Initiation of hepatitis B virus genome replication ... - Semantic Scholar

Journal of General Virology (2008), 89, 1819–1828

DOI 10.1099/vir.0.83659-0

Initiation of hepatitis B virus genome replication and production of infectious virus following delivery in HepG2 cells by novel recombinant baculovirus vector J. Lucifora,1,2 D. Durantel,1,2,3 L. Belloni,4,5 L. Barraud,4 S. Villet,1,2 I. E. Vincent,1,2 S. Margeridon-Thermet,1,2 O. Hantz,1,2 A. Kay,1,2 M. Levrero5,6 and F. Zoulim1,2,3 Correspondence

1

F. Zoulim

2

INSERM, U871, 151 Cours Albert Thomas, 69003 Lyon, France Universite´ Lyon 1, IFR62 Lyon Est, 69008 Lyon, France

[email protected]

3

Hospices Civils de Lyon, Hoˆtel Dieu Hospital, 69002 Lyon, France

4

Department of Internal Medicine and Laboratory of Gene Expression, Fondazione A. Cesalpino, University of Rome La Sapienza, Rome, Italy

5

Laboratoire Associe´ INSERM, U785, Villejuif, France

6

Eurofins-Viralliance, BioAlliance Pharma SA, Paris, France

Received 13 December 2007 Accepted 9 April 2008

One of the major problems in gaining further insight into hepatitis B virus (HBV)/host-cell interactions is to improve the existing cellular models for the study of HBV replication. The first objective of this study was to improve the system based on transduction of HepG2 cells with a recombinant baculovirus to study HBV replication. A new HBV recombinant baculovirus, BacHBV-1.1, in which the synthesis of pre-genomic RNA is driven by a strong mammalian promoter, was generated. Transduction with this new recombinant baculovirus led to higher levels of HBV replication in HepG2 cells compared with levels obtained with previously described baculovirus vectors. The initiation of a complete HBV DNA replication cycle in Bac-HBV-1.1-transduced HepG2 cells was shown by the presence of HBV replicative intermediates, including covalently closed circular DNA (cccDNA). Only low levels of cccDNA were detected in the nucleus of infected cells. Data showed that cccDNA resulted from the recycling of newly synthesized nucleocapsids and was bound to acetylated histones in a chromatin-like structure. HBV particles released into the supernatant of transduced HepG2 cells were infectious in differentiated HepaRG cells. Several Bac-HBV-1.1 baculoviruses containing HBV strains carrying mutations conferring resistance to lamivudine and/or adefovir were constructed. Phenotypic analysis of these mutants confirmed the results obtained with the transfection procedures. In conclusion, an improved cell-culture system was established for the transduction of replication-competent HBV genomes. This will be useful for future studies of the fitness of HBV mutants.

INTRODUCTION Hepatitis B virus (HBV) remains a major public health problem, with more than 350 million people chronically infected worldwide with a greater risk of developing severe liver disease, including decompensated cirrhosis and hepatocellular carcinoma. To stop or delay disease progression, several therapeutic options are possible based on several drugs approved by the US Food and Drug Administration, i.e. alpha interferon and four nucleos(t)ide analogues: lamivudine (LAM), adefovir dipivoxil (ADV), entecavir and telbivudine. Whilst alpha interferon has low efficacy and is associated with numerous side effects 0008-3659 G 2008 SGM

Printed in Great Britain

(Karayiannis, 2003), nucleos(t)ide analogues potently and specifically inhibit HBV replication by acting on its reverse transcriptase. The viral response induced by these drugs is associated with a significant improvement in biochemical and histological markers in most patients (Lau et al., 2000; Marcellin et al., 2003). HBV is a small enveloped and partially double-stranded DNA virus. In the nucleus of infected cells, the 3.2 kb viral relaxed circular (RC) DNA is converted into covalently closed circular DNA (cccDNA) that serves as a transcription template for both viral mRNAs and 3.5 kb pre-genomic RNA (pgRNA) synthesis. Histone and 1819

J. Lucifora and others

non-histone proteins are bound to cccDNA, resulting in a chromatin-like structure termed the viral minichromosome (Bock et al., 2001; Newbold et al., 1995). A recent study has demonstrated that HBV replication is regulated by the acetylation status of the cccDNA-bound H3/H4 histones (Pollicino et al., 2006). After transcription, the pgRNA, which is a 1.1-unit RNA copy of the genome, is reverse transcribed within the viral nucleocapsid to produce the 3.2 kb viral genomic RC-DNA (Seeger & Mason, 2000).

The overall objective of this work was to improve and characterize this system to study HBV replication, with particular focus on cccDNA formation, HBV virion secretion and infectivity, and resistance to antiviral drugs. The improvement in the technology is illustrated by its usefulness in performing phenotypic studies with clinically relevant and low-level-replicating HBV mutant strains.

cccDNA is a major factor in virus persistence because of its long half-life (Zhu et al., 2001). To clear infection and cccDNA in chronically infected patients, long-term treatments are necessary. Unfortunately, long-term administration of nucleos(t)ide analogues favours the emergence of resistant strains bearing drug-specific mutations within the viral polymerase gene (Zoulim, 2004). Thus, mutations conferring resistance to LAM have been mapped within the C and B domains of the viral reverse transcriptase (rtM204I/V and rtV173L or rtL180M, respectively), whereas mutations conferring resistance to ADV have been mapped within the B and D domains (rtA181V and rtN236T, respectively) (Allen et al., 1998; Delaney et al., 2003; Peters et al., 2004; Villeneuve et al., 2003). Moreover, complex polymerase gene mutants conferring multi-drug resistance have also been described (Villet et al., 2006, 2007).

Cell culture. HepG2 (ATCC), HepaRG (Gripon et al., 2002) and

The most relevant cell-culture models to study HBV resistance to nucleoside analogues in phenotypic assays would be human primary hepatocytes and HepaRG cells because they support a complete HBV life cycle. However, both of these have limitations in terms of accessibility, reproducibility and yield of HBV replication (Gripon et al., 1988, 2002) because of the low percentage of infected cells and the large amounts of input virus that need to be used to infect a reasonable proportion of cells. Thus, transfection of hepatoma cell lines (HepG2 or Huh7) with plasmids containing the complete viral genome is usually used to study HBV resistance to drugs (Brunelle et al., 2005; Durantel et al., 2005; Villet et al., 2006, 2007). Such a system is effective for the study of HBV strains that have a high replication rate, but is limited for those that have a low replication rate. Alternatively, stable cell lines, similar to the HepG2.2.15 model, can be designed for relevant HBV mutants (Qi et al., 2007; Sun & Nassal, 2006). Another possibility that can be used to trigger intracellular HBV replication is to use systems based on the transduction of HepG2 cells with recombinant adenovirus or baculovirus carrying the HBV genome (Delaney & Isom, 1998; Ren & Nassal, 2001). Transduction of hepatoma cells with HBV-loaded baculovirus has led to higher HBV replication levels compared with both transfection approaches and stable cell lines such as HepG2.2.15. Moreover, cccDNA was synthesized in transduced HepG2 cells, although the origin of its formation was not clearly explained (Delaney & Isom, 1998). 1820

METHODS HepG2.2.15 (Sells et al., 1987) cells were maintained in William’s medium (Invitrogen) supplemented with 10 % fetal calf serum (FCS), 50 U penicillin/streptomycin ml21 (Invitrogen), 2 mM GlutaMAX (Invitrogen), 5 mg bovine insulin ml21 and 561025 M hydrocortisone hemisuccinate (Roche Diagnostics) at 37 uC in humidified incubators at 5 % CO2. For differentiation, HepaRG cells were maintained for 2 weeks in standard medium and then for at least a further 2 weeks in medium supplemented with 1.8 % DMSO (Sigma). Spodoptera frugiperda (Sf9) insect cells were cultured in suspension or as monolayers at 28 uC in Sf-900 II medium (Invitrogen) containing 5 % (maintenance) or 10 % (infection) FCS (Perbio). Baculovirus construction and transduction of mammalian cells. The recombinant transfer vector pTriEx-HBV-1.1 (Fig. 1a),

which contains a 1.16unit-length HBV genome (genotype D, serotype ayw) and enables the synthesis of pgRNA under the control of the chicken b-actin promoter, was constructed as described previously (Durantel et al., 2004). The recombinant transfer vector pTriEx-HBV-1.3 was obtained by cloning a 1.36unit-length HBV genome (genotype D, serotype ayw) into pTriEx (Novagen). pTriExHBV-1.1 or pTriEx-HBV-1.3 was co-transfected with linearized baculovirus DNA (BacVector-1000 DNA kit; Novagen) into Sf9 cells using Cellfectin reagent according to the manufacturer’s instructions (Invitrogen). Recombinant viruses were isolated using standard procedures (King, 1992). Ten clones were picked for each construct, amplified and functionally screened for their ability to trigger HBV replication after transduction into hepatoma cells. Baculovirus stock production and plaque assays for virus titration were performed as described previously (King, 1992). Before use in transduction experiments, baculoviruses were concentrated 100-fold by ultracentrifugation. A baculovirus containing an HBV genome carrying a mutation (rtM204P) in the catalytic domain of the polymerase gene (Bac-HBV-1.1-YPDD) was obtained using the same procedure, starting from a transfer vector in which the mutation was inserted by mutagenesis using a commercially available kit (QuikChange Site-directed Mutagenesis kit; Stratagene) using forward primer 59-TTTGGCTTTCAGCTATCCGGATGATGTGGTATTGGGGGCC-39 (mutated nucleotides underlined) and the corresponding reverse primer. This amino acid exchange in the polymerase sequence (YMDD to YPDD) also resulted in an exchange in the surface protein (IWMM to IRMM). The previously described procedure for baculovirus transduction of HepG2 (Delaney & Isom, 1998) was modified slightly. Briefly, HepG2 cells were transduced at 80 % confluency (16105–26105 cells cm22). Baculovirus was diluted in medium to achieve the appropriate m.o.i. and was adsorbed to HepG2 cells at 37 uC with gentle rocking every 15 min to ensure an even distribution. After 1 h, the inoculum was removed and the HepG2 cells were washed with PBS and cultured as described above. Antiviral drugs and treatment. LAM and ADV (Gilead Sciences) were resuspended in sterile water to a final concentration of 10 mM, aliquoted and frozen at 220 uC to avoid repeated freezing and thawing. Drug dilutions were freshly prepared in medium at the time

Journal of General Virology 89

Improvement of the HBV baculovirus/HepG2 system of the experiments. HepG2 cells were supplied with medium containing the indicated concentration of drug at the specified time after transduction with HBV baculovirus. Analysis of viral DNA. Purification of HBV DNA from intracellular

core particles and analysis/quantification by Southern blotting with a radioactive probe were performed as described previously (Seigneres et al., 2002; Summers et al., 1990). HBV DNA was also quantified by real-time PCR using the primers 59-GCTGACGCAACCCCCACT-39 (forward) and 59-AGGAGTTCCGCAGTATGG-39 (reverse). An iCycler MyiO thermocycler (96-well format; Bio-Rad) was used with an iQ SYBR Green Supermix kit. Isolation and detection of cccDNA. Protein-free viral DNA

