The Hep3B hepatocellular carcinoma cell line was purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and was cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (Gibco, Gaithersburg, MD, USA). EGCG, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and Hoechst 33342 were obtained from Sigma (St. Louis, MO, USA). Compound C was purchased from Calbiochem (San Diego, CA, USA). Monoclonal antibodies specific for p-AMPKα1 (Thr172), AMPKα1, p-GSK3β (Ser9), GSK3β, p-β-catenin (Ser33/37), p-β-catenin (Ser552) and β-catenin were purchased from Cell Signaling Technology (Beverly, MA, USA). Antibody against lamin B1 was purchased from Santa Cruz (San Diego, CA, USA) and the β-actin antibody was obtained from Sigma.

Cells seeded into 96-well microplates at 4×10³ cells/well, were incubated with test compounds at the indicated concentrations for the indicated time periods. Following incubation with the test compound, the medium was removed, and the cells were then incubated with 10 µl MTT solution (5 mg/ml MTT in PBS) for 1 h. The samples were solubilized in DMSO. The purple formazan dye, converted from MTT by viable cells was quantified by absorbance at 595 nm.

Apoptosis was measured using a FITC-Annexin V apoptosis detection kit (BD Pharmingen, San Diego, CA, USA) or Hoechst 33342 chromatin staining dye. For Annexin V/PI staining after treatment with selenium, cells were harvested by trypsinization, washed with ice-cold phosphate-buffered saline (PBS), and suspended in a binding buffer at a density of 1×10⁶ cells/ml. Cells were stained with Annexin V-FITC and propidium iodide (PI) and analyzed by flow cytometry (Becton-Dickinson Biosciences, Franklin Lakes, NJ, USA). To examine chromatin condensation, cells were stained with 10 µM Hoechst 33342 for 30 min and fixed with 3.7% formaldehyde for 15 min. Changes in chromatin condensation were observed by fluorescence microscopy (Olympus Optical Co., Tokyo, Japan).

Cells were seeded into six-well plates and treated with test compounds. Total proteins were extracted using a RIPA lysis buffer [50 mM Tris-HCl (pH 8.0), 1% NP-40, 0.5% sodium deoxycholate, 150 mM NaCl, 1 mM PMSF] and subjected to western blot analysis with specific antibodies. The proteins were then visualized by enhanced chemiluminescence (Intron, Kyunggi, Korea) and detected using a LAS4000 chemiluminescence detection system (Fuji, Tokyo, Japan).

Cells were seeded into six-well plates and treated with test compounds. Cytoplasmic and nuclear proteins were extracted using ProteoExtract^® Subcellular Proteome Extraction kit (Calbiochem) and subjected to western blot analysis with specific antibodies. The proteins were then visualized by enhanced chemiluminescence (Intron) and detected using a LAS4000 chemiluminescence detection system.

The cells were seeded into a 12-well plate with cover glasses. Following treatment at the indicated time and dose, the cells were fixed in 3.7% formaldehyde for 20 min at room temperature (RT) and were permeabilized in 0.2% Triton X-100 for 20 min at RT. Then cells were blocked with 1% bovine serum albumin for 1 h. Next, the cells were incubated overnight with primary antibody against either AMPKα1 or β-catenin. After washing, the cells were incubated with Alexa546-conjugated anti-rabbit IgG and Alexa 488-conjugated anti-mouse IgG (both from Molecular Probes, Eugene, OR, USA) for 1 h at RT. Next, cell nuclei were stained with 10 µM Hoechst 33342 for 10 min and then observed with a confocal microscope (Carl Zeiss, Thornwood, NY, USA).

Specific siRNAs, targeting AMPKα1 (PRKAA1) and GSK3β, and non-specific control siRNAs were purchased from Dharmacon (Chicago, IL, USA). For transient transfection, cells were seeded at a density of 5×10⁴ cells/ml in antibiotic-free medium, and siRNAs were transfected using the DharmaFECT 4 transfection reagent (Dharmacon) according to the manufacturer's instructions. After incubation for 72 h, the cells were analyzed via immunofluorescence staining or western blotting.

Five-week-old male BALB/c nu/nu mice were obtained from SLC (Tokyo, Japan) and were housed in sterile filter-topped cages. Hep3B hepato-carcinoma cells (1×10⁶ cells/150 µl) were subcutaneously injected into the left flank of the mice. One week after the injection of Hep3B cells, selenium was dissolved in PBS and administered intraperitoneally (30 mg/kg/day) for 20 days. The control animals were injected with vehicle (PBS) alone. Tumor size was measured using a caliper at 2-day intervals, and the volume was calculated by the modified formula V = 1/2 (length x width²). After the 18 day treatment, tumors were removed and frozen in liquid nitrogen for western blot analysis or fixed with formalin for immunohistochemistry. All animal experiments were approved by the Ethics Committee for Animal Experimentation, Hannam University.

Tumor specimens from mice were fixed in 10% formaldehyde, embedded in paraffin and sectioned into 5-µm thick slices. Sections were deparaffinized with xylene and dehydrated with 98% ethanol. Serial sections were stained using standard immunoperoxidase techniques with primary antibodies against β-catenin (1:50) and p-AMPKα1 (1:50). For epitope retrieval, specimens were microwave-treated for 25 min before incubation with primary antibodies. Pre-immune serum was used as a negative control for immunostaining, and positive-staining was visualized with diaminobenzidine, followed by a light counter-staining with hematoxylin. All findings were evaluated by a pathologist blinded to the treatment conditions, and samples were evaluated on the basis of stain intensity and percentage of reactive cells. Images of representative results were recorded.

Cell viability and tumor volume data were statistically analyzed using an unpaired t-test (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant result.

A large body of evidence suggests that β-catenin is often aberrantly overexpressed in hepatocellular carcinoma. IGF-1 has recently been identified as being capable of increasing β-catenin expression and its transcriptional activity via phosphorylation of GSK3β (10). We examined this effect of IGF-1 on hepatocellular carcinoma growth through regulation of the GSK3β/β-catenin survival pathway. Our results showed that IGF-1 effectively increased cell growth in a time-dependent manner (Fig. 1A). We analyzed the molecular changes in the control and selenium-treated cells. The phosphorylation of β-catenin on the Ser552 residue and GSK3β were significantly increased, and phosphorylation of β-catenin on the Ser33/37 residue was significantly decreased in a time-dependent manner (Fig. 1B). To examine whether selenium exerts inhibitory effects on β-catenin and GSK3β, we analyzed changes in their phosphorylation and expression. Selenium at concentrations of >500 µM effectively inhibited IGF-1-stimulated β-catenin and phosphorylation of GSK3β and enhanced AMPK phosphorylation (Fig. 2A). We investigated the effects of selenium on the β-catenin trans-location from the cytosol to the nucleus. We discovered that selenium decreased IGF-1-increased β-catenin translocation in the nucleus at 6 h, whereas no marked difference occurred in the expression of β-catenin in the cytosol (Fig. 2C). To analyze the localization pattern of β-catenin in this system further, we immunostained β-catenin and detected it using fluorescence microscopy. Consistent with western blot results, β-catenin decreased in the nucleus after selenium treatment for 6 h (Fig. 2B). Taken together, we inferred that selenium may redistribute IGF-1-increased β-catenin protein in the cytoplasm and nuclei.

We investigated whether selenium-reduced β-catenin expression in Hep3B cells is associated with the activation of AMPK. We examined the effects of selenium on the activity of β-catenin in AMPK or GSK3β siRNA-transfected Hep3B cells. To examine whether selenium-reduced β-catenin levels are AMPK dependent, we determined the effects of selenium on β-catenin after knockdown of AMPK using siRNA in Hep3B cells. Selenium did not regulate β-catenin and GSK3β in the absence of AMPK. However, selenium regulated β-catenin in the absence of GSK3β (Fig. 3A). β-catenin was regulated not only by the GSK3β signaling pathway but also by the PI3K/Akt pathway. Thus, selenium regulates β-catenin via a GSK3β-independent pathway. Furthermore, the immunostaining results showed that selenium had no effect on β-catenin localization in AMPK-transfected Hep3B cells (Fig. 3B).

Next, we examined the direct relationship of AMPK/β-catenin using immunoprecipitates. Co-immunoprecipitation/western blot experiments performed with Hep3B cells showed identical results (Fig. 3C). To further characterize the specificity of the β-catenin-AMPK interaction, we performed additional experiments with antibodies specific for β-catenin and AMPK isoforms in the lysate of untreated or selenium-treated cells. Treatment with selenium led to the appearance of β-catenin or AMPK in the Hep3B cell lysate (Fig. 3C). These results suggest that selenium directly regulates the interaction between β-catenin and AMPK.

To examine whether selenium exerts anticancer activity in Hep3B cells, we examined the effect of selenium on cell proliferation and apoptosis. Selenium effectively inhibited cell growth in a dose-dependent manner (Fig. 4A) and induced apoptosis, as measured by Annexin V/PI and Hoechst 33342 staining (Fig. 4B).

Numerous studies have identified AMPK as a central factor inducing apoptosis in various cancer cells. Our prior study together with that of others has also implicated AMPK as a key regulator of selenium-induced apoptosis in cancer cells (15). To further validate whether AMPK inhibition by compound C is associated with cell proliferation and apoptosis of hepatocel-lular carcinoma cells. As shown in Fig. 4C, inhibition of AMPK by compound C treatment abolished the growth-stimulatory effects of IGF-1. Furthermore, fluorescence-activated cell sorter results revealed that the apoptotic rate in compound C-treated cells declined from 73.61 to 22.71% after treatment compared with that in selenium-treated cells in the control group (Fig. 4D). All of these data argue for the critical involvement of AMPK in selenium-induced apoptosis in Hep3B cells.

We built an in vivo hepato-carcinoma xenograft tumor model by inoculating Hep3B cells into male nude mice subcutaneously. We primarily discovered that treatment with selenium (3 mg/kg·day) markedly attenuated tumor growth without adverse effects on body weight and activity compared with that in the control group (Fig. 5A). We analyzed the molecular changes in the control and selenium-treated cells. The phosphorylation of GSK3β and expression of β-catenin were significantly suppressed, but the phosphorylation of AMPK was significantly increased in the selenium treatment group relative to that in the control group (Fig. 5B). Immunohistochemical analysis confirmed that the control cells had high levels of β-catenin, and the selenium-treated cells had reduced β-catenin levels. Furthermore, tumors in the selenium treatment group showed strongly increased p-AMPK levels (Fig. 5C).