These authors contributed equally to this work.
Hepatitis B virus replication in primary macaque hepatocytes: Crossing the species barrier toward a new small primate model†
Article first published online: 19 FEB 2010
Copyright © 2010 American Association for the Study of Liver Diseases
Volume 51, Issue 6, pages 1954–1960, June 2010
How to Cite
Lucifora, J., Vincent, I. E., Berthillon, P., Dupinay, T., Michelet, M., Protzer, U., Zoulim, F., Durantel, D., Trepo, C. and Chemin, I. (2010), Hepatitis B virus replication in primary macaque hepatocytes: Crossing the species barrier toward a new small primate model. Hepatology, 51: 1954–1960. doi: 10.1002/hep.23602
Potential conflict of interest: Nothing to report.
- Issue published online: 23 MAY 2010
- Article first published online: 19 FEB 2010
- Accepted manuscript online: 19 FEB 2010 12:00AM EST
- Manuscript Accepted: 29 JAN 2010
- Manuscript Received: 31 JUL 2009
The development of new anti–hepatitis B virus (HBV) therapies, especially immunotherapeutic approaches, has been limited by the lack of a primate model more accessible than chimpanzees. We have previously demonstrated that sylvanus and cynomolgus macaques are susceptible to in vivo HBV infection after intrahepatic HBV DNA inoculation. In this study, we evaluated the susceptibility of primary macaque hepatocytes (PMHs) to HBV infection with a highly efficient HBV genome–mediated transfer system via a recombinant baculovirus (Bac-HBV). Freshly prepared PMHs, isolated from macaque liver tissue by collagenase perfusion, were transduced with Bac-HBV, and intermediates of replication were followed for 9 days post-transduction. Evidence of HBV replication (hepatitis B surface antigen secretion, viral DNA, RNA, and covalently closed circular DNA) was detected from day 1 to day 9 post-transduction. HBV markers were dose-dependent and still detectable at a multiplicity of infection of 10. Importantly, transduced PMHs secreted all typical forms of HBV particles, as evidenced by a cesium chloride gradient as well as transmission electron microscopy. Furthermore, the Toll-like receptor 9 (TLR9) ligand was used to stimulate freshly prepared macaque peripheral blood mononuclear cells to generate TLR9-induced cytokines. We then demonstrated the antiviral effects of both TLR9-induced cytokines and nucleoside analogue (lamivudine) on HBV replication in transduced PMHs. Conclusion: Baculovirus-mediated genome transfer initiated a full HBV replication cycle in PMHs; thus highlighted both the baculovirus efficiency in crossing the species barrier and macaque susceptibility to HBV infection. Moreover, our results demonstrate the relevance of thus system for antiviral compound evaluations with either nucleoside analogues or inhibitory cytokines. Cynomolgus macaques are readily available, are immunologically closely related to humans, and may therefore represent a promising model for the development of new immunotherapeutic strategies. (HEPATOLOGY 2010)
Despite an effective vaccine, hepatitis B virus (HBV) infection remains a major public health problem because more than 350 million people are chronic HBV carriers worldwide and have a greater risk of developing severe liver diseases such as cirrhosis and hepatocellular carcinoma.1 Currently available therapies are restricted to the use of standard/pegylated interferon-alpha (IFN-α) or nucleos(t)ide analogues such as lamivudine, but both are still unsatisfactory. Indeed, despite its immunomodulatory and antiviral properties, IFN-α is effective in less than 30% of chronic carriers, and severe adverse side effects restrict its use. In contrast, nucleos(t)ide analogues are well tolerated and strongly suppress viral replication, but they require indefinite therapy, which leads to the emergence of drug-resistant mutants.2
Several nonprimate animals, such as Pekin ducks and American woodchucks, and their HBV-related viruses are available for studying HBV replication and screening putative antiviral compounds.3, 4 Moreover, Tupaia belangeri (a small mammal related to primates) has been successfully inoculated with human HBV and can develop a transient acute infection. However, the infection is rapidly resolved because of seroconversion to antibody to hepatitis B e antigen and antibody to hepatitis B surface antigen.5
The development of a new animal model susceptible to HBV infection that is closer to humans would be a very useful tool for anti-HBV drug evaluation and for the improvement of anti-HBV therapies. In this respect, monkey species are very attractive, and chimpanzees were the first animals described for their ability to develop an acute HBV infection after the inoculation of sera from HBV-infected patients.6, 7 However, as this species is now protected, chimpanzees no longer represent an appropriate model for an antiviral research program. Some other great apes such as gibbons8 and baboons9 have demonstrated their susceptibility to HBV infection. More recently, we have suggested that macaques could represent a useful new primate model for the study of HBV because we have demonstrated that HBV can successfully replicate in Macaca sylvanus intrahepatically inoculated with an HBV DNA plasmid construct.10
The aims of our work were to confirm in vitro that human HBV can replicate in liver macaque cells and to demonstrate the relevance of macaque models for antiviral therapy evaluations.
Materials and Methods
Cell Preparation and Culture.
Primary hepatocytes were isolated from the livers of cynomolgus macaques (Macaca fascicularis) as previously described.11 Primary macaque hepatocytes (PMHs) were next maintained in William's medium (Invitrogen) supplemented with 10% fetal calf serum (Perbio), 50 U/mL penicillin/streptomycin (Invitrogen), 2 mM GlutaMAX (Invitrogen), 5 μg/mL bovine insulin, 5 × 10−5 M hydrocortisone hemisuccinate (Roche Diagnostics, Boehringer Mannheim), and 1.8% dimethyl sulfoxide (Sigma). Macaque livers were kind gifts from INSERM UMR-S 864 (Bron, France) and commissariat à l'Energic atomique (CEA) (Paris, France).
Peripheral Blood Mononuclear Cell (PBMC) Isolation and In Vitro Stimulation.
PBMCs were separated by centrifugation on Lymphoprep (Abcys, France). Freshly isolated PBMCs were cultured at 5 × 106 cells/mL in Roswell Park Memorial Institute 1640 medium (Life Technologies, France) supplemented with 10% fetal calf serum (Hyclone, VWR, France), 2 mM L-glutamine (Invitrogen), and antibiotics (penicillin/streptomycin; Life Technologies, France). Cells were stimulated for 24 hours with Toll-like receptor 9 (TLR9) ligand at 2 μM [oligodeoxynucleotide (ODN) CpG 2216 or control ODN GpC 2216C]. Supernatants were collected, and the production of IFN-α was determined by human IFN-α enzyme-linked immunosorbent assay (Abcys). Macaque blood from healthy animals was a gift from INSERM UMR-S 864.
Baculovirus Transduction of Mammalian Cells and HBV Intermediate Analyses.
HBV recombinant baculovirus constructions, stock productions, titrations, concentrations, and transductions of mammalian cells were described previously.12 Protein-free viral DNA [covalently closed circular DNA (cccDNA) and baculovirus DNA] was separated from protein-linked viral DNA (relaxed circular, linear, and single-strand intermediates) by potassium chloride precipitation.13 cccDNA detection after rolling circle amplification and analyses of HBV RNA synthesis and hepatitis B surface antigen (HBsAg) secretion were performed as described previously.12, 14, 15 Quantification of intracellular and extracellular HBV DNA was performed with real-time polymerase chain reaction and analyzed by the second derivative maximum method with relative quantification software (LightCycler, Roche Diagnostics) with the following primers: HBV1844 (5′-GTTGCCCGTTTGTCCTCTAATTC-3′) and HBV1745 (5′-GGAGGGATACATAGAGGTTCCTTGA-3′).16, 17
Analysis of HBV Particles.
The supernatant from Bac-HBV–transduced PMHs was clarified by centrifugation for 5 minutes at 5000g and filtered (0.45 μM) before being concentrated 200 times with Centricon 70 columns (Millipore). One-tenth of the concentrated supernatant was subjected to a cesium chloride (CsCl) gradient to analyze the quality of the product as described previously,16, 17 and one-tenth was analyzed by transmission electron microscopy as previously described.12
HBV Replication Can Be Triggered in PMHs by Transduction with HBV Recombinant Baculoviruses.
