Recombination in Human Swine and Avian Influenza
http://www.recombinomics.com/News/03270804/Human_Swine_Avian.html
Recombinomics Commentary 18:20
March 27, 2008
Specifically, because our analysis is necessarily based on viral consensus sequences rather than the myriad individual viral molecules that characterize any infection, it is equally plausible that the ‘recombinants’ detected here in fact represent cases of mixed infection in individual hosts followed by the amplification and sequencing of different viral molecules, thereby producing laboratory-generated artificial recombinants. Hence, to demonstrate conclusively the occurrence of homologous recombination in influenza A virus it will be necessary either to clone (or plaque purify) and sequence multiple viral genomes from an individual host and demonstrate the presence of the recombinant and both parental genotypes within the sample (1), or to show that recombinant sequences form a distinct circulating lineage, with readily identifiable parents, that is transmitted among multiple individuals in a population (30).
The above comments are from the ahead of press publication, “Homologous Recombination is Very Rare or Absent in Human Influenza A Virus”. Although the paper presented data showing that the likelihood that the short regions of recombination in NA H3N2 human influenza was due to chance was less than a billion to one, the authors relied on lab artifact to maintain their position that homologous recombination doesn’t happen in influenza in general, or in human influenza in particular.
However, the requirements delineated in their discussion have already been met for human, swine, and avian influenza, so going through the public data is worthwhile.
Although the paper excluded the best evidence for clear cut homologous recombination in human influenza, the sequences are public and meet the above requirements. All six sequences from six individuals in South Korea have the same recombination. All have a version of human H3N2 human influenza in circulation worldwide at the time. However, each isolate has a 1991 H3N2 HA sequence in the center third of the gene, thus meeting the requirement of the same recombinant transmitted among multiple individuals in a population. Since all six sequences were generated in the same lab, ad hoc arguments for lab error could be developed, but would have to include contamination of all six isolates with a 1991 lab sequence resulting in the contaminant replacing the center of the 2002 sequence resulting in three similar sequences and three additional sequences where 2002 sequences nested inside of the 1991 sequence. Although such a scenario would be possible, it would be highly unlikely.
The authors cited a Recombinomics preprint at Nature Precedings, which met the “proof” criteria for swine influenza. Recombination evidence was presented for all 8 gene segments, but the most compelling examples were for PB2 or PA and involved sequences from two closely related 1977 isolates from Tennessee. The data for PB2 had clear evidence for recombination, involving exact matches for large portions of the gene, even though the Canadian swine isolates were from 2003/2004, while one set of parental sequences were from 1977.
One of the clearest examples of nesting can be seen in three of the isolates, 11112, 57561, and 56626. All three have sequences which exactly match a 1998 North Carolina sequence between positions 755 and 1594. This sequence remains intact in 11112, but the 1977 sequence from Tennessee is nested in the other two sequences between positions 1006 and 1326. All of the above are exact matches and represent just a subset of the recombination events depicted in the PB2 sequences. Thus, the shared recombination events are present in multiple swine isolates and would once again satisfy the requirement of the same recombination event in multiple isolates. Once again, these sequences were generated by a single lab, but generation of this data through contamination would require contamination with multiple isolates, including the two mentioned above as well as a 2002 isolate from Korea. Similar large sets of contaminating sequences would be required to generate the recombination events depicted in the other gene segments.
In addition to the large regions of recombination in the human and swine isolates described above, another Recombinomics paper at Nature Procedings provides evidence for recombination between closely related H5N1 avian NA sequences depicted by a single nucleotide polymorphisms, G743A. These data include plaque purification of isolates from one host providing evidence that the same polymorphism was acquired by two distinct clones present in a single host. The two consensus sequences, each represented by multiple clones, differed from each other at 11 positions. One set of sequences differed from one of the Gharbiya cluster sequences at just two positions and one of the two positions was G743A. The other set of sequences matched two other isolates, and differed from related sequences isolated months earlier and differed at three positions, one of which was G743A. Thus, both distinct sequences had acquired the same change, even though the number of changes was only 2 or 3 positions in the two instances. The likelihood that both sequences made the same copy error at the same time is remote.
This type of coincidence is even more remote when the same change was found on multiple additional genetic backgrounds in Egypt, Kuwait, Russia, Ghana, and Nigeria. Moreover, G743A was also on sequences related to the Kuwait sequence in the Czech Republic, multiple locations in Germany, and Krasnodar. Like the first two examples, the acquisitions on the various backgrounds were small numbers (2-6 positions) of changes between the 2007 sequences and closely related sequences from 2006 which did not have the change.
Thus, the above description delineate homologous recombination in human, swine, and avian influenza which are unlikely to be due to lab error or contamination.
