Recombinomics Commentary 13:45
April 17, 2012
With all four mutations, the virus spread even more easily. Kawaoka exposed six more uninfected ferrets to sick peers in adjacent cages. Within a week, he had found signs of the virus in all of them.
Three of these mutations are new, at least in public databases.
The above comments describe the effects of the four H5 changes (N158D, N224K, Q226L, T318I) that were associated with ferret to ferret transmission via aerosol in the Kawaoka experiments. However, the claim that only one of the four changes (N158D) is present in public data bases is false. N224K is present in waterfowl in Egypt (A/duck/Egypt/10185SS/2010) and Vietnam (A/Muscovy duck/Vietnam/NCVD-11/2007), while T318I is present in waterfowl in China (A/mallard/Xuyi/10/2005, A/spotbill duck/Xuyi/18/2005, A/duck/Guangxi/53/2002).
Although Q226L has never been reported in natural H5, the ability of N224K to substitute for the other receptor binding domain (RBD) change, G228S, raises concerns that H5N1 is naturally evolving toward human transmission, since N224K has been found in natural sequences.
The presence of this RBD change in a duck in Egypt, A/duck/Egypt/10185SS/2010, is of particular concern because H5N1 in Egypt is clade 2.2 and therefore has already fixed N158D in H5 and E627K on PB2, which was also present or introduced in transmission studies by the CDC or Ron Fouchier (and use of an H1N1pdm09 PB2 in the Kawaoka studies substituted for the E627K change).
Moreover, the above sequence is from clade 2.2.1 G, which also has S133del and T155I, which has been linked to a seasonal H1N1 profile via bioinformatic analysis. This association has also increased concerns because all public human H5N1 sequences from Egypt since mid-2009 are clade 2.2.1 G, which have S133del, T155L, N158D and PB2 E627K.
In addition, recently released H5N1 sequences from Egypt have PB1 and PB2 acquisitions of significant stretches of seasonal H1N1, H1N1pdm09, and H3N2v, supporting recombination between H5N1 and influenza sequences that are linked to efficiently transmitting human outbreaks in the past or present.
The changes however, are not being carefully analyzed due to a lack of transparency, coupled with limit sequencing. NAMRU-3 does the sequencing of human H5N1 cases in Egypt, and there is virtually no data on sequences of internal genes. Although NAMRU-3 sequences are typically limited to HA and NA, these sequences have also been withheld from public databases. The last release was in mid-2010 and included sequences from 2009 to mid 2010. This, release of most of those sequences was delayed for 3-12 months. However, the delay time has increased, since no human H5N1 sequences have been released in almost two years. 2011 sequences were released to WHO, and these sequences are represented in phylogenetic trees in WHO updates on pandemic vaccine targets, but these trees only represent a subset of sequences, and do not provide information on the acquisition of polymorphisms such as H5 N224K or T318I or changes in internal genes.
The potential for such H5 acquisitions is high because wild birds fly into the region each year because of overlapping flyways in Egypt. Moreover, clear examples of recombination have been noted in H5N1 in Egypt, which included an silent NA acquisition, G743A, as well as S133del coupled with T155I.
The concurrent acquisition of G743A provided compelling evidence for recombination since it appeared on multiple clade 2.2 backgrounds in Egypt, as well as Kuwait, Russia, Ghana, and Nigeria in early 2007 (followed by spread throughout Europe in 2007).
The acquisition of S133del couple with T155I played a role in the seasonal H1N1 bioinformatic profile, and the role of recombination was highlighted by the acquisition of both changes in Egypt clade 2.2 and Indonesian clade 2.1.
Thus, the accumulation of significant changes in Egypt H5, coupled with a lack of transparency, increases pandemic concerns.
No comments:
Post a Comment