Emergence of Uva Lake H5N1 in Europe
Recombinomics Commentary 19:29
January 15, 2008
The strain of the virus identified is very similar to that causing the cases confirmed in the Czech Republic, Romania and Poland mid to late last year. It is less similar to that found in East Anglia in November.
The above comments by DEFRA on H5N1 sequences from the mute swan infections in Dorset (see satellite map), indicate all H5N1 outbreaks in Europe since the summer of 2007 have been the Uva Lake strain (a sub-clade of clade 2.2.3). Sequences have been made public of outbreaks in late 2006 in South Korea, as well as 2007 outbreaks in Germany, Krasnodar (Russia), and Romania. Comments by FLI indicate the recent outbreaks in northeastern Germany are Uva Lake, and now comments by DEFRA indicate the outbreaks in Kuwait, Czech Republic, France, England, and Poland were also the Uva Lake sub-clade.
Therefore it is worth reviewing the emergence of the Uva Lake strain in Europe. In late 2005 and early 2006, the Qinghai strain of H5N1 was reported in approximately 50 countries west of China. The Qinghai strain has been designated Clade 2.2, and the strain was sub-divided into three additional sub-clades based on sequence data generated through the NIAID influenza sequencing project on samples provided by Ilaria Capua from Italy as described in Genome Analysis Linking Recent European and African Influenza (H5N1) Viruses. Clade 2.2.1 included isolates from Egypt, southern Germany, Italy, Mongolia, and some regions in sub-Sahara Africa. Clade 2.2.2 included isolates from northern Germany, Denmark, Sweden, Scotland, and Nigeria. Clade 2.2.3 included isolates from India, Afghanistan, Italy, and Iran.
The Uva Lake strain was isolated in wild birds in the summer of 2006 at Uva Lake, the largest lake in Mongolia. Although the outbreak was massive, killing thousands of wild birds, neither Mongolia nor Russia filed OIE reports. However, H5N1 was isolated in both countries and sequences from grebes and a duck in Tyva were published (see satellite map), as were sequences from a whooper swan and common golden eye from Mongolia. Although the sequences were clearly Clade 2.2.3, they were distinct from clade 2.2.3 sequences isolated elsewhere.
In the fall of 2006 South Korea reported an H5N1 outbreak, which was followed by an early 2007 outbreak in Japan. The sequences from the South Korean outbreak have now been published and they are closely related to the Uva Lake sequences. However, there were no reports of Uva Lake sequences to the west of Uva Lake in the latter half of 2006.
In early 2007 there was an H5N1 outbreak in Kuwait. Although it has been almost a year since that outbreak was reported, no sequences have been released. However, comments from DEFRA and Weybridge indicated the Kuwait sequences were 99.5% identical to sequences from a subsequent outbreak in the Czech Republic in the summer of 2007. The Czech outbreak was quickly followed by multiple wild bird outbreaks in Germany. FLI reports on those outbreaks indicated each outbreak (in Bavaria, Saxony, and Thuringen) was distinct, but most closely related to the sequences from Uva Lake. Recently these sequences were made public, confirming the close relationship. Outbreaks in Germany were followed by outbreaks in France, which was followed by an outbreak in Krasnodar in September.
Full sequences from all eight gene segments were quickly published from Krasnodar, and those sequences (from a chicken and whooper swan) were also closely related to Uva Lake. That outbreak was followed by the outbreak in free range turkeys in Suffolk, England. DEFRA comments indicated the H5N1 from that outbreak was related to the Czech sequences. Outbreaks were the reported at multiple locations in Poland, and as indicted by DEFRA, these were also related to Uva Lake.
Romania also reported an outbreak and quickly released HA sequences from three species (chicken, duck, and cat) and all three sequences were identical and closely related to Uva Lake, but distinct from the published sequences from Russia and Germany. FLI subsequently indicated the outbreak in northeastern Germany was also related to the earlier outbreaks in Germany. Comments haven’t been made on the sequences from Rostov, but the proximity to Krasnodar suggests the Rostov sequences are also Uva Lake related, which is also likely for sequences from Saudi Arabia and Pakistan.
Thus, it appears that the H5N1 emerging out of Qinghai Lake in China, Cheny Lake in Russia, and Erhel Lake in Mongolia in the summer of 2005 was repeated at Uva Lake in Mongolia in the summer of 2006. Like the earlier outbreak, detection of H5N1 was uncommon in the fall but was widely reported in the following year. Most of the reported outbreaks from the Qinghai strain were in early 2006. In early 2007 there were few reports of the Uva Lake strain, but reports became increasingly common beginning in the summer of 2007.
Although it is unclear what is circulating in Africa, the widespread outbreaks of the Uva Lake strain in the summer and fall of 2007 in Europe and the Middle East suggests reports of H5N1 in early 2008 will be common.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings