H5N1 Tracing To Suffolk England
Recombinomics Commentary
November 15, 2007
Scientists are working on the theory that the virus is related to an outbreak of bird flu in domestic poultry in Germany and the Czech Republic earlier this year, suggesting it may have been spread by migrating birds – a claim hotly denied by the RSPB, which said there was no evidence of the disease in the wild populations currently in the UK.
The above comments reflect many media reports on the genetic composition of the H5N1 in Suffolk, as well as denials by wildlife conservation groups on the presence of H5N1 in wild birds in the UK. Therefore, a review of the genetics and surveillance of the H5N1 is useful.
The H5N1 is the Qinghai (clade 2.2) strain that was first reported at Qinghai Lake in May, 2005. Although this strain was the broadly defined “Asian” strain that was first reported in 1996 in a goose in Guangdong Province, it had a number of markers in each of its eight gene segments which distinguished it from earlier isolates. These changes included the novel HA cleavage site of GERRRKKR (instead of the more common RERRRKKR), and a change in the PB2 gene (E627K), which had never been previously found H5N1 in from birds. The E627K change was of concern, because that change was in all human influenza A isolates, dating back to the 1918 pandemic. The change was associated with optimal replication at 34C in contrast to the avian form, which generated optimal replication at 41C, the body temperature of birds. The change was also associated with increased virulence in mammals, which have a body temperature of 37C, with temperatures lower in the nose and throat, especially during the winter when human influenza infections peak.
Initially, there was thought that the lethality of the H5N1 at Qinghai Lake might lead top the virus burning itself out. However, in the summer of 2005 H5N1 was reported at Chany Lake in Novosibirsk, Siberia where H5N1 had not been reported previously. Sequence analysis showed that the H5N1 was the Qinghai strain, and detection of the strain in a health great crested grebe demonstrated that it could be carried in asymptomatic wild birds. In addition to the wild birds, H5N1 was found at farms surrounding the lake, as well as farms across the border in Kazakhstan.
Shortly thereafter H5N1 was also reported in wild birds at Erhel Lake on Mongolia. In both instances H5N1 had not been reported previously and the sequences from the isolates showed that they were the Qinghai strain. In Mongolia, wildlife conservationists helped collect samples, and although samples from dead birds were positive, samples from live birds were not. The detection failures in the live birds were used by the conservation groups to predict that the H5N1 would not spread.
However, the H5N1 had already spread from Qinghai Lake in central China to Chany Lake in Russia and Erhel Lake in Mongolia. These locations, like Qinghai Lake, were located at the intersection of major wild bird flyways. Moreover, testing of hunter killed birds in Siberia identified over 25 wild bird species that were infected with Qinghai H5N1. Thus, spread to Europe, the Middle East, and Africa was anticipated, as birds flew south from summer locations in Siberia and Mongolia.
These concerns were realized in the fall of 2005 when H5N1 was reported in the Volga Delta and Danube Delta. Moreover, H5 was detected in a healthy teal in the Nile Delta in December, 2005. Although the virus wasn’t isolated, repeated extractions of the sample led to HA and NA sequences, which were Qinghai and most closely related to H5N1 that was subsequently isolated in Austria. Although detection of H5N1 west of China in 2005 was limited, all countries with H5N1 were detecting the “Asian” stain for the first time, and all were then Qinghai strain.
After fatal infections in Turkey were confirmed in early 2006, additional human cases were reported in Iraq, Azerbaijan, Egypt, and Djibouti. All infections were Qinghai H5N1 as were bird outbreaks throughout Europe, the Middle East, and Africa. In many countries all positives were in wild birds, but only dead or dying wild birds were positive. The failure to detect H5N1 in live birds raised serious surveillance issues.
Many of the positives were sequenced and the isolates had a number of regional markers that allowed the Qinghai strain to be grouped into sub-clades. In the spring/summer of 2006 H5N1 was again found in Qinghai province, but there was also a massive outbreak at the Siberia/Mongolia border near Tyva. This outbreak rivaled the Qinghai outbreak and H5N1 was isolated from grebes, swans, and golden eyes.
In the fall of 2006 and early 2007 the reporting of H5N1 in Europe, the Middle East, and Africa was less frequent as were the number of human cases, which were limited to Egypt and Nigeria. In Egypt the more recent isolates had the Egyptian regional markers from the preceding season, but the sequences were more complex. In Europe H5N1 was found in Hungary followed by England. The sequences between the outbreaks in Hungary and England were 99.96% identical. Although Qinghai isolates usually have identities higher than 99%, the 99.96% raised the possibility that the outbreak in England was linked to the outbreak in Hungary. This possibility was supported by the linkage of the same turkey producer to both locations and the shipment of product from Hungary to England.
However, this was the exception that proves the rule because other outbreaks were similar (99% identity) but the similarity did not reach 99.94% because wild birds are infected with various Qinghai strains with specific regional markers.
In the summer of 2007 outbreaks were reported in the Czech Republic and Germany. The initial outbreak in the Czech Republic was in domestic poultry, but a virtually identical sequence was found in a wild bird, suggesting the poultry outbreak was linked to wild birds.
In Germany, the initial positives were in wild birds and were widespread. The first positives were in Nuremberg, but that was followed by outbreaks in Saxony, Thuringia, and Saxony Anhalt. Although all of these sequences were most closely related to the 2006 outbreak in Tyva/Mongolia, each location had a slightly different sequence, as expected from wild bird isolates. Eventually H5N1 was also detected in a farm in Bavaria and further testing of farms that were commercially linked found H5N1 antibodies in asymptomatic ducks.
Thus, the sequence data supported multiple introductions by wild birds with H5N1 that was closely related to the earlier Tyva/Mongolia outbreak, but there were consistent differences between isolates from different locations in Germany.
The outbreaks in the Czech Republic, Germany, and France were followed by an outbreak in domestic poultry in Krasnodar. Full sequences from that outbreak have been made public, and those sequences are also most closely related to the public sequences from Tyva/Mongolia, and based on descriptions from sequencing agencies in Germany and England, would also be closely related to sequences in the Czech Republic, Germany, and England.
Therefore, introductions by wild birds in England would generate sequences that were similar to the Krasnodar sequence, but the identity would not be in the greater than 99.9% range.
Release of the sequences in Suffolk would be useful.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings