Homologous recombination between different genotypes of hepatitis B virus

Gene 260 (2000) 55–65 www.elsevier.com/locate/gene

Homologous recombination between different genotypes of hepatitis B virus Viacheslav Morozov *, Maria Pisareva, Mikhail Groudinin Influenza Institute, Russian Academy of Medical Science, Prof. Popov Str. 15/17, St. Petersburg 197376, Russia Received 20 April 2000; received in revised form 28 August 2000; accepted 26 September 2000 Received by J. Svoboda

Abstract Phylogenetic analysis was used to examine the evolutionary relationships among 99 complete HBV sequences. Analysis revealed nine viral genomes clustered with different genotypes depending on genome region analyzed. This discordance indicated that recombination events occurred during HBV history. The putative breakpoints between genomes of different genotypes have been mapped. Six mosaic genomes representing B/C hybrids were isolated in East Asia and three A/D hybrids in Italy. At least some recombinant strains appear to be fully viable and possess high evolutionary potential. As a result, B/C recombinants overspread through the East Asia region. They were found among the isolates from Japan, China and Indonesia. Our results suggest that recombination is a significant and relatively frequent event in the evolution of HBV genome. A possible mechanism and the implications of recombination for the natural history of HBV, clinically important properties, and phylogenetic reconstruction are discussed. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Bootscanning; Mixed infection; Mosaic genome; Phylogenetic trees; Recombinant region

1. Introduction Hepatitis B virus (HBV ) belongs to the family of hepadnaviruses. It is a partially double-stranded DNA virus with compact circular genome of about 3200 base pairs (bp) containing four overlapping open reading frames P, preS/S, preC/C and X. The virus associated with acute hepatitis, chronic hepatitis, and development of hepatocellular carcinoma causes considerable morbidity and mortality over the world. In the countries where HBV carrier rates reach 10%, HBV infection may account for 3% of total mortality (Maynard, 1990). HBV sequences have been classified on the basis of their phylogenetic relationships into seven genetic groups, termed A–G, based on an intergroup divergence of 8% or greater of complete nucleotide sequence (Okamoto et al., 1988; Norder et al., 1994). These groups have distinct geographical distribution. Genotype A is prevalent in Northern Europe, however Abbreviations: cccDNA, covalently closed circular DNA; DR1, direct repeat 1; HBV, hepatitis B virus. * Corresponding author. Tel.: +7-812-234-4251; fax: +7-812-234-6200. E-mail address: [email protected] ( V. Morozov)

it is also found in North and South America and Africa. Genotypes B and C have been evolved in East Asia. Genotype D spreads worldwide, but predominates in the Mediterranean area. Genotype E is found mainly in the western part of Sub-Saharan Africa. Genotype F has been originated from the native population of the New World (Norder et al., 1993). Recently, a new genotype G was found in the samples from USA and France (Stuyver et al., 2000). It has been assumed that the outcome of hepatitis B infection depends on HBV genotype (Mayerat et al., 1999). Therefore, variations in the genome of HBV could explain the geographical differences in the natural history of hepatitis B. However, there is little data to show a causal relationship. For example, genotype A differs from other genotypes by rare circulation of HBe(−) mutants in nature (Li et al., 1993). Genetic heterogeneity has been shown to represent an important element in viral pathogenesis. Some of the mutations identified till now suggest a contribution to viral latency, low level HBV infection, the severity of liver disease and vaccine escape (Blum, 1993). HBV has a very high rate of diversity. In fact, HBV exists as ‘quasispecies’ — a population of related, however divergent,

0378-1119/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 0 0 ) 0 0 42 4 - 8

56

V. Morozov et al. / Gene 260 (2000) 55–65

Table 1 HBV sequences used in this study ID

Accession no.

Genotype

Authors

HPBVCG HPBADW1 HPBADW2 HPBADW3 HPBADRC HPBCG HPBHBVA HPBE88A HPBH2B HPBHBCAGA HPBHBCAGB HPBHBCAGC HPBHBCAGD HPBHBCAGE HPBHBCAGF HPBHBCAGG HPBHBCAGH HPBA11A HPBAYW HPBETNC HPBMUT HPBADRA HPBHBVAA HPBADR1CG HPBCGADR HPBADWZ HPBADWZCG S50225 HBVADW HBVADR HBVADR4 XXHEPAV HBVADW2 HEHBVAYR HBVADRM HBVPREX HBVAYWMCG HBVAYWC HBVAYWCI HBVAYWE HBVDNA HBVADW4A HBVXCPS HBVORFS HHVCCHA HHVBBAS HHVBFFOU HHVBF HHVBE4 HHVBC HBVPRES12 HBVGEN1 HBVGEN2 AB014360–AB014365 AB014366 AB014367–AB014369 AB014370 AB014371–AB014399 AF043593 AF043594 AF100308 AF100309 HUMPRECX S65868 HBVAYWGEN HBV012207 D50519 D50521 D50522

D00220 D00329 D00330 D00331 D00630 D12980 D16665 D16666 D16667 D23677 D23678 D23679 D23680 D23681 D23682 D23683 D23684 D50489 J02203 L08805 L27106 M12906 M32138 M38454 M38636 M54923 M57663 S50225 V00866 V00867 X01587 X02496 X02763 X04615 X14193 X52939 X59795 X65257 X65258 X65259 X68292 X69798 X70185 X72702 X75656 X75657 X75658 X75663 X75664 X75665 X85254 Z35716 Z35717 AB014360–AB014365 AB014366 AB014367–AB014369 AB014370 AB014371–AB014399 AF043593 AF043594 AF100308 AF100309 AJ012207 D50519 D50521 D50522 L13994 S65868 Y07587

B B B C C C C C B B B C C C C C C D C D C D C C B A A A C C D A C C C D D D D D F A D C E F F E C D D A C B C A C D D B B A A D A C B B

Vaudin et al., 1988 Okamoto et al., 1988 Okamoto et al., 1988 Okamoto et al., 1988 Ono et al., 1983 Mukaide et al., 1992 Uchida et al., 1994 Uchida et al., 1995 Uchida et al., 1995 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Horikita et al., 1994 Uchida et al., 1995 Galibert et al., 1979 Ogata et al., 1993 Hasegawa et al., 1994 Kobayashi et al., 1984 Tong et al., 1990 Renbao et al., 1987 Kim et al., 1988 Satrosoewignjo et al., 1987 Estacio et al., 1988 Wands et al., 1992 Ono et al., 1983 Ono et al., 1983 Fujiyama et al., 1983 Bichko et al., 1985 Valenzuela et al., 1980 Okamoto et al., 1986 Rho et al., 1989 Loncarevic et al., 1990 Lai et al., 1991 Lai et al., 1992 Lai et al., 1992 Lai et al., 1992 Lai et al., 1992 Naumann et al., 1993 Preisler-Adams et al., 1993 Preisler-Adams et al., 1993 Norder et al., 1994 Norder et al., 1994 Norder et al., 1994 Norder et al., 1994 Norder et al., 1994 Norder et al., 1994 Lai et al., 1995 Plucienniczak et al., 1994 Plucienniczak et al., 1994 Takahashi et al., 1999 Takahashi et al., 1999 Takahashi et al., 1999 Takahashi et al., 1999 Takahashi et al., 1999 Gunther et al., 1998 Gunther et al., 1998 Wen, 1998 Wen, 1998 Asahina et al., 1995 Dai et al., 1993 Stoll-Becker et al., 1996 Heermann et al.,1998 Asahina et al., 1995 Asahina et al., 1995 Asahina et al., 1995

57

V. Morozov et al. / Gene 260 (2000) 55–65

viruses within the same individual. The mutation rate of HBV genome estimated by different authors varies from 10−5 to 10−4 nucleotide substitutions per site per year (Okamoto et al., 1987; Georgi-Geisberger et al., 1992). This value is intermediate between DNA and RNA viruses because HBV replicates similarly to retroviruses through reverse transcription of RNA intermediate by reverse transcriptase, which lacks a proofreading function, resulting in nucleotide substitutions, deletions and insertions. In addition, recently homologous recombination as a new source of variation has been documented in HBV (Tran et al., 1991; Georgi-Geisberger et al., 1992). In particular, one case of recombination between different HBV genotypes has been reported (Bollyky et al., 1996). It raises the question of how frequently such recombination does occur, and whether recombinant viruses differ in their pathogenicity. Here we report nine intergroup recombinant HBV genomes and localize crossover points between regions originated from different genotypes.

2. Data and methods Nucleotide sequences were obtained from the GenBank/EMBL/DDBJ databases. The accession numbers of 99 complete sequences are indicated in Table 1. Multiple alignments were obtained using Clustal W ( Thompson et al., 1994) with minor manual adjustment. The full-genome alignment was split into sequential separate files of 200 nucleotides with a 50 nucleotide step. Pairwise evolutionary distances were estimated by Kimura’s two-parameter method ( Kimura, 1980) with transition-to-transvertion ratio of 2.0. Phylogenetic trees were constructed for each subalignment by the neighborjoining methods implemented by the PHYLIP software package (Felsenstein, 1993) or the MEGA program ( Kumar et al., 1993). The statistical strength of the neighbor-joining methods was assessed by bootstrap resampling with 500 replicates. We assume that a recombination event has taken place if the sequence clusters

with one genotype in one region of the genome, and another genotype in the other region. In order to reveal breakpoints, the potential recombinant sequences were subjected to a bootscanning procedure (Salminen et al., 1995). The SimPlot program ( Ray, 1999) was used to identify phylogenetically informative sites supporting alternative tree topologies ( Robertson et al., 1995). This was performed by considering four sequences at a time: a putative recombinant sequence, two potential parental sequences, and an outgroup. Each informative site supports one of three possible phylogenetic trees. The distribution of sites supported the clustering of putative recombinant, with one parental sequence or the other, analyzed by maximum chi-squared method to identify breakpoints.

3. Results and discussion 3.1. Evidence for mosaicism Initially, we derived evolutionary trees from the complete genome and three ORFs of HBV: S, C and X. Phylogenetic analysis of 53 complete sequences revealed seven viral genomes clustered with different genotypes depending on analyzed ORF ( Fig. 1, Fig. 2A). The results are summarized in Table 2. To estimate crossover points of recombination events we constructed phylogenetic trees for different parts of the HBV genome as shown in Section 2. The multiplesequence alignment was split into subalignments of 200 nucleotides and trees were obtained separately for each subalignment. The HBV genome was scanned along the alignment with an increment of 50 nucleotides, searching for changes in tree topology. No recombinant sequences were detected, except seven, which were revealed because of contradictory phylogenetic relationships for three ORFs. The recombinant region in D00330, D00331, M54923 and D16665 covered the C gene entirely, so that the 5∞- and 3∞-ends of the C gene were arranged within crossover regions. One breakpoint of X65259

Table 2 Summary of the genotyping result obtained from different ORFs of putative HBV recombinants GenBank accession no.

Authors

Complete genome

S

X

C

Features

X65258 X65259 X68292 D00330 D00331 M54923 Af100308 Af100309 D16665

Lai et al., 1992a Lai et al., 1992a Lai et al., 1992a Okamoto et al., 1988 Okamoto et al., 1988 Satrosoewignjo et al., 1987 Wen, 1998a Wen, 1998a Uchida et al., 1994

D D D B B B B B C

A D D B B B B B C

A A A B B B B B C

D D A C C C C C B

overlapping PCR clones overlapping PCR clones overlapping PCR clones full-length clone full-length clone full-length clone full-length clone full-length clone full-length clone

a Strains that were presented by direct submission to GenBank.

V. Morozov et al. / Gene 260 (2000) 55–65

Fig. 1. Phylogenetic trees based on two sets of complete HBV genome sequences.

58

Fig. 2. Phylogenetic trees based on the three different ORFs (S, C and X ) of two sets of HBV sequences. The numbers at nodes indicate the percentage of bootstrap replications supporting the clusters (75% and higher are shown). Brackets indicate the HBV genotypes. The mosaic genomes are represented in boxes. Horizontal branch lengths are drawn to scale.

V. Morozov et al. / Gene 260 (2000) 55–65

59

60

V. Morozov et al. / Gene 260 (2000) 55–65

V. Morozov et al. / Gene 260 (2000) 55–65

61

Fig. 4. Diagramic representation of the mosaic structure of the putative recombinant HBV sequences. Nucleotides are numbered from a hypothetical EcoRI site.

was within the proximal part of the S gene and another was within the distal part of the core gene. X65258 and X68292 had multiple crossover points and the recombinant region consisted of several segments which covered parts of the P, S, X and C genes (Fig. 3). To augment these analyses, we constructed the trees for an additional 46 complete HBV sequences and examined the phylogenetic relationships using the same procedure as stated above ( Fig. 2B). We revealed two extra B/C recombinant sequences: AF100308 and AF100109. Both of them contained recombinant regions, which covered the C gene completely, as shown for the other B/C type hybrid genomes. To localize the crossover points more precisely, we applied the SimPlot program (Ray, 1999) to identify the distribution of phylogenetically informative sites using a four-sequence alignment including recombinant sequence, one representative of each of two parental

sequences, and an outgroup. The parental sequences were chosen on the basis of their similarity to the recombinant within each of two separate regions clustered with different genotypes. The following reference sequences were chosen: D23677 (genotype B) and D12980 (genotype C ) for B/C recombinants; X02763 (genotype A) and X02496 (genotype D) for A/D recombinants. F genotype sequence, X75658, was used as the outgroup. Fig. 4 depicts the resolution of the mosaic structure of nine putative recombinant HBV genomes by bootscanning. Bootstrap values were calculated for a window of 300 nucleotides moved in a step size of 50 nucleotides. Depending on the region, recombinant genomes were clustered with two parental genotypes alternately with bootstrap values greater than 70%. The most likely breakpoints were located by the method of chi-square maximization. As all the conclusions drawn from the analysis of informative sites depend on the

Fig. 3. Bootstrap plots illustrating the likelihood of clustering of putative B/C recombinant sequences (AF100308, AF100309, D00330, D00331, M54923, D16665) with reference sequences for genotype B (D23677, dark line) and genotype C (D12980, light line); and A/D recombinant sequences ( X65258, X65259, X68292) with reference sequences for genotype A ( X02763, light line) and genotype D ( X02496, dark line). X75658 has been used as the outgroup.

62

V. Morozov et al. / Gene 260 (2000) 55–65

assumption that reference sequences by themselves did not arise due to cross-genotype recombination, we used several additional sets of reference sequences to confirm our results. These data are summarized in Fig. 5. Thus we revealed nine intertype recombinant sequences of HBV. Six of them represented B/C hybrid genomes originated from South-East Asia, where B and C genotypes were co-circulating, and therefore superinfection of individuals by different genomic groups could occur. D16665, originated from Japan, was the only one clustered with B genotype in the core gene region. The other B/C hybrids had core genes clustered with C genotype. They were divided into two subgroups: one originated from Japan and China, the other from Indonesia. In our study five out of 12 HBV sequences which were classified on the basis of analysis of complete sequence as B genotype in fact exhibited B/C hybrid genomes. Such discordance between trees obtained from different HBV ORFs could be seen in the phylogenetic reconstruction provided by Norder et al., but the possibility of cross-type recombination has not been discussed in this report (Norder et al., 1994). Bollyky et al. have reported the recombination event between B and D genotypes and considered D00329 to be a recombinant genome (Bollyky et al., 1996). In our opinion this sequence was the only one representative of genotype B, in that analysis and the high per cent of B/C recombinant genomes (considered by Bollyky et al. as true type B genomes) within the B group have obscured the real relations within this group. Three out of nine recombinant genotypes represented A/D hybrids. All of them came from Italy, where A and D represented 90% of cases of HBV infection (Mayerat et al., 1999). Probably, they originated from the same source as a result of several recombinant events. The identification of intertype recombination suggests the existence of a relatively high frequency of HBV mixed infections. This statement is supported by Gerner et al., who had found that hepatitis B virus could change genotype from A to D after seroconversion to anti-HBe (Gerner et al., 1998). Since it is unlikely that a new genotype could evolve under selection within such a short period of time, this observation may indicate that the infection with different genotypes of HBV can occur simultaneously. On the other hand, mixed infection by two different genotypes may be the reason for numerous artifacts produced at any stage of laboratory procedures and data processing. Especially, great care should be taken in interpreting data employing PCR products. PCR co-amplification of two different genotype sequences may lead to the formation of recombinant DNA molecules as a result of annealing of incompletely extended primers to an alternative form of template strand. Artifacts may also be the result of inaccurate reconstruction of complete sequences from a PCR fragment (Learn

et al., 1996). In our case six full-genome nucleotide sequences, representing the putative B/C recombinant genotype, were sequenced from full-length genomic DNA of HBV clones. This strongly supports the suggestion that these intertype mosaic genomes could not be considered an artifact of in vitro manipulation. In contrast, each of three putative A/D recombinant sequences were assembled from nine overlapping PCR clones. Unfortunately, the authors did not specify any additional information concerning the sequencing and reconstruction methods. Therefore we cannot completely exclude the possibility that recombinant events were artifacts that occurred during PCR amplification or reconstruction of the complete sequence from PCR fragments. 3.2. Is intermolecular genome recombination provided by non-random mechanism? Computer-assisted analyses of hepadnavirus and retrovirus genomes suggest their common evolutionary origin (Miller and Robinson, 1986). Besides, the DNA molecular of HBV is generated through reverse transcription of an RNA intermediate similar to retrovirus replication (Summers and Mason, 1982). Reverse transcription serves as an important mechanism for generating variations in retroviral genomes because of errorprone DNA synthesis by reverse transcriptase, which lacks a proofreading function. In addition, recombination can occur during retroviral replication as a consequence of packaging of two RNA genome copies into each virion and reverse transcriptase switches between templates (Goodrich and Duesberg, 1990). However, HBV replication takes place inside the nucleocapsid and, unlike retroviruses, this process involves only one copy of the RNA pregenome. It seems there is no chance for homologous recombination. Nevertheless, additional to our theoretical research, the homologous recombination has been documented during the course of chronic HBV infection ( Tran et al., 1991; Georgi-Geisberger et al., 1992). The comparison of our data and data concerning HBV integration into the cellular DNA reported in the literature reveals similar regularity in distribution of recombinant sites. Eight out of nine intertype genomes reported here contained the breakpoints, which lie in the vicinity of DR1 and encapsidation signal of the HBV pregenome. A high prevalence of this region for integration has also been observed in the majority of integrants. Hino et al. found that the fragment of HBV DNA covering nucleotides 1855–1915 is indispensable for enhancement of in vitro recombination (Hino et al., 1991). Pineau et al. reported that the region encompassing nucleotides 1600–2000 reaches a recombination site density almost five-fold higher than the remaining part of the genome (Pineau et al., 1998).

V. Morozov et al. / Gene 260 (2000) 55–65

63

Fig. 5. Distribution of phylogenetically informative sites within the HBV genomes alignment. Columns with numbers indicate base positions in the alignment. The second and third columns indicate informative sites within the putative parental genotypes represented as consensus sequences. (A) The representation informative sites within the alignment of genotype groups B, C and B/C intergroup recombinants. (B) Informative sites around mosaic areas of A/D recombinant genomes compared to consensus sequences of genotypes A and D. It should be noted that the fragment of X65259 (nucleotides 1760–1999) seems to be clustered with genotype D. However, it cannot be excluded that this segment originated from a precursor representing genotype A in the result of double mutations in position 1861 (CT in codon 15) and 1899 (stop in codon 28). The coexistence of these mutations in genotype A has been described previously (Rodriguez-Frias et al., 1995). (C ) The alignment X68292 with consensus sequences of two putative parental genotypes A and D. Localization of recombinant breakpoint in the nt 2359 of X68292 is shown by an arrow.

64

V. Morozov et al. / Gene 260 (2000) 55–65

The second hot spot of the recombination arranges near the 3∞-end of the core gene. In case of X68292 the breakpoint can be estimate precisely in the position 2359 bp, because of localization within the six base insertion of A genotype (Fig. 5C ). The other six genomes contain the crossover point between position 2444–2486 (2437–2479 for D16665). The last and the least cluster include three breakpoints, which lie within the 3∞-end of the S gene. This group contains only A/D recombinant. These findings support the existence of a non-random mechanism of HBV integration and intermolecular genome recombination. We can speculate that both of these processes came through the common intermediate. Disbalance between HBV gene expression and viral replication may lead to accumulation of intermediate forms. It can result in the increase of probability of recombination between the episomal DNA and integration of HBV genome into the host genome. In the studies that used a primary duck hepatitis model, envelope proteins have been found to regulate covalently closed circular DNA (cccDNA) amplification. cccDNA is the major form of viral DNA found in the nucleus, which serves as a template for transcription of viral RNA. Mutants defective in preS/S protein accumulated high levels of cccDNA compared with the wild type (Summers et al., 1991). This could be a trigger mechanism for the intermolecular recombination. Yang et al. described a process, which was called illegitimate replication ( Yang and Summers, 1995). In this pathway viral DNA replication goes through linear DNA intermediates which can efficiently convert to cccDNA by non-homologous recombination between the two ends. Besides, the products of this recombination are oligomeric forms in which monomers are joined near the ends in random orientation. The efficiency of non-homologous recombination of monomeric and oligomeric linear DNA suggests that they can be the precursors of integrated hepadnovirus DNA ( Yang and Summers, 1995). Thus recombination of linear DNA can be supposed to lead to the appearance of mosaic HBV genomes, in case of superinfection by divergent viral strains. 3.3. B/C recombinant genomes are widespread in East Asia Because of the compact genome of HBV with overlapping open reading frames, it seems that most recombinations would be fatal for the fertility of the virus. However, at least some recombinant viruses appear to be fully viable. For example, B/C recombinants overspread through the East Asia region and were found among the isolates from Japan, China and Indonesia. There are some data which suggest that a pathogenic difference between B and C genotypes may exist.

Genotype B compared with genotype C could be associated with a faster transition through the immunoactive stage and a faster HBe seroconversion, while C is associated with more pronounced liver inflammation (Lindh et al., 1999). A Japanese study supports this finding. Noguchi et al. reported that adr carriers (i.e. genotype C ) are more often HBeAg positive than the carriers infected with adw strain (genotype B is more common in Japan) (Noguchi et al., 1994). In the light of these findings, it is very interesting to know how recombinant viruses differ in their natural history and pathogenicity.

4. Conclusions (1) The recombinant events described here have several important implications. Recombination can exert an influence on clinically important properties more dramatically than the steady accumulation of separate mutations. It could provide a powerful mechanism to escape from immunity and could be implicated in persistence of hepatitis B virus. (2) Phylogenetic reconstruction of HBV history should be provided carefully to examine whether genomes are mosaic. Moreover, extrapolating the results based on the partial sequences may not reveal the real evolutionary relationships of HBV. (3) Recently a possible association between HBV genotype and clinical outcome has been investigated in a number of studies. However, our results suggest a large share of B/C recombinant genomes among genotype B. Therefore, it should be taken into consideration that B/C recombinant genomes could obscure the real genotype-related differences in the pathogenicity of HBV in East Asia where genotype B overspread.

References Blum, H.E., 1993. Hepatitis B virus: significance of naturally occurring mutants. Intervirology 35, 40–50. Bollyky, P.L., Rambaut, A., Harvey, P.H., Holmes, E.C., 1996. Recombination between sequences of hepatitis B virus from different genotypes. J. Mol. Evol. 42, 97–102. Felsenstein, J., 1993. PHYLIP (Phylogeny Inference Package), Vers. 3.5c. Georgi-Geisberger, P., Berns, H., Loncarevic, I.F., Yu, Z.-Y., Tang, Z.-Y., Zentgraf, H., Schroder, C.H., 1992. Mutations on free and integrated hepatitis B virus DNA in a hepatocellular carcinoma: footprint of homologous recombination. Oncology 49, 386–395. Gerner, P.R., Friedt, M., Oettinger, R., Lausch, E., Wirth, S., 1998. The hepatitis B virus seroconversion to anti-HBe is frequently associated with HBV genotype changes and selection of preS2-defective particles in chronically infected children. Virology 245, 163–172. Goodrich, D.W., Duesberg, P.H., 1990. Retroviral recombination during reverse transcription. Proc. Natl. Acad. Sci. USA 87, 2052–2056. Hino, O., Tabata, S., Hotta, Y., 1991. Evidence for increased in vitro

V. Morozov et al. / Gene 260 (2000) 55–65 recombination with insertion of human hepatitis B virus DNA. Proc. Natl. Acad. Sci. USA 88, 9248–9252. Kimura, M., 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111–120. Kumar, S., Tamura, K., Nei, M., 1993. MEGA (Molecular Evolutionary Genetics Analysis), Vers. 1.02, Pennsylvania State University. Learn, G.H., Korber, B.T.M., Foley, B., Hahn, B.H., Wolinsky, S.M., Mullins, J.I., 1996. Maintaining the integrity of human immunodeficieny virus sequence databases. J. Virol. 70, 5720–5730. Li, J.S., Tong, S.P., Wen, Y.M., Vitvitski, L., Zhang, Q., Trepo, C., 1993. Hepatitis B virus genotype A rarely circulates as an HBeminus mutant: possible contribution of a single nucleotide in the precore region. J. Virol. 67, 5402–5410. Lindh, M., Hannoun, C., Dhillon, A.P., Norkrans, G., Horal, P., 1999. Core promoter mutations and genotypes in relation to viral replication and liver damage in east asian hepatitis B virus carriers. J. Infect. Dis. 179, 775–782. Mayerat, C., Mantegani, A., Frei, P.C., 1999. Does hepatitis B virus (HBV ) genotype influence the clinical outcome of HBV infection? J. Viral Hepat. 6, 299–304. Maynard, J.E., 1990. Hepatitis B: global importance and need for control. Vaccine 8, Suppl., S18–20, S21–23. Miller, R.H., Robinson, W.S., 1986. Common evolutionery origin of hepatitis B virus and retroviruses. Proc. Natl. Acad. Sci. USA 83, 2531–2535. Noguchi, A., Hayashi, J., Nakashima, K., Hirata, M., Ikematsu, H., Kashiwagi, S., 1994. HBsAg subtypes among HBsAg carriers in Okinawa, Japan. Evidence of an important relationship in seroconversion from HBeAg to anti-HBe. J. Infect. 28, 141–150. Norder, H., Hammas, B., Lee, S.D., Bile, K., Courouce, A.M., Mushahwar, I.K., Magnius, L., 1993. Genetic relatedness of hepatitis B viral strains of diverse geographical origin and natural variations in the primary structure of the surface antigen. J. Gen. Virol. 74, 1341–1348. Norder, H., Courouce, A.M., Magnius, L.O., 1994. Complete genomes, phylogenetic relatedness, and structural proteins of six strains of the hepatitis B virus, four of which represent two new genotypes. Virology 198, 489–503. Okamoto, H., Imai, M., Kametani, M., Nakamura, T., Mayumi, M., 1987. Genomic heterogeneity of hepatitis B virus in a 54-year-old

65

woman who contracted the infection through materno-fetal transmission. Jpn. J. Exp. Med. 57, 231–236. Okamoto, H., Tsuda, F., Sakugawa, H., Sastrosoewignjo, R.I., Imai, M., Miyakawa, Y., Mayumi, M., 1988. Typing hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J. Gen. Virol. 69, 2575–2583. Pineau, P., Marchio, A., Mattei, M.-G., Kim, W.-H., Youn, J.-K., Tiollais, P., Dejean, A., 1998. Extensive analysis of duplicatedinverted hepatitis B virus integrations in human hepatocellular carcinoma. J. Gen. Virol. 79, 591–600. Ray, S.C., 1999. SimPlot for Windows, Vers. 2.5. Robertson, D.L., Hahn, B.H., Sharp, P.M., 1995. Recombination in AIDS viruses. J. Mol. Evol. 40, 249–259. Rodriguez-Frias, F., Buti, M., Jardi, R., Cotrina, M., Viladomiu, L., Esteban, R., Guardia, J., 1995. Hepatitis B virus infection: precore mutants and its relation to viral genotypes and core mutations. Hepatology 22, 1641–1647. Salminen, M.O., Carr, J.K., Burke, D.S., McCutchan, F.E., 1995. Identification of breakpoints in intergenotypic recombinants of HIV-1 by bootscanning. AIDS Res. Human Retrovir. 11, 1423–1425. Stuyver, L., De Gendt, S., Van Geyt, C., Zoulim, F., Fried, M., Schinazi, R.F., Rossau, R., 2000. A new genotype of hepatitis B virus: complete genome and phylogenetic relatedness. J. Gen. Virol. 81, 67–74. Summers, J., Mason, W.S., 1982. Replication of the genome of a hepatitis B-like virus by reverse transcription of an RNA intermediate. Cell 29, 403–415. Summers, J., Smith, P.M., Huang, M.J., 1991. Morphogenetic and regulatory effects of mutations in envelop proteins of an avian hepadnovirus. J. Virol. 65, 1310–1317. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. Tran, A., Kremsdorf, D., Capel, F., Housset, C., Dauguet, C., Petit, M.A., Brechot, C., 1991. Emergence of and takeover by hepatitis B virus (HBV ) with rearrangements in the pre-S/S and pre-C/C genes during chronic HBV infection. J. Virol. 65, 3566–3574. Yang, W., Summers, J., 1995. Illegitimate replication of linear hepadnovirus DNA through nonhomologous recombination. J. Virol. 69, 4029–4036.