miR‐200b ameliorates myofibroblast transdifferentiation in precancerous oral submucous fibrosis through targeting ZEB2

Abstract Oral submucous fibrosis (OSF) is a progressive scarring disease. MicroRNA‐200b (miR‐200b) has been reported as a tumour suppressor, but its role in the precancerous OSF remains unknown. In this study, we investigated the impact of miR‐200b on myofibroblastic differentiation activity. Arecoline is a major areca nut alkaloid and has been employed to induce the elevated myofibroblast activity in human buccal mucosal fibroblasts (BMFs). Treatment of arecoline in BMFs dose‐dependently reduced gene expression of miR‐200b, which corresponded with the decreased expression of miR‐200b in fBMFs. The arecoline‐induced myofibroblast activities were abolished by overexpression of miR‐200b in BMFs, and the same results were observed in fBMFs. In addition, α‐SMA was inhibited by an increase in miR‐200b. We further demonstrated that miR‐200b‐mediated decrease in ZEB2 led to down‐regulation of α‐SMA, vimentin. Loss of miR‐200b resulted in enhanced collagen contraction and migration capabilities, and knockdown of ZEB2 reversed these phenomena. Lastly, we showed the expression of miR‐200b was significantly less and ZEB2 was markedly higher in OSF tissues. These results suggested that down‐regulation of miR‐200b may contribute to the pathogenesis of areca quid‐associated OSF through the regulation of ZEB2 and myofibroblast hallmarks.

motile mesenchymal cells, and it is mediated by various key transcription factors, including Snail (SNAI1), Slug (SNAI2), zinc-finger E-box-binding (ZEB) and other basic helix-loop-helix transcription factors. 10,11 Also, intermediate filaments, such as vimentin, play a critical role in the induction of mesenchymal cell shape and motile behaviour. 12 Our previous findings have shown that arecolineinduced myofibroblast transdifferentiation is mediated by ZEB1, 13 and inhibition of EMT transcription factor suppressed the pathogenesis of areca quid-induced OSF through down-regulation of myofibroblast activity. 14 Hence, approaches to decrease the EMT regulating factors may be a promising strategy to inhibit the activation of myofibroblasts after arecoline stimulation, therefore preventing OSF pathogenesis.
MicroRNAs (miRNAs) are small, endogenous non-coding RNAs of 21-25 nucleotides in length that mediate the post-transcriptional regulation of gene expression through partial complementary binding to the 3 0 untranslated region (UTR) in target transcripts. 15,16 Aberrant expression of miRNAs in fibrotic diseases has been reported in various studies (see Review 17 ), and numerous miRNAs have been suggested to modulate EMT process, thereby contributing or inhibiting tissue fibroses. [18][19][20] For instance, the EMT-inducing transcriptional factors ZEB1 has been found to be a target of miRNA-192 and inhibition of miR-192 resulted in decreased collagen expression as well as reduced renal fibrosis. 21 In addition, accumulating studies have demonstrated that miR-200 family reduces fibrosis by inhibiting EMT and preventing the deposition of ECM. [22][23][24] Also, it has been revealed that miR-200b is associated with cancer metastasis, 25 chemoresistance 26 and prognosis. 27 Hence, the impact of miR-200b on precancerous OSF needs to be clarified.
In this study, we explored the functional role of miR-200b in the pathogenesis of areca quid-associated OSF and the associated mechanism. We tested the expression of miR-200b in fibrotic cells and tissues, and investigated its contribution in arecoline-induced and fibrotic myofibroblast activity, including higher collagen gel contractility and migration ability. Moreover, we assessed the expression of EMT-associated factors in accordance with overexpression of miR-200b and examined the relationship between miR-200b and ZEB2 in BMFs. Our data revealed the anti-fibrotic potential of miR-200b in preventing the progression of OSF.

| Chemicals and cell culture
Arecoline and collagen solution from bovine skin were purchased from Sigma-Aldrich (St. Louis, MO, USA). TGFb type I receptor inhibitor, SB431542, was purchased from EMD Millipore (Billerica, MA, USA). Human TGFb1 was obtained from R&D Systems (Minneapolis, MN, USA). All methods applied in this study were carried out in accordance with the approved guidelines from the Institutional Review Board of Chung Shan Medical University Hospital and informed written consent was obtained from each individual prior to commencing the study. Tissue specimens from areca quid chewers were collected from Department of Dentistry, Chung Shan Medical University Hospital, Taichung, Taiwan. Biopsy specimens were taken from the histologically normal oral mucosa and fibrotic mucosa at the time of surgical third molar extraction. Fibroblast cultures were grown and maintained by using the explant method as described previously. 28 Cell cultures between the third and eighth passages were used in this study. 2 9 10 5 BMFs were suspended in 0.5 mL of 2 mg/mL collagen solution (Sigma-Aldrich) and then added into 24-well plate followed by incubation at 37°C for 2 hours for the polymerization of collagen cell gels. After detaching gels from wells, the gels were further incubated in 0.5 mL MEMa medium with or without arecoline for 48 hours.

