Human Avian H7N2 Reassortant Raises Pandemic Concerns
Recombinomics Commentary 22:22
April 28, 2008
The recent release of H7N2 sequences from a patient infected in New York in late 2003, A/New York/107/2003(H7N2), raises transparency / surveillance issues. Although the four month delay in detecting the case was explained in early 2004, it remains unclear why it took four more years to release sequences from seven of the eight gene segments, and it is also unclear why the PB2 sequence has not been released.
The released sequences are clearly avian for the H7, N2, and NP genes. The other four genes, PB1, PA, MP, NS are from a human H3N2 circulating in 2003, the time of the H7N2 infection. Since the polymerase A and polymerase B1 genes were human, it seems likely that the polymerase B2 (PB2) gene would also be human, which is cause for concern. All PB2 sequences from human H3N2 have E627K, which allows the virus to replicate faster at lower temperatures (33 C). There has only been one reported bird flu fatality that was not H5N1. It also was H7 (H7N7) and had E627K. Thus H7N2 with human PB2 would likely grow more efficiently in humans, which could cause problems since H7 is already efficient transmitted human to human, but levels are low and disease is usually mild. However, like the NY case, several H7N2 cases in England last year were hospitalized, and media reports described a severe clinical presentation.
The PB2 gene also commonly acquires human polymorphisms via recombination. The H5N1 PB2 from the vaccine resistant isolate in Israel had a number of newly acquired human polymorphisms raisiong the possibility that the PB2 in the H7N2 isoalte was a recombinant.
Moreover, detection of H7 remains a major problem. Although Canada reported widespread low path H5 infections in 2005 and 2006, they failed to detect H7 in any wild birds, even though there have been H7N3 outbreaks in the British Columbia in 2005 and Saskatchewan in 2007. Similarly, England failed to detect H7 in wild birds prior to H7N3 and H7N2 outbreaks there in 2006 and 2007.
Moreover, lab confirmation of H7 in patients in England was problematic. More humans than birds had symptoms, but lab confirmation, including index cases who owned affected farms, tested negative. These cases had symptoms, which was unusual for late spring in England in 2007.
The presence in human and avian genes in the H7N2 isolate indicates that the same host was infected with both viruses. Since one virus was human H3N2 in circulation in New York in 2003, it is likely that the dual infection was in a human host, but since H7 is readily transmitted human to human, the reassortment may have been in a human host that subsequently infected the H7N2 positive patient. The source of the infection was not determined and the case did not have a history of contact with birds.
Thus, although the H7N2 was isolated from a patient in New York 4 years ago, the poor surveillance in wild birds and humans raises concerns that other reassortants are in circulation in birds or humans.
The PB2 sequence and an explanation for the four year delay in the release of the data would be useful.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings