Hepatitis B virus X protein induces apoptosis by enhancing translocation of Bax to mitochondria

Authors


Abstract

Hepatitis B virus X protein (HBx) is essential for viral replication and plays an important role in viral pathogenesis. HBx transactivates many viral and cellular genes and participates in cellular signal transduction pathways, proliferation, and apoptosis. In the present study, we report that HBx induces apoptosis by enhancing the translocation of Bax to mitochondria, followed by inducing the loss of mitochondrial membrane potential and release of cytochrome C. In addition, Bcl-2, inhibitor of Bax, rescues the disruption of mitochondrial membrane potential and DNA fragmentation induced by serum starvation in HepG2-X cells expressing HBx. We also found that HBx binds directly to Bax and interferes with the interaction between Bax and 14-3-3ε to enhance the translocation of Bax to mitochondria. Taken together, our data suggest that HBx induces apoptosis by interacting with Bax and enhancing its translocation to mitochondria. © 2008 IUBMB IUBMB Life, 60(7): 473–480, 2008

INTRODUCTION

Human hepatitis B virus (HBV) is one of the causative pathogens of acute hepatitis, chronic hepatitis, cirrhosis, and hepatocellular carcinoma1. Among the HBV proteins, hepatitis B virus X protein (HBx) is a regulatory protein with multiple functions. However, the role of HBx in productive HBV infection still remains controversial2. Some reports suggested that HBx contributed to HBV replication3, 4, while others showed that HBV replication was not dependent on HBX5. HBx transactivates various cellular and viral promoters via protein–protein interaction, with cellular factors that bind to recognition sequences6. HBx has been implicated in HBV pathogenesis and cancer development through its ability to deregulate cell cycle check points, to activate the expression of some oncogenes and inflammatory cytokines, and to induce liver tumors in transgenic mouse strains7.

HBx is also involved in the apoptotic destruction of liver cells during viral infection8. It has been reported that HBx interacts with various cellular signaling factors to enhance apoptosis9, 10. HBx transactivates Fas ligand gene expression to sensitize UV-induced apoptosis11. In addition, HBx is associated with caspase activation and mitochondrial dysfunction12. However, in some studies, HBx has been reported to inhibit apoptosis induced by p53 or TGF-β13. Therefore, the role of HBx in apoptosis is still controversial, and the precise mechanism by which HBx induces apoptosis remains largely unknown.

Recent reports have focused on mitochondria, which play an important role in apoptosis. Mitochondria undergo major changes in membrane integrity due to apoptotic signals14, and these changes cause the redistribution of numerous apoptogenic factors such as cytochrome C, AIF, and Smac/DIABLO in the intermembrane space, which is tightly regulated by Bcl-2 family proteins.

The Bcl-2 family of proteins includes antiapoptotic members such as Bcl-2 and Bcl-xL, proapoptotic members such as Bax and Bak, and various BH3 proteins. Among them, Bax can promote mitochondrial release of cytochrome C, whereas Bcl-2 can suppress this process15. While Bax is localized in the cytoplasm, apoptotic stimulation results in its translocation to mitochondria16. Indeed, a previous report showed that free Bax released from 14-3-3 proteins translocates to the mitochondria17. The 14-3-3 proteins interact with various cellular proteins involved in apoptotic signaling such as Bax, Bad, and ASK1.

Although HBx-induced apoptosis has been reported to be associated with mitochondrial dysfunction, its direct involvement in Bax-mediated mitochondrial dysfunction remains unknown. In this study, we examined the role of HBx in mitochondrial translocation of Bax during its induction of apoptosis. Particularly, we found that HBx directly interacts with Bax and interferes with association between Bax and 14-3-3ε, which inhibits the translocation of Bax to mitochondria17. These results indicate a regulatory mechanism for HBx in mitochondria-mediated apoptosis.

MATERIALS AND METHODS

Cell Lines and Transfection

HepG2 cells and HepG2-X cells were cultured in MEM-α containing 10% (v/v) fetal bovine serum (FBS) and antibiotic–antimycotics at 37 °C. HepG2, a human hepatoblastoma cell line, was purchased from American Type Culture Collection. HepG2-X cells were constructed as previously described11. To obtain HepG2-X cells, HepG2 cells were transfected with pRSV/X plasmid-expressing HBx, selected in media containing G418 for 3 weeks, and analyzed by Western blot assay. Transient transfection of HepG2 cells was performed by the calcium phosphate method. Human embryonic kidney HEK293T cells were transiently transfected by PEI transfection method.

Plasmids

The human Bax gene was cloned into pGEX 4T-1 and pCDNA3 to construct plasmids pGBAX and pBax, respectively. The HBx gene of HBV was cloned into pGEX 4T-1, pRSETB, pCDNA, and pFlag-CMV to construct plasmids pGHBx, pRHBx, pHBx, and pFlg-HBx, respectively. The human Bcl2 gene (gift of Dr. Yutaka Eguchi) was cloned into the pCDNA3 to construct pBcl2. Plasmid pHA14-3-3ε was a gift from Dr. Cheol O. Joe.

Antibodies

For Western blot analyses, rabbit polyclonal antibody was produced against HBx purified from E. coli. Anti-Bax antibody was purchased from Oncogene Inc. (San Diego, CA), and anti-HA, anti-Flag, anti-GST, anti-β-actin, anti-14-3-3ε, and anti-cytochrome C antibodies were purchased from Sigma Chemical Co. (St. Louis, MO).

Cell Viability Assay

The amount of dead cells was determined as previously described11. Briefly, cells (7.5 × 105/100-mm plate) were cultured in MEM-α containing 0% or 10% FBS without changing media. At 24-h intervals, cells were harvested and stained with 0.4% Trypan Blue solution. Viable cells were counted with a hematocytometer. A minimum of 200 cells in six separate fields were scored, and the cell viability was calculated in percentages.

Hoechst 33258 Staining

For Hoechst 33258 staining, cells were plated onto chamber slides (Nunc Inc.). After cells were cultured in media containing 0% or 10% FBS for 72 h, the slides were fixed with 100% methanol at −20 °C for 20 min and stained at 37 °C for 20 min in 1 μg/mL solution of Hoechst 33258. By fluorescent microscopy, apoptotic cells with condensed, marginated, and fragmented nuclei were counted.

Measurement of Caspase 3 Activity

Briefly, cells were cultured, harvested, and resuspended in extraction buffer on ice for 30 min. After centrifugation (20,000g), the supernatant was diluted in an assay buffer and incubated with caspase-3 substrate at 37 °C for 30 min. Caspase 3 activity was determined by measuring the fluorescence intensity with L5508 Perkin Elmer spectrometer.

Measurement of Mitochondrial Transmembrane Potential

Mitochondrial membrane potential was measured with the cation dye rhodamine123 (Rh123) that partitions into the mitochondria. Depolarization of mitochondrial membrane potential causes the release of Rh123 from the mitochondria and a decrease in intracellular fluorescence. HepG2 cells and HepG2-X cells were cultured in MEM-α containing 0% or 10% FBS for 72 h. Cells were harvested and resuspended in HBS buffer. Rh123 was added to the cell suspension at a final concentration of 10 μM. The cells were stained at 37 °C in the dark for 30 min, washed, and resuspended in HBS buffer. Mitochondrial membrane potential was analyzed with a Becton Dickinson fluorescence cytometry using CellQuest software. The laser was adjusted to emit at a wavelength of 480 nm, and a 530-nm long-pass filter was used10.

DNA Fragmentation Assay

DNA fragmentation assays were performed as previously described11. Briefly, cells were harvested after culturing for 72 h. Then, both adherent and floating cells were resuspended in lysis buffer and treated with a solution containing RNase A and proteinase K. After phenol/chloroform extraction, the DNA was precipitated, subjected to 1.5% agarose gel electrophoresis, and visualized by ethidium bromide staining.

Preparation of Mitochondrial Fraction

Cells were cultured for 72 h, harvested, and resuspended in buffer A (20 mM HEPES-KOH, pH 7.5, 10 mM KCl, 1.5 mM MgCl2, 1 mM Na2-EDTA, 1 mM Na2-EGTA, 1 mM dithiothreitol, 0.1 mM PMSF) containing 250 mM sucrose. Cells were homogenized and centrifuged at 750g for 10 min at 4 °C. Supernatants were further centrifuged at 10,000g. The resulting mitochondria-containing pellets were resuspended in buffer A containing 250 mM sucrose to yield the mitochondrial fraction. Supernatants from the 10,000g spin were further centrifuged at 100,000g for 1 h at 4 °C to produce the cytosol fraction.

Affinity Purification of Bax and HBx Expressed in E. coli

E. coli-expressed Bax and HBx were isolated and purified as previously described18. Plasmids pGBax and pGHBx were transformed into E. coli BL21. E. coli induced by IPTG were lysed with 0.5 mg/mL lysozyme in lysis buffer, sonicated on ice, and centrifuged at 28,000g for 10 min at 4 °C. The supernatant was incubated with glutathione–Sepharose 4B (Amersham) at 4 °C for 12 h. GST-Bax and GST-HBx bound to the resin were concentrated by TCA precipitation. For His-HBx, E. coli transformed with pRHBx was induced by IPTG, lysed, spun down, and suspended in a binding buffer containing 6 M urea. His-HBx was purified by His-binding metal chelation resin chromatography and dialyzed for 12 h. All the dialysis steps were carried out at 4 °C.

Mitochondrial Localization of Bax and HBx in vitro

Mitochondria-containing supernatants prepared from HepG2 cells were incubated with purified GST, GST-Bax, and His-HBxat 30 °C for 3 h. Reaction mixtures were spun down, dissolved in SDS-containing buffer, and subjected to Western blot analysis. To investigate mitochondrial localization of HBx, the supernatants were incubated with GST and GST-HBx at 30 °C for 3 h.

Immunoprecipitations

Cells were lysed in lysis buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS) in the presence of a mixture of protease inhibitors. Equal amounts of protein were then incubated overnight with various antibodies at 4 °C. The immunoprecipitated complexes were isolated using protein-A Sepharose (Amersham), and the samples were analyzed by SDS-PAGE. After electrotransfer to nitrocellulose membrane, Western blot analysis was performed.

RESULTS

Effects of HBx on Induction of Apoptosis

First, we investigated the viability of HepG2 cells and HepG2-X cells in culture media containing either 10% or 0% serum without changing the media (Fig. 1A). While HepG2-X cells showed slightly decreased cell viability in 10% serum compared to HepG2 cells, the viability of HepG2-X cells was ∼20–30% lower than that of HepG2 cells in 0% serum. These data suggest that HBx has a more negative effect on cell viability during serum starvation. We subsequently tested whether the cell death elevated by HBx was apoptotic. Apoptosis was analyzed by nuclear condensation assays using Hoechst 33258 dye. Nuclear condensation was more abundant in HepG2-X cells than in HepG2 cells in media containing either 10% or 0% serum (Fig. 1B). Analysis of caspase 3 activity, which is another important indicator of apoptotic cell death, also showed that HBx increased caspase 3 activity in media with 10% serum, with further increases observed during serum starvation (Fig. 1C). These results demonstrate that HBx induces apoptosis in the hepatoma cell line and that the apoptotic effect of HBx is more evident during serum starvation.

Figure 1.

HBx induces apoptosis in hepatoblastoma cells. (A) The percentage of viable cells was determined at each time point by subtracting the number of cells stained with Trypan blue from the total number of cells. The data shown are means ± SD of values obtained from four independent experiments. (B) The percentage of DNA condensation was determined by subtracting the number of cells stained with Hoechst 33258 dye from the total number of cells. (C) Caspase 3 activities were determined in cytosolic extracts of HepG2 cells and HepG2-X cells that were cultured for 72 h.