| Collagen contraction assay
Contraction of the gels was photographed and measured using Ima-geJ software (NIH) to calculate their areas. 29

| Migration assay
Cell migration assay was conducted using 24-well plate Transwell â system with a polycarbonate filter membrane of 8-lm pore size (Corning, Acton, MA, USA) assay kit as previously described. 30 The migrated cancer cells were then visualized and counted from 5 randomly selected fields under 100-fold magnification using an inverted microscope.

| Wound healing assay
Cells were seeded into 6-well culture dishes. Wounds were introduced to the confluent monolayer of cells with a sterile 200 lL plastic pipette tip to create a denuded area. Cell movement towards the centre of the wound area was photographed at 0 and 48 hours under a microscope. 14

| Western blot analysis
Western blot analysis was conducted as previously described. 30 The primary antibodies against a-SMA, ZEB2, vimentin and Slug  MiR-200b overexpression plasmid construct was generated according to our previous method.

| Apoptotic assay
Apoptotic cells were detected with an Annexin V-APC kit (Calbiochem, Darmstadt, Germany) according to manufacturer's guidelines. After staining, the cells incubated with 20 lg/mL propidium iodide (PI) were analysed by FACS Calibur apparatus (Becton Dickinson, San Diego, CA, USA).

| Statistical analysis
Quantitative data were presented as mean AE SD and analysed with Student's t test performed with SPSS Statistics version 13.0. P value < .05 was considered as statistically significant.

| Mir-200b is significantly down-regulated in arecoline-stimulated BMFs and fBMFs
Arecoline is a major areca nut alkaloid and has been implicated in the pathogenesis of OSF. 3 Our previous study has demonstrated that arecoline could induce myofibroblast transdifferentiation in human primary buccal mucosal fibroblasts (BMFs). 13 qPCR analysis revealed that the expression of miR-200b reduced in both BMFs as the concentration of arecoline increased ( Figure 1A). Likewise, F I G U R E 1 miR-200b is down-regulated in arecoline-stimulated BMFs and fBMFs. The expression of miR-200b was assessed in BMFs in response to various concentration of arecoline exposure (A); the expression of miR-200b was compared between normal buccal mucosal fibroblasts (BMFs) and fibrotic BMFs performed with qRT-PCR (B); C, cells were pretreated with TGF-ß type I receptor inhibitor, SB431542 (10 lmol/L), followed by treatment with arecoline (20 mg/mL) or TGF-ß1 (5 ng/mL) for 2 h. miR-200b expression was measured using qRT-PCR analysis. *P < .05 compared with no treatment control; **P < .01 compared with BMFs; # P < .05 compared to arecoline-treated, TGF-b1treated or combined-treatment groups primary cultivated fibroblasts from OSF tissues (fBMFs) displayed a significantly lower expression of miR-200b in comparison with pair normal BMFs ( Figure 1B). To examine whether the inhibition of miR-200b by arecoline in BMFs was through TGF-ß signalling, we pretreated the BMFs with SB431542 (10 lmol/L), TGF-ß type I receptor inhibitor, followed by arecoline or TGF-ß administration. As expected, SB431542 treatment significantly prevented the arecolineor TGF-ß1-inhibited miR-200b expression in BMFs ( Figure 1C).
These results showed that the alteration of miR-200b after arecoline stimulation was via TGF-ß signalling and may be associated with the OSF development; therefore, we conducted the following experiments to investigate the functional role of miR-200b in myofibroblast characteristics.

| Overexpression of miR-200b successfully hinders the arecoline-induced myofibroblast activities
Upon injury, fibroblasts become activated to migrate into the injured site and differentiate into contractile myofibroblasts for tissue healing. 6 It also has been demonstrated that treatment of areca nut extract dose-dependently increases the collagen contractility. 31 Therefore, collagen gel contraction and migration capacities have been commonly employed to study the activity of myofibroblasts. 14,29 As expected, BMFs exhibited increased contractility in response to arecoline exposure, whereas overexpression of miR-200b counteracted it (Figure 2A). In addition, we observed that the arecoline-enhanced migration activity of BMFs was repressed by overexpression of miR-200b ( Figure 2B).

