The present study was approved by the Ethics Board of the Institute of the First Affiliated Hospital of Nanchang University (Nanchang, China). The ethics board also supervised and examined the whole process of the present study. All participants agreed to join the present study and provided written informed consent.

Cancer tissue samples and matched normal samples (distance from tumor, ≥5 cm) were obtained from 82 GC patients (43 males and 39 females; mean age, 58.3 years, age range, 33–85) who underwent surgical resection from February 2010 to November 2012 at the Department of General Surgery, First Affiliated Hospital of Nanchang University (Nanchang, China). The samples were immediately stored in liquid nitrogen following removal from patients and detailed clinical data were collected. The histological grade was assessed according to the tumor-node-metastasis (TNM) system (23), established by the Union for International Cancer Control. All patients were monitored every 3 months. The follow-up period ranged from 2 to 80 months (mean duration, 24.5 months). None of the patients received any additional treatment prior to surgery.

Human GC cell lines (SGC7901, HGC27 and BGC823) were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The human gastric mucosa GES1 and HFE145 cell lines were stored in the Central Laboratory of the Center for Experimental Medicine, the First Affiliated Hospital of Nanchang University, Nanchang, China), which were obtained from American Type Culture Collection (Manassas, MA, USA). All cell lines were cultured in Dulbecco's modified Eagle's medium (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.) and 100 U/ml penicillin at 37°C with 5% CO₂. The cells in the exponential phrase were used for experiments.

Total RNA was extracted from cells or tissues using a standard Trizol protocol (Invitrogen; Thermo Fisher Scientific, Inc.). To detect the expression of mRNAs, reverse transcription was conducted using the GoScriptTM Reverse Transcription system (Promega Corporation, Madison, WI, USA) according to the manufacturer's protocol. To detect the level of miRNA expression, reverse transcription was conducted using the EnergicScript^® cDNA Synthesis kit (ShineGene Molecular Biotech, Inc., Shanghai, China). Subsequently, the RT-qPCR detection system FTC-2000 (Funglyn Biotech Inc., Toronto, Canada) was used to measure the levels of mRNA expression and miRNA with Shine-SYBR^® Real Time qPCR MasterMix kit (ShineGene Molecular Biotech, Inc.). The expression of mRNA and miRNA was normalized to endogenous controls β-actin and U6 small nuclear RNA. Relative fold changes were calculated by the 2^−ΔCq method or the 2^−ΔΔCq method (24). The sequences of the primers are listed as follows: BRMS1 forward, 5′-CAGCCTCCAAGCAAAGACAC-3′ and reverse 5′-GCGGCGTCGCTCATAGTC-3′ and miR-125a-5p forward, 5′-GCTCCCTGAGACCCT-3′ and reverse, 5′-GAGCAGGCTGGAGAA. The thermocycler conditions were as follows: 94°C for 4 min, then 35 cycles of 94°C for 20 sec, 60°C for 30 sec, and 72°C for 30 sec.

The human miR-125a-5p plasmids, miR-control plasmids, anti-miR-125a-5p and anti-miR-control were designed and generated by Shanghai GenePharma Co., Ltd., Shanghai, China (Shanghai, China). The siRNA small-interfering (si)RNA (#S1 sequence, 5′-GGAAUAAGUACGGAAUGUGA-3′) that targets the BRMS1 gene was synthesized by Guangzhou RiboBio Co., Ltd., (Guangzhou, China) as previously described (25). Additionally, the coding sequences [without the 5′-untranslated region (UTR) of BRMS1] was amplified by PCR and cloned into the pcDNA3.1 (+) vector (Shanghai GenePharma Co., Ltd). The thermocycler conditions were as follows: 94°C for 5 min, 94°C for 1 min and 50°C for 1 min, followed by 5 cycles of 72°C for 1 min, followed by 25 cycles of 94°C for 1 min, 65°C for 1 min and of 72°C for 1 min, All constructs were verified by sequencing. The miR-125a-5p plasmids, miR-control plasmids, anti-miR-125a-5p and anti-miR-control (40 nM) were transfected using Lipofectamine^® 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) into cells in 12-well plates when 80–90% confluence was reached. A total of 24 h following transfection, the cells were used to perform migration and invasion assays. The RNA analysis and protein analysis were performed at 48 h following transfection. All sequences and primers used in this section were listed in Table I.

The invasive and migratory capacity of the cells was estimated using Transwell chambers (diameter, 6.5 mm, membrane pore size, 8 µm; Corning Incorporated, Corning, NY, USA). To conduct invasion assays, the membranes were coated with 1 mg/ml Matrigel (BD Biosciences, Franklin Lakes, NJ USA), Matrigel was not used for migration assays. The cells (migration assay, 5×10⁴ cells; invasion assay, 1×10⁵ cells) were suspended in 200 µl serum-free medium, and the cells were added to the upper chamber. Then, 600 µl 20% FBS-DMEM was added to the lower chamber. Following incubation at 37°C for 24 h, the non-migrating or non-invading cells were removed with cotton swabs. Finally, invaded cells on the lower side of the filter were fixed with 4% paraformaldehyde for 15 min and stained with 0.1% crystal violet (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) at 37°C for 30 min. The cells were counted in five different fields with a microscope (Olympus Corporation, Tokyo, Japan).

MiR-125a-5p target genes were identified using four web-based bioinformatics algorithms: microRNA.org (http://www.microrna.org/microrna/home.do), MicroCosm Targets (http://www.ebi.ac.uk/enright-srv/microcosm/htdocs/targets/v5/info.html), TargetScan Human (http://www.targetscan.org/vert_50/) and miRTar (http://mirtar.mbc.nctu.edu.tw/human/), which predict miRNA-binding sites based on complementarity to the nucleotide sequence of the miRNA (all accessed on November 10th, 2014). The algorithms used identified highly complementary sites. The results of miRTar indicated that the 5′UTR of BRMS1 binds to miR-125a-5p with a high score.

The 5′-UTR of the BRMS1 segment was amplified by PCR and inserted into the pHY-LV-Report 3.1 vector (http://www.hanyinbt.com; Shanghai, China). The thermocycler conditions were as follows: 94°C for 2 min, followed by 35 cycles at 94°C for 20 sec, 68°C for 40 sec and 72°C for 2 min The mutant of the seed region of the putative miR-125a-5p binding sites in the BRMS1 5′UTR was generated using a QuikChange Site-Directed Mutagenesis kit (Agilent Technologies, Inc., Santa Clara, CA, USA). HGC27 cells (8×10³ cells) were seeded into the 96-well plates 24 h prior to transfection. A mixture of pHY-LV-5′-UTR, negative miR-control (miR-NC) or 40 nM miR-125a-5p plasmids were co-transfected with Renilla into HGC27 cells using Lipofectamine^® 2000 reagent (Invitrogen, Thermo Fisher Scientific, Inc.). 24 h later, the luciferase activity was measured using the Dual Luciferase assay (Promega Corporation). The Renilla reporter vector was used as an internal control to assess the efficiency of transfection. The primer sequences were listed in Table I.

Total protein was extracted from gastric tissues and cells by Total Protein Extraction kit (Nanjing KeyGen Biotech Co., Ltd., Nanjing, China) according to the manufacturer's protocol and using the BAC kit (Tiangen Biotech Co., Ltd., Beijing, China) to detect the concentration of the proteins. The proteins were separated by 12% SDS-PAGE and transferred onto polyvinyl fluoride (PVDF) membranes (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The membranes were blocked with 5% non-fat dried milk for 1 h at room temperature. Then, the membranes were incubated with the primary monoclonal antibody against BRMS1 (1:500; Sigma-Aldrich; Merck KGaA; catalog no. WH0025855M1, Merck KGaA) or β-actin (1:1,000; Cell Signaling Technology, Inc., catalog no. 8H10D10) overnight at 4°C. After washing the membranes with TBST (TBS with 0.1% Tween-20) three times, the membranes were incubated for 2 h at room temperature, with horseradish peroxidase-conjugated secondary antibody (goat anti-mouse IgG; Abcam, Cambridge, UK; catalog no. ab97023) and the ECL Western Blotting Analysis system (GE Healthcare, Chicago, IL, USA) was used to detect the levels of expression of the target proteins. Band intensities were quantified using Image-Pro Plus software (version, 6.0; Media Cybernetics, Inc., Rockville, MD, USA).

All experiments in the present study were repeated at least three times. The data are presented as the mean ± standard deviation and P<0.05 was considered to indicate a statistically significant difference. SPSS (version, 19.0; IBM Corp., Armonk, NY, USA) software was used for statistical analysis. Differences between the groups were estimated using the χ², Student's t-test and one-way analysis of variance with a Student-Newman-Keuls post-hoc test. Survival was evaluated using the Kaplan-Meier method, and the correlation between miR-125a-5p and BRMS1 protein expression level was evaluated using Pearson's correlation.

To determine the level of miR-125a-5p expression in GC, three malignant human GC cell lines (SGC7901, HGC27 and BGC823) and two normal gastric mucosa cell lines (GES1 and HFE145), as well as 82 pairs of cancer tissues and matched normal tissues from patients with GC were used to perform RT-qPCR analysis. It was observed that the levels of miR-125a-5p were significantly lower in the GC cell lines compared with the expression in normal gastric mucosa cell lines (Fig. 1A), whereas no statistical difference in miR-125a-5p expression was indicated between the two normal gastric mucosa cell lines. In patient tissues, miR-125a-5p expression was lower compared with the matched normal tissues (P<0.01; Fig. 1B). In addition, based on clinical progression, the expression of miR-125a-5p was markedly decreased in patients with peritoneal metastasis compared with patients without peritoneal metastasis (P=0.0421; Fig. 1C).

To investigate the associations between miR-125a-5p expression and clinical pathological characteristics, the data of 82 patients was collected from the Pathology Department of the First Affiliated Hospital of Nanchang University, and the detailed information is listed in Table II. The results showed that the expression of miR-125a-5p was significantly associated with lymph node metastasis (P=0.034), peritoneal dissemination (P=0.030) and advanced TNM stage (P=0.025). However, no associations were identified between miR-125a-5p expression and other clinical pathological characteristics. Notably, Kaplan-Meier survival curves of patients with GC indicate that the overall survival rates of GC patients with low expression of miR-125a-5p was significantly shorter compared with patients with high miR-125a-5p expression (P=0.0092; Fig. 1D).

Previously, the association between miR-125a-5p and GC metastasis was observed (as aforementioned) and the expression of miR-125a-5p was downregulated in patients with gastric cancer, it was hypothesized that a lower miR-125a-5p expression in GC tissues was involved in invasion and metastasis of GC cells. In order to investigate the hypothesis, SGC7901 and HGC27 cells were successfully transfected with miR-125a-5p to generate an miR-125a-5p-overexpression model. The miR-control was also transfected. Invasion and migration analyses were then performed. The results showed that the migratory (P<0.05; Fig. 2A) and invasive (P<0.05; Fig. 2B) capacity of SGC7901 and HGC27 cells transfected with miR-125a-5p were significantly lower compared with those of the control group. By contrast, BGC823 cells, with high levels of endogenous miR-125a-5p expression, were transfected with anti-miR-125a-5p. Migration and invasion analyses were subsequently performed. The migration (P<0.001; Fig. 2C) and invasion (P<0.01; Fig. 2D) analyses indicated that the downregulation of miR-125a-5p was able to accelerate the movement of BGC823 cells from the upper chamber to the lower chamber. The level of miR-125a-5p expression in GC cells following transfection is shown in Fig. 1. Taken together, the data was able to confirm the hypothesis that the level of miR-125a-5p expression affects the migratory and invasive abilities of GC cells.

