Hepatocyte-Like Cells Transdifferentiated from a P
http://www.100md.com
病菌学杂志 2005年第20期
Department of Pathology and Department of Microbiology & Immunology, WHO Collaborating Center for Tropical Diseases and Center for Biodefense and Emerging Infectious Diseases, University of Texas Medical Branch, Galveston, Texas 77555-0609
Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, United Kingdom
ABSTRACT
Recently, a rat pancreatic cell line (AR42J-B13) was shown to transdifferentiate to hepatocyte-like cells upon induction with dexamethasone (Dex). The aim of this study is to determine whether transdifferentiated hepatocytes can indeed function like bona fide liver cells and support replication of hepatotropic hepatitis B virus (HBV). We stably transfected AR42J-B13 cells with HBV DNA and examined the expression of hepatocyte markers and viral activities in control and transdifferentiated cells. A full spectrum of HBV replicative intermediates, including covalently closed circular DNA (cccDNA) and Dane particles, were detected only after induction with Dex and oncostatin M. Strikingly, the small envelope protein and RNA of HBV were increased by 40- to 100-fold upon induction. When HBV RNAs were examined by primer extension analysis, novel core- and precore-specific transcripts were induced by Dex which initiated at nucleotide (nt) 1820 and nt 1789, respectively. Most surprisingly, another species of core-specific RNA, which initiates at nt 1825, is always present at almost equal intensity before and after Dex treatment, a result consistent with Northern blot analysis. The fact that HBV core protein is dramatically produced only after transdifferentiation suggests the possibility of both transcriptional and translational regulation of HBV core antigen in HBV-transfected AR42J-B13 cells. Upon withdrawal of Dex, HBV replication and gene expression decreased rapidly—less than 50% of the cccDNA remained detectable in 1.5 days. Our studies demonstrate that the transdifferentiated AR42J-B13 cells can function like bona fide hepatocytes. This system offers a new opportunity for basic research of virus-host interactions and pancreatic transdifferentiation.
INTRODUCTION
Transdifferentiation is the conversion of one differentiated cell type to another (60). For example, the appearance of hepatic foci in the pancreas was demonstrated in animal models following the administration of a methionine-deficient diet in combination with a carcinogen, treatment with the peroxisomal proliferator ciprofibrate, or a copper-deficient diet (43, 44, 46). It has also been shown that transplanted pancreatic cells are able to repopulate the liver parenchymal structure (12) and are sufficient to rescue mice deficient in fumarylacetoacetate hydrolase (64). These reports suggest that pancreatic cells have the potential to convert into hepatocyte-like cells during chronic injury and regeneration, a phenomenon reflecting the close developmental relationship between these two tissues (16, 69).
The AR42J cell line is a pancreatic tumor cell line derived from azaserine-treated rats (31, 36). It has been shown recently that AR42J cells and the subclone AR42J-B13 can be transdifferentiated into hepatocyte-like cells upon dexamethasone (Dex) treatment (48). Oncostatin M (OSM), in the presence of Dex, can enhance the hepatic transdifferentiation. The transdifferentiated hepatocytes exhibited a hepatic phenotype as judged by expression of several liver-specific markers, synthesis of acute phase proteins upon cytokine treatment (27, 59) and presence of phase I detoxification enzyme activity (35).
It is intriguing that most of the currently available hepatoma cell lines are not replication permissive for hepatitis B virus (HBV). For example, among seven different rat hepatoma cell lines available at the American Type Culture Collection, only McA-RH7777 (and its clonal derivative Q7 cells) can support wild-type HBV replication (49). Similarly, among a total of nine different human hepatoma cell lines, only the hepatoma cell line Huh7 (5) and the hepatoblastoma cell line HepG2 (55) can support wild-type HBV replication (54). Cross-species replication of HBV in the rodent system has also been reported (21, 22, 49, 50). Here, we determined if transdifferentiated rat hepatocytes can function like bona fide liver cells to support HBV replication. Our studies show unequivocally that HBV can indeed replicate in transdifferentiated hepatocytes. The establishment of an inducible transdifferentiation system will enable further investigation of the mechanism underlying virus-host interactions at the transcriptional and posttranscriptional levels. The potential clinical significance of the transdifferentiation system between two developmentally closely related organs should not be overlooked.
MATERIALS AND METHODS
Standard procedures for Southern, Northern, and Western blot analyses were as described previously (56).
Cell culture and production of HBV-producing stable transfectants. AR42J-B13 cells (48) were maintained in Dulbecco's modified Eagle's medium (low glucose [1 g/liter]; GIBCO BRL, Rockville, MD) containing penicillin, streptomycin, and 10% fetal bovine serum (Sigma, St. Louis, MO) at 37°C in an atmosphere of 5% CO2-95% air. Dexamethasone (Sigma Co.) was prepared as a stock solution (1 mM) in ethanol and was added into the medium every 2 to 3 days. OSM (R&D Systems, Inc., Minneapolis, MN) was added as a solution in phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin, at a final concentration of 10 ng/ml, together with 1 μM Dex. Stable transfectants of AR42J-B13 cells were generated using a tandem dimer of HBV DNA (ayw subtype) in a pSV2Neo vector (15, 49). A tandem dimer, instead of a 1.2-mer, was chosen here because, to date, only the viral particles produced from a tandem dimer configuration of HBV DNA have been proven to be infectious in chimpanzees (1, 50). Before transfection, AR42J-B13 cells were treated with or without 1 μM Dex and 10 ng/ml OSM (Dex plus OSM) for 7 days. Two micrograms of plasmid DNA was transfected using the FuGENE 6 transfection protocol (Roche Co.). Stable transfectants were selected in medium containing 1 mg/ml G418 (Life Technologies Co.). After 4 to 7 weeks, clones were picked, expanded, and maintained initially in medium containing G418 (1 mg/ml). For the past 18 months, B13-1 and B13-28 have been maintained in the absence of G418.
Preparation of cccDNA. For isolation of covalently closed circular DNA (cccDNA), cells were lysed with 0.5% Nonidet P-40, and nuclei were collected by low-speed centrifugation (5,000 rpm for 5 min). The preparation of cccDNA followed closely a modified alkali lysis procedure for the isolation of plasmid DNA (14). The cccDNA-containing samples were diluted in an equal volume of 0.1 N NaOH and incubated at 4°C for 10 min to irreversibly denature noncovalently closed, double-stranded DNA species. The DNA was neutralized by adding 3 M potassium acetate (pH 5.2) to a final concentration of 0.6 M. Single-strand DNA was efficiently removed by phenol extraction. The double-strand cccDNA remaining in the aqueous phase was recovered by ethanol precipitation. The cccDNA was analyzed using a 1% agarose gel in 40 mM Tris-acetate-1 mM EDTA buffer.
Primer extension analysis. The primer extension protocol was modified as described previously (29). Briefly, a 5'-end-labeled oligonucleotide (1930AS, 5'-GAGAGTAACTCCACAGTAGCTCC-3') was annealed at 65°C for 10 min with total cytoplasmic RNA (25 to 30 μg) in a buffer containing 43% formamide, 1 mM EDTA, 40 mM piperazine-N,N'-bis(2-ethanesulfonic acid) (pH 7.8), and 400 mM NaCl. This mixture was then cooled down to room temperature, and 20 μl of 3 M NaOAc, 150 μl diethyl pyrocarbonate-treated water, and 400 μl of ethanol were added for precipitation. After centrifugation, the pellet was dissolved in 12 μl diethyl pyrocarbonate-treated water and reverse transcription was performed at 42°C for 1.5 h with 20 U of Moloney murine leukemia virus reverse transcriptase (RT; New England Biolabs, Beverly, MA), 2 μl of RT buffer, 1 μl of 10 mM deoxynucleoside triphosphate mixture, 2 μl of 0.1 M dithiothreitol, 10 U of RNasin, and 1 μg of actinomycin D. The reaction was terminated by adding 1 μl of 0.5 M EDTA, pH 7.8, and 1 μl of RNase A for 30 min at 37°C. A 200-μl volume of NTE (10 mM Tris-HCl, 1 mM EDTA, 0.1 M NaCl) was added to the reaction mixture, followed by phenol-chloroform extraction and ethanol precipitation. After centrifugation, the pellet was dissolved in 9 μl of Tris-EDTA, and 6 μl of sequencing loading buffer was added and run on a 6% polyacrylamide sequencing gel.
Immunofluorescence analysis and antisera. For immunofluorescent staining, cells were cultured on glass coverslips, rinsed with PBS twice, fixed with 4% paraformaldehyde in PBS for 30 min at room temperature, and then permeabilized with 0.1% (vol/vol) Triton X-100 in PBS for 30 min and incubated in 2% blocking buffer (Roche, Indianapolis, IN) for 1 h. The cells were then incubated sequentially with primary and secondary antibodies (Table 1). After immunostaining, coverslips were mounted on slides in gelvatol medium containing 4'-6-diamidino-2-phenylindole (DAPI; 500 ng/ml in PBS). Images were collected using a Zeiss confocal microscope (LSM 510) and processed with Adobe Photoshop software.
Electron microscopy of core particles. Standard negative staining was used to prepare capsid particles for electron microscopy. The procedure of negative staining of capsid particles was as described previously (49).
HBeAg and HBsAg assays. Enzyme-linked immunosorbent assays (ELISAs) for HBV e antigen (HBeAg) and HBV surface antigen (HBsAg) were performed according to the vendor's protocols (International Immuno-Diagnostics Co., Foster City, CA).
RESULTS
Detection of HBsAg and HBeAg in the medium of B13-1 and B13-28 cells by ELISA. To investigate whether the transdifferentiated AR42J-B13 cells can function like bona fide hepatocytes, we attempted to perform transient transfection using the HBV tandem dimer genome in AR42J-B13 cells. For reasons unclear, no significant secretion of HBeAg was detected in the medium by ELISA (data not shown). Stable transfection was thus initiated using cells with or without prior Dex and OSM treatment. Neomycin-resistant colonies were selected in the presence of Dex and OSM using G418 (49). Among a total of 19 neomycin-resistant clones, clone B13-3 is the only one that did not secrete any detectable HBeAg or HBV DNA synthesis, while clone B13-10 is the only clone which exhibited detectable HBeAg and HBV DNA synthesis before induction (data not shown). The remaining 17 clones shared similar properties in HBeAg secretion and HBV DNA synthesis (data not shown). Here, we chose to focus our characterizations on clones B13-1 and B13-28. Both clones are representative of the other 15 clones in terms of their tight inducibility to secrete HBeAg and HBsAg and to initiate HBV DNA synthesis after Dex plus OSM treatment.
Briefly, media obtained from B13-1 and B13-28 cultures were collected on days 3, 5, and 7 after induction with Dex plus OSM and were subjected to ELISAs for both HBsAg and HBeAg (Fig. 1) (56). Total cell number counts from each dish were used for normalization of the ELISA readings. As shown in Fig. 1A, we noted that the HBeAg secretion after the induction in both B13-1 and B13-28 cells was increased by approximately eightfold on day 3 and four- to fivefold on day 7. To our surprise, the HBsAg titer was increased by 40- to 100-fold upon induction in both B13-1 and B13-28 cells (Fig. 1B). The twofold increase of secreted HBsAg of B13-1 cells from day 3 to day 7 probably reflected a doubling of pancreatic cells converting to hepatocyte-like cells (Fig. 1B).
Coexpression of HBcAg and hepatic glutamine synthetase. We examined whether the intracellular expression of HBV core antigen (HBcAg) and the liver marker glutamine synthetase can be detected in transdifferentiated B13-1 (Fig. 2A) and B13-28 (Fig. 2B) cells by immunofluorescent staining. As expected, untreated B13-1 and B13-28 cells did not express either HBcAg or glutamine synthetase (Fig. 2A, panels e and f, and B, panels k and l). In contrast, after induction, cells appeared to heterogeneously express HBcAg (green) and glutamine synthetase (red). Some cells coexpressed both HBcAg and glutamine synthetase, as shown by the yellow color in overlaid images (Fig. 2A, panel c, and B, panel i). Yet, there were cells expressing glutamine synthetase only (remained red in overlaid images) or HBcAg only (remained green in overlaid images). It remains to be investigated whether the heterogeneity of expression patterns observed here is physiologically relevant to the differential expression of glutamine synthetase in vivo (e.g., perivenous versus periportal hepatocytes) (59).
Immunofluorescent staining of HBsAg and a liver marker. Similarly, we observed the heterogeneity of expression patterns of HBsAg and another liver marker, transferrin, in transdifferentiated B13-1 and B13-28 cells. HBsAg (green) and transferrin (red) could be detected in B13-1 (Fig. 3A) and B13-28 (Fig. 3B) cells on day 5 or day 7 postinduction. There was always a substantial fraction of cells that would express transferrin at various levels. Positive transferrin staining was observed, and HBsAg stain was all over the cells. However, most HBsAg was expressed only in transferrin-expressing cells, though the level of transferrin varied. During the period between day 5 and 7, cell size expansion occurred, as is evident by comparing the same scale bar (20 μm) used in all pictures (Fig. 3). As in parental AR42JB13 cells (27, 48), uninduced B13-1 and B13-28 cells tended to be smaller in size, while the transdifferentiated hepatocytes tended to be enlarged and more flattened (Fig. 3).
Coexpression of liver-enriched transcription factors and HBsAg. Liver-specific gene expression is mainly regulated by a number of liver-enriched transcription factors (11, 30). As shown in Fig. 4, we examined the expression of HBsAg or HBcAg (in red) and four different liver-enriched transcription factors (in green), including C/EBP-, C/EBP-?, HNF 4, and HNF 3?. The nuclei were counterstained in blue with DAPI. Before induction with Dex plus OSM, there was no expression of C/EBP, C/EBP?, and nuclear HNF 4 in either B13-1 (data not shown) or B13-28 (Fig. 4a, g, and m) cells. In contrast to C/EBP, C/EBP?, and nuclear HNF 4, the expression of HNF 3? was observed prior to induction (Fig. 4s). After treatment with Dex plus OSM, most of B13-1 and B13-28 cells expressed C/EBP (Fig. 4f), C/EBP? (Fig. 4l), HNF 3? (Fig. 4x), and HNF 4 in the nucleus (Fig. 4r). The expression of HBsAg correlates best with HNF 4 (Fig. 4p, q, and r). The constitutive expression pattern of HNF 3? serves as a good control for the inducible expression of the other three transcription factors in B13-1 and B13-28 cells.
Western blot analysis of HBcAg, liver, and pancreatic markers. To confirm the results from immunofluorescent staining (Fig. 2 to 4), we performed Western blot analysis on B13-1 and B13-28 cells before and after conversion, using antibodies specific for HBcAg, 1-antitrypsin, -amylase, and -tubulin (Fig. 5). As predicted, HBcAg and 1-antitrypsin were induced upon Dex plus OSM treatment, and -tubulin was unaffected by the treatment. Intriguingly, unlike the parental AR42J-B13 cells (data not shown), we were unable to detect any -amylase in B13-1 and B13-28 cells without induction. However, when the parental AR42J-B13 cells were examined, we were able to detect -amylase before induction (data not shown). Therefore, B13-1 and B13-28 cells appeared to be different from their parental AR42J-B13 cells in their basal expression of -amylase.
Analysis of HBV-specific RNAs before and after transdifferentiation. Upregulation of HBsAg (Fig. 1 and 3) and HBcAg (Fig. 2 and 5) could result from the transcriptional stimulation of Dex plus OSM on these two viral genes. To examine this possibility, we carried out Northern blot analysis on B13-1 and B13-28 cells before and after Dex plus OSM treatment. As shown in Fig. 6A, the most dramatic effect induced by Dex plus OSM was observed on the 2.1-kb pre-S2/S RNA. In contrast, no significant effect was observed on the 2.3-kb pre-S1 RNA, and a mild change was observed in the 3.5-kb RNA, which consists of the pregenomic RNA (pgRNA; core-specific RNA) and the precore RNA. These two RNA species are structurally related and can be distinguished from each other at their 5' ends. To resolve these two closely related RNA species, we performed RNA primer extension analysis using the total intracellular RNA prepared from both B13-1 and B13-28 cells with or without Dex plus OSM. As shown in Fig. 6B, precore RNA initiates at nucleotide (nt) 1785 in both B13-1 and B13-28 cells before induction. After induction, a new species of precore RNA initiates at nt 1789, which is absent in untreated cells. These results are consistent with the ELISA data for HBeAg (Fig. 1).
Most surprisingly, there is only one species of core-specific RNA before induction which initiates at nt 1825. After induction, there are two core-specific RNA species which initiate at nt 1820 and nt 1825, respectively. The existence of core-specific RNA before induction is therefore confirmed by both primer extension and Northern blot analyses.
Southern blot analysis of HBV DNA replication before and after induction. Although B13-1 and B13-28 cells are capable of expressing HBV RNA and protein upon induction (Fig. 1 to 6), it remained unclear whether HBV can indeed replicate in this system. To address this issue, we conducted Southern blot analysis of viral DNAs prepared from B13-1 and B13-28 cells before and after hepatic conversion. HBV DNA isolated from Qs21 was included again as a positive control. As shown in Fig. 7A, characteristic replication patterns of HBV DNA were observed, including the full-length single-strand (SS) and relaxed circular (RC) replicative intermediates on days 3, 5, and 7 postinduction. We noted that the full-length RC form in this system tends to be more abundant than other HBV-producing systems, such as Qs21 (49) (lighter exposure data not shown).
HBV replicates via an RNA intermediate. The pgRNA is transcribed from the cccDNA template in the nucleus (63). To determine if cccDNA can be detected in B13-1 and B13-28 cells after induction, we analyzed the cccDNA by Southern blot analysis. As shown in Fig. 7B, an HBV-specific signal could be detected at a position expected for cccDNA from both Dex plus OSM-treated B13-1 and B13-28 cells. Restriction enzyme EcoRI is known to cut once in the HBV genome of the ayw subtype (49). When the extracted cccDNA preparation was digested with EcoRI before loading, this putative cccDNA band upshifted to a 3.2-kb position on the 1.2% agarose gel, which is where a linearized full-length HBV genome would band (Fig. 7B, lane 3).
To see if intracellular HBV capsids can be secreted into the medium of B13-1 and B13-28 systems, we analyzed the virions by density gradient centrifugation (data not shown). Fractions corresponding to the expected density of HBV virions (around 1.24 g/cm3) were collected and dialyzed to remove CsCl (56). Virion-associated DNA was extracted and analyzed by Southern blotting. As expected from wild-type HBV, the mature genome is preferentially exported (Fig. 7C).
Electron microscopic examination of secreted HBV viral and subviral particles in the medium of transdifferentiated B13-28 cells. The results in Fig. 7C were also confirmed by transmission electron microscopy. As shown in Fig. 8, in addition to tubular and spherical subviral particles, we can detect Dane-like particles with an electron-dense core, concentric shape, and approximately 42 nm in diameter. This result suggests that the hepatocyte-like cells of transdifferentiated B13-1 and B13-28 can indeed support virion assembly and secretion of mature genome similar to an authentic hepatocyte.
Continuous presence of Dex plus OSM is required for HBV replication and gene expression in B13-1 and B13-28 cells. When we removed Dex and OSM from the medium, the levels of secreted HBsAg and HBeAg (Fig. 9A) were downregulated. Moreover, RC and SS DNA also decreased rapidly (Fig. 9B). Similarly, cccDNA was reduced to less than 50% on day 2 and almost undetectable by day 5 after withdrawal (Fig. 9C). Consistent with the ELISA results for HBsAg and HBeAg in Fig. 9A, intracellular HBsAg (p24 and gp27) and HBcAg decreased significantly by Western blot analysis (Fig. 9D). The decrease of viral replication and gene expression was also paralleled by the decrease of 1-antitrypsin and -amylase, as measured by Western blot analysis (Fig. 9D). Again, -tubulin remained unaffected 7 days post-Dex withdrawal. Similar to -tubulin, HNF 3? remained constant after withdrawal (Fig. 10D). In contrast, transcription factors C/EBP , C/EBP ?, and HNF 4 disappeared rapidly and were barely detectable on day 3 after withdrawal (Fig. 10A to C).
DISCUSSION
Transdifferentiation from pancreatic cells to hepatocyte-like cells was first discovered more than two decades ago (44, 46). We demonstrate here for the first time that transdifferentiated hepatocyte-like cells can fully support the fastidious HBV gene expression, replication, cccDNA formation, assembly, and secretion of virions and subviral particles. In all assays, the effect of Dex treatment alone is similar to, yet much slower than, Dex plus OSM treatment (data not shown). Based on these results, B13 cells can be induced to undergo transdifferentiation to bona fide hepatocytes.
Advantages and limitations of transdifferentiated hepatocytes. Although the original main purpose of this study was to provide the most rigorous proof of the authenticity of the transdifferentiated hepatocytes, a useful side-product of this project is the established stable HBV-producing system. It is perhaps worth comparing the B13-1 and B13-28 cells with the most commonly used HBV-producing systems of human Huh7 or HepG2 (5, 47, 55, 61, 68) or rodent (22, 49, 50) cell lines. A tetracycline-inducible HBV-producing system has also been reported (28). One advantage of the B13-1 and B13-28 cells resides in their inducible and synchronized HBV replication upon treatment with physiological ligands (Dex and OSM). Such an inducible feature provides an advantage to synchronize the cell population and thus improve the detection sensitivity in experimental assays for the host hepatocytes. Another advantage of our current transdifferentiated hepatocyte system which is absent in the human HepG2 and Huh7 system is the possibility to extend our future research to an immunocompetent rodent model in vivo. One major limitation of B13-1 and B13-28 cells is that they are unlikely to be infectible with HBV due to the known species barrier of infection between rodents and humans.
Direct or indirect effect of glucocorticoid treatment on HBV replication and HBsAg expression. HBV contains five promoters and two enhancers (20). The pre-S1 promoter, which is responsible for the 2.3-kb viral RNA, is not responsive to Dex plus OSM treatment (Fig. 6A). In contrast, as shown by the results of Northern blot analysis (Fig. 6A), the 40- to 100-fold increased production of HBsAg protein in B13-1 and B13-28 cells by Dex plus OSM treatment (Fig. 1 and 9D) is mainly caused by the specific increase of the 2.1-kb pre-S2/S RNA. This stimulatory effect of Dex plus OSM on HBsAg transcription could be due to the presence of a glucocorticoid response element in the HBV genome. Alternatively, it could indirectly result from increased levels of transcription factors which are involved in HBsAg transcription (37). Although steroid-free chemotherapy decreased the risk of HBV reactivation in HBV carriers with lymphoma (7), this in vivo observation by itself does not argue one way or the other if the Dex effect in vitro is direct or indirect.
A direct effect of Dex on HBV or simian virus 540 (SV40) enhancer/promoter elements appeared unlikely because (i) we observed no apparent effect on the production of HBsAg when HBV-transfected HepG2, Huh7, or Q7 cells were treated with Dex (data not shown); (ii) there is no well-matched glucocorticoid response element consensus motif (5'-AGAACANNNTGTTCT-3') on the HBV or SV40 genome (data not shown); (iii) only a limited effect of Dex on HBV expression in the transient-transfection assay has been reported in the literature (9, 62). Instead of a simple direct effect, we are in favor of the possibility that the Dex effect on HBV could be predominantly indirect and pleiotropic. When AR42J-B13 cells were treated with Dex, cells were reprogrammed toward hepatic phenotypes (48). Indeed, Dex has been reported to have a potent effect on the phenotypic maturation of hepatocytes via growth arrest (38). Many host factors are known to be important for liver development (11, 16, 30, 69). Dex is known to induce expression of HNF 4 and C/EBP-, which are essential factors for hepatocyte differentiation (Fig. 4) (37). In summary, the effect of Dex plus OSM on HBV replication and gene expression in B13-1 and B13-28 cells is most likely indirect, which is not mutually exclusive with the remote possibility of a direct effect.
Is the expression of HBsAg liver specific? While the expression of HBsAg is generally not believed to be liver specific (15, 51, 52), there are reports that HBsAg mRNA or protein is preferentially expressed in more differentiated hepatocyte cell lines (6, 17) or in the liver in animal models (see further discussions below) (3, 4, 8). Interestingly, consistent with our current findings in vitro (Fig. 1, 3, and 6), Dex has been shown to enhance the expression of HBsAg in HBV transgenic mice (19). Ubiquitous transcription factors SP1 and NF-Y are known to bind and transactivate the pre-S2/S promoter (34, 42). Although a liver-specific transcription factor HNF 3 can also bind to the pre-S2/S promoter, it functions as a negative regulator for HBsAg production (57). It is also interesting that HNF 3? is present before or after induction (Fig. 4D and 10D). Therefore, the presence of HNF 3? does not appear to interfere with HBV replication in the presence of Dex plus OSM. Furthermore, C/EBP- and HNF 4 are not known to have any effect on the pre-S2/S promoter, despite their known effects on the precore/core promoters and HBV replication (24, 32, 33, 41, 58, 70). If the liver specificity of the dramatic HBsAg expression were determined by liver-specific enhancers I and/or II of HBV, then it would be hard to reconcile with the modest increase of HBeAg mRNA and protein in transdifferentiated hepatocytes (Fig. 1 and 6). The molecular mechanisms of the liver-specific expression of HBsAg in transdifferentiated hepatocytes in vitro and transgenic mice in vivo merit further investigation.
Extrahepatic expression of hepadnaviruses. As discussed above, the pre-S2/S promoter is more active in well-differentiated hepatocytes. However, extrahepatic gene expression of HBV has also been commonly observed in animal models. For example, HBV gene expression in the kidney has been observed in transgenic mice by at least three independent groups (3, 8, 18). Similarly, viral antigen expression of duck HBV was found in both islets and acini of the pancreas in early embryos and young ducks (25). Indeed, the most common extrahepatic tissues for hepadnaviral gene expression in ducks and woodchucks include spleen, kidney, and pancreas (26, 39). It is possible that some transcription factors are common in these different organs. AR42J-B13 cells were derived from AR42J cells, which in turn originated from a rat pancreatic acinar tumor (31, 36). However, in our B13-1 and B13-28 cells, we are unable to detect HBV DNA replication or viral antigen expression without Dex plus OSM treatment.
A positive transcriptional control and a negative translational control for the gene expression of HBcAg. In the absence of Dex plus OSM, HBV core protein was not detectable by Western blotting (Fig. 5 and 9D) or immunofluorescence assays (Fig. 2), but in the presence of Dex plus OSM, core protein was strongly induced. The absence of detectable core protein in the untreated B13-1 and B13-28 cells does not result from the absence of core-specific mRNA before induction. In fact, core/precore-specific RNAs can be detected in both B13-1 and B13-28 cells before induction by Northern blot analysis (Fig. 6A). To resolve core-specific RNA from the structurally related precore-specific RNA, we performed primer extension analysis (Fig. 6B). We detected novel core-specific transcript in Dex-treated cells, which initiate at nt 1820. Interestingly, another species of core-specific RNA, which initiates at nt 1825, is preexisting before induction and persists at a constant level after induction (Fig. 6B). A similar phenomenon was observed for the precore-specific RNAs. The Dex-inducible species of precore-specific RNA initiates at nt 1789, while the preexisting precore RNA species initiates at nt 1785. In the literature, core-specific RNA initiates around nt 1820 ± 2 (45, 67, 68), and precore-specific RNA initiates around nt 1785 ± 3 (20).
The precore protein is known to be a precursor to HBeAg. The preexisting precore RNA must therefore be responsible for the HBeAg detectable by ELISA before induction (Fig. 1A). What appears to be most intriguing is the preexisting core-specific RNA which initiates at nt 1825, 5 nt downstream from the putative 5' end of the Dex-inducible core-specific RNA. Apparently, this preexisting core-specific RNA is not engaged in the translation of core protein before induction. Since the AUG codon for translational initiation of core protein is at nt 1903, the preexisting core RNA species has a sufficient length of 5'-untranslated region for ribosomal scanning. By inference, the Dex-inducible core-specific RNA is most likely to be responsible for the synthesis of core protein after induction. At present, we entertain the hypotheses that the preexisting core RNA is negatively regulated at the translational level in B13-1 and B13-28 cells before induction (Fig. 11). This phenomenon is somewhat reminiscent of a so-called uncoupled phenotype of bipotential liver cell lines (53). Despite the existence of hepatocyte transcription factors (HNF 1, GATA4, and HNF 4), no hepatocyte functions (albumin or apolipoprotein) were detected.
As for the Dex-induced core RNA, the simplest interpretation is that it is caused by a Dex-induced de novo transcriptional initiation at nt 1820. The fact that the 5' end of the Dex-inducible core RNA is only 5 nt away from the 5' end of the preexisting core RNA raised the question of whether this new species of core RNA could be under the control of posttranscriptional processing. For example, if the 5' end of the preexisting core RNA is somehow blocked in untreated cells, such blockage for translational initiation could also interfere with the progression of reverse transcriptase to the very end of its mRNA template during the in vitro primer extension reaction. If we further assume that the Dex-induced transdifferentiation occurs only at a maximal efficiency near 50%, then the other 50% of the uninduced population of B13-1 or B13-28 cells should still contain only the preexisting core RNA. In this scenario, the core RNA profile of posttranscriptional regulation would be similar to that of a simple transcriptional control. However, judging from the relative intensity between Dex-induced versus preexisting core RNA species in transdifferentiated hepatocytes, a mechanism of posttranscriptional processing appears unlikely. In summary, while the preexisting core RNA appears to be under negative translational control before induction, it remains to be investigated if the Dex-induced core RNA is subject to transcriptional or posttranscriptional control. In either case, it remains to be resolved if it is a direct or indirect effect of Dex plus OSM on the synthesis or processing of core-specific RNA.
Half-lives of HBV RC, SS, and cccDNAs. Because cccDNA in the nucleus is the template for the synthesis of pgRNA, it plays a critical role in the life cycle of HBV (63). It is generally believed that cccDNA is primarily responsible for persistent infection of HBV in the liver (65). The withdrawal of Dex plus OSM from the culture medium of B13-28 cells allowed us to assess the rate of disappearance of HBV DNA replicative intermediates in transdifferentiated hepatocytes. In some systems, the half-lives of cccDNA of hepadnaviruses are very long, ranging from 33 to 70 days or so (2, 13, 71). As quantitated from the signals in Fig. 9C, the cccDNA level from each dish on day 2 after Dex withdrawal was reduced to approximately 45% of the level of day zero, suggesting that the apparent half-life of cccDNA in this system is approximately 36 h. Surprisingly, unlike the cccDNA, the RC and SS DNAs were still detectable on day 7 (Fig. 9B). The actual half-life of cccDNA in B13-28 cells could even be shorter if we assume that RC and SS DNAs can still mature into cccDNA after Dex withdrawal. The shorter half-life of cccDNA (per dish) in our system is more comparable to the reported half-lives of 1 to 5 days in other systems (10, 14, 23, 40). It has been reported that elimination of cccDNA in vivo probably involves both noncytolytic and cytolytic immune-mediated mechanisms (66). In our reversed transdifferentiation system, withdrawal of Dex plus OSM alone, without any drug, cytokine treatment, or cell death, rapidly eliminated cccDNA. Absence of Dex plus OSM relieved the host hepatocytes from growth arrest and thus probably converted the quiescent hepatocytes back to proliferative pancreatic cells.
ACKNOWLEDGMENTS
We thank colleagues in the Shih lab for careful reading of the manuscript and Eugene Knutson and Tom Albrecht at the Optical Imaging Center of UTMB for the help in confocal laser scanning microscopy.
This work was funded by NIH grants R01 CA 70336 and CA 84217 to C.S. and by the Medical Research Council and the Biotechnology and Biological Sciences Research Council to D.T.
REFERENCES
Acs, G., M. S. Sells, R. H. Purcell, P. Price, R. Engle, M. Shapiro, and H. Popper. 1987. Hepatitis B virus produced by transfected Hep G2 cells causes hepatitis in chimpanzees. Proc. Natl. Acad. Sci. USA 84:4641-4644.
Addison, W. R., K. A. Walters, W. W. S. Wong, J. S. Wilson, D. Madej, L. D. Jewell, and D. L. J. Tyrrell. 2002. Half-life of the duck hepatitis B virus covalently closed circular DNA pool in vivo following inhibition of viral replication. J. Virol. 76:6356-6363.
Araki, K., J. Miyazaki, O. Hino, N. Tomita, O. Chisaka, K. Matsubara, and K. Yamamura. 1989. Expression and replication of hepatitis B virus genome in transgenic mice Proc. Natl. Acad. Sci. USA 86:207-211.
Babinet, C., H. Farza, D. Morello, M. Hadchouel, and C. Pourcel. 1985. Specific expression of hepatitis B surface antigen (HBsAg) in transgenic mice. Science 230:1160-1163.
Chang, C. M., K. S. Jeng, C. P. Hu, S. J. Lo, T. S. Su, L. P. Ting, C. K. Chou, S. H. Han, E. Pfaff, and J. Salfeld. 1987. Production of hepatitis B virus in vitro by transient expression of cloned HBV DNA in a hepatoma cell line. EMBO J. 6:675-680.
Chang, H. K., and L. P. Ting. 1989. The surface gene promoter of the human hepatitis B virus displays a preference for differentiated hepatocytes. Virology 170:176-183.
Cheng, A. L., C. A. Hsiung, I. J. Su, P. J. Chen, M. C. Chang, C. J. Tsao, W. Y. Kao, W. C. Uen, C. H. Hsu, H. F. Tien, T. Y. Chao, L. T. Chen, and J. Whang-Peng. 2003. Steroid-free chemotherapy decreases risk of hepatitis B virus (HBV) reactivation in HBV-carriers with lymphoma. Hepatology 37:1320-1328.
Choo, K. B., L. N. Liew, K. Y. Chong, R. H. Lu, and W. T. Cheng. 1991. Transgenome transcription and replication in the liver and extrahepatic tissues of a human hepatitis B virus transgenic mouse. Virology 182:785-792.
Chou, C. K., L. H. Wang, H. M. Lin, and C. W. Chi. 1992. Glucocorticoid stimulates hepatitis B viral gene expression in cultured human hepatoma cells. Hepatology 16:13-18.
Civitico, G. M., and S. A. Locarnini. 1994. The half-life duck hepatitis B virus supercoiled DNA in congenitally infected primary hepatocyte cultures. Virology 203:81-89.
Costa, R. H., V. V. Kalinichenko, A. X. Holterman, and X. Wang. 2003. Transcription factors in liver development, differentiation, and regeneration. Hepatology 38:1331-1347.
Dabeva, M. D., S. G. Hwang, S. R. Vasa, E. Hurston, P. M. Novikoff, D. C. Hixson, S. Gupta, and D. A. Shafritz. 1997. Differentiation of pancreatic epithelial progenitor cells into hepatocytes following transplantation into rat liver. Proc. Natl. Acad. Sci. USA 94:7356-7361.
Dandri, M., M. R. Burda, H. Will, and J. Petersen. 2000. Increased hepatocyte turnover and inhibition of woodchuck hepatitis B virus replication by adefovir in vitro do not lead to reduction of the closed circular DNA. Hepatology 32:139-146.
Delaney, W. E., IV, T. G. Miller, and H. C. Isom. 1999. Use of the hepatitis B virus recombinant baculovirus-HepG2 system to study the effects of (–)-beta-2',3'-dideoxy-3'-thiacytidine on replication of hepatitis B virus and accumulation of covalently closed circular DNA. Antimicrob. Agents Chemother. 43:2017-2026.
Dubois, M. F., C. Pourcel, S. Rousset, C. Chany, and P. Tiollais. 1980. Excretion of hepatitis B surface antigen particles from mouse cells transformed with cloned viral DNA. Proc. Natl. Acad. Sci. USA 77:4549-4553.
Duncan, S. A. 2003. Mechanisms controlling early development of the liver. Mech. Dev. 120:19-33.
Faktor, O., T. De-Medina, and Y. Shaul. 1988. Regulation of hepatitis B virus S gene promoter in transfected cell lines. Virology 162:362-368.
Farza, H., M. Hadchouel, J. Scotto, P. Tiollais, C. Babinet, and C. Pourcel. 1988. Replication and gene expression of hepatitis B virus in a transgenic mouse that contains the complete viral genome. J. Virol. 62:4144-4152.
Farza, H., A. M. Salmon, M. Hadchouel, J. L. Moreau, C. Babinet, P. Tiollais, and C. Pourcel. 1987. Hepatitis B surface antigen gene expression is regulated by sex steroids and glucocorticoids in transgenic mice. Proc. Natl. Acad. Sci. USA 84:1187-1191.
Ganem, D., and R. Schneider. 2001. Hepadnaviridae: the viruses and their replication, p. 2923-2970. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA.
Guidotti, L. G., B. Matzke, H. Schaller, and F. V. Chisari. 1995. High-level hepatitis B virus replication in transgenic mice. J. Virol. 69:6158-6169.
Guilhot, S., T. Miller, G. Cornman, and H. C. Isom. 1996. Apoptosis induced by tumor necrosis factor-alpha in rat hepatocyte cell lines expressing hepatitis B virus. Am. J. Pathol. 148:801-814.
Guo, J. T., M. Pryce, X. Wang, I. M. Barrasa, J. Hu, and C. Seeger. 2003. Conditional replication of duck hepatitis B virus in hepatoma cells. J. Virol. 77:1885-1893.
Guo, W., M. Chen, T. S. Yen, and J. H. Ou. 1993. Hepatocyte-specific expression of the hepatitis B virus core promoter depends on both positive and negative regulation. Mol. Cell. Biol. 13:443-448.
Halpern, M. S., S. B. McMahon, W. S. Mason, and A. P. O'Connell. 1986. Viral antigen expression in the pancreas of DHBV-infected embryos and young ducks. Virology 150:276-282.
Hosoda, K., M. Omata, K. Uchiumi, F. Imazeki, O. Yokosuka, Y. Ito, K. Okuda, and M. Ohto. 1990. Extrahepatic replication of duck hepatitis B virus: more than expected. Hepatology 11:44-48.
Kurash, J. K., C. N. Shen, and D. Tosh. 2004. Induction and regulation of acute phase proteins in transdifferentiated hepatocytes. Exp. Cell. Res. 292:342-358.
Ladner, S. K., M. J. Otto, C. S. Barker, K. Zaifert, G. H. Wang, J. T. Guo, C. Seeger, and R. W. King. 1997. Inducible expression of human hepatitis B virus (HBV) in stably transfected hepatoblastoma cells: a novel system for screening potential inhibitors of HBV replication. Antimicrob. Agents Chemother. 41:1715-1720.
Le Pogam, S., P. K. Chua, M. Newman, and C. Shih. 2005. Exposure of RNA templates and encapsidation of spliced viral RNA are influenced by the arginine-rich domain of human hepatitis B virus core antigen (HBcAg 165-173). J. Virol. 79:1871-1887.
Locker, J. 2001. Tissue-specific regulation by transcription factors, p. 237-262. In J. Locker (ed.), Transcription factors, 1st ed. BIOS Scientific Publishers Ltd., Oxford, England.
Longnecker, D. S., H. S. Lilja, J. French, E. Kuhlmann, and W. Noll. 1979. Transplantation of azaserine-induced carcinomas of pancreas in rats. Cancer Lett. 7:197-202.
Lopez-Cabrera, M., J. Letovsky, K. Q. Hu, and A. Siddiqui. 1990. Multiple liver-specific factors bind to the hepatitis B virus core/pregenomic promoter: trans-activation and repression by CCAAT/enhancer binding protein. Proc. Natl. Acad. Sci. USA 87:5069-5073.
Lopez-Cabrera, M., J. Letovsky, K. Q. Hu, and A. Siddiqui. 1991. Transcriptional factor C/EBP binds to and transactivates the enhancer element II of the hepatitis B virus. Virology 183:825-829.
Lu, C. C., M. Chen, J. H. Ou, and T. S. Yen. 1995. Key role of a CCAAT element in regulating hepatitis B virus surface protein expression. Virology 206:1155-1158.
Marek, C. J., G. A. Cameron, L. J. Elrick, G. M. Hawksworth, and M. C. Wright. 2003. Generation of hepatocytes expressing functional cytochromes P450 from a pancreatic progenitor line in vitro. Biochem. J. 370:763-769.
Mashima, H., H. Shibata, T. Mine, and I. Kojima. 1996. Formation of insulin-producing cells from pancreatic acinar AR42J cells by hepatocyte growth factor. Endocrinology 137:3969-3976.
Michalopoulos, G. K., W. C. Bowen, K. Mule, and J. Luo. 2003. HGF-, EGF-, and dexamethasone-induced gene expression patterns during formation of tissue in hepatic organoid cultures. Gene Express. 11:55-75.
Michalopoulos, G. K., W. C. Bowen, K. Mule, and D. B. Stolz. 2001. Histological organization in hepatocyte organoid cultures. Am. J. Pathol. 159:1877-1887.
Ogston, C. W., E. M. Schechter, C. A. Humes, and M. B. Pranikoff. 1989. Extrahepatic replication of woodchuck hepatitis virus in chronic infection. Virology 169:9-14.
Peek, S. F., P. J. Cote, J. R. Jacob, I. A. Toshkov, W. E. Hornbuckle, B. H. Baldwin, F. V. Wells, C. K. Chu, J. L. Gerin, B. C. Tennant, and B. E. Korba. 2001. Antiviral activity of clevudine [L-FMAU, (1-(2-fluoro-5-methyl-beta, L-arabinofuranosyl) uracil)] against woodchuck hepatitis virus replication and gene expression in chronically infected woodchucks (Marmota monax). Hepatology 33:254-266.
Pei, D., and C. Shih. 1991. An "attenuator domain" is sandwiched by two distinct transactivation domains in the transcription factor C/EBP. Mol. Cell. Biol. 11:1480-1487.
Raney, A. K., H. B. Le, and A. McLachlan. 1992. Regulation of transcription from the hepatitis B virus major surface antigen promoter by the Sp1 transcription factor. J. Virol. 66:6912-6921.
Rao, M. S., V. Subbarao, and J. K. Reddy. 1986. Induction of hepatocytes in the pancreas of copper-depleted rats following copper repletion. Cell. Differ. 18:109-117.
Reddy, J. K., M. S. Rao, S. A. Qureshi, M. K. Reddy, D. G. Scarpelli, and N. D. Lalwani. 1984. Induction and origin of hepatocytes in rat pancreas. J. Cell Biol. 98:2082-2090.
Roychoudhury, S., A. F. Faruqi, and C. Shih. 1991. Pregenomic RNA encapsidation analysis of eleven missense and nonsense polymerase mutants of human hepatitis B virus. J. Virol. 65:3617-3624.
Scarpelli, D. G., and M. S. Rao. 1981. Differentiation of regenerating pancreatic cells into hepatocyte-like cells. Proc. Natl. Acad. Sci. USA 78:2577-2581.
Sells, M. A., M. L. Chen, and G. Acs. 1987. Production of hepatitis B virus particles in Hep G2 cells transfected with cloned hepatitis B virus DNA. Proc. Natl. Acad. Sci. USA 84:1005-1009.
Shen, C. N., J. M. Slack, and D. Tosh. 2000. Molecular basis of transdifferentiation of pancreas to liver. Nat. Cell Biol. 2:879-887.
Shih, C., L. S. Li, S. Roychoudhury, and M. H. Ho. 1989. In vitro propagation of human hepatitis B virus in a rat hepatoma cell line. Proc. Natl. Acad. Sci. USA 86:6323-6327.
Shih, C., M. Y. Yu, L. S. Li, and J. W. Shih. 1990. Hepatitis B virus propagated in a rat hepatoma cell line is infectious in a primate model. Virology 179:871-873.
Standring, D. N., W. J. Rutter, H. E. Varmus, and D. Ganem. 1984. Transcription of the hepatitis B surface antigen gene in cultured murine cells initiates within the presurface region. J. Virol. 50:563-571.
Stenlund, A., D. Lamy, J. Moreno-Lopez, H. Ahola, U. Pettersson, and P. Tiollais. 1983. Secretion of the hepatitis B virus surface antigen from mouse cells using an extra-chromosomal eucaryotic vector. EMBO J. 2:669-673.
Strick-Marchand, H., and M. C. Weiss. 2002. Inducible differentiation and morphogenesis of bipotential liver cell lines from wild-type mouse embryos. Hepatology 36:794-804.
Suk, F. M., M. H. Lin, M. Newman, S. Pan, S. H. Chen, J. D. Liu, and C. Shih. 2002. Replication advantage and host factor-independent phenotypes attributable to a common naturally occurring capsid mutation (I97L) in human hepatitis B virus. J. Virol. 76:12069-12077.
Sureau, C., J. L. Romet-Lemonne, J. I. Mullins, and M. Essex. 1986. Production of hepatitis B virus by a differentiated human hepatoma cell line after transfection with cloned circular HBV DNA. Cell 47:37-47.
Tai, P. C., F. M. Suk, W. H. Gerlich, A. R. Neurath, and C. Shih. 2002. Hypermodification and immune escape of an internally deleted middle-envelope (M) protein of frequent and predominant hepatitis B virus variants. Virology 292:44-58.
Tang, H., and A. McLachlan. 2002. Mechanisms of inhibition of nuclear hormone receptor-dependent hepatitis B virus replication by hepatocyte nuclear factor 3?. J. Virol. 76:8572-8581.
Tang, H., and A. McLachlan. 2001. Transcriptional regulation of hepatitis B virus by nuclear hormone receptors is a critical determinant of viral tropism. Proc. Natl. Acad. Sci. USA 98:1841-1846.
Tosh, D., C. N. Shen, and J. M. Slack. 2002. Differentiated properties of hepatocytes induced from pancreatic cells. Hepatology 36:534-543.
Tosh, D., and J. M. W. Slack. 2003. How cells change their phenotype. Nat. Rev. Mol. Cell Biol. 3:187-194.
Tsurimoto, T., A. Fujiyama, and K. Matsubara. 1987. Stable expression and replication of hepatitis B virus genome in an integrated state in a human hepatoma cell line transfected with the cloned viral DNA. Proc. Natl. Acad. Sci. USA 84:444-448.
Tur-Kaspa, R., R. D. Burk, Y. Shaul, and D. A. Shafritz. 1986. Hepatitis B virus DNA contains a glucocorticoid-responsive element. Proc. Natl. Acad. Sci. USA 83:1627-1631.
Tuttleman, J. S., C. Pourcel, and J. Summers. 1986. Formation of the pool of covalently closed circular viral DNA in hepadnavirus-infected cells. Cell 47:451-460.
Wang, X., M. Al-Dhalimy, E. Lagasse, M. Finegold, and M. Grompe. 2001. Liver repopulation and correction of metabolic liver disease by transplanted adult mouse pancreatic cells. Am. J. Pathol. 158:571-579.
Werle-Lapostolle, B., S. Bowden, S. Locarnini, K. Wursthorn, J. Petersen, G. Lau, C. Trepo, P. Marcellin, Z. Goodman, W. E. Delaney, IV, S. Xiong, C. L. Brosgart, S. S. Chen, C. S. Gibbs, and F. Zoulim. 2004. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 126:1750-1758.
Wieland, S. F., H. C. Spangenberg, R. Thimme, R. H. Purcell, and F. V. Chisari. 2004. Expansion and contraction of the hepatitis B virus transcriptional template in infected chimpanzees. Proc. Natl. Acad. Sci. USA 101:2129-2134.
Will, H., W. Reiser, T. Weimer, E. Pfaff, M. Buscher, R. Sprengel, R. Cattaneo, and H. Schaller. 1987. Replication strategy of human hepatitis B virus. J. Virol. 61:904-911.
Yaginuma, K., Y. Shirakata, M. Kobayashi, and K. Koike. 1987. Hepatitis B virus (HBV) particles are produced in a cell culture system by transient expression of transfected HBV DNA. Proc. Natl. Acad. Sci. USA 84:2678-2682.
Zaret, K. S. 2001. Hepatocyte differentiation: from the endoderm and beyond. Curr. Opin. Genet. Dev. 11:568-574.
Zheng, Y., J. Li, and J. H. Ou. 2004. Regulation of hepatitis B virus core promoter by transcription factors HNF1 and HNF4 and the viral X protein. J. Virol. 78:6908-6914.
