Topological Glycan Changes Targeted By H5N1
Recombinomics Commentary 23:37
January 6, 2008
The senior author of the study said it gives scientists a way to spot avian flu viruses that are becoming more transmissible to and among people.
The above comment by the senior author of the Nature Biotechnology paper, "Glycan topology determines human adaptation of avian H5N1 virus hemagglutinin", places the emphasis on changes in the receptor binding domain of the H5N1, not on changes in receptors of hosts, which is one of the popular widely circulated genetic predisposition media myths..
The sequences of H5N1 have already yielded receptor binding domain changes that increase affinity for 2.6 gal human receptors, but the assays did not represent the true receptors very faithfully, so human and avian receptor differences required high concentrations of virus. Consequently, differences were highly dependent on conditions, and minor variations could create conflicting results. Similarly, the high concentrations of virus reduces the ability to detect intermediate changes, which are involved in the evolution of H5N1.
Thus, H5N1 that had acquired intermediate changes produced larger clusters, but the changes didn’t achieve the affinities in seasonal flu, so the larger clusters did not lead to a full blown pandemic.
However, H5N1 evolves via recombination, so combinations of changes that produce intermediate changes can lead to synergies, which produce more dramatic effects when these changes are combined.
The new assay should be more quantitative and there should be a continuum of binding affinities defined by various combinations of changes.
The association of these changes with large cluster is strong for the Qinghai strain (clade 2,2). The clusters linked to this sub-clade were associated with a series of receptor binding domain changes and in some cases the H5N1 had acquired two such changes, as seen in Iraq (N186S and Q196R) and Egypt (V223I and V230I). Moreover, one of the changes, S227N, was associated with patients in Turkey as well as Egypt. However, because these changes lead to different tissue tropisms, the levels of these changes can vary in isolates from the upper or lower respiratory tract, which can generate difference in isolates from cluster members, as was seen for S227N in Turkey, and is the likely cause of the failure to find receptor binding domain changes in Pakistan.
These failures highlight additional limits of assays used by the WHO and consultants who draw conclusions and set policy. The assays are largely confirmatory which work well when tissues from multiple dead or dying host are available, but tend to produce abysmal results when samples are limited, which is frequently the case in clinical situations.
Thus, in Pakistan, H5N1 was isolated from one in ten infected patients, and in Turkey, H5N1 was isolate from four in 21 infected patients. Moreover, the borderline confirmatory tests work best on strong positives, and like the older receptor binding assays, produce conflicting data.
This new assay for receptor binding domain changes should serve as a reminder that policy and conclusions based on negative data are likely to be hazardous to the world’s health.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings