Copy number alterations (CNA) and mRNA data for the ZBTB7A and PTEN genes in 441 cases of human gastric adenocarcinoma were downloaded from TCGA database (24,25). Z-score indicates ZBTB7A mRNA expression levels. For the analysis of overall patient survival, the ZBTB7A expression data, along with the survival data, were divided into two groups, ‘ZBTB7A low expression’ and ‘ZBTB7A high expression’ based on the median expression level of ZBTB7A.

The gastric adenocarcinoma cell line SGC-7901 was bought from the Type Culture Collection of the Chinese Academy of Sciences, (Shanghai, China). SGC-7901 cells were cultured in RPMI-1640 medium (cat no. 10-013-CVR; Corning Incorporated, Corning, NY, USA) with 10% fetal bovine serum (FBS, cat no. VS500T; Ausbian, Vian-Saga Company, Shanghai, China; http://www.viansaga.com/h-pd-1.html#_pp=2_731) and 1% penicillin/streptomycin. 293 cells (cat no. R79007; Thermo Fisher Scientific, Inc., Waltham, MA, USA) were cultured in DMEM (cat no. R10-017-CVR; Corning Incorporated) with 10% FBS. All cells were cultured at 37°C in a humidified incubator (MCO-15A; Sanyo, Osaka, Japan) containing 5% CO₂.

Total RNA was isolated with the SuperfecTRI™ reagent (cat no. 3101-100; Shanghai Pufei Biotech Co., Ltd., Shanghai, China). A total of 1 µg RNA was reverse transcribed into cDNA using the Moloney-murine leukemia virus kit (cat no. M1705; Promega Corporation, Madison, WI, USA). qPCR was performed in a Real-Time PCR machine system (cat no. MX3000p; Agilent Technologies, Inc., Santa Clara, CA, USA), using cDNA and SYBR Master mixture (cat no. DRR041B; Takara Biotechnology Co., Ltd., Dalian, China) (26–27). The following cycling conditions were used: One cycle for 30 sec at 95°C, 40 cycles for 5 sec at 95°C and 30 sec at 60°C, then one cycle of dissociation including 15 sec at 95°C, 30 sec at 60°C and 15 sec at 95°C. GAPDH was used as an endogenous control. The primer sequences of GAPDH and ZBTB7A genes were as follows: GAPDH forward, 5′-TGACTTCAACAGCGACACCCA-3′ and reverse; 5′-CACCCTGTTGCTGTAGCCAAA-3′. ZBTB7A forward, 5′-CATCTGCGAGAAGGTCATCC-3′ and reverse 5′-TGTCCTGCCTGGTGAAGC-3′ (26,28).

The pGV115-ZBTB7A-FLAG-green fluorescent protein (GFP)-puro plasmid (20 µg; Shanghai GeneChem Co., Ltd., Shanghai, China) was constructed by inserting a full-length human cDNA of ZBTB7A-FLAG gene into a pGV115-GFP-puro plasmid vector. The pGV115-ZBTB7A-FLAG-GFP-puro plasmid along with another two lentiviral packaging plasmids pHelper1.0 and pHelper2.0 (15 µg each; both from Shanghai GeneChem Co., Ltd.) was cotransfected into 293 cells using Lipofectamine™ reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The lentiviral supernatant was collected, concentrated and purified in the 48–72 h following cotransfection. The SGC-7901 cell line was treated with an equal concentration of 2×10⁸ (PFU/ml) of lentiviral supernatant for transfection. Cells were observed for GFP expression under a fluorescence microscope after 72 h viral transfection.

Protein was extracted using 2X radioimmunoprecipitation assay lysis buffer (cat no. WB-0071; Dingguo Bio Co., Ltd, Shanghai, China) from whole cells. The protein concentration was measured using a bicinchoninic acid protein assay kit (cat no. P0010S). The cell lysate was separated using 10% SDS-PAGE with loading 30 µg protein and then transferred onto a polyvinylidene difluoride membrane (cat no. IPVH00010; EMD Millipore, Billerica, MA, USA) at 4°C and 300 mA for 150 min and blotted with 5% milk in 1X TBST buffer at room temperature for 1 h. Membranes were then blotted with diluted primary antibodies at 4°C overnight for GAPDH (1:5,000; cat no. SC-32233; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), FLAG (1:3,000; cat no. F1804; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and ZBTB7A (1:2,000; cat no. Ab175918; Abcam, Cambridge, UK). Survivin-3FLAG-GFP was used as FLAG positive control. A goat-anti-rabbit secondary antibody (1:5,000; cat no. sc-2004; Santa Cruz Biotechnology, Inc.) were then incubated at room temperature for 1.5 h. Subsequently, Pierce™ ECL western blotting substrate was added for exposure (Thermo Fisher Scientific, Inc.). Each western blot analysis was performed at least three times independently.

A total of 2,000 healthy cells/well were seeded into a 96-well plate (cat no. 3599; Corning Incorporated) with 100 µl medium. A total of 20 µl 5 mg/ml MTT reagent (cat no. JT343; Genview, Beijing, China) was added to each well ~4 h prior to detection. Next, the culture medium was removed and 100 µl/well dimethyl sulfoxide was added. Following 5 min incubation, the optical density of the cells was analyzed at 490/570 nm emission/absorption wavelength on a Tecan infinite machine (cat no. M2009PR; Tecan Group, Ltd., Mannedorf, Switzerland).

Cells were seeded into 6-cm dishes with 4 ml medium following lentiviral transfection. Then cells were collected after 3 days. Transfected cells were trypsinized, washed, fixed for 1 h at 4°C with 75% ethanol and incubated with propidium iodide (PI) dye (cat no. P4170; Sigma-Aldrich; Merck KGaA) for 30 min at room temperature. Stained cells were measured for cell cycle phase distribution using a flow cytometer (Guava^® easyCyte HT; EMD Millipore). Cell cycle data was analyzed using FlowJo software (version 7.6.1; FlowJo LLC, Ashland, OR, USA). The cell cycle assay was repeated three times independently.

Cells were seeded into 6-well plates with 2 ml medium following transfection and were harvested 2 days later. Cells were stained using the Annexin V-APC&PI Apoptosis Detection kit (cat no. 88-8007; eBioscience; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Cells were stained to measure apoptosis using flow cytometry software (version 7.6.1; FlowJo LLC). The apoptosis assay was repeated three times independently.

An equal number of 3×10⁴ cells/well were seeded into a 96-well plate following transfection. The cell monolayer was scratched in a straight line in each well. The line was marked and images were captured under phase-contrast microscope (Zeiss; XDS-100). Cells were cultured for 8 and 24 h. Following incubation, images were retaken in the same region centered on the line. The width was measured and recorded at 0, 8 and 24 h. The migratory rate was calculated as [(width at 0 h - width at 8 or 24 h)/width at 0 h]. The rate of migration was analyzed. The scratch assay was repeated four times independently.

All results were analyzed using GraphPad Prism software (version 5; GraphPad software, Inc., La Jolla, CA, USA) and data were presented as the mean ± standard error of the mean. The data were analyzed using a Student's t-test for comparisons between two groups. Multiple comparison tests were performed using two-way analysis of variance (ANOVA) and Bonferroni post-hoc tests to analyze the data from the cell cycle distribution and migration assays. One-way ANOVA was used to analyze the data from the apoptosis data, followed by Tukey's test. Log-rank test and Kaplan-Meier estimators were performed. P<0.05 was considered to indicate a statistically significant difference.

As frequent chromosomal deletions at the ZBTB7A gene locus have been reported in a number of different types of human cancer (18–20), in the present study the ZBTB7A gene was investigated in human gastric adenocarcinoma. CNA, mRNA expression and overall survival data from 441 patients were downloaded from the TCGA provisional dataset. A total of 8 patients (1.8%) presented with a homozygous deletion of ZBTB7A and 156 patients (35.37%) exhibited hemizygous deletions (Fig. 1A and B). A total of 37.17% of patients with gastric adenocarcinoma exhibited a ZBTB7A gene deletion, with 56.56% exhibiting no deletion (Fig. 1B). The ZBTB7A mRNA expression level of the two gene deletion groups was significantly decreased compared with the normal diploid group (Fig. 1C). The overall survival data were analyzed and it was demonstrated that the ZBTB7A high expression group exhibited a median survival of 57.39 months, which was significantly increased compared with 28.71 months in the ZBTB7A low expression group (Fig. 1D). These results implied that low expression of ZBTB7A was associated with a poor median survival. The data indicated that ZBTB7A may function as potential tumor suppressor in gastric adenocarcinoma. To further investigate this, a gain-of-function experiment was performed for ZBTB7A, and its impact on cell proliferation, apoptosis and migration in the GC cell line SGC-7901 was investigated.

To produce a cell line that overexpressed ZBTB7A, SGC-7901 cells were transfected using a lentiviral vector. GFP expression by the negative control (NC) and overexpression (OE) group confirmed that transfection was successful (data not shown). The amplification curve of the RT-qPCR results confirmed that the mRNA level of the OE group reached a peak more rapidly than the NC group (Fig. 2A). The ZBTB7A/GAPDH mRNA expression analysis demonstrated that the level of mRNA in the OE group was ~16 times that of the NC group (Fig. 2B), suggesting that ZBTB7A mRNA was successfully over-expressed. In the western blot analysis, endogenous ZBTB7A expression was equal in the control (CON), NC and OE groups (Fig. 2C). Ectopic ZBTB7A-FLAG protein was successfully expressed only in OE group with Survivin-3FLAG-GFP serving as positive control (Fig. 2D). The data indicated that the ZBTB7A overexpression cell line was successfully constructed. These cells were used for further assays.

To investigate whether gain-of-function of ZBTB7A affects GC cell proliferation or the cell cycle, a cell cycle and MTT assay were used. In the MTT assay, no significant difference in proliferation between the CON, NC and OE groups was detected (data not shown), which suggested that overexpression of ZBTB7A may not affect cell proliferation. An increased proportion of cells were in the S phase in the OE group compared with the CON or NC groups according to the cell cycle assay (Fig. 3A). Statistical analyses were performed following three repeats. A decreased percentage of cells in OE group were demonstrated to be in the G1 phase compared with the CON (P=0.0007) and NC (P=0.0021) groups, with no difference between the CON and NC groups (Fig. 3B). The P-values for the S phase in the OE group were CON (P=0.0015) and NC (P=0.0022), with no difference between CON and NC groups (Fig. 3B). No difference was detected in the percentage of cells in the G2 phase between the three groups (Fig. 3B). Overexpression of ZBTB7A in the SGC-7901 cell line induced an abnormal number of cells to arrest in the S phase of the cell cycle but without significant impact on cell proliferation.

To further investigate the impact on cell death or apoptosis of overexpressing ZBTB7A, an apoptosis assay was used. In the NC group, there were ~90.8% GFP-positive cells and 92.78% GFP-positive cells in the OE group were detected, with only 0.9% GFP-positive cells in CON group (Fig. 4A). The apoptosis percentage of CON, NC and OE groups was 3.06±0.27, 2.11±0.26 and 4.69±0.12%, respectively (Fig. 4B). Statistical analysis identified that percentage of apoptotic cells in the OE group was significantly increased compared with the CON group (P=0.0007) and NC group (P=0.0001). These data indicated that gain-of-function of ZBTB7A in SGC-7901 cell promoted cell apoptosis.

The migratory and invasive ability is another key characteristic of GC cells, which facilitates metastasis to other organs and results in a poor prognosis in patients (29,30). In the present study, the migration assay test was used to investigate the potential impact of upregulation of ZBTB7A on GC cell migration (Fig. 5). The migratory rate of OE group was 0.17±0.02, in contrast with 0.28±0.03 of NC group and 0.24±0.03 of CON group at 8 h (Fig. 5B). Following 24 h, the migratory rate of the CON, NC and OE groups were 0.46±0.04, 0.49±0.03 and 0.35±0.02, respectively (Fig. 5B). Another three repeats were conducted independently. Statistical analysis demonstrated that the migratory rate of cells in the OE group at 8 and 24 h time points were significantly decreased compared with the CON group and NC group (Fig. 5B), suggesting that overexpression of ZBTB7A repressed GC cell migration.