|Home||Founder||What's New||In The News||Contact Us|
|Paradigm Shift Intervention Monitoring||Commentary
More H5N1 Receptor Binding Domain Changes
Recombinomics Commentary 13:35
April 12, 2008
The triple N158S/Q226L/N248D HA mutation (which eliminates a glycosylation site at position 158) caused a switch from avian to human receptor specificity.
The above comments from a recent paper describe a new combination of changes that affect receptor binding specificity in clade 1 H5N1. These data acknowledge that various combinations in H5N1 can affect this specificity and raise additional concerns that the rapid change of H5N1 can lead to the acquisition and fixing of these changes and lead to more efficient transmission of H5N1 to humans.
Clade 1 isolates have tandem glycosylation sites at positions 158 and 159. However, the vast majority of clade 2.2 isolates have eliminated both of these sites reducing the number of required receptor binding changes from 3 to 2. However, different clades may use different combinations to change specificties, so the absence of glycosylation at both positions may create additional combinations.
Clade 2.2 outbreaks have already produced clusters of cases in the Middle East. The first reported clade 2.2 human infection involved S227N, which is adjacent to Q226L described in the paper. Although this predicted change was not found in an isolate from the sister of the index case, the change was found in other human isolate from Turkey. In Turkey there were 21 lab confirmed cases, but only 12 of the 21 were confirmed in England, and sequences from only four of the 12 have been published. Thus, 2 of the 4 human H5N1 sequences from Turkey have S227N. Almost all 21 lab confirmed cases in Turkey were from clusters, suggesting the S227N was widespread, but degradation of samples limited confirmation and isolation of the H5N1 from that 2006 outbreak. S227N was also found in one of the Egyptian isolates in 2006 as well as 2007.
In addition to changes at position 227, other receptor binding domain changes have been found in patients from Egypt. The Gharbiya cluster, in late 2006, lead to the isolation of H5N1 from two patients. Both had V223I and M230I. These changes bracket the change described at position 226. Although human sequences have not been released for this season, poultry isolates have both V223I and M230I. Last season all human isolates with M230I were from fatal cases. In addition, isolates from vaccinated flocks in Egypt have another receptor binding domain change, M230V.
In addition to the receptor binding domain changes in Egypt and Turkey, additional changes have been reported in human H5N1 from Iraq. In Iraq the human H5N1 isolates had two receptor binding domain changes, N186S and Q196R, while human isolates from Azerbaijan had one change, N186K. The isolates from Iraq and Azerbaijan, like the isolates from Turkey and Egypt were from clusters which represented more efficient transmission to humans and/or human to human transmission.
In addition to the elimination of the glycosylation site at position 158, clade 2.2 isolates have PB1 E627K, which increases polymerase activity at 33 C, the temperature of a human nose or throat in the winter. Moreover, M230I is present in seasonal flu (H1N1, H3N2 , influenza B) signaling selective advantage for growth in humans.
Thus, the latest paper describes an additional path to the conversion of H5N1 from an avian to a human specificity, raising concerns of additional combinations of changes that create an altered binding specificity, which when combined with changes described above, are cause for concern.
Recombinomics Paper at Nature Precedings