As we failed to obtain satisfactory HBV replication after direct infection with HBV particles or transfection with an HBV-replicating competent plasmid (data not shown), a baculovirus vector was used to deliver the HBV genome into PMHs. Indeed, we and others have demonstrated the ability of baculovirus vectors to deliver the HBV genome into a high number of cells and trigger the production of huge amounts of HBV proteins, RNA, DNA, and infectious particles without any toxicity.12, 18 A 1.1-genome-length HBV recombinant baculovirus (Bac-HBV-1.1-WT) in which pregenomic RNA expression is under the control of a strong mammalian promoter was used to transduce freshly prepared PMHs at different multiplicities of infection (MOIs). Intermediates of HBV replication were analyzed at various time points post-transduction (PT). HBV RNA and secreted HBsAg were detected 24 hours PT and remained detectable until the end of the experiment on day 9 PT (Fig. 1A,B). HBV DNA replicative intermediates, including cccDNA, were also detected from 24 hours PT until day 9 (Fig. 1C,D). Moreover, levels of HBV replication clearly increased with the MOI. Similar results were obtained with PMHs transduced with Bac-HBV-1.3-WT, in which HBV expression is driven by the natural HBV promoters (data not shown); this suggests that these cells are indeed competent for HBV replication and that host restriction is probably located at the step of viral entry within the cell. Moreover, quantification of intracellular HBV DNA at an MOI of 50 demonstrated that the magnitude of HBV replication in PMHs was 106 HBV DNA copies per well; it reached a peak around day 3 PT and lasted more than 9 days (Fig. 1E). In comparison, the transduction of HepG2 cells was more efficient because intracellular HBV DNA quantification was 2 log higher than that in PMHs (Fig. 1E). Moreover, by using a recombinant baculovirus for green fluorescent protein, we found that the transduction efficiency at an MOI of 50 was clearly higher in HepG2 cells versus PMHs, with more than 70% and 50% of the cells transduced, respectively (data not shown). When we looked at extracellular HBV DNA after PMH transduction, we observed a peak of secretion on day 6 PT with 8 × 105 copies per well (Fig. 1F). Finally, the results also showed that the amount of the intracellular baculovirus genome was maximum on day 1 PT, and it decreased thereafter but was still detectable by Southern blot analysis on day 9 PT (Fig. 1C). The differences in the kinetics of accumulation and clearance of both baculoviral and HBV nucleic acids suggest that the half-life of encapsidated HBV DNA is longer than that of the baculoviral genome, as previously described.12
HBV Particles Are Produced by PMHs Transduced with an HBV Recombinant Baculovirus.
It has previously been shown that HepG2 cells transduced with HBV baculoviruses produce infectious HBV particles.12, 18 The supernatant from HBV baculovirus–transduced PMHs was first characterized by CsCl gradient analyses. Most HBV DNA was found in fractions with a density range of 1.24 and 1.20 g/cm3 (fractions 9-12; Fig. 2A). This indicates that most HBV DNA was released within Dane particles, which have been shown to peak at 1.21 g/cm3 in isopycnic density analysis.18 Electron microscopy analyses confirmed the presence of Dane particles as well as spheres and filamentous particles (Fig. 2B). Altogether, these data showed that HBV-transduced PMHs secreted newly produced HBV particles in 3 typical forms.19 Concentrated supernatant from HBV baculovirus–transduced PMHs was then used to inoculate freshly prepared PMHs or HepaRG cells,20 but none of these cells showed convincing signals of HBV infection within 15 days after inoculation (data not shown).
PMHs Transduced with HBV Recombinant Baculoviruses Are Suitable for Antiviral Compound Evaluation.
The last step was the evaluation of the potentiality of our system for antiviral therapy testing. First, the antiviral activity of lamivudine was tested. Nucleos(t)ide analogues such as lamivudine are able to specifically inhibit HBV viral polymerase activity and thus prevent the secretion of mature HBV particles, whereas other steps of the viral life cycle, such as entry or antigen secretion, are not targeted by such a treatment.21 As expected, HBV DNA secretion (Fig. 3A), but not HBsAg (Fig. 3B), was inhibited by lamivudine treatment in a dose-dependent manner in PMHs transduced with Bac-HBV-1.1-WT; this was demonstrated by residual replication of 49%, 36%, and 31% with drug concentrations of 0.1, 0.5, and 1 μM, respectively.
Second, as human and macaque components of the innate immune system closely resemble each other, human reagents can be used,22 and there is an opportunity to test an immunotherapeutic strategy in particular through the use of the promising TLR9 ligand. Indeed, CpG ODNs are synthetic agonists of TLR9 and potent inducers of innate (IFN-α, IFN-β, IFN-γ, and TNF-α) and adaptive immunity (T helper 1 CD4 and CD8 T cell responses).23 Moreover, we recently demonstrated that CpG-induced cytokines strongly inhibited HBV viral intermediates of replication as well as HBsAg and hepatitis B e antigen secretion from HBV-transduced or HBV-infected cells.24 Thus, to address the adequacy of our system for testing such an antiviral strategy, PBMCs were isolated from two different macaques (named RU and Orion) and stimulated with the CpG ODN (2216) or ODN control (2216C) for 24 hours. Supernatants from stimulated cells, which were shown to contain at least IFN-α (Fig. 4A), were used to treat PMHs transduced with Bac-HBV-1.1-WT. The results clearly showed the potency of such an antiviral effect because intracellular encapsidated HBV DNA and HBsAg secretion was inhibited in a dose-dependent manner, and it decreased even below the detection level at the highest concentration of CpG-induced cytokines (Fig. 4B,C).
In this study, we have demonstrated that transduced PMHs support a complete HBV replication cycle, and we have provided further evidence for the susceptibility of monkey hepatocytes to HBV replication and confirmed our previous in vivo observations after intrahepatic HBV transfection.10 Therefore, a baculoviral delivery system, in comparison with transfection, allows the induction of high rates of HBV replication in PMHs, although both the transduction efficiency and the levels of HBV markers were lower than those observed in the HepG2 cell line.12 The reason that HBV infection in the wild appears to be restricted to apes remains unclear at this stage. Given the highly selective distribution of HBV infection in Old World species and the efficient HBV replication levels obtained through baculovirus delivery, we can hypothesize that the species restriction may be due to the viral receptor specificity at the hepatocyte surface rather than the cell machinery itself. Indeed, HBV replication can be induced in unsusceptible mice with transfection, transduction, or hydrodynamic injection of viral DNA.25 Surprisingly, we were not able to infect de novo PMHs or HepaRG cells with the HBV particles produced after transduction, despite the demonstration of complete viral particle secretion. However, hepatoma cells are not susceptible to HBV infection in vitro, and only a low level of HBV replication can be obtained after the infection of susceptible primary human hepatocytes or HepaRG.26, 27 This lack of infectiosity in vitro may be evaluated by attempts to infect macaques in vivo. Throughout this study, we have demonstrated the antiviral effects of both TLR9-induced cytokines and a nucleoside analogue (lamivudine) on HBV replication in transduced PMHs. This is of major importance because human and macaque immune systems are closely related and so macaques may become an important model for evaluating the efficiency and side effects of immunotherapies. Indeed, as the phylogenic distance between humans and macaques enables the use of human reagents, it provides the opportunity to undertake immune manipulation, particularly through the promising TLR ligands.23 Moreover, we have previously demonstrated that the combination of the TLR9 ligand with nucleoside analogues represents an interesting immunotherapeutic strategy,24 and this may be applied to the macaque model.22 When we consider (1) our previous demonstration that intrahepatic transfection of HBV DNA induces hepatitis in cynomolgus macaques, (2) the present work showing that PMHs support a complete HBV replication cycle associated with the secretion of Dane particles, and (3) our ongoing and future in vivo experiments in cynomolgus macaques evaluating hepatitis induction with either intrahepatic inoculation of Bac-HBV-1.1-WT or inoculation of HBV particles produced in PMHs, we are confident of the possibility of establishing an HBV infection in macaques by serial in vivo passages of virus produced either in vitro (PMHs) or in vivo (serum from animals inoculated with intrahepatic Bac-HBV-1.1-WT). In conclusion, the opportunity to infect macaques in vivo may allow the establishment of a new small primate model for HBV immunobiology and the further development of innovative antiviral strategies.
- 15Rolling circle amplification, a powerful tool for genetic and functional studies of complete hepatitis B virus genomes from low-level infections and for directly probing covalently closed circular DNA. Antimicrob Agents Chemother 2008; 52: 3068-3073., , , , , , et al.