by increased transfection by E2 in reporter assays (Soares et al. 2004) or that the NOTCH1 antibody might not have recognized NOTCH1 after cleavage (Rizzo et al. 2008). Nevertheless, one cannot ignore that many studies have demonstrated that ERα levels are decreased while ERβ levels are dramatically increased in ectopic lesions and ESCs in EMS (Xue et al. 2007, Bulun et al. 2012) in spite of our confirmation of the positive role of ERα in cell invasion. Compared to normal endometrium, in eutopic endometrium, ERα levels were also found to be higher or unchanged (Prentke et al. 1992, Cavallini et al. 2011, Pellegrini et al. 2012, Zhang et al. 2016b ). Although ERβ frequently exerts a suppressive effect on cell invasion in tumours (Thomas & Gustafsson 2011), it was recently confirmed to promote the invasion of epithelial cells via the involvement in TNF-α induced epithelial mesenchymal transition (EMT) in EMS. Therefore, we speculated that the upregulation of ERβ in EMS may exert compensatory effects for ERα decrease to promote cell invasion, which leads to more invasive ectopic ESCs. Further investigations into the effect of ERβ on Notch signalling and cell invasion in EMS are needed in the future.
Nuclear ERs classically act as transcription factors by binding to the EREs of target genes to regulate their expression. Our study confirmed that oestrogen could activate NOTCH1 and JAG1 transcription by binding to the EREs in their promoters. This result was consistent with Soares’ observation for JAG1 (Soares et al. 2004). However, in our study, we found three EREs for NOTCH1 that worked cooperatively towards transcriptional activation, while no functional EREs were found in Soares’ study. The difference in these results may be due to differences in the EREs that were found or differences in cell context. Moreover, oestrogen had a stronger pro-transcription effect on JAG1 than on NOTCH1 (Figs 5A, 6B and C), even though more EREs were found in the NOTCH1 gene, indicating that other regulatory mechanism may be involved in the ERα-mediated regulation of JAG1. For example, nuclear ERs can regulate target genes by recruiting coactivators that bind to the promoters of target genes in the absence of ERE binding (Bjornstrom & Sjoberg 2005).
MMP9 and VEGF are well-known metastasis-related genes that regulate cell invasion by promoting the degradation of ECM and interacting with receptors expressed by tumour cells respectively. Moreover, the fragments resulted from the MMP9-mediated cleavage of ECM can also regulate cell invasion directly or affect other cellular signalling mechanisms that modulate cell invasion (Egeblad & Werb 2002). Both MMP9 and VEGF are downstream effectors of Notch signalling in the regulation of tumour invasion (Wang et al. 2006, Hafeez et al. 2009). NOTCH1 might directly regulate the expression of MMP9 by enhancing its promoter activity or affect MMP9 level via a posttranslational stabilization mechanism (Hafeez et al. 2009). Our group previously confirmed that MMP9 and VEGF are involved in the oestrogen-induced invasiveness of ESCs in EMS (Xiong et al. 2015). The results of this study are consistent with our previous results. Moreover, knockdown of NOTCH1 abrogated oestrogen-induced cell invasion and upregulation of MMP9 and VEGF, indicating that oestrogen regulates the invasion of ESCs via Notch signalling and the expression of MMP9 and VEGF. Notably, Notch signalling alone can regulate the invasiveness of ESCs via MMP9 and VEGF. Thus, we provide evidence showing that oestrogen enhances the invasive capacity of ESCs by activating Notch signalling to upregulate the expression of MMP9 and VEGF, thereby leading to increased invasiveness in endometrial cells.
Our finding that Notch activity mediated oestrogen-induced cell invasion in EMS also has clinical implications. It is well known that medical treatments based on oestrogen depletion are usually associated with high recurrence rates and unexpected side effects. Combinatorial treatment, namely oestrogen depletion concomitant with Notch blockage, may produce synergistic effects and efficiently inhibit the development and metastasis of EMS. It may also be possible that Notch inhibitors or other knockdown techniques are, on their own, alternative nonhormonal treatments for EMS. Moreover, the expression levels of key Notch molecules are higher in ectopic than in normal endometria, indicating that Notch components could be utilized as biomarkers for the diagnosis of EMS.
Supplementary data
This is linked to the online version of the paper at https://doi.org/10.1530/REP-18-0326.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the National Natural Science Foundation of China (Grant No. 81471439) and National Youth Fund (Grant No. 81701417).
Acknowledgements
The authors appreciate the editing assistance provided by Professor Zhibing Zhang to an earlier version of this manuscript.
References
Bednarz-Knoll N, Efstathiou A, Gotzhein F, Wikman H, Mueller V, Kang Y & Pantel K 2016 Potential involvement of jagged1 in metastatic progression of human breast carcinomas. Clinical Chemistry 62 378–386. (https://doi.org/10.1373/clinchem.2015.246686)
Bjornstrom L & Sjoberg M 2005 Mechanisms of estrogen receptor signaling: convergence of genomic and nongenomic actions on target genes. Molecular Endocrinology 19 833–842. (https://doi.org/10.1210/me.2004-0486)
Bourlev V, Volkov N, Pavlovitch S, Lets N, Larsson A & Olovsson M 2006 The relationship between microvessel density, proliferative activity and expression of vascular endothelial growth factor-A and its receptors in eutopic endometrium and endometriotic lesions. Reproduction 132 501–509. (https://doi.org/10.1530/rep.1.01110)
Bray SJ 2006 Notch signalling: a simple pathway becomes complex. Nature Reviews Molecular Cell Biology 7 678–689. (https://doi.org/10.1038/nrm2009)
Bray SJ 2016 Notch signalling in context. Nature Reviews Molecular Cell Biology 17 722–735. (https://doi.org/10.1038/nrm.2016.94)
Brown DM, Lee HC, Liu S, Quick CM, Fernandes LM, Simmen FA, Tsai SJ & Simmen RCM 2018 Notch-1 signaling activation and progesterone receptor expression in ectopic lesions of women with endometriosis. Journal of the Endocrine Society 2 765–778. (https://doi.org/10.1210/js.2018-00007)
Bulun SE, Zeitoun KM, Takayama K & Sasano H 2000 Estrogen biosynthesis in endometriosis: molecular basis and clinical relevance. Journal of Molecular Endocrinology 25 35–42. (https://doi.org/10.1677/jme.0.0250035)
Bulun SE, Monsavais D, Pavone ME, Dyson M, Xue Q, Attar E, Tokunaga H & Su EJ 2012 Role of estrogen receptor-beta in endometriosis. Seminars in Reproductive Medicine 30 39–45. (https://doi.org/10.1055/s-0031-1299596)
Burns KA, Rodriguez KF, Hewitt SC, Janardhan KS, Young SL & Korach KS 2012 Role of estrogen receptor signaling required for endometriosis-like lesion establishment in a mouse model. Endocrinology 153 3960–3971. (https://doi.org/10.1210/en.2012-1294)
Capaccione KM & Pine SR 2013 The Notch signaling pathway as a mediator of tumor survival. Carcinogenesis 34 1420–1430. (https://doi.org/10.1093/carcin/bgt127)
Cavallini A, Resta L, Caringella AM, Dinaro E, Lippolis C & Loverro G 2011 Involvement of estrogen receptor-related receptors in human ovarian endometriosis. Fertility and Sterility 96 102–106. (https://doi.org/10.1016/j.fertnstert.2011.04.032)
da Costa e Silva Rde C, Moura KK, Ribeiro Júnior CL & Guillo LA 2016 Estrogen signaling in the proliferative endometrium: implications in endometriosis. Revista da Associacao Medica Brasileira 62 72–77. (https://doi.org/10.1590/1806-9282.62.01.72)
Di Carlo C, Bonifacio M, Tommaselli GA, Bifulco G, Guerra G & Nappi C 2009 Metalloproteinases, vascular endothelial growth factor, and angiopoietin 1 and 2 in eutopic and ectopic endometrium. Fertility and Sterility 91 2315–2323. (https://doi.org/10.1016/j.fertnstert.2008.03.079)
Dong C, Ye DX, Zhang WB, Pan HY, Zhang ZY & Zhang L 2015 Overexpression of c-fos promotes cell invasion and migration via CD44 pathway in oral squamous cell carcinoma. Journal of Oral Pathology and Medicine 44 353–360. (https://doi.org/10.1111/jop.12296)
Dutertre M & Smith CL 2000 Molecular mechanisms of selective estrogen receptor modulator (SERM) action. Journal of Pharmacology and Experimental Therapeutics 295 431–437.
Egeblad M & Werb Z 2002 New functions for the matrix metalloproteinases in cancer progression. Nature Reviews Cancer 2 161–174. (https://doi.org/10.1038/nrc745)
Emonard H & Grimaud JA 1990 Matrix metalloproteinases. A review. Cellular and Molecular Biology 36 131–153. (https://doi.org/10.1177/10454411930040020401)
Fan JZ, Yang L, Meng GL, Lin YS, Wei BY, Fan J, Hu HM, Liu YW, Chen S & Zhang JK et al. 2014 Estrogen improves the proliferation and differentiation of hBMSCs derived from postmenopausal osteoporosis through notch signaling pathway. Molecular and Cellular Biochemistry 392 85–93. (https://doi.org/10.1007/s11010-014-2021-7)
Ferrero S, Remorgida V, Maganza C, Venturini PL, Salvatore S, Papaleo E, Candiani M & Maggiore ULR 2014 Aromatase and endometriosis: estrogens play a role. Annals of the New York Academy of Sciences 1317 17–23. (https://doi.org/10.1111/nyas.12411)
Giudice LC & Kao LC 2004 Endometriosis. Lancet 364 1789–1799. (https://doi.org/10.1016/S0140-6736(04)17403-5)
Gonzalez-Foruria I, Santulli P, Chouzenoux S, Carmona F, Chapron C & Batteux F 2017 Dysregulation of the ADAM17/Notch signalling pathways in endometriosis: from oxidative stress to fibrosis. Molecular Human Reproduction 23 488–499. (https://doi.org/10.1093/molehr/gax028)
Guan YT, Huang YQ, Wu JB, Deng ZQ, Wang Y, Lai ZY, Wang HB, Sun XX, Zhu YL & Du MM et al. 2016 Overexpression of chloride channel-3 is associated with the increased migration and invasion ability of ectopic endometrial cells from patients with endometriosis. Human Reproduction 31 986–998. (https://doi.org/10.1093/humrep/dew034)
Hafeez BB, Adhami VM, Asim M, Siddiqui IA, Bhat KM, Zhong W, Saleem M, Din M, Setaluri V & Mukhtar H 2009 Targeted knockdown of Notch1 inhibits invasion of human prostate cancer cells concomitant with inhibition of matrix metalloproteinase-9 and urokinase plasminogen activator. Clinical Cancer Research 15 452–459. (https://doi.org/10.1158/1078-0432.CCR-08-1631)
Huhtinen K, Desai R, Stahle M, Salminen A, Handelsman DJ, Perheentupa A & Poutanen M 2012 Endometrial and endometriotic concentrations of estrone and estradiol are determined by local metabolism rather than circulating levels. Journal of Clinical Endocrinology and Metabolism 97 4228–4235. (https://doi.org/10.1210/jc.2012-1154)
Kitawaki J, Kado N, Ishihara H, Koshiba H, Kitaoka Y & Honjo H 2002 Endometriosis: the pathophysiology as an estrogen-dependent disease. Journal of Steroid Biochemistry and Molecular Biology 83 149–155. (https://doi.org/10.1016/S0960-0760(02)00260-1)
Li Y, Wang X, Wang X, Wan L, Liu Y, Shi Y, Zhang L, Fang Z & Wei Z 2018 PDCD4 suppresses proliferation, migration, and invasion of endometrial cells by inhibiting autophagy and NF-kappaB/MMP2/MMP9 signal pathway. Biology of Reproduction 99 360–372. (https://doi.org/10.1093/biolre/ioy052)
Lichtenberger BM, Tan PK, Niederleithner H, Ferrara N, Petzelbauer P & Sibilia M 2010 Autocrine VEGF signaling synergizes with EGFR in tumor cells to promote epithelial cancer development. Cell 140 268–279. (https://doi.org/10.1016/j.cell.2009.12.046)
Liu H, Wang J, Wang H, Tang N, Li Y, Zhang Y & Hao T 2015 Correlation between matrix metalloproteinase-9 and endometriosis. International Journal of Clinical and Experimental Pathology 8 13399–13404.
Machado DE, Berardo PT, Palmero CY & Nasciutti LE 2010 Higher expression of vascular endothelial growth factor (VEGF) and its receptor VEGFR-2 (Flk-1) and metalloproteinase-9 (MMP-9) in a rat model of peritoneal endometriosis is similar to cancer diseases. Journal of Experimental and Clinical Cancer Research 29 4. (https://doi.org/10.1186/1756-9966-29-4)
Mitsuhashi Y, Horiuchi A, Miyamoto T, Kashima H, Suzuki A & Shiozawa T 2012 Prognostic significance of Notch signalling molecules and their involvement in the invasiveness of endometrial carcinoma cells. Histopathology 60 826–837. (https://doi.org/10.1111/j.1365-2559.2011.04158.x)
Pan H, Zhang P, Li JR, Wang H, Jin MF, Feng C & Huang HF 2016 c-Fos-regulated matrix metalloproteinase-9 expression is involved in 17β-estradiol-promoted invasion of human endometrial stromal cell. Current Molecular Medicine 16 266–275. (https://doi.org/10.2174/1566524016666160225153454)
Pancewicz J & Nicot C 2011 Current views on the role of Notch signaling and the pathogenesis of human leukemia. BMC Cancer 11 502. (https://doi.org/10.1186/1471-2407-11-502)
Paw I, Carpenter RC, Watabe K, Debinski W & Lo HW 2015 Mechanisms regulating glioma invasion. Cancer Letters 362 1–7. (https://doi.org/10.1016/j.canlet.2015.03.015)
Pellegrini C, Gori I, Achtari C, Hornung D, Chardonnens E, Wunder D, Fiche M & Canny GO 2012 The expression of estrogen receptors as well as GREB1, c-MYC, and cyclin D1, estrogen-regulated genes implicated in proliferation, is increased in peritoneal endometriosis. Fertility and Sterility 98 1200–1208. (https://doi.org/10.1016/j.fertnstert.2012.06.056)
Prentke A, Randall BJ, Weddell A, McGill A, Henry L, Horne CH & Thomas EJ 1992 Ovarian steroid receptor expression in endometriosis and in two potential parent epithelia: endometrium and peritoneal mesothelium. Human Reproduction 7 1318–1325. (https://doi.org/10.1093/oxfordjournals.humrep.a137848)
Rizzo P, Miao H, D'Souza G, Osipo C, Song LL, Yun J, Zhao H, Mascarenhas J, Wyatt D & Antico G et al. 2008 Cross-talk between notch and the estrogen receptor in breast cancer suggests novel therapeutic approaches. Cancer Research 68 5226–5235. (https://doi.org/10.1158/0008-5472.CAN-07-5744)
Ruiz-Palmero I, Simon-Areces J, Garcia-Segura LM & Arevalo MA 2011 Notch/neurogenin 3 signalling is involved in the neuritogenic actions of oestradiol in developing hippocampal neurones. Journal of Neuroendocrinology 23 355–364. (https://doi.org/10.1111/j.1365-2826.2011.02110.x)
Sampson JA 1927 Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. American Journal of Obstetrics and Gynecology 14 422–469. (https://doi.org/10.1016/S0002-9378(15)30003-X)
Soares R, Balogh G, Guo S, Gartner F, Russo J & Schmitt F 2004 Evidence for the Notch signaling pathway on the role of estrogen in angiogenesis. Molecular Endocrinology 18 2333–2343. (https://doi.org/10.1210/me.2003-0362)
Thomas C & Gustafsson JÅ 2011 The different roles of ER subtypes in cancer biology and therapy. Nature Reviews Cancer 11 597–608. (https://doi.org/10.1038/nrc3093)
Wang Z, Banerjee S, Li Y, Rahman KM, Zhang Y & Sarkar FH 2006 Down-regulation of Notch-1 inhibits invasion by inactivation of nuclear factor-kappaB, vascular endothelial growth factor, and matrix metalloproteinase-9 in pancreatic cancer cells. Cancer Research 66 2778–2784. (https://doi.org/10.1158/0008-5472.CAN-05-4281)
Wei Y, Zhang Z, Liao H, Wu L, Wu X, Zhou D, Xi X, Zhu Y & Feng Y 2012 Nuclear estrogen receptor-mediated Notch signaling and GPR30-mediated PI3K/AKT signaling in the regulation of endometrial cancer cell proliferation. Oncology Reports 27 504–510. (https://doi.org/10.3892/or.2011.1536)
Witz CA, Monotoya-Rodriguez IA & Schenken RS 1999 Whole explants of peritoneum and endometrium: a novel model of the early endometriosis lesion. Fertility and Sterility 71 56–60. (https://doi.org/10.1016/S0015-0282(98)00400-2)
Xiong W, Zhang L, Yu L, Xie W, Man Y, Xiong Y, Liu H & Liu Y 2015 Estradiol promotes cells invasion by activating beta-catenin signaling pathway in endometriosis. Reproduction 150 507–516. (https://doi.org/10.1530/REP-15-0371)
Xiong Y, Liu Y, Xiong WQ, Zhang L, Liu HW, Du Y & Li N 2016 Hypoxia-inducible factor 1α-induced epithelial-mesenchymal transition of endometrial epithelial cells may contribute to the development of endometriosis. Human Reproduction 31 1327–1338. (https://doi.org/10.1093/humrep/dew081)
Xue Q, Lin Z, Cheng YH, Huang CC, Marsh E, Yin P, Milad MP, Confino E, Reierstad S & Innes J et al. 2007 Promoter methylation regulates estrogen receptor 2 in human endometrium and endometriosis. Biology of Reproduction 77 681–687. (https://doi.org/10.1095/biolreprod.107.061804)
Yland J, Mathyk B & Young SL 2018 Steroid hormones and endometriosis. Current Women’s Health Reviews 14 117–126. (https://doi.org/10.2174/1573404813666170621080058)
Yoshida S, Harada T, Mitsunari M, Iwabe T, Sakamoto Y, Tsukihara S, Iba Y, Horie S & Terakawa N 2004 Hepatocyte growth factor/met system promotes endometrial and endometriotic stromal cell invasion via autocrine and paracrine pathways. Journal of Clinical Endocrinology and Metabolism 89 823–832. (https://doi.org/10.1210/jc.2003-030874)
Zhang L, Xiong WQ, Xiong Y, Liu HW, Li N, Du Y & Liu Y 2016a Intracellular Wnt/beta-catenin signaling underlying 17beta-estradiol-induced matrix metalloproteinase 9 expression in human endometriosis. Biology of Reproduction 94 70. (https://doi.org/10.1095/biolreprod.115.135574)
Zhang L, Xiong WQ, Xiong Y, Liu HW & Liu Y 2016b 17Beta-estradiol promotes vascular endothelial growth factor expression via the Wnt/beta-catenin pathway during the pathogenesis of endometriosis. Molecular Human Reproduction 22 526–535. (https://doi.org/10.1093/molehr/gaw025)
Zhou L, Wang DS, Li QJ, Sun W, Zhang Y & Dou KF 2012 Downregulation of the Notch signaling pathway inhibits hepatocellular carcinoma cell invasion by inactivation of matrix metalloproteinase-2 and -9 and vascular endothelial growth factor. Oncology Reports 28 874–882. (https://doi.org/10.3892/or.2012.1880)
