The leaves of K. septemlobus were obtained from Gurye Wild Flower Institute (Gurye, Korea) and authenticated by Professor S.H. Hong, Department of Biochemistry, Dongeui University College of Korean Medicine (Busan, Korea). The dried leaves (50 μg) were cut into small pieces, ground into a fine powder, and then soaked with 500 ml 70% ethanol (500 ml) for 2 days. The extracted liquid was filtered twice through Whatman No. 3 filter paper to remove any insoluble materials and was then concentrated using a rotary evaporator (Rikakikai Co., Tokyo, Japan). The resulting extract (EEKS) was dissolved in dimethylsulfoxide (DMSO; Sigma-Aldrich Co., St. Louis, MO, USA) to a final concentration of 200 mg/ml (extract stock solution) and was subsequently diluted with medium to the desired concentration prior to use.

HepG2 HCC cells and Chang liver cells (an immortalized non-tumor cell line derived from normal liver tissue) were purchased from the American Type Culture Collection (Manassas, MD, USA). The cells were cultured at 37°C in humidified air with 5% CO₂ in RPMI-1640 medium (Gibco-BRL, Gaithersburg, MD, USA), supplemented with 10% fetal bovine serum (FBS, Gibco-BRL), 100 U/ml of penicillin, and 100 mg/ml of streptomycin (Sigma-Aldrich Co.), with or without added EEKS.

Cells were incubated with EEKS, with or without z-Val-Ala-Asp-fluoromethylketone (z-VAD-fmk, a caspase-3 inhibitor; Calbiochem, San Diego, CA, USA) or dorsomorphin dihydrochloride (compound C, an AMPK inhibitor; Tocris Bioscience, Bristol, UK) at serial concentrations. The culture medium was then aspirated and the cells were washed with phosphate-buffered saline (PBS), followed by incubation with 0.5 mg/ml of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich Co.) at 37°C for 2 h. After incubation, the supernatant was removed and the cells were treated with DMSO to dissolve the formazan reaction product. The concentration of formazan was determined by measuring the absorbance at 540 nm using an enzyme-linked immunosorbent assay (ELISA) reader (Molecular Devices, Sunnyvale, CA, USA).

Apoptosis was assessed by washing the cells with PBS, followed by fixation in 3.7% paraformaldehyde (Sigma-Aldrich Co.) in PBS for 10 min at room temperature. The fixed cells were washed, permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature, and then incubated with 2.5 μg/ml 4,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich Co.) for 10 min at room temperature. The cells were washed twice with PBS and images were then captured using a fluorescence microscope (Carl Zeiss, Jena, Germany).

The control and EEKS-treated cells were harvested and washed with PBS, and the pellets were lysed for 1 h on ice in lysis buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 5 mM trypsin-ethylenediaminetetraacetic acid (EDTA), and 0.5% Triton X-100. Following centrifugation at 19,000 x g for 30 min at 4°C, DNA in the supernatant was extracted in an equal volume of neutral phenol : chloroform : isoamylalcohol (25:24:1, v/v/v, Sigma-Aldrich Co.). Equal volumes of DNA samples were then separated by electrophoresis in 1.5% agarose gel at 40 V, stained with 0.1 μg/ml ethidium bromide (EtBr, Sigma-Aldrich Co.), and visualized under ultraviolet light.

After treatment with EEKS, the cells were harvested, washed twice with ice-cold PBS, and fixed with 75% ethanol at 4°C for 30 min. The cells were then stained with 5 μl Annexin V-fluorescein isothiocyanate (FITC) (R&D Systems, Minneapolis, MN, USA) and 5 μl propidium iodide (PI, Sigma-Aldrich Co.). The samples were incubated for 15 min at room temperature in the dark, and then the degree of apoptosis was quantified by flow cytometry as a percentage of the Annexin V-positive and PI-negative (Annexin V⁺/PI⁻ cells) cells (Becton-Dickinson, San Jose, CA, USA).

After treatment with different concentrations of EEKS, the cells were lysed in buffer containing 40 mM Tris (pH 8.0), 120 mM NaCl, 0.5% Nonidet P-40, 0.1 mM sodium orthovanadate, 2 μg/ml aprotinin, 2 μg/ml leupeptin and 100 μg/ml phenymethylsulfonyl fluoride. In a parallel experiment, the mitochondrial and cytosolic fractions were isolated using a mitochondrial fractionation kit (Active Motif, Carlsbad, CA, USA) according to the manufacturer's instructions. The Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA) was used according to the manufacturer's instructions to determine the protein concentrations. For western blot assays, equal amounts of protein (30–50 μg/lane) were subjected to electrophoresis on sodium dodecyl sulfate (SDS)-polyacrylamide gels, and then transferred onto nitrocellulose membranes (Schleicher and Schuell, Keene, NH, USA). The membranes were blocked with Tris-buffered saline (10 mM Tris-Cl, pH 7.4) containing 0.5% Tween-20 and 5% non-fat dry milk for 1 h at room temperature and incubated with the primary antibodies overnight. Membranes were then washed with PBS and incubated with the secondary antibody conjugated to horseradish peroxidase (Amersham Co., Arlington Heights, IL, USA) for 1 h at room temperature. Immunoreactivity was detected by using the enhanced chemiluminescence (ECL) western blotting detection system (Amersham Co.) according to the manufacturer's instructions. The primary antibodies used in this study were as follows: actin (1:1,000, sc-7120; rabbit polyclonal, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), caspase-3 (1:1,000, sc-7272; mouse monoclonal, Santa Cruz Biotechnology, Inc.), caspase-8 (1:1,000, sc-7890; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), caspase-9 (1:1,000, sc-7885; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), PARP (1:1,000, sc-7150; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), XIAP (1:1,000, sc-11426; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), cIAP-1 (1:1,000, sc-7943; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), cIAP-2 (1:1,000, sc-7944; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), TRAIL (1:500, sc-7877; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), DR4 (1:1,000, sc-7863; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), DR5 (1:1,000, sc-65314; mouse monoclonal, Santa Cruz Biotechnology, Inc.), Fas (1:1,000, sc-715; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), FasL (1:1,000, sc-957; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), Bcl-2 (1:1000, sc-509; mouse monoclonal, Santa Cruz Biotechnology, Inc.), Bax (1:1,000, sc-493; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), Bid (1:500, sc-11423; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), cytochrome c (1:500, sc-7159; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), COX IV (1:1,000, sc-376731; mouse monoclonal, Santa Cruz Biotechnology, Inc.), AMPK (1:1,000, sc-25792; rabbit polyclonal, Santa Cruz Biotechnology, Inc.), p-AMPK (1:500, 2535; rabbit polyclonal, Cell Signaling Technology, Inc., Boston, MA, USA), ACC (1:1,000, sc-30212; rabbit polyclonal, Santa Cruz Biotechnology, Inc.) and p-ACC (1:500, 3661; rabbit polyclonal, Cell Signaling Technology, Inc.)

The MMP values were determined using the dual-emission potential-sensitive probe, 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-imidacarbocyanine iodide (JC-1, Sigma-Aldrich Co.), which is internalized and concentrated by respiring mitochondria and can therefore reflect MMP changes in cells. Briefly, cells were fixed and permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature, and then incubated with 10 μM JC-1 for 30 min at 37°C in the dark. Subsequently, the cells were washed with PBS to remove unbound dye, and the amount of JC-1 retained by 10,000 cells per sample was measured at 488 and 575 nm using a flow cytometer.

The activities of the caspases were assessed by marking cells with colorimetric assay kits (R&D Systems), which utilize synthetic tetrapeptides (Asp-Glu-Val-Asp (DEAD) for caspase-3; Ile-Glu-Thr-Asp (IETD) for caspase-8; and Leu-Glu-His-Asp (LEHD) for caspase-9) labeled with p-nitroaniline (pNA) that is linked to the end of the caspase-specific substrate. Briefly, the cells were washed with PBS and lysed in the supplied lysis buffer, according to the manufacturer's instructions. The supernatants were collected and incubated at 37°C for 2 h in the dark with the supplied reaction buffer containing dithiothreitol and DEAD-pNA, IETD-pNA, or LEHD-pNA as substrates. The reactions were measured by changes in absorbance at 405 nm using an ELISA reader.

The experiments were repeated three times, and the results were expressed as means ± standard deviation (SD). A one-way analysis of variance (ANOVA), followed by Dunnett's t-test, was applied to assess the statistical significance of the difference among the study groups. A value of p<0.05 was considered to be statistically significant.

The effects of EEKS on cell viability were evaluated by stimulating the HepG2 cells with various concentrations of EEKS for 24 h, and performing an MTT assay. As shown in Fig. 1A, EEKS caused significant concentration-dependent decreases in cell viability in HepG2 cells. The results of an additional experiment using Chang liver cells, conducted to examine the effect of EEKS on the proliferation of normal cells, are shown in Fig. 1B. EEKS concentrations ≤500 μmg/ml did not induce cytotoxicity, whereas 600 μg/ml EEKS significantly reduced cell viability. Therefore, in this study, conditions for subsequent experiments were chosen that induced no cytotoxicity in Chang liver cells.

The possibility that the growth inhibitory activity of EEKS was caused by an induction of apoptosis was examined based on three established criteria for assessment of apoptosis. First, morphological changes in HepG2 cells were determined using DAPI staining, as shown in Fig. 2A. A significant number of cells showed chromatin condensation, loss of nuclear construction, and formation of apoptotic bodies following EEKS treatment, whereas these features were not observed in control cells. Second, treatment with EEKS induced progressive accumulation of fragmented DNA in a concentration-dependent manner, which appeared as a typical ladder pattern of DNA fragmentation, another hallmark of apoptosis (Fig. 2B). In addition, flow cytometry analysis confirmed an increased accumulation of Annexin V-positive cells following EEKS treatment, indicating a greater degree of apoptosis in cells treated with EEKS (Fig. 2C). These results suggested that EEKS-induced apoptosis in HepG cells contributed to its anti-proliferative and cytotoxic effects.

The molecular mechanism of EEKS-induced apoptosis in HepG2 cells was examined by determining whether EEKS affected activation of caspase cascades-key executioners of apoptosis. Western blot analyses revealed that the expression levels of pro-caspase-8 and -9, upstream activators of caspase-3/-7 in the extrinsic and intrinsic pathways, respectively, were downregulated in a concentration-dependent manner (Fig. 3A). In addition to the reduced expression of pro-caspase-8 and -9, the level of pro-caspase-3 was also concentration-dependently reduced in response to EEKS treatment, and correspondingly, EEKS treatment led to progressive proteolytic cleavage of PARP (116-kDa), a known caspase-3 substrate (47), to yield an 85-kDa cleaved fragment. In addition, equal amounts of proteins from lysates from cells treated with EEKS were assayed to assess in vitro caspase activity. As shown in Fig. 3B, treatment with EEKS significantly increased the activity of caspase-3, -8 and -9.

We further confirmed the involvement of activation of caspases in the apoptosis induced by EEKS by pretreating the cells with a pan-caspase inhibitor, z-VAD-fmk, for 1 h, followed by EEKS treatment. As shown in Fig. 4A and B, pre-treatment with z-VED-fmk significantly prevented the appearance of apoptotic features such as chromatin condensation and DNA fragmentation. Flow cytometry analysis and MTT assay also revealed that the pan-caspase inhibitor suppressed EEKS-induced apoptosis and reduction in viability in HepG2 cells (Fig. 4C and D). These results clearly indicate that EEKS-triggered apoptosis may be mediated through caspase-dependent extrinsic and intrinsic pathways in HepG2 cells. In addition, EEKS treatment reduced the expression of members of the inhibitor of the apoptosis proteins (IAP) family, such as XIAP, cIAP-1, and cIAP-2 (Fig. 5), which bind to caspases and lead to their inactivation (48,49). This finding indicated that EEKS-induced activation of caspases was associated with a reduction in IAP family proteins.

We further explored the induction of apoptosis of HepG2 cells by EEKS by detecting the expression levels of DR-mediated apoptosis regulators and Bid, a BH3 pro-apoptotic protein, which might be involved in the regulation of extrinsic apoptosis pathway. Western blot analysis revealed that the levels of DR ligands, such as the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Fas ligand (FasL), were relatively unchanged in response to EEKS treatment; however, the levels of DR4 and DR5, but not Fas, increased markedly in a concentration-dependent manner (Fig. 3). A subsequent immunoblotting analysis revealed that exposure of HepG2 cells to EEKS caused a progressive accumulation of truncated Bid (tBid) (Fig. 4A), presumably resulting from truncation by activated caspase-8. These results indicate that the cytotoxic effects induced by EEKS could be mediated through DR-mediated apoptosis, thereby accentuating cross-talk between the intrinsic and extrinsic apoptosis pathways.

We next evaluated the effect of EEKS on the mitochondrial apoptosis signaling pathway. Notably, the total levels of pro-apoptotic Bax and anti-apoptotic Bcl-2 proteins remained unchanged in response to EEKS treatment (Fig. 6A); however, EEKS treatment decreased the cytosolic levels of Bax, whereas its mitochondrial levels were increased significantly after treatment with increasing concentrations of EEKS (Fig. 6B). An indicative stage in the intrinsic apoptotic pathway is the depolarization of the mitochondrial membrane and the subsequent increase in permeability of the outer membrane, followed by pore formation and the release of proapoptotic molecules including cytochrome c into cytosol (50,51). The present results demonstrated that treatment with increasing concentrations of EEKS significantly reduced the MMP levels (Fig. 6C), suggesting depolarization of the mitochondria by EEKS. Further characterization of the mitochondrial-mediated apoptotic effect of EEKS, by analyzing the release of cytochrome c, revealed that EEKS treatment promoted a concentration-dependent increase in the release of cytochrome c from mitochondria into the cytosol (Fig. 4B), suggesting that EEKS also activated the mitochondria-mediated intrinsic apoptosis pathway in HepG2 cells.

Several reports have demonstrated that activation of AMPK leads to the induction of apoptosis in numerous human cancer cell types (29–31). We therefore investigated whether the phosphorylation of AMPK is induced by EEKS. As demonstrated in Fig. 7A, when compared with the basal level, EEKS treatment led to increased levels of phosphorylation of AMPK (Thr172) in a concentration-dependent manner, without inducing significant changes in the total protein levels. Consistent with the activation of AMPK, the phosphorylation of acetyl-CoA carboxylase (ACC) at Ser79 (the best-characterized site phosphorylated by AMPK) (52) also increased after EEKS administration, indicating that the AMPK pathway in HepG2 cells was activated in the presence of EEKS.

We further confirmed the relationship between apoptosis induction and AMPK activation in EEKS-stimulated HepG2 cells by examining the effects of compound C, a specific AMPK inhibitor. As shown in Fig. 7B, treatment with compound C blocked phosphorylation of AMPK as well as ACC and, interestingly, prevented cleavage of caspases (caspase-3, -8 and -9) and degradation of PARP in EEKS-treated HepG2 cells, implying a linkage between caspase activation and AMPK activation. In line with these observations, results in Fig. 8 demonstrated that AMPK inhibition significantly reduced the increase in apoptosis induced by EEKS (Fig. 8A–C) and the cell viability loss (Fig. 8D). These results suggested that EEKS-induced apoptosis and loss of viability were mediated by activation of AMPK, and that AMPK was probably upstream of caspase activation in the signaling pathway involved in this process.