Rab27A promotes cellular apoptosis and ROS production by regulating the miRNA‐124‐3p/STAT3/RelA signalling pathway in ulcerative colitis

Abstract Ulcerative colitis (UC) is a multifactorial inflammatory disease, and increasing evidence has demonstrated that the mechanism of UC pathogenesis is associated with excessive cellular apoptosis and reactive oxygen species (ROS) production. However, their function and molecular mechanisms related to UC remain unknown. In this study, Rab27A mRNA and protein were proven to be overexpressed in intestinal epithelial cells of UC patients and DSS‐induced colitis mice, compared with control (P < 0.05). And Rab27A silencing inhibits inflammatory process in DSS‐induced colitis mice (P < 0.05). Then, it was shown that knockdown of Rab27A suppressed apoptosis and ROS production through modulation of miR‐124‐3p, whereas overexpression of Rab27A promoted apoptosis and ROS production in LPS‑induced colonic cells. In addition, enhanced expression of miR‐124‐3p attenuated apoptosis and ROS production by targeting regulation of STAT3 in LPS‑induced colonic cells. Mechanistically, we found Rab27A reduced the expression and activity of miR‐124‐3p to activate STAT3/RelA signalling pathway and promote apoptosis and ROS production in LPS‑induced colonic cells, whereas overexpression of miR‐124‐3p abrogated these effects of Rab27A. More importantly, animal experiments illustrated that ectopic expression of Rab27A promoted the inflammatory process, whereas overexpression of miR‐124‐3p might interfere with the inflammatory effect in DSS‐induced colitis mice. In summary, Rab27A might modulate the miR‐124‐3p/STAT3/RelA axis to promote apoptosis and ROS production in inflammatory colonic cells, suggesting that Rab27A as a novel therapeutic target for the prevention and treatment of UC patients.


| INTRODUC TI ON
Ulcerative colitis (UC) is a chronic inflammatory bowel disease (IBD) that affects both children and adults, and is characterized by periods of relapse followed by periods of remission. 1 Because of the westernized dietary lifestyle, the incidence of UC in several Asian countries, especially in China, is rising rapidly. 2,3 However, although dysregulation in intestinal epithelial cells, such as TNF-α and IL-1β, 4,5 has been widely reported in UC, the molecular basis and pathophysiology of UC are not completely understood.
MicroRNAs (miRNAs) are small non-coding RNAs that bind to corresponding sequences in the 3'-untranslated region (UTR) of reciprocal objective mRNAs, thus inhibiting the synthesis of proteins. 6 Increasing evidence suggests that miRNAs are widely dysregulated in UC, potentially impacting UC pathogenesis, diagnosis and therapy. [6][7][8] For example, Wu et al found that miR-206 regulates TNF-α and IL-8 in active human UC and dextran sodium sulphate (DSS)-induced experimental colitis in mice. 8 Min et al illustrated that miR-155 overexpression enhanced UC inflammatory activity by down-regulating the expression of FOXO3a, a key inhibitor of the NF-κB signalling pathways. 7 In the past decade, a number of Rab proteins have been demonstrated to be involved in the initiation, development and progression of IBD, such as Rab7b, Rab11 and Rab13. [9][10][11] Rab27A, belonging to the Ras superfamily of monomeric G proteins, localizes to distinct cellular compartments and regulates specific steps of intracellular membrane trafficking. 12 Recently, Tang et al showed that Rab27A could directly bind to miR-124-3p to inhibit tumorigenesis in osteoclastogenesis. 13 Although Rab27A has been reported to be up-regulated in clinical UC patients, 14 its function and molecular mechanisms related to UC remain unknown. In the present study, we observed that Rab27A mRNA and protein expression levels were increased in both human UC patients and DSS-induced colitis mice. Subsequent results demonstrated that knockdown of Rab27A suppressed cellular apoptosis and reactive oxygen species (ROS) production in colonic inflammatory cells.
Mechanistically, Rab27A could regulate the miR-124-3p/STAT3/RelA axis to promote apoptosis and ROS production in ulcerative colitis.  Table 1. The diagnosis of UC was based on standard clinical characteristic and histological criteria, colonoscopy feature, and pathological results. 15 The patients were obtained before initiation of anti-inflammatory treatment. The samples were embedded in paraffin for immunofluorescence analysis or immediately used for RT-qPCR or Western blotting.

| Animal studies
The animal research was approved by the local ethics committee of Renji Hospital, Shanghai, China. Male C57BL/6 mice [age 7-8 weeks old, weight 20-22 g] were regularly treated with 2.5% dextran sulphate sodium (DSS, MW 40-50 kDa; MP Biomedicals, USA) in drinking water for 1 month, after which intestinal mucosa was harvested for further analysis. The severity of colitis was scored by recording standard parameters, including colonic length, inflammatory cell infiltration and histological score. To evaluate the function of Rab27A, the C57BL/6 mice were intracolonically administered 40 μg of Lv-shRab27A or a control Lv-shRNA on 1 and 15 days using the previously reported. 16 Briefly, the appropriate amount of Lv-shRab27A and its control were resuspended in 100 μL of Opti-MEM with 2 μL of Lipofectamine 3000.

| Histological analysis
The human/mice colonic tissue was stained with H&E, and histological score was blindly established by per Obermeier et al 17 The intestinal mucosal damage was graded on the following 0-4 scale: 0-none; 1-minimal loss of goblet cells; 2-extensive loss of goblet cells; 3-minimal loss of crypts and extensive loss of goblet cells; and 4-extensive loss of crypts. Inflammatory infiltration was graded on the following 0-4 scale: 0-none; 1-infiltrate around crypt bases; 2-infiltrate in muscularis mucosa; 3-extensive infiltrate in muscularis mucosa, with oedema; and 4-infiltration of the submucosa. The histological activity index (HAI) was designated as the sum of the mucosa and infiltration scores, resulting in the total HAI score ranging from 0 (unaffected) to 8 (severe colitis).

| Isolation of intestinal epithelial cells (IECs)
The mucosa was collected from human/mice intestines at 4°C and immediately stored at −80°C. The frozen mucosal tissues were homogenized with an OmniTH homogenizer (Beijing Pioneer Trading Co., Ltd., China) at homogenization buffer (50 mmol\L Tris-HCl, pH 7.2) containing Na 3 VO 4 and a protease inhibitor cocktail (Sigma-Aldrich, USA). After ultrasonic treatment, the homogenate was centrifuged at 2500 × g for 5 min. The above supernatant was isolated as total intestinal epithelial proteins, and protein concentrations then were measured by a Bio-Rad Protein Assay (Hercules, CA, USA).

| Cell lines and plasmid transfection
The HT-29 and Caco-2 colonic cell lines were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China).
The cells were distributed in 6-well plates to approximately 50%-70% confluence and were transfected the next day with plasmid at a concentration of 100 nmol\L in DMEM (GenePharma, Shanghai, China) using Lipofectamine 3000 (Invitrogen, USA), according to the manufacturer's instructions.

| RNA isolation and real-time quantitative PCR (RT-qPCR)
According to the manufacturer's instructions, total RNA from cell lines and tissue samples was extracted using the TRIzol reagent (Invitrogen, USA). cDNA was synthesized using a microRNA Reverse

| Western blotting
All the proteins were separated by 10% SDS-PAGE and transferred onto a PVDF membrane (Millipore, Bedford, MA). The membranes were blocked for 90 minutes in TBS containing 0.1% Tween 20 and 5% non-fat powdered milk and then incubated first with primary antibodies against Rab27A (ab55667, 1:1500, Abcam, UK), STAT3

| Flow cytometry analysis
Cell apoptosis and ROS production were quantified by flow cytome-

TA B L E 2 Quantitative Real-time PCR primers used in this study
construct was compared with that of the pmirGLO vector group.
Each experiment was repeated in triplicates.

| Statistical analysis
The statistical differences were analysed using the Student's t test between two groups or chi-squared testing between multiple groups by using SPSS 22.0 statistical package software (SPSS, Chicago, IL).
The value of P < 0.05 was considered statistically significant.

| Rab27A mRNA and protein expression levels were up-regulated in IECs from UC patients and experimental animal
To examine the expression status of Rab27A in UC tissues, we isolated IECs from UC patients. Rab27A mRNA and protein expression levels were markedly up-regulated in UC tissues compared with those in matched normal tissues ( Figure 1A,B). Moreover, immunohistochemical analysis revealed that Rab27A protein in the inflamed colonic mucosa of UC tissues was highly expressed, especially in the epithelial layer, compared with that in the control colonic tissue mucosa ( Figure 1C). Next, we established a DSS-induced colitis mouse model that has been used extensively to study the pathogenesis of ulcerative colitis ( Figure 1D-F). As shown in Figure 1G-H, Rab27A mRNA and protein were extremely overexpressed in IECs of DSSinduced mice. Taken together, these data indicate that Rab27A is dysregulated in ulcerative colitis.

| Knockdown of Rab27A reduced UC progression in inflammatory colonic cells
Lipopolysaccharide (LPS), the major outer membrane constituent of Gram-negative bacteria, stimulates production of pro-inflammatory cytokines such as IL-1β and TNF-α. 18 To identify the regulatory role of Rab27A in inflammatory colonic cells, HT-29 and Caco-2 cells were induced using 10 ng/mL LPS to establish ulcerative colitis cell models, which were confirmed by detecting the inflammatory factors

| Knockdown of Rab27A suppressed UC progression in DSS-induced colitis mice
To improve our understanding of the functions of Rab27A in UC, we  Figure 3D). Furthermore, Lv-shRab27A could reduce cellular apoptosis ( Figure 3E) and ROS production ( Figure 3F) in the Lv-shRab27A/ DSS group. Therefore, these results suggested that knockdown of Rab27A might inhibit the pathogenesis of ulcerative colitis.

| Rab27A directly interacts with miR-124-3p in ulcerative colitis
In recent decades, the knowledge of miRNAs in UC has expanded, indicating that miRNAs play an important role in regulating inflammatory processes. 7,8 In this study, we have been suggested that miRNA regulated Rab27A expression in UC patients. We used three bioinformatic software programs (miRDB, microRNA.org and TargetScan) to predict miRNAs that could potentially target Rab27A. We identified 6 miRNAs as candidates that could potentially bind to the Rab27A  Figure S2). Similarly, the levels of Rab27A were found to inversely correlate with those of miR-124-3p but not miR-96-5p in UC tissues ( Figure 4C). Subsequently, we revealed that miR-124-3p expression was notably increased by treatment with Lv-shRab27A and significantly repressed by treatment with Lv-Rab27A in inflammatory colonic cells, whereas miR-96-5p expression observed no changes ( Figure 4D-E). Next, we constructed two luciferase reporter plasmids with either a wild-type or mutant Rab27A 3'-UTR (mutant miR-124-3p-binding site) ( Figure 4F). The miR-124-3p mimics notably suppressed the luciferase reporter activity of the wild-type Rab27A 3'-UTR, whereas no significant difference was showed in the activity of the mutant Rab27A 3'-UTR ( Figure 4G). Overall, we speculated that miR-124-3p was the regulatory miRNA responsible for Rab27A 3'-UTR activity in colonic inflammatory cells.
Flow cytology assays revealed that the down-regulation of Rab27A greatly inhibited the cellular apoptosis or ROS production, and miR-124-3p inhibitors could counteract these effects ( Figure 5A,B). In addition, miR-124-3p mimics promoted the effect of Lv-shRab27A, which inhibited the cellular apoptosis or ROS production ( Figure   S3A,B). What's more, up-regulation of Rab27A increased the cellular apoptosis and ROS production compared with those of control groups, but increased Lv-Rab27A-Mut (mutant miR-124-3p-binding site) expression did not enhance these effects ( Figure 5C,D).

| Rab27A promoted UC progression via the STAT3/RelA signalling pathway
It is reported that the STAT3/RelA signalling pathway promotes intracellular ROS production to aggravate disease progression in UC. 22,23 Thus, to further illustrate the potential mechanism involved in Rab27A-associated exacerbated progression of UC, we determined the expression levels of the STAT3/RelA signalling pathway in LPS-treated IECs. Western blotting analysis showed that the STAT3 and RelA expression levels were significantly reduced when F I G U R E 4 Rab27A directly interacts with miR-124-3p in ulcerative colitis. A: Bioinformatic software was used to predict miRNAs potentially targeting Rab27A. B: The levels of miR-124-3p and miR-96-5p in UC and normal tissues were determined by RT-qPCR. C: Rab27A expression was inversely correlated with miR-124-3p but not miR-96-5p levels in UC tissues. D-E: miR-124-3p and miR-96-5p expression levels in colonic inflammatory cells in the presence of Lv-shRab27A (D) or Lv-Rab27A (E) were detected by RT-qPCR. F: The miR-124-3p target site in the 3'-UTR of Rab27A mRNA was predicted, and the mutated site in the 3'-UTR of Rab27A is also shown. G: Luciferase activity was detected after cotransfection of the Rab27A 3'-UTR or its mutant form and miR-124-3p mimics or control mimics in colonic inflammatory cells. Data are presented as means ± SD of three independent experiments. *P < .05, **P < .01, ***P < .001 Rab27A was knocked down ( Figure 7A). Moreover, overexpression of Rab27A caused the opposite results ( Figure 7B). In addition, Lv-shSTAT3 treatment in LPS-IECs interfered with the increase in STAT3 and RelA expression caused by Lv-Rab27A ( Figure 7C). Lv-STAT3 reversed the effect of Lv-shRab27A, which alleviated STAT3 and RelA expression ( Figure 7D). Therefore, these data indicate that Rab27A may promote UC progression via the STAT3/RelA signalling pathway.

| D ISCUSS I ON
In this current study, we illustrated the function and mechanism of Rab27A in inflammatory colonic cells and found that Rab27A mRNA and protein were frequently up-regulated in UC tissues and DSSinduced mouse model. Our findings demonstrated that knockdown of Rab27A reduced apoptosis and ROS production in colonic epithelial cells. Furthermore, we clarified that Rab27A stimulated the STAT3/RelA signalling pathway by binding with miR-124-3p to promote the progression of ulcerative colitis (Figure 9).
In previous studies, Rab27A was shown to regulate tumour cellular proliferation and apoptosis in a number of malignant tumours, such as lung cancer, 24 pancreatic carcinoma 25 and colorectal cancer. 26,27 Moreover, Rab27A regulates inflammatory responses and consequently contributes to neutrophil functions. [28][29][30] In this study, we demonstrated that Rab27A competitively binds miRNA-124-3p to regulate the STAT3/RelA signalling pathway in LPS-induced colonic cells by dual-luciferase reporter assay and Western blotting.
In conclusion, the significant increase in Rab27A mRNA and protein identified in this study make it a potential candidate as a biomarker for UC in the future. In addition to the insights into the pathology of this disease, we found a new pathway in the mechanism of UC: Rab27A, by regulating miR-124-3p, can activate the STAT3/ RelA signalling pathway, which may provide novel therapeutic approaches with great impact in ulcerative colitis.

ACK N OWLED G EM ENTS
This work was supported by the grant from National Natural Science Foundation of China (no. 81873555, 81802308, 81672347 and 81702300).

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sets used and analysed during the current study are available from the corresponding author on reasonable request.