This study was approved by the Institutional Ethics Committee of the Cancer Hospital of Harbin Medical University. Written informed consent was also obtained from each patient. Fifty-four pairs of gastric cancer tissues and normal gastric tissues were collected from patients with gastric cancer who underwent surgery at the Cancer Hospital of Harbin Medical University between June 2014 and January 2016. None of these patients included in this research received chemotherapy or radiotherapy prior to surgical resection. Upon resection, tissue samples were immediately snap-frozen in liquid nitrogen and then stored at −80°C.

Five human gastric cancer cell lines, including AGS, MKN-45, SGC-7901, BGC-823, and MGC-803, were purchased from Institute of Biochemistry and Cell Biology at the Chinese Academy of Sciences (Shanghai, China). One human gastric epithelial cell line GES-1 were obtained from American Type Culture Collection (Manassas, VA, USA). All cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (both from Gibco; Thermo Fisher Scientific, Waltham, MA, USA), 100 U/ml penicillin and 100 µg/ml streptomycin in a humidified incubator with 5% CO₂ atmosphere at 37°C.

miR-744 mimics and negative control miRNA mimics (miR-NC) were bought form GenePharma Co., Ltd. (Shanghai, China). The pCMV6-BDNF plasmid and blank vector (pCMV6) were purchased from OriGene Technologies, Inc. (Rockville, MD, USA). For transfection studies, cells were seeded into 6-well plates at a density of 5×10⁵ cells/well. After incubation overnight, cells were transfected with miRNAs or vectors using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Then, the plates were incubated in a humidified atmosphere with 5% CO₂ at 37°C.

Total RNA from tissues or cells was extracted using TRIzol reagent (Thermo Fisher Scientific) according to the manufacturer's protocol. For detection of miR-744, the first-strand cDNA was synthesized using the TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems; Thermo Fisher Scientific). Quantitative PCR (qPCR) was performed on an Applied Biosystems 7500 Sequence Detection system (Thermo Fisher Scientific) using TaqMan MicroRNA PCR kit (Applied Biosystems; Thermo Fisher Scientific), with U6 as an internal control. For detection of brain-derived neurotrophic factor (BDNF) mRNA, total RNA was reverse-transcribed with a PrimeScript RT Reagent kit (Takara Biotechnology Co., Ltd., Dalian, China) and qPCR was conducted with SYBR Premix Ex Taq™ kit (Takara Biotechnology Co., Ltd.), with GAPDH as an internal control. The sequences of the primers were as follows: miR-744 forward, 5′-AATGCGGGGCTAGGGCTA-3′, and reverse, 5′-GTGCAGGGTCCGAGGT-3′; U6 forward, 5′-CGCTTCGGCAGCACATATAC-3′ and reverse, 5′-TTCACGAATTTGCGTGTCAT-3′; BDNF forward, 5′-AGCCTCCTCTTCTCTTTCTGCTGGA-3′ and reverse, 5′-TCCCGCCCGACATGTCCACT-3′; GAPDH forward, 5′-GACTCATGACCACAGTCCATGC-3′ and reverse, 5′-AGAGGCAGGGATGATGTTCTG-3′. Each sample was performed in triplicate and calculated using the 2^−ΔΔCt method (23).

Cell proliferation was analyzed with CCK-8 assay (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer's instructions. Briefly, cells were plated into 96-well plates at a density of 3×10³ cells per well. Cells were transfected with miRNAs or vectors. After incubation of 0, 24, 48, and 72 h, 10 µl CCK-8 regent was added into each well and incubated at 37°C for additional 2 h. The optical density (OD) at 450 nm was determined with a multifunction microplate reader (BioTek, Winooski, VT, USA). Each experiment was performed in triplicate and repeated three times.

Cell invasion assay was performed using a Transwell chamber (8.0-µm pores) coated with Matrigel (both from BD Biosciences, San Jose, CA, USA). After transfection 48 h, transfected cells were collected, and a number of 5×10⁴ transfected cells were plated on the upper Transwell chambers. DMEM medium containing 10% FBS was added into the lower chamber as a chemoattractant. Subsequent to 24 h incubation, the cells remaining in the upper chamber were removed with a cotton swab. The invasive cells were fixed with 100% methanol and stained with 0.5% crystal violet. The invasive cells were imaged and counted using an inverted microscope (Olympus Corporation, Tokyo, Japan). Each experiment was repeated at least three times.

Total protein was extracted from cells or tissue samples using radioimmunoprecipitation (RIPA) lysis buffer supplemented with protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA). We detect the protein concentration by using a BCA protein assay kit (Beyotime, Nanjing, China). Subsequently, equal amounts of proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA). After that, the membranes were blocked in 5% nonfat milk in Tris-based saline Tween-20 (TBST) for 1 h at room temperature, and further incubated overnight at 4°C with mouse anti-human BDNF monoclonal antibody (sc-546; 1:1,000 dilution) or mouse anti-human GAPDH monoclonal antibody (sc-47724; 1:1,000 dilution) (both from Santa Cruz Biotechnology, Santa Cruz, CA, USA). After being washed three times for 10 min, the membranes were further incubated with corresponding horseradish peroxidase (HRP)-conjugated secondary antibody (1:5,000 dilution; sc-2005; Santa Cruz Biotechnology) for 1 h at room temperature. Finally, the protein bands were shown with the application of ECL Immunoblot Detection system (Pierce Biotechnology, Inc., Rockford, IL, USA). GAPDH was used as a loading control, and protein expression was measured using Quantity One software (Bio-Rad, Hercules, CA, USA).

The potential targets of miR-744 were analyzed using TargetScan (http://www.targetscan.org/index.html) and and miRanda (http://www.microrna.org/microrna/).

For luciferase reporter assays, pMIR-BDNF-3′-UTR wild type (Wt) vector and pMIR-BDNF-3′-UTR mutant (Mut) vector were obtained from GenePharma Co., Ltd. Cells were seeded into 24-well plates and cotransfected with pMIR-BDNF-3′-UTR Wt or pMIR-BDNF-3′-UTR Mut, and miR-744 mimics or miR-NC using Lipofectamine 2000 reagent. Following incubation at 37°C for 48 h, the firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay system (Promega, Madison, WI, USA), according to the manufacturer's protocol. Renilla luciferase was chosen as the normalization. Three independent experiments were performed.

Data were presented as mean ± standard deviation, and analyzed using (SPSS) version 18.0 (SPSS Inc., Chicago, IL, USA). The differences between two groups were analyzed using Students t-test, or assessed by one-way ANOVA when there were more than two groups. A two-tailed value of P less than 0.05 was considered statistically significant.

To determine whether miR-744 expression is associated with gastric cancer, we measured the miR-744 expression of five human gastric cancer cell lines (AGS, MKN-45, SGC-7901, BGC-823 and MGC-803) and that of a human gastric epithelial cell line (GES-1). Results showed that miR-744 was significantly downregulated in the gastric cancer cell lines compared with that in GES-1 (Fig. 1A, P<0.05). In addition, we detected miR-744 expression in fifty-four pairs of gastric cancer tissues and normal gastric tissues. As shown in Fig. 1B, gastric cancer tissues had lower miR-744 expression than normal gastric tissues (P<0.05).

To clarify the potential clinical significance of miR-744 in gastric cancer, all patients were divided into two groups according to the median miR-744 value: Low-miR-744 group (n=28) and high-miR-744 group (n=26). As shown in Table I, low miR-744 expression was significantly correlated with lymph node metastasis (P=0.030), invasive depth (P=0.005), and TNM staging (P=0.006) in gastric cancer patients. However, miR-744 expression was not significantly associated with age (P=0.377), gender (P=0.607), tumor size (P=0.599) and differentiation (P=0.761).

To investigate the biological relevance of miR-744 downregulation on gastric cancer progression, we transfected the gastric cancer cell lines SGC-7901 and BGC-823 with either miR-744 mimics or miR-NC. Transfection efficiency was confirmed in both cell lines using RT-qPCR (Fig. 2A and B, P<0.05). The effect of miR-744 overexpression on gastric cancer cell proliferation was determined using CCK-8 assay. We observed that the proliferation of SGC-7901 and BGC-823 cells was inhibited by miR-744 upregulation compared with that of cells transfected with miR-NC (Fig. 2C). Moreover, Transwell cell invasion assay revealed that the restoration of miR-744 expression repressed the invasion capacities of SGC-7901 and BGC-823 cells (Fig. 2D, P<0.05). These results suggest that miR-744 inhibits the malignant behaviour of gastric cancer cells.

To elucidate the underlying mechanism by which miR-744 affects the biological functions of gastric cancer cells, the potential targets of miR-744 were determined using bioinformatic prediction. Among these predicted targets, NKD1 (20), Bcl-2 (24), ARHGAP5 (21), and c-Myc (25) was identified as direct targets. In this study, BDNF was selected for further validation (Fig. 3A) because of its vital role in gastric cancer progression (26). Luciferase reporter assays were conducted to test whether BDNF is a target of miR-744. SGC-7901 and BGC-823 cells were transfected with pMIR-BDNF-3′-UTR wild type (Wt) or pMIR-BDNF-3′-UTR mutant (Mut), along with miR-744 mimics or miR-NC. As shown in Fig. 3B, miR-744 markedly decreased the relative luciferase activities of BDNF-3′-UTR Wt in both SGC-7901 and BGC-823 cells (P<0.05), whereas the cells transfected with BDNF-3′-UTR Mut did not exhibit decreased luciferase activities in the presence of miR-744. Furthermore, we investigated the regulatory effects of miR-744 on BDNF expression by measuring the mRNA and protein levels of BDNF in SGC-7901 and BGC-823 cells transfected with miR-744 mimics or miR-NC. RT-qPCR and Western blot analyses displayed that miR-744 delivery obviously reduced BDNF expression in SGC-7901 and BGC-823 cells at both mRNA (Fig. 3C, P<0.05) and protein (Fig. 3D, P<0.05) levels. Collectively, these results demonstrate that BDNF is a direct target of miR-744 in gastric cancer.

We detected BDNF expression in gastric cancer and normal gastric tissues to further explore the association between miR-744 and BDNF in gastric cancer. RT-qPCR and Western blot analyses indicated that BDNF expression was upregulated in gastric cancer tissues compared with normal gastric tissues at both mRNA (Fig. 4A, P<0.05) and protein (Fig. 4B, P<0.05) levels. Spearman's correlation analysis further showed that miR-744 expression negatively correlated with BDNF mRNA expression in gastric cancer tissues (Fig. 4C; r=−0.6833, P<0.0001).

After identifying BDNF as a direct target of miR-744 in gastric cancer, we next focused on whether BDNF could mediate the biological roles of miR-744 in gastric cancer. SGC-7901 and BGC-823 cells were transfected with miR-NC, miR-744 mimics or miR-744 mimics combined with pCMV6-BDNF. Western blot analysis confirmed that the ectopic expression of miR-744 suppressed BDNF protein expression, whereas co-transfection with pCMV6-BDNF could recover the BDNF expression in SGC-7901 and BGC-823 cells (Fig. 5A, P<0.05). Importantly, the restoration of BDNF expression could reverse the inhibited proliferation (Fig. 5B, P<0.05) and invasion (Fig. 5C, P<0.05) imposed by miR-744 in SGC-7901 and BGC-823 cells. These results indicate that miR-744 inhibits the proliferation and invasion of gastric cancer cells, at least in part, through BDNF downregulation.