The second RNA segment of the influenza virus genome encodes the PB1 protein – part of the viral RNA polymerase – and, in some strains, a second protein called PB1-F2. The latter protein is believed to be an important determinant of influenza virus virulence. The absence of a full-length PB1-F2 protein has been suggested as one possible determinant for the low pathogenicity of the 2009 influenza H1N1 pandemic strain. Analysis of the evolutionary history of PB1-F2 suggests that it does not contribute significantly to viral fitness – the ability of the virus to replicate.
PB1-F2 binds to mitochondria, leading to a release of cytochrome c and induction of apoptosis in CD8 T-cells and alveolar macrophages. The protein increases the severity of primary viral and secondary bacterial infections in mice, and is associated with the high pathogenicity of avian H5N1 and the 1918 H1N1 pandemic virus.
The PB1-F2 protein is not produced in cells infected with the 2009 H1N1 strain because there are three stop codons at nucleotide positions 12, 58, and 88. The PB1 segment of the 2009 H1N1 strain is related to PB1 of H1N2 and H3N2 swine viruses from 1998 and human H3N2 viruses. Curiously, all the relatives of the 2009 H1N1 strain in swine and in humans encode a complete PB1-F2 protein. A truncated PB1-F2 is encoded by the genome of classical swine H1N1 viruses and human H1N1 viruses since 1947. But 96% of the avian influenza virus sequences deposited in NCBI as of 2007 encode the full length version of the protein.
Because the full-length PB1-F2 protein is not encoded in the genome of many influenza viruses, its evolutionary role and contribution to the fitness of the virus is unclear. To answer these questions, the evolution of PB1-F2 was compared with PB1 and two other open reading frames of similar size within the same RNA segment that are not translated into protein.
PB1-F2 is complete in all H1N1 human isolates before 1947, when a stop codon appeared which leads to production of a shorter version of the protein – 57 amino acids. If the complete protein conferred a functional advantage to the virus, a change in the evolutionary rates of the human H1N1 PB1-F2 proteins should have occurred in 1947. No such change is observed.
Results of sequence analysis reveal that the PB1-F2 open reading frame is as conserved, and maintained as a full-length protein, as other non-coding regions of the same RNA segment and of a randomly generated PB1 segment. These observations, and the fact that PB1-F2 is truncated in many virus isolates, suggest that the evolutionary role of PB1-F2 in animal hosts is minimal. Why the full length protein is produced by some viruses – and unfortunately leads to higher virulence – remains a puzzle.
Trifonov, V., Racaniello, V., & Rabadan, R. (2009). The Contribution of the PB1-F2 Protein to the Fitness of Influenza A Viruses and its Recent Evolution in the 2009 Influenza A (H1N1) Pandemic Virus PLoS Currents: Influenza