Oncogenic virus proteins often target to tumor suppressor p53 during virus life cycle. In the case of hepatitis C virus (HCV) core protein, it has been shown to affect p53-dependent transcription. Here, we further characterized the in vitro and in vivo interactions between HCV core protein and p53 and showed that these two proteins colocalized in subnuclear granular structures and the perinuclear area. By use of a reporter assay, we observed that while low level of HCV core protein enhanced the transactivational activity of p53, high level of HCV core protein inhibited this activity. In both cases, however, HCV core protein increased the p53 DNA-binding affinity in gel retardation analyses, likely due to the hyperacetylation of p53 Lys373 and Lys382 residues. Additionally, HCV core protein, depending on its expression level, had differential effects on the Ser15 phosphorylation of p53. Moreover, HCV core protein could rescue p53-mediated suppressive effects on both RNA polymerase I and III transcriptions. Collectively, our results indicate that HCV core protein targets to p53 pathway via at least three means: physical interaction, modulation of p53 gene regulatory activity and post-translational modification. This feature of HCV core protein, may potentially contribute to the HCV-associated pathogenesis.
Hepatitis C virus (HCV) infection mostly leads to a number of liver diseases, including chronic infection, cirrhosis, and hepatocellular carcinoma (HCC) (Saito et al., 1990; Bukh et al., 1993). The molecular mechanisms of HCV persistent infection and pathogenesis are still obscure. Extensive studies have been conducted to elucidate the interference between the virus products and host cellular factors. In this aspect, the core protein of HCV, which exhibits pleiotropic functions, appears to be noted (for reviews, see Lai and Ware, 2000; Ray and Ray, 2001). HCV core protein is located at the N-terminal portion of the HCV polyprotein and is highly conserved among various HCV subtypes (Santolini et al., 1994). Full-length HCV core protein consists of 191 amino acids, and can be phosphorylated in vivo (Shih et al., 1995). Immunofluorescence analyses indicate that it localizes at both nuclear and cytoplasmic regions (Barba et al., 1997; Yasui et al., 1998; Chen et al., 2003). Apart from acting as the basic building block of viral nucleocapsid, it also interacts with a variety of cellular factors, interfering with their normal functions (Lai and Ware, 2000; Ray and Ray, 2001). For instance, HCV core protein has been shown to target several cellular transcription factors such as hnRNP K (Hsieh et al., 1998), LZIP (Jin et al., 2000), 14-3-3 (Aoki et al., 2000), p21/WAF1 (Wang et al., 2000), and RNA helicase CAP-Rf (You et al., 1999), resulting in the alterations in the responsive transcription regulatory activities. Moreover, HCV core protein is able to cooperate with ras oncogene in the transformation of rodent fibroblasts under certain conditions (Chang et al., 1998), to influence host cell growth and proliferation through different mechanisms (Aoki et al., 2000; Cho et al., 2001; Erhardt et al., 2002), to promote immortalization of primary human hepatocytes (Ray et al., 2000), and to cause HCC formation at the late stage of transgenic mice (Moriya et al., 1998). Together, a growing body of evidence suggests the direct involvement of HCV core protein in the HCV-associated pathogenesis and carcinogenesis.
Over half of human tumors have been associated with p53 mutations, indicating its pivotal role as a tumor suppressor protein (Hollstein et al., 1991). It is widely recognized that the sequence-specific DNA binding, transcriptional activation, regulation of DNA replication, and the capacity to induce cellular growth arrest and apoptosis are critical for p53 functions (Levine, 1997; Agarwal et al., 1998; Vogelstein et al., 2000). Oncogenic viral proteins such as SV40T antigen (Sheppard et al., 1999), adenovirus E1B protein (Martin and Berk, 1998), and hepatitis B virus X protein (Truant et al., 1995) have been shown to interact with p53 and alter its functions through distinct mechanisms. Generally, regulation of p53 functions involves mainly the post-translational modifications, including phosphorylation, acetylation, sumoylation, and ubiquitination of p53 (Jayaraman and Prives, 1999; Jimenez et al., 1999), thus establishing sophisticated regulatory cascades. Through the interference with regulatory components of p53 or the direct targeting to p53, oncogenic virus proteins can modulate the p53 functions and mediate the escape from p53-mediated cellular surveillance. Interestingly, this may also be the case for HCV. Previous studies indicate that of 10 viral proteins encoded by the HCV genome, HCV core protein, and nonstructural protein NS3 and NS5A physically interact with p53, albeit their effects on p53 transcriptional activity vary and are not conclusive yet (Ishido and Hotta, 1998; Ray et al., 1998; Lu et al., 1999; Otsuka et al., 2000; Arima et al., 2001; Kwun et al., 2001; Majumder et al., 2001). In the case of HCV core protein, the evidence of in vivo association and cellular localization is not available. To this end, we further examined the physical interaction of these two proteins and the effects of HCV core protein on p53-dependent transactivation. In this work, we presented evidence showing that HCV core protein and p53 associate both in vitro and in vivo, and colocalize at the perinuclear region and the subnuclear granular structures. Moreover, different from the findings by Lu et al. (1999) and Otsuka et al. (2000) that p53 transactivational activity is enhanced by HCV core protein, our results show that while low level of HCV core protein cooperates with p53 to transactivate the p53-responsive promoter activity, higher level of HCV core protein on the other hand downregulates the p53-mediated transactivation. Surprisingly, regardless of the activation or suppression of p53-dependent transcription, HCV core protein enhances the in vitro p53 DNA-binding affinity, presumably attributed by the induction of hyperacetylation of p53 at Lys373 and Lys382 residues. Additionally, Ser15 phosphorylation of p53 is differentially regulated in HepG2 cells transiently expressing HCV core protein, whereas the Ser392 residue of p53 is constitutively phosphorylated and is not affected. Together, these findings indicate that HCV core protein targets to p53 pathway via at least three different means, that is, physical interaction, deregulation of p53-mediated transactivation, and alteration of p53 post-translational modifications. Notably, p53 is a general repressor for both RNA polymerase (Pol) I- and Pol III-dependent transcriptions (Cairns and White, 1998; Zhai and Comai, 2000), while here we demonstrate that HCV core protein can rescue the p53-mediated suppression on both RNA polymerase systems. Since the activities of RNA Pol I and Pol III are strictly modulated in response to signals controlling cell growth and proliferation (Larminie et al., 1998; Paule and White, 2000), this feature of HCV core protein to relieve the p53-mediated suppression is consistent with its functional roles in growth and proliferation promotion and likely the HCV-associated oncogenesis.
Characterization of HCV core protein–p53 interaction
To further characterize the interaction between HCV core protein and tumor suppressor p53, GST pull-down analysis was performed using GST-HCV core protein variants harboring full-length (GST/HCVc195) and various length of the N-terminal fragments of HCV core protein (Figure 1b). As shown in Figure 1a, in vitro synthesized 35S-Met-labeled p53 was specifically retained by all GST/HCV core variants (lanes 4–7), except GST/HCVc24 (lane 3) and control GST molecules (lane 2). Therefore, N-terminal 50-amino-acid fragment of HCV core protein is sufficient to bind p53. Next, a reciprocal experiment using in vitro synthesized HCV core protein and a series of GST/p53 variants harboring full-length and specific domains of p53 (Figure 1e) was performed to delineate the HCV core protein-interacting domain(s) on p53. Results shown in Figures 1c and d indicated that, in contrast to the GST/p53-truncated variants lacking p53 N-terminal 75 or 160 amino-acid residues (GST/p53ΔN75 or GST/p53ΔN160), all GST/p53-truncated variants lacking their C-terminal portion (GST/p53ΔC30, GST/p53ΔC55, GST/p53ΔC75 or GST/p53ΔC150) failed to pull down HCV core protein, suggesting that the integrity of p53 C-terminal region, but not that of the N-terminus, was essential for the interaction to HCV core protein. When a series of GST/p53 variants containing only the C-terminal 30, 55, and 75 amino-acid residues of p53 (see Figure 1e) were used for pull-down assay, only the GST/p53-C75 that comprises the p53 C-terminal last 75 amino-acid residues, but not GST/p53-C30 or GST/p53-C55, associated with HCV core protein (Figure 1d, lanes 8–10). Together, our results support the finding that HCV core protein and p53 interact in vitro. Furthermore, we showed that the N-terminal 50-amino-acid fragment of HCV core protein and the last C-terminal 75-amino-acid segment of p53 (residues 319–393) are involved in HCV core protein–p53 association (Figure 1b and e).
The interaction between HCV core protein and p53 was further investigated by the in vivo co-immunoprecipitation (co-IP) analysis. The nuclear extracts from HCV core protein-producing HepG2 cells (G2/C) were immunoprecipitated with anti-p53 antibody-conjugated agarose resins, and the resulted immunoprecipitates were analysed by Western blotting with antibody against HCV core protein. Results shown in Figure 2a indicated that HCV core protein was co-precipitated with cellular p53 (lane 4), suggesting that these two proteins form a specific complex in the nucleus compartment. Additionally, when we examined the association of three C-terminally truncated p53 variants (p53ΔC30, p53ΔC55, and p53ΔC75) with HCV core protein in p53-null H1299 cells using similar co-IP experiments, our results indicated that, in consistent with the in vitro studies, only the full-length p53 protein is able to associate with HCV core protein in vivo (Figure 2b, lane 2). Taken together, our experiments demonstrate that HCV core protein forms a specific complex with p53 both in vitro and in vivo. Notably, the C-terminal last 75-residue segment of p53, which encompasses both tetramerization domain and basic regulatory domain (Ko and Prives, 1996; Jayaraman et al., 1997), is essential for the interaction with HCV core protein.
Colocalization of HCV core protein and p53
Next, we investigated the subcellular localization of HCV core protein and p53 by performing indirect immunofluorescence microscopy. HeLa cells transfected with p53 (pCEP4-p53) and/or HCV core protein expression vectors (pSRα/HCVc195) were fixed, reacted with anti-HCV core serum and anti-p53 antibody, and observed under confocal microscopy (see Materials and methods). When expressed alone, HCV core proteins were present mainly throughout the cytoplasm (Figure 3a; panels a and b), and the exogenous p53 showed a typical diffuse nuclear staining (Figure 3a, panels c and d). Interestingly, when these two proteins coexpressed in HeLa cells, they colocalized at the perinuclear region and nuclear granular structures (Figure 3b, panels a–c; yellow color). The confocal immunofluorescence analysis thus provides further evidence that HCV core protein interacts with p53 at distinct cellular localizations.
HCV core protein differentially modulates the transactivational activity of p53
To study the effects of HCV core protein on p53-mediated transactivation, a luciferase reporter construct (p53CON; Chen et al., 1993a) controlled by two copies of synthetic p53-responsive elements upstream the heat-shock protein 70 (hsp70) TATA minimal promoter was used to monitor the activity of p53-dependent transcription. Increasing amounts of HCV core protein expression constructs (pSRα/HCVc195 or pSRα/HCVc101) and a fixed level of p53 expression constructs (pMT-p53) along with the p53CON were cotransfected into HuH7 cells. When expressed at low level, full-length HCV core protein (pSRα/HCVc195) modestly coactivated the p53-dependent reporter activity (2.6–1.6 fold; Figure 4a, lanes 2 and 3). On the other hand, higher level of full-length HCV core protein repressed the reporter activity to 60–12% of that when p53 was transiently expressed alone (Figure 4a, lanes 4 and 5). Interestingly, the C-terminal truncated variant of HCV core protein (pSRα/HCVc101; 1–101 amino-acid residues of HCV core protein) did not show any effect on the reporter activity, regardless its expression level (Figure 4a). Western blot analysis indicated that the level of p53 expression was not altered by either HCV core protein (Figure 4a, bottom panel; HCV Core 101 and HCV Core 195). Therefore, the dual effects on p53-responsive reporter gene activity were not due to the availability of p53 in the presence of HCV core protein. Moreover, HCV core protein alone had no significant effect on p53CON reporter gene activity in the absence of exogenous p53 (data not shown).
To determine whether the dual effects of HCV core protein on the p53-dependent transactivational activity were also applicable in other cell lines, we further examined the effects of HCV core protein in several cell lines with various status of p53 using two different HCV core protein expression constructs, pECE/HCVC-KF and pSRα/HCVc195, for transfection. As evidenced by the immunoblot shown in Figure 4b, the expression level of HCV core protein in pSRα/HCVc195-transfected HepG2 cells is at least 7–10 times higher than those of pECE/HCVC-KF-transfected cells. The two expression constructs were then transfected into HuH7 cells separately along with a fixed level of pMT-p53 and p53CON reporter gene. While less amount of HCV core protein was expressed (Figure 4c, left panel; as the case in pECE/HCVC-KF-transfected cells), the p53-responsive reporter activity was enhanced about threefold. In cells that express high level of HCV core protein (Figure 4c, right panel; as the case in pSRα/HCVc195-transfected cells), the reporter activity was reduced to 17% of that in control cells. Essentially, the same conclusion was obtained when using HepG2 (wild-type p53, without exogenous p53), Hep3B (p53-null), and H1299 (p53-null) cells with endogenous or exogenous p53 for study, albeit the activation or repression fold varied with the weak transactivation observed in pECE/HCVC-KF-transfected HepG2 and H1299 cells. This implies that the transcription regulatory functions of HCV core protein on p53 are not restricted to a particular cell line. Furthermore, to assess the effect of HCV core protein on cellular p53 target gene expression, we examined the expression levels of p53 downstream target genes, p21/WAF1 (El-Deiry et al., 1993) and proliferating cell nuclear antigen (PCNA; Jackson et al., 1994; Shivakumar et al., 1995), in HepG2 cells transiently expressing HCV core protein. As shown in Figure 4b, we found a significant decrease in p21/WAF1 expression in cells transiently expressing a high level of HCV core protein (compare lanes 3 and 4), and modest increase in cells expressing low level of HCV core protein (compare lanes 1 and 2). As noted, the decrease of p21/WAF1 expression in pSRα/HCVc195-transfected HepG2 cells was not due to a decrease of endogenous p53 protein (Figure 4b). However, under similar experimental conditions, the expression of PCNA, which is negatively regulated by p53, was not affected by HCV core protein. In conclusion, our results suggest that HCV core protein has the ability to differentially modulate p53-mediated transactivation depending on the level of HCV core protein and the promoter architectures of p53-responsive genes.
In vitro DNA-binding activity of p53 is enhanced by HCV core protein
A number of studies indicate that C-terminal regulatory domain of p53 plays a key role in modulating its sequence-specific DNA-binding activity (Gu and Roeder, 1997 and references within). Since the C-terminal last 75 amino-acid residues of p53 are responsible for the interaction with HCV core protein, it is likely that this interaction may affect p53 DNA-binding activity. To test this possibility, gel mobility shift assays (EMSA) were performed. As shown in Figure 5, nuclear extracts prepared from HepG2 cells or from HepG2 cells transfected with control vector pSRα or pECE produced a single p53-DNA shift in the EMSA gel (Figure 5; lanes 2, 3, and 5). This shift band was sequence specific because it could be competed by a 50-fold molar excess of unlabeled oligonucleotides containing the wild-type p53 binding site (lane 8), but not by oligonucleotides containing the mutated p53-binding site (lane 7). Surprisingly, irrespective of low (pECE/HCVC-KF-transfected cells) or high (pSRα/HCVc195-transfected cells) levels of HCV core protein transiently expressed in HepG2 cells, the p53 DNA-binding activity was enhanced about threefold (Figure 5; compare lanes 3 and 4, lanes 5 and 6). As p53 protein level was not altered by HCV core protein (Figure 4b), the enhancement of p53 DNA-binding activity cannot be ascribed as a change in cellular p53 level.
HCV core protein differentially regulates the post-translational modifications of p53
It is well established that post-translational modifications such as phosphorylation, acetylation, and other kinds of covalent modifications are the major mechanisms regulating the biological functions of p53 in response to a variety of stimuli (Giaccia and Kastan, 1998 and references therein). Our results above showed that HCV core protein biphasically regulates p53 transactivational activity, and augments p53 DNA-binding affinity in vitro. However, the mechanism(s) of HCV core protein's dual effects on p53-dependent transactivation cannot be fully explained by the increase in p53 DNA-binding activity. Recent studies have shown that acetylation catalysed by p300/CBP and PCAF (p300/CBP-associated factor) at specific Lys residues in the C-terminus of p53 potentiates p53 sequence-specific DNA-binding activity and influences interactions with other cellular factors that bind to this domain (Prives and Manley, 2001). To examine the effect of HCV core protein on the acetylation status of cellular p53, we performed p53 IP–Western analysis in HepG2 cells transiently expressing HCV core protein. The levels of immunoprecipitated p53 were normalized by immunoblotting (Figure 6a, bottom), and p53 acetylation was assessed using specific antiacetylated p53 antibodies. As shown in Figure 6a, the acetylation of p53 at Lys373 and Lys382 was upregulated in HepG2 cells expressing either low (pECE/HCVC-KF-transfected cells) or high level (pSRα/HCVc195-transfected cells) of HCV core protein (compare lanes 1 and 2, lanes 3 and 4), whereas the Lys320 acetylation of p53 was not affected (top). Since the hyperacetylation of C-terminal region of p53 has been implicated in the enhancement of p53 sequence-specific DNA binding in vitro, presumably, HCV core protein-induced hyperacetylation of p53 at Lys373 and Lys382 of p53 may lead to the increase of p53 DNA-binding activity. Besides acetylation, p53 can be phosphorylated by various cellular kinases at multiple Ser and Thr residues in the N- and C-terminal region of p53 in response to different cellular traumas (Appella and Anderson, 2000,2001). To investigate the effect of HCV core protein on the phosphorylation states of p53, similar IP–Western analysis was performed in the transfected HepG2 cells. The immunoprecipiates were then analysed with either anti-phospho-Ser15 p53 (Figure 6b, top) or anti-phospho-Ser392 p53 (middle). As indicated, the Ser392 residue of p53 was constitutively phosphorylated in HepG2 cells transfected with either pECE/HCVC-KF or pSRα/HCVc195 (Figure 6b, middle). Interestingly, while the Ser15 phosphorylation of p53 was slightly enhanced in cells expressing low level of HCV core protein, the Ser15 phosphorylation of p53 was substantially suppressed when high level of HCV core protein was expressed (Figure 6b, top, compare lanes 1 and 2, lanes 3 and 4). Together, these findings indicate that HCV core protein can induce p53 hyperacetylation at Lys373 and Lys382 residues, and biphasically alter the Ser15 phosphorylation status depending on the expression level of HCV core protein. Presumably, these collective effects of the HCV core protein-induced post-translational modifications on endogenous p53 may establish the differential regulation of p53 gene transactivational activity.
HCV core protein rescues the p53-mediated suppression of RNA Pol I- and III-dependent promoter activities
Recent studies by Cairns and White, 1998 and Zhai and Comai, 2000 indicate that p53 not only transregulates a variety of protein-coding genes but also acts as a general repressor for both RNA Pol I and Pol III transcriptions. To test whether HCV core protein could affect the p53-mediated suppression on both RNA polymerase systems, an RNA Pol I- (prHu3-Luc, a human ribosomal RNA promoter-driven reporter; Figure 7a) and an RNA Pol III-dependent (pArg-tRNA-Luc, a Drosophila melanogaster Arg-specific tRNA promoter-driven reporter; Figure 7b) luciferase reporters were used in the transient reporter assay. As shown in Figure 7a, our in vivo reporter analysis indicated that both RNA Pol I- and Pol III-dependent reporter gene activities were repressed by increasing amounts of exogenous p53 in HeLa cells (Figures 7a and b; lanes 2–5; approximately ∼50% suppression in either reporter activity). By cotransfection of increasing levels of HCV core protein and a fixed amount of p53 expression plasmids into HeLa cells, the suppressive effects of p53 on both RNA Pol I- and Pol III-dependent promoter were gradually incapacitated (lanes 6–10), while expression of HCV core protein alone significantly upregulated both RNA Pol I- and Pol III-dependent reporter gene activity (lane 11). Thus, our results indicate that HCV core protein has the potential to activate RNA Pol I and Pol III transcriptions even in the presence of functional p53.
In this study, we demonstrated that HCV core protein and tumor suppressor p53 form a physical complex both in vitro and in vivo (Figures 1–2), and both colocalize at the perinuclear region and nuclear granular structures. In comparison to the findings by Lu et al. (1999) and Otsuka et al. (2000), our reporter analyses indicate that HCV core protein has dual effects on p53-mediated transactivation depending on viral protein concentration: inhibition at high level of expression and coactivation by low level of HCV core protein (Figure 4). We also showed that the acetylation and phosphorylation states of p53 are significantly altered in cells expressing HCV core protein (Figure 6). Other than its transcription regulatory functions on RNA Pol II transcription system, the suppressive effects of p53 on both RNA Pol I and Pol III transcription systems are relieved by HCV core protein (Figure 7). More importantly, overexpression of HCV core protein alone is able to activate the activities of two RNA Pol transcription systems (Figure 7, lane 11). Since rapid proliferating cells such as tumor cells require more efficient rRNA and tRNA transcription to achieve the elevated rates of cellular biosynthesis (Larminie et al., 1998), activation of RNA Pol I and Pol III transcriptions by HCV core protein is likely one of mechanisms for its functional roles in promoting cell proliferation and causing liver oncogenesis. Conclusively, HCV core protein has the potential to fine tune p53 functions via at least three means: physical interaction, modulation of p53 transcriptional activity, and post-translational modifications.
Earlier studies have shown that the extreme C-terminal 75 amino acids of p53, which comprise the tetramerization domain, three NLSs, one NES, and the regulatory domain, play important roles in the regulation of p53 biological functions (Ko and Prives, 1996; Jayaraman et al., 1997). Besides HCV core protein, other two HCV-encoded viral products, nonstructural protein 3 (NS3; Ishido and Hotta, 1998) and nonstructural protein 5A (NS5A; Majumder et al., 2001; Lan et al., 2002), have been recently demonstrated capable of interacting with the C-terminus of p53 and thereby deregulate p53 normal functions. Both NS3 and NS5 viral proteins, when being overexpressed in cells, cause the downregulation of p53-dependent transactivational activity. It is known that the extreme C-terminal domain of p53 interacts with numerous viral proteins, cellular transcription factors, and a wide range of cellular kinases, including HBV X protein (Truant et al., 1995), adenovirus E4orf6 (Dobner et al., 1996), TBP (Martin et al., 1993), TFIIH (Xiao et al., 1994), PCAF (Liu et al., 2000), and CKII (Filhol et al., 1992). Through the physical interactions with these protein factors, the functional status of p53 is finely regulated. While three HCV viral products target to the same region of p53, the combinatory effects have not been directly examined. It is also noted that, among three HCV viral proteins, only HCV core protein is shown to differentially regulate p53 transactivation activity and post-translational modification.
Although the gene transactivation function of p53 is biphasically regulated by HCV core protein (Figure 4), the in vitro DNA-binding activity of p53 is increased in the presence of either low or high level of HCV core protein (Figure 5). Moreover, the mobility of p53-DNA complexes is not altered by HCV core protein. Conceivably, HCV core protein may not associate with the p53–DNA complex, or the interaction of HCV core protein and the p53–DNA complex is relatively transient. Since the level of cellular p53 in HepG2 cells is not altered, this enhancement of p53 DNA-binding activity by HCV core protein is likely mediated through other mechanisms, such as post-translational modifications. Consistent with this notion, we showed that the Lys373 and Lys382 residues, but not Lys320, are hyperacetylated in HepG2 cells transiently expressing either low or high level of HCV core protein (Figure 6a). It has been demonstrated that Lys373 and Lys382 residues of p53 are inducibly acetylated by p300/CBP acetyltransferase, whereas Lys320 residue is acetylated by PCAF (Gu and Roeder, 1997; Sakaguchi et al., 1998; Liu et al., 1999). In views that the acetylation at the C-terminal Lys residues of p53 is associated with its in vitro DNA-binding activity, presumably the HCV core protein-induced hyperacetylation of p53 can lead to the enhanced p53 DNA-binding activity, which is in consistent with our observations obtained in the EMSA experiments (Figure 5). Interestingly, our preliminary protein interaction study shows that HCV core protein interact with p300/CBP acetyltransferase both in vitro and in vivo (unpublished observations), thus raising the possibility that HCV core protein may directly modulate the p300/CBP acetyltransferase activity. It would be interesting to know whether the complex containing HCV core protein, p53, and p300 forms in vivo, and whether HCV core protein-induced p53 hyperacetylation influences the interaction of p53 to other cellular factors.
In comparison to acetylation, phosphorylaiton at Ser15 residue of p53 is modestly enhanced in the presence of low-level HCV core protein, while this residue is significantly repressed in the presence of high-level of HCV core protein (Figure 6b). These observations suggest that the states of p53 Ser15 phosphorylation are functionally related to its gene transactivation activity. Cellular kinases, such as DNA-PK, ATM, ATR, hCHK1, and hCHK2, are capable of phosphorylating Ser15 residue of p53 when in response to a variety of cellular traumas and signals (for a review, see Appella and Anderson, 2000,2001). Recent studies have shown that the hyperphosphorylation of p53 at Ser15 residue does not significantly alter the binding of p53 to TBP (Chen et al., 1993b), TAFII31 (Lu and Levine, 1995), and MDM2 (Shieh et al., 1997; Unger et al., 1999) in vitro, whereas it facilitates and stabilizes the association with p300/CBP coactivators in vitro, consequently resulting in the activation of p53-mediated gene transcription (Dumaz and Meek, 1999). With respect to these observations, it is possible that the positive and negative regulation on the Ser15 phosphorylation of p53 in the presence of HCV core protein determines the capability for p53 to recruit, alternatively to exclude, p300/CBP coactivators or other transcription cofactors to the p53-responsive promoter. It is not yet clear as to which cellular kinase(s) or phsophatase(s) that mediates the Ser15 phosphorylation or dephosphorylation of p53 is interfered and regulated by HCV core protein. As also noted, additional phosphorylation events occur on p53 simultaneously. The possibility that HCV core protein has the potential to modulate other phosphorylation events cannot be formally ruled out. For instance, cellular kinases, Raf-1 and JNK, which are functionally activated in cells expressing HCV core protein (Shrivastava et al., 1998; Aoki et al., 2000), have been demonstrated to phosphorylate p53 in vitro and in vivo in response to stressful conditions such as DNA damages (Jamal and Ziff, 1995; Miline et al., 1995). Since the functional consequence of increased p53 DNA-binding activity does not correlate with the inhibition of p53 gene transactivational activity in cells expressing high-level HCV core protein, the dual regulation of p53 Ser15 phosphorylation is likely one of the mechanisms to differentially modulate p53 gene transactivation functions by HCV core protein.
Several previous reports (Barba et al., 1997; Yasui et al., 1998) together with this study have shown that HCV core protein localizes predominantly at the cytoplasmic region and to a less extent in the nucleus as granular nuclear dots. While the overexpressed p53 show homogeneous nuclear staining, we found that p53 and HCV core protein mostly coexist at the nuclear granular spots and the perinuclear region of HeLa cells (Figure 3). Recently, functional association between p53 and promyelocytic leukemia protein (PML) has been characterized, and these two proteins predominantly colocalize at the PML-nuclear bodies (PML-NBs; Lain et al., 1999; Fogal et al., 2000). Relocalization of p53 into PML-NBs is important for its gene transactivational activity, presumably by modulating p53 modifications such as SUMOylation and acetylation (Fogal et al., 2000). PML-NBs are cell cycle-regulated matrix-associated subnuclear structures that appear as punctate foci in the interphase nuclei. It has been suggested that PML-NBs involve in many different cellular functions such as transcription regulation, growth suppression, and apoptosis (Seeler and Dejean, 1999). Since the cellular codistribution of HCV core protein and p53 exhibits a similar nuclear topology as that of PML and p53, it is tempting to speculate that these three proteins may colocalize at the PML-NBs. Two observations support this hypothesis: (i) HCV core protein interacts with the acetyltransferase p300, which is one of the component of PML-NB and associates with PML protein (Seeler and Dejean, 1999; Zhong et al., 2000; Jensen et al., 2001), and modulates p53 acetyaltion; (ii) Cellular level of PML is affected by DNA or RNA virus infection, and PML-NBs are disrupted during the early phases of virus infection. Particularly, the upregulation of PML expression is also found associated with the inflammatory disorders such as hepatitis (Terris et al., 1995). In spite of these clues, the direct evidence whether nuclear HCV core protein exists in PML-NB and functionally implicates in PML-NB functions still awaits to be established.
The fact that many virus oncoproteins bind to p53 and affect its functions indicates that the interference of p53 normal functions is an important step in the virus-mediated oncogenic transformation. While p53 functional inactivation is detected in over half of human tumors, p53 mutations are rare in the early stages of HCC (Hirohashi, 1991; Teramoto et al., 1992). The most likely scenario for the predisposition of HCV-associated HCC is to deregulate the host gene expression and to disturb the tumor suppressive systems. Thus, the interaction between tumor suppressor p53 and HCV core protein is of particular interest. The differential modulation of p53 post-translational modifications and transcription regulatory activity by HCV core protein is surely an important factor in the multistep progression of HCV-associated liver traumas.
Materials and methods
The plasmid p53CON that contains two copies of p53-responsive element upstream of the hsp70 minimal TATA promoter sequences was used as the luciferase reporter plasmid for the detection of p53-dependent transactivation (Chen et al., 1993a). The RNA Pol I-dependent reporter, prHu3-Luc, was constructed by subcloning the Klenow-filled, HindIII/BstEII-digested fragment (0.6 kb) of prHu3-CAT (Zhai et al., 1997) into the SmaI site of pGL2-Basic (Promega Biotech) with a correct orientation. The RNA Pol III-dependent luciferase reporter, pArg-tRNA-Luc, was constructed by cloning the HindIII-digested fragment (0.5 kb) of pArgMaxi (Dingermann et al., 1983) into the HindIII site of pGL2-Basic (Promega Biotech) with a correct orientation. A DNA fragment containing full-length p53 gene was subcloned into pMT at HindIII (filled-in) and EcoRI (filled-in) sites to generate the p53 expression vector pMT-p53. pCEP4-p53, pCEP4-p53CD30, pCEP4-p53CD55, and pCEP4-p53CD75, derivatives of pCEP4 (Invitrogen Biotech), directed the expression of full-length, and C-terminal 30, 55, and 75 amino-acid-deleted variants of p53 in mammalian cells, respectively (Hsu et al., 1995). Plasmids pGST/p53 (1–393 aa), pGST/p53ΔC75 (1–318 aa), pGST/p53ΔN75 (76–393 aa), pGST/p53ΔC150 (1–243 aa), and pGST/p53ΔN160 (161–393 aa) were used for expressing GST-p53 variants in Escherichia coli (Sang et al., 1994). To generate other deletion GST-p53 variants (pGST/p53ΔC30 (1–363 aa), pGST/p53ΔC55 (1–318 aa), pGST/p53-C30 (364–393 aa), pGST/p53-C55 (339–393 aa), and pGST/p53-C75 (319–393 aa)), DNA fragments containing the specific domains of p53 were PCR amplified and cloned in frame into EcoRI and XhoI sites of pGEX-5X-1 plasmids (Amersham Phamarcia Biotech). The productions of low and high level of full-length HCV core protein in mammalian cells were mediated by the expression vector pECE/HCVc-KF and pSRα/HCVc195, respectively (Shih et al., 1993; You et al., 1999). Plasmids pGST/HCVc24, pGST/HCVc50, pGST/HCVc101, pGST/HCVc122, and pGST/HCVc195 used in this study had been described elsewhere (Shih et al., 1995; Chen et al., 1997). The plasmids pGEM-p53 and pET23a/HCVc were used as the DNA templates for generating in vitro synthesized p53 and HCV core protein, respectively (Chen et al., 1993a; You et al., 1999).
Cell culture, transfection, and luciferase assay
Human hepatoma cell lines HuH-7, HepG2, Hep3B, and human cervical carcinoma HeLa cells were cultured at 37°C, 5% CO2 in Delbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS). Human lung carcinoma cell line H1299 was grown in RPMI medium containing 10% FBS. HepG2 cells that constitutively expressing HCV core protein (G2/C) and the parental cells (G2) were kindly provided by Dr LH Hwang (National Taiwan University). For transient transfections, cells seeded onto 10-cm2 or six-well dishes were transfected with appropriate amounts of plasmid DNA by calcium phosphate co-precipitation method (Shih et al., 1993) or by SuperFect transfection reagent (Qiagen Biotech, Hilden, Germany). At 48 h post-transfection, cells were harvested and processed for the luciferase assay (You et al., 1999) and Western blot analysis.
In vitro binding assay
GST fusion HCV core protein and p53 were expressed in E. coli DH5α, and purified using glutathione–Sepharose resins as described by the manufacturer (Amersham Phamarcia Biotech). In vitro synthesized 35S-Met-labeled p53 (pGEM-p53 as the DNA template) and HCV core proteins (pET23a/HCVc) were produced by the TNT kit (Promega Biotech). For each binding reaction, Sepharose-bound GST fusion proteins (1 μg) were incubated with the in vitro translated products in 1 × binding buffer (50 mM Tris-HCl (pH 8.0), 100 M NaCl, 5 mM MgCl2, 20% glycerol) for 4 h or overnight at 4°C. The resins were then extensively washed with washing buffer (1 × binding buffer + 0.5% NP-40) and analysed by SDS–PAGE. Signals were revealed by autoradiography.
Nuclear extracts and cytoplasmic fractions were prepared as described previously (You et al.,. 1999). Briefly, cell pellets were lysed in the hypotonic buffer (20 mM HEPES (pH 7.4), 1 mM MgCl2, 10 mM KCl, 0.5% NP-40, 0.5 mM DTT, and protease inhibitor cocktail (Complete; Boehringer)), and the soluble cytosol fractions were obtained after centrifugation. Nuclear extracts were recovered from the resulting nuclei pellets by resuspension in the high-salt buffer (20 mM HEPES (pH 7.4), 20% glycerol, 0.4 M NaCl. 1 mM MgCl2, 10 mM KCl, 0.5 mM DTT, and protease inhibitor cocktail). As for the preparation of whole-cell lysates, cell pellets were lysed in lysis buffer (1 × PBS, 0.5% NP-40, protease inhibitor cocktail) for 30 min on ice, followed by passing through a 26G needle for 10 times. Supernatant was collected as whole-cell lysates by centrifugation. Cell extracts, either nuclear or whole-cell lysates, were incubated with 20 μl anti-p53 (DO-1) antibody-conjugated agarose beads (Oncogene Biotech) in 1 × binding buffer for 4 h at 4°C. The immunoprecipitated complexes were extensively washed to reduce nonspecific contaminants, and analysed by SDS–PAGE, followed by immunoblotting using enhanced chemiluminescence. Anti-p53 antibodies (DO-1 and PAb421) and anti-phospho p53 (α-P-Ser15, and α-P-Ser392) antibodies were purchased from Oncogene Biotech. Anti α-tubulin (B7) and polyclonal anti-p53 (FL-393) antibodies were purchased from Santa Cruz Biotech, and anti-acetylated p53 antibodies (α-Ac-Lys320 and α-Ac-Lys373, 382) were purchased from Upstate Biotech. Human anti-HCV core protein sera were used for the detection of HCV core protein (Shih et al., 1995).
Nuclear extracts (NE) were prepared as described above, and 20 μg of proteins were used in each reaction. Reactions were performed in a binding buffer that contains 20 mM HEPES (pH 7.5), 50 mM NaCl, 1.5 mM MgCl2, 10% glycerol, 5 mM DTT, 100 μg/ml BSA, and 0.05 μg/μl poly dI-dC (Amersham Phamarcia Biotech). Reaction mixtures were preincubated at room temperature for 10 min before 32P-labeled DNA probes (wild-type p53 consensus sequences: 5′-IndexTermGGACATGCCCGGGCATGTCC-3′; 0.125 pmol) and anti-p53 antibody (PAb 421, Oncogene Biotech; 200 ng) were added, and further incubated at room temperature for 30 min. Competition experiments were carried out by adding 50-fold molar excess of wild-type or mutant (mutant-type p53 consensus sequences: 5′-IndexTermGGAGATTCCCGGGGATTTCC-3′) competitors along with 32P-labeled probes. Reaction mixtures were then loaded onto a native 4% polyacrylamide gel containing 2.5% glycerol and electrophoresed in 0.5 × TBE buffer at 350 V for 1–2 h at 4°C. Radioactive intensity of the shift band was analysed by a PhosphorImager (Molecular Dynamics).
HeLa cells grown on the glass coverslips were transfected with p53 (pCEP4-p53) and HCV core protein expression vectors (pSRα/HCVc195) either alone or in combination. After 24 h, cells were fixed with acetone/methanol solution (1 : 1; −20°C), and then reacted with monoclonal anti-p53 antibody (DO-1; Oncogene Biotech) and human anti-HCV core sera, followed by reacting with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG Ab and rhodamine-conjugated goat anti-human IgG Ab (Jackson ImmnuoResearch Laboratories). Finally, the specimens were observed by laser-scanning confocal microscopy.
Agarwal ML, Taylor WR, Chernov MV, Chernova OB and Stark GR . (1998). J. Biol. Chem., 273, 1–4.
Aoki H, Hayashi J, Moriyama M, Arkawa Y and Hino O . (2000). J. Virol., 74, 1736–1741.
Appella E and Anderson CW . (2000). Pathol. Biol., 48, 227–245.
Appella E and Anderson CW . (2001). Eur. J. Biochem., 268, 2764–2772.
Arima N, Kao CY, Licht T, Padmanabhan R, Sasaguri Y and Padmanabhan R . (2001). J. Biol. Chem., 276, 12675–12684.
Barba G, Harper F, Harada T, Kohara M, Goulinet S, Matsuura Y, Eder G, Schaff Z, Chapman MJ, Miyamura T and Brechot C . (1997). J. Virol., 94, 1200–1205.
Bukh J, Miller RH, Kew MC and Purcell RH . (1993). Proc. Natl. Acad. Sci. USA, 90, 1848–1851.
Cairns CA and White RJ . (1998). EMBO J., 17, 3112–3123.
Chang J, Yang SH, Hwang SB, Hahn YS and Sung YC . (1998). J. Virol., 72, 3060–3065.
Chen CM, You LR, Hwang LH and Lee YHW . (1997). J. Virol., 71, 9417–9426.
Chen JY, Funk WD, Wright WE, Shay JW and Minna JD . (1993a). Oncogene, 8, 2159–2166.
Chen SY, Kao CF, Chen CM, Shih CM, Hsu MJ, Chao CH, Wang SH, You LR and Lee YHW . (2003). J. Biol. Chem., 278, 591–607.
Chen X, Farmer G, Zhu H, Prywes R and Prives C . (1993b). Genes Dev., 7, 1837–1849.
Cho JW, Baek WK, Suh SI, Yang SH, Chang J, Sung YC and Suh MH . (2001). Liver, 21, 137–142.
Dingermann T, Sharp S, Schaack J and Soll D . (1983). J. Biol. Chem., 258, 10395–10402.
Dobner T, Horikoshi N, Rubenwolf S and Shenk T . (1996). Science, 272, 1470–1473.
Dumaz N and Meek DW . (1999). EMBO J., 18, 7002–7010.
El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein B . (1993). Cell, 75, 817–825.
Erhardt A, Hassan M, Heintges T and Haussinger D . (2002). Virology, 292, 272–284.
Filhol O, Baudier J, Delphin C, Loue-Mackenbach P, Chambaz EM and Cochet C . (1992). J. Biol. Chem., 267, 20577–20583.
Fogal V, Gostissa M, Sandy P, Zacchi P, Sternsdorf T, Jensen K, Pandolfi PP, Will H, Schneider C and Del Sal G . (2000). EMBO J., 19, 6185–6195.
Giaccia AJ and Kastan MB . (1998). Genes Dev., 12, 2973–2983.
Gu W and Roeder RG . (1997). Cell, 90, 595–606.
Hirohashi S . (1991). Princess Takamatsu Symp., 22, 87–93.
Hollstein M, Sidransky D, Vogelstein B and Harris CC . (1991). Science, 253, 49–53.
Hsieh TY, Matsumoto M, Chou HC, Schneider R, Hwang SB, Lee AS and Lai MMC . (1998). J. Biol. Chem., 273, 17651–17659.
Hsu YS, Tang FM, Liu WL, Chuang JY, Lai MY and Lin YS . (1995). J. Biol. Chem., 270, 6966–6974.
Ishido S and Hotta H . (1998). FEBS Lett., 438, 258–262.
Jackson P, Ridgway P, Rayner J, Noble J and Braithwaite A . (1994). Biochem. Biophys. Res. Commun., 203, 133–140.
Jamal S and Ziff EB . (1995). Oncogene, 10, 2095–2101.
Jayaraman L, Freulich E and Prives C . (1997). Methods Enzymol., 283, 245–256.
Jayaraman L and Prives C . (1999). Cell. Mol. Life Sci., 55, 76–87.
Jensen K, Shiels C and Freemont PS . (2001). Oncogene, 20, 7223–7233.
Jimenez GS, Khan SH, Stommel JM and Wahl GM . (1999). Oncogene, 18, 7656–7665.
Jin DY, Wang HL, Zhou Y, Chun ACS, Kibler KV, Hou YD, Kung HF and Jeang KT . (2000). EMBO J., 19, 729–740.
Ko LJ and Prives C . (1996). Genes Dev., 10, 1054–1072.
Kwun HJ, Jung EY, Ahn JY, Lee MN and Jang KL . (2001). J. Gen. Virol., 82, 2235–2241.
Lai MMC and Ware CF . (2000). Curr. Top. Microbiol. Immunol., 242, 117–134.
Lain S, Midgley C, Sparks A, Lane EB and Lane DP . (1999). Exp. Cell Res., 248, 457–472.
Lan KH, Sheu ML, Hwang SJ, Yen SH, Chen SY, Wu JC, Wang YJ, Kato N, Omata M, Chang FY and Lee SD . (2002). Oncogene, 21, 4801–4811.
Larminie CG, Alzuherri HM, Cairns CA, McLees A and White RJ . (1998). J. Mol. Med., 76, 96–103.
Levine AJ . (1997). Cell, 88, 323–331.
Liu L, Scolnick DM, Trievel RC, Zhang HB, Marmorstein R, Halazonetis TD and Berger SL . (1999). Mol. Cell. Biol., 19, 1202–1209.
Liu Y, Colosimo AL, Yang XJ and Liao D . (2000). Mol. Cell. Biol., 20, 5540–5553.
Lu H and Levine AJ . (1995). Proc. Natl. Acad. Sci. USA, 92, 5154–5158.
Lu W, Lo SY, Chen M, Wu KJ, Fung YKT and Ou JH . (1999). Virology, 264, 134–141.
Majumder M, Ghosh AK, Steele R, Ray R and Ray RB . (2001). J. Virol., 75, 1401–1407.
Martin DW, Subler MA, Munoz RM, Brown DR, Deb SP and Deb S . (1993). Biochem. Biophys. Res. Commun., 195, 428–434.
Martin ME and Berk AJ . (1998). J. Virol., 72, 3146–3154.
Miline DM, Campbell L, Campbell DG and Meek DW . (1995). J. Biol Chem., 270, 5511–5518.
Moriya K, Fujie H, Shintani Y, Yotsuyanagi H, Tsutsumi T, Ishibashi K, Matsuura Y, Kimura S, Miyamura T, Koike K and Kato N . (1998). Nat. Med., 4, 1065–1067.
Otsuka M, Kato N, Lan KH, Yoshida H, Kato J, Goto T, Shiratori Y and Omata M . (2000). J. Biol. Chem., 275, 34122–34130.
Paule MR and White RJ . (2000). Nucleic Acids Res., 28, 1283–1298.
Prives C and Manley JL . (2001). Cell, 107, 815–818.
Ray RB, Meyer K and Ray R . (2000). Virology, 271, 197–204.
Ray RB and Ray R . (2001). FEMS. Micro. Lett., 202, 149–156.
Ray RB, Steele R, Meyer K and Ray R . (1998). Gene, 208, 331–336.
Saito I, Miyamura T, Ohbayashi A, Harada H, Katayama T, Kikuchi S, Watanabe Y, Koi S, Onji M, Ohta Y, Choo Q, Houghton M and Kuo G . (1990). Proc. Natl. Acad. Sci. USA, 87, 6547–6549.
Sakaguchi K, Herrera JE, Saito S, Miki T, Bustin M, Vassilev A, Anderson CW and Appella E . (1998). Genes Dev., 12, 2831–2841.
Sang BC, Chen JY, Minna J and Barbosa MS . (1994). Oncogene, 9, 853–859.
Santolini E, Migliaccio G and La Monica N . (1994). J. Virol., 68, 3631–3641.
Seeler JS and Dejean A . (1999). Curr. Opin. Genet. Dev., 9, 362–367.
Sheppard HM, Corneillie SI, Espiritu C, Gatti A and Liu X . (1999). Mol. Cell. Biol., 19, 2746–2753.
Shieh SY, Ikeda M, Taya Y and Prives C . (1997). Cell, 91, 325–334.
Shih CM, Chen CM, Chen SY and Lee YHW . (1995). J. Virol., 69, 1160–1171.
Shih CM, Lo SJ, Miyamura T, Chen SY and Lee YHW . (1993). J. Virol., 67, 5823–5832.
Shrivastava A, Manna SK, Ray R and Aggarwal BB . (1998). J. Virol., 72, 9722–9728.
Shivakumar CV, Brown DR, Deb S and Deb SP . (1995). Mol. Cell. Biol., 15, 6785–6793.
Teramoto T, Satonaka K, Kitazawa S, Fujimori T, Hayashi K and Maeda S . (1992). Caner Res., 54, 231–235.
Terris B, Baldin V, Dubois S, Deggot C, Flejou JF, Henin D and Dejean D . (1995). Cancer. Res., 55, 1590–1597.
Truant R, Antunovic J, Greenblatt J, Prives C and Cromlish JA . (1995). J. Virol., 69, 1851–1859.
Unger T, Juven-Gershon T, Moallem E, Berger M, Vogt Sinov R, Lozano G, Oren M and Haupt Y . (1999). EMBO J., 18, 1805–1814.
Vogelstein B, Lane D and Levine AJ . (2000). Nature, 408, 307–310.
Wang F, Yoshida I, Takamatsu M, Fujita T, Oka K and Hotta H . (2000). Biochem. Biophys. Res. Commun., 273, 479–484.
Xiao H, Pearson A, Coulombe B, Truant R, Zhang S, Regier JL, Triezenberg SJ, Reinberg D, Flores O and Ingles CJ . (1994). Mol. Cell. Biol., 14, 7013–7024.
Yasui K, Wakita T, Kyoko TK, Funahashi SI, Ichikawa M, Kajita T, Moradpour D, Wands JR and Kohara M . (1998). J. Virol., 72, 6048–6055.
You LR, Chen CM, Yeh TS, Tsai TY, Mai RT, Lin CH and Lee YHW . (1999). J. Virol., 73, 2841–2853.
Zhai W and Comai L . (2000). Mol. Cell. Biol., 20, 5930–5938.
Zhai W, Tuan JA and Comai L . (1997). Genes Dev., 11, 1605–1617.
Zhong S, Salomoni P and Pandolfi PP . (2000). Nat. Cell Biol., 2, E85–E90.
We thank Dr LH Hwang for generously providing HepG2 cells that constitutively express HCV core protein. We also thank Dr L Comai for prHu3-CAT, Dr DL Johnson for pArgMaxi, and Dr YS Lin for pCEP4-p53 variants. This work was supported by the following grants to YHW Lee: NHRI-EX90-9002BL, NHRI-EX91-9002BL, and NHRI-EX92-9002BL from the National Health Research Institute, and in part by Grants NSC89-2320-B-010-141, NSC90-2320-B-010-083, and NSC91-2320-B-010-072 from the National Science Council.
About this article
Molecular Mechanisms of Hepatocarcinogenesis Following Sustained Virological Response in Patients with Chronic Hepatitis C Virus Infection
International Journal of Molecular Sciences (2018)
Reviews in Medical Virology (2018)
Modeling hepatitis C virus protein and p53 interactions in hepatocytes: Implications for carcinogenesis
Mathematical Biosciences (2018)