Single- and repeat-dose toxicity of IDX14184, a nucleotide prodrug with antiviral activity for hepatitis C viral infection, in mice, rats, and monkeys.

The single- and repeat-dose toxicity profile of IDX14184, a novel guanosine nucleotide prodrug with antiviral activity against hepatitis C viral infection, was characterized following once daily oral administration for durations up to 13, 26, and 32 weeks in mouse, rat, and cynomolgus monkey, respectively. The heart, liver, kidney, skeletal muscles, and lower gastrointestinal tract (cecum, colon, and/or rectum) were identified as the primary toxicity targets in these nonclinical species. The mouse was relatively insensitive to IDX14184-induced cardiac toxicity and hepatotoxicity. The rat was very sensitive to IDX14184-induced skeletal muscle, liver, heart, and lower gastrointestinal tract toxicity but relatively insensitive to kidney toxicity. The monkey is a good animal species to detect IDX14184-induced toxicity in the cardiac and skeletal muscles, and in the liver and kidney, but not lower gastrointestinal tract toxicity. The toxicity profile of IDX14184 was most appropriately characterized in rats and monkeys. The conduct of a series of cardiac size and function assessments during a non-rodent toxicology study using echocardiography proved great utility in this work. IDX14184 clinical development was eventually terminated due to suboptimal efficacy and regulatory concerns on potential heart and kidney injury in patients, as seen with a different guanosine nucleotide prodrug, BMS-986094.

In vitro activity and resistance profile of samatasvir, a novel NS5A replication inhibitor of hepatitis C virus.

The hepatitis C virus (HCV) nonstructural 5A (NS5A) protein is a clinically validated target for drugs designed to treat chronic HCV infection. This study evaluated the in vitro activity, selectivity, and resistance profile of a novel anti-HCV compound, samatasvir (IDX719), alone and in combination with other antiviral agents. Samatasvir was effective and selective against infectious HCV and replicons, with 50% effective concentrations (EC50s) falling within a tight range of 2 to 24 pM in genotype 1 through 5 replicons and with a 10-fold EC50 shift in the presence of 40% human serum in the genotype 1b replicon. The EC90/EC50 ratio was low (2.6). A 50% cytotoxic concentration (CC50) of >100 µM provided a selectivity index of >5 × 10(7). Resistance selection experiments (with genotype 1a replicons) and testing against replicons bearing site-directed mutations (with genotype 1a and 1b replicons) identified NS5A amino acids 28, 30, 31, 32, and 93 as potential resistance loci, suggesting that samatasvir affects NS5A function. Samatasvir demonstrated an overall additive effect when combined with interferon alfa (IFN-α), ribavirin, representative HCV protease, and nonnucleoside polymerase inhibitors or the nucleotide prodrug IDX184. Samatasvir retained full activity in the presence of HIV and hepatitis B virus (HBV) antivirals and was not cross-resistant with HCV protease, nucleotide, and nonnucleoside polymerase inhibitor classes. Thus, samatasvir is a selective low-picomolar inhibitor of HCV replication in vitro and is a promising candidate for future combination therapies with other direct-acting antiviral drugs in HCV-infected patients.

Monoval-LdC: efficient prodrug of 2'-deoxy-beta-L-cytidine (L-dC), a potent and selective anti-HBV agent.

In order to improve the oral bioavailability of LdC, valinyl esters were prepared as prodrugs. We report here the syntheses of the 3'-mono-, 5'-mono, and 3',5'-di-O-valinyl esters of LdC. The comparison of their ease of synthesis, their physicochemical properties, as well as their pharmacokinetic parameters in cynomologus monkeys has revealed 3'-mono-O-valinyl derivative as the most promising of the studied prodrugs. This compound is being developed as a new anti-HBV agent.

Antiviral beta-L-nucleosides specific for hepatitis B virus infection.