| MiR-200b reduces the characteristics of myofibroblasts
To confirm the role of miR-200b in myofibroblast activation, we investigated whether these increased activities and myofibroblast marker, a-SMA, would be suppressed by elevated expression of miR-200b. By collagen contraction assay, we demonstrated that overexpression of miR-200b significantly inhibited the highly contractile phenotype in BMFs from fibrotic oral cavity tissues ( Figure 3A).
With regard to the impact of miR-200b on their migration capacity, we observed a decreased cell migration performed with transwell ( Figure 3B) and wound healing ( Figure 3C) assays. Moreover, we showed that miR-200b reduced the expression level of a-SMA, indicating the anti-fibrotic effect of miR-200b ( Figures 3D and S1).

ZEB2-regulated expression of a-SMA and vimentin in fBMFs
As a direct target of miR-200b, 32 Figures 4C and S2). Besides, we showed that knockdown of ZEB2 led to down-regulation of a-SMA and vimentin in fBMF (Figures 4D and S3).

BMFs and increases the expression of fibrosis markers
To further confirm the functional involvement of miR-200b in ZEB2mediated oral fibrosis, we treated BMFs with knockdown sponge (Spg. miR-200b) and found the elevated expression of fibrosis-associated markers, including Slug and a-SMA, in comparison with cells treated with control sponge (Spg. Ctl.) (Figures 5A and S4). Nevertheless, the increased expression levels of Slug and a-SMA were abolished by knockdown of ZEB2 ( Figure 5A). We showed that silencing of endogenous miR-200b induced higher migration ability ( Figure 5B), collagen gel contractility ( Figure 5C) and wound healing capacity ( Figure 5D) in BMFs, whereas knockdown of ZEB2 counteracted these responses ( Figure 5B-D). Altogether, these findings suggested that blockage of the endogenous miR-200b resulted in increased expression of Slug and a-SMA via ZEB2 in BMFs, leading to increased myofibroblast activity.

| Overexpression of miR-200b expression induces apoptosis in fBMFs
Furthermore, we found that overexpression of miR-200b increased the mean number of apoptotic cells in fBMFs ( Figure S5A). miR-200b suppressed the anti-apoptotic genes Bcl-2 and Bcl-xl but elevated the Bax gene in fBMFs by real-time RT-PCR analysis ( Figure S5B).  OSF specimens than normal group ( Figure 6A). On the other hand, the relative gene expression of ZEB2 was up-regulated in OSF tissues ( Figure 6B). Linear regression analysis reveals a negative correlation between miR-200b expression and ZEB2 mRNA expression in clinical OSF specimen ( Figure 6C). Over the past few years, the mechanism for the anti-fibrotic effects of miR-200b has been examined and discussed in several fibrotic diseases. It has been shown that miR-200b ameliorated TGF-b1-induced fibrosis in colorectal epithelial cells. 37 In hypertrophic scars, miR-200b has been proved to decrease the expression of fibrosis markers, such as collagen type 1 and a-SMA, as well as ZEB1 in fibroblasts. 38 Another study showed that miR-200b affected hypertrophic scarring through regulating the fibroblasts proliferation and apoptosis by affecting the collagen synthesis, fibronectin expression and TGF-b1/a-SMA signalling. 39 As for pulmonary fibrosis, it has been found that the expression of miR-200 was reduced in fibrotic lungs, and down-regulation of miR-200 may contribute to EMT and enhance pulmonary fibroblast accumulation. 22 In contrast, overexpression of miR-200b markedly attenuated TGF-b1-induced expression of fibronectin and a-SMA in lung fibroblasts. 22 Furthermore, overexpressing miR-200b in intestinal epithelial cells led to inhibition of EMT characterized by down-regulation of vimentin and up-regulation of E-cadherin through targeting ZEB1. 40 One of the recent studies showed that miR-200b/c exerts a protective effect in the LPS-induced early pulmonary fibrosis by targeting ZEB1/2, which may be associated with the inhibition of p38 MAPK and F I G U R E 6 The expression of miR-200b is down-regulated and ZEB2 is up-regulated in OSF tissues. The relative expression levels of miR-200b (A) and ZEB2 (B) were detected and compared between normal (N) and OSF tissues **P < .01 compared with normal mucosal tissues. C, Total RNA was extracted from tissues of OSF patients and the expression of miR-200b and ZEB2 was determined by qRT-PCR methods and analysed with Spearman rank correlation test TGF-b/smad3 signalling. 41

CONF LICT OF I NTEREST
The authors declare that they have no conflict of interests.