GA (purity, 99.5%; melting point, 251°C), MTT, dimethyl sulfoxide (DMSO), rhodamine (RH)-123 and 2,7-dichlorfluorescein-diacetate were obtained from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco's modified Eagle medium (DMEM), RPMI-1640 and fetal bovine serum (FBS) were purchased from Gibco Life Technologies (Carlsbad, CA, USA). Penicillin and streptomycin sulfate were purchased from GE Healthcare Life Sciences (Logan, UT, USA). Mouse anti-B-cell lymphoma (Bcl)-2, apoptosis regulator Bcl-2-like protein 4 (Bax) and cytochrome c immunoglobulin G (IgG) monoclonal antibodies and anti-β-actin IgG polyclonal antibody were purchased from BD Biosciences (San Jose, CA, USA). Horseradish peroxidase-conjugated anti-mouse IgG was purchased from OriGene Technologies, Inc. (Beijing, China). The acridine orange staining kit, Annexin V binding apoptosis assay kit and caspase activity assay kit were purchased from Nanjing KeyGen Biotech Co., Ltd. (Nanjing, China). GA was dissolved in DMSO to produce a stock solution (10 mg/ml), and diluted to various concentrations with double-distilled water containing 10% DMSO.

The HepG2 and SMMC-7721 human HCC cell lines were obtained from the Cell Bank of the Chinese Academy of Science (Beijing, China). Cells were cultured in DMEM medium containing 10% FBS, 100 U/ml penicillin and 100 µg/ml streptomycin sulfate. The HL-7702 human normal hepatocyte cell line was obtained from the Chinese Academy of Science Type Culture Collection (Shanghai, China) and incubated in RPMI-1640 medium supplemented with 10% FBS. All cells were incubated at 37°C in a humidified incubator (Thermo Fisher Scientific, Waltham, MA, USA) in an atmosphere of 5% CO₂. Cells were detached from the monolayer using 0.25% trypsin (Sigma-Aldrich) for 5 min once cells had grown to near confluence.

Cell viability was measured using MTT assays (21). HepG2, SMMC-7721 and HL-7702 cells were seeded at a density of 2×10⁴ cells/well in 96-well plates and allowed to adhere for 24 h. Subsequently, the medium was replaced with 200 µl complete medium (DMEM or RPMI-1640 medium supplemented with 10% FBS), supplemented with various concentrations of GA (0, 6.25, 12.5, 25.0, 50.0 and 100.0 µg/ml). Cells were incubated for 24, 48 and 72 h, and following incubation, 20 µl MTT (5 mg/ml) was added to each well. Cells were incubated for an additional 4 h, the medium was removed, the formazan crystals were dissolved in 150 µl DMSO and the absorbance values were measured at a wavelength of 490 nm using a microplate reader (Synergy™ H4 Hybrid Microplate Reader; RayBiotech, Norcross, GA, USA). The relative cell viability IC₅₀ (concentration inducing 50% cytotoxicity) value following 48 h of treatment was extrapolated from linear regression analysis of experimental optical density (OD) data using WinNonlin software (version 5.2; Pharsight Corp., Princeton, NJ, USA) (22). The experiment was repeated in triplicate.

To investigate the survival of cells treated with GA, colony formation assays were performed as previously described (23). HepG2 and SMMC-7721 cells were seeded (300 cells/well) into a 24-well plate and allowed to adhere for 24 h. Cells were incubated in culture medium containing GA (0, 2.0, 4.0, 6.0, 8.0 and 10.0 µg/ml) for ~6 days. Following incubation, the cells were fixed with methanol (Beijing Dingguo Changsheng Biotech Co., Ltd., Beijing, China) and stained using a Giemsa staining kit (Nanjing KeyGen Biotech Co., Ltd.) and colonies (>50 cells) were counted using an Olympus IX71 inverted microscope (Olympus Corp., Toyko, Japan).

Apoptotic morphology was detected using an Annexin V binding apoptosis kit, according to the manufacturer's instructions (Nanjing KeyGen Biotech Co., Ltd.). In brief, 5×10⁵ SMMC-7721 cells/well were seeded into a 6-well plate containing a cover slip, and cultured overnight at 37°C in a humidified incubator.. Cells were treated with 0, 6.25, 12.5 and 25.0 µg/ml GA for 48 h. The treated cells were washed with phosphate-buffered saline (PBS; Beijing Dingguo Changsheng Biotech Co., Ltd.), fixed in 10% paraformaldehyde (Beijing Dingguo Changsheng Biotech Co., Ltd.) for 10 min and stained with reagent I (Giemsa) from the kit. Typical apoptotic changes to the nuclei were observed using light microscopy (CHK-213; Olympus Corp., Tokyo, Japan), and apoptotic cells were identified according to criteria from a previous study (24).

Apoptotic morphology was detected using acridine orange staining, according to the manufacturer's instructions. Briefly, 5×10⁵ SMMC-7721 cells/well were seeded into a 6-well plate and cultured overnight. Cells were treated with 0, 25.0 and 50.0 µg/ml GA for 48 h. Treated cells were subsequently harvested and stained with acridine orange (100 µg/ml) for 15 min in the dark. Any morphological changes to the nuclei were observed using fluorescence microscopy (Eclipse TS100, Nikon Corp., Tokyo, Japan).

In order to evaluate oligonucleosomal fragmentation, genomic DNA was extracted and examined using agarose gel electrophoresis as previously described (25). SMMC-7721 cells (5×10⁵ cells/well) were seeded into 6-well plates and treated with various concentrations of GA (0, 6.25, 12.5 and 25.0 µg/ml) for 48 h. Subsequently, the cells were collected and the DNA from treated and untreated cells was extracted using a DNA ladder sample preparation kit (Nanjing KeyGen Biotech Co., Ltd.), according to the manufacturer's instruction. DNA samples were separated using 2% agarose gel electrophoresis (Beijing Dingguo Changsheng Biotech Co., Ltd.) at 50 V for 3 h, and subsequently visualized with ethidium bromide using a Gel Documentation system (Far Gene Pouyesh, Tehran, Iran).

The proportion of apoptotic cells was measured using Annexin V-FITC/PI double staining, according to the manufacturer's instructions. SMMC-7721 cells (5×10⁵ cells/well) were seeded into a 6-well plate and cultured overnight. Cells were treated with GA (0, 6.25, 12.5 and 25.0 µg/ml) for 48 h, subsequently harvested and washed twice with ice-cold PBS, and treated according to the manufacturer's instructions. Following harvesting and washing, cells were immediately analyzed using flow cytometry (FCM; COULTER® EPICS® XL™ Flow Cytometer; Beckman Coulter, Inc., Brea, CA, USA).

The levels of caspase-3, −8 and −9 activity were measured using the caspase activity assay kit according to the manufacturer's instructions. Briefly, following GA treatment (0, 6.25, 12.5, 25.0, and 50.0 µg/ml) for 48 h, ~3×10⁵ SMMC-7721 cells were harvested and washed twice with PBS, and subsequently treated with the caspase activity assay kit according to the manufacturer's instructions. Data were obtained in triplicate and expressed as the mean ± standard deviation. The activity of caspase was normalized and expressed as OD_test / OD_control × 100%.

The change in MMP in SMMC-7721 cells was analyzed by flow cytometry using RH-123 staining (26). Briefly, following 48 h of GA treatment (0, 6.25, 12.5 and 25.0 µg/ml), ~3×10⁵ SMMC-7721 cells were harvested and stained with PBS containing 3 µg/ml RH-123 at 37°C, for 30 min in the dark. Stained cells were subsequently washed with ice-cold PBS and the RH-123 fluorescence was detected by FCM.

The production of ROS in SMMC-7721 cells was detected by FCM using dichloro-dihydro-fluorescein diacetate (DCFH-DA; Sigma-Aldrich) as a probe according to previous methodology (27). Briefly, SMMC-7721 cells (5×10⁵ cells/well) were seeded into a 6-well plate and cultured overnight. Cells were treated with GA (0, 6.25, 12.5 and 25.0 µg/ml) for 2 h, and subsequently harvested, washed twice with PBS and incubated with 10 µM DCFH-DA for 30 min at 37°C. The intracellular ROS generation converted non-fluorescent DCFH-DA to the highly fluorescent compound 2,7-dichlorfluorescein (DCF). DCF fluorescence intensity was determined in 10,000 cells using FCM. The proportion of cells in the right quadrant with a higher intensity of DCF fluorescence represented the cells with increased ROS levels.

The expression levels of Bax, Bcl-2 and cytochrome c proteins were measured by western blotting according to the standard protocol (28). SMMC-7721 cells (2×10⁶⁾ were seeded into a 10-cm plate and cultured overnight. Following treatment with GA (0, 6.25, 12.5 and 25.0 µg/ml) for 48 h, whole-cell protein or mitochondrial fractions were isolated using the protein extraction kit (Nanjing KeyGen Biotech Co., Ltd.). Protein concentration was measured using the bicinchoninic acid protein assay kit (Nanjing KeyGen Biotech Co., Ltd.). Protein (50 µg) was separated using 12% SDS-PAGE with the Mini-PROTEAN® system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and subsequently transferred onto a polyvinylidene difluoride membrane (Gene Tech Biotechnology Co., Ltd., Shanghai, China). The membranes were incubated with the specific aforementioned antibodies (1:500 dilution) against Bax, Bcl-2 and cytochrome c for 2 h at 37°C. Following washing with Tris-buffered saline and Tween 20, the membranes were incubated with peroxidase-conjugated IgG goat anti-mouse or anti-rabbit secondary antibody (1:1,000 dilution; OriGene Technologies, Inc.) for 30 min at 37°C, respectively. Immunoreactive proteins were visualized using 3,3-diaminobenzidine staining. β-actin served as a control.

All experimental data are expressed as the mean ± standard deviation, and reproducibility was confirmed in at least three separate experiments. Statistical analysis was performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL, USA).

Fig. 1 illustrates the chemical structure of GA. The present study initially investigated the cytotoxic effects of GA on HepG2 and SMMC-7721 cells by MTT assay. As demonstrated in Fig. 2A and B, the proliferation of HepG2 and SMMC-7721 cells was markedly decreased following GA treatment, in a dose- and time-dependent manner. The IC₅₀ values of GA were 28.5±1.6 µg/ml for HepG2 cells and 22.1±1.4 µg/ml for SMMC-7721 cells at 48 h. This indicated that GA effectively inhibited the proliferation of HepG2 and SMMC-7721 cells, and the anti-proliferative effect of GA on SMMC-7721 cells exceeded the anti-proliferative effect on HepG2 cells, with significantly lower IC₅₀ values of GA for SMMC-7721 compared with HepG2 (P<0.05). As a control, the cytotoxic effects of GA on a HL-7702 human normal hepatocyte line were also investigated. As demonstrated in Fig. 2C, the IC₅₀ value was 80.9±4.6 µg/ml for HL-7702 cells at 48 h. This value was 3-fold higher than the IC₅₀ value observed in HCC cells, which indicated that GA was able to selectively inhibit the growth of HCC cells. Furthermore, the present study investigated the effects of GA on the survival rate of HepG2 and SMMC-7721 cells by colony formation assay. As indicated in Fig. 2D, GA treatment resulted in a significant inhibition in the colony formation ability of HepG2 and SMMC-7721 cells in a dose-dependent manner.

To determine the mode of cell death observed, morphological changes in the nucleus of SMMC-7721 cells following GA treatment were analyzed using Giemsa and acridine orange staining. As demonstrated in Fig. 3, the plasma membrane, nuclear envelope and nucleolus were complete and distinct in untreated SMMC-7721 cells. By contrast, the nuclei of SMMC-7721 cells treated with GA (6.25, 12.5, 25 µg/ml) for 48 h exhibited condensed and fragmented chromatin, indicative of apoptosis.

The present study also investigated the induction of apoptosis in SMMC-7721 cells via the occurrence of DNA fragmentation. It was observed that DNA extracted from untreated SMMC-7721 cells exhibited no fragmentation, while DNA from GA treated cells demonstrated DNA laddering, which occurred as a result of endonuclease action at sites between nucleosomes (Fig. 4). In order to confirm the apoptotic effect of GA on SMMC-7721 cells, Annexin-V FITC/PI apoptosis assays were performed. The results revealed that GA induced apoptosis in SMMC-7721 cells (Fig. 5). The early apoptotic cell proportions in the lower right quadrant (FITC⁺/PI⁻) were 11.8±0.8, 17.7±0.7 and 27.8±0.8% for SMMC-7721 cells treated with 6.25, 12.5 and 25.0 µg/ml GA, respectively. The late apoptotic or necrotic cell proportions in the upper right quadrant (FITC⁺/PI⁺) were 1.8±0.4, 10.2±0.6 and 19.2±0.5%, respectively. These results suggest that GA partially inhibited the proliferation of SMMC-7721 cells via the induction of apoptosis.

To assess whether the GA-induced apoptosis of SMMC-7721 cells was caspase-dependent, the catalytic activity of caspase-3, −8 and −9 was examined. The results (Fig. 6A and B) demonstrated that GA upregulated the activity of caspase-3 and caspase-9 in SMMC-7721 cells in a dose-dependent manner. When SMMC-7721 cells were treated with 12.5 µg/ml GA for 48 h, the relative activities of caspase-3 and −9 were increased 2.0- and 1.5-fold, respectively. When SMC-7721 cells were treated with 25 µg/ml GA for 48 h, the relative activities of caspase-3 and −9 were increased 3.0- and 2.5-fold, respectively. When SMMC-7721 cells were treated with 50.0 µg/ml GA for 48 h, the relative activities of caspase-3 and −9 were increased 3.0- and 2.0-fold, respectively. This decrease in activity compared with the cells treated with 25.0 µg/ml GA may be due to the presence of more necrotic cells. However, the activity of caspase-8 was not markedly altered compared with that of untreated cells (Fig. 6C).

Caspase activation is primarily triggered by a change in MMP, and apoptosis is also associated with loss of the MMP in a number of cases. Therefore, the present study measured the MMP of SMMC-7721 cells using RH-123 dye as marker, and flow cytometric analysis revealed depletion of the MMP. Untreated SMMC-7721 cells possessed intact plasma and a normal MMP, and the cell RH-123 fluorescence was 95.7±0.4%. However, in SMMC-7721 cells treated with 6.25, 12.5 and 25.0 µg/ml GA for 48 h, the levels of RH-123 fluorescence were 86.5±1.4, 77.1±2.2 and 67.7±2.8%, respectively; this fluorescence was statistically different from the untreated group (P<0.05; Fig. 7A). The percentage of SMMC-7721 cells containing quenched RH-123 increased with increasing GA concentration, and the reduction in RH-123 levels indicated a loss of the MMP. These results revealed that GA induced mitochondrial depolarization and a reduction of the MMP in SMMC-7721 cells.

ROS generation may correspond with a reduction in the MMP, and ROS have been observed to be involved in the induction of apoptosis in a number of systems. In order to determine whether ROS production was involved in GA-induced apoptosis of SMMC-7721 cells, the present study detected the ROS levels in GA-treated SMMC-7721 cells using DCFH-DA. Flow cytometric analysis revealed that untreated SMMC-7721 cells possessed low levels of endogenous ROS (2.6±0.2%), whereas treatment with GA significantly enhanced intracellular ROS levels. The levels of ROS generation reached 11.3±0.8, 20.1±2.3 and 47.4±2.8%, when SMMC-7721 cells were treated with 6.25, 12.5 and 25.0 µg/ml GA, respectively, for 2 h; this ROS generation was significantly different from the untreated group (P<0.05; Fig. 7B). These results indicated that treatment with GA elevated intracellular ROS levels in SMMC-7721 cells.

Mitochondria possess a critical role in caspase-dependent cell apoptosis, via the release of cytochrome c (29). Bax is also known to be associated with mitochondrial function. Following translocation to the mitochondrial membrane, Bax initiates cytochrome c release and subsequently activates caspase during apoptosis (30). In order to investigate whether cytochrome c, Bax and Bcl-2 expression were involved in the apoptosis of SMMC-7721 cells, the present study measured the cytochrome c levels in the cytosol and mitochondria, and investigated Bax and Bcl-2 expression using western blotting. The results of these investigations revealed that Bcl-2 protein expression was downregulated and Bax expression was upregulated, and the levels of cytochrome c in the cytosol were increased, accompanied by a reduction in the levels of cytochrome c in mitochondria following treatment with various concentrations of GA for 48 h (Fig. 8).