HepG2 cells were purchased from the Cellbank of Shanghai Institutes for Life Science, Chinese Academy of Sciences (Shanghai, China) and were maintained at 37°C in an atmosphere of 5% CO₂. Cell culture reagents were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA) unless otherwise indicated. The cells were grown in complete culture medium (high-glucose) consisting of Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% heat-inactivated foetal bovine serum (FBS) and 1% penicillin/streptomycin.

Based on the ATF3 gene sequence in GenBank (NM_001674; http://www.ncbi.nlm.nih.gov/nuccore), primers that specifically matched ATF3 mRNA were designed (forward, GAGGATCCCCGGGTACCGGTC GCCACCATGATGCTTCAACACCCAGG; reverse, TCCTT GTAGTCCATACCGCTCTGCAATGTTCCTTCT; 587 bp). The lentiviral vector GV287-EGFP (Shanghai GeneChem Co., Ltd., Shanghai, China) was used to construct the ATF3 mRNA-expressing vector. The non-transfected group and mock vehicle group transfected with an empty vector without ATF3 mRNA were used as the negative control (NC). The GV287-EGFP vector was linearized by digestion with the restriction endonucleases AgeI and BamHI, and the inserted sequences were ligated with an In-Fusion PCR Cloning kit (Clontech Laboratories, Inc., Mountainview, CA, USA) to produce the human ATF3 mRNA vectors (LV-ATF3-EGFP and LV-NC-EGFP). DNA sequencing was used to verify all inserted sequences. Using Lipofectamine^® 2000 (Invitrogen: Thermo Fisher Scientific, Inc.), 293T cells (Cellbank of Shanghai Institutes for Life Science, Chinese Academy of Sciences) were co-transfected with 20 µg ATF3 mRNA vectors and the 15 µg pHelper1.0 and 10 µg pHelper2.0 plasmids (Addgene, Inc., Cambridge, MA, USA) for 48 h to generate lentiviruses. The viral titres were determined, and the lentiviral particles 2×10⁹ TU/ml were used to infect HepG2 2×10⁵ cells at 37°C for 72 h. Successfully transfected cells, which expressed EGFP, were selected and grown in cell culture for subsequent investigation.

HepG2 cells were lysed in ice-cold lysis buffer [150 mM NaCl, 100 mM Tris (pH 8.0), 1% Tween-20, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml trypsin, and 10 µg/ml leupeptin]. The lysate was left on ice for 20 min and then centrifuged at 4°C, 1,200 × g for 10 min. The clarified supernatant was collected, and the protein concentration was measured by using a Biotek protein assay kit (Elx800; BioTek Instruments, Winooski, VT, USA). Protein lysate (60 µg per well), was separated by 12% SDS-PAGE and transferred to polyvinylidene difluoride membranes. Membranes were incubated with 5% non-fat milk for 1 h at room temperature and then with mouse monoclonal anti-human ATF3 antibody (1:100 dilution, ab58668; Abcam, Cambridge, MA, USA) at 4°C overnight. The membranes were washed and stained with a horseradish-peroxidase-conjugated secondary antibody (at 1:1,000 dilution, A0216, Beyotime Institute of Biotechnology, Shanghai, China). Proteins were visualized using Enhanced Chemiluminescence Plus system (Beyotime Institute of Biotechnology) and exposed to autoradiography film (Kodak, Rochester, NY, USA). Blots with mouse monoclonal β-Actin antibody (at 1:1,000 dilution, AA128; Beyotime Institute of Biotechnology) were similarly generated to ensure that equal amounts of protein were loaded in the wells.

HepG2 cells were divided into three groups: An overexpression group (OE); a mock vehicle-transfected negative control group (NC); and a non-transfection group (control).

Cell proliferation was measured using an MTT colorimetric assay from Beijing Dingguo Changsheng Biotechnology Co. Ltd. (Beijing, China). All the cells were seeded in 96-well plates and cultured with complete culture medium. Cell proliferation was measured on days 1–5 at 490 nm in an ELIASA plate reader (Elx800; BioTek Instruments) following the addition of dimethyl sulfoxide to each well. The cell proliferation rate was calculated relative to the optical density at 490 nm of day 1.

Transwell migration assays were performed in 24-well dishes with cell culture plate inserts (Corning Incorporated, Corning, NY, USA). The upper chamber had a 3 µm pore size and was coated with a polycarbonate membrane, with 100 µl serum-free medium/well, and the plates were incubated at 37°C for 1–2 h. The lower chamber consisted of 600 µl DMEM medium supplemented with 10% FBS. Actively growing cells (1×10⁵ cells) were diluted in 100 µl serum-free medium supplemented with 0.1% bovine serum albumin (Absin Biochemical Company, Shanghai, China), added to the upper chamber and incubated at 37°C for 16 h. Following incubation, the medium from the upper chamber was removed, and the cells on the upper surface of the membrane were rinsed and removed with a cotton swab. The upper chambers were stained at 37°C for 20 min with Giesma solution (Sigma-Aldrich; MercK KGaA, Darmstadt, Germany). Images of the migrated cells were captured by an inverted fluorescence microscope (magnification, ×100, Olympus Corporation, Tokyo, Japan), and their optical densities at 570 nm were measured by ELIASA microplate reader (Elx800; Bio Tek Instruments).

HepG2 cells were collected following infection with the lentiviruses for 5 days; then, the cells were washed twice with cold PBS and binding buffer. The collected cells were resuspended in staining buffer and then added into Annexin V-allophycocyanin (APC) staining solution and incubated for 10–15 min at room temperature in the dark. Flow cytometric analysis was performed using a FACSCalibur instrument (BD Biosciences, Franklin Lakes, NY, USA). The results were analysed by FlowJo7.6 software (FlowJo LLC, Ashland, OR, USA).

HepG2 cells (third to fourth generation) were collected following infection with the lentivirus; then, the collected cells were washed and resuspended with cold PBS. Subsequent to fixation with 70% ethyl alcohol at 4°C overnight, the cell suspension was added into 50 µg/ml propidium iodide staining solution and incubated for 30 min at 37°C in the dark. Flow cytometric analysis was performed using a FACSCalibur instrument, and the results were analysed using FlowJo7.6 software (FlowJo LLC).

The statistical package SPSS 19.0 was used to perform the statistical analyses. P<0.05 was considered to indicate a statistically significant difference.

The optical density (OD) 490 value, OD570/OD490 value, apoptotic rate and cell percentages were representative of the MTT, Transwell, apoptosis and cell cycle results, respectively. All results presented as mean ± standard deviation. Tumour-associated variables in the Transwell and apoptosis results were examined using a one-way analysis of variance (ANOVA) to compare the differences among groups, and followed by a Least Significant Difference test between groups. For the variables associated with the MTT and cell cycle assays, statistical significance was determined according to an ANOVA with a randomized block design, followed by a Least Significant Difference test.

To examine the biological activity of ATF3, ATF3 was cloned into the lentiviral vector GV287-EGFP. Stable pools of HepG2 cells containing full-length ATF3 were generated. The overexpression of ATF3 was confirmed in these cells by fluorescence microscope and western blot analysis. Relative to the NC and control groups, the OE group exhibited a marked increase in ATF3 protein expression (Fig. 1).

The optical density at 490 nm on the first day was considered the reference coefficient. The percent viability was calculated relative to this value as the fold-change (OD490/fold) from the second day to the fifth day. The MTT-OD490 values were representative of cell growth activity. The results indicated that the value in the OE group was significantly decreased compared with that in the NC and control groups beginning from the third day following lentiviral vector infection (P<0.05), whereas no significant differences between the NC and control groups were observed (P=0.637; Fig. 2). These data indicated that the overexpression of ATF3 may inhibit the growth of HepG2 cells.

To examine the effect of cell proliferation on the number of cells that migrated across the Transwell membrane, the optical density at 570 nm was calculated relative to the corresponding MTT-OD490 values. The OD570/OD490 values were representative of the numbers of cells that migrated through the chamber membrane. The results indicated no significant differences in the number of cells that migrated in the OE, NC and control groups, and there were no significant differences between the various groups (P>0.05; Fig. 3). These data indicated that the overexpression of ATF3 had no significant effects on the migration of HepG2 cells.

The APC fluorescence intensity was examined by flow cytometry in the FL4 channel, and the fluorescence intensity of EGFP, which was used as a reporter gene, was measured in the FL1 channel. The percentage of apoptotic cells, which was representative of the cell apoptotic rate, was calculated by comparing these 2 fluorescence intensities. The results demonstrated that the apoptotic rate in the OE group was significantly increased compared with that in the NC and control groups (P<0.05), whereas there was no significant difference between the NC and control groups (P=0.058; Fig. 4). These data suggest that the overexpression of ATF3 may increase the apoptosis of HepG2 cells.

The numbers of cells in each phase of the cell cycle were detected by flow cytometry using a FACSCalibur instrument. The cell proportions in various phases of the cell cycle were calculated to demonstrate the specific cell growth states. The results indicated that the proportion of cells in the G0/G1 phase in the OE group was significantly increased compared with that in the NC and control groups (P<0.05), whereas the proportions of cells in the S and G2/M phases were not significantly different compared with the other groups (P>0.05). There was no significant difference between the NC and control groups in any of the phases of the cell cycle (P=0.183). The sum of the S and G2/M cell proportions was representative of cell proliferation. The results were the same as those of the MTT assay, which suggested that the sum of the S and G2/M cell proportions in the OE group was significantly decreased compared with that in the NC and control groups (P<0.05), whereas there was no significant difference between the NC and control groups (P=0.167; Fig. 5). These data indicated that the overexpression of ATF3 inhibited cell proliferation by inducing cell cycle arrest at the G0/G1 phase.