Phylogenetic relationships of aquatic birnaviruses based on deduced amino acid sequences of genome segment A cDNA

Aquatic birnaviruses, such as infectious pancreatic necrosis virus (IPNV), cause serious diseases in a variety of fish species used worldwide in aquaculture and have also been isolated from a variety of healthy fish and shellfish species. These viruses exhibit a high degree of antigenic heterogeneity and variation in biological properties such as pathogenicity, host range, and temperature of replication. To better understand genetic and biological diversity among these viruses, the nucleotide and deduced amino acid sequences were determined from cDNA of the large open reading frame (ORF) of genome segment A of the 9 type strains of Serogroup A and 4 other representative strains of Serotype A1, the predominant serotype in the United States. In addition, nucleotide and deduced amino acid sequences were determined for the VP2 coding region of a variety of isolates representing 5 of the 9 serotypes. VP2 is the major outer capsid protein of aquatic birnaviruses. RT-PCR was used to amplify a 2904 bp cDNA fragment including all but a few bp of the large ORF of genome segment A or a 1611 bp fragment representing the entire VP2 coding region. Nucleotide and deduced amino acid sequences were determined from the PCR products. Pairwise comparisons were made among our data and 2 other aquatic birnavirus sequences previously published. Several hypervariable regions were identified within the large ORF. The most divergent pair of viruses exhibited a similarity of 80.1% in the deduced amino acid sequence encoded by the large ORF. Genomic relationships revealed in a phylogenetic tree constructed from comparison of the deduced amino acid sequences of the large ORF demonstrated that these viruses were clustered into several genogroups. Phylogenetic comparison of the deduced amino acid sequences of the VP2 coding region of 28 aquatic birnavirus isolates, including the type strains of all 9 serotypes, demonstrated 6 genogroups, some of which were comprised of several genotypes. The most divergent pair of viruses exhibited a similarity of 81.2% in the deduced amino acid sequence from the VP2 coding region. In contrast to previous studies of much shorter genomic sequences within the C-terminus-pVP2/NS junction coding region, these genogroups based on the entire large ORF or the VP2 coding region generally correlated with geographical origin and serological classification. Isolates from the major Canadian serotypes were more closely related to the European isolates than to isolates from the United States.