Detection and Characterization of Homologues of Human Hepatitis Viruses and Pegiviruses in Rodents and Bats in Vietnam

Van Nguyen, Dung; Van Nguyen, Cuong; Bonsall, David; Ngo, Tue Tri; Carrique-Mas, Juan; Pham, Anh Hong; Bryant, Juliet E.; Thwaites, Guy; Baker, Stephen; Woolhouse, Mark; Simmonds, Peter

Amplicons derived from PCR screening experiments were all successfully sequenced and complete or near complete genome sequences were determined for representatives of the viruses by PCR or deep sequencing. After exclusion of primers, the obtained screening sequences from each targeted virus clustered in one or two clades of those with high identity and were most closely related to sequences previously reported from rodents or bats from the US [17, 18] , China [7] and Vietnam [16] . Rodent hepacivirus sequences (360 nucleotides) formed two well supported clades (Figure 2a) . Clade 1 included all of 35 sequences from Rhizomys pruinosus which shared 84.5-100% pairwise nucleotide identity while three sequences (nucleotide identity of 89-99%) from Rattus losea and Rattus argentiventer grouped in clade 2. These clades differed from each other by mean distances of 39.6% and 33.2% at nucleotide and amino acid levels, respectively. While the four Vietnamese bamboo rat hepacivirus genomes were highly similar to each other (<12% nucleotide and <3% amino acid distances in the complete coding region (cds)), they were remarkably different from the closest sequence (KC815310) with the corresponding distances of 40% and 36%, respectively ( Table 2 ). The amino acid p-distances of the obtained hepacivirus sequences and KC815310 in the regions 1123-1566 and 2536-2959 (numbered relative to M62321) were 30% and 32%, respectively. The newly identified hepaciviruses therefore could be provisionally classified as members of a new hepacivirus species (Figure 2b ) [32] . The other bamboo rat hepaciviruses in clade 1 may also belong to this new virus species on the basis of the high sequence identity in this clade. Similarly, although complete genome sequences were not determined for hepaciviruses in clade 2, they likely belong to species Hepacivirus G due to their low amino acid p-distances (7.6-8.4%) in the screening region to KJ950938. The 5' untranslated region sequences of these hepaciviruses contain two miR-122 seed sites (CACUCC), which were located 51 nucleotides from each other, as consistent with the suspected hepatotropism of the viruses.