MiR-876-5p regulates gastric cancer cell proliferation, apoptosis and migration through targeting WNT5A and MITF

MicroRNAs (miRNAs) are reported to play critical roles in various cancers. Recently, mounting miRNAs are found to exert oncogenic or tumor inhibitory role in gastric cancer (GC), however, their potential molecular mechanism in GC remains ill-defined. Currently, we aimed to elucidate the functional and mechanistic impacts of a novel miRNA on GC cellular process. The significant down-regulation of miR-876-5p in GC cells attracted our attention. In function, we performed gain-of-function assays and found that miR-876-5p overexpression repressed proliferative, anti-apoptotic and migratory abilities and epithelial–mesenchymal transition (EMT) of GC cells. By applying bioinformatics prediction and mechanism experiments, we verified that miR-876-5p could double-bind to the 3′ untranslated regions (3′UTRs) of Wnt family member 5A (WNT5A) and melanogenesis associated transcription factor (MITF), thus regulating their mRNA and protein levels. Both WNT5A and MITF were highly expressed in GC cells. Additionally, we conducted loss-of-function assays and confirmed the oncogenic roles of WNT5A and MITF in GC. Finally, rescue assay uncovered a fact that miR-876-5p suppressed GC cell viability and migration, but induced cell apoptosis via targeting WNT5A and MITF. Taken together, we might offer a valuable evidence for miR-876-5p role in GC development.


Introduction
As one of the most frequent cancerous occurring in both male and female, gastric cancer (GC) is ranked third in cancer-related mortality all around the world [1,2]. According to the previous studies, a majority of GC patients were diagnosed in the advanced stage, thereby missing the optimal treatment time for radical gastrectomy [3,4]. In spite of great advances in surgical and comprehensive therapies, the clinical effect of GC patients remains to be improved [5,6]. It is well-known that genetic and epigenetic dysregulations are implicated in GC onset and progression [7,8]. Hence, the investigation of novel diagnostic and prognostic biomarkers in GC has been a research hotspot.
Recent advances in high-throughput genome sequencing technologies have provided a large amount of evidence that most of human transcriptome are non-coding RNAs (ncRNAs) without apparent protein-coding potential, including long ncRNAs, microRNAs (miRNAs), circular RNAs, pseudogenes, etc. [9]. Copious reports reveal that miRNAs are widely involved in various cancerous biological behaviors and pathological processes, including GC [10][11][12]. By completely or incompletely base-pairing with the 3 untranslated region (3 UTR) of target messenger RNA (mRNA), miRNA triggers either mRNA degradation or translational repression [13][14][15]. Compelling evidence has indicated that miRNA is emerging as an indispensable modulator of cellular process, including cell proliferation, migration, apoptosis, differentiation, angiogenesis and immune response in various human diseases, especially human cancers [16][17][18][19][20]. Previously published studies have revealed that aberrantly expressed miRNAs are implicated in GC tumorigenesis, progression and drug resistance [21]. Ding et al. reported that miR-375 frequently down-regulated in GC suppressed cell proliferation through targeting JAK [22]. Shang et al. pointed out that miR-508-5p affects multidrug resistance of GC via regulating ABCB1 and ZNRD1, thereby forming an miR-508-5p/ZNRD1/ABCB1 regulatory loop [23]. Liu et al. unveiled a factor that miR-30a inhibits helicobacter pylori-related GC biological process by double-targeting COX-2 and BCL9 [24]. In addition to its antitumor role, miRNAs may also exert oncogenic effects on GC cellular processes, as exemplified by miR-27a which is commonly overexpressed in GC and promotes proliferation and metastasis of GC cells via inhibiting PHLPP2 and activating AKT/GSK3β pathway [25].
In the current study, we aimed to exploring the biological function and underlying mechanistic involvement of a novel miRNA in GC. Firstly, we selected miR-876-5p as a study focus after qRT-PCR examination. The following gain-of-function assays uncovered that miR-876-5p impaired proliferative, anti-apoptotic and migratory capacities of GC cells, acting as a tumor suppressor. Mechanistically, both Wnt family member 5A (WNT5A) and melanogenesis-associated transcription factor (MITF) were verified as the targets of miR-876-5p in GC. Importantly, WNT5A and MITF regulated positively GC cell viability and migration. Finally, rescue assays suggested that miR-876-5p regulated GC progression through targeting WNT5A and MITF.

5-Ethynyl-2 -deoxyuridine-labeling assay
The 5-ethynyl-2 -deoxyuridine (EdU)-labeling assay was performed to detect GC cell proliferation ability. MGC803 and MKN-45 cells were incubated in 96-well (4 × 10 4 cells per well) plates. Thereafter, cells were treated with 50 μM EdU and incubated for 2 h at 37 • C. After washing with phosphate buffered saline (PBS) for two times, cells were fixed with 4% paraformaldehyde for 30 min at room temperature. Thereafter, cells were washed with PBS after treating with 50 μl glycine (2 mg/ml), decolorizing for 5 min. Cells were permeabilized by 0.5% Triton X-100 for 5 min. 100 μl 1× Apollo reaction cocktail was added and incubated for 30 min. Subsequently, 4,6-diamidino-2-phenylindole (DAPI) was utilized to stain the cell nuclei for 15 min. A fluorescence microscope (Olympus) was used to obtain images. EdU assay was performed more than two times.

Flow cytometer analysis
Treated GC cells were trypsinized, washed twice with cold PBS and resuspended in Annexin V Binding Buffer, followed by double staining with FITC Annexin V and propidiumiodide (PI) using the Annexin V FITC apoptosis detection kit (BD Bioscience, CA, U.S.A.) as instructed by the manufacturer. The apoptotic cells were examined on a flow cytometer (BD Bioscience).

Transwell assay
Transwell assay was conducted and perform to investigate GC cell migratory ability. In brief, 200 μl MGC803 and MKN-45 were seed into the upper Transwell chamber (8-μm pores, Corning, NY, U.S.A.). The lower chamber contained 500 μl DMEM mixed with 10% FBS (Gibco, U.S.A.). After incubating for 24 h, non-migrating cells on the upper surface of the chamber were removed using cotton swab. Migratory cells on the lower surface of the chamber were blocked with paraformaldehyde for 30 min and stained with 0.1% crystal violet for 20 min. Finally, migratory cells were determined using a microscopy (×200 magnification; Nikon, Japan) and counted in five random fields. This experiment was performed at least three times.

Immunofluorescence staining
Treated GC cells were placed into 24-well plates and fixed with 4% paraformaldehyde and permeabilized with 100 μM digitonin, and blocked by 1% BSA for 30 min, followed by overnight incubation at 4 • C with primary antibodies against E-cadherin

Luciferase reporter assay
Luciferase reporter assay was performed in MGC803 and MKN-45 cells. The wild type of WNT5A and MITF (WNT5A-WT and MITF-WT) and mutant type of WNT5A and MITF (WNT5A-MUT and MITF-MUT) were synthesized and purchased from Sangon Biotech (Shanghai, China). Thereafter, WNT5A-WT/MUT and MITF-WT/MUT were respectively co-transfected into MGC803 and MKN-45 cells along with miR-876-5p or miR-NC using Lipofectamine2000 (Invitrogen) in accordance with the user guide. After 48-h transfection, cells were reaped. The luciferase activity of RNAs was detected by the dual luciferase reporter assay system (Promega, Madison, Wisconsin, U.S.A.) and normalized to that of the Renilla luciferase. Luciferase reporter assay was carried out at least three times.

Xenografts experiments
Female 6-week-old athymic BABL/c nude mice commercially obtained from the Animal Center of Guangdong Province (Guangzhou, China) were randomly distributed to four groups (three per group) with different transfections. The logarithmic phase MGC803 cells (5 × 10 4 ) with different transfections were injected into tail vein of the nude mice with sterile 28 gauge needles for detecting the tumorigenesis. The tumors were carefully removed from the bodies of the killed mice after 28 days. Tumor weight and volume were measured, respectively.

Statistical analysis
All data were obtained from at least three replications. Data were showed as mean + − standard deviation (SD). The statistical analyses were performed using SPSS version 17.0 software (Abbott Laboratories, Chicago, IL, U.S.A.). A two-sided Student's t-test and one-way analysis of variance were utilized to ascertain differences between two groups or more than two groups. P < 0.05 was considered statistical significance.

MiR-876-5p targeted WNT5A and MITF in GC cells
Through the intersection of five bioinformatics prediction tools (PITA, miRanda, mircoT, miRmap and RNA22) from Starbase, we screened four putative targets (PALLD, WNT5A, GNG7 and MITF) for miR-876-5p. Venn diagram was shown in Figure 2A. Results of qRT-PCR disclosed that expression levels of both WNT5A and MITF were significantly weakened by miR-876-5p overexpression in MGC803 and MKN-45 cells, while no significant change in levels of PALLD and GNG7 was observed on introduction of miR-876-5p mimics ( Figure 2B). Thus we selected WNT5A and MITF as main targets for subsequent assays. The binding sites between miR-876-5p and WNT5A were displayed in Figure 2C. Luciferase reporter assay revealed that miR-876-5p mimics lessened the luciferase activity of wild type  Figure 2D). Additionally, pull-down assay showed that the biotinylated miR-876-5p (Bio-miR-876-5p-WT) probe enhanced the level of WNT5A than Bio-miR-NC and Bio-miR-876-5p-Mut probes ( Figure 2E). Importantly, WNT5A was noticeably up-regulated in GC cells ( Figure 2F). Result from western blotting indicated that miR-876-5p overexpression repressed WNT5A protein level in GC cells ( Figure 2G). Moreover, the binding sites between miR-876-5p and MITF were shown in Figure 2H. Similarly, luciferase reporter and RNA pull down assays confirmed the binding of miR-876-5p to the 3 UTR of MITF in MGC803 and MKN-45 cells (Figure 2I,J). The up-regulated expression of MITF was found in GC cell lines ( Figure 2K). More importantly, miR-876-5p modulated negatively MITF protein level in GC cells ( Figure 2L). Thus, we identified WNT5A and MITF as targets of miR-876-5p in GC.

Both WNT5A and MITF functioned as oncogenes in regulating GC biological process
Subsequently, we intended to further evaluate the bio-function of WNT5A and MITF in GC cells. Using specific siR-NAs (si-WNT5A#1/2/3 and si-MITF#1/2/3), we silenced expressions of WNT5A and MITF in MGC803 and MKN-45 cells, wherein si-WNT5A#1 and si-MITF#2 exhibited the best knockdown efficacy, respectively (Figures 3A and 4A). CCK-8 assay disclosed a fact that GC cell viability was remarkably weakened by si-WNT5A#1/2/3 and si-MITF#1/2/3, especially by si-WNT5A#1 and si-MITF#1 (Figures 3B and 4B). Next, we estimated the effect of si-WNT5A#1 and si-MITF#1 on GC cell apoptosis and migration. As shown in Figures 3C,D and 4C  that the silenced WNT5A and MITF expressions noticeably enhanced the level of Bcl-2 protein, but lessened the level of Bax protein (Figures 3E and 4E). Transwell assay indicated that migratory capacity of MGC803 and MKN-45 cells was suppressed by the depleted WNT5A and MITF expressions ( Figures 3F and 4F). Moreover, we detected the alteration of EMT-associated protein levels by Western blot and immunofluorescence. It was found that depletions of WNT5A and MITF reversed EMT into MET through strengthening E-cadherin expression and reducing levels of mesenchymal markers ( Figures 3G,H and 4G,H). Hence, we concluded that both WNT5A and MITF exerted pro-tumorigenesis role in GC cellular process.

MiR-876-5p inhibited GC cellular process via targeting WNT5A and MITF in vitro
At length, we designed and performed a series of functional rescue assays to confirm whether miR-876-5p could modulate GC biological behaviors through WNT5A and MITF. Significantly, the enhanced WNT5A and MITF expressions promoted GC cell proliferation in miR-876-5p-overexpressed MGC803 cell line ( Figure 5A). Caspase-3 activity and cell apoptosis was facilitated by miR-876-5p mimics, but mitigated by overexpression of WNT5A and MITF ( Figure 5B,C). In addition, the reduced level of Bcl-2 by miR-876-5p up-regulation was impaired through transfection MGC803 cell with pcDNA-WNT5A and pcDNA-MITF, whereas we observed an opposite impact on Bax protein level ( Figure 5D). MiR-876-5p mimics inhibited GC cell migration, but the enhanced WNT5A and MITF expressions abrogated this inhibitory effect ( Figure 5E). Importantly, overexpression of WNT5A and MITF accelerated EMT process in miR-876-5p-up-regulated MGC803 cell ( Figure 5F,G). Through all findings, we found that miR-876-5p regulated negatively GC cellular process via targeting WNT5A and MITF.

MiR-876-5p restrained GC cell proliferation via targeting WNT5A and MITF in vivo
To investigate whether miR-876-5p affected GC development in vivo, treated MGC803 cells were injected into the tail vein of nude mice. Tumors derived from different groups were observed and photographed. Tumors from the mice injected with miR-876-5p-transfected MGC803 cells were smaller than that derived from mimics control-transfected mice; in miR-876-5p and si-WNT5A or si-MITF group, overexpression of WNT5A or MITF mitigated the inhibitory effects of miR-876-5p on tumor size ( Figure 6A). Consistently, compared with the mimics control group, the tumor volume and tumor weight in the miR-876-5p mimics group were significantly reduced, which can be rescued by ectopic expression of WNT5A or MITF ( Figure 6B,C).

Discussion
Although the early diagnosis and therapy of GC were rapidly developed over the past decades, most of the patients with GC were faced with metastasis recurrence [31][32][33]. It is important for GC patients to search out effective new biomarkers. Recently, miRNA is emerging as a vital regulator, participating in GC development. miRNA-876-5p has been frequently reported as a tumor suppressor in different types of cancer. In hepatocellular carcinoma (HCC), miR-876-5p was revealed to enhance sensitivity to sorafenib and restrain HCC cell proliferation, migration and EMT progression by repression on its target genes [34][35][36]. In addition, miR-876-5p was reported to assume an antitumor role in lung cancer [37], head and neck squamous cell carcinoma [38], and osteosarcoma [39]. The other mature isoform of miR-876, miR-876-3p has been reported to suppresse the durg-resistance of GC cells [40]. However, effect of miR-876-5p on GC progression is still unknown. Hence, we focused on miR-876-5p in the present study.
In our study, up-regulation of miR-876-5p in three GC cell lines was determined by qRT-PCR. In addition, gain-of-function assays uncovered a fact that miR-876-5p overexpression significantly inhibited cell proliferation,  migration and EMT progress, but induced more apoptotic cells in GC. Noticeably, miR-876-5p acted as a tumor repressor in GC via regulating cell proliferation, migration and apoptosis.
To probe into the downstream target genes of miR-876-5p, we applied bioinformatics prediction tools. Among all candidate targets, qRT-PCR disclosed that only WNT5A and MITF were obviously down-regulated in response to miR-876-5p overexpression in GC cells. Of significance, we found the up-regulated WNT5A and MITF expressions. Moreover, it was observed that WNT5A and MITF functioned as targeted mRNAs of miR-876-5p, thus exerting positively functional role in affecting GC cellular process (proliferation, migration and apoptosis). Previous studies have unveiled the oncogenic role of WNT5A and MITF in GC. Kurayoshi et al. unmasked that WNT5A exacerbates the aggressiveness of GC cells [41]. Xue et al. reported that WNT5A is overexpressed in GC and integral to ZEB1-induced progression of GC [42]. Yingduan et al. uncovered that MITF as an oncogene in GC cells was repressed via chromatin modification by ZNF382 [43]. According to the existing studies, WNT5A was identified as a non-canonical ligand in Wnt family, with high expression in several human malignancies including GC [44][45][46]. It is noteworthy that WNT5A could activate the β-catenin-mediated Wnt signaling pathway, thereby driving cancerous progression [47]. Furthermore, as described in previous reports, MITF was similarly linked to Wnt/β-catenin signaling pathway [48], which is considered as one of representative target genes of β-catenin [49]. Therefore, whether miR-876-5p can modulate GC initiation and development via modulating Wnt/β-catenin signaling pathway remains to be further explored in the future.
Summing up the above, we reported the inhibitory role of miR-876-5p in affecting GC biological behaviors, including cell proliferation, migration, EMT process and apoptosis. In mechanism, miR-876-5p targeted WNT5A and MITF, thus exerting functional regulation in GC cells. To our knowledge, this original study might be the first report for the role of miR-876-5p in GC, providing a potential biomarker for GC therapy.