A total of 10 GC tumor samples and their corresponding adjacent tissues were collected from patients (45–60 years old) at The Third Affiliated Hospital of Harbin Medical University form February 2015 to August 2017 (Harbin, China). Informed consent was obtained from all patients and the present study was approved by the Ethics Committee of Harbin Medical University. All tissue samples were immediately frozen and preserved in liquid nitrogen until further use.

GC cell lines, including AGS, SGC7901, MGC803 and BGC823, and human gastric epithelial GES-1 cells were purchased from the Cell Bank of Chinese Academy of Sciences (Shanghai, China). Cells were cultured in Dulbecco's Modified Eagle's medium (DMEM; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) with 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.), 50 U/ml penicillin and 50 µg/ml streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.). All cells were maintained at 37°C in a humidified incubator at 5% CO₂.

The sequences of the miR-92a mimics (cat. no. miR10000092-1-5), mimics-negative control (NC; cat. no. miR01101-1-5), miR-92a inhibitor (cat. no. miR20000092-1-5) and inhibitor-NC (miR02201-1-5) were obtained from Guangzhou Ribobio Co., Ltd. (Guangzhou, China). miR-92a mimics or inhibitor (20 nM) was used for transfection with by Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) with serum-free medium according to experiments request. The sequences were as follows: mimics-NC, 5′-UUUGUACUACACAAAAGUACUG-3′ and 3′-AAACAUGAUGUGUUUUCAUGAC-5′; miR-92a mimics, 5′-UAUUGCACUUGUCCCGGCCUGU-3′ and 3′-AUAACGUGAACAGGGCCGGACA-5′; inhibitor-NC, 5′-CAGUACUUUUGUGUAGUACAAA-3′ and miR-92a inhibitor, 5′-ACAGGCCGGGACAAGUGCAAUA-3′. SGC7901 cells (8×10³) were seeded into 6 cm dishes 24 h prior to transfection. miR-92a, inhibitor or inhibitor-NC were transfected using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) with serum-free medium according to the conditions of the subsequent experiments. After 5 h, the cell medium was replaced with complete medium. The cell lysates were harvested 48 h post-transfection.

Cells and tissues were lysed and total RNA was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocols and transcribed into cDNA using Superscriptase II (Invitrogen; Thermo Fisher Scientific Inc.). qPCR was performed using Power SYBR Green PCR master Mix (Thermo Fisher Scientific, Inc.). The following primer sets were used for ING2 detection. ING2, forward 5′-GCAGCAACTGTACTCGTCG-3′, reverse, 5′-GACTCCACGCACTCAAGGTA-3′; and β-actin, forward 5′-CATGTACGTTGCTATCCAGGC-3′, reverse, 5′-CTCCTTAATGTCACGCACGAT-3′.

miR-92a expression was measured using the TaqMan MicroRNA assay kit (Thermo Fisher Scientific, Inc.). In brief, 1 µg of RNA was reverse transcribed into cDNA using an miR-92a specific stem-loop primer at 25°C for 10 min, 37°C for 120 min, 85°C for 5 min, and 4°C for 5 min. qPCR with miR-92a specific primers and Taqman probes was performed on an ABI7500 real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.), over 35 cycles of 95°C, 30 sec at 60°C and 72°C for 35 sec. The data were analyzed using 2^−ΔΔCq method (15). The sequence of miR-92a and U6 were as follows: miR-92a, forward, 5′-GCTGAGTATTGCACTTGTCCCG-3′, reverse, 5′-GTGTCGTGGAGTCGGCAA-3′ and U6, forward 5′-CTCGCTTCGGCAGCACA-3′ and reverse, 5′-AACGCTTCACGAATTTGCGT-3′.

SGC7901 cells were transfected with miR-92a inhibitor or inhibitor-NC. Cells (5×10³) were incubated with AO/EB mixing solution for 5 min (Beijing Solarbio Science and Technology Co., Ltd., Beijing, China) at room temperature. A total of eight fields were randomly analyzed to examine the cellular morphology alterations under a fluorescence microscope (magnification, ×200). The percentage of apoptotic cells was calculated by the following formula: Apoptotic rate (%) = number of apoptotic cells/number of all cells counted.

Wild-type ING2 (ING2-WT) and mutant ING2 (ING2-Mut) 3′-untranslated region (UTR) sequences were separately inserted into the SpeI and HindIII sites of pMIR-REPORT Luciferase vectors (Ambion; Thermo Fisher Scientific, Inc.). SGC7901 cells were seeded in 6-well plates and transfected with the specific vectors using Lipofectamine 2000 for 48 h. Luciferase activity was assessed using the Dual Luciferase-reporter 1000 assay system immediately after transfection (Promega Corporation, Madison, WI, USA). Renilla activity was used for normalization.

Cells were lysed with radioimmunoprecipitation assay lysis buffer containing a protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland). Protein concentration was measured using a BCA Protein assay kit (Beyotime Institute of Biotechnology, Shanghai, China). Equal amounts of protein (40 µg) were separated by 12.5% SDS-PAGE and transferred to a polyvinylidene difluoride transfer membrane (Thermo Fisher Scientific, Inc.). Following blocking with 5% skim milk for 2 h at room temperature, the blots were probed with primary antibodies against β-actin (cat. no. 3700, dilution: 1:20,00), B-cell lymphoma 2(Bcl-2)-associated X (Bax; cat. no. 2772, dilution: 1:1,000), Bcl-2-associated death promoter (Bad; cat. no. 9292, dilution: 1:500), p53 (cat. no. 2524, dilution: 1:500) and cleaved caspase-3 (cat. no. 9953, dilution: 1:1,000) (Cell Signaling Technology, Inc., Dallas, TX, USA), proliferating cell nuclear antigen (PCNA) (cat. no. sc-25280, dilution: 1:1,000), cyclin dependent kinase (CDK)4 (Cat. sc-70832, dilution: 1:1,000) and CDK6 (cat. no. sc-7961, dilution: 1:1,000) (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and ING2 (cat. no. ab109504, dilution: 1:1,000) (Abcam, Cambridge, UK). Following washing with PBST and incubating with rabbit or mouse secondary antibodies (Cell Signaling Technology, Inc.), the blots were visualized by an enhanced chemiluminescence reagent (GE Healthcare, Chicago, IL, USA). Three individual experiments were performed.

SGC7901 cells transfected with miR-92a inhibitor or inhibitor-NC were seeded into 96-well plates with a density of 5×10³ per well and cultured for 4 days. Cell viability was assessed with a CCK-8 kit (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) at days 0, 2 and 4 at 450 nm with a microplate reader (Thermo Fisher Scientific, Inc.). To assess the cell viability upon doxorubicin treatment, the cells transfected with miR-92a inhibitor or inhibitor-NC were seeded into 96-well plates with a density of 5×10³ per well and treated with various concentrations 5, 10, 20, 50, 100, 200, 300 and 500 nM) of doxorubicin for 24 h. Cell viability was measured at 450 nm with a microplate reader.

A total of 8×10² SGC7901 cells transfected with miR-92a inhibitor or inhibitor-NC were counted and seeded into 6 cm dishes. After 10 days of culturing, the colonies were stained with 0.1% crystal violet in 20% methanol for 15 min at room temperature. The samples were imaged and the numbers of visible colonies were counted.

SGC7901 cells were seeded on cover slips and transfected with miR-92a inhibitor or inhibitor-NC. After 48 h, the cells were fixed with 4% PFA at room temperature for 10 min and then incubated with Ki-67 antibody (cat. no. 9129, Cell Signaling Technology, Inc.) or γH2AX (cat. no. 7631, Cell Signaling Technology, Inc.) for 1 h and then incubated with Alexa-488 conjugated anti-rabbit IgG (cat. no. 4412, Cell Signaling Technology, Inc.) at room temperature for 20 min. H2AX expression was analyzed following treatment with 80 nM doxorubicin for 24 h at 37°C. The cells were then counterstained with 1 µg/ml DAPI at room temperature for 10 min to stain the cell nuclei. All cover slips were mounted using Prolong Diamond Antifade Mountant (Applied Biosystems; Thermo Fisher Scientific, Inc.). Nine random fields per slips were captured for analysis using a fluorescence microscope (magnification, ×200).

Potential miRNA-target gene interactions were predicted using www.Targetscan.org; release 7.2) and microrna.org.

The SGC7901 cell line was continuously exposed to increasing doses of doxorubicin (5, 10 and 15 µM) at 37°C in a humidified incubator at 5% CO₂ (Aladdin Biochemical Technology Co. Shanghai, China) for ~9 weeks. The starting dose was 5 µM and this was increased to 10 µM after 4 weeks, to 15 µM after another 4 weeks and continued at 15 µM for the last 4 weeks. The established resistant SGC7901 cell line was then maintained in DMEM medium with 10% (v/v) FBS and 10 µM doxorubicin.

Data were obtained from at least 3 independent experiments and are presented as the mean ± standard deviation. For the clinical tissue test, the data was evaluated by a paired Student's t test, and data from the other experiments were evaluated by unpaired Student's t-test or analysis of variance followed by a Tukey's post-hoc test to compare multiple groups. P<0.05 was considered to indicate a statistically significant difference. Statistical values were calculated using SPSS 19.0 software (IBM Corp., Armonk, NY, USA) and illustrated using the GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA).

To determine the expression of miR-92a in GC, 10 GC tissues and their corresponding adjacent tissues were collected. Via RT-qPCR, miR-92a was observed to be upregulated by ~6.1-fold in GC tissues than in adjacent tissues (Fig. 1A). In addition, the expression of miR-92a was analyzed in several GC cell lines. As presented in Fig. 1B, the expression levels of miR-92a were significantly increased in all GC cell lines compared with in human gastric epithelial GES-1 cells. These observations suggest a potential role of miR-92a in GC. As SGC7901 cells expressed the highest levels of miR-92a, further experiments were conducted using this cell line.

To investigate the function of miR-92a in GC, SGC7901 cells were transiently transfected with an miR-92a inhibitor or mimics to suppress or upregulate the expression of miR-92a, respectively. The suppression and overexpression efficiencies were confirmed by RT-qPCR at 48 h post-transfection (Fig. 2A). A CCK-8 assay was performed to analyze viable cells at day 0, 2 and 4 post-transfection. The data revealed that suppression of miR-92a significantly suppressed cell viability compared with in the inhibitor-NC-transfected group by a CCK-8 assay (Fig. 2B). The colony formation assay revealed significantly fewer visible colonies upon miR-92a knockdown (Fig. 2C and D). Furthermore, Ki-67 immunofluorescence staining exhibited notably lower signals in miR-92a inhibitor-transfected SGC7901 cells compared with in inhibitor-NC-transfected cells (Fig. 2E). Consistently, western blotting demonstrated that the protein expression levels of PCNA, CDK4 and CDK6 were notably reduced, while p53 was upregulated (Fig. 2F). All these results indicate that miR-92a contributes to the maintenance of proliferation in SGC7901 cells.

To investigate the role of miR-92a in cell apoptosis, we subjected miR-92a inhibitor-transfected SGC7901 cells to AO/EB staining. The results revealed that >50% of apoptotic cells were counted upon miR-92a inhibitor transfection; however, <5% of apoptotic cells were counted upon inhibitor-NC transfection (Fig. 3A and B). In addition, proapoptotic proteins, including Bax, Bad and cleaved caspase-3, were determined to be increased via the suppression of miR-92a compared with in control cells (Fig. 3C).

Conventionally, miRNAs suppress gene expression by binding to the 3′-UTR sequence of target mRNA (16). By employing bioinformatics databases (TargetScan, microrna.org), we observed that ING2 was a potential target of miR-92a (Fig. 4A). Additionally, ING2 expression at the mRNA and protein levels was notably increased following miR-92 inhibitor transfection compared with the control (Fig. 4B and C). To investigate whether ING2 was a direct target of miR-92a, a luciferase reporter construct containing the ING2-WT or ING2-Mut of ING2. Co-transfection of the miR-92a mimic and ING2-WT resulted in increased luciferase activity (Fig. 4D); however, miR-92a inhibitor did not markedly alter the luciferase activities in the ING2-Mut group (Fig. 4D). Thus, these findings support the result that ING2 is a direct target of miR-92a.

It was proposed that suppression of miR-92a led to reduced proliferation and the induction of apoptosis in SGC7901 cells via upregulation of ING2; thus, whether suppression of miR-92a sensitizes SGC7901 cells to doxorubicin treatment was investigated. Compared with in inhibitor-NC-transfected cells, miR-92a inhibitor-transfected cells were more sensitive to doxorubicin-induced growth inhibition as determined by the CCK-8 assay (Fig. 5A). According to Fig. 5A, the half-maximal inhibitory concentration (IC₅₀) of inhibitor-NC transfected SGC7901 cells was ~147.6 nM; however, a notably lower IC₅₀ (~82.1 nM) was calculated in SGC7901 cells following knockdown of miR-92a. The colony formation assay also demonstrated that suppression of miR-92a markedly reduced SGC7901 cell survival upon doxorubicin treatment, compared with control cells (Fig. 5B). The cytotoxic effect of doxorubicin is known to occur via the induction of DNA damage (16). Bioinformatics analysis revealed that ING2 was a potential target of miR-92a (Fig. 4A). Additionally, we reported that ING2 at the mRNA and protein levels was increased by miR-92 inhibitor transfection (Fig. 4B and C). To validate whether ING2 was a direct target of miR-92a, a luciferase reporter construct containing ING2-WT or ING2-Mut of the ING2 was generated. Co-transfection of the miR-92a mimics and ING2-WT resulted in increased luciferase activity (Fig. 4D). However, the miR-92a inhibitor did not change the luciferase activities in the ING2-Mut group (Fig. 4D). Thus, these findings support ING2 as a direct target of miR-92a. Thus, DNA damage foci formation was examined by γH2AX immunofluorescence staining. We observed that miR-92a downregulation was associated with numerous DNA damage foci than in control cells (Fig. 5C). These data suggest that suppression of miR-92a sensitizes SGC7901 cells to doxorubicin treatment.