Outbreak of highly pathogenic avian influenza caused by Asian lineage h5n1 virus in turkeys in Great Britain in January 2007

UNTIL recently, it appeared that the epidemiology of avian influenza consisted of the perpetuation of low pathogenicity avian influenza (LPAI) viruses of all 16 haemagglutinin (H) subtypes in wild birds, where they caused little or no disease and spread on occasion to poultry. Very occasionally, introductions of LPAI viruses of the H5 or H7 subtype into poultry resulted in the mutation of these viruses to virulent viruses that caused highly pathogenic avian influenza (HPAI). This is how the four HPAI outbreaks in poultry in Great Britain between 1959 and 2006 were thought to have occurred. These HPAI outbreaks were one in chickens in Scotland in 1959, caused by virus of H5N1 subtype, and three in turkeys in Norfolk, one in 1963 caused by virus of H7N3 subtype, one in 1979 caused by H7N7 virus, and the most recent in 1991 caused by a virus of H5N1 subtype (Alexander 1992, Alexander and others 1993). Since the mid-1990s, the epidemiology of HPAI has become more complicated. In 1996, the progenitor virus of the Asian lineage H5N1 HPAI virus was reported in geese in Guandong Province, China (Xu and others 1999). This virus and its descendents proceeded to spread to poultry in most countries in south-east Asia, becoming endemic in several of them. Following reports of the spread of the Asian lineage H5N1 HPAI virus to wild birds, the virus was reported in wild birds and poultry throughout Asia, Europe and into Africa (Alexander 2007), probably as a result of movements of infected poultry or poultry products and infected wild birds. There had been two isolated incursions of the Asian lineage H5N1 HPAI virus into Great Britain before 2007. The first occurred in October 2005 in a quarantine facility in England. Statutory postimportation investigations of deaths in captive caged birds, supposedly from Taiwan, showed them to be the result of Asian lineage H5N1 HPAI virus infection (DEFRA 2005). This virus was genetically closest to contemporaneous viruses isolated in China. The second case, in April 2006, involved a dead whooper swan (Cygnus cygnus) found in the sea at Cellardyke, Fife, which proved to be infected (Blissit 2007). This H5N1 virus was genetically closest to viruses isolated from swans and other wild birds in Germany, including those infecting birds on islands in the Baltic Sea. This short communication describes the outbreak of HPAI caused by an Asian lineage H5N1 virus in turkeys in Holton, Suffolk, that began at the end of January 2007, and the characterisation of the causative virus. The outbreak occurred on a large commercial meat turkey site consisting of 24 houses, two of which were empty, stocked with 159,000 birds. There was also a slaughter and meat processing plant adjacent to the site. Suspicion of notifiable disease was reported to DEFRA following a very sudden deterioration in health, with rapidly increasing mortality, of sevento eight-week-old turkeys in only one house (house 10), with an original placement of 7119 day-old poults. Mortality and cull figures for the five days beginning January 29 were eight, 71, 186, 860 and 1580, that is, in this period 38 per cent of the birds originally placed had died or had been culled. The clinical picture was of exceptionally high morbidity, with over 90 per cent of birds sitting down, some with the legs out to one side, some birds showing a fine head tremor, and others were moribund. The flock was very quiet. There was virtually no feeding or drinking, and the birds were uninterested in people. At postmortem examination of typically affected turkeys, there were no specific changes suggestive of avian influenza virus infection. Carcases were generally dehydrated, although there was no macroscopic evidence of striking kidney pathology. Gizzards were small and flabby, and contained more litter than feed; crops were empty. There was no upper respiratory tract involvement, and, grossly, the lungs appeared unremarkable. A few birds showed a mild airsacculitis, mainly of the lower abdominal air sacs. Livers appeared congested. Spleens were enlarged, pale and mottled. There were no haemorrhages in the internal organs or any mucosal surfaces of the respiratory or alimentary tracts. Suspicion of notifiable disease was reached only due to the unexplained high and escalating mortality, and exceptionally high morbidity. At the time of reporting to DEFRA, and up to the time of culling of all the birds on the site, there was no evidence of clinical disease in any other house on the premises. Ten carcases, 17 cloacal swabs and 20 blood samples were submitted from the affected turkey house to the Veterinary Laboratories Agency (VLA) – Weybridge, and examined using the diagnostic methods prescribed in Directive 2005/94/EC and the EU diagnostic manual for avian influenza (CEC 2006a, b). The swabs, tissues and samples from the carcases were tested for the presence of infectious avian influenza virus by inoculation of nineto 10-day-old embryonated specific pathogenfree fowls’ eggs. The samples were also tested by real-time reverse transcriptase-PCR (rRT-PCR) to detect the presence of the influenza A matrix gene (Spackman and others 2002), and any positives were further tested by rRT-PCR to detect H5 or H7 genes (Slomka and others 2007). The rRT-PCR results were positive for H5 and, subsequently, virus was isolated. This was shown to have the HA0 cleavage site amino acid motif PQGERRRKKR*GLF, which is typical of the Asian lineage H5N1 HPAI viruses seen in Europe in 2006. The isolated virus was identified as H5N1 subtype using conventional techniques (Alexander and Spackman 1981) and subjected to an intravenous pathogenicity index test (CEC 2006a, b), in which a maximum value of 3·00 was recorded. Following the decision to cull all the turkeys on the site, 20 oropharyngeal swabs, 20 cloacal swabs and 20 blood samples were submitted to VLA – Weybridge from turkeys in each of the other 21 populated houses. Despite the lack of clinical signs, HPAI H5N1 virus was detected in samples in three further houses by rRT-PCR, and H5N1 HPAI virus isolates were obtained from birds in these three houses. None of the blood samples from the 21 houses was positive for H5 antibodies. Phylogenetic analysis of the nucleotide sequences of the HA1 gene of the viruses isolated showed them to be of the Asian H5N1 HPAI lineage and the Lake Qinghai sublineage, that is, closely related to the viruses isolated from poultry and wild birds in Europe in 2006. The nucleotide sequence of the HA1 gene of a HPAI H5N1 virus isolated from farmed domestic geese in Hungary in January 2007 was identical to the viruses isolated from the Suffolk outbreak. Sequences of the whole genome (13,558 nucleotides) of representative viruses A/turkey/England/250/07 (virus isolated from house 10), A/goose/Hungary/2823/07 (derived from the first reported Hungarian outbreak in January 2007) and Veterinary Record (2007) 161, 100-101