Introduction
Up to 10% of the estimated 40 million individuals infected with HIV type 1 worldwide are co-infected with hepatitis B virus (HBV) [1–3] . In the era of highly active antiretroviral therapy (HAART), liver failure has emerged as a major cause of death in HIV/HBV co-infected individuals, who are at increased risk of liver-related mortality and cirrhosis compared to HIV or HBV mono-infected individuals [4,5] .
The reason(s) for increased progression of liver disease in HIV/HBV co-infected individuals are unclear. The host immune response is an important mediator of liver disease in chronic hepatitis B, but this cannot explain the more aggressive course seen in many HIV/HBV co-infected patients since their immune systems are suppressed. In patients receiving HAART, however, it is possible that restoration of the immune response is a factor in liver disease progression, although this is yet to be confirmed. In general, the HBV is not cytopathic; however, in immunosuppressed patients, including HIV-infected individuals, a more fulminant form of HBV has been described, fibrosing cholestatic hepatitis (FCH) [6,7] . In FCH, the histology reveals ballooning degeneration and very high intracellular levels of viral proteins and nucleic acids consistent with the notion of direct cytolysis due to the HBV itself.
Liver disease progression in individuals chronically infected with HBV is associated with the accumulation of numerous mutant viral genomes, in particular viruses with deletions in the core/precore and envelope genes [8–12] . Deletions in the HBV core and envelope genes have also been identified in HBV-mono-infected individuals on immunosuppressive drug therapy, such as renal and liver transplant recipients who have progressive liver disease [10,13,14] . Since many of these deletions abrogate translation of the HBV core protein, the absence of viral nucleocapsid rendered these mutants incapable of autonomous replication in vitro . Replication of the mutant viruses was, however, rescued when the HBV core protein was supplied in trans [13,14] .
In HIV/HBV co-infected individuals, increased HBV replication and lower rates of HBeAg seroconversion compared to HBV mono-infected individuals [15–17] suggests that (1) HIV infection alters the cell micro-environment to favour HBV replication and/or (2) the HBV genome may differ in HIV co-infected individuals. To date, sequence analysis of the HBV genome in HIV co-infected individuals has been limited to the polymerase gene, the overlapping envelope gene, and part of the basal core promoter [18,19] and has not included the HBV core, precore, or the complete HBx genes. The main aims of this study were to (1) sequence the complete HBV genomes from HIV co-infected individuals compared with HBV mono-infected individuals and (2) analyse in more detail novel mutations that were more common in HIV/HBV co-infected individuals.
Materials and methods
Cohort 1: full genome analysis
To enable complete HBV genome analysis, serum was initially obtained from 10 HIV-infected individuals (P1 to P10) from Australia and the USA who were chronically infected with HBV genotype A, the dominant genotype in HIV/HBV co-infected individuals in the USA and Australia [18] . All individuals were HBeAg positive and were chosen from a previously characterized and recently described cohort of HIV/HBV co-infected individuals [18] on the basis of serum availability and co-infection with HBV genotype A. Six individuals were from the USA (MACS cohort; P4 to P9), three from Sydney (P1, P2, P10) and one from Melbourne (P3). Serum samples were obtained [a pre-lamivudine sample was not available for patient P1] prior to antiviral therapy and again during lamivudine (LMV) therapy. HBV encoding LMV-resistance mutations had emerged in all patients (previously described by [18] ). To enable comparison of the HBV genomes from HIV/HBV co-infected and HBV mono-infected individuals, serum samples were also obtained from five randomly selected Australians with chronic HBV genotype A mono-infection (P11 to P15), prior to and during LMV therapy. All serum samples were negative for hepatitis C virus and hepatitis delta virus and all participants were male. Clinical information including HIV RNA and HBV DNA levels, alanine aminotransferase (ALT) levels, CD4 cell counts and HBeAg status was obtained using the methods described in [20] (Table 1 ). The complete antiretroviral drug history was obtained for seven of the ten patients.
Table 1: Clinical data prior to (pre) and during (post) lamivudine (LMV) therapy.
DNA extraction and hepatitis B virus genomic-length polymerase chain reaction
DNA was extracted from serum as described previously [20] and genomic-length HBV DNA was amplified by PCR with the PicoMaxx high fidelity PCR amplification system (Stratgene, La Jolla, California, USA) using the amplification primers and conditions described in [21] .
Sequence analysis
Genomic-length amplicons from patients in cohort 1 were sequenced using the ABI platform (MicroMon, Monash University) [22] and contiguous sequences were generated using the SeqScape program (ABI, Scoresby, Victoria, Australia). Mutations in HBV genes and regulatory regions were initially identified by comparing all sequences to a genotype A reference sequence (Acc. No. X02763) using the SeqHepB program [23,24] . The sequences were subsequently compared to HBV genomic length sequences obtained from 5 HBV-mono-infected individuals infected with HBV genotype A (P11 to P15) and nine additional genotype A sequences obtained from GenBank (Accession numbers A98547760, A00142825, A00508092, A03529306, A03536255, A055078434, A93571959 and A98154208). For phylogenetic analysis, genomic length nucleotide sequences and the deduced amino acid sequences of the HBV polymerase gene was aligned with HBV genotype A reference sequences from subtypes A1, A2 and A3 (accession numbers X02763, X70185, Z72478, AY128092, AJ012207 and M57663) using ClustalX [25] and bootstrapped.
Neighbour-joining phylogenetic trees were constructed using the Phylip Seqboot, DNADist/ProtDist, Neighbour and Consense programs (http://evolution.gs.washington.edu/phylip.html ). Maximum-likelihood analysis was performed using BioEdit [26] and trees were generated using Treeview [27] . HBV sequence variability in HIV/HBV co-infected and HBV mono-infected individuals was compared using Mann–Whitney non-parametric analysis.
Clonal analysis of mutations in cohort 1
To determine the frequency of a novel precore/core deletion mutation (termed –1G) in the HIV/HBV co-infected individuals and HBV mono-infected individuals from cohort 1, the region encompassing the mutation was amplified by PCR, cloned and sequenced. PCR was performed using either PicoMaxx high fidelity PCR (Stratagene) or Expand Hi Fi (Roche, Castle Hill, New South Wales, Australia) polymerase mixes and ten clones were sequenced from each patient.
Cohort 2: precore/core gene analysis
To determine the abundance of the –1G deletion mutation in other HIV/HBV co-infected individuals, HBV DNA sequences encompassing the mutation were amplified from an additional 39 HIV/HBV co-infected and 57 HBV mono-infected individuals infected with HBV genotype A and analysed by direct sequencing. The genotype of all HBV isolates was determined using the SeqHepB program [23] . Fifteen of the HIV/HBV co-infected individuals were from the cohort described in [18] and 24 were from a previously uncharacterized group of HIV/HBV co-infected individuals who were selected on the basis of HBsAg positivity at baseline, as well as being PCR negative for hepatitis C virus and hepatitis delta virus. The HBV mono-infected individuals were all from Australia and samples were obtained prior to and during LMV treatment where available. The HBV DNA concentration was determined using the methods described in [20] . The association between identification of the –1G mutation and HBV DNA concentration in patients within cohorts 1 and 2 was investigated using Mann–Whitney nonparametric analysis.
Results
Hepatitis B virus sequence variability (cohort 1)
Analysis of genomic-length nucleotide sequences amplified from ten HBeAg positive HIV co-infected individuals infected with HBV genotype A (P1 to P10) identified a maximum of 1.3% nucleotide variability among all sequences prior to LMV therapy and 1.4% variability following between 9 and 74 months therapy (data not shown). HBV sequences from individual patients only varied by a maximum of 0.5% in pre and post LMV therapy samples. The sequences from five HBV-mono-infected individuals (P11 to P15) varied by up to 3.7%, with a maximum of 0.8% variability in HBV sequences from individual patients following between 34 and 47 months LMV therapy. Although the sample sizes were small, the difference in sequence variability between the ten HIV/HBV co-infected and the five HBV mono-infected individuals was statistically significant (P < 0.002).
Phylogenetic analysis of the deduced amino acid sequences of the polymerase gene showed that all HBV sequences amplified from HIV/HBV co-infected (Fig. 1 ) and HBV mono-infected individuals (results not shown) grouped within the A2 (Ae, European/North American) subtype of HBV genotype A, with the observed topology receiving strong bootstrap support. The sequences from HIV/HBV co-infected and mono-infected individuals were distributed throughout the A2 clade and there was no significant clustering of either group. Similar tree topologies were obtained using neighbour-joining and maximum likelihood algorithms, indicating the observed topologies were valid (results not shown).