It has been verified that miRNAs generally regulate the expression of target genes to regulate cellular processes associated with cancer, including metastasis (26). Therefore, bioinformatics websites, including microRNA.org, MicroCosm Targets, TargetScan Human and miRTar were used to predict whether miR-125a-5p target BRMS1 genes. Bioinformatics analysis indicated that there is a putative binding site for miR-125a-5p in the 5′-UTR of BRMS1 mRNA (Fig. 3A). To investigate whether BRMS1 is an exact target of miR-125a-5p, the full-length wild-type (Wt-BRMS1-5′UTR) and mutant BRMS1 5′-UTRs (Mt-BRMS1-5′UTR) were amplified and directly fused to the pHY-LV-Report 3.1 vector downstream of the luciferase reporter gene (Fig. 3A). Then, luciferase assays were performed, and HGC27 cells were co-transfected with the vector and either with miR-NC or miR-125a-5p. Luciferase activity was significantly decreased in the group that transfected with miR-125a-5p and Wt-BRMS1-5′UTR compared with the other three groups (P<0.05; Fig. 3B).

To determine the association between miR-125a-5p expression and BRMS1 protein levels, the expression of BRMS1 in 82 GC patient tissues was also detected. The mean levels of BRMS1 protein were also decreased in GC samples compared with the expression in normal gastric samples (Fig. 4A and B). Subsequently, linear correlation analysis was performed to analyze the correlation between BRMS1 protein expression and the levels of miR-125a-5p in GC samples. The findings revealed that the level of BRMS1 protein was positively correlated with the levels of miR-125a-5p (P<0.01; Fig. 4C). In addition, in order to verify that miR-125a-5p is able to regulate BRMS1 protein expression in GC cells, western blot assays were performed to detect the levels of BRMS1 protein in GC cell lines. Compared with GES1 and HFE145 cells, the expression of BRMS1 protein was significantly decreased in GC cells (Fig. 4D).

To further verify the effects of miR-125a-5p on the regulation of BRMS1 expression, RT-qPCR and western blotting were performed to detect the relative mRNA and protein expression level in various GC cell lines (SGC7901, HGC27, BGC823). The results indicated that overexpression of miR-125a-5p (Fig. 5A) increased BRMS1 mRNA and protein expression in SGC7901 (Fig. 5B) and HGC27 cells (Fig. 5C). By contrast, marked suppression of BRMS1 expression was observed in inhibitor-treated BGC823 cells compared with untreated cells and negative control-treated cells (Fig. 5D).

Considering the aforementioned results, whether BRMS1 is a functional target of endogenous miR-125a-5p, which affects GC cells migration and invasion, was investigated. BGC823 cells were transfected separately with si-BRMS1 and anti-miR-125a-5p, and BRMS1 expression was analyzed by RT-qPCR and western blotting. The transfection efficiency of si-BRMS1 and anti-miR-125a-5p was detected (Fig. 6A and B). Furthermore, it was observed that there was no statistical significant difference in the ability to downregulate endogenous BRMS1 expression between anti-miR-125a-5p and si-BRMS1.

Migration and invasion assays revealed that knockdown of BRMS1 mRNA was able to significantly increase the migration and invasion of the BGC823 cells, and similar results were observed when there is a low expression of miR-125a-5p in GC cells (Fig. 6C and D). Moreover, when BRMS1 expression was restored, it was able to counteract the inhibitory effects of anti-miR-125a-5p in BGC823 cells (Fig. 6A-D). Taken together, these results confirmed that miR-125a-5p is able to affect the migration and invasion of GC cells through regulating BRMS1 expression.