Until 1997, avian influenza viruses, even the highly pathogenic strains (HPAI) that cause high mortality in chickens and turkeys, were thought to pose little direct threat to humans, needing first to mix genetic information (or recombine), thought most likely to occur in pigs, with influenza viruses already adapted to humans. This assessment changed when 18 people in direct contact with poultry in Hong Kong were directly infected with a virus, killing six. The virus carried the distinctive surface markers H5 and N1. The H5N1 virus has since spread widely throughout SE Asia, reaching Europe and Africa, and continues to infect humans infrequently (about 200 cases with a continued high mortality rate of 51%) as well as zoo animals and cats. It can clearly spread widely by infection of migratory birds and continues to evolve and diversify, threatening to develop into a human pandemic strain, either directly by mutation or by recombination. Measures are being taken to improve protection against emergence of a possible human pandemic virus (which may or may not retain the H5 and N1 markers) but it is recognised that control of the virus in poultry and birds is important to reducing the risk of the emergence of such a pandemic strain. Standard practise in the UK and EU in the event of focal outbreaks of HPAI is containment and slaughter. However, in the face of widespread HPAI (H5N2) in Mexico in the mid-1990's, vaccination was introduced and was partially successful, in so far as it eradicated HPAI, but it failed to eradicate low pathogenicity strains (LPAI). This campaign was a landmark in that as well as using conventional vaccines based on killed influenza viruses of the same or 'homologous' type as the epidemic strain, a genetically modified Fowlpox virus coding for the H5 part of the influenza virus was used for the first time as a so called 'recombinant vaccine'. In the face of the current H5N1 epidemic, several countries have introduced vaccination programmes. In China, three types of vaccine have been used: (i) conventional, mismatched or 'heterologous' (H5N2) killed influenza vaccine, (ii) a killed, LPAI derivative of H5N1, engineered into a safer backbone by genetic modification, and (iii) a recombinant Fowlpox virus expressing both H5 and N1. In France, as the recombinant Fowlpox virus and the genetically modified H5N1 have not been licensed, a conventional, heterologous, killed vaccine (the H5N2 used as a homologous vaccine in Mexico) has been used. With the speed of evolution of the H5N1 strain, the use of older versions of H5 in vaccines, especially that from the H5N2 strain used in Mexico, causes some concern. Current vaccines protect vaccinated birds but do not always prevent them become infected and a quarter of those infected birds become carriers. As vaccines becomes less effective due to evolution of the virus, there are fears that the vaccine's inability to stop bird-to-bird transmission will contribute to a vicious circle of increased virus variation. The conventional and recombinant Fowlpox virus vaccines induce good, protective antibody responses but are not so good at inducing the 'cellular responses' necessary to allow the body to clear itself of infected cells. Work in mice has shown that better cellular responses would be expected not against H5 and N1 but from the internal and non-structural proteins of influenza virus so we will make recombinant Fowlpox viruses based on these proteins as well as on H5 and N1. We may also need to adjust the balance of antibody versus cellular responses. This could be done by incorporating chicken immune regulators into the recombinant Fowlpox virus vaccines, and we will try this in the experimental situation. However we think it would be preferable to achieve this adjustment by changing some of the Fowlpox virus's own immune regulators; we already have demonstrated this approach can work against a different virus disease of poultry.
Recombinant Fowlpox viruses (rFWPV) are playing an important role in China in attempts to control high pathogenicity avian influenza H5N1 virus in poultry. rFWPV expressing H5 were first employed in Mexico during the campaign to control high pathogenicity H5N2, with almost a billion doses used. New recombinants, derived in China to express H5 and N1 homologous for the current panzootic, are being used in a massive vaccination campaign (some 5 billion poultry). Although rFWPV are reported to be effective in controlling disease, it is not clear how effective they are at inducing cellular immunity and controlling spread. It is likely that spread will be most effectively controlled by induction of cellular as well as improved humoral/mucosal immunity. We intend to investigate methods of (a) improving induction of humoral/mucosal immunity and (b) driving cell-mediated responses after vaccination with rFWPV. Cell-mediated immunity will probably require expression of internal and non-structural proteins (nucleoprotein will not be included as its absence is important in differential diagnosis). Coexpression of host-derived cytokines by rFWPV has been shown to affect the humoral/cellular balance of the immune response, improving vaccine efficacy in certain cases. We therefore aim to generate a spectrum of responses for each arm of the immune system by coexpression of a variety of host cytokines (and one viral cytokine) so that we can optimise the desired response. Significant improvement in efficacy may alternatively be achieved by deletion of FWPV-encoded immunomodulators that bind host cytokines, without the safety concerns attracted by expression of host cytokines. We will therefore investigate the effect of deleting two such viral immunomodulators from the rFWPV vectors. The nature of the humoral and cellular immune responses induced by the various modified recombinants will be assayed in vivo, both before and after challenge with highly pathogenic H5N1.