Three simple, related nucleosides, beta-L-2'-deoxycytidine (LdC), beta-Lthymidine (LdT), and beta-L-2'-deoxyadenosine (LdA), have been discovered to be potent, specific and selective inhibitors of the replication hepatitis B virus (HBV), as well as the closely related duck and woodchuck hepatitis viruses (WHV). Structure-activity relationship analysis indicates that the 3'-OH group of the beta-L-2'-deoxyribose of the beta-L-2'-deoxynucleoside confers specific anti-hepadnavirus activity. The simple nucleosides had no effect on the replication of 15 other RNA and DNA viruses, and did not inhibit human DNA polymerases (alpha, beta and gamma) or compromise mitochondrial function. The nucleosides are efficiently converted intracellularly into active triphosphate metabolites that have a long half-life. Once-daily oral administration of these compounds in the woodchuck efficacy model of chronic HBV infection reduced viral load by as much as 10(8) genome equivalents/ml serum and there was no drug-related toxicity. In addition, a decline in WHV surface antigen (WHsAg) paralleled the decrease in viral load. This class of nucleosides displays an excellent overall safety profile. The first compound, LdT, has already entered clinical trials and LdC, currently being developed as a prodrug, is expected to enter the clinic in the near future. These compounds have the potential for use in combination therapy with the goal of achieving superior viral suppression and diminishing the onset of resistance.

Virus-specific cofactor requirement and chimeric hepatitis C virus/GB virus B nonstructural protein 3.

GB virus B (GBV-B) is closely related to hepatitis C virus (HCV) and causes acute hepatitis in tamarins (Saguinus species), making it an attractive surrogate virus for in vivo testing of anti-HCV inhibitors in a small monkey model. It has been reported that the nonstructural protein 3 (NS3) serine protease of GBV-B shares similar substrate specificity with its counterpart in HCV. Authentic proteolytic processing of the HCV polyprotein junctions (NS4A/4B, NS4B/5A, and NS5A/5B) can be accomplished by the GBV-B NS3 protease in an HCV NS4A cofactor-independent fashion. We further characterized the protease activity of a full-length GBV-B NS3 protein and its cofactor requirement using in vitro-translated GBV-B substrates. Cleavages at the NS4A/4B and NS5A/5B junctions were readily detectable only in the presence of a cofactor peptide derived from the central region of GBV-B NS4A. Interestingly, the GBV-B substrates could also be cleaved by the HCV NS3 protease in an HCV NS4A cofactor-dependent manner, supporting the notion that HCV and GBV-B share similar NS3 protease specificity while retaining a virus-specific cofactor requirement. This finding of a strict virus-specific cofactor requirement is consistent with the lack of sequence homology in the NS4A cofactor regions of HCV and GBV-B. The minimum cofactor region that supported GBV-B protease activity was mapped to a central region of GBV-B NS4A (between amino acids Phe22 and Val36) which overlapped with the cofactor region of HCV. Alanine substitution analysis demonstrated that two amino acids, Val27 and Trp31, were essential for the cofactor activity, a finding reminiscent of the two critical residues in the HCV NS4A cofactor, Ile25 and Ile29. A model for the GBV-B NS3 protease domain and NS4A cofactor complex revealed that GBV-B might have developed a similar structural strategy in the activation and regulation of its NS3 protease activity. Finally, a chimeric HCV/GBV-B bifunctional NS3, consisting of an N-terminal HCV protease domain and a C-terminal GBV-B RNA helicase domain, was engineered. Both enzymatic activities were retained by the chimeric protein, which could lead to the development of a chimeric GBV-B virus that depends on HCV protease function.

In vitro inhibition of hepadnavirus polymerases by the triphosphates of BMS-200475 and lobucavir.

The guanosine analogs BMS-200475 and lobucavir have previously been shown to effectively suppress propagation of the human hepatitis B virus (HBV) and woodchuck hepatitis virus (WHV) in 2.2.15 liver cells and in the woodchuck animal model system, respectively. This repression was presumed to occur via inhibition of the viral polymerase (Pol) by the triphosphate (TP) forms of BMS-200475 and lobucavir which are both produced in mammalian cells. To determine the exact mode of action, BMS-200475-TP and lobucavir-TP, along with several other guanosine analog-TPs and lamivudine-TP were tested against the HBV, WHV, and duck hepatitis B virus (DHBV) polymerases in vitro. Estimates of the 50% inhibitory concentrations revealed that BMS-200475-TP and lobucavir-TP inhibited HBV, WHV, and DHBV Pol comparably and were superior to the other nucleoside-TPs tested. More importantly, both analogs blocked the three distinct phases of hepadnaviral replication: priming, reverse transcription, and DNA-dependent DNA synthesis. These data suggest that the modest potency of lobucavir in 2.2.15 cells may be the result of poor phosphorylation in vivo. Kinetic studies revealed that BMS-200475-TP and lobucavir-TP competitively inhibit HBV Pol and WHV Pol with respect to the natural dGTP substrate and that both drugs appear to bind to Pol with very high affinities. Endogenous sequencing reactions conducted in replicative HBV nucleocapsids suggested that BMS-200475-TP and lobucavir-TP are nonobligate chain terminators that stall Pol at sites that are distinct yet characteristically two to three residues downstream from dG incorporation sites.

Efficacy of the carbocyclic 2'-deoxyguanosine nucleoside BMS-200475 in the woodchuck model of hepatitis B virus infection.

Daily oral treatment with the cyclopentyl 2'-deoxyguanosine nucleoside BMS-200475 at doses ranging from 0.02 to 0.5 mg/kg of body weight for 1 to 3 months effectively reduced the level of woodchuck hepatitis virus (WHV) viremia in chronically infected woodchucks as measured by reductions in serum WHV DNA levels and endogenous hepadnaviral polymerase activity. Within 4 weeks of daily therapy with 0.5 or 0.1 mg of BMS-200475 per kg, endogenous viral polymerase levels in serum were reduced about 1,000-fold compared to pretreatment levels. Serum WHV DNA levels determined by a dot blot hybridization technique were comparably decreased in these treated animals. In the 3-month study, the sera of animals that had undetectable levels of WHV DNA by the dot blot technique were further analyzed by a highly sensitive semiquantitative PCR assay. The results indicate that BMS-200475 therapy reduced mean WHV titers by 10(7)- to 10(8)-fold, down to levels as low as 10(2) to 10(3) virions/ml of serum. Southern blot hybridization analysis of liver biopsy samples taken from animals during and after BMS-200475 treatment showed remarkable reductions in the levels of WHV DNA replicative intermediates and in the levels of covalently closed circular viral DNA. WHV viremia in BMS-200475-treated WHV carriers eventually returned to pretreatment levels after therapy was stopped. These results indicate that BMS-200475 should be evaluated in clinical trials for the therapy of chronic human hepatitis B virus infections.

Generation of replication-competent hepatitis B virus nucleocapsids in insect cells.

The double-stranded DNA genome of human hepatitis B virus (HBV) and related hepadnaviruses is reverse transcribed from a pregenomic RNA by a viral polymerase (Pol) harboring both priming and RNA- and DNA-dependent elongation activities. Although hepadnavirus replication occurs inside viral nucleocapsids, or cores, biochemical systems for analyzing this reaction are currently limited to unencapsidated Pols expressed in heterologous systems. Here, we describe cis and trans classes of replicative HBV cores, produced in the recombinant baculovirus system via coexpression of HBV core and Pol proteins from either a single RNA (i.e., in cis) or two distinct RNAs (in trans). Upon isolation from insect cells, cis and trans cores contained Pol-linked HBV minus-strand DNA with 5' ends mapping to the authentic elongation origin DR1 and also plus-strand DNA species. Only trans cores, however, were highly active for the de novo priming and reverse transcription of authentic HBV minus strands in in vitro endogenous polymerase assays. This reaction strictly required HBV Pol but not the epsilon stem-loop element, although the presence of one epsilon, or better, two epsilons, enhanced minus-strand synthesis up to 10-fold. Compared to unencapsidated Pol enzymes, encapsidated Pol appeared to be (i) highly processive, able to extend minus-strand DNAs of 400 nucleotides from DR1 in vitro, and (ii) more active for HBV plus-strand synthesis. These observations suggest possible contributions to the replication process from the HBV core protein. These novel core reagents should facilitate the analysis of HBV replication in its natural environment, the interior of the capsid, and also fuel the development of new anti-HBV drug screens.

Identification of BMS-200475 as a potent and selective inhibitor of hepatitis B virus.

BMS-200475 is a novel carbocyclic 2'-deoxyguanosine analog found to possess potent and selective anti-hepatitis B virus (anti-HBV) activity. BMS-200475 is distinguished from guanosine by replacement of the natural furanose oxygen on the sugar moiety with an exo carbon-carbon double bond. In the HepG2 stably transfected cell line 2.2.15, BMS-200475 had a 50% effective concentration (EC50) of 3.75 nM against HBV, as determined by analysis of secreted HBV DNA. Structurally related compounds with adenine, iodouracil, or thymine base substitutions were significantly less potent or were inactive. Direct comparison of the antiviral activities of BMS-200475 with those of a variety of other nucleoside analogs, including lamivudine (EC50 = 116.26 nM), demonstrated the clearly superior in vitro potency of BMS-200475 in 2.2.15 cells. Intracellular HBV replicative intermediates were uniformly reduced when cells were treated with BMS-200475, but rebounded after treatment was terminated. The concentration of BMS-200475 causing 50% cytotoxicity in 2.2.15 cell cultures was 30 microM, approximately 8,000-fold greater than the concentration required to inhibit HBV replication in the same cell line. Treatment with BMS-200475 resulted in no apparent inhibitory effects on mitochondrial DNA content.

Establishment of an in vitro assay to characterize hepatitis C virus NS3-4A protease trans-processing activity.

An in vitro cleavage system was established to measure HCV NS3 protease trans-processing activity. This system utilizes purified NS3-4A protein from baculovirus, purified substrates expressed by in vitro transcription and translation and defined buffer components. The 41-residue substrates, 5A/5B and 4A/4B, were processed efficiently in trans by wild-type NS3 but not by a catalytically inactive mutant protease; radiolabel sequencing confirmed that NS3-mediated cleavage occurred at the correct cysteine/serine sites, thereby authenticating this system. Two striking features of this in vitro assay are: (1) analogous 4B/5A and 3/4A substrates cannot be processed in trans under the same conditions, and (2) in vitro cleavage of the 5A/5B and 4A/4B sites is highly dependent on the presence of NS4A, which we show is not the case in vivo.

Assembly and antigenicity of hepatitis B virus core particles.

Recent studies in Xenopus oocytes and other systems have led to an understanding of the HBV capsid, or core particle, assembly process. Nascent HBV core polypeptides rapidly dimerize. Accumulation of free dimers to a signature concentration (approximately 0.8 microM) then triggers a highly cooperative capsid assembly reaction. This dimer-to-capsid transition is accompanied by a switch from HBe to HBc antigenicity and appears to be nucleated by interaction between core protein and RNA: deletion of a protamine-like RNA binding domain at the C-terminus of the core protein markedly increases the concentration of dimers needed to drive capsid assembly. The simple assembly pathway seen for HBV capsids mirrors that of R17 bacteriophage.

Ribonucleoprotein complex formation by the human hepatitis B virus polymerase.

Human hepatitis B virus (HBV) polymerase (pol or RT), when expressed in Xenopus oocytes upon injection of synthetic minimal pol RNA (RT RNA), assembles into a higher molecular weight complex with the characteristics of a ribonuclear protein (RNP) complex. In vitro RNA competition binding data suggest that RT RNA is preferentially packaged into this complex even though it lacks the authentic viral encapsidation signal, epsilon, and viral capsid protein sequences. Consistent with this finding, the in vitro polymerase reaction performed in pol-expressing oocyte extracts generates primarily HBV-specific DNAs even when the pol template is challenged with a coinjected non-HBV competitor RNA. These results suggest that interaction between pol and its cognate RNA can be mediated by sequences other than the known packaging elements. We speculate that HBV RNP complexes containing at least polymerase and viral RNA may play a role in viral nucleocapsid assembly and may help to segregate HBV reverse transcription from the cellular milieu in vivo.

A protease-sensitive hinge linking the two domains of the hepatitis B virus core protein is exposed on the viral capsid surface.

Core particles of hepatitis B virus are assembled from dimers of a single 185-residue (subtype adw) viral capsid or core protein (p21.5) which possesses two distinct domains: residues 1 to 144 form a minimal capsid assembly domain, and the arginine-rich, carboxyl-terminal residues 150 to 185 form a protamine-like domain that mediates nucleic acid binding. Little is known about the topography of the p21.5 polypeptide within either the p21.5 capsids or dimers. Here, using site-specific proteases and monoclonal antibodies, we have defined the accessibility of p21.5 residues in dimers and capsids assembled from wild-type and mutant hepatitis B virus core proteins in Xenopus oocytes and in vitro. The data reveal the protamine region to be accessible to external reagents in p21.5 dimers but largely cryptic in wild-type capsids. Strikingly, in capsids the only protease target region was a 9-residue peptide covering p21.5 residues Glu-145 to Asp-153, which falls largely between the two core protein domains. By analogy with protease-sensitive interdomain regions in other proteins, we propose that this peptide constitutes a hinge between the assembly and nucleic acid binding domains of p21.5. We further found that deletion or replacement of the terminal Cys-185 residue greatly increased surface exposure of the protamine tails in capsids, suggesting that a known disulfide linkage involving this residue tethers the protamine region inside the core particles. We propose that disruption of this disulfide linkage allows the protamine region to appear transiently on the surface of the core particle.

Phenotypic mixing between different hepadnavirus nucleocapsid proteins reveals C protein dimerization to be cis preferential.

Hepadnaviruses encode a single core (C) protein which assembles into a nucleocapsid containing the polymerase (P) protein and pregenomic RNA during viral replication in hepatocytes. We examined the ability of heterologous hepadnavirus C proteins to cross-oligomerize. Using a two-hybrid assay in HepG2 cells, we observed cross-oligomerization among the core proteins from hepatitis B virus (HBV), woodchuck hepatitis virus, and ground squirrel hepatitis virus. When expressed in Xenopus oocytes, in which hepadnavirus C proteins form capsids, the C polypeptides from woodchuck hepatitis virus and ground squirrel hepatitis virus, but not duck hepatitis B virus, can efficiently coassemble with an epitope-tagged HBV core polypeptide to form mixed capsids. However, when two different core mRNAs are coexpressed in oocytes the core monomers show a strong preference for forming homodimers rather than heterodimers. This holds true even for coexpression of two HBV C proteins differing only by an epitope tag, suggesting that core monomers are not free to diffuse and associate with other monomers. Thus, mixed capsids result from aggregation of different species of homodimers.

Stability governs the apparent expression of "particulate" hepatitis B e antigen by mutant hepatitis B virus core particles.

The p21.5 capsid or core protein of hepatitis B virus carries two distinct classes of epitopes. Core (HBc) epitopes are found exclusively on the surface of the 28-nm viral icosahedral capsids or core particles, while HBe epitopes are normally expressed only by subparticulate forms of the core protein. Recent studies have suggested that a "particulate" form of HBe is expressed on the surface of capsid particles assembled from p17, a truncated core protein that lacks the carboxy-terminal protamine-like region of p21.5 and hence the ability to bind and encapsidate RNA. In this report we have used epitope-specific ELISAs in conjunction with capsids assembled from a series of carboxy-terminally truncated core proteins to address the mechanistic basis for particulate HBe. Specifically, we sought to test the idea that particulate HBe expression might be linked to the loss of RNA binding. However, our results strongly suggest that expression of HBe by mutant core particles is a result of their intrinsic instability which increases sharply when RNA binding is lost. We show that core particles assembled from mutant core proteins lacking Cys residues also express HBe, again because of capsid instability. We report mild conditions that can induce the dissociation of the mutant capsids and discuss our findings in terms of the factors that control capsid stability.

Recombinant human hepatitis B virus reverse transcriptase is active in the absence of the nucleocapsid or the viral replication origin, DR1.

The double-stranded DNA genome of hepatitis B virus (HBV) is reverse transcribed from the viral pregenome RNA template by a virally encoded reverse transcriptase enzyme (RT) that possesses both priming and elongation activities. Prior efforts have failed to express an active form of HBV RT outside the nucleocapsid in animal cells or to release it from viral nucleocapsids, thus restricting the characterization of this important enzyme. Here, we have engineered epitope-tagged HBV RT proteins and expressed them in Xenopus oocytes via a synthetic RT mRNA which does not include the viral capsid protein or the known initiation site for viral DNA synthesis, DR1. We demonstrate the production of an immunoprecipitable 96-kDa HBV RT protein and show, using a simple in vitro RT assay, that oocyte lysates containing this protein possess an activity that (i) catalyzes an RNA-dependent deoxynucleotide triphosphate polymerization reaction by using an as-yet-unidentified RNA template and (ii) is sensitive to the RT inhibitors actinomycin D and phosphonoformate. Experiments with the chain terminator ddATP suggest that a significant amount of chain elongation occurs in our in vitro reaction. Electrophoretic analysis reveals a heterogeneous array of RT reaction products with sizes ranging from about 100 bases to far larger than that of the input RT mRNA. These products appear to contain covalently bound protein, consistent with the notion that the RT protein may have primed their synthesis. We conclude that HBV RT activity can be uncoupled from both the nucleocapsid and the replication origin, DR1. Our results raise the possibility that unless HBV employs novel mechanisms to regulate its constitutively active RT, cellular RNAs may be reverse transcribed during HBV infection, with potential implications for the development of HBV-related liver cancer. The use of the oocyte system should facilitate studies of HBV RT, including the development of HBV RT inhibitors for antiviral therapy.

A micromolar pool of antigenically distinct precursors is required to initiate cooperative assembly of hepatitis B virus capsids in Xenopus oocytes.

Assembly of hepatitis B virus capsid-like (core) particles occurs efficiently in a variety of heterologous systems via aggregation of approximately 180 molecules of a single 21.5-kDa core protein (p21.5), resulting in an icosahedral capsid structure with T = 3 symmetry. Recent studies on the assembly of hepatitis B virus core particles in Xenopus oocytes suggested that dimers of p21.5 represent the major building block from which capsids are generated. Here we determined the concentration dependence of this assembly process. By injecting serially diluted synthetic p21.5 mRNA into Xenopus oocytes, we expressed different levels of intracellular p21.5 and monitored the production of p21.5 dimers and capsids by radiolabeling and immunoprecipitation, by radioimmunoassay, or by quantitative enzyme-linked immunosorbent assay analysis. The data revealed that (i) p21.5 dimers and capsids are antigenically distinct, (ii) capsid assembly is a highly cooperative and concentration-dependent process, and (iii) p21.5 must accumulate to a signature concentration of approximately 0.7 to 0.8 microM before capsid assembly initiates. This assembly process is strikingly similar to the assembly of RNA bacteriophage R17 as defined by in vitro studies.

Hepatitis B virus capsid particles are assembled from core-protein dimer precursors.

Our studies on the assembly of hepatitis B virus capsids or core particles in Xenopus oocytes have demonstrated that unassembled p21.5 core proteins ("free p21.5") provide a pool of low-molecular-mass precursors for core-particle assembly. Here we have characterized this material. Free p21.5 sedimented through gradients of 3-25% sucrose (wt/vol) as a single protein species of approximately 40 kDa, corresponding to a p21.5 dimer. On nonreducing SDS/polyacrylamide gels, free p21.5 migrated as disulfide-linked p21.5 dimeric species of 35 and 37 kDa. Truncated core proteins lacking most or all of the 36-amino acid protamine region at the p21.5 carboxyl terminus were also found to behave as disulfide-linked dimers with appropriately reduced molecular masses. Our experiments failed to reveal monomeric core proteins or stable intermediates between dimers and capsids along the assembly pathway. We conclude that hepatitis B virus core particles are most likely assembled by aggregating 90 (or possibly 180) disulfide-linked p21.5 dimers. We discuss similarities between the assembly of hepatitis B virus capsids and simple T = 3 plant virus and bacteriophage structures.

RNA- and DNA-binding activities in hepatitis B virus capsid protein: a model for their roles in viral replication.

The hepatitis B virus capsid or core protein (p21.5) binds nucleic acid through a carboxy-terminal protamine region that contains nucleic acid-binding motifs organized into four repeats (I to IV). Using carboxy-terminally truncated proteins expressed in Escherichia coli, we detected both RNA- and DNA-binding activities within the repeats. RNA-binding and packaging activity, assessed by resolving purified E. coli capsids on agarose gels and disclosing their RNA content with ethidium bromide, required only the proximal repeat I (RRRDRGRS). Strikingly, a mutant in which four Arg residues replaced repeat I was competent to package RNA, demonstrating that Arg residues drive RNA binding. In contrast, probing immobilized core proteins with 32P-nucleic acid revealed an activity which (i) required more of the protamine region (repeats I and II), (ii) appeared to bind DNA better than RNA, and (iii) was apparently modulated by phosphorylation in p21.5 derived from Xenopus oocytes. Deletion analysis suggested that this activity may depend on an SPXX-type DNA-binding motif in repeat II. Similar motifs found in repeats III and IV may also function to bind DNA. On the basis of these observations, together with a reinterpretation of recent studies showing that capsid protein mutants cause defects in viral genome replication, we propose a model suggesting that hepadnavirus capsid proteins participate directly in the intracapsid reverse transcription of RNA into DNA. In this model, repeat I binds RNA whereas the distal repeats are progressively recruited to bind elongating DNA strands. The latter motifs may be required for replication to be energetically feasible.