Fig. 1: Unrooted neighbour-joining phylogenetic tree of the deduced HBV polymerase amino acid sequences amplified from ten HIV co-infected individuals (A to J) and 8 HBV genotype A reference sequences. The sequences grouped into the A1 and A3 subtypes respectively are circled, with sequences in subtype A2 indicated by the solid line on the right and side of the cladeogram. The accession numbers of the reference sequences are indicated. Bootstrap values (1000 replicates) at the major nodes are indicated.
Identification of a novel hepatitis B virus core/precore mutation (cohort 1)
Despite the high level of nucleotide sequence conservation, a number of important mutations were identified in HBV coding sequences and regulatory elements. Analysis of genomic-length HBV sequences from ten HIV/HBV co-infected individuals and five HBV mono-infected individuals identified a –1G deletion in the HBV precore and core genes of six HIV/HBV co-infected individuals and three HBV mono-infected individuals respectively (Table 2 ). All individuals harbouring the –1G mutation were HBeAg positive. The –1G deletion was detected in the six HIV/HBV co-infected individuals during LMV therapy and HAART and was also detected in four of these individuals prior to LMV therapy and HAART (Table 2 ). Although the presence of the –1G mutation was not associated with HIV RNA levels, CD4 cell counts, or ALT levels (result not shown), this clinical information was not available for all patients and the sample sizes were too small to permit statistical analysis. The –1G mutation was, however, associated with high HBV viral DNA concentrations (P = 0.002, see later). Of the five HBV mono-infected individuals selected for genomic-length analysis, direct sequencing showed that the –1G mutation was present at low levels in three individuals prior to LMV therapy and was not present in any individuals following LMV treatment (Table 2 ). Of these three individuals, one of them (P13) was anti-HBc negative, another had severe liver disease that later required a liver transplant (P14), and the disease status of the third individual (P15) was unknown. The amplification of sequences harbouring the –1G mutation with two different polymerase/proof-reading enzymes confirmed the mutation was not an artifact of the PCR itself (data not shown).
Table 2: Detection of –1G mutation by direct sequencing and cloning prior to (pre) and during (post) lamivudine (LMV) therapy or HAART.
The –1G deletion was located in a homopolymeric string of guanosine (G) nucleotides between HBV core nucleotides 184 and 190 (core codons 63 and 64) and introduced a frameshift that terminated the coding sequence of the HBV core and precore genes (Fig. 2 ). This mutation was located in a major CD4 T-cell epitope (Fig. 2 ) and the molecular weight of the deduced truncated core gene product was 7.5 kDa, compared to 21 kDa for the full-length core protein. Similarly, the mutation also truncated the HBV precore protein. The HBV precore protein (p25) is a 25 kDA protein that is subsequently processed into 22 kDa (p22) and 17 kDa (HBeAg) proteins, the latter being secreted from the cell. Premature termination of p25 and p22 as a result of the –1G mutation would result in cellular proteins of 10.5 kDa and 8.4 kDa, respectively. The N-terminal sequences of P22 and HBeAg are identical, suggesting that any secreted truncated peptide resulting from the –1G mutation would also be 8.4 kDa in size; however, the absence of C-terminal precore sequences important for secretion [28] suggests it is unlikely the truncated precore protein would be secreted, although this is yet to be determined experimentally.
Fig. 2: The location of the –1G deletion in the deduced HBV precore and core genes. The top panel shows location of homopolymeric G sequence in relation to characterized T cell and B cell epitopes
[15–17] and the deduced gene products encoded by the HBV precore and core genes. The lower panel shows the deduced truncated precore and core gene products that would result from the –1G deletion.
Clonal analysis of –1G mutation in cohort 1
To determine the abundance of the –1G mutation relative to wild-type HBV sequence on individual virus genomes, the region was amplified by PCR and cloned, using serum-derived DNA extracted prior to and/or during lamividuine therapy. The –1G deletion was detected in clones from four of six HIV/HBV co-infected individuals prior to LMV therapy and HAART and all six individuals post LMV therapy and HAART. The –1G mutation was the dominant sequence relative to wild-type HBV in four HIV/HBV co-infected individuals, with 50 to 90% of clones harbouring the –1G mutation (Table 2 ). Clones harbouring wild-type HBV sequence were, however, always detected but usually as a minor population with the dominant –1G deletion population. No clones harbouring the –1G mutation were obtained if the mutation was not visible by direct sequencing. Although the –1G mutation was observed at low levels by direct sequencing in three HBV mono-infected individuals prior to LMV treatment, clones harbouring the mutation were only obtained from two patients. The mutation was detected in one clone from one HBV mono-infected individual and four clones from the other. Cloned DNA from the third HBV mono-infected individual did not harbour the –1G deletion, however some clones had either an eight nucleotide deletion (10%, one clone) or a 26 nucleotide insertion in the BCP (30%, three clones) that did not disrupt the HBV core gene open reading frame (data not shown).
Prevalence of the –1G mutation in additional HIV/hepatitis B virus co-infected persons (cohort 2)
To determine the abundance of the –1G mutation in additional HIV/HBV co-infected and HBV mono-infected individuals, direct sequencing of HBV genomes was performed on HBV sequences from an additional 39 HIV/HBV co-infected individuals and 57 HBV mono-infected individuals infected with HBV genotype A. The mutation was not detected in any of the additional 57 HBV mono-infected individuals. Overall, including the ten HIV/HBV co-infected and five HBV mono-infected patients analysed by genomic-length sequencing and cloning from cohort 1 (P1 to P15), the –1G deletion was detected in 34% of samples from HIV/HBV co-infected individuals and 5% of samples from HBV mono-infected individuals (P < 0.001, data not shown). In the 49 HIV/HBV co-infected individuals, the –1G mutation was detected in approximately 20% of (5/26) sequences prior to LMV therapy and 40% (20/48) of sequences during lamivudine therapy. In 62 HBV mono-infected individuals, the –1G mutation was only detected in three patients prior to LMV treatment and was not detected during LMV therapy.
Association of the –1G mutation with hepatitis B virus DNA concentration
Mann–Whitney rank sum analysis comparing the DNA concentrations in 58 samples from persons in cohorts 1 and 2 infected with –1G mutant or wild-type HBV, showed that the HBV DNA concentration was higher in patients harbouring the –1G mutation than in patients infected with wild-type HBV alone (P = 0.002) (Table 3 ). The –1G mutation was predominantly detected in individuals with a HBV DNA concentration of ≥1 × 106 IU/ml (16/22 individuals), with a maximum DNA concentration of 1 × 1010 IU/ml, although it was also detected in one individual with a HBV DNA concentration of 6.7 × 104 IU/ml and two individuals with a HBV DNA concentration of ≥ 4.5 × 105 IU/ml (not shown). In persons without the –1G mutant HBV, HBV DNA concentrations ranged from 8 × 102 to 1.7 × 1010 IU/ml (not shown).
Table 3: Mann–Whitney non-parametric analysis of hepatitis B virus (HBV) DNA concentration compared with detection of the –1G mutant.
Additional hepatitis B virus mutations identified by genomic length sequencing
In addition to previously described mutations in the HBV basal core promoter and polymerase/envelope genes in HIV/HBV co-infected individuals [18] (data not shown), genomic-length sequencing of ten HIV/HBV co-infected individuals from cohort 1 identified additional mutations in the HBV genome (Table 4 ). A number of these mutations were detected exclusively in the HIV/HBV co-infected individuals and included:
Table 4: Hepatitis B virus (HBV) mutations that emerged during lamivudine (LMV) therapy in 10 HIV individuals (P1 to P10) co-infected with HBV genotype A.
Core An cL60I T-cell epitope mutation emerged during LMV therapy in the HBV core gene of one HIV/HBV co-infected individual and an cE77Q mutation emerged in two others. These mutations were also observed in HBV mono-infected individuals following LMV therapy, however a cI59V core gene mutation was detected exclusively in two HIV/HBV co-infected individuals prior to and during LMV treatment. The cI59V and cL60I mutations were located in the CD4 T-cell epitope that harboured the –1G deletion (Fig. 2 ).
HBV polymerase, HBx and regulatory sequences . An RNaseH rhT119S polymerase mutation detected in one HIV/HBV co-infected individual (P2) also altered HBx at xD48E. A number of mutations were also identified in key HBV regulatory elements, including the basal core promoter (BCP), preS1 promoter, and the X promoter. BCP A1762T/G1764A mutations were detected prior to LMV in one HIV/HBV co-infected individual (P6) and four HBV mono-infected individuals (P11, P12, P14, P15) and following LMV therapy these mutations were detected in two HIV/HBV co-infected individuals (P2, P6) and three HBV mono-infected individuals (P12, P14, P15). HBV encoding G1786T, C1788G, and A1792T BCP mutations emerged during LMV therapy in one individual (P4) and these mutations also altered HBx at xR138M, xH139G and xK140I respectively.
The possible role of LMV or other antiretroviral drugs in the emergence of these mutations requires additional investigation.
Discussion
We have identified a novel –1G mutation in the HBV core and precore genes that was more abundant in HIV/HBV co-infected individuals than patients with HBV mono-infection. The –1G mutant HBV was always detected in the presence of wild-type HBV and was the dominant virus in some HIV/HBV co-infected individuals. The –1G mutation was associated with high HBV DNA concentrations, suggesting it may have a high replication phenotype similar to HBV core mutants identified in some immunosuppressed HBV mono-infected individuals [13,14] . This is yet to be determined experimentally. Phylogenetic analysis showed that all HBV sequences from HIV/HBV co-infected and HBV mono-infected individuals from Australia and the USA were in the A2/Ae subgenotype. This was not surprising as the A2 subgenotype is prevalent in the USA and Europe and the A1 and A3 HBV genotype A subtypes are not common in Australia, and are most common in southern Asia/sub-Saharan Africa and West Africa respectively [29–31] . Although common in HIV/HBV co-infected individuals [18] , HBV genotype A is the least common genotype associated with HBV mono-infection in Australia [32] . The high degree of HBV sequence conservation in the HIV/HBV co-infected individuals relative to HBV mono-infected individuals was, however, unexpected. Sequences varied by a maximum of only 1.3% across the HBV genome, irrespective of geographic location (Australia or the USA), and duration of LMV treatment. This high degree of sequence conservation may reflect the relatively recent acquisition of HBV by HIV-infected individuals, although further studies are required with additional genomic-length HBV sequences to determine if this degree of HBV sequence conservation is typical for HIV/HBV co-infected individuals.
The –1G deletion mutation resulted in premature termination of the deduced HBV precore and core genes and from all the samples analysed by direct sequencing from the different cohorts, was present in 34% of HIV/HBV co-infected individuals and 5% of HBV mono-infected individuals. This mutation leads to a premature stop codon at amino acid 64 that could alter HBV pathogenesis through one of two mechanisms. First, this altered protein may lead to differences in virulence of the virus, as the absence of a complete core gene means the virus could only replicate in the presence of wild-type HBV. This is analogous to core deletion mutant viruses detected in renal transplant recipients, which are replication defective, but can be ‘rescued’ by supplying the HBV core protein in trans [13,14] . Experiments are in progress to determine the replication phenotype of the –1 G deletion mutant. Second, the protein is truncated at a major CD4 T-cell epitope identified between amino acids 50 and 69 (aa50–69) [33] . Mutations in T-cell epitopes have previously been shown to influence the T-cell-mediated immune response to HBV infection and be associated with HBV immune escape and viral persistence. Interestingly, an cI59V aa50-69 major CD4 T-cell epitope HBV mutation was also identified in two HIV/HBV co-infected individuals (P2 and P9) and an cL60I major CD4 T-cell epitope HBV mutation was identified in one HIV/HBV co-infected individual (P8) and one HBV mono-infected individual (P12, data not shown). The affect of these and the –1G mutation on the T-cell-mediated immune response is presently unknown.
The –1G mutation was identified in 5% (three patients) of HBV mono-infected individuals, although the mutation was only detected at low levels prior to LMV therapy. Although we detected the –1G mutation prior to and during LMV therapy in HIV/HBV co-infected individuals and prior to therapy in HBV mono-infected individuals, the small sample size precluded statistical analysis of any possible role for LMV in selection of this mutation. Mutations in homopolymer nucleotide stretches have been identified previously in T4 bacteriophage, polyomavirus and drug resistant isolates of herpes simplex virus (HSV) [34–37] . The detection of the –1G mutation prior to LMV therapy in approximately 30% of individuals with the mutation, however, suggests that it is unlikely to be related to drug resistance and thus it would be more likely to be associated with immune-escape. The presence of the –1G mutation was also unrelated to treatment with other antiretroviral drugs, including nucleoside/tide analogues, non-nucleoside analogues and protease inhibitors. Cloning showed that the –1G mutation was always present as a minor quasispecies in HBV mono-infected individuals, whereas in HIV/HBV co-infected individuals the mutation was detected prior to and during LMV treatment and cloning showed that the mutant genome often dominated the quasi-species pool. Other HBV core deletion mutants have been identified in immunosuppressed HBV mono-infected individuals, such as renal transplant recipients [8–10] , although the –1G mutation has not been reported previously.
The HBV precore gene is the precursor of the secreted HBeAg [38,39] and mutations in the HBV basal core promoter and precore gene that reduce or eliminate HBeAg are associated with progressive liver disease [40] . Although the –1G mutation abrogates precore synthesis, all HIV/HBV co-infected and HBV mono-infected individuals infected with HBV harbouring the –1G mutation in the current study were, however, HBeAg positive. This was most likely because wild-type HBV was at least 10% (one clone) of the quasi-species pool, which would appear to be sufficient to produce enough precore protein to detect HBeAg in circulation. Although all of the patients in the current study with the –1G mutation were HBeAg positive, it remains to be determined if the levels of HBeAg were lower in individuals infected with –1G mutant HBV, as the currently available HBeAg assays do not reliably or conveniently permit quantification.
In addition to mutations detected in the HBV precore/core genes, mutations also emerged in the HBx, envelope and the polymerase genes of some HIV/HBV co-infected individuals. A number of mutations also emerged in key HBV regulatory regions, including in the transcription regulators of the HBx, core and envelope genes. One individual harboured three BCP mutations at position 1786, 1788 and 1792 that altered the overlapping HBx gene and a second patient harboured an RNaseH mutation that altered the overlapping HBx promoter sequence. HBx is a modest transactivating protein associated with liver disease progression and hepatocellular carcinoma, although its role in HBV replication and pathogenesis remains contentious (reviewed in [41] ). It remains to be determined whether the mutations we observed in the HBx gene and regulatory sequences play any role in disease progression in HIV/HBV co-infected individuals.
The identification of HBV encoding novel precore/core deletion mutants, as well as the emergence of drug resistant viruses [18,19] , in HIV/HBV co-infected individuals may provide one explanation for the increased liver disease observed in these patients. The –1G deletion mutants may be able to replicate to high levels under reduced immunological pressure, in the presence of small amounts of wild-type virus, although this is yet to be shown experimentally. Reduced immunological pressure may not, however, be the only explanation for the emergence of these mutations, as we observed –1G mutant sequences in co-infected individuals prior to and during HAART, including individuals with CD4 cell reconstitution. Experiments are also planned to analyse the T-cell and TLR-2 mediated response to the –1G mutant precore/core proteins. We are also characterizing the life cycle of the HBV –1G deletion mutants in vitro and quantifying the HBeAg status in HIV/HBV co-infected and HBV mono-infected individuals. This will tell us whether the –1G mutant has a high replication phenotype, similar to viruses encoding the precore G1896A stop-codon mutuation [42] and/or whether it is retained within the hepatocyte. Together, these studies will assist us to elucidate the role of the –1G mutation in liver disease progression in HIV-/HBV co-infected individuals.
Acknowledgements
This work was supported by NIH grant no AIO60449. We thank Tim Shaw for assistance with the statistical analysis.
This work was supported by NIH grant no AIO60449 (C.L.T., P.A.R. and S.A.L.). Some of the data presented in this manuscript was collected by the Multicenter AIDS Cohort Study (MACS), USA, funded by the National Institute of Allergy and Infectious Diseases, the National Cancer Institute and the National Heart, Lung and Blood Institute.
References
1. Alter MJ. Epidemiology of viral hepatitis and HIV co-infection. J Hepatol 2006; 44:S6–S9.
2. Cooley L, Sasadeusz J. Clinical and virological aspects of hepatitis B co-infection in individuals infected with human immunodeficiency virus type-1. J Clin Virol 2003; 26:185–193.
3. Lincoln D, Petoumenos K, Dore GJ. HIV/HBV and HIV/HCV coinfection, and outcomes following highly active antiretroviral therapy. HIV Med 2003; 4:241–249.
4. Thio CL, Seaberg EC, Skolasky R Jr, Phair J, Visscher B, Munoz A, Thomas DL. HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter Cohort Study (MACS). Lancet 2002; 360:1921–1926.
5. Bica I, McGovern B, Dhar R, Stone D, McGowan K, Scheib R, Snydman DR. Increasing mortality due to end-stage liver disease in patients with human immunodeficiency virus infection. Clin Infect Dis 2001; 32:492–497.
6. Hung YB, Liang JT, Chu JS, Chen KM, Lee CS. Fulminant hepatic failure in a renal transplant recipient with positive hepatitis B surface antigens: a case report of fibrosing cholestatic hepatitis. Hepatogastroenterology 1995; 42:913–918.
7. Mason AL, Wick M, White HM, Benner KG, Lee RG, Regenstein F,
et al . Increased hepatocyte expression of hepatitis B virus transcription in patients with features of fibrosing cholestatic hepatitis. Gastroenterology 1993; 105:237–244.
8. Gunther S, Baginski S, Kissel H, Reinke P, Kruger DH, Will H, Meisel H. Accumulation and persistence of hepatitis B virus core gene deletion mutants in renal transplant patients are associated with end-stage liver disease. Hepatology 1996; 24:751–758.
9. Penna A, Chisari FV, Bertoletti A, Missale G, Fowler P, Giuberti T,
et al . Cytotoxic T lymphocytes recognize an HLA-A2-restricted epitope within the hepatitis B virus nucleocapsid antigen. J Exp Med 1991; 174:1565–1570.
10. Preikschat P, Gunther S, Reinhold S, Will H, Budde K, Neumayer HH,
et al . Complex HBV populations with mutations in core promoter, C gene, and pre-S region are associated with development of cirrhosis in long-term renal transplant recipients. Hepatology 2002; 35:466–477.
11. Stuyver LJ, Locarnini SA, Lok A, Richman DD, Carman WF, Dienstag JL, Schinazi RF. Nomenclature for antiviral-resistant human hepatitis B virus mutations in the polymerase region. Hepatology 2001; 33:751–757.
12. Tai PC, Banik D, Lin GI, Pai S, Pai K, Lin MH,
et al . Novel and frequent mutations of hepatitis B virus coincide with a major histocompatibility complex class I-restricted T-cell epitope of the surface antigen. J Virol 1997; 71:4852–4856.
13. Marschenz S, Endres AS, Brinckmann A, Heise T, Kristiansen G, Nurnberg P,
et al . Functional analysis of complex hepatitis B virus variants associated with development of liver cirrhosis. Gastroenterology 2006; 131:765–780.
14. Gunther S, Piwon N, Jung A, Iwanska A, Schmitz H, Will H. Enhanced replication contributes to enrichment of hepatitis B virus with a deletion in the core gene. Virology 2000; 273:286–299.
15. Colin JF, Cazals-Hatem D, Loriot MA, Martinot-Peignoux M, Pham BN, Auperin A,
et al . Influence of human immunodeficiency virus infection on chronic hepatitis B in homosexual men. Hepatology 1999; 29:1306–1310.
16. Mai AL, Yim C, O'Rourke K, Heathcote EJ. The interaction of human immunodeficiency virus infection and hepatitis B virus infection in infected homosexual men. J Clin Gastroenterol 1996; 22:299–304.
17. Gilson RJ, Hawkins AE, Beecham MR, Ross E, Waite J, Briggs M,
et al . Interactions between HIV and hepatitis B virus in homosexual men: effects on the natural history of infection. Aids 1997; 11:597–606.
18. Matthews GV, Bartholomeusz A, Locarnini S, Ayres A, Sasaduesz J, Seaberg E,
et al . Characteristics of drug resistant HBV in an international collaborative study of HIV-HBV-infected individuals on extended lamivudine therapy. Aids 2006; 20:863–870.
19. Cooley L, Ayres A, Bartholomeusz A, Lewin S, Crowe S, Mijch A,
et al . Prevalence and characterization of lamivudine-resistant hepatitis B virus mutations in HIV-HBV co-infected individuals. Aids 2003; 17:1649–1657.
20. Chin R, Shaw T, Torresi J, Sozzi V, Trautwein C, Bock T,
et al . In vitro susceptibilities of wild-type or drug-resistant hepatitis B virus to (-)-beta-D-2,6-diaminopurine dioxolane and 2′-fluoro-5-methyl-beta-L-arabinofuranosyluracil. Antimicrob Agents Chemother 2001; 45:2495–2501.
21. Gunther S, Li BC, Miska S, Kruger DH, Meisel H, Will H. A novel method for efficient amplification of whole hepatitis B virus genomes permits rapid functional analysis and reveals deletion mutants in immunosuppressed patients. J Virol 1995; 69:5437–5444.
22. Ayres A, Locarnini S, Bartholomeusz A. HBV genotyping and analysis for unique mutations. Methods Mol Med 2004; 95:125–149.
23. Yuen KWL, Ayres A, Littlejohn M, Colledge D, Edgley A, Maskill WJ,
et al . SEQHEPB: A sequence and analysis program and relational database system for chronic hepatitis B antiviral research.
Antiviral Research in press.
24. Bartholomeusz A, Locarnini SA, Ayres A, Thompson G, Sozzi V. Molecular modelling of hepatitis B virus polymerase and adefovir resistance identifies three clusters of mutations. Hepatology 2004; 40(Suppl 1A):A185.
25. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 1997; 25:4876–4882.
26. Hall T. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp 1999; 41:95–98.
27. Page RD. TreeView: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 1996; 12:357–358.
28. Carlier D, Jean-Jean O, Fouillot N, Will H, Rossignol JM. Importance of the C terminus of the hepatitis B virus precore protein in secretion of HBe antigen. J Gen Virol 1995; 76(Pt 4):1041–1045.
29. Bowyer SM, van Staden L, Kew MC, Sim JG. A unique segment of the hepatitis B virus group A genotype identified in isolates from South Africa. J Gen Virol 1997; 78(Pt 7):1719–1729.
30. Hannoun C, Soderstrom A, Norkrans G, Lindh M. Phylogeny of African complete genomes reveals a West African genotype A subtype of hepatitis B virus and relatedness between Somali and Asian A1 sequences. J Gen Virol 2005; 86:2163–2167.
31. Sugauchi F, Kumada H, Sakugawa H, Komatsu M, Niitsuma H, Watanabe H,
et al . Two subtypes of genotype B (Ba and Bj) of hepatitis B virus in Japan. Clin Infect Dis 2004; 38:1222–1228.
32. Bell SJ, Lau A, Thompson A, Watson KJ, Demediuk B, Shaw G,
et al . Chronic hepatitis B: recommendations for therapy based on the natural history of disease in Australian patients. J Clin Virol 2005; 32:122–127.
33. Ferrari C, Bertoletti A, Penna A, Cavalli A, Valli A, Missale G,
et al . Identification of immunodominant T cell epitopes of the hepatitis B virus nucleocapsid antigen. J Clin Invest 1991; 88:214–222.
34. Andrei G, Balzarini J, Fiten P, De Clercq E, Opdenakker G, Snoeck R. Characterization of herpes simplex virus type 1 thymidine kinase mutants selected under a single round of high-dose brivudin. J Virol 2005; 79:5863–5869.
35. Okada Y, Streisinger G, Owen JE, Newton J, Tsugita A, Inouye M. Molecular basis of a mutational hot spot in the lysozyme gene of bacteriophage T4. Nature 1972; 236:338–341.
36. Sasadeusz JJ, Tufaro F, Safrin S, Schubert K, Hubinette MM, Cheung PK, Sacks SL. Homopolymer mutational hot spots mediate herpes simplex virus resistance to acyclovir. J Virol 1997; 71:3872–3878.
37. Wilson JB, Hayday A, Courtneidge S, Fried M. A frameshift at a mutational hotspot in the polyoma virus early region generates two new proteins that define T-antigen functional domains. Cell 1986; 44:477–487.
38. Locarnini S, Shaw T, Dean J, Colledge D, Thompson A, Li K,
et al . Cellular response to conditional expression of the hepatitis B virus precore and core proteins in cultured hepatoma (Huh-7) cells. J Clin Virol 2005; 32:113–121.
39. Messageot F, Salhi S, Eon P, Rossignol JM. Proteolytic processing of the hepatitis B virus e antigen precursor. Cleavage at two furin consensus sequences. J Biol Chem 2003; 278:891–895.
40. Hunt CM, McGill JM, Allen MI, Condreay LD. Clinical relevance of hepatitis B viral mutations. Hepatology 2000; 31:1037–1044.
41. Bouchard MJ, Schneider RJ. The enigmatic X gene of hepatitis B virus. J Virol 2004; 78:12725–12734.
42. Scaglioni PP, Melegari M, Wands JR. Posttranscriptional regulation of hepatitis B virus replication by the precore protein. J Virol 1997; 71:345–353.