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

Abstract

Rodents and bats are now widely recognised as important sources of zoonotic virus infections in other mammals, including humans. Numerous surveys have expanded our knowledge of diverse viruses in a range of rodent and bat species, including their origins, evolution, and range of hosts. In this study of pegivirus and human hepatitis-related viruses, liver and serum samples from Vietnamese rodents and bats were examined by PCR and sequencing. Nucleic acids homologous to human hepatitis B, C, E viruses were detected in liver samples of 2 (1.3%) of 157 bats, 38 (8.1%), and 14 (3%) of 470 rodents, respectively. Hepacivirus-like viruses were frequently detected (42.7%) in the bamboo rat, Rhizomys pruinosus, while pegivirus RNA was only evident in 2 (0.3%) of 638 rodent serum samples. Complete or near-complete genome sequences of HBV, HEV and pegivirus homologues closely resembled those previously reported from rodents and bats. However, complete coding region sequences of the rodent hepacivirus-like viruses substantially diverged from all of the currently classified variants and potentially represent a new species in the Hepacivirus genus. Of the viruses identified, their routes of transmission and potential to establish zoonoses remain to be determined.

Keywords: Vietnam; bats; hepatitis viruses; homologues; pegiviruses; rodents.

Figure 1

Box plot of rodent hepacivirus RNA concentration with each dot representing one sample.

Figure 2

Maximum likelihood trees of (a) the screening fragment and (b) the amino acid region 1123–1566 of hepaciviruses using best-fitting models of rtREV+G+I and LG+G+I, respectively. The magnified box shows the two Vietnamese rodent hepacivirus clades. Labels for sequences obtained in this study are highlighted in bold. Bootstrap support values of ≥ 70% are shown. The trees were drawn to scale; bar, estimated number of substitutions per site.

Figure 3

Maximum likelihood trees of (a) the screening fragment and (b) the ORF1 of hepeviruses using the best-fitting models of LG+G+I and LG+G+F, respectively. Rodent species are shown. Labels for sequences obtained in this study are highlighted in bold. Bootstrap support values of ≥ 70% are shown. The trees were drawn to scale; bar, estimated number of substitutions per site.

Figure 4

Maximum likelihood trees of (a) the screening fragment and (b) the polymerase of hepadnaviruses using the best-fitting models of JTT+G and JTT+G+I+F, respectively. Labels for sequences obtained in this study are highlighted in bold. Bootstrap support values of ≥ 70% are shown. The trees were drawn to scale; bar, estimated number of substitutions per site.

Figure 5

Maximum likelihood trees of (a) the screening fragment and (b) the region 888–1635 of pegiviruses using the best-fitting model of LG+G+I+F. Labels for sequences obtained in this study are highlighted in bold. Bootstrap support values of ≥ 70% are shown. The trees were drawn to scale; bar, estimated number of substitutions per site.