In total, 32 pairs of human gastric tissue samples were obtained from patients who underwent surgical resection in the Department of Gastrointestinal Surgery at the First Hospital of Wenzhou Medical College (Wenzhou, China) between 2008 and 2012. Patients were aged between 42 and 80 years, with a mean age of 65 years, and were diagnosed with GC based on histopathological evaluation. The matched normal adjacent tissue was obtained from a segment of the resected specimens that was the farthest from the tumor (i.e., >5 cm). The samples were snap-frozen in liquid nitrogen and stored at −80°C. None of the patients received chemotherapy or radiotherapy prior to surgical excision. In addition, the tumor histological grade was staged using the TNM staging of the International Union against Cancer/American Joint Committee on Cancer (AJCC) system (2002) (14). The study was then approved by the Research Ethics Committee of Wenzhou Medical College, and written informed consent was obtained from all patients.

Total RNA was extracted from the specimens using TRIzol reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) following the manufacturer's instructions. In order to remove potentially contaminated DNA, the extracted RNA samples were further purified using DNase I reagent (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. A NanoDrop2000 spectrophotometer (Thermo Fisher Scientific, Inc.) detected RNA purity by measuring the OD₂₆₀/OD₂₈₀ ratio and calculating the RNA concentration and purity. Only an OD₂₆₀/OD₂₈₀ ratio >1.8 was considered suitable for subsequent experiments. At last, the integrity of the RNA was checked by 1.0% denaturing agarose gel electrophoresis (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Xue et al (15) reported a method of detecting miRNA by stem-loop qPCR, where GAPDH RNA was used as an endogenous control. Using this approach, the expression of miR-187 was analyzed in tissue samples. The first-strand cDNA synthesis was performed with the ReverTra Ace^® qPCR RT kit (cat. no. FSQ-101; Toyobo Co., Ltd., Osaka, Japan) using 0.5 µg total RNA as the template and specific reverse primers at 16°C for 30 min, 42°C for 30 min and 95°C for 5 min of reverse transcription. The resulting cDNA was amplified by PCR using miRNA specific primers with SYBR^® Green Real-Time PCR Master Mix (Toyobo Co., Ltd.) and was performed using an Eppendorf Mastercycler Ep Realplex (SABiosciences; Qiagen Sciences, Inc., Gaithersburg, MD, USA). The primers used for qPCR are listed in Table I. PCR parameters were as follows: 95°C for 5 min, followed by 40 cycles of 95°C for 10 sec, 60°C for 20 sec and 72°C for 20 sec. At the end of the PCR cycles, a melting curve analysis was performed, and PCR products were analyzed using 2.5% PAGE electrophoresis.

The relative expression level of miR-187 in tissues was compared to the endogenous control or paired normal tissues. Thus, the 2^−ΔCq and 2^−ΔΔCq methods were used to evaluate the relative expression levels of the target (16). In these analyses; ΔCq=Cq_miR-187-Cq_GAPDH, and ΔΔCq=[(Cq_miR-187-Cq_GAPDH)_tumor]-[(Cq_miR-187-Cq_GAPDH)_control]. When the data were calculated by the ΔΔCq method, the value of the relative expression ratio <1.0 was considered to be low expression in cancer relative to the control, and a ratio >1.0 was considered to indicate a high level of expression. Moreover, all qPCR assays were run in triplicate.

MGC-803 is a poorly differentiated GC cell-line and was purchased from the Institute of Biochemistry and Cell Biology at the Chinese Academy of Sciences (Shanghai, China). MGC-803 cells were cultured in RPMI-1640 medium (Hyclone; GE Healthcare Life Sciences, Chalfont, UK) and were all supplemented with 10% fetal bovine serum (Hyclone; GE Healthcare Life Sciences), 50 U/ml penicillin-G (Invitrogen; Thermo Fisher Scientific, Inc.) and 50 µg/ml streptomycin sulfate (Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C and 5% CO₂ in the air.

The miRNA mimics were designed and synthesized by the GenePharma Company (Shanghai, China). miR-187 mimics (Table II) were an RNA duplex and negative control (NC) RNA duplex and were non-homologous to any human genomic sequences. All the oligonucleotide sequences used in this experiment are listed in Table II. The mimic transfection was performed with Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer's instructions. The GC cells were seeded into 6-well plates and were grown to 60-70% confluence before transfection. Moreover, transfection efficiency was then monitored using qPCR.

The effect of miR-187 on the proliferation of GC cells was evaluated using the CCK-8 assay kit (Dojindo Molecular technologies, Inc., Kumamoto, Japan), according to the manufacturers instructions. MGC-803 cells were plated into 96-well plates at a density of 8×10³ cells/well in three parts. After 24 h of static culture, the cells were transfected with miR-187 mimics and the negative control (NC) using Lipofectamine 2000. The cellular proliferation capacity for the untransfected blank control (Blank), cells transfected with miR-187 mimic NCs and cells transfected with miRNA-187 mimics were measured using the CCK-8 assay. Moreover, the transfected miRNA-187 mimics group was set to final concentrations of 20, 40 and 80 nM. The cells were then incubated for 24, 48 and 72 h at 37°C in an atmosphere of 5% CO₂ in the air. Next, a total 10 µl of the CCK-8 solution was added to each well, and the cells were incubated for a further 4 h. The absorbance (A) values were determined using a spectrophotometer (EPOCH2; BioTeK Instruments, Inc., Winooski, VT, USA) at a wavelength of 450 nm, and the experiment was repeated three times. The inhibition rate of cell proliferation=1-(A_{450 nm} of the study group/A_{450 nm} of the control group) × 100%. In addition, the relative activity of the cells=A_{450 nm} of the study group/A_{450 nm} of the control group.

For the cell cycle analysis, cells were seeded into 6-well plates at a density of 2×10⁵ cells/well, and transfected according to the protocol described above. The experimental groups and time points were set as described for the proliferation assay above. The cells were obtained after being cultured for an appropriate time by trypsinization, and were then pooled with the floating cells, centrifuged at 1,000 × g for 5 min at 4°C and stored at −20°C overnight. DNA staining solution that contained (PI) iodide and RNaseA reagent [MultiSciences (Lianke) Biotech Co., Ltd., Hangzhou, China] were added to the cells. Moreover, cell cycle analysis was performed by fluorescence-activated cell sorting flow cytometry using CellQuest software 1.0 (BD Biosciences, San Jose, CA, USA).

miR-187 mimics (5′-UCGUGUCUUGUGUUGCAGCCGG-3′; Applied Biological Materials, Inc., Richmond, BC, Canada) or negative control oligonucleotides (5′-UUCUCCGAACGUGUCACGUTT-3′; Applied Biological Materials, Inc.) at a concentration of 40 nM were transfected into GC MGC-803 cells using the Lipofectamine™ 2000 Transfection Reagent kit (cat. no. 11668019; Invitrogen; Thermo Fisher Scientific, Inc.) and cells were harvested 48 h after transfection. Proteins were extracted and separated on a 12% SDS-PAGE gel and transferred onto nitrocellulose membranes (Bio-Rad Laboratories, Inc.). The membrane was then blocked with 5% non-fat milk proteins and incubated with anti-MAD2L2 (cat. no. ab180579; 1:2,000), anti-STOML2 (cat. no. ab37531; 1:2,000; each Abcam, Cambridge, UK) or anti-β-actin antibody (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). After being washed extensively, a goat anti-mouse secondary antibody (Pierce Biotechnology, Inc., Rockford, IL, USA) was added to the system. The proteins were detected using enhanced chemiluminescence reagents (cat. no. 32106; Pierce Biotechnology, Inc.). Protein expression levels were quantified using Image-Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA).

In order to verify the direct interaction of miR-187 to the target genes MAD2L2 and STOML2, human mRNA sequence were cloned into the pMIR-reporter construct (Ambion; Thermo Fisher Scientific, Inc.) in order to synthesize the luciferase reporter construct. Wild-type and mutant MAD2L2 and STOML2 mRNA fragments were amplified and sub-cloned into HindIII sites of the luciferase reporter, and luciferase reporter assays were performed as previously described (17). MGC803 cells were co-transfected with 50 nM single-stranded miRNA mimics or internal control oligonucleotides, then plated into 24-well plates with 10 ng pRL-TK (Promega Corporation, Madison, WI, USA) and 50 ng firefly luciferase reporter using the JetPRIME reagent (Polyplus-transfection) according to the manufacturer's instructions. Cells were collected at 48 h after transfection (performed as aforementioned) and analyzed using the Dual-Luciferase Reporter Assay System (Promega Corporation).

Paired paraffin-embedded tissue sections (n=20) were cut into 5-µm sections, deparaffinized in xylene and rehydrated in graded series of ethanols followed by heat-induced epitope retrieval in citrate buffer (pH 6.0). The expression levels of MAD2L2 and STOML2 were detected using polyclonal antibodies targeted to MAD2L2 (cat. no. ab180579; 1:50) and STOML2 (cat. no. ab37531; 1:50; each Abcam). Next, samples were incubated with horseradish peroxidase-conjugated secondary goat anti-mouse antibodies (cat. no. SPKIT-C7; Fuzhou Maixin Biotech Co., Ltd., Fuzhou, China) and visualized with 3,3′-diaminodbenzidine (Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China). Images obtained were processed using Image-Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA).

The expression of miR-187 was calculated for paired groups (i.e., the tumor and paired non-tumor) were compared using non-parametric Wilcoxon tests. The statistical significance of correlation between the expression of miR-187 and the clinicopathological parameters was calculated using non-parametric tests (Mann-Whitney U test between two groups and Kruskall-Wallis test for three or more groups). The association between miR-187 expression and prognosis was analyzed using the Kaplan-Meier survival curve statistics. In the Kaplan-Meier survival curve, high expression was defined as the fold-change of >1 and low expression levels as <1 compared to the probability of survival by the Log-rank test. Statistical analysis was performed using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). Differences were considered statistically significant at a value of P<0.05.

PCR products of miR-187 and GAPDH cDNA revealed a single band at the appropriate position (67 bp for miR-187 and 101 bp for GAPDH) on the electrophoretic gel (Fig. 1A). In addition, the melting curves of the products were sharply defined curves with a narrow peak (Fig. 1B). The combination of melting curves and gel electrophoresis confirmed PCR specificity. Furthermore, the expression of miR-187 in all 32 pairs of GC tissues and their matched non-cancerous tissues was detected using qPCR. As shown in Fig. 1C, the expression levels of miR-187 were lower in gastric tumors than in normal control tissues. The median relative expression levels of miR-187 was 0.031 (25-75th percentile, 0.013-0.102) in tumor samples, and was compared to that in non-tumor control samples, which was set at 0.086 (25-75th percentile, 0.028-0.192). Moreover, the difference in expression of miR-21 between the tumor and the control samples was statistically significant (P=0.037, Wilcoxon test).

Τhe association between miR-187 expression and the clinical and pathological characteristics of GC were explored further (Table III). The downregulated expression levels of miR-187 were associated with cell differentiation (P=0.042) and TNM stage (P=0.034) in GC patients. However, no significant association was identified between the expression of miR-187 and the demographic or clinical variables, including gender, age, tumor location, tumor size, depth of tumor invasion and lymph node metastasis (P>0.05). According to the relative expression levels of miR-187 of paired tumor and normal tissues, specimens with a relative expression level <1 were set as Group One, and specimens with a relative value ≥1 were set as Group Two. A Kaplan-Meier survival analysis illustrated that the cohort with higher expression levels of miR-187 demonstrated greater rates of survival than those with lower levels of expression of miR-187 (Fig. 2). Moreover, the 5-year survival rate for Group One was 66.7%, while that for Group Two was 13.5%.

The significant reduction in miR-187 expression in GC samples prompted us to explore the possible biological roles of miR-187 in the mechanism of tumorigenesis. Thus, the effect of miR-187 expressing again was investigated on the proliferative capacity of the GC cell-line. Compared with that of the cells transfected with either the negative or blank control, MGC-803 cells that were transiently transfected with miR-187 mimics had a significant growth inhibition to different degrees (P<0.05; Fig. 3A). Moreover, the cell vitality of GC cells was markedly associated with different transfected concentrations and was time-dependent (P<0.05; Fig. 3A). In order to investigate further whether inhibition of MGC-803 proliferation reflected cell cycle arrest, the progression of cell cycle phases was analyzed by PI staining and flow cytometry. The results revealed that MGC-803 cells that were dose-dependently transfected with miR-187 mimics had an evident effect on cell cycle arrest at the G₀/G₁ phase (Fig. 3B and C) as compared with the blank and the negative control group (P<0.05). However, the proportion of cells that were transfected with different doses of miR-187 mimics had little effect on the cell cycle phases (P>0.05).

As miRNAs function mainly through the inhibition of target genes, the target of miR-187 that functions in the pathogenesis of GC was further analyzed. Based on previously published CLASH data, which provided direct experimental data of miRNA-targeted pairs (18), MAD2L2, STOML2 and tubulin, γ 1 (TUBG1) were identified as potential targets of miR-187. In order to confirm whether these genes that had been regulated by miR-187 were involved in the pathogenesis of GC, MGC-803 cells were transfected with miR-187 mimics or negative control oligonucleotides, and the protein expression levels of the genes were examined by western blotting. Data indicated that the expression of MAD2L2 and STOML2 were consistently and substantially downregulated by the ectopic expression of miR-187, whereas TUBG1 expression was not significantly affected by miR-187 (Fig. 4A).

In order to verify the interaction between miR-187, MAD2L2 and STOML2, luciferase reporter assays were performed in MGC803 cells. The luciferase reporter plasmid with sequences of MAD2L2 and STOML2 mRNA or mutant sequences were co-transfected into MGC803 cells for 48 h with miR-187 or NC, respectively. Moreover, the luciferase activity was measured in transfected cells. The results demonstrated that the reporter plasmid of MAD2L2 and STOML2 mRNA caused a significant decrease in the luciferase activity in cells that were transfected with miR-187. By contrast, the luciferase activity of the reporter plasmid with mutant sequences of MAD2L2 and STOML2 did not change (Fig. 4B). Furthermore, GC tissues with a low miR-187 demonstrated much higher MAD2L2 and STOML2 expression when compared with paired normal gastric tissues (Fig. 5). Altogether, the results suggested that MAD2L2 and STOML2 was the target of miR-187 in GC cells.