Commentary
Concurrent Acquisition of H274Y Tamiflu Resistance in Japan
Recombinomics Commentary 19:00
July 11, 2008
The preliminary analysis suggests that the resistant mutants do not share a single origin and further genotypic analysis is ongoing.
The above comment from the last WHO update entitled "Influenza A(H1N1) virus resistance to oseltamivir" confirms the analysis of public H1N1 sequences as well as the sequences represented on the NA phylogenetic tree of recent isolates from Japan.
The sudden appearance of oseltamivir resistance worldwide has startled influenza "experts" because the appearance was sudden and not linked to oseltamivir treatment of seasonal flu. The resistance was limited to one position, H274Y, on H1N1, which is the same dominant change in H5N1. This change first began to appear in public sequences from the 2006/2007 season in isolates from the United States. Those isolates were from the New Caledonia strain, and similar results were not reported elsewhere, although sequences from the United States make up the vast majority of public NA sequences.
However, early this year Norway reported H274Y levels exceeding 50% of H1N1 NA sequences, which sounded alarm bells worldwide, in part because oseltamivir was a front line anti-viral commonly used to treat H5N1 patients and also to be used as a blanket to prophylacticly treat wide areas with confirmed H5N1. H274Y was subsequently reported in H5N1 infected wild birds, which was then followed by the increase in H274Y in the 2006/2007 season, and the explosion of cases in the 2007/2008 season.
Analysis of public sequences from the United States indicated that the New Caledonia strain of 2006/2007 was replaced by the Solomon Island/3 and Brisbane/59 strains in 2007/2008. The H274Y in the US in the most recent season was limited to the Brisbane strain, but the analysis of the sequences identified three independent introductions. The largest group was found throughout the United States and matched sequences from England and Turkey. The latest data from Japan indicates it also is closely related to sequences from Norway, France, and Japan. These isolates could all come from a common introduction and are shaded and labeled "Northern EU-like" in the phylogram from Japan.
However, in the United States, two isolates from Hawaii were one a separate branch, with additional isolates from Hawaii and California which did not have H274Y, indicating the Hawaiian samples, which are listed in the phylogram and labeled "Hawaii-like". However, the branch with the two positives from Hawaii also has an isolate from Miyagi once again showing that the acquisition of H274Y was an independent event that happened after this branch was formed. Similarly, seven additional isolates from Japan form a closely related branch, but only four of the seven isolates on that branch have H274Y. Similarly another closely related branch have five isolate that have H274Y. The isolates from Yokohama are epidemiologically linked but they are also related to an isolates from Yamagata, which also doesn't have H274Y, again indicating that these branches represent additional introductions.
Similarly, in the United States, a recent isolate from Florida is on teh same seperate branch with other isoaltes from Florida, but only one of the three Florida sequences has H274Y.
The phylogenetic tree from Japan has additional examples of H274Y mapping to more branches, which contain isolates without H274Y, again indicating that the H274Y isolates on those branches have also acquired H274Y independently.
This simultaneous acquisitions of the same polymorphisms onto multiple genetic backgrounds is similar to the results described for another NA polymorphisms, G743A, found on multiple clade 2.2 H5N1 genetic backgrounds. These acquisitions also appeared concurrently on multiple genetic backgrounds found in divergent locations, which can be most easily explain by recombination with a common source.
Thus, the acquisition of H274Y onto multiple genetic backgrounds provides additional support for acquisition via recombination from a common source. In the case of H274Y, the initial acquisition may have come from H5N1 in 2006/2007, but the explosion of cases this season is likely due to movement of H274Y from one Brisbane/59 genetic background to another as was seen in the case of G743A on clade 2.2 genetic backgrounds.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings