DDB2 Induces Nuclear Accumulation of the Hepatitis B Virus X Protein Independently of Binding to DDB1

ABSTRACT The hepatitis B virus (HBV) X protein (HBx) is critical for the life cycle of the virus. HBx associates with several host cell proteins including the DDB1 subunit of the damaged-DNA binding protein DDB. Recent studies on the X protein encoded by the woodchuck hepadnavirus have provided correlative evidence indicating that the interaction with DDB1 is important for establishment of infection by the virus. In addition, the interaction with DDB1 has been implicated in the nuclear localization of HBx. Because the DDB2 subunit of DDB is required for the nuclear accumulation of DDB1, we investigated the role of DDB2 in the nuclear accumulation of HBx. Here we show that expression of DDB2 increases the nuclear levels of HBx. Several C-terminal deletion mutants of DDB2 that fail to bind DDB1 are able to associate with HBx, suggesting that DDB2 may associate with HBx independently of binding to DDB1. We also show that DDB2 enhances the nuclear accumulation of HBx independently of binding to DDB1, since a mutant that does not bind DDB1 is able to enhance the nuclear accumulation of HBx. HBV infection is associated with liver pathogenesis. We show that the nuclear levels of DDB1 and DDB2 are tightly regulated in hepatocytes. Studies with regenerating mouse liver indicate that during late G1 phase the nuclear levels of both subunits of DDB are transiently increased, followed by a sharp decrease in S phase. Taken together, these results suggest that DDB1 and DDB2 would participate in the nuclear functions of HBx effectively only during the late-G1 phase of the cell cycle.

Chronic infection with hepatitis B virus (HBV) is believed to be one of the key risk factors for the development of hepatocellular carcinoma (2,11,26,33,43). Despite the existence of successful vaccination programs, human HBV continues to be a major health problem, affecting around 350 million people worldwide (16). HBV belongs to the hepadnavirus family, which includes rodent viruses, such as woodchuck hepatitis virus (WHV) and ground squirrel hepatitis virus, and the distantly related duck hepatitis virus. The X gene of mammalian HBV encodes a small (17-kDa) multifunctional protein named HBx, which shares no similarity with any other known viral or cellular protein (25). HBx has been shown to affect a number of cellular processes (reviewed in reference 1). Among the best documented functions, transcriptional transactivation of several cellular and viral promoters (reviewed in references 7, 34, and 48) and stimulation of the apoptotic pathway (4,8,19,31,42) have received a wide range of experimental support. HBx has also been reported to inhibit the transactivation function and cause cytoplasmic sequestration of p53 (13,44,46).
Previous studies have shown that HBx binds to the DDB1 subunit (125 kDa) of UV-damaged DNA binding protein (DDB) (3,20,40). DDB has been implicated in DNA repair (10,14,15,17,18,28,30,32,45). Mutational analysis of HBx has demonstrated a partial correlation between the reduction of repair activity in cells expressing HBx and the ability of HBx to bind DDB1 (3). However, with transgenic mice it has been shown that expression of HBx does not affect the frequency of spontaneous mutations (24). A recent study indicated a more significant function of the HBx-DDB1 interaction in the virus life cycle. It was shown that X protein mutants that fail to bind DDB1 are also impaired in stimulating viral replication and productive infection in vivo (38). Furthermore, the interaction of HBx with DDB1 was correlated with the ability of the X protein to localize in the nucleus (39).
Our previous studies have indicated that the 48-kDa DDB2 subunit of DDB plays a critical role in nuclear accumulation of the DDB1 subunit. Naturally occurring mutants of DDB2 failed to enhance nuclear accumulation of DDB1 (36). Because the interaction with DDB1 is important for the nuclear accumulation of HBx, we sought to determine the role of DDB2 in that process. In this report, we show that DDB2 indeed plays a role in the nuclear accumulation of HBx. Interestingly, we observed that DDB2 binds HBx and that the interaction may occur independent of DDB1. Mutational analysis indicated that the WD motif in DDB2 is important for both binding and nuclear accumulation of the HBx protein.
Because HBV is a liver pathogen, we examined whether DDB nuclear expression was altered in a liver injury model. We show that the nuclear levels of DDB1 and DDB2 are tightly regulated in hepatocytes. Studies with regenerating mouse liver indicate that the nuclear levels of both subunits of DDB increase in late-G 1 phase, followed by a decrease in S phase. Our results suggest that DDB participates in the nuclear functions of HBx only during the late-G 1 phase of the cell cycle.
Expression plasmids. Constructs expressing T7-tagged DDB2 C-terminal mutants were generated by PCR, in which the upstream primers contained sequences encoding the T7 epitope in frame with the first ATG of the DDB2 cDNA. The sequence of the upstream primer (p48 upstream primer) has been described before (36). Downstream primers are as follows: for T7p48 (1-380), ACGTCTAGATTACCCTGAGTTTCCATCG; for T7p48 (1-320), GCATCTA GACTACCACTGGGAAGCAGA; for T7p48 (1-260), CGATCTAGATCATG TGGCCAGGAACCAAT; and for T7p48 , CGATCTAGATTAGATG GTGTCTGAGCT. Amplified fragments were cloned as KpnI (5Ј)-XbaI (3Ј) fragments into pCDNA3. The T7p48WD deletion construct was made in two steps. The region between amino acids 1 and 238 was first amplified using the upstream primer with the T7 epitope (p48 upstream primer) described above and the downstream primer CGATCTAGAATTCCAAAGCTCTTTGCCG (p48wd2). Similarly, the region between amino acids 278 and 427 was amplified using the upstream primer GTACTCTAGAGCCAGCTTCCTCTACTCGCT (p48wd3) and the downstream primer ATCGGATCCTCACTTCCGTGTCCTGGCT (p48wd4BamHI). An XbaI site was engineered into the primers designated p48wd2 and p48wd3. Following PCR amplification and digestion with XbaI, the fragments were ligated and a heat-inactivated aliquot of the ligation mix was used to perform PCR using the p48 upstream primer and the p48wd4BamHI primer as described above. The amplified product was cloned as a KpnI (5Ј-BamHI (3Ј) fragment into pCDNA3.
The p125/V5-His construct was made by first amplifying the p125 cDNA by PCR using TTGGTACCACCATGTCGTACAACTACGTG as the upstream primer and GGTCTAGAGATCCGAGTTAGCTCCT as the downstream primer and then cloning it in frame with the V5 epitope in the pcDNA3.1/V5-HisA vector (Invitrogen) at the KpnI and XbaI sites. The plasmid expressing HBx (pCMV-X) was generated in our laboratory as described earlier (36). pCMV-X-FLAG was cloned into pCDNA3 using EcoRI and XbaI sites in frame with a FLAG epitope tag.
Preparation of cytosolic and nuclear extracts. Cytosolic and nuclear extracts were prepared from transfected cells by the method described by Dignam et al. (9). Briefly, the harvested cells were washed with phosphate-buffered saline (PBS) and suspended in 2 volumes of hypotonic buffer, and membranes were disrupted by 30 strokes of a 2-ml Konte tissue grinder. Nuclei were pelleted by centrifugation at 500 ϫ g for 5 min. The supernatant served as the cytosolic fraction, and the nuclear pellet was extracted with high-salt buffer (0.5 M KCl). Extracted nuclear proteins were obtained after centrifuging down the debris at 10,000 ϫ g for 10 min.
Immunoprecipitation and Western blot analysis. Cells were harvested after DNA transfection. Harvested cells were washed twice with PBS, and cell extracts were prepared by incubation in 2 volumes of a lysis buffer containing 10 mM 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS), 2 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM sodium vanadate, 1 mM sodium fluoride, 1 mM dithiothreitol (DTT), 5 g of aprotinin/ml, and 5 g of leupeptin/ml in PBS for 1 h at 4°C. The lysate was centrifuged at 10,000 ϫ g for 10 min, and equal amounts (1 mg) of the extracts were subjected to immunoprecipitation with T7 antibody conjugated to beads (Amersham Pharmacia Biotech). The immunoprecipitates were eluted with gel-loading buffer at room temperature for 10 min. boiled, and subjected to Western blot assays. Western blotting was performed by using anti-rabbit or anti-mouse Fab fragments conjugated to horseradish peroxidase (Amersham) and ECL Western blot detection reagents (Amersham) according to the manufacturer's instructions. Monoclonal V5 antibodies were purchased from Invitrogen, and T7 antibodies were obtained from Novagen. An antibody against cdk2 was from Santa Cruz Biotechnology. Polyclonal peptide antibodies specific for DDB1 and DDB2 were raised in our laboratory as described previously (36).
Immunostaining. U2OS or HepG2 cells were grown in plates containing coverslips and were transfected with plasmids (5 g) expressing FLAG-tagged HBx either alone or in combination with T7-tagged wild-type DDB2 or mutant DDB2 (1-380) or DDB2 (WD⌬). Coverslips containing transfected cells were fixed with methanol, blocked with 5% goat serum in PBS, and probed with fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies against the FLAG-epitope (Sigma Chemical Co.). Finally, coverslips were washed and mounted on glass slides by using 5 l of the Vectashield mounting medium (Vector Laboratories, Burlingame, Calif.).
Immunofluorescence was detected and images were taken using a CLSM 510 microscope (Zeiss) and a 63ϫ Acrophlan water immersion objective. DNA transfections. Transient transfections were carried out by the calcium phosphate coprecipitation method as described previously (36). The total concentration of the DNA for transfection was maintained at 20 g/100-mm-diameter plate by addition of empty vector DNA. HepG2 cells were transfected by using Lipofectamine 2000 (Gibco BRL) according to the manufacturer's instructions.
PHx surgery and immunohistochemical staining. Two-month-old wild-type CD-1 mice were subjected to partial-hepatectomy (PHx) operations to induce liver regeneration as described previously (47). Briefly, a midventral laparotomy was performed on each mouse under anesthesia, and two-thirds of the liver was surgically resected (removal of left lateral, left median, and right median liver lobes). At each of several time points following PHx operation (24,32,36,40, and 44 h), two mice were sacrificed by CO 2 asphyxiation, and regenerating livers were dissected. Regenerating livers were divided into two portions: one was used to isolate nuclear protein extracts; the other was paraffin embedded, and 5-m sections were prepared with a microtome and then used for immunohistochemical staining (47). Paraffin wax was removed from liver sections with xylene, sections were rehydrated with decreasing graded ethanol washes, and microwave retrieval was used to enhance antigenic activity as described previously (47). DDB1 and DDB2 peptide antibodies were diluted 1:50 and used for immunohistochemical detection with the ABC kit and DAB peroxidase substrate purchased from Vector Laboratories.
Preparation of nuclear extracts from regenerating mouse liver. Nuclear extracts were prepared from less than 1 g of regenerating mouse liver using a modified procedure described by Slomiany et al. (41). Regenerating liver was dissected at various time points following PHx, washed with polyamine buffer, and then minced with a razor blade. Polyamine buffer consisted of 10 mM HEPES (pH 7.9), 10 mM KCl, 750 M spermidine, 150 M spermine, 1 mM EDTA, 0.2 mM PMSF, and 1 mM DTT with protease inhibitor cocktail (2 g of each of the protease inhibitors aprotinin, leupeptin, and pepstatin [Sigma]/ml). The minced liver was subjected to Dounce homogenization 20 times in polyamine buffer (polyamine/liver ratio, 2:1 [vol/vol]) and brought to 0.1% NP-40. Following a 10-min incubation, nuclei were collected by centrifugation (at 500 ϫ g for 10 min) in an Eppendorf tube, and the supernatant was decanted. The crude nuclear pellet was resuspended in 600 l of polyamine buffer, and 600 l of polyamine buffer containing 30% sucrose was carefully layered at the bottom of the tube. Nuclei were pelleted through this sucrose pad by centrifugation at 6,000 ϫ g for 10 min (4°C). The supernatant was discarded, and the nuclear pellet was resuspended in 500 l of low-salt buffer. Nuclear proteins were extracted by slow addition of 500 l of high-salt buffer with continuous mixing, and the tube was rocked for 30 min at 4°C. Chromatin was removed by centrifugation at 10,000 ϫ g for 20 min (4°C), and the supernatant nuclear extract was frozen in small aliquots at Ϫ80°C. The low-salt buffer consisted of 10 mM HEPES (pH 7.9), 0.2 mM PMSF, 10 mM KCl, 1.5 mM MgCl 2 , and 1 mM DTT with protease inhibitor cocktail. The high-salt buffer consisted of 20 mM HEPES (pH 7.9), 500 mM KCl, 0.2 mM EDTA, 20% glycerol, 1.5 mM MgCl 2 , 20 g of PMSF/ml, and 1 mM DTT with protease inhibitor cocktail.

Coexpression of DDB2 increases nuclear accumulation of HBx.
Interaction with DDB1 has been implicated in the nuclear localization of HBx (39). Since DDB2 has been shown to facilitate nuclear localization of DDB1 (36), we wanted to investigate whether DDB2 has any role to play in nuclear accumulation of HBx. To accomplish this, we cotransfected human oesteosarcoma (U2OS) cells with plasmids expressing HBx along with T7-tagged DDB2 and DDB1 as well as with T7-epitope-tagged DDB2 alone. U2OS cells can be efficiently transfected by the calcium phosphate method, and we have used transfected U2OS cells successfully to obtain nuclear and cytosolic fractions (36). Under the transfection conditions used in these experiments, transfected cells overexpress the DDB polypeptides approximately 30-fold (data not shown). Transfected cells were lysed by Dounce homogenization in hypotonic buffer; nuclei were pelleted and then treated with highsalt buffer to extract the nuclear proteins (see reference 9). The supernatant obtained after centrifugation of the nuclei was taken as the cytosolic fraction. Cytosolic and nuclear extracts were subjected to Western blot analysis, and blots were probed with anti-HBx antiserum. Western blot analysis of subcellular fractions revealed that upon overexpression of both subunits of DDB, the level of HBx in the nuclear fraction was significantly increased (Fig. 1, upper panel). Similar results were obtained when HBx was coexpressed with DDB2 alone (Fig. 1, lower panels). However, expression of the DDB1 subunit alone had only a marginal effect on nuclear HBx levels (Fig. 1, lower  panels). This is consistent with our previous observation that DDB1, when expressed alone, is found predominantly in the cytoplasm (36). These results suggest that DDB2 assisted in the accumulation of HBx in the nucleus.
DDB2 interacts with HBx independently of binding to DDB1. It has been well established that DDB1 associates with HBx (3,20,40). Because an increase in nuclear accumulation of HBx was observed with the expression of DDB2 alone, we sought to determine whether DDB2 associates with HBx. However, it is possible that HBx associates with DDB2 through an interaction with the endogenous DDB1. Therefore, in order to investigate whether DDB2 associates with HBx independently or through DDB1, we analyzed several mutants of DDB2 for the ability to bind DDB1 and HBx. Several T7tagged deletion mutants of DDB2 (1-380, 1-320, 1-260, and WD⌬) were transiently coexpressed with HBx or DDB1 in U2OS cells. The WD⌬ mutant of DDB2 lacks the WD motif located between residues 238 and 278. Extracts of the transfected cells were subjected to immunoprecipitation with a monoclonal antibody against T7 that was covalently linked to Sepharose beads. Immunoprecipitates were released by washing the beads with sodium dodecyl sulfate-protein gel sample loading buffer and were subjected to Western blot analysis for the presence of HBx or DDB1. T7 epitope antibody was found to coprecipitate HBx from extracts of cells expressing wild-type DDB2 and the C-terminal deletion mutants of DDB2 except for the WD deletion mutant (Fig. 2.). This mutational analysis suggests that the WD motif of DDB2 is essential for HBx association. To study the interaction between these mutant DDB2 proteins and DDB1, the mutants were coexpressed with V5 epitope-tagged DDB1 in U2OS cells. Equal amounts of cell lysates were immunoprecipitated with monoclonal antibodies directed against the T7 epitope. The presence of DDB1 in the immunoprecipitates was detected by analysis of Western blots with anti-V5 monoclonal antibodies (Invitrogen). Signal for V5-tagged DDB1 was specifically detected only in immunoprecipitates obtained from lysates of cells transfected with wildtype DDB2 (Fig. 3), suggesting that the DDB2 mutants examined had lost the ability to interact with DDB1, unlike their native counterpart. Thus, the C-terminal mutants of DDB2, which are unable to bind DDB1, still retain the ability to associate with HBx, providing evidence that HBx interacts with DDB2 independently of DDB1. The immunoprecipitation experiments for which results are shown in Fig. 2 and 3 were carried out under a very stringent condition because HBx is FIG. 1. DDB2 overexpression stimulates nuclear accumulation of HBx. A plasmid expressing HBx (5 g) was transfected into U2OS cells either with empty vector or with a plasmid(s) expressing either T7-tagged DDB1 (5 g), T7-tagged DDB2 (5 g), or both. Crude cytosolic and nuclear extracts were prepared as described in Materials and Methods. Portions (100 g) of the extracts were analyzed by Western blot assays. Blots were probed with polyclonal antibodies against HBx (3) as described in Materials and Methods. cdk2 levels were assayed as a loading control.
FIG. 2. The WD motif in DDB2 is important for binding to HBx. U2OS cells were cotransfected with FLAG epitope-tagged HBx plasmids and either T7-tagged wild-type DDB2 or a T7-tagged deletion mutant of DDB2 (1-380, 1-320, 1-260, or WD⌬). (Top) Total cell extracts of transfected cells were subjected to immunoprecipitation with T7 antibody conjugated to beads, followed by a Western blot assay. The blot was probed with antibodies against HBx (3). Co-IP, coimmunoprecipitate. (Center and bottom) To assay for expression of the HBx and DDB2 proteins, extracts were subjected to Western blot analysis using antibodies against HBx and the T7 epitope, respectively. A mutant of DDB2 that does not bind DDB1 retains the ability to stimulate nuclear accumulation of HBx. In order to further explore the role of DDB2-HBx interaction with regard to nuclear localization of HBx, we performed immunostaining studies using both the oesteosarcoma cell line U2OS (Fig. 4) and the hepatocyte cell line HepG2 (Fig. 5). Cells, grown on coverslips, were transfected with plasmids expressing FLAGtagged HBx either alone or with wild-type or mutant DDB2. The transfected cells were subjected to immunostaining with FITC-labeled monoclonal antibodies against the FLAG epitope. The subcellular localization of HBx in cells expressing FLAGtagged HBx was examined by immunofluorescence confocal microscopy. Cells transfected with HBx alone exhibited specific staining in the cytoplasm (Fig. 4 and 5). However, cells cotransfected with HBx and wild-type DDB2 displayed strong nuclear staining for HBx. Similar fluorescence patterns were observed for cells expressing HBx along with the mutant DDB2 (1-380), which does not bind DDB1 (Fig. 4 and 5). Consistent with the coimmunoprecipitation studies (Fig. 2), the HBx protein fluorescence pattern remained cytosolic upon overexpression of the WD deletion mutant of DDB2, which does not bind HBx. This supports the hypothesis that this DDB2 mutant, which fails to associate with HBx, is also defective in facilitating its nuclear accumulation.
The subunits of DDB accumulate in the nuclei of heptocytes duringthe late G 1 phase of the cell cycle. Infection with HBV results in a persistent liver infection leading to chronic inflammatory liver injury and repair. Therefore, we used regenerating mouse liver to look at the expression of the DDB proteins. Our studies with cell culture indicated that DDB2 is detected in the nucleus mainly during the mid-to late-G 1 phase of the cell cycle (A. Nag, T. Bondar, and P. Raychaudhuri, unpublished data). Mice were subjected to PHx, liver tissue was harvested at different time points following the operation, and expression of the DDB1 and DDB2 proteins was detected during hepatocyte proliferation by immunohistochemical staining using affinitypurified peptide antibodies specific for DDB1 and DDB2 (Fig.  6.). While these antibodies failed to detect any significant nuclear staining for the DDB subunits in resting adult livers ( Fig.  6A and G), nuclear accumulation of the DDB subunits was easily detected in regenerating liver sections (Fig. 6). The highest levels of nuclear DDB1 and DDB2 staining in hepatocytes were found at 28 to 36 h post-PHx, times shown in previous experiments to correspond to the late-G 1 phase of the cell cycle prior to the initiation of S phase (see Fig. 6) (47). DNA replication in regenerating liver cells peaks at 40 h after PHx (47). The signals are specific, as evidenced by the fact that we did not detect any signal with another antibody against DDB1 that works well in Western blots but not in immunostaining experiments (data not shown). To obtain further evidence for the changes in the nuclear levels of the DDB polypeptides, we prepared nuclear extracts from regenerating liver tissues. The nuclear extracts were assayed for DDB1 and DDB2 by using affinity-purified antibodies in Western blot assays (Fig. 7). To control for loading, an appropriate part of the blot was also probed for cdk2. Consistent with the immunohistochemical analysis, the Western blot assay demonstrated a clear increase in the nuclear levels of both DDB1 and DDB2 before the peak of DNA synthesis (40 h) (see reference 47). These results suggest that the nuclear levels of the DDB proteins are tightly regulated in hepatocytes and that they are abundant only during the late-G 1 phase of the cell cycle.

DISCUSSION
Several lines of evidence have indicated that HBx is required for the replication and life cycle of HBV. However, the molecular mechanisms by which HBx supports viral replication FIG. 3. The C-terminal region of DDB2 is essential for binding to DDB1. U2OS cells were transfected with V5-tagged DDB1 expression plasmids and either T7-tagged wild-type DDB2 or a T7-tagged Cterminal deletion mutant of DDB2 (1-380, 1-320, 1-260, 1-200, or WD⌬). (Top) Extracts of the transfected cells were subjected to immunoprecipitation using T7 antibodies. Immunoprecipitates were analyzed by a Western blot assay. The blot was probed with horseradish peroxidase-linked V5 antibody (Invitrogen) to detect coprecipitating DDB1. Co-IP, coimmunoprecipitate. (Center and bottom) Extracts were also tested for expression of DDB1 and DDB2 by probing the blots with V5 and T7 antibodies, respectively. The star indicates a nonspecific band.
FIG. 4. DDB2 overexpression facilitates nuclear accumulation of HBx in U2OS cells. U2OS cells were grown on plates containing coverslips and were transfected with plasmids (5 g) expressing FLAG-tagged HBx either alone or in combination with either T7-tagged wild-type DDB2 or the indicated T7-tagged mutant. Coverslips containing transfected cells were fixed with methanol and probed with FITC-conjugated monoclonal antibodies against the FLAG epitope as described in Materials and Methods. Immunofluorescence was detected, and images were taken, using a CLSM 510 microscope. Panels on the left (L) represent immunofluorescing cells; panels on the right (R) represent their overlap with the respective phase-contrast micrograph. and productive infection have remained unclear. It has been shown that HBx can associate with numerous host cell transcription factors and stimulate transcription from a wide variety of promoters (7,29,34,48). This transcription activation function is very similar to that of the adenovirus 13S E1A gene product, which is essential for a lytic infection (27). E1A is required for efficient expression of the viral genes that are essential for adenovirus replication (27). Interestingly, it has been shown that HBx can partially replace the activity of E1A in the context of adenovirus DNA replication (35). HBx, like E1A, is also proapoptotic (5). It has been proposed that the apoptotic function of HBx might play a role in spreading the infection (40).
Recent studies on HBx have focused on its interaction with the DDB1 subunit of the UV-damaged DNA-binding protein DDB. Mutational studies on the X protein (WHx) encoded by WHV have indicated a correlation between binding to DDB1 and ability to stimulate viral replication (38). Moreover, it has been shown that DDB1 stimulates X-mediated activation of transcription. Also, WHx mutants that are deficient in binding to DDB1 are also impaired in the ability to stimulate transcription. A similar correlation was also observed when the mutants were analyzed in an apoptosis assay (39). These studies suggested a significant role for WHx-DDB1 interaction in viral replication. DDB1 is believed to be a functional partner of DDB2, a 48-kDa protein that is mutated in xeroderma pigmentosum group E (14). DDB2 remains associated with DDB1, and the DDB1-DDB2 complex has a high affinity for UV-damaged DNA; however, a clear role of this complex in DNA repair has yet to be established.
One of the key functions of DDB2 is to stimulate nuclear accumulation of DDB1 (36). DDB1 is found both in the nucleus and in the cytoplasm (36). However, when DDB1 is expressed alone, it is detected mainly in the cytoplasm (36), which is consistent with the fact that DDB1 has no recognizable nuclear localization signal. DDB2, on the other hand, possesses three nuclear localization signals and is predominantly a nuclear protein (23,36). Our previous studies indicated that DDB2 increases the nuclear import of DDB1. Studies by Sitterlin et al. (39) suggested that the interaction between WHx and DDB1 might be important for the nuclear localization of WHx. However, we failed to detect any significant increase in the nuclear level of HBx by coexpressing DDB1 (Fig. 1). We predicted that DDB2 might enhance nuclear accumulation of the DDB1-HBx complex. Coexpression of DDB2 clearly provided evidence for that. Surprisingly, analysis of DDB2 mutants indicated that mutants that do not bind DDB1 were still able to enhance the nuclear accumulation of HBx. Further analyses demonstrated that DDB2 could interact with HBx independently of binding to DDB1. The interaction was dependent on the retention of the WD motif of DDB2; a deletion mutant lacking the WD motif failed to bind HBx or enhance its nuclear localization. The DDB2-mutants used in this study do not associate with DDB1, and therefore these mutants are not expected to reconstitute DDB function, which involves a complex of both subunits. Although the mutational analysis suggests that DDB2 binds HBx independently of binding to DDB1, endogenous DDB2 always remains associated with DDB1. Moreover, overexpression of HBx does not disrupt the interaction between DDB1 and DDB2 (data not shown). Therefore, it is likely that HBx simultaneously binds to both subunits of DDB to form a ternary complex. Based on our previous study (36) and results presented here, we propose that DDB2 plays a key role in the nuclear accumulation of the DDB-HBx complex.
Because HBV is a liver pathogen, we examined nuclear expression of the DDB subunits in mouse livers. The expression studies indicated that the nuclear levels of DDB1 and DDB2 are quite low in resting liver cells. Upon PHx, the remaining liver tissue regenerates and induces hepatocytes to proliferate. Analysis of proliferating liver cells demonstrated increases in the nuclear levels of both DDB1 and DDB2 at late-G 1 phase and several hours before the peak of DNA replication. Moreover, levels of DDB1 and DDB2 fell significantly during the S phase, suggesting that the nuclear levels of DDB1 and DDB2 are tightly regulated in the liver. If DDB1 and DDB2 are indeed physiological partners of HBx in carrying out its nuclear functions (such as transcription), we propose that the nuclear functions of HBx will be limited by the regulated expression of DDB1 and DDB2. However, it is also possible that HBx stimulates expression of DDB1 and DDB2 in the host cell. HBx has been shown to activate the ras-raf pathway (1, 3a). The ras-raf pathway is also involved in the proliferation response after PHx (6), which leads to accumulation of DDB1 and DDB2 in the nucleus. Therefore, we speculate that HBx, by stimulating the ras-raf pathway, increases the levels of DDB1 and DDB2, which in turn assist HBx in performing its transcription activation function. Clearly, further work will be necessary to investigate whether DDB1 and DDB2 are indeed induced by the ras-raf pathway.
It has been shown that in HBV-infected cells, the majority of HBx localizes in the cytoplasm (1, 3a, 37). A recent study indicated that HBx possesses a nuclear export signal and that in the presence of leptomycin B, an inhibitor of Crm1-dependent nuclear export, HBx accumulates in the nucleus (12). The presence of the nuclear export signal in HBx, coupled with the observation that DDB2 is expressed only during late-G 1 phase, would explain the predominantly cytoplasmic localization of HBx. The DDB2-mediated accumulation of HBx during late-G 1 phase may be critical for viral replication, since the host cell synthesizes the replication proteins and enzymes during that phase of the cell cycle. It is also possible that HBx, by associating with DDB, modifies its function. For example, DDB1 has been shown to associate with the V protein encoded by the paramyxovirus simian virus 5, and that interaction correlates with a delay in cell cycle progression (21,22). DDB has FIG. 5. DDB2 overexpression enhances nuclear accumulation of HBx in HepG2 cells. HepG2 cells were grown on plates containing coverslips and were transfected with plasmids (5 g) expressing FLAG-tagged HBx either alone or in combination with either T7-tagged wild-type DDB2 or the indicated T7-tagged mutant. Cells on coverslips were fixed and probed with FITC-conjugated monoclonal antibodies against the FLAG epitope. Immunofluorescence was detected, and images were taken, using a CLSM 510 microscope. Panels on the left (L) represent immunofluorescing cells; panels on the right (R) represent their overlap with the respective phase-contrast micrograph.
FIG. 6. Hepatocytes exhibit elevated expression of the DDB subunits in nuclei following liver injury. Two-month-old wild-type CD-1 mice were subjected to a two-thirds hepatectomy. Regenerating livers were harvested at the indicated times following PHx, and portions of liver tissue were fixed in 4% paraformaldehyde, paraffin embedded, sectioned with a microtome, and used for immunohistochemical staining with affinity-purified peptide antibodies specific for DDB1 and DDB2 as described in Materials and Methods. been implicated in global genomic repair (14,15,45) as well as in transcription of cell cycle genes (Nag et al., unpublished). Further studies will be important in determining how DDB2mediated nuclear accumulation influences the replication cycle of HBV and the transcriptional activity of the HBx protein. FIG. 7. Changes in levels of DDB proteins in regenerating mouse liver nuclear extracts. After PHx, mice were sacrificed at the indicated time points, and regenerating livers were harvested. Livers were washed with polyamine buffer and minced with a razor blade, and nuclear extracts were prepared as described in Materials and Methods. Five-hundred-microgram portions of the nuclear extracts were assayed for DDB1, DDB2, and cdk2 using the respective antibodies as described in Materials and Methods.