Integrative approach to influenza virus surveillance characterization and control

Insights into virulence: viral factors contributing to the emergence of NAI-resistant or virulent influenza.


Neuraminidase is involved in the propagation and virulence of influenza viruses. This glycoprotein is a homotetramer of 4 identical polypeptide, each monomer contains about 470 AA arranged in four domains (N-term cytoplasmic sequence, a membrane anchoring hydrophobic TM domain, a thin stalk variable in length and a globular head that contains the active site) [Air GM et al, Proteins, 1989]. The nine NAs are divided in two groups: group 1 (N1, N4, N5 and N8) and group 2 (N2, N3, N6, N7 and N9). As surface glycoproteins, the NA is interacting with the HA. The HA is a receptor-binding protein, binding sialic acid while NA is a receptor-destroying enzyme, allowing nascent infectious particles to be released from the cell surface [Palese P et al, Virology, 1974].


An appropriate HA-NA balance is required for good viral fitness. Early X-ray structure of NA heads showed that the active site of the NA (the receptor-destroying pocket) is highly conserved in spatial as well as in sequence properties. Eleven conserved AA make contact with the substrate while another 6 conserved AA form a second shell that holds the active site residues in place [Tulip WR et al, J Mol Biol, 1991; Burmeister WP et al, EMBO J, 1992]. This rigid, conserved and stable enzyme active site has been an excellent target for drug design and development of neuraminidase inhibitors (NAIs). Since the discovery of NAIs our team has characterized substitutions in the NA responsible for resistance to NAI [Ferraris et al., 2008; Escuret et al., 2008; Casalegno et al., 2010]. NA bearing substitutions responsible for resistance to NAI result into viruses with low fitness because of impaired or abolished sialidase activity (disruption of the catalytic pocket, dissociation of the tetramer into inactive monomers, absence of NA at the surface of the virus) [Richard et al., 2011]. However, a fit seasonal H1N1 in 2007/2008 emerged with the H275Y substitution related to NAI resistance thanks to association with additional substitutions that compensated for the decreased fitness. To better understand the conditions that can permit the emergence of influenza viruses resistant to NAI we want to better analyze the structure and constraints of the neuraminidases bearing specific substitutions related to NAI resistance, and determine if NA from the same phylogenetic group display similar resistance patterns according to the substitution observed. For this purpose, we shall work on neuraminidases of avian representatives viruses, and introduce substitutions in positions known to confer oseltamivir-resistance in the N1 (H275Y) and N2 (E119V and R292K) of human viruses. Through a RG approach, we shall introduce specific substitutions in NA genes by directed mutagenesis, produce influenza viruses with the corresponding substituted NA, and check for NAI susceptibility and sialidase catalytic activity (Vmax, Km, Ki). For example, we plan to analyse the impact of the H275Y substitution (N1 numbering) in the 3 NAs of group1, to confirm if the change in NAI susceptibility and sialidase activity can be inferred to all Group 1 NA. Similarly, we shall study the E119V, E119G and R292K substitutions (N2 numbering) for group 2 NAs, with the same objectives. In addition, we shall perform a structural analysis of all the mutated NAs. While the structure of wild type NAs are known, those with H275Y, E119V E119G or R292K, substitutions have not been resolved yet in most NA subtypes. For this purpose, we have a close collaboration with the virology division at the NIMR in London that will perform the structural analysis of the glycoproteins. In addition, we will determine if resistance associated to the E119 substitutions (E119V and E119G) results from impaired sialidase activity or tetramer dissociation, as speculated by others. To carry out this analysis, we have all the tools required for the determination of the enzymatic properties of NA (Vmax, Km, Ki) and for the visualization and quantification of the NA spikes at the virion surface (cryo-EM analysis, MALDI-TOF mass spec) [Casalegno et al., 2010, Moules, V et al., 2011; Pourceau et al., 2011]. This understanding of the mechanisms of resistance can provide valuable information on the NA folding pathway, and the possible role of thee fifth beta sheet of the Na propeller structure in the stabilization of the homotetramer.



The emergence of viruses with a NAI-resistant phenotype may result from combination of low affinities of both HA and Na to α2,3 or α2,6 sialic acids. A balance between these receptor-binding and receptor-destroying activities is required for good virus fitness and transmission. We have already characterized the HA-NA balance of such NAI-R viruses [Richard et al., PLoSOne 2012], and we recently showed that HA binding properties can be wrongly associated to virulence when not analyzed in the context of NA-HA balance, glycoprotein surface densities, and genetic content. These works confirmed that looking solely at the HA or NA protein functions out of a viral context provides only limited, although valuable, information. Regarding this balance, we recently described that the addition of 222-119 mutations in the NA of H3N2 viruses can result in the lack of infectious particles or the production of viruses with defective Na [Richard et al., AAC 2011], but we (and others), also showed that the 222 or the 247 positions may play an important role in restoring fitness in NAI-R otherwise impaired NA mutants [Richard et al., PLoSOne 2012]. Our objective is to better analyze this HA-NA balance in genetically characterized viruses, with NA and HA displaying well defined affinities to the two receptors (α2,3 and α2,6 sialic acids).