Pandemic Influenza Sequences in Companion Dogs in Korea
Recombinomics Commentary 10:14
April 3, 2008
Partial HA and NA sequences from A/canine/Korea/GCVP01/2007(H3N2) have been released at Genbank. A BLAST analysis of the partial sequences confirm that both sequences are most closely related to bird H3N2 sequences from Korea (see list of the top 100 matches for HA and NA here and here).
However, regions of the sequences can also be found in human seasonal flu. These matches are with pandemic flu. The H3 sequence has regions that are present in H3N2 pandemic strains from 1968, while the N2 sequence has regions that are present in H2N2 pandemic strains from 1957.
The conservation of these sequences for 40-50 years coupled with the detection in H3N2 infected companion dogs in Korea are cause for concern. Although many of the infected dogs died, the recovery of others and the evidence for dog to dog transmission raises concerns that these avian sequences are widely dispersed in companion dogs and provide a reservoir of older sequences which lead to rapid changes in both seasonal and pandemic flu.
The slower evolution of human sequences in other species is easily seen in swine. The H1N1 Canadian swine sequences from 2003/2004 have a human PB1. However, the human PB1 is most closely related to human PB1 sequences from the mid 90’s. This slower evolution of human sequences in swine is common, because the virus in swine is selected for growth in swine, not humans.
Thus, the presence of earlier human sequences in birds or dogs is not unexpected, but avian H3N2 in companion dogs creates opportunities for co-infections and acquisition of these earlier sequences by contempory sequences via recombination. Recent data has demonstrated statistically significant recombination involving short influenza sequences in multiple genes, and the limited dataset and experimental design precluded detection in all eight gene segments. The high rate of recombination between closely related sequences generates short stretches of recombination, which frequently look like single nucleotide polymorphisms whioch are mistakenly thought to be de novo point mutations.
Moreover, since three different serotypes (H3N2, H3N8, H5N1) have recently been found in dogs, the potential for additional serotypes in dogs (or cats) and exchanges of genetic information is high.
These data also highlight serious surveillance shortfalls and a possible increase in trans-species jumps by avian influenza. Screening of archives serum samples for H3N2 antibodies in dogs and cats would be useful.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings