Recepteur d'origine nantais contributes to the development of endometriosis via promoting epithelial‐mesenchymal transition of a endometrial epithelial cells

Abstract Endometriosis is a benign, chronic inflammatory disease that commonly occurs in reproductive‐aged women. Epithelial‐mesenchymal transition (EMT) of endometrial epithelial cells plays an important role in the development of endometriosis. Recepteur d'origine nantais (RON), a receptor tyrosine kinase, has been reported to promote EMT and progression in tumours. However, whether and how RON mediates the EMT and endometriosis development is not known. Here, we found that RON activation could improve the migratory and invasive capabilities, change cellular morphologies, and decrease expression of E‐cadherin and increase expression of N‐cadherin in endometrial epithelial cells. Inhibition or knockdown of RON expression suppressed the migration and invasion of endometrial epithelial cells. Our studies also indicated that RON played its part in endometrial epithelial cells through protein kinase B (Akt) and mitogen‐activated protein kinase (MAPK) pathways. Treatment with a RON inhibitor could decrease the number of ectopic lesions in a mouse model of endometriosis and mediate expression of EMT markers in endometriotic lesions. These data suggest that RON contributed to endometriosis development by promoting EMT of endometrial epithelial cells. Therefore, RON may be a new therapeutic target for endometriosis.

most popular. 4,5 Migration and invasion of endometrial cells underpin the RM theory, and these two processes have been demonstrated to be implicated in epithelial-mesenchymal transition (EMT). [6][7][8][9] During EMT, cells lose their epithelial features and acquire the characteristics of mesenchymal cells. 10,11 When EMT occurs, the expression of mesenchymal markers (eg N-cadherin) is increased.
However, the expression of epithelial molecular markers (eg E-cadherin) is decreased. 12,13 In addition, the expressions of cell polarity (such as Par3) and tight junction (such as Occludin) proteins are decreased, which destroys the epithelial cell polarity and tight junctions between cells. [14][15][16] As a result, the abilities of migration, invasion and anti-apoptosis are increased. 10,11,17 Recently, several scholars have shown that EMT can make the endometrial cells from RM obtain the ability of migration, invasion and anti-apoptosis. These features are beneficial for endometrial implantation in the abdominal cavity, suggesting an important role of EMT in endometriosis development. 18,19 Oestrogen and hypoxia are currently considered to be two inducers of EMT for endometrial cells originating from RM, but how EMT is triggered in endometriosis is still unclear. 20,21 The receptor tyrosine kinase, recepteur d'origine nantais (RON, also known as macrophage-stimulating protein receptor, MST1R), belongs to the MET family of receptors 22 and its ligand is macrophage-stimulating protein (MSP). 23,24 RON is expressed mainly in various types of human epithelial tissues including endometrial epithelial cells. Overexpression of RON has been proved to be associated with the progression, metastasis, survival and prognosis of various types of cancers including cancer of the pancreas, colon, lung, breast and ovary. [25][26][27][28] In women with endometriosis, increased expression of RON has also been found in endometriotic lesions, suggesting that RON is implicated in endometriosis pathogenesis. 29,30 Recent studies have shown that RON activation can promote the migration and invasion of epithelial cancer cells in vitro. [31][32][33] Endometriosis is a benign disease, but it has malignant biological behaviour, including proliferation, migration and invasion. 34 Nevertheless, whether RON contributes to the development and progression of endometriosis by promoting EMT merits investigation.
We hypothesized that RON is involved in endometriosis pathogenesis by promoting the EMT of endometrial cells. First, we measured RON expression in endometriotic lesions from ovarian endometrioma. Second, we induced RON expression in endometrial cells in vitro. Third, we detected the effect of BMS 777607 (RON inhibitor) on endometriosis development in mice. In so doing, we explored the role of RON in endometriosis.

| Ethics statement
The scheme of this study was approved by the Human Ethics Committee of the Women's Hospital, School of Medicine, Zhejiang University in Zhejiang, China (approval number: 20160115). All patients gave written informed consent for their participation.
Animal experiments were carried out according to programmes approved by the ethics committee of Zhejiang University (NO. ZJU20190127).

| Collection of clinical samples
Twenty-eight patients with an ovarian endometrioma (seven women provided their ectopic endometria, twelve women provided their eutopic endometria and nine women provided both of their ectopic and eutopic endometria) and thirteen women without endometriosis who provided control endometria were recruited from Women's Hospital (Zhejiang University, School of Medicine).
Patients with endometriosis had mainly a late-stage (III and IV) ovarian endometrioma. All recruited women had regular menstrual cycles and none had not received any hormonal treatments for ≥3 months at the time of surgery. The endometrial samples used in this study were in the proliferative phase of the menstrual cycle, which was confirmed by the last menstrual period and endometrial histopathology.

| Immunohistochemical (IHC) staining
Endometrial tissues were preserved in 4% formaldehyde and embedded in paraffin. Then they were cut into sections (thickness = 3 μm) using a microtome. After dewaxing, sections were incubated with antibodies to measure expression of RON and EMT markers in endometrial tissues: anti-RON (1:600; ab52927; Abcam), anti-Par3 were examined by two independent observers unaware of the sample's background. IHC scores were utilized based on the previously studies, 35,36 0 represented no staining; one represented staining of 1%-25%; two represented staining of 25%-50%; three represented staining of 50%-100%. Also, staining intensity was graded: zero represented no staining, one denoted 'weak' staining, two reflected 'moderate' staining, and three represented 'high' staining. The sum of the percentage score and the intensity score was represented as the immunostaining score. Finally, expression of these proteins was defined 'low' (score of 0-5) or 'high' (score of 5-6).

| Culture of human endometrial cells and intervention
Isolation of primary human endometrial epithelial cells was undertaken as described in our previous paper. 37 Then, cells were cul-

| Migration assays and invasion assays
We employed 24-well Transwell™ plates (pore size, 8 μm; diameter, 6.5 mm) to carry out assays on migration and invasion assays. Human

| Transfection of small interfering (si) RNA
Three siRNAs targeted to the RON (siRON) gene were purchased from GenePharma. CEECs were seeded onto 12-well plates and transfected with siRON (10 pmol/L) or negative control using LipofectamineTM RNAiMAX (Invitrogen) according to the manufacturer's protocols. Cells were used for migration assays and invasion assays after 48 hours of transfection, and cells transfected for 96 hours were used for Western blotting.

| Animals and treatment
Eighteen female BALB/c mice, 6-8 weeks, were used in this research.

| Statistical analyses
Statistical analyses were carried out using IBM SPSS 17.0 (IBM).
The Student's t test was used to determine the differences between two measurements. One-way ANOVA analysis or Kruskal's test was employed to evaluate the multiple comparisons. Fisher's exact test was used to analyse the categorical variables. P < .05 was considered significant.

| RON was overexpressed in ovarian endometriotic lesions
Previously, we revealed that RON was overexpressed in secretory ovarian endometriotic lesions. 30 In this study, we investigated the RON expression in control endometria (n = 13), eutopic endometria (n = 21) and ovarian endometriotic lesions (n = 16), and all of them were proliferative endometria.
Immunohistochemical staining showed that the RON expression in ovarian endometriotic lesions was higher than that in control and eutopic endometrial tissues in the proliferative phase ( Figure 1).
The score of RON expression in ovarian endometriotic lesions

| EMT occurs in endometrial epithelial cells of ovarian endometriotic lesions
Several scholars have been proposed that EMT participates in endometriosis. 9 To investigate whether EMT occurs in ovarian and ovarian endometriotic lesions (n = 16). The IHC results are illustrated in Figure 2, and expression profile of the EMT markers in endometrial tissues and the IHC score is displayed in Table 1.
Par3 is an important protein for maintaining apical-basal polarity, which is first lost during EMT. 39 Epithelial cells of control endometria presented the highest staining of Par3 compared with that of eutopic and ectopic endometria ( Figure 2, Table 1). The IHC score of ectopic endometria was lower than that of eutopic endometria (Table 1).
Strong staining of N-cadherin was found in 62.5% of endometriotic lesions. Only weak staining appeared in control and eutopic endometrial tissues, and the IHC score of endometriotic lesions was also the highest among them ( Figure 2, Table 1 Table 1) and their IHC score was lower than that of control and eutopic endometrial tissues.
Serving as a tight junction protein for epithelial cells, Occludin exhibited similar staining to that observed for E-cadherin. Only 12.5% of endometriotic lesions had strong staining, whereas the percentage of strong staining in control and eutopic endometria were 92.3% and 66.7% respectively ( Figure 2, Table 1). Taken together, these data explained that the epitheliums of ovarian endometriotic lesions had gone through the EMT.

| RON induced EMT in endometrial epithelial cells
Based on the data mentioned above, we realized that RON expression The cells changed from cubes to spindles, the contacts between cells tend to be lost and cell clusters were scattered ( Figure 3A), which illustrated that the cells had undergone the morphological changes seen typically in EMT. Next, we measured expression of EMT markers in CEECs. Expressions of N-cadherin and ZEB2 were increased, whereas expression of E-cadherin was reduced, in MSP treated cells ( Figure 3B).
To further evaluate the effect of RON on endometrial epithelial cells, we used siRON to transfected CEECs. Expression of N-cadherin protein was down-regulated by siRON and that of E-cadherin protein level was up-regulated ( Figure 3C). Upon stimulation with MSP, the number of CEECs that migrated was 82 ± 14, which was increased significantly compared with control group (57 ± 10), whereas the number of cells was decreased significantly after treated with BMS 777607 (33 ± 10) ( Figure 4A).

| RON promotes the migration and invasion of endometrial epithelial cells
Similar results were seen in the invasion assay: MSP could enhance the invasion of CEECs remarkably, and BMS 777607 inhibited this process ( Figure 4B). Transfection with RON siRNA could also reduce the migration and invasion ability of CEECs ( Figure 4C,D). These data shown above suggested that RON increased the migration and invasion of endometrial epithelial cells through the promotion of EMT in endometriosis. TA B L E 1 Expression profile and IHC score of the EMT markers in endometrial tissues

| RON promoted EMT through Akt and MAPK signalling pathways in human endometrial epithelial cells
Macrophage-stimulating protein could induce EMT in human endometrial epithelial cells and enhance the migration and invasiveness of CEECs, but the signalling involved in these changes is not known.  Figure 5B). Taken together, these data suggested that RON may promote EMT in endometriosis through Akt and MAPK pathways.

| RON inhibitor suppressed the development of endometriosis via the restriction of EMT in vivo
To further study the role of RON in endometriosis, we constructed

| D ISCUSS I ON
We discovered that both RON and EMT are associated with endometriosis development, and that RON can promote migratory and Previously, we showed RON expression in ovarian endometriotic lesions to be significantly higher than that in peritoneal and deeply infiltrating endometriotic lesions. 30 We speculated that the role of RON played in ovarian endometrioma may be greater, so we focused mainly on ovarian endometrioma in the present study.
Importantly, due to the use of different antibodies purchased from different companies, the IHC score of RON in the present study was greater than that in our previous study, but the trend of RON expression in different tissues was consistent with that in our previous study.
Epithelial-mesenchymal transition includes a loss of polarity, damage to cell adhesion and the acquisition of migratory ability. 13 We used N-cadherin, E-cadherin, Occludin and Par3 as markers to study EMT of ovarian endometrioma. N-cadherin expression was We showed, in a preliminarily manner, that RON could induce EMT in endometrial cells. It is worth noting that compared with N-cadherin, the change of E-cadherin expression due to RON is less. The role of E-cadherin in endometriosis is controversial: some studies have displayed its expression to be decreased in endometriosis, 40,41 but others have denied such a decrease or even shown an increase. 19,[42][43][44] Repression of E-cadherin transcription has been considered as a key step of EMT for many years, but an increasing number of studies has shown that the role of E-cadherin in EMT is complicated.
Recently, the concept of an 'EMT spectrum' has been proposed, [45][46][47] whereby EMT is a series of graded processes with intermediate EMT.
Intermediate EMT, also known as 'partial EMT', is the intermediate stage of EMT, and the characteristics of the epithelium and mesenchyme coexist. 48 In addition, some studies have pointed out that high expression of E-cadherin in endometriotic glands may be due to endometriosis being a benign disease 21  times experiments were repeated and the values were expressed as mean ± SD (*P < .05, **P < .01, ***P < .001 compared with control group, # P < .05, ## P < .01, ### P < .001 compared with MSP group). Three times experiments were repeated and the values were expressed as mean ± SD (*P < .05, **P < .01, ***P < .001 compared with control group) RAS-MAPK and PI3K/AKT are the main signalling pathways in the action of RON. 33   key molecules that RON regulates during EMT in endometriosis, and to find better treatments for endometriosis.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.