(cccDNA and baculovirus DNA) was separated from protein-linked viral DNA (RC, linear and single-stranded intermediates) by KCl precipitation as described previously (Summers et al., 1990). To eliminate the baculovirus genome, enzyme digestion was performed with HindIII (New England Biolabs), which cuts several times in the baculovirus DNA, followed by a plasmid-safe DNase (Epicentre) digestion performed by adding 2.5 mM ATP, 16 reaction buffer and 10 U plasmid-safe DNase for 2 h at 37 uC. The DNase was inactivated at 70 uC for 30 min. The digested sample was used as the matrix for rolling-circle amplification (RCA) (Lasken & Egholm, 2003; Margeridon-Thermet et al., 2006). The RCA product (2 ml) was digested in a final volume of 10 ml with 3 U SpeI (New England Biolabs), which cuts only once in the HBV genome. Digestion products were analysed by Southern blotting. Hybridization was performed with a specific S gene oligonucleotide probe labelled with [32P]dCTP by terminal deoxynucleotide transferase (Fermentas). Analysis of viral RNA. Total RNA was extracted from cells using Extract All reagent according to the manufacturer’s instructions (EuroBio). Viral RNA was then analysed by electrophoresis in 1.2 % agarose gels, followed by Northern blot analysis. Hybridization was performed using an HBV DNA-specific probe covering the whole genome. Analysis of secreted HBV antigens. Detection of HBV surface

antigen (HBsAg) and the amounts produced were analysed using a commercially available ELISA kit (MONOLISA Ag HBs PLUS; BioRad) according to the manufacturer’s instructions. Supernatant collected from the cells was clarified by centrifugation at 10 000 g for 5 min and stored at 220 uC until analysed. Analysis of secreted HBV virions by transmission electron microscopy. Concentrated supernatant from the cells was mounted

onto Formvar-coated, carbon-stabilized copper grids (200 mesh). Grids were either stained directly with 1 % uranyl acetate or submitted to immunogold staining using a rabbit anti-HBs primary antibody and a goat anti-rabbit IgG conjugated to 5 nm gold particles (EM.GAR5; British BioCell International) prior to uranyl acetate staining. Grids were observed under a 1200EX transmission electron microscope (JEOL). Infection of HepaRG cells. Infection of differentiated HepaRG cells was performed in medium containing 4 % PEG 8000 (Sigma) for 8 h at 37 uC as described previously (Gripon et al., 2002). At the end of the incubation, cells were washed with medium and cultured as described above. Chromatin immunoprecipitation (ChIP) assays. ChIP assays to

detect cccDNA-bound acetylated histone H3 (AcH3) and acetylated histone H4 (AcH4) were performed as described previously (Pollicino et al., 2006). Real-time ChIP results were expressed as the percentage of the ‘input’. ‘Input’ was the result of the real-time PCR performed with cccDNA-specific primers on the starting chromatin material http://vir.sgmjournals.org

from each experimental condition, including controls and samples, of the ChIP assay. In selected experiments, chromatin immunoprecipitates were digested with HindIII to eliminate the baculovirus genome prior to plasmid-safe DNase treatment for 1 h at 37 uC and then subjected to PCR analysis with cccDNA-specific primers or with primers designed to amplify the baculovirus vector DNA.

RESULTS Improvement of the recombinant HBV baculovirus system to initiate high levels of HBV genome replication Previous reports have described the use of a 1.36genomelength HBV recombinant baculovirus to trigger HBV replication in HepG2 cells. With this system, a higher level of HBV replication was obtained in HepG2 cells compared with that obtained in HepG2.2.15 cells (Delaney & Isom, 1998). Our objective was to improve the efficiency of such a system further, in particular to study poorly replicating mutant strains. To this end, a 1.16genome-length HBV recombinant baculovirus (Bac-HBV-1.1-WT) was generated. In this construct, expression of the pgRNA was placed under the control of a strong mammalian promoter, whereas in Bac-HBV-1.3-WT it was driven by the natural HBV promoter (Fig. 1a). As a control for direct comparison, Bac-HBV-1.3-WT was also constructed using the same HBV sequence. To compare the efficiency of both constructs to initiate an HBV replication cycle, HepG2 cells were transduced at an m.o.i. of 100 with each baculovirus vector. Encapsidated viral DNA and viral RNA were isolated at various time points post-transduction (p.t.), whilst culture medium was collected to analyse HBsAg production. In HepG2 cells transduced with Bac-HBV-1.1-WT, the amount of HBV transcripts was maximal at day 1 p.t. and declined afterwards until the last day of analysis (day 12 p.t.; Fig. 1b). In HepG2 cells transduced with Bac-HBV-1.3WT, HBV transcripts were barely detectable at day 1 p.t. by Northern blotting, reached a peak level between days 3 and 6 p.t. and declined thereafter (Fig. 1b). As synthesis of HBV transcripts was higher in Bac-HBV-1.1-WT-transduced HepG2 cells, the subsequent amounts of encapsidated replicative intermediates and secreted HBsAg were also higher in Bac-HBV-1.1-WT-transduced HepG2 cells compared with those observed in Bac-HBV-1.3-WT-transduced HepG2 cells (Fig. 1c and d). HBV encapsidated replicative intermediates and secreted HBsAg remained detectable after day 12 p.t. for up to 1 month in HepG2 cells that were not split (data not shown). The amount of intracellular baculovirus genome was maximal at day 1 p.t. and decreased thereafter, with kinetics mirroring those of pgRNA, but was still detectable by Southern blotting at day 27 p.t. (data not shown). The differences in kinetics of accumulation and clearance of both baculoviral and HBV nucleic acids suggested that the half-life of encapsidated HBV DNA was longer than that of pgRNA or the baculoviral genome. It was noteworthy that clearance of 1821


Fig. 1. Improvement of the baculovirus system. (a) Maps of the HBV recombinant baculoviruses. (b–e) HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT and Bac-HBV-1.3-WT. At various times p.t., viral RNA was isolated and analysed by Northern blotting (b), medium was collected and submitted to HBsAg detection (c), encapsidated DNA was isolated and analysed by Southern blotting (d) and a mitochondrial toxicity test (MTT) was performed (e) (absorbance values were normalized with non-transduced cells at each time point). Mock, non-transduced cells.

encapsidated HBV DNA was faster when transduced cells were split every 3–4 days (data not shown). Finally, it is important to note that, despite the high level of HBV replication, no significant cytotoxicity was observed with either baculovirus vector (Fig. 1e), thus confirming that differences observed in replication kinetics after transduction with these baculoviruses vectors were not due to differences in toxicity. Collectively, these results demonstrated the superiority of Bac-HBV-1.1-WT over Bac-HBV-1.3-WT in initiation of HBV genome replication in HepG2 cells. Initiation of HBV genome replication is followed by the formation of transcriptionally active cccDNA in transduced HepG2 cells Delaney & Isom (1998) reported that cccDNA formation takes place in Bac-HBV-1.3-transduced HepG2 cells. Similar experiments were performed with HepG2 cells obtained from ATCC. To detect low levels of cccDNA, a sensitive method based on RCA was used (MargeridonThermet et al., 2006). HepG2 cells were transduced and cccDNA was isolated at various times p.t. and amplified by 1822

RCA after complete elimination of baculovirus genomes by HindIII and plasmid-safe DNase digestions. cccDNA was detected at day 1 p.t. and remained detectable up to the last day of the assay (day 12 p.t.; Fig. 2a). To demonstrate the specificity of the method for cccDNA detection, RCA was performed on samples digested by EcoRI (which cuts once in the HBV genome and cccDNA) or HindIII (no digestion sites in the HBV genome). As expected, the cccDNA signal was lost when viral DNA was linearized by EcoRI prior to RCA (Fig. 2b). The next objective was to gain insight into the origin of cccDNA formation. We tested two main hypotheses. First, we tested whether cccDNA could be formed by nucleocapsid recycling back towards the nucleus, as described previously in the duck HBV model (Tuttleman et al., 1986). Secondly, we tested whether cccDNA could be formed by recombination from the baculovirus vector between the repeated sequences of HBV (Fig. 1a). HepG2 cells transduced with Bac-HBV-1.1-WT were treated with a high concentration of LAM. If cccDNA is formed by nucleocapsid recycling, LAM administration would also prevent cccDNA formation. Our results showed that high concentrations of LAM, which completely eliminates Journal of General Virology 89

Improvement of the HBV baculovirus/HepG2 system

acetylation of bound H4 histones was observed in BacHBV-1.1-WT-transduced HepG2 cells treated with LAM (Fig. 3d), consistent with the observation that cccDNA is not formed in LAM-treated cells. These results suggested that the cccDNA formed in HepG2 cells transduced with Bac-HBV-1.1-WT was probably transcriptionally active and used as a template for HBV RNA synthesis. HepG2 cells transduced with recombinant baculovirus vectors produce infectious HBV virions

Fig. 2. cccDNA formed in HepG2 cells transduced with BacHBV-1.1-WT detected by RCA. HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT. cccDNA was isolated at various times p.t. and submitted directly to RCA (a) or digested by EcoRI (cutting once) or HindIII (no digestion sites in the HBV genome) prior to RCA (b). The RCA product was digested with SpeI (cutting once in the HBV genome) and analysed by Southern blotting. Mock, non-transduced cells.

wild-type HBV replication, prevented cccDNA formation (Fig. 3a), suggesting that HBV DNA synthesis is a prerequisite for the formation of cccDNA by nucleocapsid recycling under our experimental conditions. To demonstrate this point further, another baculovirus (Bac-HBV1.1-YPDD), containing an HBV genome carrying a mutation in the catalytic domain (rtM204P) that eliminates polymerase activity, was used. No formation of cccDNA was observed with this mutant, confirming that cccDNA most probably originates from intracellular nucleocapsid recycling towards the nucleus. Next, we asked whether cccDNA formed in Bac-HBV-1.1WT-transduced HepG2 cells could be used as a template for HBV RNA synthesis. The acetylation status of cccDNAbound histones H3 and H4 was analysed. HepG2 cells were transduced and formaldehyde-cross-linked nuclear chromatin was immunoprecipitated with specific anti-AcH3 and anti-AcH4 antibodies. The specificity of the cccDNA ChIP assay in the context of Bac-HBV-1.1-WT-transduced HepG2 cells was supported by the different kinetics of baculovirus- and cccDNA-bound H4 acetylation and the observation that HindIII digestion of anti-AcH4 chromatin immunoprecipitates eliminated the baculovirus-specific PCR signal without affecting the cccDNA-specific PCR signal (Fig. 3b). A cccDNA-specific real-time PCR assay (Werle-Lapostolle et al., 2004) was then performed to quantify cccDNA in the anti-AcH3 histone and anti-AcH4 histone ChIPed material. At day 1 p.t., approximately 30 and 8 % of the cccDNA formed contained AcH3 and AcH4 histones, respectively (Fig. 3c). The acetylation of bound H3 and H4 histones remained detectable at day 3 p.t. but disappeared at day 6 p.t. Moreover, no significant http://vir.sgmjournals.org

It has been shown previously that HepG2 cells transduced with Bac-HBV-1.3 produced HBV particles that peaked at 1.21 g cm23 in isopycnic density gradient analysis (Delaney & Isom, 1998). Supernatant from HepG2 cells transduced with Bac-HBV-1.1-WT was characterized by electron microscopy analyses and immunogold staining in comparison with that produced in stably transfected HepG2.2.15 cells (Sells et al., 1987). The results revealed the presence of Dane particles, spheres and filamentous particles in concentrated supernatant from HepG2 cells transduced with Bac-HBV-1.1-WT (Fig. 4a), as well as in supernatant from HepG2.2.15 cells. Of note with regard to the infectivity experiments described below, no baculoviruses were observed in the same concentrated supernatants. To investigate whether HBV virions produced in HepG2 cells transduced with Bac-HBV-1.1-WT were infectious, HepaRG cells were inoculated with the corresponding concentrated supernatant. As a positive control, supernatant from HepG2.2.15 cells was also used. In addition, we ruled out the presence of residual baculoviruses that could have been responsible for HepaRG transduction using microscopy and functional assays (data not shown). The amount of HBsAg production in the supernatant of inoculated HepaRG cells was followed for 12 days and cells were lysed to analyse HBV RNA by Northern blotting 12 days post-inoculation. HBsAg and HBV RNA synthesis were detectable not only in HepaRG cells infected by supernatant from HepG2.2.15 cells, but also in cells infected by supernatant from HepG2 cells transduced with Bac-HBV-1.1-WT (Fig. 4b). These results indicated that HBV virions produced from HepG2 cells transduced with Bac-HBV-1.1-WT were infectious. Use of the improved recombinant HBV baculovirus system to study a complex mutant with low replication capacity It has been shown previously that some complex and clinically relevant HBV mutants have significantly lower replication capacity levels (Brunelle et al., 2005; Villeneuve et al., 2003; Villet et al., 2006, 2007). In these cases, phenotyping analyses are difficult to perform with a transfection system. Thus, the baculovirus system could be very useful, as levels of replication are higher. To test this hypothesis, mutations conferring resistance to LAM 1823


Fig. 3. cccDNA is formed by nucleocapsid recycling in HepG2 cells transduced with Bac-HBV-1.1-WT and bound to AcH3 and AcH4 histones. (a) HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT or with Bac-HBV-1.1-YPDD and immediately treated or not with 100 mM LAM. RC-DNA and cccDNA were isolated at various times p.t. and analysed directly by Southern blotting (top panel) or submitted to RCA followed by Southern blotting (bottom panel), respectively. (b) HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT. ChIP was performed at various times p.t. using an anti-AcH4-specific antibody. After the reverse cross-linking step, chromatin immunoprecipitates were digested with HindIII, treated with plasmid-safe DNase and then subjected to PCR analysis with primers specific for the pre-core/core promoter (preC/C Pr) of HBC cccDNA or primers designed to amplify the baculovirus vector DNA (Baculo). IgG indicates immunoprecipitation with non-specific immunoglobulins (Santa Cruz Biotechnology) as a negative control. (c) HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT. ChIP was performed at various times p.t. using antibodies specific to AcH3 and AcH4. Selective primers were used to amplify the pre-C/C promoter of HBV cccDNA. AcH3- and AcH4-bound cccDNA is expressed as a percentage of the input. (d) HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT and immediately treated or not with 100 mM LAM. ChIP was performed at various times p.t. using antibodies specific to AcH4. Selective primers were used to amplify the pre-C/C promoter of HBV cccDNA. AcH4-bound cccDNA is expressed as a percentage of the input. Mock, non-transduced cells.

(rtL180M+M204V), ADV (rtN236T) and to both drugs (rtL180M+M204V+N236T) that were engineered previously into the sequence used for the construction of wildtype baculoviruses (Brunelle et al., 2005) were used to derive corresponding baculovirus vectors, i.e. Bac-HBV1.1-NT (containing the rtN236T substitution), Bac-HBV1.1-LMMV (containing the rtL180M+M204V substitutions) and Bac-HBV-1.1-LMMVNT (containing the rtL180M+M204V+N236T substitutions). First, HepG2 cells were transduced with these baculoviruses to study the replication ability of these HBV mutant strains. DNA from intracellular core particles was extracted at 2 days p.t. and submitted to Southern blot and real-time PCR analysis (Fig. 5a). As reported previously using transfection of plasmids (Brunelle et al., 2005), the replication rate of the mutants was reduced in the following order: WT 1824

(100 %).rtN236T (43 %).rtL180M+rtM204V (30 %). rtL180M+rtM204V+rtN236T (7 %). Secondly, the mutant HBV/baculovirus vectors were used to study the susceptibility of mutants to LAM and ADV. As expected, the rtL180M+M204V and rtN236T strains were susceptible to ADV and LAM, respectively, whereas the WT strain was susceptible to both drugs. Both the rtN236T and rtL180M+M204V+N236T strains were less susceptible to ADV, with 1.7- and 7.8-fold resistance, respectively, and both the rtL180M+M204V and rtL180M+M204V+ N236T strains were resistant to LAM, with .33-fold resistance (Fig. 5b). Interestingly, cccDNA was detected by RCA, performed at 48 h p.t., for each mutant strain, even with the low-level-replicating rtL180M+rtM204V+ rtN236T mutant. As expected, LAM administration prevented the formation of cccDNA in HepG2 cells Journal of General Virology 89


DISCUSSION Stably transfected hepatocyte-derived cell lines such as HepG2.2.15 (Sells et al., 1987) or transient transfection of plasmid containing the HBV genome in HepG2 or HuH7 cells are the models currently used for in vitro studies of HBV replication (Brunelle et al., 2005; Durantel et al., 2005). It has been demonstrated that HBV recombinant baculovirus vectors can also be used to transduce HepG2 cells in order to initiate high levels of HBV replication and study HBV resistance to nucleoside analogues (Delaney & Isom, 1998; Delaney et al., 1999, 2003). To study virus replication and HBV mutants with a low replication capacity, we designed a new HBV recombinant baculovirus vector (Bac-HBV-1.1-WT) in which HBV pgRNA synthesis is driven by a strong mammalian promoter. This vector differed from the previously described Bac-HVB-1.3-WT that contained 1.36unitlength HBV genomes and in which the pgRNA was driven by the endogenous HBV promoter (Delaney & Isom, 1998). Another advantage of our vector was that the rate of pgRNA synthesis was identical, regardless of the HBV genome cloned, as shown previously (Durantel et al., 2004). It is noteworthy that the construction of transfer vectors used for the production of a given baculovirus has been optimized previously, thus facilitating the generation of novel recombinant vectors harbouring relevant HBV genomes (Durantel et al., 2004). HBV replication was found to be higher in HepG2 cells transduced with BacHBV-1.1-WT compared with cells transduced with BacHVB-1.3-WT. This improved viral genome replication was most likely due to increased synthesis of pgRNA. Despite a higher replication rate, it was not possible to produce a persistent infection in HepG2 cells. Fig. 4. Baculovirus system allowing production of HBV virions that can infect HepaRG cells. HepG2 and HepG2.2.15 cells were seeded at 80 % confluency in 75 cm2 bottles (9¾106 cells) and allowed to attach for 5 h. HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT. The supernatant of transduced HepG2 cells and HepG2.2.15 cells was collected every 3 days for 15 days, filtered (0.45 mm pore size) and concentrated 50-fold by ultracentrifugation from 75 ml supernatant. (a) Supernatants were analysed by transmission electron microscopy. Specific labelling of HBV particles was performed using an anti-HBs antibody and is shown in the inserts for each preparation. (b) Concentrated supernatants from HepG2 cells transduced with Bac-HBV-1.1WT or from HepG2.2.15 cells were used to infect HepaRG cells. Supernatant from HepaRG cells was collected 6–12 days postinoculation for HBsAg analysis and HBV RNA was analysed 12 days post-inoculation by Northern blotting. Mock, supernatant from non-transduced cells.

transduced with baculovirus loaded with WT or rtN236T strains, but not in cells transduced with baculovirus loaded with LAM-resistant strains (rtL180M+M204V and rtL180M+M204V+N236T) (Fig. 5c). http://vir.sgmjournals.org

Delaney & Isom (1998) studied HBV virion secretion in HepG2 cells transduced with their HBV recombinant baculovirus by isopycnic centrifugation in a CsCl gradient and fractionation. They showed that HBV DNA and HBsAg co-migrated at a density that has been reported for Dane particles. The results of our experiments using electron microscopy analysis showed that HepG2 cells transduced with Bac-HBV-1.1-WT produced Dane particles, spheres and filamentous particles usually found in the sera of patients (Ganem, 1991). Furthermore, we provided new data showing that the supernatant from HepG2 cells transduced with Bac-HBV-1.1-WT contained HBV particles that were infectious and could initiate HBV replication in HepaRG cells. Therefore, this system may prove useful for studies of the infectivity and fitness of clinically relevant HBV mutants. An important issue in HBV infection concerns the synthesis of cccDNA, as this is the template for viral RNA synthesis in vivo and plays a crucial role in virus persistence. Delaney & Isom (1998) demonstrated that cccDNA could be formed after transduction of HepG2 with HBV recombinant baculovirus. However, two main questions remained largely unanswered: what was the 1825


Fig. 5. Analysis of HBV mutant strains using the improved baculovirus system. HepG2 cells were transduced at an m.o.i. of 100 with Bac-HBV-1.1-WT, Bac-HBV-1.1-NT, Bac-HBV-1.1-LMMV or Bac-HBV-1.1-LMMVNT. (a) At 48 h p.t., cells were lysed and encapsidated DNA was isolated and analysed by Southern blotting (left) or real-time PCR (right). The values obtained for each mutant were divided by the WT value, which was assumed to represent 100 % virus replication. Each value is the mean±SD of at least three independent experiments. (b) HepG2 cells were treated immediately after transduction with six different concentrations of LAM or ADV. The medium was removed and replaced by fresh medium containing the drug at different concentrations every day for 3 days. DNA from intracellular core particles was extracted and submitted to Southern blotting. The panels represent different autoradiograms with different exposure times. The HBV DNA hybridization signal was quantified by PhosphorImager and ImageQuant software and the results are shown in (b) as EC50 (drug concentration for which HBV replication has decreased by 50 %), EC90 (drug concentration for which HBV replication has decreased by 90 %) and fold resistance (mutant EC50/WT EC50). Values are means±SD of three independent experiments, each performed in duplicate. (c) At 48 h p.t., cells were lysed and cccDNA was isolated at various time p.t. and submitted to RCA. Mock, non-transduced cells; NT, Bac-HBV-1.1-NT; LMMV, Bac-HBV-1.1-LMMV; LMMVNT, Bac-HBV-1.1-LMMVNT.

1826



origin of the cccDNA synthesized, and was this cccDNA transcriptionally active and therefore responsible for the synthesis of viral RNAs? Our results showed that low levels of cccDNA could be detected after transduction with baculoviruses loaded with WT or mutant HBV strains. The overall amount of cccDNA detected in both cases was lower than that detected by Delaney & Isom (1998), as it was not detectable by Southern blot analysis. One may hypothesize that differences in cell-line batches and experimental conditions may explain this discrepancy. The results of experiments with LAM treatment suggested that cccDNA is most probably formed by nucleocapsid recycling to the nucleus. This was consistent with the results of experiments using constructs expressing a polymerase-inactive mutant, which ruled out the possibility that cccDNA could be formed by recombination from the baculovirus vector. It has been shown that cccDNA can persist in cell nuclei as a stable chromatin-like episome associated with AcH3 and Ac H4 histones when replication is active (Pollicino et al., 2006). Using a ChIP/cccDNA assay described recently (Pollicino et al., 2006), we found that H3 and H4 histones bound to cccDNA were acetylated, suggesting that cccDNA formed in HepG2 cells transduced with Bac-HBV-1.1-WT may be transcriptionally active. However, the acetylation of cccDNA-bound H3 and H4 histones was only transient in transduced HepG2 cells. This may explain why a persistent HBV replication could not be produced in transduced HepG2 cells, as well as the absence of selection of HBV resistant mutants when suboptimal doses of drugs were administrated to transduced HepG2 cells (data not shown). With the generation of additional 1.16genome-length HBV recombinant baculovirus vectors carrying mutations conferring resistance to LAM, ADV or both, we demonstrated that this technology is also relevant for studying virus resistance to nucleoside analogues in so-called phenotypic studies. Thus, the strong mammalian promoter in the Bac-HBV-1.1 vector led to a higher level of HBV genome replication than that obtained with the Bac-HBV1.3 vector in which the pgRNA synthesis is driven by the endogenous HBV promoter, without affecting the order of replication levels of the HBV mutants and their susceptibility to nucleosides analogues (Brunelle et al., 2005). Moreover, cccDNA formation occurred after transduction of HepG2 cells with each HBV mutant baculovirus, including the low-level-replicating rtL180M+rtM204V+ rtN236T strain. Taken together, these results demonstrate the usefulness of baculovirus transduction for the study of low-level-replicating HBV mutant strains. This approach may also prove useful for the production of HBV mutant particles in vitro to perform proper fitness studies both in vitro and in vivo. In conclusion, the results of our study recombinant HBV baculovirus vector 1.16unit-length HBV genome may initiate HBV replication following transduction of http://vir.sgmjournals.org

show that a encoding a high levels of HepG2 cells,

including the formation of transcriptionally active cccDNA and the production of infectious HBV particles. This system may be of interest for in vitro studies of the fitness of antiviral-resistant mutants.

ACKNOWLEDGEMENTS We would like to thank Dr Marie-Anne Petit (INSERM, U871) for the anti-HBs antibody and Simone Peyrol (IFR62, Lyon Est) for help with electron microscopy analyses. This work was supported by a grant from the European Community (ViRgil LSHM-CT-2004-503359) and the French National Agency for Research against AIDS and viral hepatitis (ANRS).

REFERENCES Allen, M. I., Deslauriers, M., Andrews, C. W., Tipples, G. A., Walters, K. A., Tyrrell, D. L., Brown, N. & Condreay, L. D. (1998). Identification

and characterization of mutations in hepatitis B virus resistant to lamivudine. Lamivudine Clinical Investigation Group. Hepatology 27, 1670–1677. Bock, C. T., Schwinn, S., Locarnini, S., Fyfe, J., Manns, M. P., Trautwein, C. & Zentgraf, H. (2001). Structural organization of the

hepatitis B virus minichromosome. J Mol Biol 307, 183–196. Brunelle, M. N., Jacquard, A. C., Pichoud, C., Durantel, D., CarroueeDurantel, S., Villeneuve, J. P., Trepo, C. & Zoulim, F. (2005).

Susceptibility to antivirals of a human HBV strain with mutations conferring resistance to both lamivudine and adefovir. Hepatology 41, 1391–1398. Delaney, W. E., IV & Isom, H. C. (1998). Hepatitis B virus replication

in human HepG2 cells mediated by hepatitis B virus recombinant baculovirus. Hepatology 28, 1134–1146. Delaney, W. E., IV, Miller, T. G. & Isom, H. C. (1999). Use of the

hepatitis B virus recombinant baculovirus-HepG2 system to study the effects of (2)-b-29,39-dideoxy-39-thiacytidine on replication of hepatitis B virus and accumulation of covalently closed circular DNA. Antimicrob Agents Chemother 43, 2017–2026. Delaney, W. E., IV, Yang, H., Westland, C. E., Das, K., Arnold, E., Gibbs, C. S., Miller, M. D. & Xiong, S. (2003). The hepatitis B virus

polymerase mutation rtV173L is selected during lamivudine therapy and enhances viral replication in vitro. J Virol 77, 11833–11841. Durantel, D., Carrouee-Durantel, S., Werle-Lapostolle, B., Brunelle, M. N., Pichoud, C., Trepo, C. & Zoulim, F. (2004). A new strategy for

studying in vitro the drug susceptibility of clinical isolates of human hepatitis B virus. Hepatology 40, 855–864. Durantel, D., Brunelle, M. N., Gros, E., Carrouee-Durantel, S., Pichoud, C., Villet, S., Trepo, C. & Zoulim, F. (2005). Resistance

of human hepatitis B virus to reverse transcriptase inhibitors: from genotypic to phenotypic testing. J Clin Virol 34 (Suppl. 1), S34–S43. Ganem, D. (1991). Assembly of hepadnaviral virions and subviral

particles. Curr Top Microbiol Immunol 168, 61–83. Gripon, P., Diot, C., Theze, N., Fourel, I., Loreal, O., Brechot, C. & Guguen-Guillouzo, C. (1988). Hepatitis B virus infection of adult

human hepatocytes cultured in the presence of dimethyl sulfoxide. J Virol 62, 4136–4143. Gripon, P., Rumin, S., Urban, S., Le Seyec, J., Glaise, D., Cannie, I., Guyomard, C., Lucas, J., Trepo, C. & Guguen-Guillouzo, C. (2002).

Infection of a human hepatoma cell line by hepatitis B virus. Proc Natl Acad Sci U S A 99, 15655–15660. 1827

J. Lucifora and others Karayiannis, P. (2003). Hepatitis B virus: old, new and future

approaches to antiviral treatment. J Antimicrob Chemother 51, 761–785.

Seigneres, B., Pichoud, C., Martin, P., Furman, P., Trepo, C. & Zoulim, F. (2002). Inhibitory activity of dioxolane purine analogs on

a Laboratory Guide. London: Chapman & Hall.

wild-type and lamivudine-resistant mutants of hepadnaviruses. Hepatology 36, 710–722.

Lasken, R. S. & Egholm, M. (2003). Whole genome amplification:

Sells, M. A., Chen, M. L. & Acs, G. (1987). Production of hepatitis B

abundant supplies of DNA from precious samples or clinical specimens. Trends Biotechnol 21, 531–535.

virus particles in Hep G2 cells transfected with cloned hepatitis B virus DNA. Proc Natl Acad Sci U S A 84, 1005–1009.

Lau, D. T., Khokhar, M. F., Doo, E., Ghany, M. G., Herion, D., Park, Y., Kleiner, D. E., Schmid, P., Condreay, L. D. & other authors (2000).

Summers, J., Smith, P. M. & Horwich, A. L. (1990). Hepadnavirus

King, L. A. & Possee, R. D. (1992). The Baculovirus Expression System:

Long-term therapy of chronic hepatitis B with lamivudine. Hepatology 32, 828–834. Marcellin, P., Chang, T. T., Lim, S. G., Tong, M. J., Sievert, W., Shiffman, M. L., Jeffers, L., Goodman, Z., Wulfsohn, M. S. & other authors (2003). Adefovir dipivoxil for the treatment of hepatitis B e

antigen-positive chronic hepatitis B. N Engl J Med 348, 808–816. Margeridon, S., Durantel, S., Chemin, I., Barraud, L., Zoulim, F., Tre´po, C. & Kay, A. (2006). Rolling circle amplification technology for

the study of HBV variants. J Clin Virol 36, S51–S52. Newbold, J. E., Xin, H., Tencza, M., Sherman, G., Dean, J., Bowden, S. & Locarnini, S. (1995). The covalently closed duplex form of the

hepadnavirus genome exists in situ as a heterogeneous population of viral minichromosomes. J Virol 69, 3350–3357. Peters, M. G., Hann Hw, H., Martin, P., Heathcote, E. J., Buggisch, P., Rubin, R., Bourliere, M., Kowdley, K., Trepo, C. & other authors (2004). Adefovir dipivoxil alone or in combination with lamivudine

in patients with lamivudine-resistant Gastroenterology 126, 91–101.

chronic

hepatitis

B.

Pollicino, T., Belloni, L., Raffa, G., Pediconi, N., Squadrito, G., Raimondo, G. & Levrero, M. (2006). Hepatitis B virus replication is

regulated by the acetylation status of hepatitis B virus cccDNA-bound H3 and H4 histones. Gastroenterology 130, 823–837. Qi, X., Xiong, S., Yang, H., Miller, M. & Delaney, W. E., 4th (2007). In

vitro susceptibility of adefovir-associated hepatitis B virus polymerase mutations to other antiviral agents. Antivir Ther 12, 355–362. Ren, S. & Nassal, M. (2001). Hepatitis B virus (HBV) virion and

covalently closed circular DNA formation in primary tupaia hepatocytes and human hepatoma cell lines upon HBV genome transduction with replication-defective adenovirus vectors. J Virol 75, 1104–1116. Seeger, C. & Mason, W. S. (2000). Hepatitis B virus biology.

Microbiol Mol Biol Rev 64, 51–68.

1828

envelope proteins regulate covalently closed circular DNA amplification. J Virol 64, 2819–2824. Sun, D. & Nassal, M. (2006). Stable HepG2- and Huh7-based human

hepatoma cell lines for efficient regulated expression of infectious hepatitis B virus. J Hepatol 45, 636–645. Tuttleman, J. S., Pourcel, C. & Summers, J. (1986). Formation of the

pool of covalently closed circular viral DNA in hepadnavirus-infected cells. Cell 47, 451–460. Villeneuve, J. P., Durantel, D., Durantel, S., Westland, C., Xiong, S., Brosgart, C. L., Gibbs, C. S., Parvaz, P., Werle, B. & other authors (2003). Selection of a hepatitis B virus strain resistant to adefovir in a

liver transplantation patient. J Hepatol 39, 1085–1089. Villet, S., Pichoud, C., Villeneuve, J. P., Trepo, C. & Zoulim, F. (2006).

Selection of a multiple drug-resistant hepatitis B virus strain in a liver-transplanted patient. Gastroenterology 131, 1253–1261. Villet, S., Ollivet, A., Pichoud, C., Barraud, L., Villeneuve, J. P., Trepo, C. & Zoulim, F. (2007). Stepwise process for the development

of entecavir resistance in a chronic hepatitis B virus infected patient. J Hepatol 46, 531–538. Werle-Lapostolle, B., Bowden, S., Locarnini, S., Wursthorn, K., Petersen, J., Lau, G., Trepo, C., Marcellin, P., Goodman, Z. & other authors (2004). Persistence of cccDNA during the natural history of

chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 126, 1750–1758. Zhu, Y., Yamamoto, T., Cullen, J., Saputelli, J., Aldrich, C. E., Miller, D. S., Litwin, S., Furman, P. A., Jilbert, A. R. & Mason, W. S. (2001).

Kinetics of hepadnavirus loss from the liver during inhibition of viral DNA synthesis. J Virol 75, 311–322. Zoulim, F. (2004). Mechanism of viral persistence and resistance to nucleoside and nucleotide analogs in chronic hepatitis B virus infection. Antiviral Res 64, 1–15.