Zhu, Y., T. Yamamoto, J. Cullen, J. Saputelli, C. E. Aldrich, D. S. Miller, S. Litwin, P. A. Furman, A. R. Jilbert, and W. S. Mason. 2001. Kinetics of hepadnavirus loss from the liver during inhibition of viral DNA synthesis. J. Virol. 75:311-322.(Robert Yung-Liang Wang, C)
Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, United Kingdom
ABSTRACT
Recently, a rat pancreatic cell line (AR42J-B13) was shown to transdifferentiate to hepatocyte-like cells upon induction with dexamethasone (Dex). The aim of this study is to determine whether transdifferentiated hepatocytes can indeed function like bona fide liver cells and support replication of hepatotropic hepatitis B virus (HBV). We stably transfected AR42J-B13 cells with HBV DNA and examined the expression of hepatocyte markers and viral activities in control and transdifferentiated cells. A full spectrum of HBV replicative intermediates, including covalently closed circular DNA (cccDNA) and Dane particles, were detected only after induction with Dex and oncostatin M. Strikingly, the small envelope protein and RNA of HBV were increased by 40- to 100-fold upon induction. When HBV RNAs were examined by primer extension analysis, novel core- and precore-specific transcripts were induced by Dex which initiated at nucleotide (nt) 1820 and nt 1789, respectively. Most surprisingly, another species of core-specific RNA, which initiates at nt 1825, is always present at almost equal intensity before and after Dex treatment, a result consistent with Northern blot analysis. The fact that HBV core protein is dramatically produced only after transdifferentiation suggests the possibility of both transcriptional and translational regulation of HBV core antigen in HBV-transfected AR42J-B13 cells. Upon withdrawal of Dex, HBV replication and gene expression decreased rapidly—less than 50% of the cccDNA remained detectable in 1.5 days. Our studies demonstrate that the transdifferentiated AR42J-B13 cells can function like bona fide hepatocytes. This system offers a new opportunity for basic research of virus-host interactions and pancreatic transdifferentiation.
INTRODUCTION
Transdifferentiation is the conversion of one differentiated cell type to another (60). For example, the appearance of hepatic foci in the pancreas was demonstrated in animal models following the administration of a methionine-deficient diet in combination with a carcinogen, treatment with the peroxisomal proliferator ciprofibrate, or a copper-deficient diet (43, 44, 46). It has also been shown that transplanted pancreatic cells are able to repopulate the liver parenchymal structure (12) and are sufficient to rescue mice deficient in fumarylacetoacetate hydrolase (64). These reports suggest that pancreatic cells have the potential to convert into hepatocyte-like cells during chronic injury and regeneration, a phenomenon reflecting the close developmental relationship between these two tissues (16, 69).
The AR42J cell line is a pancreatic tumor cell line derived from azaserine-treated rats (31, 36). It has been shown recently that AR42J cells and the subclone AR42J-B13 can be transdifferentiated into hepatocyte-like cells upon dexamethasone (Dex) treatment (48). Oncostatin M (OSM), in the presence of Dex, can enhance the hepatic transdifferentiation. The transdifferentiated hepatocytes exhibited a hepatic phenotype as judged by expression of several liver-specific markers, synthesis of acute phase proteins upon cytokine treatment (27, 59) and presence of phase I detoxification enzyme activity (35).
It is intriguing that most of the currently available hepatoma cell lines are not replication permissive for hepatitis B virus (HBV). For example, among seven different rat hepatoma cell lines available at the American Type Culture Collection, only McA-RH7777 (and its clonal derivative Q7 cells) can support wild-type HBV replication (49). Similarly, among a total of nine different human hepatoma cell lines, only the hepatoma cell line Huh7 (5) and the hepatoblastoma cell line HepG2 (55) can support wild-type HBV replication (54). Cross-species replication of HBV in the rodent system has also been reported (21, 22, 49, 50). Here, we determined if transdifferentiated rat hepatocytes can function like bona fide liver cells to support HBV replication. Our studies show unequivocally that HBV can indeed replicate in transdifferentiated hepatocytes. The establishment of an inducible transdifferentiation system will enable further investigation of the mechanism underlying virus-host interactions at the transcriptional and posttranscriptional levels. The potential clinical significance of the transdifferentiation system between two developmentally closely related organs should not be overlooked.
MATERIALS AND METHODS
Standard procedures for Southern, Northern, and Western blot analyses were as described previously (56).
Cell culture and production of HBV-producing stable transfectants. AR42J-B13 cells (48) were maintained in Dulbecco's modified Eagle's medium (low glucose [1 g/liter]; GIBCO BRL, Rockville, MD) containing penicillin, streptomycin, and 10% fetal bovine serum (Sigma, St. Louis, MO) at 37°C in an atmosphere of 5% CO2-95% air. Dexamethasone (Sigma Co.) was prepared as a stock solution (1 mM) in ethanol and was added into the medium every 2 to 3 days. OSM (R&D Systems, Inc., Minneapolis, MN) was added as a solution in phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin, at a final concentration of 10 ng/ml, together with 1 μM Dex. Stable transfectants of AR42J-B13 cells were generated using a tandem dimer of HBV DNA (ayw subtype) in a pSV2Neo vector (15, 49). A tandem dimer, instead of a 1.2-mer, was chosen here because, to date, only the viral particles produced from a tandem dimer configuration of HBV DNA have been proven to be infectious in chimpanzees (1, 50). Before transfection, AR42J-B13 cells were treated with or without 1 μM Dex and 10 ng/ml OSM (Dex plus OSM) for 7 days. Two micrograms of plasmid DNA was transfected using the FuGENE 6 transfection protocol (Roche Co.). Stable transfectants were selected in medium containing 1 mg/ml G418 (Life Technologies Co.). After 4 to 7 weeks, clones were picked, expanded, and maintained initially in medium containing G418 (1 mg/ml). For the past 18 months, B13-1 and B13-28 have been maintained in the absence of G418.
Preparation of cccDNA. For isolation of covalently closed circular DNA (cccDNA), cells were lysed with 0.5% Nonidet P-40, and nuclei were collected by low-speed centrifugation (5,000 rpm for 5 min). The preparation of cccDNA followed closely a modified alkali lysis procedure for the isolation of plasmid DNA (14). The cccDNA-containing samples were diluted in an equal volume of 0.1 N NaOH and incubated at 4°C for 10 min to irreversibly denature noncovalently closed, double-stranded DNA species. The DNA was neutralized by adding 3 M potassium acetate (pH 5.2) to a final concentration of 0.6 M. Single-strand DNA was efficiently removed by phenol extraction. The double-strand cccDNA remaining in the aqueous phase was recovered by ethanol precipitation. The cccDNA was analyzed using a 1% agarose gel in 40 mM Tris-acetate-1 mM EDTA buffer.
Primer extension analysis. The primer extension protocol was modified as described previously (29). Briefly, a 5'-end-labeled oligonucleotide (1930AS, 5'-GAGAGTAACTCCACAGTAGCTCC-3') was annealed at 65°C for 10 min with total cytoplasmic RNA (25 to 30 μg) in a buffer containing 43% formamide, 1 mM EDTA, 40 mM piperazine-N,N'-bis(2-ethanesulfonic acid) (pH 7.8), and 400 mM NaCl. This mixture was then cooled down to room temperature, and 20 μl of 3 M NaOAc, 150 μl diethyl pyrocarbonate-treated water, and 400 μl of ethanol were added for precipitation. After centrifugation, the pellet was dissolved in 12 μl diethyl pyrocarbonate-treated water and reverse transcription was performed at 42°C for 1.5 h with 20 U of Moloney murine leukemia virus reverse transcriptase (RT; New England Biolabs, Beverly, MA), 2 μl of RT buffer, 1 μl of 10 mM deoxynucleoside triphosphate mixture, 2 μl of 0.1 M dithiothreitol, 10 U of RNasin, and 1 μg of actinomycin D. The reaction was terminated by adding 1 μl of 0.5 M EDTA, pH 7.8, and 1 μl of RNase A for 30 min at 37°C. A 200-μl volume of NTE (10 mM Tris-HCl, 1 mM EDTA, 0.1 M NaCl) was added to the reaction mixture, followed by phenol-chloroform extraction and ethanol precipitation. After centrifugation, the pellet was dissolved in 9 μl of Tris-EDTA, and 6 μl of sequencing loading buffer was added and run on a 6% polyacrylamide sequencing gel.
Immunofluorescence analysis and antisera. For immunofluorescent staining, cells were cultured on glass coverslips, rinsed with PBS twice, fixed with 4% paraformaldehyde in PBS for 30 min at room temperature, and then permeabilized with 0.1% (vol/vol) Triton X-100 in PBS for 30 min and incubated in 2% blocking buffer (Roche, Indianapolis, IN) for 1 h. The cells were then incubated sequentially with primary and secondary antibodies (Table 1). After immunostaining, coverslips were mounted on slides in gelvatol medium containing 4'-6-diamidino-2-phenylindole (DAPI; 500 ng/ml in PBS). Images were collected using a Zeiss confocal microscope (LSM 510) and processed with Adobe Photoshop software.
Electron microscopy of core particles. Standard negative staining was used to prepare capsid particles for electron microscopy. The procedure of negative staining of capsid particles was as described previously (49).
HBeAg and HBsAg assays. Enzyme-linked immunosorbent assays (ELISAs) for HBV e antigen (HBeAg) and HBV surface antigen (HBsAg) were performed according to the vendor's protocols (International Immuno-Diagnostics Co., Foster City, CA).
RESULTS
Detection of HBsAg and HBeAg in the medium of B13-1 and B13-28 cells by ELISA. To investigate whether the transdifferentiated AR42J-B13 cells can function like bona fide hepatocytes, we attempted to perform transient transfection using the HBV tandem dimer genome in AR42J-B13 cells. For reasons unclear, no significant secretion of HBeAg was detected in the medium by ELISA (data not shown). Stable transfection was thus initiated using cells with or without prior Dex and OSM treatment. Neomycin-resistant colonies were selected in the presence of Dex and OSM using G418 (49). Among a total of 19 neomycin-resistant clones, clone B13-3 is the only one that did not secrete any detectable HBeAg or HBV DNA synthesis, while clone B13-10 is the only clone which exhibited detectable HBeAg and HBV DNA synthesis before induction (data not shown). The remaining 17 clones shared similar properties in HBeAg secretion and HBV DNA synthesis (data not shown). Here, we chose to focus our characterizations on clones B13-1 and B13-28. Both clones are representative of the other 15 clones in terms of their tight inducibility to secrete HBeAg and HBsAg and to initiate HBV DNA synthesis after Dex plus OSM treatment.
Briefly, media obtained from B13-1 and B13-28 cultures were collected on days 3, 5, and 7 after induction with Dex plus OSM and were subjected to ELISAs for both HBsAg and HBeAg (Fig. 1) (56). Total cell number counts from each dish were used for normalization of the ELISA readings. As shown in Fig. 1A, we noted that the HBeAg secretion after the induction in both B13-1 and B13-28 cells was increased by approximately eightfold on day 3 and four- to fivefold on day 7. To our surprise, the HBsAg titer was increased by 40- to 100-fold upon induction in both B13-1 and B13-28 cells (Fig. 1B). The twofold increase of secreted HBsAg of B13-1 cells from day 3 to day 7 probably reflected a doubling of pancreatic cells converting to hepatocyte-like cells (Fig. 1B).
Coexpression of HBcAg and hepatic glutamine synthetase. We examined whether the intracellular expression of HBV core antigen (HBcAg) and the liver marker glutamine synthetase can be detected in transdifferentiated B13-1 (Fig. 2A) and B13-28 (Fig. 2B) cells by immunofluorescent staining. As expected, untreated B13-1 and B13-28 cells did not express either HBcAg or glutamine synthetase (Fig. 2A, panels e and f, and B, panels k and l). In contrast, after induction, cells appeared to heterogeneously express HBcAg (green) and glutamine synthetase (red). Some cells coexpressed both HBcAg and glutamine synthetase, as shown by the yellow color in overlaid images (Fig. 2A, panel c, and B, panel i). Yet, there were cells expressing glutamine synthetase only (remained red in overlaid images) or HBcAg only (remained green in overlaid images). It remains to be investigated whether the heterogeneity of expression patterns observed here is physiologically relevant to the differential expression of glutamine synthetase in vivo (e.g., perivenous versus periportal hepatocytes) (59).
Immunofluorescent staining of HBsAg and a liver marker. Similarly, we observed the heterogeneity of expression patterns of HBsAg and another liver marker, transferrin, in transdifferentiated B13-1 and B13-28 cells. HBsAg (green) and transferrin (red) could be detected in B13-1 (Fig. 3A) and B13-28 (Fig. 3B) cells on day 5 or day 7 postinduction. There was always a substantial fraction of cells that would express transferrin at various levels. Positive transferrin staining was observed, and HBsAg stain was all over the cells. However, most HBsAg was expressed only in transferrin-expressing cells, though the level of transferrin varied. During the period between day 5 and 7, cell size expansion occurred, as is evident by comparing the same scale bar (20 μm) used in all pictures (Fig. 3). As in parental AR42JB13 cells (27, 48), uninduced B13-1 and B13-28 cells tended to be smaller in size, while the transdifferentiated hepatocytes tended to be enlarged and more flattened (Fig. 3).
Coexpression of liver-enriched transcription factors and HBsAg. Liver-specific gene expression is mainly regulated by a number of liver-enriched transcription factors (11, 30). As shown in Fig. 4, we examined the expression of HBsAg or HBcAg (in red) and four different liver-enriched transcription factors (in green), including C/EBP-, C/EBP-?, HNF 4, and HNF 3?. The nuclei were counterstained in blue with DAPI. Before induction with Dex plus OSM, there was no expression of C/EBP, C/EBP?, and nuclear HNF 4 in either B13-1 (data not shown) or B13-28 (Fig. 4a, g, and m) cells. In contrast to C/EBP, C/EBP?, and nuclear HNF 4, the expression of HNF 3? was observed prior to induction (Fig. 4s). After treatment with Dex plus OSM, most of B13-1 and B13-28 cells expressed C/EBP (Fig. 4f), C/EBP? (Fig. 4l), HNF 3? (Fig. 4x), and HNF 4 in the nucleus (Fig. 4r). The expression of HBsAg correlates best with HNF 4 (Fig. 4p, q, and r). The constitutive expression pattern of HNF 3? serves as a good control for the inducible expression of the other three transcription factors in B13-1 and B13-28 cells.
Western blot analysis of HBcAg, liver, and pancreatic markers. To confirm the results from immunofluorescent staining (Fig. 2 to 4), we performed Western blot analysis on B13-1 and B13-28 cells before and after conversion, using antibodies specific for HBcAg, 1-antitrypsin, -amylase, and -tubulin (Fig. 5). As predicted, HBcAg and 1-antitrypsin were induced upon Dex plus OSM treatment, and -tubulin was unaffected by the treatment. Intriguingly, unlike the parental AR42J-B13 cells (data not shown), we were unable to detect any -amylase in B13-1 and B13-28 cells without induction. However, when the parental AR42J-B13 cells were examined, we were able to detect -amylase before induction (data not shown). Therefore, B13-1 and B13-28 cells appeared to be different from their parental AR42J-B13 cells in their basal expression of -amylase.
Analysis of HBV-specific RNAs before and after transdifferentiation. Upregulation of HBsAg (Fig. 1 and 3) and HBcAg (Fig. 2 and 5) could result from the transcriptional stimulation of Dex plus OSM on these two viral genes. To examine this possibility, we carried out Northern blot analysis on B13-1 and B13-28 cells before and after Dex plus OSM treatment. As shown in Fig. 6A, the most dramatic effect induced by Dex plus OSM was observed on the 2.1-kb pre-S2/S RNA. In contrast, no significant effect was observed on the 2.3-kb pre-S1 RNA, and a mild change was observed in the 3.5-kb RNA, which consists of the pregenomic RNA (pgRNA; core-specific RNA) and the precore RNA. These two RNA species are structurally related and can be distinguished from each other at their 5' ends. To resolve these two closely related RNA species, we performed RNA primer extension analysis using the total intracellular RNA prepared from both B13-1 and B13-28 cells with or without Dex plus OSM. As shown in Fig. 6B, precore RNA initiates at nucleotide (nt) 1785 in both B13-1 and B13-28 cells before induction. After induction, a new species of precore RNA initiates at nt 1789, which is absent in untreated cells. These results are consistent with the ELISA data for HBeAg (Fig. 1).
Most surprisingly, there is only one species of core-specific RNA before induction which initiates at nt 1825. After induction, there are two core-specific RNA species which initiate at nt 1820 and nt 1825, respectively. The existence of core-specific RNA before induction is therefore confirmed by both primer extension and Northern blot analyses.
Southern blot analysis of HBV DNA replication before and after induction. Although B13-1 and B13-28 cells are capable of expressing HBV RNA and protein upon induction (Fig. 1 to 6), it remained unclear whether HBV can indeed replicate in this system. To address this issue, we conducted Southern blot analysis of viral DNAs prepared from B13-1 and B13-28 cells before and after hepatic conversion. HBV DNA isolated from Qs21 was included again as a positive control. As shown in Fig. 7A, characteristic replication patterns of HBV DNA were observed, including the full-length single-strand (SS) and relaxed circular (RC) replicative intermediates on days 3, 5, and 7 postinduction. We noted that the full-length RC form in this system tends to be more abundant than other HBV-producing systems, such as Qs21 (49) (lighter exposure data not shown).
HBV replicates via an RNA intermediate. The pgRNA is transcribed from the cccDNA template in the nucleus (63). To determine if cccDNA can be detected in B13-1 and B13-28 cells after induction, we analyzed the cccDNA by Southern blot analysis. As shown in Fig. 7B, an HBV-specific signal could be detected at a position expected for cccDNA from both Dex plus OSM-treated B13-1 and B13-28 cells. Restriction enzyme EcoRI is known to cut once in the HBV genome of the ayw subtype (49). When the extracted cccDNA preparation was digested with EcoRI before loading, this putative cccDNA band upshifted to a 3.2-kb position on the 1.2% agarose gel, which is where a linearized full-length HBV genome would band (Fig. 7B, lane 3).
To see if intracellular HBV capsids can be secreted into the medium of B13-1 and B13-28 systems, we analyzed the virions by density gradient centrifugation (data not shown). Fractions corresponding to the expected density of HBV virions (around 1.24 g/cm3) were collected and dialyzed to remove CsCl (56). Virion-associated DNA was extracted and analyzed by Southern blotting. As expected from wild-type HBV, the mature genome is preferentially exported (Fig. 7C).
Electron microscopic examination of secreted HBV viral and subviral particles in the medium of transdifferentiated B13-28 cells. The results in Fig. 7C were also confirmed by transmission electron microscopy. As shown in Fig. 8, in addition to tubular and spherical subviral particles, we can detect Dane-like particles with an electron-dense core, concentric shape, and approximately 42 nm in diameter. This result suggests that the hepatocyte-like cells of transdifferentiated B13-1 and B13-28 can indeed support virion assembly and secretion of mature genome similar to an authentic hepatocyte.
Continuous presence of Dex plus OSM is required for HBV replication and gene expression in B13-1 and B13-28 cells. When we removed Dex and OSM from the medium, the levels of secreted HBsAg and HBeAg (Fig. 9A) were downregulated. Moreover, RC and SS DNA also decreased rapidly (Fig. 9B). Similarly, cccDNA was reduced to less than 50% on day 2 and almost undetectable by day 5 after withdrawal (Fig. 9C). Consistent with the ELISA results for HBsAg and HBeAg in Fig. 9A, intracellular HBsAg (p24 and gp27) and HBcAg decreased significantly by Western blot analysis (Fig. 9D). The decrease of viral replication and gene expression was also paralleled by the decrease of 1-antitrypsin and -amylase, as measured by Western blot analysis (Fig. 9D). Again, -tubulin remained unaffected 7 days post-Dex withdrawal. Similar to -tubulin, HNF 3? remained constant after withdrawal (Fig. 10D). In contrast, transcription factors C/EBP , C/EBP ?, and HNF 4 disappeared rapidly and were barely detectable on day 3 after withdrawal (Fig. 10A to C).
DISCUSSION
Transdifferentiation from pancreatic cells to hepatocyte-like cells was first discovered more than two decades ago (44, 46). We demonstrate here for the first time that transdifferentiated hepatocyte-like cells can fully support the fastidious HBV gene expression, replication, cccDNA formation, assembly, and secretion of virions and subviral particles. In all assays, the effect of Dex treatment alone is similar to, yet much slower than, Dex plus OSM treatment (data not shown). Based on these results, B13 cells can be induced to undergo transdifferentiation to bona fide hepatocytes.
Advantages and limitations of transdifferentiated hepatocytes. Although the original main purpose of this study was to provide the most rigorous proof of the authenticity of the transdifferentiated hepatocytes, a useful side-product of this project is the established stable HBV-producing system. It is perhaps worth comparing the B13-1 and B13-28 cells with the most commonly used HBV-producing systems of human Huh7 or HepG2 (5, 47, 55, 61, 68) or rodent (22, 49, 50) cell lines. A tetracycline-inducible HBV-producing system has also been reported (28). One advantage of the B13-1 and B13-28 cells resides in their inducible and synchronized HBV replication upon treatment with physiological ligands (Dex and OSM). Such an inducible feature provides an advantage to synchronize the cell population and thus improve the detection sensitivity in experimental assays for the host hepatocytes. Another advantage of our current transdifferentiated hepatocyte system which is absent in the human HepG2 and Huh7 system is the possibility to extend our future research to an immunocompetent rodent model in vivo. One major limitation of B13-1 and B13-28 cells is that they are unlikely to be infectible with HBV due to the known species barrier of infection between rodents and humans.
Direct or indirect effect of glucocorticoid treatment on HBV replication and HBsAg expression. HBV contains five promoters and two enhancers (20). The pre-S1 promoter, which is responsible for the 2.3-kb viral RNA, is not responsive to Dex plus OSM treatment (Fig. 6A). In contrast, as shown by the results of Northern blot analysis (Fig. 6A), the 40- to 100-fold increased production of HBsAg protein in B13-1 and B13-28 cells by Dex plus OSM treatment (Fig. 1 and 9D) is mainly caused by the specific increase of the 2.1-kb pre-S2/S RNA. This stimulatory effect of Dex plus OSM on HBsAg transcription could be due to the presence of a glucocorticoid response element in the HBV genome. Alternatively, it could indirectly result from increased levels of transcription factors which are involved in HBsAg transcription (37). Although steroid-free chemotherapy decreased the risk of HBV reactivation in HBV carriers with lymphoma (7), this in vivo observation by itself does not argue one way or the other if the Dex effect in vitro is direct or indirect.
A direct effect of Dex on HBV or simian virus 540 (SV40) enhancer/promoter elements appeared unlikely because (i) we observed no apparent effect on the production of HBsAg when HBV-transfected HepG2, Huh7, or Q7 cells were treated with Dex (data not shown); (ii) there is no well-matched glucocorticoid response element consensus motif (5'-AGAACANNNTGTTCT-3') on the HBV or SV40 genome (data not shown); (iii) only a limited effect of Dex on HBV expression in the transient-transfection assay has been reported in the literature (9, 62). Instead of a simple direct effect, we are in favor of the possibility that the Dex effect on HBV could be predominantly indirect and pleiotropic. When AR42J-B13 cells were treated with Dex, cells were reprogrammed toward hepatic phenotypes (48). Indeed, Dex has been reported to have a potent effect on the phenotypic maturation of hepatocytes via growth arrest (38). Many host factors are known to be important for liver development (11, 16, 30, 69). Dex is known to induce expression of HNF 4 and C/EBP-, which are essential factors for hepatocyte differentiation (Fig. 4) (37). In summary, the effect of Dex plus OSM on HBV replication and gene expression in B13-1 and B13-28 cells is most likely indirect, which is not mutually exclusive with the remote possibility of a direct effect.
Is the expression of HBsAg liver specific? While the expression of HBsAg is generally not believed to be liver specific (15, 51, 52), there are reports that HBsAg mRNA or protein is preferentially expressed in more differentiated hepatocyte cell lines (6, 17) or in the liver in animal models (see further discussions below) (3, 4, 8). Interestingly, consistent with our current findings in vitro (Fig. 1, 3, and 6), Dex has been shown to enhance the expression of HBsAg in HBV transgenic mice (19). Ubiquitous transcription factors SP1 and NF-Y are known to bind and transactivate the pre-S2/S promoter (34, 42). Although a liver-specific transcription factor HNF 3 can also bind to the pre-S2/S promoter, it functions as a negative regulator for HBsAg production (57). It is also interesting that HNF 3? is present before or after induction (Fig. 4D and 10D). Therefore, the presence of HNF 3? does not appear to interfere with HBV replication in the presence of Dex plus OSM. Furthermore, C/EBP- and HNF 4 are not known to have any effect on the pre-S2/S promoter, despite their known effects on the precore/core promoters and HBV replication (24, 32, 33, 41, 58, 70). If the liver specificity of the dramatic HBsAg expression were determined by liver-specific enhancers I and/or II of HBV, then it would be hard to reconcile with the modest increase of HBeAg mRNA and protein in transdifferentiated hepatocytes (Fig. 1 and 6). The molecular mechanisms of the liver-specific expression of HBsAg in transdifferentiated hepatocytes in vitro and transgenic mice in vivo merit further investigation.
Extrahepatic expression of hepadnaviruses. As discussed above, the pre-S2/S promoter is more active in well-differentiated hepatocytes. However, extrahepatic gene expression of HBV has also been commonly observed in animal models. For example, HBV gene expression in the kidney has been observed in transgenic mice by at least three independent groups (3, 8, 18). Similarly, viral antigen expression of duck HBV was found in both islets and acini of the pancreas in early embryos and young ducks (25). Indeed, the most common extrahepatic tissues for hepadnaviral gene expression in ducks and woodchucks include spleen, kidney, and pancreas (26, 39). It is possible that some transcription factors are common in these different organs. AR42J-B13 cells were derived from AR42J cells, which in turn originated from a rat pancreatic acinar tumor (31, 36). However, in our B13-1 and B13-28 cells, we are unable to detect HBV DNA replication or viral antigen expression without Dex plus OSM treatment.
A positive transcriptional control and a negative translational control for the gene expression of HBcAg. In the absence of Dex plus OSM, HBV core protein was not detectable by Western blotting (Fig. 5 and 9D) or immunofluorescence assays (Fig. 2), but in the presence of Dex plus OSM, core protein was strongly induced. The absence of detectable core protein in the untreated B13-1 and B13-28 cells does not result from the absence of core-specific mRNA before induction. In fact, core/precore-specific RNAs can be detected in both B13-1 and B13-28 cells before induction by Northern blot analysis (Fig. 6A). To resolve core-specific RNA from the structurally related precore-specific RNA, we performed primer extension analysis (Fig. 6B). We detected novel core-specific transcript in Dex-treated cells, which initiate at nt 1820. Interestingly, another species of core-specific RNA, which initiates at nt 1825, is preexisting before induction and persists at a constant level after induction (Fig. 6B). A similar phenomenon was observed for the precore-specific RNAs. The Dex-inducible species of precore-specific RNA initiates at nt 1789, while the preexisting precore RNA species initiates at nt 1785. In the literature, core-specific RNA initiates around nt 1820 ± 2 (45, 67, 68), and precore-specific RNA initiates around nt 1785 ± 3 (20).
The precore protein is known to be a precursor to HBeAg. The preexisting precore RNA must therefore be responsible for the HBeAg detectable by ELISA before induction (Fig. 1A). What appears to be most intriguing is the preexisting core-specific RNA which initiates at nt 1825, 5 nt downstream from the putative 5' end of the Dex-inducible core-specific RNA. Apparently, this preexisting core-specific RNA is not engaged in the translation of core protein before induction. Since the AUG codon for translational initiation of core protein is at nt 1903, the preexisting core RNA species has a sufficient length of 5'-untranslated region for ribosomal scanning. By inference, the Dex-inducible core-specific RNA is most likely to be responsible for the synthesis of core protein after induction. At present, we entertain the hypotheses that the preexisting core RNA is negatively regulated at the translational level in B13-1 and B13-28 cells before induction (Fig. 11). This phenomenon is somewhat reminiscent of a so-called uncoupled phenotype of bipotential liver cell lines (53). Despite the existence of hepatocyte transcription factors (HNF 1, GATA4, and HNF 4), no hepatocyte functions (albumin or apolipoprotein) were detected.
As for the Dex-induced core RNA, the simplest interpretation is that it is caused by a Dex-induced de novo transcriptional initiation at nt 1820. The fact that the 5' end of the Dex-inducible core RNA is only 5 nt away from the 5' end of the preexisting core RNA raised the question of whether this new species of core RNA could be under the control of posttranscriptional processing. For example, if the 5' end of the preexisting core RNA is somehow blocked in untreated cells, such blockage for translational initiation could also interfere with the progression of reverse transcriptase to the very end of its mRNA template during the in vitro primer extension reaction. If we further assume that the Dex-induced transdifferentiation occurs only at a maximal efficiency near 50%, then the other 50% of the uninduced population of B13-1 or B13-28 cells should still contain only the preexisting core RNA. In this scenario, the core RNA profile of posttranscriptional regulation would be similar to that of a simple transcriptional control. However, judging from the relative intensity between Dex-induced versus preexisting core RNA species in transdifferentiated hepatocytes, a mechanism of posttranscriptional processing appears unlikely. In summary, while the preexisting core RNA appears to be under negative translational control before induction, it remains to be investigated if the Dex-induced core RNA is subject to transcriptional or posttranscriptional control. In either case, it remains to be resolved if it is a direct or indirect effect of Dex plus OSM on the synthesis or processing of core-specific RNA.
Half-lives of HBV RC, SS, and cccDNAs. Because cccDNA in the nucleus is the template for the synthesis of pgRNA, it plays a critical role in the life cycle of HBV (63). It is generally believed that cccDNA is primarily responsible for persistent infection of HBV in the liver (65). The withdrawal of Dex plus OSM from the culture medium of B13-28 cells allowed us to assess the rate of disappearance of HBV DNA replicative intermediates in transdifferentiated hepatocytes. In some systems, the half-lives of cccDNA of hepadnaviruses are very long, ranging from 33 to 70 days or so (2, 13, 71). As quantitated from the signals in Fig. 9C, the cccDNA level from each dish on day 2 after Dex withdrawal was reduced to approximately 45% of the level of day zero, suggesting that the apparent half-life of cccDNA in this system is approximately 36 h. Surprisingly, unlike the cccDNA, the RC and SS DNAs were still detectable on day 7 (Fig. 9B). The actual half-life of cccDNA in B13-28 cells could even be shorter if we assume that RC and SS DNAs can still mature into cccDNA after Dex withdrawal. The shorter half-life of cccDNA (per dish) in our system is more comparable to the reported half-lives of 1 to 5 days in other systems (10, 14, 23, 40). It has been reported that elimination of cccDNA in vivo probably involves both noncytolytic and cytolytic immune-mediated mechanisms (66). In our reversed transdifferentiation system, withdrawal of Dex plus OSM alone, without any drug, cytokine treatment, or cell death, rapidly eliminated cccDNA. Absence of Dex plus OSM relieved the host hepatocytes from growth arrest and thus probably converted the quiescent hepatocytes back to proliferative pancreatic cells.
ACKNOWLEDGMENTS
We thank colleagues in the Shih lab for careful reading of the manuscript and Eugene Knutson and Tom Albrecht at the Optical Imaging Center of UTMB for the help in confocal laser scanning microscopy.
This work was funded by NIH grants R01 CA 70336 and CA 84217 to C.S. and by the Medical Research Council and the Biotechnology and Biological Sciences Research Council to D.T.
REFERENCES
Acs, G., M. S. Sells, R. H. Purcell, P. Price, R. Engle, M. Shapiro, and H. Popper. 1987. Hepatitis B virus produced by transfected Hep G2 cells causes hepatitis in chimpanzees. Proc. Natl. Acad. Sci. USA 84:4641-4644.
Addison, W. R., K. A. Walters, W. W. S. Wong, J. S. Wilson, D. Madej, L. D. Jewell, and D. L. J. Tyrrell. 2002. Half-life of the duck hepatitis B virus covalently closed circular DNA pool in vivo following inhibition of viral replication. J. Virol. 76:6356-6363.
Araki, K., J. Miyazaki, O. Hino, N. Tomita, O. Chisaka, K. Matsubara, and K. Yamamura. 1989. Expression and replication of hepatitis B virus genome in transgenic mice Proc. Natl. Acad. Sci. USA 86:207-211.
Babinet, C., H. Farza, D. Morello, M. Hadchouel, and C. Pourcel. 1985. Specific expression of hepatitis B surface antigen (HBsAg) in transgenic mice. Science 230:1160-1163.
Chang, C. M., K. S. Jeng, C. P. Hu, S. J. Lo, T. S. Su, L. P. Ting, C. K. Chou, S. H. Han, E. Pfaff, and J. Salfeld. 1987. Production of hepatitis B virus in vitro by transient expression of cloned HBV DNA in a hepatoma cell line. EMBO J. 6:675-680.
Chang, H. K., and L. P. Ting. 1989. The surface gene promoter of the human hepatitis B virus displays a preference for differentiated hepatocytes. Virology 170:176-183.
Cheng, A. L., C. A. Hsiung, I. J. Su, P. J. Chen, M. C. Chang, C. J. Tsao, W. Y. Kao, W. C. Uen, C. H. Hsu, H. F. Tien, T. Y. Chao, L. T. Chen, and J. Whang-Peng. 2003. Steroid-free chemotherapy decreases risk of hepatitis B virus (HBV) reactivation in HBV-carriers with lymphoma. Hepatology 37:1320-1328.
Choo, K. B., L. N. Liew, K. Y. Chong, R. H. Lu, and W. T. Cheng. 1991. Transgenome transcription and replication in the liver and extrahepatic tissues of a human hepatitis B virus transgenic mouse. Virology 182:785-792.
Chou, C. K., L. H. Wang, H. M. Lin, and C. W. Chi. 1992. Glucocorticoid stimulates hepatitis B viral gene expression in cultured human hepatoma cells. Hepatology 16:13-18.
Civitico, G. M., and S. A. Locarnini. 1994. The half-life duck hepatitis B virus supercoiled DNA in congenitally infected primary hepatocyte cultures. Virology 203:81-89.
Costa, R. H., V. V. Kalinichenko, A. X. Holterman, and X. Wang. 2003. Transcription factors in liver development, differentiation, and regeneration. Hepatology 38:1331-1347.
Dabeva, M. D., S. G. Hwang, S. R. Vasa, E. Hurston, P. M. Novikoff, D. C. Hixson, S. Gupta, and D. A. Shafritz. 1997. Differentiation of pancreatic epithelial progenitor cells into hepatocytes following transplantation into rat liver. Proc. Natl. Acad. Sci. USA 94:7356-7361.
Dandri, M., M. R. Burda, H. Will, and J. Petersen. 2000. Increased hepatocyte turnover and inhibition of woodchuck hepatitis B virus replication by adefovir in vitro do not lead to reduction of the closed circular DNA. Hepatology 32:139-146.
Delaney, W. E., IV, T. G. Miller, and H. C. Isom. 1999. Use of the hepatitis B virus recombinant baculovirus-HepG2 system to study the effects of (–)-beta-2',3'-dideoxy-3'-thiacytidine on replication of hepatitis B virus and accumulation of covalently closed circular DNA. Antimicrob. Agents Chemother. 43:2017-2026.
Dubois, M. F., C. Pourcel, S. Rousset, C. Chany, and P. Tiollais. 1980. Excretion of hepatitis B surface antigen particles from mouse cells transformed with cloned viral DNA. Proc. Natl. Acad. Sci. USA 77:4549-4553.
Duncan, S. A. 2003. Mechanisms controlling early development of the liver. Mech. Dev. 120:19-33.
Faktor, O., T. De-Medina, and Y. Shaul. 1988. Regulation of hepatitis B virus S gene promoter in transfected cell lines. Virology 162:362-368.
Farza, H., M. Hadchouel, J. Scotto, P. Tiollais, C. Babinet, and C. Pourcel. 1988. Replication and gene expression of hepatitis B virus in a transgenic mouse that contains the complete viral genome. J. Virol. 62:4144-4152.
Farza, H., A. M. Salmon, M. Hadchouel, J. L. Moreau, C. Babinet, P. Tiollais, and C. Pourcel. 1987. Hepatitis B surface antigen gene expression is regulated by sex steroids and glucocorticoids in transgenic mice. Proc. Natl. Acad. Sci. USA 84:1187-1191.
Ganem, D., and R. Schneider. 2001. Hepadnaviridae: the viruses and their replication, p. 2923-2970. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA.
Guidotti, L. G., B. Matzke, H. Schaller, and F. V. Chisari. 1995. High-level hepatitis B virus replication in transgenic mice. J. Virol. 69:6158-6169.
Guilhot, S., T. Miller, G. Cornman, and H. C. Isom. 1996. Apoptosis induced by tumor necrosis factor-alpha in rat hepatocyte cell lines expressing hepatitis B virus. Am. J. Pathol. 148:801-814.
Guo, J. T., M. Pryce, X. Wang, I. M. Barrasa, J. Hu, and C. Seeger. 2003. Conditional replication of duck hepatitis B virus in hepatoma cells. J. Virol. 77:1885-1893.
Guo, W., M. Chen, T. S. Yen, and J. H. Ou. 1993. Hepatocyte-specific expression of the hepatitis B virus core promoter depends on both positive and negative regulation. Mol. Cell. Biol. 13:443-448.
Halpern, M. S., S. B. McMahon, W. S. Mason, and A. P. O'Connell. 1986. Viral antigen expression in the pancreas of DHBV-infected embryos and young ducks. Virology 150:276-282.
Hosoda, K., M. Omata, K. Uchiumi, F. Imazeki, O. Yokosuka, Y. Ito, K. Okuda, and M. Ohto. 1990. Extrahepatic replication of duck hepatitis B virus: more than expected. Hepatology 11:44-48.
Kurash, J. K., C. N. Shen, and D. Tosh. 2004. Induction and regulation of acute phase proteins in transdifferentiated hepatocytes. Exp. Cell. Res. 292:342-358.
Ladner, S. K., M. J. Otto, C. S. Barker, K. Zaifert, G. H. Wang, J. T. Guo, C. Seeger, and R. W. King. 1997. Inducible expression of human hepatitis B virus (HBV) in stably transfected hepatoblastoma cells: a novel system for screening potential inhibitors of HBV replication. Antimicrob. Agents Chemother. 41:1715-1720.
Le Pogam, S., P. K. Chua, M. Newman, and C. Shih. 2005. Exposure of RNA templates and encapsidation of spliced viral RNA are influenced by the arginine-rich domain of human hepatitis B virus core antigen (HBcAg 165-173). J. Virol. 79:1871-1887.
Locker, J. 2001. Tissue-specific regulation by transcription factors, p. 237-262. In J. Locker (ed.), Transcription factors, 1st ed. BIOS Scientific Publishers Ltd., Oxford, England.
Longnecker, D. S., H. S. Lilja, J. French, E. Kuhlmann, and W. Noll. 1979. Transplantation of azaserine-induced carcinomas of pancreas in rats. Cancer Lett. 7:197-202.
Lopez-Cabrera, M., J. Letovsky, K. Q. Hu, and A. Siddiqui. 1990. Multiple liver-specific factors bind to the hepatitis B virus core/pregenomic promoter: trans-activation and repression by CCAAT/enhancer binding protein. Proc. Natl. Acad. Sci. USA 87:5069-5073.
Lopez-Cabrera, M., J. Letovsky, K. Q. Hu, and A. Siddiqui. 1991. Transcriptional factor C/EBP binds to and transactivates the enhancer element II of the hepatitis B virus. Virology 183:825-829.
Lu, C. C., M. Chen, J. H. Ou, and T. S. Yen. 1995. Key role of a CCAAT element in regulating hepatitis B virus surface protein expression. Virology 206:1155-1158.
Marek, C. J., G. A. Cameron, L. J. Elrick, G. M. Hawksworth, and M. C. Wright. 2003. Generation of hepatocytes expressing functional cytochromes P450 from a pancreatic progenitor line in vitro. Biochem. J. 370:763-769.
Mashima, H., H. Shibata, T. Mine, and I. Kojima. 1996. Formation of insulin-producing cells from pancreatic acinar AR42J cells by hepatocyte growth factor. Endocrinology 137:3969-3976.
Michalopoulos, G. K., W. C. Bowen, K. Mule, and J. Luo. 2003. HGF-, EGF-, and dexamethasone-induced gene expression patterns during formation of tissue in hepatic organoid cultures. Gene Express. 11:55-75.
Michalopoulos, G. K., W. C. Bowen, K. Mule, and D. B. Stolz. 2001. Histological organization in hepatocyte organoid cultures. Am. J. Pathol. 159:1877-1887.
Ogston, C. W., E. M. Schechter, C. A. Humes, and M. B. Pranikoff. 1989. Extrahepatic replication of woodchuck hepatitis virus in chronic infection. Virology 169:9-14.
Peek, S. F., P. J. Cote, J. R. Jacob, I. A. Toshkov, W. E. Hornbuckle, B. H. Baldwin, F. V. Wells, C. K. Chu, J. L. Gerin, B. C. Tennant, and B. E. Korba. 2001. Antiviral activity of clevudine [L-FMAU, (1-(2-fluoro-5-methyl-beta, L-arabinofuranosyl) uracil)] against woodchuck hepatitis virus replication and gene expression in chronically infected woodchucks (Marmota monax). Hepatology 33:254-266.
Pei, D., and C. Shih. 1991. An "attenuator domain" is sandwiched by two distinct transactivation domains in the transcription factor C/EBP. Mol. Cell. Biol. 11:1480-1487.
Raney, A. K., H. B. Le, and A. McLachlan. 1992. Regulation of transcription from the hepatitis B virus major surface antigen promoter by the Sp1 transcription factor. J. Virol. 66:6912-6921.
Rao, M. S., V. Subbarao, and J. K. Reddy. 1986. Induction of hepatocytes in the pancreas of copper-depleted rats following copper repletion. Cell. Differ. 18:109-117.
Reddy, J. K., M. S. Rao, S. A. Qureshi, M. K. Reddy, D. G. Scarpelli, and N. D. Lalwani. 1984. Induction and origin of hepatocytes in rat pancreas. J. Cell Biol. 98:2082-2090.
Roychoudhury, S., A. F. Faruqi, and C. Shih. 1991. Pregenomic RNA encapsidation analysis of eleven missense and nonsense polymerase mutants of human hepatitis B virus. J. Virol. 65:3617-3624.
Scarpelli, D. G., and M. S. Rao. 1981. Differentiation of regenerating pancreatic cells into hepatocyte-like cells. Proc. Natl. Acad. Sci. USA 78:2577-2581.
Sells, M. A., M. L. Chen, and G. Acs. 1987. Production of hepatitis B virus particles in Hep G2 cells transfected with cloned hepatitis B virus DNA. Proc. Natl. Acad. Sci. USA 84:1005-1009.
Shen, C. N., J. M. Slack, and D. Tosh. 2000. Molecular basis of transdifferentiation of pancreas to liver. Nat. Cell Biol. 2:879-887.
Shih, C., L. S. Li, S. Roychoudhury, and M. H. Ho. 1989. In vitro propagation of human hepatitis B virus in a rat hepatoma cell line. Proc. Natl. Acad. Sci. USA 86:6323-6327.
Shih, C., M. Y. Yu, L. S. Li, and J. W. Shih. 1990. Hepatitis B virus propagated in a rat hepatoma cell line is infectious in a primate model. Virology 179:871-873.
Standring, D. N., W. J. Rutter, H. E. Varmus, and D. Ganem. 1984. Transcription of the hepatitis B surface antigen gene in cultured murine cells initiates within the presurface region. J. Virol. 50:563-571.
Stenlund, A., D. Lamy, J. Moreno-Lopez, H. Ahola, U. Pettersson, and P. Tiollais. 1983. Secretion of the hepatitis B virus surface antigen from mouse cells using an extra-chromosomal eucaryotic vector. EMBO J. 2:669-673.
Strick-Marchand, H., and M. C. Weiss. 2002. Inducible differentiation and morphogenesis of bipotential liver cell lines from wild-type mouse embryos. Hepatology 36:794-804.
Suk, F. M., M. H. Lin, M. Newman, S. Pan, S. H. Chen, J. D. Liu, and C. Shih. 2002. Replication advantage and host factor-independent phenotypes attributable to a common naturally occurring capsid mutation (I97L) in human hepatitis B virus. J. Virol. 76:12069-12077.
Sureau, C., J. L. Romet-Lemonne, J. I. Mullins, and M. Essex. 1986. Production of hepatitis B virus by a differentiated human hepatoma cell line after transfection with cloned circular HBV DNA. Cell 47:37-47.
Tai, P. C., F. M. Suk, W. H. Gerlich, A. R. Neurath, and C. Shih. 2002. Hypermodification and immune escape of an internally deleted middle-envelope (M) protein of frequent and predominant hepatitis B virus variants. Virology 292:44-58.
Tang, H., and A. McLachlan. 2002. Mechanisms of inhibition of nuclear hormone receptor-dependent hepatitis B virus replication by hepatocyte nuclear factor 3?. J. Virol. 76:8572-8581.
Tang, H., and A. McLachlan. 2001. Transcriptional regulation of hepatitis B virus by nuclear hormone receptors is a critical determinant of viral tropism. Proc. Natl. Acad. Sci. USA 98:1841-1846.
Tosh, D., C. N. Shen, and J. M. Slack. 2002. Differentiated properties of hepatocytes induced from pancreatic cells. Hepatology 36:534-543.
Tosh, D., and J. M. W. Slack. 2003. How cells change their phenotype. Nat. Rev. Mol. Cell Biol. 3:187-194.
Tsurimoto, T., A. Fujiyama, and K. Matsubara. 1987. Stable expression and replication of hepatitis B virus genome in an integrated state in a human hepatoma cell line transfected with the cloned viral DNA. Proc. Natl. Acad. Sci. USA 84:444-448.
Tur-Kaspa, R., R. D. Burk, Y. Shaul, and D. A. Shafritz. 1986. Hepatitis B virus DNA contains a glucocorticoid-responsive element. Proc. Natl. Acad. Sci. USA 83:1627-1631.
Tuttleman, J. S., C. Pourcel, and J. Summers. 1986. Formation of the pool of covalently closed circular viral DNA in hepadnavirus-infected cells. Cell 47:451-460.
Wang, X., M. Al-Dhalimy, E. Lagasse, M. Finegold, and M. Grompe. 2001. Liver repopulation and correction of metabolic liver disease by transplanted adult mouse pancreatic cells. Am. J. Pathol. 158:571-579.
Werle-Lapostolle, B., S. Bowden, S. Locarnini, K. Wursthorn, J. Petersen, G. Lau, C. Trepo, P. Marcellin, Z. Goodman, W. E. Delaney, IV, S. Xiong, C. L. Brosgart, S. S. Chen, C. S. Gibbs, and F. Zoulim. 2004. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 126:1750-1758.
Wieland, S. F., H. C. Spangenberg, R. Thimme, R. H. Purcell, and F. V. Chisari. 2004. Expansion and contraction of the hepatitis B virus transcriptional template in infected chimpanzees. Proc. Natl. Acad. Sci. USA 101:2129-2134.
Will, H., W. Reiser, T. Weimer, E. Pfaff, M. Buscher, R. Sprengel, R. Cattaneo, and H. Schaller. 1987. Replication strategy of human hepatitis B virus. J. Virol. 61:904-911.
Yaginuma, K., Y. Shirakata, M. Kobayashi, and K. Koike. 1987. Hepatitis B virus (HBV) particles are produced in a cell culture system by transient expression of transfected HBV DNA. Proc. Natl. Acad. Sci. USA 84:2678-2682.
Zaret, K. S. 2001. Hepatocyte differentiation: from the endoderm and beyond. Curr. Opin. Genet. Dev. 11:568-574.
Zheng, Y., J. Li, and J. H. Ou. 2004. Regulation of hepatitis B virus core promoter by transcription factors HNF1 and HNF4 and the viral X protein. J. Virol. 78:6908-6914.
Zhu, Y., T. Yamamoto, J. Cullen, J. Saputelli, C. E. Aldrich, D. S. Miller, S. Litwin, P. A. Furman, A. R. Jilbert, and W. S. Mason. 2001. Kinetics of hepadnavirus loss from the liver during inhibition of viral DNA synthesis. J. Virol. 75:311-322.(Robert Yung-Liang Wang, C)