Down‐regulation of miRNA‐27b‐3p suppresses keratinocytes apoptosis in oral lichen planus

Abstract Oral lichen planus (OLP) is considered a precancerous lesion with no known cure. Recent studies reported that abnormal regulation of apoptosis was involved in the pathogenesis of OLP. Next generation sequencing was used to screen the candidate microRNAs and genes in biopsies from patients with OLP and healthy mucosa. Human oral keratinocytes were transfected into the related oligonucleotides of miR‐27b‐3p/cyclophilin D and their control groups. Apoptosis was detected by TdT‐mediated dUTP nick end labelling and flow cytometry. The levels of mRNA and protein were detected by quantitative PCR, Western blots, and enzyme‐linked immunosorbent assays, respectively. Luciferase assays were performed to detect the luciferase activities of miR‐27b‐3p and cyclophilin D. Here, we showed that basal epithelium apoptosis was reduced and the miR‐27b‐3p levels were decreased in clinical OLP samples. We also found that down‐regulation of miR‐27b‐3p inhibited epithelial keratinocyte apoptosis by up‐regulating cyclophilin D expression. Moreover, cyclophilin D increased the protein stability of Bcl2 through direct binding, and Bcl2 suppressed caspase9/3 activation and cytochrome C release. Taken together, these data showed that miR‐27b‐3p regulated keratinocyte apoptosis through cyclophilin D/Bcl2 signalling, suggesting the miR‐27b‐3p regulated the pathogenesis of OLP.

Although T cells are considered responsible for triggering apoptosis of basal epithelial cells via various pathways, [9][10][11] vacuolar degeneration does not unequivocally indicate apoptosis. 12 Moreover, down-regulation of apoptosis markers (ie active caspase 3 and Bax) 4,12,13 and upregulation of anti-apoptotic markers (Bcl2) 14 are frequently found in the basal cell layer of OLP. Thus, we proposed that abnormal apoptosis in the basal keratinocytes participates in the pathogenesis of OLP.
It is well-known that apoptosis is mainly initiated by intrinsic or extrinsic pathways, which eventually regulate the activity of apoptotic proteins and/or anti-apoptotic proteins, such as the proteins of the Bcl-2 family. [15][16][17] MicroRNAs (miRNAs) are small, non-coding, single-stranded RNAs that bind to complementary mRNA sequences and negatively regulate gene expression by repressing translation or destabilizing the target mRNAs. 16 A single miRNA could broadly affect hundreds of target genes. [17][18][19] Previous studies have reported that apoptosis is regulated by miRNAs in OLP through targeting apoptosis-related proteins, 20,21 although the mechanism by which mRNAs are regulated in OLP is not clear. 3 Increasing evidence has shown that miRNAs are involved in the regulation of OLP pathogenesis. 22, 23 Ma et al 24 reported that approximately 70 miRNAs were significantly up-or down-regulated in OLP patients compared to healthy volunteers. Therefore, the current study aimed to clarify the role and mechanism of miRNAs in regulating apoptosis of OLP. Here, we showed that miR-27b-3p was significantly downregulated in oral biopsy specimens from OLP patients by using miRNA microarrays 25 and RNA sequencing, and down-regulation of miR-27b-3p reduced basal apoptosis in OLP.
Moreover, we also found that cyclophilin D (CypD) was a new target of miR-27b-3p. In summary, we demonstrated a regulatory role of miR-27b-3p in OLP keratinocyte apoptosis.

| Sample collection
Biopsy samples were collected from 25

| TUNEL in situ detection of DNA fragmentation
In situ cell death detection kit (Roche, Berlin, Germany) (TUNEL technology) was used to analyse apoptosis in OLP and control samples.
After deparaffinizing and dehydration, sections were incubated with proteinase K and 30% H 2 O 2 , respectively, followed by incubation with complete labelling reaction buffer and antibody solution at 37°C for 1 hour. The antigen-antibody interaction was visualized by fluorescence labelling. Sections were then counter stained with DAPI, and visualized under a fluorescence microscope. At least five fields in each slide were analysed for TUNEL-positive cells in each group. The evaluation was performed by an investigator blinded to the groups studied.
Blocking was performed for 3 hours at 37°C with 2% goat serum and 2 mg/mL bovine serum albumin in phosphate-buffered saline (PBS) and alkaline phosphatase-conjugated Fab anti-DIG antibody (Roche, Mannheim, Germany). Staining was performed using NBT/BCIP (Roche), with Nuclear Fast Red used as the counterstain. The level of miR-27b-3p expression was measured using Image Pro-Plus software, version 6.0 (Media Cybernetics, Rockville, CA). 28

| Plasmid construction and dual luciferase reporter assay
The full-length 3′-UTR of the human CypD gene was amplified by PCR using human genomic DNA as a template. The sense primer was 5′-TTTGAGCTCTCCTCACGACCTCATTTCTGGG-3′, and the antisense primer was 5′-TTTCTCGAGGCTCAGTAAAGATCAGCTCCAA-3′.
Site-directed mutagenesis was performed to generate a mutant CypD 3′-UTR that contained mutations in the conserved miR-27b-3p binding site. In the mutant 3′-UTR of the CypD gene, the nucleotide sequence complementary to nucleotides 2-5 of the miR-27b-3p binding site (ACUGUGA) was mutated to the sequence found in miR-27b-3p (TCAGAGT). The HEK 293T cells were seeded in 12-well plates at a density of 2 × 10 5 cells/well and co-transfected with 200 ng of pmirGLO-cypd 3′-UTR (wild-type or mutant) plasmid and 3000 ng of F I G U R E 3 The miR-27b-3p regulates CypD expression. (A) Screening of miR-27b-3p target genes by integrative analyses using the web-based prediction and RNA sequencing data. (B) The verification of six candidate genes using the 3′-UTR luciferase assay. Plasmids containing wild-type 3′-UTRs of all six genes were separately co-transfected with the miR-27b-3p mimics or scrambled miRNA. (C) The effects of the miR-27b-3p on wild-type UTR or mutant UTR were determined by the 3′-UTR luciferase assay. The key complementary sequence of CypD 3′-UTR to miR-27b-3p was highlighted, and the nucleotides in red were mutated to their complementary nucleotides. (D) The effects of miR-27b-3p mimics or inhibitors on CypD expression in HOK cells were detected by q-PCR and (E) Western blot analyses, respectively, and were quantitated (right). Results are representative of three independent experiments and are presented as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 Scr-miR or miR-27b-3p mimics using the Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's protocol. After 48 hours of incubation, the luciferase activity was measured using the dual-luciferase reporter assay system (Promega, Beijing, China).
Synthetic mimics or inhibitors were transfected into cell cultures using Lipofectamine 2000 (Invitrogen) to promote or inhibit miR-27b-3p activity, respectively. Negative controls were used to validate both reactions. The final concentration of the mimics and inhibitors was 100 nmol/L and 200 nmol/L, respectively.

| Cell viability assays
To test the optimal concentration of etoposide for the induction of apoptosis in HOK cells, cell viability was analysed by the median effect equation to obtain the IC 50 value based on all the data points of the cytotoxicity-concentration curve. Cells (5000 cells/well) were seeded into a 96-well plate and treated with etoposide at different concentrations (10×, 1×, 1:10, 1:100, 1:1,000, and 1:10,000); the cells were labelled with CCK8 reagent (Dojindo Molecular Technologies, Tokyo, Japan) for 2 hours. Cell viability was measured at an absorbance at 450 nm using a microplate reader (Model 450; Bio-Rad Laboratories, Hercules, CA).
Based on the survival ratio of cells, the IC 50 value was obtained by the integration of data from the cells treated with the etoposide gradient, thereby generating quantitative measures of protection.

| RNA isolation and qPCR
Total RNA was extracted from the cultured cells using TRIzol ® reagent (Invitrogen) according to the manufacturer's instructions. The reverse transcription of miR-27b-3p was performed using the TaqMan ® MicroRNA RT kit (Invitrogen). The primer sequence used for the cDNA synthesis was GTCGTATCCAG TGC AGGG TCCG AGG TAT TCGC AC TGG ATACG ACGC AG A A .  F I G U R E 5 CypD was up-regulated in oral lichen planus (OLP) and was a prognostics marker for oral squamous carcinoma. (A) The protein levels of CypD were detected in cultured primary epithelial cells from OLP and healthy control (HC) mucosa by Western blots and further analysed by densitometry. The data were normalized to β-actin. (B, C and D) CypD, Bcl2, and caspase3 expression were analysed by immunohistochemical staining of OLP and healthy control (HC) tissues. The expressions were further analysed by densitometry. Left, magnification is ×200; Right, magnification is ×400. (E) The overall survival of oral squamous carcinoma patients was analysed. The patients were divided into two groups according to low expression (n = 149) and high expression (n = 370) of CypD. *P < 0.05

| Protein extraction and Western blot analysis
The membranes were blocked with 5% non-fat dry milk in TBST for 1 hour and incubated overnight at 4°C in 5% BSA in TBS with antibodies against CypD (Santa Cruz Biotechnology, Santa Cruz, CA), Bcl2, Apoptosis Antibody Sampler Kit (for PARP, cleaved caspase 3, cleaved caspase 9, and caspase 9) (Cell Signaling Biotechnology, Beverly, MA). The antibody against β-actin (Santa Cruz Biotechnology) was used for the internal control.

| Apoptosis assay
Apoptosis was assessed using annexin-V-propidium iodide (PI).  buffer containing a protease inhibitor cocktail, and disrupted with a glass tissue grinder. Homogenates were centrifuged at 600 g at 4°C for 5 minutes, and the resulting supernatants were transferred to 0.5 mL conical tubes, and further centrifuged at 11 000 g at 4°C for 10 minutes, the resulting precipitates were the isolated mitochondria.

| Co-immunoprecipitation
Samples were homogenized in ice-cold 'immunoprecipitation buffer' containing 20 mmol/L Tris-HCl, pH 7.5, 150 mmol/L NaCl, and 1.5% Nonidet P-40 supplemented with protease inhibitors. The extracts (500 µg protein per sample) were pre-cleared with protein A/G beads and then mixed with non-specific IgG (1 µg) or polyclonal mouse anti-Bcl2 antibody (Cell Signaling Biotechnology), or anti-CypD (Santa Cruz Biotechnology) overnight at 4°C, followed by the addition of 40 µL of protein A/G-agarose (Santa Cruz Biotechnology) for 3 hours at 4°C.
Immune complexes were washed four times in an 'immunoprecipitation wash buffer' (100 mmol/L Tris-HCl, pH 7.5, 100 mmol/L NaCl, 0.1% Triton X-100) and resuspended in 2× Laemmli buffer (Santa Cruz Biotechnology). The inputs were also mixed with 2× Laemmli buffer, and then the immunoprecipitation reactions and inputs were boiled for 10 minutes and pelleted in a centrifuge.
The supernatants were subjected to Western blot analysis as described above.

| Immunohistochemistry
Paraffin-embedded clinical samples were deparaffinized, rehydrated, and blocked for endogenous peroxidase activity with 3% H 2 O 2 /methanol. After rinsing, tissue sections were blocked for non-specific binding with 1% goat serum in PBS. Tissue slides were incubated with primary antibodies at 4°C overnight, followed by incubation with peroxidase-labelled anti-mouse secondary antibodies. Finally, antibody binding was visualized with a DAB substrate (ImmunoPure ® ; Pierce Biotechnology, Waltham. MA).

| Statistical analysis
The data are presented as the mean ± SEM of at least three independent experiments. The samples were analysed using a two-tailed unpaired Student's t test, unless otherwise noted, and multiple group comparisons were made using one-way analyses of variance (ANOVA). P < 0.05 were considered statistically significant. P < 0.05, <0.01, and < 0.001 are indicated with one, two, and three asterisks (*, **, or ***), respectively. Graph Pad Prism software (San Diego, CA, USA; version 6.03) was used for data analyses.

| MiR-27b-3p was significantly downregulated in the epithelial of OLP patients
To determine what miRNAs are involved in OLP, the miRNA profiling was first performed by using miRNA microarrays (3HC vs 3OLP) in oral biopsy specimens of OLP patients. As shown in Figure 1, 38 miRNAs were found to be down-regulated more than 2-fold in OLP specimens, compared to that of healthy control specimens. Furthermore, the miRNA expression was examined in OLP specimens by using RNA sequencing. Forty-three miRNAs were found to be down-regulated more than 2-fold in OLP specimens as shown by RNA sequencing (2HC vs 2OLP) ( Figure 1). By comparatively analysing two datasets from microarrays and RNA sequencing, the miR-27b-3p was the most significantly down-regulated miRNA (more than 3.6-fold) among six commonly down-regulated miRNAs (Figure 1 and Figure S1A).
The miR-27b-3p expression was further verified in oral biopsies of OLP patients by using quantitative PCR (qPCR) and in situ hybridization (ISH) (Figure 1,C). Both methods showed that the miR-27b-3p levels decreased in OLP tissues compared to that of healthy control tissues. Moreover, the ISH data clearly revealed that the miR-27b-3p was mainly expressed in the epithelial layer, and was down-regulated in OLP tissues, compared to healthy tissues ( Figure 1).
To investigate the overall apoptosis levels in OLP tissues, haematoxylin and eosin staining (H＆E) and TUNEL (TdT-mediated dUTP nick end labelling) assays were performed. As shown in  (Figure 1 and Figure S1B).
Moreover, the apoptotic cells were mainly located in the basal layer of OLP specimens, suggesting that apoptosis was suppressed in the basal layer.

| Down-regulation of miR-27b-3p suppressed keratinocytes apoptosis
To investigate the role of miR-27b-3p in OLP tissues, miR-27b-3p was either overexpressed or knocked down in human oral keratinocytes (HOKs), mediated by lentiviruses infection. The HOK cells are immortalized epithelial cells from the basal layer of oral mucosa tissues. 29 The survival ability of keratinocytes was first analysed in empty vector-transfected control cells and miR-27b-3p overexpressing/knockdown cells. As shown in Figure 2, knockdown of miR-27b-3p promoted cell growth; in contrast, overexpression of miR-27b-3p inhibited cell growth.
To further investigate the regulation of miR-27b-3p on apoptosis, apoptotic signalling was determined in the HOK cells overexpressing or depleted of miR-27b-3p. The cleaved caspase 3 levels were determined in the HOK cells treated with etoposide, a compound known to trigger the caspase 3-mediated apoptotic pathway. 30,31 As shown in Figure 2, etoposide treatment in-
To further determine what genes were targets of miR-27b-3p, the 3′-UTR luciferase assay was performed. As shown in Figure  When miR-27b-3p was stably overexpressed or knocked down by lentiviruses, CypD protein levels were also decreased or increased in HOK cells, respectively ( Figure 3). Taken together, these results showed that miR-27b-3p negatively regulated CypD expression in keratinocytes cells.

| CypD suppressed apoptosis by increasing Bcl2 stability
To determine the effects of CypD on apoptosis in keratinocytes, annexin V turnover ratios were analysed by flow cytometry in HOK cells expressing or depleted of CypD. As shown in Figure  To determine the mechanism by which CypD regulated keratinocyte apoptosis, co-immunoprecipitation was performed using CypD antibody to determine whether CypD regulated apoptosis by interacting with Bcl2 (a major anti-apoptotic protein). 31 As shown in

| CypD was up-regulated in OLP tissues and was associated with the prognosis of oral squamous carcinoma
To determine whether CypD was a novel biomarker of OLP, the protein level of CypD was first analysed by Western blotting in primary keratinocytes from patients with OLP and from healthy controls. As shown in Figure 5, CypD was up-regulated in primary keratinocytes from OLP patients, compared to that from healthy controls. The immunohistochemistry staining assay results also showed that protein levels of CypD were up-regulated in OLP tissues ( Figure 5), which was consistent with the data using miR-27b-3p. In addition, the immunohistochemistry staining assay showed increased Bcl2 and caspase 3 expression levels in OLP tissues and there was statistical significance ( Figure 5,D), which further confirmed the corresponding results using miR-27b-3p in vitro, except that there was no statistical significance in caspase 9 expression level ( Figure S1D), which may be due to the interference of various apoptotic regulatory pathways.
Because the OLP potentially develops into malignant oral squamous carcinoma at a late stage, finding a prognostic marker is critical for OLP patients. To further determine whether CypD is a prognostic marker for oral squamous carcinoma, survival analyses were performed on patients with high expression or low expression of CypD.
As shown in Figure 5, patients with low expression of CypD survived for a longer period than those with high expression of CypD, suggesting that CypD was a prognostic marker for oral squamous carcinoma.

| D ISCUSS I ON
OLP is thought to undergo premalignant transformation, 33 35 showed a significant decrease in miR-27b in OLP, which was consistent with our results and the previous observations. 25 We also observed colloid bodies and decreased apoptosis in OLP epithelium, and further found that apoptotic cells were mainly In our study, we found that CypD was a downstream target of miR-27b-3p that was up-regulated in the epithelial layer of OLP.
CypD, also known as peptidyl prolyl isomerase F, regulates the mitochondrial permeability transition pore opening, 39  Bcl2 in the OLP epithelium, with the latter inhibiting cytochrome C release and caspase 3/9 activation, which was supported by the results from Eliseev et al. 31 Our results demonstrated that miR-27b-3p down-regulation as well as CypD up-regulation inhibited basal keratinocyte apoptosis, suggesting that miR-27b-3p/CypD-medicated apoptosis may interfere with the pathogenesis of OLP. Jia et al 46 demonstrated that histone deacetylase 6 regulated miR-27b that suppressed proliferation, promoted apoptosis, and targeted oncogenic MET, implicating the potential application of miR-27b for prognostic prediction and therapy of diffuse large B-cell lymphoma. Moreover, miR-27b-3p also inhibited the inflammatory response by reducing the expression/secretion of inflammatory factors, such as IL-1A, JAK2, IL-6, and IL-1B, 47 suggesting that besides apoptosis, miR-27b-3p/CypD signalling may play multiple roles in oral mucosal tissue; however, the specific mechanism needs further study in the future.
In summary, our study showed that miR-27b-3p regulated keratinocyte apoptosis of the basal epithelial in OLP. MiR-27b-3p down-regulation directly increased CypD protein levels; the latter inhibited keratinocyte apoptosis by binding to Bcl2. Further study of the miR-27b-3p/CypD signalling complex may lead to a better understanding of molecular pathways involved in the pathogenesis of OLP and epithelial apoptosis, thus facilitating the development of more effective therapies. However, more studies are necessary to thoroughly elucidate the role of miR-27b-3p in OLP.