HBx Induces Loss of Mitochondrial Membrane Potential

HBx has been reported to induce apoptosis through either the mitochondria-mediated pathway12, 19 or the mitochondria-independent pathway20. To determine whether HBx induces apoptosis through the mitochondria-mediated pathway, the mitochondrial membrane potentials of HepG2-X cells were measured by flow cytometry using the cationic fluorescent dye rhodamine123. As shown in Fig. 2A, HepG2-X cells showed about 5% loss of mitochondrial membrane potential compared to HepG2 cells in 10% serum and a marked loss of potential (about 30% loss) in serum starvation compared to HepG2 cells, suggesting that HBx induces mitochondrial membrane dysfunction. To test whether the apoptotic activity of HBx could be reversed by Bcl2, which is known to inhibit apoptosis15, a Bcl2 expression plasmid was transfected into HepG2-X cells and then mitochondrial membrane potential and DNA fragmentation were analyzed. Bcl2 overexpression reversed the loss of mitochondrial membrane potential and the DNA fragmentation induced by HBx (Figs. 2A and 2B). These data indicate that HBx induces apoptosis through mitochondrial membrane dysfunction.

Figure 2.

HBx induces the loss of mitochondrial transmembrane potential and Bcl2 inhibits HBx-induced apoptosis. (A) Mitochondrial potential was analyzed in HepG2 cells and HepG2-X cells. Cells were cultured for 72 h and stained with rhodamine123. Mitochondrial potential was analyzed with Becton Dickinson fluorescence cytometry using CellQuest software. The effect of Bcl2 was analyzed by transient transfection of HepG2-X cells with Bcl2 expression plasmid. (B) DNA fragmentation was analyzed to determine the effect of Bcl2 upon the apoptosis of HepG2 cells and HepG2-X cells.

HBx Enhances the Translocation of Bax to Mitochondria

We next examined whether the mitochondrial membrane dysfunction induced by HBx is associated with translocation of Bax from cytosol to mitochondria, since Bax is reported to translocate into mitochondrial membrane and induce mitochondrial dysfunction following a death signal15, 21. In HepG2-X cells, Bax was localized more to mitochondria compared to HepG2 cells in both 10% serum and serum starvation, suggesting that HBx possibly enhanced the translocation of Bax to mitochondria (Fig. 3A). The translocation of Bax from cytosol to mitochondria was apparent during serum starvation. At the same time, in HepG2-X cells, the release of cytochrome C from mitochondria to cytosol is more evident, suggesting that the enhanced Bax translocation to mitochondria in the presence of HBx resulted in the release of cytochrome C from mitochondria (Fig. 3B). Therefore, it is possible that mitochondrial Bax translocation affected by HBx may be crucial for HBx-induced apoptosis.

Figure 3.

HBx enhances the translocation of Bax from cytosol to mitochondria. (A) Mitochondrial translocation of Bax was analyzed in vivo. HepG2 cells and HepG2-X cells were cultured in MEM-α with 10% or 0% serum for 72 h. The mitochondrial fraction and the cytoplasmic fraction were analyzed by Western blotting. (B) Release of cytochrome C from mitochondria to cytoplasm was analyzed. Each mitochondrial and cytoplasmic fraction of both cell types was analyzed by Western blotting using anti-cytochrome C antibody and quantified by densitometry. (C) Mitochondrial translocation of recombinant Bax enhanced by HBx was analyzed in vitro. HepG2 cells were cultured in media with 10% or 0% serum for 72 h. The mitochondria-containing supernatants of HepG2 cells were prepared and mixed with a combination of GST, GST-Bax, and His-HBx. The mitochondrial pellets were collected and subjected to Western blot analysis using anti-GST antibody (top). Purified recombinant GST-Bax and His-HBx were confirmed with anti-Bax antibody (middle) and anti-HBx antibody (bottom).

To confirm the direct evidence for Bax mitochondrial translocation enhanced by HBx, we reconstituted an in vitro mitochondria system using the recombinant proteins, His-HBx and GST-Bax, and the mitochondrial fraction of HepG2 cells. The purified recombinant HBx enhanced the mitochondrial translocation of recombinant Bax in vitro in both 10% serum and serum starvation (Fig. 3C). Taken together, it is evident that HBx enhances the translocation of Bax to mitochondria, suggesting that HBx possibly interacts with Bax either directly or indirectly for enhancement of Bax mitochondrial translocation.

HBx Interacts with Bax and Interferes with the Interaction Between 14-3-3ε and Bax

To investigate whether HBx interacts with Bax directly, the extracts of HepG2-X cells were immunoprecipitated with anti-HBx antibody, followed by Western blot analysis with anti-Bax antibody. Clearly, HBx binds directly to endogenous Bax (Fig. 4A). To further verify the interaction between HBx and Bax, HEK293T cells were transiently cotransfected with plasmids pHBx and pBax. The cell extracts were prepared, immunoprecipitated with anti-HBx antibody, and analyzed by Western blotting using anti-Bax antibody (Fig. 4B). These data confirm the specific interaction between HBx and Bax, suggesting that the mitochondrial Bax translocation enhancement by HBx may be derived from direct physical interaction between the two proteins.

Figure 4.

HBx binds directly to Bax and interferes with the association between 14-3-3ε and Bax. (A) Interaction between HBx and endogenous Bax was analyzed. The extracts of HepG2 cells and HepG2-X cells were immunoprecipitated with anti-HBx antibody and anti-Bax antibody. Then, the immunocomplexes were analyzed by Western blotting using anti-Bax antibody and anti-HBx antibody, respectively (top panel). The expressions of Bax and HBx in HepG2 cells and HepG2-X cells were confirmed by Western blot analysis (bottom panel). (B) Interaction between transiently expressed HBx and Bax was analyzed. HEK293T cells were transiently cotransfected with expression plasmids for Bax and HBx. The cell extracts were prepared and immunoprecipitated with anti-HBx antibody. The immunocomplexes were analyzed by Western blotting using anti-Bax antibody (top panel–top). The cell extracts were immunoprecipitated with anti-Bax antibody and the immunocomplexes were analyzed by anti-HBx antibody (top panel–bottom). Transiently expressed HBx and Bax were confirmed by Western blot analysis (bottom panel). (C) Interference of HBx with the interaction between transiently expressed 14-3-3ε and Bax was analyzed. HEK293T cells were transfected with a combination of expression plasmids for Bax, HA14-3-3ε, and various amounts of Flag-HBx (0, 2, 4 μg). The cell extracts were prepared, immunoprecipitated with anti-HA antibody, and analyzed by Western blotting using anti-Bax antibody (top). After stripping, the membrane was reblotted with anti-HA antibody (middle). The levels of Flag-HBx and β-actin in cell lysates were analyzed by anti-Flag and anti-β-actin antibodies (lower and bottom). (D) Interference of HBx with the interaction between endogenous 14-3-3ε and endogenous Bax was analyzed. HepG2 cells and HepG2-X cells were cultured in media with 0% or 10% serum for 72 h. The cell extracts were prepared and immunoprecipitated with anti-14-3-3ε antibody. The immunocomplexes were analyzed by Western blotting using anti-Bax antibody (top) and anti-14-3-3ε antibody (middle). Amounts of cell extracts were determined using anti-β-actin antibody (bottom).

Many studies have shown that 14-3-3 proteins interact with pro-apoptotic Bax and inhibit the translocation of Bax into mitochondria17. It is also possible that HBx might interact with 14-3-3 proteins to allow Bax to translocate to mitochondria. However, we did not find any physical interactions between HBx and 14-3-3 proteins (data not shown). Therefore, we could speculate that HBx might interact with Bax to interfere with the interaction between Bax and 14-3-3. To understand the effect of HBx on the interaction between Bax and 14-3-3 proteins, HEK293T cells were cotransfected with plasmids for Bax, HA-14-3-3ε, and Flag-HBx. The cell extracts were prepared, immunoprecipitated with HA antibody, and subjected to Western blot analysis using anti-Bax antibody (Fig. 4C). We found that HBx dose-dependently interfered with the interaction between 14-3-3ε and Bax.

To identify the interfering activity of HBx upon the interaction between endogenous 14-3-3ε and endogenous Bax, the extracts of HepG2 cells and HepG2-X cells were immunoprecipitated with anti-14-3-3ε antibody and analyzed by Western blot using anti-Bax antibody (Fig. 4D). We found that HBx interfered with the interaction between endogenous14-3-3ε and endogenous Bax. Consequently, these results suggest that HBx inhibits the association between 14-3-3ε and Bax by interacting with Bax, enhances mitochondrial translocation of Bax and release of cytochrome C, and subsequently induces apoptotic cell death.

DISCUSSION

HBx is a promiscuous regulatory protein that plays a major role in HBV pathogenesis and viral replication2. In addition, HBx has been reported to induce apoptosis by interacting with cellular signaling proteins such as c-FLIP9 and Hsp6010.

Many groups have reported possible mechanisms of HBx-induced apoptosis associated with mitochondria. HBx has been reported to affect mitochondrial membrane potential22, 23. These reports focused on its interaction with mitochondrial components such as HVDAC3, a mitochondrial ion-channel protein19. Our results showed that HBx induces cell death in HepG2 cells through loss of mitochondrial membrane potential, which was repressed by Bcl2. It is possible that HBx induces cell death in HepG2 cells through different pathways depending on various apoptotic inducers. For example, HBx sensitizes HepG2 cells to UV-induced apoptosis by activating the gene expression of Fas ligand11. Since we used HepG2-X, a clone expressing HBx, it would be desirable to test the apoptotic effect of HBx in multiclonal populations expressing HBx. In addition, our results using HepG2-X could not represent the situation that occurs in cells infected with HBV. Although we used HepG2-X to avoid possible interfering effects by other viral proteins, we need to further investigate HBx-induced apoptosis using the specific siRNA against HBx in HBV-replicating cells.

In this study, HBx induced apoptosis by enhancing the translocation of Bax to mitochondria. We determined that HBx-induced apoptosis is the result of the intermolecular interaction between HBx and Bax. In vivo, HBx binds to endogenous Bax and enhances its mitochondrial translocation. We also found that recombinant HBx enhanced the translocation of recombinant Bax to mitochondria in an in vitro system. The translocation of Bax to mitochondria is important for mitochondria-mediated apoptosis24. Many studies have reported that Bax stays in an inactive state in cytosol and, in response to death stimuli, undergoes conformational changes that expose membrane-targeting domains, resulting in its translocation to the mitochondrial membrane to induce apoptotic killing25. Therefore, it is plausible that HBx promotes the translocation of Bax by inducing conformational changes. Recently, it has been reported that Bax is associated with HBx-sensitized apoptosis induced by TRAIL, and HBx might upregulate Bax protein expression without significantly altering its gene expression24. According to our data, the upregulation of Bax by HBx may be due to possible Bax stabilization by HBx through direct interaction between HBx and Bax.

The subcellular localization of HBx is largely cytosolic26. HBx is reportedly associated with outer mitochondrial membrane19. However, the precise mechanism for mitochondrial localization of HBx is obscure. According to our results, it is plausible that Bax bound to 14-3-3ε is released by HBx in cytosol, and free Bax can subsequently move to mitochondria separately or together with HBx. The intrinsic binding affinity of HBx to mitochondrial membrane may facilitate targeting of Bax to mitochondria.

The regulation of apoptosis has been implicated in many pathological events of human diseases, especially carcinogenesis27. Therefore, the elucidation of the mechanism of HBx-induced apoptosis in association with Bax contributes to our understanding of not only HBV-derived liver diseases, but also hepatocellular carcinogenesis.

Acknowledgements

We thank Dr. Yutaka Eguchi (Osaka University, Japan) and Dr. Cheol O. Joe (KAIST, Korea) for generously providing plasmids. This work was supported by a grant (R01-2006-000-10667-0) from the Korea Science and Engineering Foundation.

Ancillary