Characterization of low-pathogenicity H5N1 avian influenza viruses from North America

Abstract

Wild-bird surveillance in North America for avian influenza (AI) viruses with a goal of early identification of the Asian H5N1 highly pathogenic AI virus has identified at least six low-pathogenicity H5N1 AI viruses between 2004 and 2006. The hemagglutinin (HA) and neuraminidase (NA) genes from all 6 H5N1 viruses and an additional 38 North American wild-bird-origin H5 subtype and 28 N1 subtype viruses were sequenced and compared with sequences available in GenBank by phylogenetic analysis. Both HA and NA were phylogenetically distinct from those for viruses from outside of North America and from those for viruses recovered from mammals. Four of the H5N1 AI viruses were characterized as low pathogenicity by standard in vivo pathotyping tests. One of the H5N1 viruses, A/MuteSwan/MI/451072-2/06, was shown to replicate to low titers in chickens, turkeys, and ducks. However, transmission of A/MuteSwan/MI/451072-2/06 was more efficient among ducks than among chickens or turkeys based on virus shed. The 50% chicken infectious dose for A/MuteSwan/MI/451072-2/06 and three other wild-waterfowl-origin H5 viruses were also determined and were between 10(5.3) and 10(7.5) 50% egg infective doses. Finally, seven H5 viruses representing different phylogenetic clades were evaluated for their antigenic relatedness by hemagglutination inhibition assay, showing that the antigenic relatedness was largely associated with geographic origin. Overall, the data support the conclusion that North American H5 wild-bird-origin AI viruses are low-pathogenicity wild-bird-adapted viruses and are antigenically and genetically distinct from the highly pathogenic Asian H5N1 virus lineage.

FIG. 1.

Nucleotide phylogenetic tree of the H5 HA1 from selected North American-origin wild aquatic bird isolates (from this report and previously reported) and selected isolates from avian and mammalian virus species rooted to A/WhooperSwan/Mongolia/244/05. Isolates used for antigenic characterization are in boldface, and isolates used in animal studies are underlined. The tree was constructed with PAUP* 4.0b10 (Sinauer Associates, Sunderland, MA) using maximum parsimony, a heuristic search, and 500 bootstrap replicates (bootstrap values are shown on tree). Canadian provinces and U.S. states are abbreviated by their standard two-letter postal codes.

FIG. 2.

Nucleotide phylogenetic tree of N1 NA from selected North American-origin wild aquatic bird isolates (from this report and previously reported) and selected isolates from avian and mammalian virus species rooted to A/WhooperSwan/Mongolia/244/05. The isolate used for antigenic characterization is in boldface, and isolates used in animal studies are underlined. The tree was constructed with PAUP* 4.0b10 (Sinauer Associates, Sunderland, MA) using maximum parsimony, a heuristic search, and 500 bootstrap replicates (bootstrap values are shown on tree). Canadian provinces and U.S. states are abbreviated by their standard two-letter postal codes.