H5N1 Mutation Media Myth
Recombinomics Commentary 20:18
December 28, 2007
"There is no suggestion that the virus has changed into a form that poses a broader risk," WHO spokesman John Rainford told AFP. "If that had been the case, we would have witnessed more cases of human transmission."
The above comments have taken the WHO spin into the stratospher. WHO consultants have already identified receptor binding domain changes in H5N1. In clade 2.2 these receptor binding domain changes have been associated with larger clusters involving human transmission. In Pakistan, most of the patients who were lab confirmed locally were linked to clusters.
The larger cluster appears to be the longest ever reported for H5N1 and was sustained for more a month. Although WHO hasn’t provided the details in a WHO situation update, the disease onset dates were cited in media reports quoting WHO officials.
Receptor binding domain changes can lead to more efficient binding to human receptors. This affinity leads to larger clusters and human to human transmission. In clade 2.2 this change is of concern because another change, PB2 E627K, has become fixed.
Prior to the outbreak at Qinghai Lake in May, 2005, E627K had never been reported in H5N1 from bird. It was found in some H5N1 from patients and was also in the only reported bird flu fatality that wasn’t H5N1, the H7N7 fatality in 2003 in the Netherlands. It also increased H5N1 virulence in mice and was associated with an increase in viral replication at lower temperatures (33 C). This temperature corresponds to the temperature of a human nose or throat in the winter. Consequently E627K is a human polymorphisms and has been found in all human flu isolates dating back to 1918.
Therefore, changes in the receptor binding domain are of concern and the association of receptor binding domain changes with clusters and human to human transmission has been clear, especially for the Qingahi strain (clade 2.2).
A prediction of such a change, S227N, in Qinghai H5N1 flying into the Middle East was predicted based on donor sequences in endemic H9N2. The first Qinghai H5N1 human case was in Turkey, and the isolate from the index case had S227N. Although the change wasn’t in the isolate from the sister of the index case, it was in a second isolate from Turkey. Thus, two of the four Turkey isolates had S227N. In Turkey 21 cases were lab confirmed and virtually all cases were from lab confirmed clusters.
The outbreak in Turkey was followed by an outbreak in Iraq. Although only three patients were H5N1 confirmed in Iraq, two were in a cluster and both had two receptor binding changes, N186S and Q196R.
The outbreak in Iraq was followed by an outbreak in Azerbaijan,and again the vast majority of the cases were in clusters and again another receptor binding domain change, N186K, was found in all human isolates from Azerbaijan.
The next human outbreak was in Egypt, and the vast majority of cases there were not clusters. However, a year ago Egypt did have a cluster of three, the largest reported there to date. H5N1 was isolated from two of the three fatal infections and both had two receptor binding domain changes, V223I and M230I. M230I was another “Human” polymorphism found in all human seasonal flu isolates, including influenza B
Thus, the association of receptor binding domain changes with larger clusters and increased human to human transmission for clade 2.2 Qinghai H5N1 is quite clear, and similar changes in the HA sequence from the sole confirmed positive in Pakistan is likely.
In Pakistan, 8 of the 9 patients testing positive locally were in clusters. Six were in the larger cluster while two were in the smaller cluster. Only one positive was not in a cluster.
Although a receptor binding domain is not a certainty (isolation procedures can select away from receptor binding domain changes present in the H5N1 from the patient), it is likely that the H5N1 from Pakistan will be clade 2.2 (Qinghai), which has been true for all H5N1 cases west of China, and the H5N1 in the patient will have a receptor binding domain change.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings