Introduction of Virulence Markers in PB2 of Pandemic Swine-Origin Influenza Virus Does Not Result in Enhanced Virulence or Transmission

In the first 6 months of the H1N1 swine-origin influenza virus (S-OIV) pandemic, the vast majority of infections were relatively mild. It has been postulated that mutations in the viral genome could result in more virulent viruses, leading to a more severe pandemic. Mutations E627K and D701N in the PB2 protein have previously been identified as determinants of avian and pandemic influenza virus virulence in mammals. These mutations were absent in S-OIVs detected early in the 2009 pandemic. Here, using reverse genetics, mutations E627K, D701N, and E677G were introduced into the prototype S-OIV A/Netherlands/602/2009, and their effects on virus replication, virulence, and transmission were investigated. Mutations E627K and D701N caused increased reporter gene expression driven by the S-OIV polymerase complex. None of the three mutations affected virus replication in vitro. The mutations had no major impact on virus replication in the respiratory tracts of mice and ferrets or on pathogenesis. All three mutant viruses were transmitted via aerosols or respiratory droplets in ferrets. Thus, the impact of key known virulence markers in PB2 in the context of current S-OIVs was surprisingly small. This study does not exclude the possibility of emergence of S-OIVs with other virulence-associated mutations in the future. We conclude that surveillance studies aimed at detecting S-OIVs with increased virulence or transmission should not rely solely on virulence markers identified in the past but should include detailed characterization of virus phenotypes, guided by genetic signatures of viruses detected in severe cases of disease in humans.The new H1N1 swine-origin influenza virus (S-OIV) recently emerged to cause the first influenza pandemic in 40 years (2). The S-OIV presumably emerged from pigs, as its genome was shown to consist of six gene segments of “triple-reassortant” swine viruses and two of “Eurasian lineage” swine viruses (9). The start of the S-OIV pandemic has been relatively mild, with a clinical spectrum ranging from mild upper respiratory tract illness to sporadic cases of severe pneumonia leading to acute respiratory distress syndrome (22). As of 15 November 2009, worldwide, more than 206 countries have reported laboratory-confirmed cases of S-OIV infection, including over 6,770 deaths (32).In previous influenza pandemics, such as the Spanish influenza pandemic of 1918 and the Hong Kong influenza pandemic of 1968, a first wave of cases of relatively mild illnesses was followed by more severe subsequent waves (29). The reason for this increased severity has remained largely unknown, but one possible explanation could be that the pandemic viruses required further adaptation to the human host, resulting in the emergence of viruses that were more virulent than those of the first wave. Such adaptive changes could occur by gene reassortment between cocirculating influenza A viruses or by mutation.In the past decade, determinants of influenza A virus virulence have been mapped using reverse genetics with a variety of pandemic, epidemic, and zoonotic influenza viruses. Mutations affecting virulence and host range have frequently been mapped to hemagglutinin (HA) and neuraminidase (NA) in relation to their interaction with sialic acids, the virus receptors on host cells (11, 18, 30). Nonstructural protein 1 (NS1) has been implicated in the virulence of highly pathogenic avian influenza (HPAI) virus H5N1 and the 1918 H1N1 virus, as the NS1 proteins of these viruses were shown to act as strong antagonists of the interferon pathways (10, 25). Furthermore, the polymerase genes, in particular the PB2 gene, have been shown to be important determinants of virulence in the HPAI H5N1 and H7N7 viruses and of transmission in the 1918 H1N1 virus (11, 21, 31). One of the most commonly identified virulence markers to date is E627K in PB2. The glutamic acid (E) residue is found generally in avian influenza viruses, while human viruses have a lysine (K), and this mutation has been described as a determinant of the host range in vitro (28). When avian viruses lacking the E627K substitution were passaged in mice, the viruses acquired the mutation spontaneously upon a single passage (15, 17). In the HPAI H5N1 and H7N7 viruses, E627K was shown to be the prime determinant of pathogenesis in mice (11, 21, 23). Given that all human and many zoonotic influenza viruses of the last century contained 627K (1), it was surprising that the S-OIV had 627E.Additionally, the aspartate (D)-to-asparagine (N) mutation at position 701 of PB2, which was shown to compensate for the absence of E627K, has also not been detected in S-OIV (27). This D701N mutation has previously been shown to expand the host range of avian H5N1 to mice and humans (3, 15) and to increase virus transmission in guinea pigs (27). Thus, S-OIV was the first known human pandemic virus with 627E and 701D, and it has been speculated that S-OIV could mutate into a more virulent form by acquiring one of these mutations, or both.On 8 May 2009, the detection of another mutation in the PB2 gene of S-OIV, an E-to-glycine (G) mutation at position 667, was reported (http://www.promedmail.org/pls/apex/f?p=2400:1000, archive no. 20090508.1722). It has previously been suggested that the E667G substitution in PB2 of HPAI H5N1 virus was under positive selection and possibly played a role in sustainable transmission in humans (14).On 28 September 2009, detection of the E627K mutation in PB2 of S-OIVs of two individuals in the Netherlands was reported (http://www.promedmail.org/pls/apex/f?p=2400:1000, archive no. 20090928.3394) and raised concern about the possible enhanced replication of the S-OIV in humans, possibly associated with increased virulence. To date, the D701N mutation in PB2 has not been reported in any of the S-OIVs sequenced, and additional viruses with mutation E627K have not been recorded, either. In contrast, viruses with E677G have been reported from the United States, Canada, Germany, the United Kingdom, Norway, and France, according to the public sequence databases.Here, the effects of the E627K, D701N, and E677G mutations in the PB2 genes of S-OIVs was investigated using genetically engineered influenza viruses based on a prototype S-OIV, A/Netherlands/602/2009. Polymerase activity was measured in minigenome assays in human 293T cells, virus replication was analyzed in Madin-Darby Canine kidney (MDCK) cells, virulence was tested in mouse and ferret models, and transmission by aerosols or respiratory droplets was tested in ferrets. In contrast to the earlier assumptions based on experience with other influenza A viruses, S-OIVs with E627K, D701N, or E677G in PB2 did not show a marked increase in virulence or transmission compared to the wild-type virus.