H5N1 Sequence Hoarding Should End
Recombinomics Commentary
November 10, 2006
Jia said the Fujian-like strain, which Guan said had emerged in March 2005, was actually the same as bird flu viruses found in Hunan in February 2004 in terms of genetic sequencing.
"Guan said he wanted to alert the world with the paper, but why didn't he report the markets with virus-carrying birds to the government if he truly believed in his findings?" Jia asked.
He said there were 10 confirmed poultry outbreaks in seven provinces of China this year, adding 95 percent of domestic birds had been vaccinated.
The above comments extended the back and forth disagreements about relatively minor points, and create distractions from the real issue, which is the release of sequences from H5N1 in China. Much of the discussion centers on splitting hairs which is driven by different definitions. Many sequences have been published, so the overall picture of H5N1 spread and evolution are clear.
The definition of the "Fujian-like" strain in the PNAS paper really centers on changes in the H5N1 sequence in or near the receptor binding domain. All parties are well aware of this novel cleavage site, LRERRRK_R, which was first reported in Nature over a year ago. The partial HA sequence of the isolate, A/duck/Fujian/1734/2005, included the full sequence of the HA cleavage site. Earlier isolates, dating back to 2003, had lost a K. These isolates were from Hunan and Taiwan. The Taiwan isolate was from a duck being smuggled in from Fujian province. Both sequences were published in 2004.
China's Ministry of Health was well aware of the 1734 sequence, because it was used in the MOH report dated January 20, 2006. Moreover, the HA cleavage site of published and unpublished human isolates showed that all had the new Fujian cleavage site. Hong Kong University also published the sequences in March, 2006 showing that isolates from Malaysia and Laos had the novel cleavage site as did sequences from wild birds in Hong Kong, published in June. Moreover, the WHO pandemic vaccine targets included one of the sequences from one of the fatal human cases in Anhui, which, like all human H5N1 cases from China, had the novel cleavage site.
Thus, all parties were well aware of the spread of the Fujian strain long before the PNAS paper was published. the paper had about 250 HA sequences with the Fujian cleavage site. However, there were approximately 150 additional sequences that had different cleavage sites, many of which were also novel.
Moreover, many of the sequences had changes in the receptor binding domain, including isolates from Hunan that had some of the changes found in the newly released sequence from Shanxi. Thus, although there are clear differences between isolates from northern and southern China, these is also recombination between these isolates to generate regions of identity.
The sequences clearly show increased genetic variation in the receptor binding domain and the HA cleavage site. These changes, or combinations of changes can have significant biological effects, which may be species specific. These small differences inpact vaccine development, diagnostic PCR tests which require primer matches, and determination of transportation and transmission of influenza genetic information, which originates in both high and low pathogenic avian influenza.
Both squabbling groups have very large amounts of unpublished sequence information for all eight H5N1 gene segments and the time to stop hoarding has past. H5N1 is rapidly evolving and becoming increasingly diverse, as is easily seen by simply comparing receptor binding domain or cleavage site changes.
However, these changes are in all eight gene segments, and the sequence data of the H5N1 isolates in the PNAS paper as well as the sequences from northern China, should be released immediately, along with other hoarded sequences such as those from H5N1 outbreaks in Europe and H5 isolates in North America, which also show recombination with H5N1 in China.
Media sources
Phylogenetic Trees