H5N1 Isolation Failures at USDA Raise Pandemic Concerns
Recombinomics Commentary
October 18, 2006
The USDA National Veterinary Services Laboratories (NVSL) confirmed the presence of H6N2 through virus isolation in a pool of five samples of the 11 samples collected from wild Green-winged Teals in the Rice Lake Conservation Area of Fulton County, Illinois. Initial screening results announced on Sept. 29 indicated that H5 and N1 subtypes might be present in the collected samples, but further testing was necessary to confirm the H and N subtypes as well as pathogenicity.
The failure to isolate the H5N1 in the Green-winged Teal samples from Illinois is cause for concern. This failure may be due in part to pooling of samples, which can lead to an isolation failure because of overgrowth by a serotype not detected in the testing of the pooled samples. A similar failure was reported for Northern pintail samples from Montana. One sample was positive for H5N1, but H5N3 was detected even though sixteen samples were H5 positive. It remains unclear if the H5N3 serotype was from samples that were H5 positive but N1 negative, or was from the H5N1 positive sample. In any even, no isolation of H5N1 in Montana was reported.
The number of isolation failures remains unclear. H5N1 has been isolated in Maryland, Michigan, and Pennsylvania. H5 was detected in a live market in New Jersey, but no mandatory OIE report was filed. Media reports suggested it was not H5N1, but all H5 infected poultry, regardless of N serotype, requires an OIE report. Similarly, media reports indicate H5 has been found in California, Washington State, and possibility Alaska, but addition detail from the USDA have not been announced. Similarly, serotypes other than H5N1 are announced only for H5N1 positive samples. It is unclear what, if anything is isolated from the H5 positive samples.
These isolation failures raise concerns that low levels of HPAI H5N1 could be missed due to pooling of samples, low sensitivity in the isolation procedure, or other technical issues including collection / transport / degradation issues.
Isolation of HPAI H5N1 from health wild birds is rare. Although such isolations have been reported in Russia, the vast majority of H5N1 isolates have come from dead birds or live poultry on farms that have fatal H5N1 infections.
Thus, although isolation of HPAI H5N1 from live wild birds is rare, the above protocol involves pooling samples, lowering the sensitivity of the isolation procedure. Repeated failures to isolate any H5N1 from H5N1 positive samples raise concerns about assurances on the detection of low path H5N1.
The USDA testing cannot exclude mixtures of low path and high path H5N1 in samples that are H5N1 positive.
H5 was detected in a dead gosling on Prince Edward Island. Only one dead goose was tested, but four died suddenly after displaying bird flu symptoms. Low path H5 rarely causes sudden death in waterfowl, although the Qinghai strain of H5N1 is known to kill waterfowl, including thousands of bar headed geese at Qinghai Lake in May, 2005. This strain has been transported and transmitted by long range migratory birds into Europe, the Middle East, and Africa. The H5N1 has been found in flyways that connect to northwestern and northeastern North America. Sequence data from North American H5 isolates have polymorphisms that are predominantly found in Asia. Similarly, North American polymorphisms have been found in the Qinghai stain in Asia.
Recent sequence data from China, shows that recombination is extensive in H5N1 infect poultry and wild birds. These recombination events have moved sequences into Indonesia and have generated swine H5N1 sequences with polymorphisms found in the Qinghai strain of H5N1 as well as H5N1 found in human isolates on Java as well as isolates from the Karo cluster.
Thus far, only four genes of an H5N2 isolate from British Columbia in August, 2005 has been released. This isolate has acquired North American swine sequences, as well as HPAI H5N1 sequences from Asia. These acquisitions indicate dual infections and associated recombination is common in H5 in North America. No other H5 sequences from 2005 isolates in Canada or any 2006 H5 isolates from the United States or Canada have been released. The explosion in the detection of H5 in Canada last year, as well as Canada and the United State this year raises concerns about the evolution of H5 in North America as well as the repeated isolation failures at the USDA.
Media Sources
Phylogenetic Trees