Primary antibodies against poly(ADP-ribose) polymerase (PARP), caspase-3, -8, -9, -10 and Omi/HtrA2 were purchased from Cell Signaling Technology, Inc. Primary antibodies against proliferating cell nuclear antigen (PCNA), survivin, Cyt c, Bcl-2, Bax and p21 were obtained from Santa Cruz Biotechnology, Inc. Matrigel was obtained from BD Biosciences. The general caspase inhibitor, Z-VAD-FMK, was purchased from R&D Systems. SH (98% purity), provided by the Hunan Zhengqing Pharmaceutical Group Co. Ltd., was directly dissolved in DMEM containing 10% fetal bovine serum for in vitro studies and formulated in physiological saline for in vivo studies.

The human HCC cell lines, Hep3B, SMMC7721, HepG2, HHCC, Bel7402, MHCC97-H and MHCC97-L, were obtained from the Shanghai Institutes for Biological Sciences (SIBS). Cells were grown in DMEM supplemented with 10% fetal bovine serum and maintained in an incubator with a humidified atmosphere containing 5% CO₂ at 37°C.

We conducted 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays as recommended by the manufacturer. Briefly, exponentially growing cells were trypsinized, seeded at an initial density of 8×10³ per well in 96-well plates and incubated in standard growth medium for 24 h prior to being subjected to different treatments for the indicated periods of time. Cells were subsequently incubated in medium containing MTT (0.5 mg/ml) for an additional 4 h. The supernatant was removed and 150 μl DMSO were added to each well to dissolve the intracellular formazan compound. The absorbance was measured with an ELISA reader (Bio-Rad, Hercules, CA, USA) at 490 nm. The relative cell viability (%) was calculated as the percentage between SH-treated cells and the untreated controls.

Cells were grown in 6-cm dishes at a density of 0.5×10⁶ cells/ml and treated with the desired concentrations of SH in complete medium for 48 h. Cells were then collected and fixed in 70% ethanol at −20°C overnight. After being suspended in 0.5 ml of propidium iodide (PI) staining solution (50 μg/ml PI, 100 μg/ml RNase and 0.2% Triton X-100 in PBS) for 30 min at room temperature in the dark, cell cycle distribution was determined by a FACSCalibur flow cytometer (Becton-Dickinson).

Proteins were prepared as described previously (26). After protein quantification, equal amounts of protein (∼30–60 μg) were electrophoretically separated by SDS-PAGE with 12% Tris-Glycine gels and transferred onto nitrocellulose membranes. After being blocked with 5% non-fat dry milk in TBST buffer (weight/volume) for 1 h, the membranes were incubated with the primary antibodies at 4°C overnight, followed by incubation with a suitable horseradish peroxidase-conjugated secondary antibody at room temperature for 1 h. Immunoreactive bands on the membranes were developed using an electrogenerated chemiluminescence (ECL) detection system followed by exposure to X-ray film.

Hoechst staining was performed using an Apoptosis-Hoechst Staining kit (Beyotime), according to the manufacturer’s instruction. Exponentially growing cells were seeded in 6-multiwell plates at a density of 5×10⁵ cells per well, maintained in culture overnight and exposed to SH at 2 μmol/ml for 48 h. Cells were fixed with 4% paraformaldehyde, stained with Hoechst 33528 and then visualized under a fluorescence microscope (Olympus).

Cells were plated on 6-cm dishes at a density of 0.5×10⁶ cells/ml, grown for 24 h and treated with increasing concentrations of SH for 48 h. Cells were then collected and centrifuged. For the detection of DNA ladders, intracellular DNA was extracted using an Apoptosis DNA Ladder Detection kit (Nanjing KeyGen Biotech. Co. Ltd.) according to the manufacturer’s instructions. DNA samples were dissolved in TE buffer (pH 8.0), separated by electrophoresis on a 1.5% agarose gel with ethidium bromide (0.5 μg/ml) and then photographed using a molecular imager (Bio-Rad). Quantitative analysis of apoptosis by flow cytometry was assessed using an Annexin V-FITC Apoptosis Assay kit (Joincare Biosciences, Zhuhai, China), as described in the manufacturer’s recommendations. At least 10,000 cells were analyzed for each sample.

We examined Δψ_m using a Mitochondrial Membrane Potential Assay kit with JC-1 (Beyotime) as described in the manufacturer’s instructions. Cells at a density of 0.5×10⁶ cells/ml were seeded on 6-cm dishes, grown for 24 h, treated with different concentrations of SH for 48 h and stained with JC-1, followed by flow cytometry for the quantitative analysis of Δψ_m. At least 10,000 events per sample were recorded.

Briefly, cells were lysed with permeabilization buffer (250 mmol/l sucrose, 20 mmol/l HEPES/KOH (pH 7.4), 1 mmol/l EGTA, 1 mmol/l EDTA, 1 mmol/l DTT, 0.1 mmol/l PMSF, 1 μg/ml chymostatin, 1 μg/ml leupeptin, 1 μg/ml antiparin and 1 μg/ml pepstatin A). The lysates were centrifuged at 500 × g for 10 min (4°C). The supernatants were further centrifuged at 13,000 × g for 30 min (4°C) and the supernatants containing cytosolic fraction were collected.

The activity of caspase-3 was assessed using a Caspase Colorimetric Assay kit (R&D Systems), following the instructions provided by the manufacturer. Briefly, cells were washed with ice-cold PBS and lysed in a lysis buffer. After being centrifuged at 10,000 × g for 10 min (4°C), cell lysates were incubated with 50 μl of reaction buffer and 5 μl of caspase-3 colorimetric substrate (DEVD-p-nitroaniline) for 2 h at 37°C. The optical density of the color reaction was spectrophotometrically measured at 405 nm.

Four-week-old male BALB/c athymic nude mice, weighing 18–23.5 g, were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. (Shanghai, China) and housed in sterile laminar flow rooms with 12-h light and dark cycles at a temperature range of 19–23°C and a humidity of 40–60% in the Laboratory Animal Centre of Xi’an Jiaotong University. All experimental procedures were conducted in accordance with the institutional guidelines for conduct and animal welfare. Mice were subcutaneously injected with 5×10⁶ of human HCC Hep3B cells suspended in 100 μl of mixture containing serum-free DMEM and matrigel (5:1) in the right hind legs. Thirty mice were randomly divided into the following 4 groups: control group (n=6), SH 50 mg/kg group (n=8), SH 100 mg/kg group (n=8) and SH 150 mg/kg group (n=8). After 24 h, mice in the control group were intraperitoneally (i.p.) injected with 0.1 ml of physiologic saline per 10 g body weight daily and mice in other 3 groups were i.p. injected with 50, 100 and 150 mg/kg doses of SH in 0.1 ml of saline daily for 24 days. Mice and their average food consumption were weighed daily to determine the effects of SH on their general health. Tumor size was also measured daily with vernier caliper and tumor volume was calculated as 0.5236 × length × width × width and expressed as mm³. At the end of the experiment, mice were sacrificed and tumors were immediately removed, weighed and then fixed in 4% paraformaldehyde.

Paraformaldehyde-fixed tissues were embedded in paraffin blocks and cut into 4-μm sections. Haematoxylin and eosin (H&E) staining was conducted according to conventional procedures. Immunohistochemistry staining was carried out using a Dako Autostainer (Dako, Carpinteria, CA, USA) as described previously (27). Briefly, the sections were dewaxed, rehydrated, subjected to antigen retrieval by pressure cooking for 5 min and incubated with serum for 15 min to block endogenous enzyme and non-specific antigens. The sections were incubated with appropriately diluted primary antibodies at 4°C overnight and then with DakoCytomation EnVision-HRP reagent for 30 min, followed by incubation with diaminobenzidine (DAB) in a dark room. After counterstaining with hematoxylin, the sections were then dehydrated. The relative protein expression was evaluated by the average percentage of positive cells (number of positive cells ×100/total number of cells) in 5 different random microscopic fields (×400) in each tumor sample.

Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) assay was performed using an In Situ Cell Death Detection kit, POD (Roche Diagnostics) according to the protocol supplied by the manufacturer. The average apoptotic index of each tumor sample was calculated as described above.

SPSS 16.0 was used to analyze statistical data. Data are expressed as the means ± SE. Statistical comparisons were performed using one-way ANOVA or the Student’s t-test. P<0.05 was considered to indicate a statistically significant difference.

The effect of SH on the growth of HCC cells were evaluated using MTT assay. We employed 7 HCC cell lines representing different degrees of differentiation and from diverse genetic backgrounds to investigate the extent at which SH inhibits the growth of HCC cells. In all the HCC cell lines, treatment with SH at 0.5, 1 and 2 μmol/ml for 24, 48 and 72 h significantly inhibited cell growth in a concentration- and time-dependent manner (Fig. 1B–H).

We hypothesized that the SH-induced cell growth inhibition may be associated with the modulation of cell cycle progression. To determine this possibility, following SH treatment, the cells were stained with PI and then subjected to flow cytometry to analyze cell cycle distribution. In the Hep3B and SMMC7721 cells, it was shown that an accumulation of cells in the G1 phase, accompanied by a decrease in the number of cells in the S phase and an increase in the number of cells in the sub-G1 phase, was caused by SH in a concentration-dependent manner following 48 h of treatment (Fig. 2A). To elucidate the mechanism of the cell cycle arrest induced by SH, we further investigated the effect of SH on the cell cycle regulatory protein, p21. Our results revealed that the protein level of p21 was significantly elevated following 48 h of SH treatment in a concentration-dependent manner (Fig. 2B).

We then assessed whether SH causes apoptotic cell death in Hep3B and SMMC7721 cells. As shown in Fig. 3A, significant morphological changes associated with apoptosis, e.g., formation of condensed and fragmented nuclei, were induced by SH. The formation of a DNA ladder pattern, a biochemical characteristic of apoptosis indicating internucleosomal DNA fragmentation, was markedly detected in a concentration-dependent manner (Fig. 3B). Correspondingly, the increase in Annexin V-positive cells, another feature of apoptosis, was also observed in the SH-treated cells. As shown in Fig. 3C, treatment with 0, 0.5, 1 and 2 μmol/ml of SH for 48 h induced 7.37, 11.72, 22.22 and 48.52% of the apoptotic cell population in the Hep3B cells and 7.45, 37.93, 49.76 and 59.90% in the SMMC7721 cells, respectively.

As the mitochondria play a key role in propagating apoptotic signaling, we examined the effect of SH on Δψ_m by JC-1 staining. Quantitative analysis using flow cytometry revealed that the percentage of cells with collapse of Δψ_m increased following treatment with SH in a concentration-dependent manner in the Hep3B and SMMC7721 cells (Fig. 4A). Additionally, we found that SH treatment gradually increased the levels of apoptogenic Cyt c and Omi/HtrA2 proteins in the cytosolic fraction in a concentration-dependent manner (Fig. 4B). As the altered Bcl-2/Bax ratio is a precursor for the release of apoptogenic proteins, we further examined whether the Bcl-2/Bax ratio decreased upon SH treatment in HCC cells. As shown in Fig. 4C, SH treatment resulted in an upregulation of Bax protein and a downregulation of Bcl-2 protein expression in a concentration-dependent manner.

Based on the above findings, we then determined whether the activation of a caspase cascade is involved in SH-induced apoptosis in Hep3B and SMMC7721 cells. Western blot analysis showed that SH treatment decreased the protein levels of precursors of caspase-8 and -10 and increased the levels of cleaved caspase-9, caspase-3 and PARP, a known substrate of caspase-3 (23,28), in a concentration-dependent manner (Fig. 5A). To gain an insight into the contribution of caspases to SH-induced apoptosis, cells were pre-treated with a general caspase inhibitor, Z-VAD-FMK, prior to SH treatment. We confirmed that pre- treatment with Z-VAD-FMK efficiently inhibited SH-induced caspase activation, which was revealed by the reduction in caspase-3 activity (Fig. 5B) and PARP cleavage (Fig. 5C). As expected, Z-VAD-FMK efficiently suppressed SH-induced apoptosis as shown by flow cytometry analysis (Fig. 5D). These results suggest that the SH-induced apoptosis in HCC cells depends on the activation of a caspase cascade.

As a key regulator of mitosis and apoptosis, survivin inhibits p21 expression and caspase activation and plays an important role in HCC tumor cell proliferation and survival (29–32). We therefore anticipated that SH may decrease the expression levels of survivin in Hep3B and SMMC7721 cells. As shown in Fig. 6, compared with the untreated controls, treatment with SH for 48 h significantly caused a concentration-dependent decrease in the level of survivin protein expression.

To further evaluate the antitumor activity of SH, we conducted an in vivo study with an athymic nude mice xenograft model by subcutaneous inoculation of human Hep3B HCC cells. All animals survived throughout the experiment. Fig. 7A shows that SH at the doses of 50, 100 and 150 mg/kg for 24 days caused a 5.51, 25.10 (P<0.05) and 50.01% (P<0.05) inhibition of tumor volume, respectively, as compared with the controls. Similarly, the tumor weight per mouse in the 50, 100 and 150 mg/kg SH-treated groups was 4.56, 23.89 (P<0.05) and 43.79% (P<0.05) less, respectively, than that in the control group treated with the vehicle (Fig. 7B). We did not find any gross sign of toxicity following SH administration, demonstrated by no obvious changes in body weight and dietary consumption throughout the study (Fig. 7C and D).

The in vivo anti-proliferative effect of SH treatment on HCC xenografts was investigated by PCNA immunostaining. Qualitative analysis showed that the number of PCNA-positive cells decreased from 82.2% in the control group to 78.5, 57.7 (P<0.05) and 38.9% (P<0.05) in the 50, 100 and 150 mg/kg SH-treated groups, respectively (Fig. 8A–C). TUNEL staining was performed to assess the in vivo apoptosis induction by SH administration. The number of apoptotic cells increased from 3.2% in the control group to 4.1, 14.7 and 32.1% (P<0.05) in the 50, 100 and 150 mg/kg SH-treated groups, respectively (Fig. 8D–F). Correspondingly, SH treatment at the doses of 100 and 150 mg/kg increased the expression levels of p21 and cleaved caspase-3 proteins, as compared with the control and 50 mg/kg SH-treated groups (Fig. 8G–L). Additionally, as expected, the decreased number of survivin-stained cells was also observed in the tumor samples from the SH-treated groups (Fig. 8M–O). Overall, these data demonstrate the potential molecular mechanisms by which SH exerts its in vivo anti-neoplastic effect on HCC.