Recombinant hepatitis C virus in experimentally infected chimpanzees

Footnotes

In a previous study, three chimpanzees were infected with HCV subgenotypes 1a and 1b following experimental inoculation with a factor VIII concentrate (lot DO56) that contained HCV subgenotypes 1a, 1b, 2b and 3a (Nainan et al., 2006). We looked for HCV recombination in these three animals CH810, CH921 and CH1433.

HCV recombinants were identified by cloning and sequencing PCR fragments derived from the E1E2 region (position 9121547). Two primer pairs were used for nested PCR: first round, 727F (5'-GGCTTCGCCGACCTCATGGGGT) and 1633R (5'-TCATTGCAGTTCAGGGCCGT); second round, 851F (5'-CCGGTTGCTCTTTCTCTATCTT) and 1619R (5'-GCAGTCCTGTTGATGTGCCA). PCR products were cloned by using a TOPO TA cloning kit for sequencing (Invitrogen Life Technologies) and cloned products were sequenced by using BigDye v3 (Applied Biosystems) and an automated sequencer (ABI 3100 Genetic Analyzer; PE Applied Biosystems).

Sequence analysis was performed by using programs from the GCG package. Phylogenetic analysis was conducted by using MEGA (Kumar et al., 2001). Similarity and bootscan analyses were performed by using the SimPlot program v3.2 (Lole et al., 1999) to identify recombinant virus variants. Phylogenetic trees were constructed by using the neighbour-joining method with distances estimated by using the Kimura two-parameter model (Felsenstein, 1993).

Analysis of HCV quasispecies in the factor VIII lot DO56 inoculum and specimens from the three chimpanzees was conducted on 96 clones obtained for each specimen source. Similarity and bootscanning analyses identified two sequences that contained subgenotype 1a and 1b sequences: (i) CH921a-02 derived from chimpanzee CH921 and (ii) CH1433a-68 from chimpanzee CH1433.

Phylogenetic analysis showed that, within CH921a-02, the sequence at position 9121321 belonged to subgenotype 1b, whereas the sequence at position 12991547 belonged to subgenotype 1a (Fig. 1a, b). The sequence at position 12991321 could be classified as either subgenotype 1a or 1b. These findings suggest that sequence CH921a-02 is recombinant, with a crossover point located within region 12991321. By the same approach, sequence CH1433a-68 could be classified as a recombinant between HCV subgenotypes 1a and 1b. The sequence at position 9121447 belonged to subgenotype 1a, the sequence at position 14291547 belonged to subgenotype 1b (Fig. 1c, d) and position 14291447 contained a crossover point. For both sequences, recombination would not change the frame of translation.

To look further for potential inter-subgenotypic recombination, two additional sets of primers for subgenotypes 1a and 1b were designed. For subgenotype 1a, the primer designs were: first-round forward primer 1aF1, 5'-TAACTCGAGTATTGTGTACGAG, reverse primer 1aR1, 5'-GAGACTAGCAAATCCAGAGGTGGTG; and forward nested primer 1aF2, 5'-CGATGCCATCCTGCACACTCC, reverse primer 1aR2, 5'-TCCAGAGGTGGTGTGGCCGGC. For subgenotype 1b, the first-round forward primer was 1bF1, 5'-CAACTCAAGCATTGTGTACGAGGCA, with reverse primer 1bR1, 5'-GACGCGAAACTGTAGGTGTTGC; the forward nested primer was 1bF2, 5'-ACATGATCATGCACACCCCCG, and the reverse nested primer was 1bR2, 5'-ACTGTAGGAGTTGCGGGCCGA. These primers were used to probe for all four combinations of HCV subgenotypes 1a and 1b (i.e. 1a/1a, 1a/1b, 1b/1b, 1b/1a) in the chimpanzees and factor VIII concentrate. The 1a/1a and 1b/1b combinations of primers amplified fragments from all four sources. However, the 1a/1b and 1b/1a combinations of primers only generated a fragment from CH921. Fragments identified with 1a/1b and 1b/1a primers were plasmid-cloned and sequenced. Among 39 clones, three (CH921-F2, CH921-F6 and CH921-F13) contained sequences from the 1a and 1b subgenotype regions (Fig. 2) and could be classified as recombinants. Two of these sequences had crossover points at a location different from that of the recombinants found in the previously described experiments. The recombinant sequence CH921-F13 had a crossover point at exactly the same location as CH921a-02. However, the 1a and 1b segments were arranged within the CH921-F13 sequence in an order different from that within CH921a-02.

Genetic recombination between HCV genotypes and subgenotypes has been reported in humans. A recombinant HCV isolate comprising subgenotype 2k sequences within the 5' untranslated region (UTR), core, envelope and part of the NS2 regions and subgenotype 1b sequences within the rest of the genome was identified in St Petersburg, Russia (Kalinina et al., 2002). Another recombinant isolate with 1b sequence within the 5' UTRcore region and 1a sequences within the NS5b region has been reported from Peru (Colina et al., 2004). Our findings of several recombinant HCV quasispecies variants in experimentally infected chimpanzees provide additional evidence for genomic recombination in this virus.

Determining the frequency of HCV recombination may add to our understanding of HCV evolution. However, each case of presumed recombination should be evaluated carefully to exclude recombination artefacts caused by PCR amplification (Meyerhans et al., 1990; Odelberg et al., 1995). Whilst it is impossible to exclude completely that some of our observed recombinant species were not PCR artefacts, a number of observations indicate that they were of natural origin. These include: (i) recombinants were not identified in the DO56 inoculum; (ii) recombinants were not detected in chimpanzee CH810, which was infected with multiple HCV subgenotypes and had a viral load similar to that of CH1433; (iii) additional recombinants were amplified with subgenotype-specific primers only from chimpanzee CH921 and not from the other two animals; (iv) using subgenotype-specific primers, the crossover point of one variant was the same as that of the original recombinant, whilst the 1a and 1b sequences were in a different order, which may be indicative of a recombination mechanism different from the polymerase strand switching suggested for PCR-induced recombination (Odelberg et al., 1995); (v) no recombinants were found in a serum-mixing experiment; and (vi) no recombinants were found in a mixing experiment with cloned 1a and 1b fragments.

The recombinant HCV genomes detected in the chimpanzees were most probably generated during the duration of infection, rather than having been transmitted from the DO56 inoculum. Although it is possible that recombinant HCV variants exist in the DO56 inoculum below the level of sensitivity of our detection methods, it is very improbable that these minor variants established infection successfully after transmission to chimpanzees. As was shown in our previous study on the same animals, only major HCV variants were detected that established productive infection (Nainan et al., 2006).

RNA viruses vary greatly in their ability to undergo recombination (Posada et al., 2002). HCV genomic recombination is considered a rare event, whereas GB virus C, a virus related closely to HCV, has been found to have a high frequency of recombination (Worobey & Holmes, 2001). The rate of recombination, calculated as the percentage of recombinant molecules in the 700 bp fragment, varied from ∼1 % (one clone out of 96) for chimpanzee CH1433 to ∼3 % (four clones out of 135) for chimpanzee CH921.

We did not analyse more than one genome region and do not know whether recombination might vary by region. If recombination has equal probability to occur within any region of the HCV genome, then the estimated recombination rate for the entire genome could be 1452 % (the studied 700 bp region represented only ∼7 % of the entire HCV genome). Recombination was detected within 1 month after experimental infection in chimpanzees.

The identification of HCV recombination is hampered by the difficulties in detection of mixed HCV infections, which must be present for recombination to occur. Existing methods to detect mixed HCV infections are not very sensitive or accurate. The reported prevalence of mixed HCV infections in humans ranges from 0 to 40 % and appears to depend on the study population and techniques used for genotype determination (Giannini et al., 1999; Hu et al., 2000; Qian et al., 2000; Viazov et al., 2000). Reliance on analysis of small HCV genome segments makes it difficult to detect genetic recombination and detection could be improved if it were known in which genome regions recombination was likely to occur. The recombination rate found in these experimentally infected chimpanzees appears to be somewhat higher than that found in humans. However, the number of animals studied does not allow an estimate of whether the recombination rate in chimpanzees differs from that in humans. Further research is required to elucidate the actual rate and factors affecting HCV recombination.