Mammalian influenza viruses are descendants of avian strains that crossed the species barrier and underwent further adaptation. Since 1997 in southeast Asia, H5N1 highly pathogenic avian influenza viruses have been causing severe, even fatal disease in humans. Although no lineages of this subtype have been established until now, such repeated events may initiate a new pandemic. As a model of species transmission, we used the highly pathogenic avian influenza virus SC35 (H7N7), which is low-pathogenic for mice, and its lethal mouse-adapted descendant SC35M . Specific mutations in SC35M polymerase considerably increase its activity in mammalian cells, correlating with high virulence in mice. Some of these mutations are prevalent in chicken and mammalian isolates, especially in the highly pathogenic H5N1 viruses from southeast Asia. These activity-enhancing mutations of the viral polymerase complex demonstrate convergent evolution in nature and, therefore, may be a prerequisite for adaptation to a new host paving the way for new pandemic viruses.

In 1997, an H5N1 influenza virus outbreak occurred in chickens in Hong Kong, and the virus was transmitted directly to humans. Because there is limited information about the avian influenza virus reservoir in that region, we genetically characterized virus strains isolated in Hong Kong during the 1997 outbreak. We sequenced the gene segments of a heterogeneous group of viruses of seven different serotypes (H3N8, H4N8, H6N1, H6N9, H11N1, H11N9, and H11N8) isolated from various bird species. The phylogenetic relationships divided these viruses into several subgroups. An H6N1 virus isolated from teal (A/teal/Hong Kong/W312/97 [H6N1]) showed very high (>98%) nucleotide homology to the human influenza virus A/Hong Kong/156/97 (H5N1) in the six internal genes. The N1 neuraminidase sequence showed 97% nucleotide homology to that of the human H5N1 virus, and the N1 protein of both viruses had the same 19-amino-acid deletion in the stalk region. The deduced hemagglutinin amino acid sequence of the H6N1 virus was most similar to that of A/shearwater/Australia/1/72 (H6N5). The H6N1 virus is the first known isolate with seven H5N1-like segments and may have been the donor of the neuraminidase and the internal genes of the H5N1 viruses. The high homology between the internal genes of H9N2, H6N1, and the H5N1 isolates indicates that these subtypes are able to exchange their internal genes and are therefore a potential source of new pathogenic influenza virus strains. Our analysis suggests that surveillance for influenza A viruses should be conducted for wild aquatic birds as well as for poultry, pigs, and humans and that H6 isolates should be further characterized.

High-pathogenic avian influenza viruses (HPAIVs) evolve from low-pathogenic precursors specifying the HA serotypes H5 or H7 by acquisition of a polybasic HA cleavage site. As the reason for this serotype restriction has remained unclear, we aimed to distinguish between compatibility of a polybasic cleavage site with H5/H7 HA only and unique predisposition of these two serotypes for insertion mutations. To this end, we introduced a polybasic cleavage site into the HA of several low-pathogenic avian strains with serotypes H1, H2, H3, H4, H6, H8, H10, H11, H14, or H15, and rescued HA reassortants after cotransfection with the genes from either a low-pathogenic H9N2 or high-pathogenic H5N1 strain. Oculonasal inoculation with those reassortants resulted in varying pathogenicity in chicken. Recombinants containing the engineered H2, H4, H8, or H14 in the HPAIV background were lethal and exhibited i.v. pathogenicity indices of 2.79, 2.37, 2.85, and 2.61, respectively, equivalent to naturally occurring H5 or H7 HPAIV. Moreover, the H2, H4, and H8 reassortants were transmitted to some contact chickens. The H2 reassortant gained two mutations in the M2 proton channel gate region, which is affected in some HPAIVs of various origins. Taken together, in the presence of a polybasic HA cleavage site, non-H5/H7 HA can support a highly pathogenic phenotype in the appropriate viral background, indicating requirement for further adaptation. Therefore, the restriction of natural HPAIV to serotypes H5 and H7 is likely a result of their unique predisposition for acquisition of a polybasic HA cleavage site.

As recently shown, mutations in the polymerase genes causing increased polymerase activity in mammalian cells are responsible for the adaptation of the highly pathogenic avian influenza virus SC35 (H7N7) to mice (G. Gabriel et al., Proc. Natl. Acad. Sci. USA 102:18590-18595, 2005). We have now compared mRNA, cRNA, and viral RNA levels of SC35 and its mouse-adapted variant SC35M in avian and mammalian cells. The increase in levels of transcription and replication of SC35M in mammalian cells was linked to a decrease in avian cells. Thus, the efficiency of the viral polymerase is a determinant of both host specificity and pathogenicity.

A promising approach to reduce the impact of influenza is the use of an attenuated, live virus as a vaccine. Using reverse genetics, we generated a mutant of strain A/WSN/33 with a modified cleavage site within its hemagglutinin, which depends on proteolytic activation by elastase. Unlike the wild-type, which requires trypsin, this mutant is strictly dependent on elastase. Both viruses grow equally well in cell culture. In contrast to the lethal wild-type virus, the mutant is entirely attenuated in mice. At a dose of 10⁵ plaque-forming units, it induced complete protection against lethal challenge. This approach allows the conversion of any epidemic strain into a genetically homologous attenuated virus.