The X protein of hepatitis B virus inhibits apoptosis in hepatoma cells through enhancing the methionine adenosyltransferase 2A gene expression and reducing S-adenosylmethionine production

Abstract

The X protein (HBx) of hepatitis B virus (HBV) is involved in the development of hepatocellular carcinoma (HCC), and methionine adenosyltransferase 2A (MAT2A) promotes the growth of liver cancer cells through altering S-adenosylmethionine homeostasis. Thus, we speculated that a link between HBx and MAT2A may contribute to HCC development. In this study, the effects of HBx on MAT2A expression and cell apoptosis were investigated, and the molecular mechanism by which HBx and MAT2A regulate tumorigenesis was evaluated. Results from immunohistochemistry analyses of 37 pairs of HBV-associated liver cancer tissues/corresponding peritumor tissues showed that HBx and MAT2A are highly expressed in most liver tumor tissues. Our in vitro results revealed that HBx activates MAT2A expression in a dose-dependent manner in hepatoma cells, and such regulation requires the cis-regulatory elements NF-κB and CREB on the MAT2A gene promoter. Electrophoretic mobility shift assay (EMSA) and chromatin immunoprecipitation (ChIP) further demonstrated that HBx facilitates the binding of NF-κB and CREB to MAT2A gene promoter. In addition, overexpression of HBx or MAT2A inhibits cell apoptosis, whereas knockdown of MAT2A expression stimulates apoptosis in hepatoma cells. Furthermore, we demonstrated that HBx reduces MAT1A expression and AdoMet production but enhances MAT2β expression. Thus, we proposed that HBx activates MAT2A expression through NF-κB and CREB signaling pathways to reduce AdoMet production, inhibit hepatoma cell apoptosis, and perhaps enhance HCC development. These findings should provide new insights into our understanding how the molecular mechanisms underline the effects of HBV infection on the production of MAT2A and the development of HCC.

FIGURE 1.

Determination of expression status of MAT2A and HBx in HBV-associated liver tumor tissues by immunohistochemistry analyses. A, representative results of immunohistochemistry analyses of 37 cases of surgically resected HBV-associated HCC tissues and their adjacent nontumorous liver tissues. Panels a and d, normal liver tissues obtained from liver traumas patients undergoing partial hepatectomy; panels b and e, HBV-associated HCC tissues; panels c and f, peritumoral noncancerous tissues. Panels a–c, immunohistochemistry analyses using antibody to MAT2A; panels d–f, immunohistochemistry analyses using antibody to HBx. B, analyses of the correlation between the expression level of MAT2A and HBx. The levels (high ++; low +; and undetectable −) of HBx in HBV-associated HCC tissues were compared with that of MAT2A, respectively (p = 0.0075, χ² test; p = 0.0061, Fisher two-tailed exact test). C, seven HBV-associated HCC tissues (samples 1–7) were selected for Western blot analyses using antibodies to MAT2A, HBx, and GAPDH proteins, respectively.

FIGURE 2.

Role of HBx in the regulation of MAT2A expression in hepatoma cells. A, RT-PCR analyses of mRNA levels in HepG2 and HepG2.2.15 cells using primers specific to the MAT2A, HBx, and GAPDH genes, respectively. B, Western blot analyses of protein expression status in HepG2 and HepG2.2.15 cells using antibodies to MAT2A, HBx, and GAPDH proteins, respectively. C, HepG2 cells were transfected with pBlue-SK or pBlue-HBV, respectively, at different times as indicated. MAT II enzyme activities were determined. D, HepG2 cells were co-transfected with a reporter plasmid pGL3-MAT2A, in which the luciferase gene is under the control of the MAT2A promoter, along with plasmids (pcDNA-S, pcDNA-preS1, pcDNA-preS2, pcDNA-HBs, pcDNA-HBx, and pcDNA-HBp) expressing each of the HBV proteins, respectively. Relative luciferase activity was determined by standard procedures. Results shown are mean ± S.D. of three experiments performed in duplicate. E, HepG2 cells were co-transfected with the reporter plasmid pGL3-MAT2A and pcDNA-HBx at different concentrations. Relative luciferase activity was determined by standard procedures. Results shown are mean ± S.D. of three experiments performed in duplicate. The protein levels were determined by Western blot using antibodies to HBx and GAPDH, respectively. F, HepG2 cells were transfected with pcDNA3.1 or plasmids expressing each of the HBV proteins, respectively. MAT II enzyme activities were determined 48 h post-transfection.

FIGURE 3.

Determination of MAT2A expression levels and MAT II activities regulated by HBx. A, RT-PCR analyses of mRNA levels in L02, HepG2, and BEL7404 cells using primers specific to the MAT2A or GAPDH gene. B, Western blot analyses of protein levels in L02, HepG2, and BEL7404 cells using antibodies to MAT2A or GAPDH proteins. C, L02, HepG2, and BEL7404 cells were transfected with pcDNA-HBx at 0, 0.25, and 0.5 μg per well. MAT II enzyme activity was measured 48 h post-transfection. Data are shown as the mean ± S.D. of three independent experiments. The expression status of HBx and GAPDH was also determined using antibodies to the two proteins, respectively.

FIGURE 4.

Requirement of cis-acting elements in the activation of MAT2A gene expression regulated by HBx. A, left panel, diagrams of reporter constructs containing the luciferase (Luc) gene under the control of the human MAT2A promoter or its serial deletions. A, right panel, HepG2 and BEL-7404 cell lines were co-transfected with pcDNA-HBx along with the constructed reporter plasmids with wild-type or mutated MAT2A promoters. Relative luciferase activity was determined. The results shown are the means ± S.D. of three experiments performed in duplicate. B, left panel, diagrams of reporter constructs containing the luciferase gene under the control of MAT2A promoter with specific mutations in cis-regulatory elements as indicated. B, right panel, HepG2 and BEL-7404 cell lines were co-transfected with pcDNA-HBx along with these constructed reporter plasmids with wild-type or mutated MAT2A promoters. Relative luciferase activity was determined. The results shown are the means ± S.D. of three experiments performed in duplicate.

FIGURE 5.

Analyses of effects of HBx on binding of CREB and NF-κB to the MAT2A promoter. A, analyses of the effect of HBx on the binding of CREB to the MAT2A promoter by EMSA. EMSA was performed with nuclear extracts of HepG2 cells transfected with control plasmid (lane 1) or pcDNA-HBx at different concentrations (lanes 2–7). CREB probe was generated by annealing single-stranded and end-labeled oligonucleotides containing the cognate MAT2A promoter region (nucleotides −385/−365). Mutated oligonucleotide (lane 2), nonspecific competitor (lane 6), or specific competitor (unlabeled CREB probe, lane 7) were used as a control. For supershift, antibody to CREB (lane 4) was incubated with nuclear extracts before being added to the reaction. B, analyses of the effect of HBx on the binding of NF-κB to the MAT2A promoter by EMSA. EMSA was performed with nuclear extracts of HepG2 cells transfected with pcDNA-HBx. NF-κB probe was generated by annealing single-stranded and end-labeled oligonucleotides containing the cognate MAT2A promoter region (nucleotides −296/−284). Mutated oligonucleotide (lane 2), nonspecific competitor (lane 6), or specific competitor (unlabeled NF-κB probe, lane 7) were used as a control. For supershift, antibody to p65 (lane 4) was incubated with nuclear extracts before being added to the reaction. Samples were electrophoresed on 5% nondenaturing polyacrylamide gels and visualized by autoradiography. Arrows indicate the shift bands or supershifted protein-DNA complexes or free probes. C, determination of the role of HBx in the binding of CREB and NF-κB to the MAT2A promoter by ChIP assays. HepG2 cells transfected with pcDNA-HBx (+) or control vector (−) were lysed and subjected to ChIP assays. The exact locations of PCR products of Chip1 and Chip2 underlie the simplified genomic structures of the MAT2A gene promoter. The results are representatives of four independent experiments.

FIGURE 6.

Determination of signaling pathways involved in the activation of MAT2A gene expression regulated by HBx. A, analyses of the effects of inhibitors of signaling components on the activation of MAT2A expression regulated by HBx. HepG2 cells were transfected with pcDNA-HBx and treated with LY294002 (10 μm), U0126 (13 μm), MG132 (4 μm), SB203580 (10 μm), GF109203 (1 μm), H89 (10 μm), SP600125 (30 μm), and PD098059 (10 μm), respectively. MAT II enzyme activity was measured, and MAT2A protein was determined by Western blot analyses 72 h post-transfection. B, HepG2 cells were transfected with pcDNA and treated with LY294002 (10 μm), U0126 (13 μm), MG132 (4 μm), SB203580 (10 μm), GF109203 (1 μm), H89 (10 μm), SP600125 (30 μm), and PD098059 (10 μm), respectively. MAT II enzyme activity was measured and MAT2A protein was determined by Western blot analyses 72 h post-transfection. C, determination of the effects of shRNAs of signaling components on the activation of MAT2A expression regulated by HBx. HepG2 cells were co-transfected with pMAT2A-Luc and pcDNA-HBx along with plasmids expressing shRNA-control, shRNA-RIG-I, shRNA-MAV, shRNA-IRAK2, shRNA-TRAF6, shRNA-IKKα, shRNA-IKKβ, shRNA-IKKi, shRNA-p50, or shRNA-p65, respectively. Luciferase activity was measured 72 h post-transfection. D, effect of MG132 on the MAT II enzyme activity and MAT2A gene expression regulated by HBx. HepG2 cells were transfected with pcDNA-HBx and treated with MG132 at different concentrations as indicated. MAT II activity was measured (top panel), and MAT2A and GAPDH mRNA levels were determined by RT-PCR (lower panel) 72 h post-transfection. E, effect of shRNA to CREB on the MAT II enzyme activity and MAT2A gene expression regulated by HBx. HepG2 cells were co-transfected with pcDNA-HBx and pshRNA-CREB. MAT II activity was measured (top panel), and MAT2A and GAPDH mRNAs were determined RT-PCR (lower panel) 72 h post-transfection.

FIGURE 7.

Effects of HBx and MAT2A on cell apoptosis and AdoMet production in hepatoma cells. HepG2 cells were plated at a density of 2 × 10⁵ cells/cm². A and B, HepG2 cells were transfected with pcDNA-3.1 (panel a), pcDNA-MAT2A (panel b), pcDNA-HBx (panel c), pMAT2A-shRNA (panel d), and pcDNA-HBx plus pMAT2A-shRNA (panel e). A, changes in nuclear morphology of transfected cells were examined by staining of nuclear DNA with DAPI and visualized under fluorescence microscopy. B, rates of apoptosis were evaluated by the determination of sub-G₁ populations of transfected cells through Guava Nexin-V assays using flow cytometry. C, rates of apoptosis were summarized from flow cytometry analyses. D, levels of AdoMet and AdoHcy produced in transfected cells were determined by reverse phase-HPLC. Results in C and D are means of three independent experiments, and the bars represent ± S.D., significantly different from control by Turkey test. E, effects of HBx on the expression of MAT genes. L02 cells were transfected with pcDNA-HBx at different concentrations as indicated. The levels of MAT1A, MAT2A, MAT2β, and GAPDH proteins were determined using antibodies to the four proteins respectively. F, effects of HBx on the change in AdoMet homeostasis. L02 cells were transfected with pcDNA-HBx at different concentrations as indicated. The levels of AdoMet and AdoHcy were measured by HPLC. Results were mean of three independent experiments, significantly different from control by Turkey test.

FIGURE 8.

Proposed model for the role of HBx in MAT2A expression, AdoMet production, cell apoptosis, and perhaps HCC development. Top, in normal liver cells, two mechanisms to maintain the high cellular AdoMet level are as follows: ① up-regulation of MAT1A expression by AdoMet with the increase in MAT I/III activity; and ② the high capacity of MAT I/III to convert dietary methionine and ATP into AdoMet. Because AdoMet down-regulates MAT2A expression and inhibits MAT II activity, the contribution of MAT II to the production of AdoMet is minimal in liver cells. AdoMet controls liver growth and also regulates apoptosis with an anti-apoptotic effect on normal hepatocytes. Bottom, during HBV infection, the viral protein HBx activates MAT2A expression and MAT II activity by enhancing the binding of NF-κB and CREB to MAT2A gene promoter. MAT2A facilitates cancer cell growth through DNA hypomethylation, and thus, it functions as a positive regulator in hepatoma growth. In addition, in gene expression, highly expressed MAT2A inhibits MAT1A expression, which is progressively silenced by a mechanism that involves the methylation of the MAT1A gene promoter and its association with hypoacetylated histones. As a result, a new lower steady state level of AdoMet is reached. Such reduction in AdoMet level releases the inhibitory effect of this molecule on proliferation, which facilitates liver regeneration. If the conditions leading to chronic HBV infection are persistent, MAT2A levels are maintained continuously high, and AdoMet levels are maintained continuously low, which result in the inhibition of apoptotic cell death and predisposes the liver to develop steatohepatitis, cirrhosis, and ultimately HCC.