Evolutionary dynamics of the N1 neuraminidases of the main lineages of influenza A viruses Mathieu Fourment a, * , Jeffrey T. Wood b , Adrian J. Gibbs c , Mark J. Gibbs d,1 a Department of Biological Sciences, Macquarie University, Sydney, NSW 2109, Australia b Statistical Consulting Unit, Graduate School, The Australian National University, Canberra, ACT 0200, Australia c Yarralumla, Canberra, ACT 2605, Australia d School of Botany and Zoology, The Australian National University, Canberra, ACT 0200, Australia article info Article history: Received 29 April 2009 Revised 22 April 2010 Accepted 23 April 2010 Available online 29 April 2010 Keywords: Phylogeny Influenza Neuraminidase Substitution rate Selection Tree imbalance Human Swine Avian abstract Influenza A virus infects a wide range of hosts including birds, humans, pigs, horses, and other mammals. Because hosts differ in immune system structure and demography, it is therefore expected that host pop- ulations leave different imprints on the viral genome. In this study, we investigated the evolutionary tra- jectory of the main lineages of N1 type neuraminidase (NA) gene sequences of influenza A viruses by estimating their evolutionary rates and the selection pressures exerted upon them. We also estimated the time of emergence of these lineages. The Eurasian (avian-like) and North American (classical) swine lineages, the human (seasonal) and avian H5N1 lineages, and a long persisting avian lineage were studied and compared. Nucleotide substitution rates ranged from 1.9 Â 10 À3 to 4.3 Â 10 À3 substitutions per site per year, with the H5N1 lineage estimated to have the greatest rate. The evolutionary rates of the H1N1 human lineage appeared to be slightly greater after it re-emerged in 1977 than before it disappeared in the 1950s. Comparing across the lineages, substitution rates appeared to correlate with the number of positively selected sites and with the degree of asymmetry of the phylogenetic trees. Some lineages had strongly asymmetric trees, implying repeated genotype replacement and narrow genetic diversity. Positively selected sites were identified in all lineages, with the H5N1 lineage having the largest number. A great number of isolates of the H5N1 lineage were sequenced in a short time period and the phylogeny of the lineage was more symmetric. We speculate that the rate and selection estimations made for this lineage could have been influenced by sampling and may not represent the long-term trends. Ó 2010 Elsevier Inc. All rights reserved. 1. Introduction Influenza A viruses cause serious diseases in a wide range of ver- tebrates. Their genomes consist of eight minus-sense RNA segments each of which encodes either one or two proteins. The haemaggluti- nin and neuraminidase proteins (HA and NA) protrude from the out- er surface of influenza virions, and are key determinants of the molecular epidemiology of the virus as they interact most actively with the immune systems of the vertebrate hosts. HAs bind the viri- ons to terminal sialic acid residues on the surface glycoproteins of the host cell. NAs catalyse the release of virions by hydrolysing those sialic acid residues (Palese and Compans, 1976). Antibodies against both proteins, especially the HA, provide immunity to infection (Webster et al., 1988), and NAs are the targets of the neuraminidase inhibitor antiviral called zanamivir and oseltamivir. No species are formally recognised in the Influenza A virus genus, instead isolates are named (subtyped) according to the serologically distinct HA and NA proteins they carry. Several virus lineages with different subtype combinations are recognised. Sixteen HA subtypes, H1 to H16, and nine NAs subtypes, N1 to N9 (Fouchier et al., 2005) have been discovered in birds, but only about half of the 144 possible com- binations of those subtypes have been isolated (Munster et al., 2007). Human isolates have either the N1 or N2 NA genes. Mutations are fixed at more than 3 Â 10 À3 per site per year in the HA gene in lineages infecting people, and are fixed almost as quickly in other genes, probably as the result of linkage (Buonagurio et al., 1986; Ferguson et al., 2003; Gorman et al., 1991; Nobusawa and Sato, 2006; Kawaoka et al., 1998; Webster et al., 1992). The trajectory of new mutations can be influenced by genetic drift and natural selec- tion. Although these processes are interconnected, their individual effects vary with population size. Influenza A virus has successfully invaded different host populations with important differences in their immune system structure, size, density, mobility and lifespan. Influenza A viruses have been subject to different selective regimes 1055-7903/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2010.04.039 * Corresponding author. Present address: Department of Pathology, University of California, San Diego, USA. E-mail addresses: m.fourment@gmail.com, mfourment@ucsd.edu (M. Fourment), jeff.wood@anu.edu.au (J.T. Wood), adrian_j_gibbs@hotmail.com (A.J. Gibbs), gibbslab@gmail.com (M.J. Gibbs). 1 Present address: Curtin, Canberra, ACT 2605, Australia. Molecular Phylogenetics and Evolution 56 (2010) 526–535 Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev