Hepatitis B Virus X Protein Inhibits Transforming Growth Factor-β-induced Apoptosis through the Activation of Phosphatidylinositol 3-Kinase Pathway*

Transforming growth factor-β (TGF-β) is a potent inducer of apoptosis in Hep 3B cells. This work investigated how hepatitis B virus X protein (HBx) affects TGF-β-induced apoptosis. Trypan blue exclusion and colony formation assays revealed that HBx increased the ID50 toward TGF-β. In the presence of HBx, TGF-β-induced DNA laddering was decreased, indicating that HBx had the ability to block TGF-β-induced apoptosis. Furthermore, HBx did not alter the expression levels of type I and type II TGF-β receptors. HBx did not affect TGF-β-induced activation of promoter activities of the plasminogen activator inhibitor-1 (PAI-1) gene. These results indicate that HBx interferes with only a subset of TGF-β activity. In the presence of phosphatidylinositol (PI) 3-kinase inhibitors, wortmannin or LY294002, the HBx-mediated inhibitory effect on TGF-β-induced apoptosis was alleviated. In addition, the tyrosine phosphorylation levels of the regulatory subunit p85 of phosphatidylinositol 3-kinase (PI 3-kinase) and PI 3-kinase activity were elevated in stable clones with HBx expression. Transactivation-deficient mutants of HBx lost their ability to inhibit TGF-β-induced apoptosis. Phosphorylation of the p85 subunit of PI 3-kinase and Akt, a downstream target of PI 3-kinase, was not observed in stable clones with transactivation-deficient HBx mutant's expression. Thus, the anti-apoptotic effect of HBx against TGF-β can be mediated through the activation of the PI 3-kinase signaling pathway, and the transactivation function of HBx is required for its anti-apoptosis activity.

Hepatitis B virus X protein (HBx) 1 has been demonstrated to function as a transcriptional transactivator of a variety of viral and cellular promoter/enhancer elements (1,2). Although not binding directly to DNA, HBx can transactivate transcription through multiple cis-acting elements including AP-1, AP-2, ATF/CREB, NF-B, C/EBP, and Egr-1 binding sites. However, the exact mechanism of transactivation still remains unsolved. Previous investigations have demonstrated that HBx interacts in the nucleus with components of the basal transcription machinery, including RPB5, a subunit of all three mammalian RNA polymerases, and several transcription factors (3)(4)(5)(6). Thus, HBx may exert its effect by mimicking the cellular coactivator function. Another proposed mechanism for HBx activity involves the activation of signal transduction pathways such as the Ras/Raf/ERK, and MEKK-1/JNK cascades, leading to the induction of AP-1, NF-B, and probably other transcription factors (7)(8)(9)(10)(11)(12). HBx has been discovered to be distributed not only in the cytoplasm but also to some extent in the nucleus of transfected cells (9). Thus, HBx may have a dual function: one, related to its cytoplasmic localization, which can mediate the activation of signal transduction pathways, and another, a nuclear function, that may account for the interaction with transcription factors and components of the transcription apparatus to enhance the binding or activity of these proteins (9). In addition to its well known transcriptional transactivation ability through interaction with different cellular targets, HBx has been reported to affect DNA repair (13)(14)(15)(16)(17), cell cycle control (18,19), and apoptosis (20 -23). Therefore, the pleiotropic activities of HBx are potentially relevant to the development of hepatocellular carcinoma.
Transforming growth factor-␤ (TGF-␤) is a potent inducer of apoptosis in hepatocytes and several hepatoma cell lines (24 -26). TGF-␤ exerts its action through transmembrane serine/ threonine kinase receptors. These receptors propagate the signal by phosphorylating the intracellular targets, Smads. Phosphorylated Smad2 or Smad3 can form a stable complex with Smad4, which then translocates to the nucleus to regulate the transcriptional response to TGF-␤ (27,28). However, the mechanism(s) whereby TGF-␤ induces apoptosis is not fully characterized. Nevertheless, induction of oxidative stress (29), activation of caspase 3 (30), and inhibition of Rb expression (26) have been implicated in mediating TGF-␤-induced apoptosis. In liver cells, insulin and insulin-like growth factor-1 (31), as well as interleukin-6 (32), all block TGF-␤-induced apoptosis. Recent studies have revealed that phosphatidylinositol 3-kinase (PI 3-kinase) and its downstream target, Akt, are responsible for the anti-apoptotic activity of these factors against TGF-␤ (32,33).
To elucidate the correlation between the HBx gene and its response to apoptotic stimuli, the effect of HBx gene expression on TGF-␤-induced apoptosis in the Hep 3B cell line was exam-ined. Cells with constitutive or inducible expression of wild or mutant HBx were generated and tested. Transactivation-proficient HBx inhibited TGF-␤-induced apoptosis. The PI 3-kinase/Akt signaling pathway was involved in the HBx-mediated anti-apoptotic effect.

EXPERIMENTAL PROCEDURES
Antibodies and Reagents-Rabbit polyclonal antibodies against HBx expressed in Escherichia coli were employed for detection of HBx. Anti-HA (clone 12CA5) was purchased from Roche Molecular Biochemicals. Dr. R.-H. Chen of the National Taiwan University provided the antibodies for the type I and type II TGF-␤ receptors. The antibody for the p85 subunit of the PI 3-kinase was purchased from Santa Cruz Biotechnology. The anti-phosphotyrosine (clone 4G10) antibody was purchased from Upstate Biotechnology. Antibodies for Akt and phospho-Akt (Ser-473) were purchased from New England BioLabs. pGFPemd-b was purchased from Packard. Finally, wortmannin and LY294002 were purchased from Sigma.
Plasmid Construction-pHBV48 (containing two tandem copies of the HBV genome, provided by Drs. H.-L. Wu and P.-J. Chen of the National Taiwan University Hospital) was employed as a template for polymerase chain reaction amplification of the HBx cDNA fragment. The 0.5-kilobase pair HBx cDNA fragment was inserted into NotI/ blunt-ended pOP13ЈCAT (Stratagene) vector, generating pOP13pX. Two oligonucleotides (HA-U, CAT GTA CCC ATA CGA TGT TCC AGA TTA CGC TCC; HA-L, CAT GGG AGC GTA ATC TGG AAC ATC GTA TGG GTA) were synthesized, annealed, and ligated into NcoI-cut pOP13pX, generating pOP13HApX. When expressed, pOP13HApX generated an N-terminal HA-tagged HBx. The tetracycline-regulated RevTet TM system (CLONTECH) was applied for inducible expression of HBx and mutants. An 0.6-kilobase pair DNA fragment containing an HBx open reading frame was isolated after NcoI and BglII digestion of pHBV48, blunt-ended with Klenow fragment, and ligated with SalI/ blunt-ended pRev-TRE (CLONTECH), generating pRT-X. After HindIII and NaeI digestion of pGFPemd-b, an 0.776-kilobase pair DNA fragment containing a GFP open reading frame was isolated, blunt-ended with Klenow fragment, and ligated with SalI/blunt-ended pRev-TRE (CLONTECH), generating pRT-GFP. pSVX WT , pSVX 7 , pSVX 61 , pSVX 69 , and pSVX 90 -91 (provided by Dr. B. L. Slagle at Baylor College of Medicine) were employed as templates for generating the desired HBx cDNA fragments by polymerase chain reaction. The 3Ј-primer was designed so that the TAG stop codon of HBx was mutated into AAG. The amplified cDNA fragment was inserted into BamHI/blunt-ended pRT-GFP, generating pRT-HBxGFP, pRT-HBx 7 GFP, pRT-HBx 61 GFP, pRT-HBx 69 GFP, pRT-HBx 90 -91 GFP. When present in cells with rtTA expression, these plasmids generated a regulatable expression of the desired HBx-GFP fusion proteins. All of the constructs were confirmed by direct DNA sequencing.
Cell Culture and Transfection-Hep 3B cells were cultured in a minimum essential medium supplemented with Earle's salt, 10% fetal calf serum, penicillin G (50 units/ml), streptomycin (50 g/ml), and fungizone (1.25 g/ml) at 37°C in a 5% CO 2 incubator. The PT67 packaging cell line was grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, penicillin G (50 units/ml), streptomycin (50 g/ml), and fungizone (1.25 g/ml) at 37°C in a 5% CO 2 incubator. Transfection was performed by the calcium phosphate method according to the procedure described previously (34). For transient transfection, cells were harvested 48 h after transfection. For selection of stable clones, Hep 3B cells were cultured in the presence of 500 g/ml G418 or 100 g/ml hygromycin for 2 weeks depending upon the expression vector utilized. Resistant clones were selected, expanded, and assayed for expression of the transfected cDNA by Western blotting or immunoprecipitation.
TGF-␤-induced Cytotoxicity Assays-2 ϫ 10 5 Hep 3B cells were seeded onto a 35-mm tissue culture plate. 24 h after inoculation, cells were washed with phosphate-buffered saline and cultured in serumfree minimum essential medium for 48 h. TGF-␤ was then added to a culture medium at various concentrations. At 48 h after treatment, cells were collected by trypsinization and suspended in minimum essential medium supplemented with 10% fetal calf serum. Viable and nonviable cells were then determined by direct counting using a hemocytometer in the presence of trypan blue. ID 50 was defined as the concentration of TGF-␤ able to reduce 50% of the viable cells. For the colony formation assay, after TGF-␤ treatment the Hep 3B cells were cultured in a complete medium for 2 additional weeks. The colonies were visualized by amido black staining.
DNA Fragmentation Analysis-Hep 3B cells, with or without TGF-␤ treatment, were collected, washed with phosphate-buffered saline, and lysed with lysis buffer (50 mM Tris-HCl, pH 7.5, 20 mM EDTA, 1% Nonidet P-40). The supernatant was collected and incubated with RNase at a final concentration of 500 g/ml for 1 h at 37°C. Subsequently, proteinase K was added to a final concentration of 500 g/ml.
The mixtures were then incubated overnight at 55°C. The DNA was extracted with phenol/chloroform, precipitated with ethanol, dissolved in TE 8.0 (10 mM Tris-HCl, pH 8.0, 1 mM EDTA), and subjected to 1.7% agarose-gel electrophoresis. Luciferase Assay-p800luc, kindly provided by Dr. R.H. Chen, contained the luciferase expression unit driven by the TGF-␤-responsive elements in the plasminogen activator-1 (PAI-1) promoter. p800luc and pRK␤gal were transfected into the designated cells. 24 h after transfection, the cells were serum-starved for 10 h and then treated with TGF-␤ (5 ng/ml) for an additional 12 h. Luciferase and ␤-galactosidase activities were quantified by the Luciferase Assay System and the ␤-galactosidase Enzyme Assay System (Promega), respectively. The luciferase activity was normalized to ␤-galactosidase activity to account for the transfection efficiency.
PI 3-Kinase Activity Assay-PI 3-kinase activities were assayed according to a procedure described elsewhere (35). Briefly, 10 7 cells were washed twice with ice-cold phosphate-buffered saline and lysed with 1 ml of lysis buffer (137 mM NaCl, 2.7 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 1% Nonidet P-40, 10% glycerol, 1 mg/ml bovine serum albumin, 20 mM Tris, pH 8.0, 2 mM orthovanadate). Cell extracts were incubated with 1 g of anti-phosphotyrosine antibody (Clone 4G10) overnight at 4°C. The immunocomplex was precipitated with 50 l of protein A-Sepharose for 1 h at 4°C, washed three times with lysis buffer, twice with LiCl buffer (0.5 M LiCl, 100 mM Tris, pH 7.6), and twice with TNE buffer (10 mM Tris, pH 7.6, 100 mM NaCl, 1 mM EDTA). The immunocomplex was preincubated on ice for 10 min with 10 l of 20 mM Hepes (pH 7.4) containing 2 mg/ml PI (Sigma). Kinase reaction was performed by the addition of 40 l of the reaction buffer (10 Ci of [␥-32 P]ATP, 20 mM Hepes, pH 7.4, 20 M ATP, 5 mM MgCl 2 ) at room temperature for 15 min. The reaction was stopped by the addition of 100 l of 1 M HCl and extracted with 200 l of a 1:1 mixture of chloroform and methanol. The radiolabeled lipids were separated by thin-layer chromatography and visualized with a PhosphorImager.

HBx Protects Hep 3B Cells from TGF-␤-induced Apoptosis-
Hep 3B cells were transfected with pOP13HApX plasmid. In this plasmid, expression of wild type HBx protein fused to an N-terminal HA epitope was driven by a Rous sarcoma viruslong terminal repeat (RSV-LTR) promoter. G418-resistant colonies were selected, expanded, and verified for expression of HA-HBx fusion protein by immunoprecipitation. Fig. 1 indicates that a HA-HBx protein of the anticipated size (18 kDa) was detected in two representative clones, Hep 3B-H4 and Hep 3B-H13, respectively. Parental Hep 3B and a selected clone, Hep 3B-H11, with no detectable expression were included herein as a negative control. Trypan blue exclusion and colony formation assays were employed to assay cell viability after treatment with TGF-␤. Fig. 2, A and B, illustrates that HBx expression resulted in a 3-4-fold elevation of ID 50 toward TGF-␤. A DNA fragmentation assay was performed to verify that the increased viability of HA-HBx-expressing cells exposed to TGF-␤ is due to their resistance to the apoptotic effect of TGF-␤. 10 h after the addition of 5 ng/ml TGF-␤, a DNA laddering pattern (indicative of internucleosomal DNA cleavage) appeared, and it became apparent at 24 h in parental Hep 3B and Hep 3B-H11 cell lines (Fig. 2C). In contrast, two independent HBx-expressing cell lines (Hep 3B-H4 and Hep 3B-H13) showed little DNA fragmentation even at 24 h after exposure to the same amount of TGF-␤. Therefore, our data indicated that HBx expression can block TGF-␤-induced apoptosis.

HBx Affects Neither the Expression level of TGF-␤ Receptors nor the Transcription Activation Ability of TGF-␤ on the PAI-1
Gene-A previous study by Oshikawa et al. (36) illustrated that mink lung epithelial cells (MvlLu) with HBx expression displayed a reduced growth inhibition response to TGF-␤, which might be related in part to decreased expression of TGF-␤ receptors. Western blot analysis was performed to examine whether the resistance to TGF-␤-induced apoptosis observed in Hep 3B cells with HBx expression is due to a cell surface loss of TGF-␤ receptors. Fig. 3A illustrates that parental Hep 3B as well as selected clones with or without HBx expression contained similar amounts of type I and II TGF-␤ receptors. Fig. 3B reveals the quantitative result after normalization with ␣-tubulin.
In addition to its ability to induce apoptosis, TGF-␤ possesses many biological activities, such as transcriptional activation or repression. The PAI-1 gene is highly induced by TGF-␤ and frequently applied as an indicator of the effects of TGF-␤ on the production of the extracellular matrix (37). A reporter plasmid containing the TGF-␤-responsive element in PAI-1 promoter was employed to evaluate the induction of the PAI-1 gene by TGF-␤ under the influence of HBx and to investigate whether HBx blocks other cellular responses to TGF-␤. In Hep 3B cells,

FIG. 3. HBx affects neither the expression level of TGF-␤ receptors nor the transcriptional activation ability of TGF-␤ on the PAI-1 gene.
A, levels of the type I and type II TGF-␤ receptors were determined by Western blot analysis. ␣-Tubulin was included as an internal control. B, after normalization with ␣-tubulin, relative amounts of type I and type II TGF-␤ receptors were compared with parental Hep 3B cells. C, PAI-1 promoter-driven luciferase reporter plasmid and ␤-galatosidase expression plasmid were co-transfected into the indicated cell lines. Luciferase activity was measured after cells were incubated with TGF-␤ for 12 h and expressed as a -fold induction relative to that from cells not treated with TGF-␤. Data are the means Ϯ S.D. of three independent experiments, each performed in duplicate. 4. Wortmannin and LY294002, two specific inhibitors of PI 3-kinase, alleviated HBx-mediated anti-apoptosis effect. Cells were treated with various agents as indicated. Cell viability was determined by trypan blue staining. The two inhibitors alone did not cause cytotoxicity (data not shown). Data are presented as the means Ϯ S.D. of two independent experiments, each performed in duplicate.

FIG. 5. Enhancement of PI 3-kinase activity by HBx.
A, induction of tyrosine phosphorylation of p85 subunit of PI 3-kinase by HBx. Cell lysates were immunoprecipitated with anti-p85 antibody. The immunocomplexes (IP) were separated by SDS-PAGE and subjected to Western blot (WB) analysis with antibody to phosphotyrosine (upper panel) or p85 (lower panel). B, PI 3-kinase activity was increased in cells with HBx expression. Cell lysates were immunoprecipitated with an anti-phosphotyrosine antibody. The immunocomplex was subjected to in vitro kinase assay using phosphatidylinositol as the substrate in the absence (lanes 1-4) or presence (lanes 5-8) of 30 nM wortmannin. The production of 32 P-labeled phosphatidylinositol 3-phosphate (PI-3-P) was analyzed by thin-layer chromatography, and its position is indicated. TGF-␤ induced a ϳ4.5-fold increase in luciferase activity. HBx expression did not significantly affect the induction (Fig. 3C). Therefore, HBx is likely to blocks a step specific to the apoptotic activity of TGF-␤.

PI 3-Kinase Is Involved in HBx-mediated Anti-apoptosis Effect-
Several studies have demonstrated that, in many circumstances, PI 3-kinase could transmit a survival signal to rescue cells from apoptosis (32,33,38,39). Two specific inhibitors of PI 3-kinase, wortmannin and LY294002, were employed to determine the involvement of PI 3-kinase in the anti-apoptotic effect of HBx. Although these two inhibitors alone did not affect cell viability or alter susceptibility toward TGF-␤ in parental Hep 3B cells (data not shown), wortmannin and LY294002 increased the susceptibility toward TGF-␤-induced toxicity in Hep 3B-H4 and Hep 3B-H13 cells with HBx expression (Fig. 4). The same result was observed using a DNA fragmentation assay (data not shown). These results indicate that in mediating its anti-apoptotic signaling against TGF-␤, HBx might work upstream of the PI 3-kinase. The ability of HBx to induce the activation of PI 3-kinase was then investigated. Fig. 5A reveals that the tyrosine phosphorylation level of the regulatory subunit of PI 3-kinase, p85, is elevated in Hep 3B cells with expression of HBx. Furthermore, HBx induced a substantial increase in PI 3-kinase activity (Fig. 5B). Taken together, our results demonstrate the critical role of PI 3-kinase in mediating the anti-apoptotic effect of HBx against TGF-␤.

Protection Effect of HBx against TGF-␤-induced Apoptosis Correlated with Its Transactivation Activity-Point mutant
HBx proteins were utilized to determine whether the transactivation function was required for the anti-apoptotic effect of HBx. Pooled cells with doxycycline-inducible expression of green fluorescence protein (GFP), HBx, HBxGFP, HBx 7 GFP, HBx 61 GFP, HBx 69 GFP, and HBx 90 -91 GFP were generated using the RevTet TM system (CLONTECH). Direct microscopic observation of green fluorescence (data not shown) or Western blot analysis confirmed expression of the desired products (Fig. 6A). The Transactivation ability of the desired HBx or HBx mutants on NF-B-responsive element-driven luciferase reporter was tested. The fusion with GFP did not affect the HBx-mediated transactivation function. Although HBx 7 GFP retained its transactivation activity, as previously reported, HBx 61 GFP, HBx 69 GFP, and HBx 90 -91 GFP lost their transactivation ability (data not shown) (15). A DNA fragmentation assay was employed to monitor the indicated HBx or mutant effects on TGF-␤-induced apoptosis. Fig. 6B illustrates that transactivation-proficient HBx (HBx, HBxGFP, and HBx 7 GFP) reduced TGF-␤-induced DNA laddering and that transactivation-deficient HBx (HBx 61 GFP, HBx 69 GFP, HBx 90 -91 GFP) lost their ability to protect cells from TGF-␤-induced apoptosis. Using a quantitative cell viability assay by trypan blue staining, the same results were observed (Fig. 6C). Therefore, the domain of HBx responsible for anti-TGF-␤-induced apoptosis overlapped with the region of HBx previously demonstrated to be important to the HBx-mediated transactivation function.
Correlation of HBx-mediated Activation of PI 3-Kinase/Akt Signaling Pathway with Its Transactivation Ability-Having demonstrated the involvement of PI 3-kinase in the anti-apoptotic signaling of HBx, the domain required for activation of PI 3-kinase was then mapped. In our system, the tyrosine phosphorylation status of the regulatory subunit of PI 3-kinase, p85, correlated with PI 3-kinase activity. Therefore, the tyro-FIG. 6. The anti-apoptotic effect of HBx correlated with its transactivation activity. A, establishment of Hep 3B cells with inducible expression of HBx or the desired HBx-GFP fusion proteins. The tetracycline-regulated RevTet TM system (CLONTECH) was utilized for inducible expression of HBx and its mutants. All of the cells with expression of the indicated products were generated according to the manufacturer's instruction as described under "Experimental Procedures." Western blot analysis with anti-HBx antibody was employed to verify the expression of the desired products in the absence (Ϫ) or presence (ϩ) of 1 g/ml doxycycline (Dox.). B, transactivation-deficient HBx mutants lost their ability to protect cells from TGF-␤-induced apoptosis. After incubation with (lower panel) or without (upper panel) TGF-␤ (5 ng/ml) for 16 h, DNA was extracted and subjected to agarosegel electrophoresis. C, viability of cells with expression of the indicated products following TGF-␤ treatment in the absence (Ϫ) or presence (ϩ) of wortmannin using trypan blue exclusion assay. sine phosphorylation level of p85 was monitored by immunoprecipitation with anti-p85 antibody followed by Western blotting employing an anti-phosphotyrosine antibody in cells with or without induction of expression of the indicated products (Fig. 7A). All mutations that affected the transactivation function of HBx revealed no elevation of the tyrosine phosphorylation level of p85. The serine/threonine kinase, Akt, a downstream target of PI 3-kinase, is a critical component of PI 3-kinase anti-apoptotic signaling pathway. Activation of Akt requires phosphorylation of Ser-473 (40). Anti-active form Akt antibody was utilized to monitor Akt status by Western blot analysis. Fig. 7B illustrates elevated phosphorylation at Ser-473 of Akt in the transactivation-proficient but not the transactivation-deficient HBx mutants. These results indicate that activation of the PI 3-kinase/Akt signaling pathway by HBx correlated with its transactivation function and played a significant role in mediating the anti-apoptotic activity.

DISCUSSION
This study has demonstrated that HBx effectively suppresses TGF-␤-induced apoptotic death of hepatoma cells. The HBx-mediated anti-apoptotic effect was not mediated through decreased expression of TGF-␤ receptors. Two specific inhibitors of PI 3-kinase, wortmannin and LY294002, blocked the anti-apoptotic effect of HBx, implying that HBx might affect the PI 3-kinase signaling pathway in mediating the effect. In cells expressing HBx, the PI 3-kinase activity and not its protein level was elevated. An increased phosphorylation of Akt at Ser-473 resulted. The anti-apoptotic mechanism of HBx was attributed, at least in part, to the activation of PI 3-kinase signaling cascades.
The effects of HBx on cell death or apoptosis have been studied by several groups. p53-dependent apoptosis was prevented by microinjection of HBx into primary fibroblasts (18). Chirillo et al. (19) demonstrated that after DNA damage, HBx induced p53-dependent apoptosis in NIH3T3 cells transiently expressing HBx. In Chang liver cells, HBx failed to induce apoptosis; however, it did sensitize cells to apoptosis triggered by TNF-␣ (20). Upon the induction of HBx expression mediated by the Cre/loxP recombination system, liver cell apoptosis was observed independently of the p53 pathway (41). The liver cells derived from a transgenic mouse were more susceptible to diverse apoptosis insults, and this phenomenon was not dependent upon p53 (42). These seemingly contradictory results of HBx on apoptotic events might be attributable to the utilization of different cells and expression systems. However, these results suggest that HBx affects the apoptotic processes by multiple mechanisms, including the inactivation of the p53 functions, interference of DNA repair ability, or modulation of cellular signaling cascades. Our findings on the modulation of PI 3-kinase signaling by HBx in mediating its anti-apoptotic effect offered a new mechanism.
The molecular mechanism by which HBx activates PI 3-kinase is addressed hereafter. Because of its well known transactivation function through the increased expression of cytokines or cognate receptors, HBx might establish an autocrine or paracrine loop. HBx was reported to stimulate the expression of cytokines (e.g. interleukin-6 (43) and insulin-like growth factor-II (44)) as well as cytokine receptors (e.g. insulin-like growth factor-I receptor (45) and epidermal growth factor receptor (46)). Interleukin-6 was demonstrated to inhibit apoptosis through the PI 3-kinase signaling pathway in hepatoma cells (32). The phosphorylated tyrosine residues, generated on receptors (e.g. epidermal growth factor receptor) or their associated substrate molecules (such as IRS-1/2 in signaling by insulin and insulin-like growth factor), form the docking sites for the Src homology-2 domains of p85. The interaction mediates the translocation of PI 3-kinase to the receptor tyrosine kinases and their substrate and assists in positioning p110, the catalytic subunit of PI 3-kinase, close to the membranes that contain the lipid substrates. Using Northern blot analysis or a ribonuclease protection assay, HBx did not alter the mRNA level of interleukin-6, TNF-␣, or interferon-␤ (data not shown). The addition of media collected from the overnight culture of HBx-expressing cells did not protect Hep 3B cells from TGF-␤induced apoptosis (data not shown). Although not excluded, the hypothesis that the observed activation of PI 3-kinase by HBx might be due to the secondary results of the primary activation of cytokines or growth factors was not favored.
Rather, HBx might work through the modulation of signaling cascades to activate PI 3-kinase. Related investigations reported that HBx modulates several signaling cascades (7)(8)(9)(10)(11)(12). Benn and Schneider (7) indicated that HBx activated Ras-GTP complex formation. Activation of Src family kinases was demonstrated to be indispensable for HBx-mediated activation of Ras (12). The activated Ras-GTP complex binds to p110, the catalytic subunit of PI 3-kinase, resulting in the activation of PI 3-kinase (47). The binding of the Src homology-3 domain of Src family kinases to a proline-rich region within the p85 of PI 3-kinase resulted in the activation of PI 3-kinase (48). Lately, HBx has been shown to interact with Jak1 and activate Jak-STAT signaling (49). Both Jak (50 -52) and STAT (53) interact with p85 and activate PI 3-kinase signaling pathway. Therefore, through the activation of Ras, Src family kinases, or JAK-STAT, HBx might be able to achieve its effect on PI 3-kinase. In addition, HBx might directly activate PI 3-kinase. In a glutathione S-transferase pull-down assay, HBx was found to be associated with p110 (data not shown). The contribution of this interaction in the observed elevation of PI 3-kinase activity requires further investigation.
It is noteworthy that wortmannin and LY294002 partially blocked the anti-apoptotic activity of HBx (Fig. 4). Although activation of the PI 3-kinase/Akt signaling pathway mediated the observed phenomenon, the interference of other molecules by HBx cannot be excluded. A recent investigation confirmed that HBx can inhibit caspase 3 activity (22). Chen and Chang (30) reported that caspase 3 was involved in TGF-␤-induced apoptosis. However, the contribution of the inhibitory effect of HBx on caspase 3 in our system remains to be elucidated.
Polyoma middle T antigen (PMT) is identified as the tumorigenic component of the polyoma virus. PMT forms a complex with pp60 c-src and PI 3-kinase, subsequently activating PI 3-kinase (54). Several studies have inferred that a PI 3-kinase signaling pathway is required for PMT-mediated tumorigenesis (35,38). Our study clearly demonstrated the activation of PI 3-kinase signaling by HBx, subsequently triggering anti-apoptotic signaling. As with PMT, inhibition of apoptosis by HBx could disrupt the normal cellular surveillance mechanism for removing damaged cells, thereby providing a clonal selective advantage for hepatocytes expressing this integrated viral gene during the early stages of human liver carcinogenesis. Mutations that affected the transactivation activity of HBx inhibited its ability to activate PI 3-kinase/Akt signaling pathway and failed to block apoptosis. These observations indicate that transactivation and anti-apoptotic activity of HBx are linked. However, Gottlob et al. (55) have reported that transactivation activity is not required for the transforming activity of HBx in REV2 cells. Therefore, additional studies are required to further define whether PI 3-kinase activation and subsequently anti-apoptotic activity are a prerequisite for HBx-mediated transformation.