Characterization of low pathogenicity avian influenza viruses isolated from wild birds in Mongolia 2005 through 2007

Abstract

Background: Since the emergence of H5N1 high pathogenicity (HP) avian influenza virus (AIV) in Asia, numerous efforts worldwide have focused on elucidating the relative roles of wild birds and domestic poultry movement in virus dissemination. In accordance with this a surveillance program for AIV in wild birds was conducted in Mongolia from 2005-2007. An important feature of Mongolia is that there is little domestic poultry production in the country, therefore AIV detection in wild birds would not likely be from spill-over from domestic poultry.

Results: During 2005-2007 2,139 specimens representing 4,077 individual birds of 45 species were tested for AIV by real time RT-PCR (rRT-PCR) and/or virus isolation. Bird age and health status were recorded. Ninety rRT-PCR AIV positive samples representing 89 individual birds of 19 species including 9 low pathogenicity (LP) AIVs were isolated from 6 species. A Bar-headed goose (Anser indicus), a Whooper swan (Cygnus cygnus) and 2 Ruddy shelducks (Tadorna ferruginea) were positive for H12N3 LP AIV. H16N3 and H13N6 viruses were isolated from Black-headed gulls (Larus ridibundus). A Red-crested pochard (Rhodonessa rufina) and 2 Mongolian gulls (Larus vagae mongolicus) were positive for H3N6 and H16N6 LP AIV, respectively. Full genomes of each virus isolate were sequenced and analyzed phylogenetically and were most closely related to recent European and Asian wild bird lineage AIVs and individual genes loosely grouped by year. Reassortment occurred within and among different years and subtypes.

Conclusion: Detection and/or isolation of AIV infection in numerous wild bird species, including 2 which have not been previously described as hosts, reinforces the wide host range of AIV within avian species. Reassortment complexity within the genomes indicate the introduction of new AIV strains into wild bird populations annually, however there is enough over-lap of infection for reassortment to occur. Further work is needed to clarify how AIV is maintained in these wild bird reservoirs.

Figure 1

Phylogenetic trees of internal protein genes. Trees include all avian influenza virus isolates collected from wild birds in Mongolia 2005-2007 and selected reference isolates. Trees are shown for all 8 segments of each isolate as follows: A) NS, B) M, C) NP, D) PA, E) PB1 and F) PB2. Trees were constructed with merged duplicate runs of BEAST v. 1.4.8 using HKY substitution, empirical base frequency, Gamma heterogeneity, codon 2 partitions, relaxed lognormal clock, Yule Process tree prior with default operators with UPGMA starting tree and MCMC length of 10⁷. Posterior values are shown at the nodes. Isolates collected during this study are shown in red font.

Figure 2

Phylogenetic trees of HA and NA genes. Trees include all the HA and NA genes from avian influenza virus isolates collected from wild birds in Mongolia 2005-2007 with selected reference isolates. Trees are shown for all 8 segments of each isolate as follows: A) H3, B) H12, C) H13, D) H16, E) N3 and F). N6. Trees were constructed with merged duplicate runs of BEAST v. 1.4.8 using HKY substitution, empirical base frequency, Gamma heterogeneity, codon 2 partitions, relaxed lognormal clock, Yule Process tree prior with default operators with UPGMA starting tree and MCMC length of 10⁷. Posterior values are shown at the nodes. Isolates collected during this study are shown in red font.

Figure 3

Location of the sampling sites in Mongolia by year