Characterization of influenza A viruses with different amino acid residues at PB2-627

Abstract

Background: Viral ribonucleoprotein (vRNP) of influenza A virus is responsible for transcription and replication of the viral genome. Residue 627 of one of its subunit, PB2, has been shown to have significant effect on host specificity, polymerase activity and viral replication. The majority of avian influenza viruses carry a glutamic acid (E) at PB2-627, while most of the human influenza viruses identified before emergence of 2009 H1N1 pandemic carry a lysine (K). The amino acid change E627K in PB2 is believed to favor avian virus adapting in mammalian hosts. Although glutamic acid and lysine are commonly found at PB2-627 in nature, some other amino acids are also possible at this residue. For example, some avian viruses possess valine and some human viruses possess arginine at this residue. However, the effects of these less frequent amino acids at PB2-627 are not fully understood. Furthermore, the possibility of emergence of virus with other amino acids at PB2-627 cannot be excluded. This prompted us to generate and characterize the recombinant viruses with a random amino acid mutation at this position. Materials and Methods: Degenerated primers were designed to generate random amino acid mutations at residue 627 of the PB2 gene in A/PR/8/1934 (PR8). Recombinant PR8 viruses with random mutations at this position were generated by reverse genetics techniques in 293T cells. The recombinant viruses generated were inoculated in the mammalian system (MDCK cells) or the avian system (10-day-old embryonated chicken eggs). The viruses harvested were plaque purified and the amino acids at PB2-627 were identified by DNA sequencing. The polymerase activities of vRNPs with different amino acids at PB2-627 were compared and the growth kinetics of the recombinant viruses were also determined in mammalian and avian cells. The virulence of the recombinant viruses were compared in balb/c mice. Results: Sequencing result of the recombinant viruses revealed that amino acids at PB2-627 are not restricted to those isolated from nature. It was observed that the recombinant viruses derived from the avian system behaved differently from those derived from the mammalian system in polymerase activity assay and in vitro growth kinetics assay. The recombinant viruses derived in the avian system generally had lower polymerase activity and slower growth rate than the wild-type virus in mammalian cells. On the other hand, some recombinant viruses derived in the mammalian system had low polymerase activity and slow growth in avian cells. Although all the recombinant viruses with different amino acids at PB2-627 replicated in mammalian cells, the virulence of the recombinant viruses in mice vary. It was found that recombinant viruses with isoleucine at PB2-627 led to more significant weight loss and higher mortality in mice than the recombinant viruses with the avian marker at PB2- 627, albeit lower than wild-type virus. Conclusions: This study demonstrated that the amino acid identity at PB2-627 is not exclusive to glutamic acid and lysine. Introducing other unnatural amino acid mutations at this position can also lead to viable viruses in mammalian and avian cells. Some of these novel amino acid mutations may also lead to highly virulent virus in mammals. Although the mechanism leading to the role of PB2-627 is yet to be elucidated, more understanding on the effect of this residue on the viral polymerase can surely contribute to the evaluation of the pandemic potential of any novel virus that may emerge in the future.