Investigating the role of segment 3 in H9N2 avian influenza virus pathogenicity

Abstract

H9N2 avian influenza viruses (AIV) are widespread in poultry populations worldwide, causing large economic losses. Reassortment events with other co-circulating AIV strains has led to an increased pathogenicity of H9N2 in poultry. However, the molecular basis of this increased pathogenicity remains largely undetermined, although previous experiments have implicated exchange of segment 3; encoding the PA subunit of the viral RNA polymerase and the PA-X virulence factor. It is important to understand the factors responsible for enhancing pathogenicity in order to improve control measures and identify risk factors. This study uses H9N2 AIV as a model to understand how evolutionary molecular changes in segment 3 (PA gene) modulate virus virulence in birds. To investigate the molecular basis of this, site directed mutagenesis was used to introduce reciprocal alterations to the PA genes of two pre- and post-reassortant H9N2 AIV strains: A/guineafowl/Hong Kong/WF10/99 (WF10) and A/chicken/Pakistan/UDL-01/08 (UDL-01) respectively. A single polymorphism adjacent to the PA endonuclease active site (K26E) was identified as having the largest impact on viral phenotype. This change did not significantly affect the transcriptional activity of the viral polymerase. However, when the mutation was introduced into viruses, the replication phenotypes (assessed via plaque size and viral titre) were switched between the WF10 and UDL-01 strains. During in vivo pathogenicity studies within the UDL-01 virus, the introduction of the E26K change altered viral replication as well as reducing overall pathogenicity in directly inoculated and contact chickens but without affecting transmission. In contact birds, the mutant virus was less able to disseminate beyond the respiratory tract to the visceral organs. Thus overall, the E26K change altered viral replication in vitro and in vivo, identifying it as a pathogenicity-determining residue. The E26K mutation lies within the protein domain common to PA and PAX. PA-X has roles in host cell protein synthesis shut off and when the shut off activity of the WF10 and UDL-01 PA-Xs were assessed, there was a marked difference in their activity. WF10 PA-X was unable to control host cell protein synthesis whereas UDL-01 PA-X had highly active host cell shut off ability. This activity could be switched with the introduction of the E26K PA mutation. Further functions of the H9N2 PA-Xs were then investigated. Polymerase transcriptional activity was also similar between the virus strains. However, removal of PA-X expression from UDL-01 led to a reduced plaque phenotype and viral replication. Removal of PA-X from WF10 had no viral fitness implications. When both the E26K change was made and PA-X expression was removed from UDL-01 segment 3, viral replication and plaque diameter was reduced, the reciprocal effect was observed with the introduction of both mutations into WF10. During in vivo pathogenicity studies with the UDL-01 strain with or without PA-X expression, loss of PA-X reduced viral shedding at earlier times post infection. Overall, these studies show that a single amino acid change within the PA gene of H9N2 avian influenza viruses is able to reduce replication and pathogenicity of these viruses in poultry via impacting upon ability of the virus to control host cell protein synthesis.