H5N1 Receptor Binding Domain Changes in Gharbiya Cluster
Recombinomics Commentary
January 11, 2007
The HA sequences from two members of the Gharbiya cluster were released by Genbank today. These sequences were deposited December 28, 2006 by NAMRU-3, one day after confirmation of the cluster in the WHO update. Once again US NAMRU-3 has set the standard for prompt release of critical sequence data.
The sequences from this cluster were important because all three cluster members died and lived 12 miles from an earlier case in Gharbiya, located in the heart of the Nile Delta. The earlier case has M230I, which was the first Qinghai isolate with that change. The change was cause for concern because all three human strains of influenza (H1N1, H3N2, and influenza B) have that polymorphism, as do serotypes efficiently transmitted from mammal to mammal such as H3N8 in dogs and horses, or H7N7 in horses and seals. Moreover, the acquisition creates a region of identity between influenza B and H5N1 at positions 226-230 (QSGRI).
The HA sequence from the 26M (A/ Egypt/14724-NAMRU3/2006) and 15F (A/ Egypt/14725-NAMRU3/2006) both have M230I suggesting the change is circulating in wild birds in the Qinghai strain in Egypt. The isolates have a number of regional markers found in H5N1 from birds or patients from the spring of 2006. Thus, the M230I was appended onto an Egyptian Qinghai genetic background. M230I has been reported in H5N1 isolates from Vietnam and Guiyang Province in China in 2005, suggesting that the polymorphisms were acquired by recombination.
Both isolates however, also have another change in the receptor binding domain, V223I. This change is also found in recent H5N1 isolates. It is in a bar-headed goose isolate from Mongolia from 2005. Mapping of newly acquired polymorphisms trace migratory paths, which include Qinghai polymorphisms found in Mongolia. However, the polymorphism is also found in a series of Shantou goose isolates from 2006, which are also clade 2, but distinct from the Qinghai strain or the Fujian strain. The appearance of V223I in the Shantou geese as well as Egyptian Qinghai isolates, again indicates these acquisitions are by recombination, which is supported by the NA sequence from the cluster.
These receptor binding domain acquisitions are cause for concern. The genetic diversity of H5N1 is increasing markedly giving rise to three distinct Clade 2 sub-clades causing human fatalities. The number of changes in or near the receptor binding domain is increasing, and as seen in the two sequences from the Gharbiyah cluster, these polymorphisms are recombining and producing novel receptor binding domain combinations.
The possible fixing of M230I is cause for concern. The Qinghai strain has already fixed E627K in PB2, which, prior to the isolate of Qinghai H5N1 in May of 2005, E627K had never been reported in H5N1 isolated from a bird. This E627K change is found in all human H1 ,H2, and H3 isolates and was tied to an increased polymerase activity at lower temperatures. The acquisition of this polymorphism by the Qinghai strain raised concern that the strain would be more active in mammals, especially during the flu season when temperatures were lower. This concern was justified when H5N1 caused human infections in multiple countries in the Middle East, including Turkey, Iraq, Egypt, Djibouti, and Azerbaijan. Many of the human isolates were associated with changes that affect receptor binding, including S227N, N186K, and Q196R. The isolates from the Gharbiya cluster add M230I and V223I to the list of changes and raise additional concerns about M230I, which is present in seasonal flu and is in all three Qinghai sequences from Egypt this season.
It will be important to sequence wild and domestic bird sequences in the region to determine how widespread the M230I is, and if V223I has also become established in H5N1 in birds in the region.
Media sources
Phylogenetic Trees