Positive selection operates continuously on hemagglutinin during evolution of H3N2 human influenza A virus
Introduction
Influenza viruses are members of the family Orthomyxoviridae (Shope, 1931), containing a segmented and single-stranded (negative-sense) RNA genome in an enveloped virion (Noda et al., 2006). The genome encodes envelope glycoproteins, matrix proteins, nonstructural proteins, nucleoproteins, and polymerase proteins. Influenza viruses are classified into types A–C (Suzuki and Nei, 2002), among which the A virus is the most pathogenic to humans, causing 3–5 million cases of severe tracheobronchitis and 0.25–0.5 million deaths worldwide during annual epidemics (World Health Organization, 2003).
Hemagglutinin (HA) and neuraminidase (NA) are the envelope glycoproteins of influenza A virus. HA is cleaved into a signal peptide, HA1, and HA2 during maturation. HA1 is the major target of humoral immunity against influenza A virus. According to the antigenic properties of HA1, which are determined by epitopes A–E, and those of NA, influenza A viruses are classified into 16 (H1–H16) and 9 (N1–N9) subtypes, respectively (World Health Organization, 1980). H1N1 and H3N2 viruses are co-circulating in humans, among which the latter is more prevalent and pathogenic than the former (Nelson and Holmes, 2007).
When phylogenetic analysis was conducted for HA1s of H3N2 human influenza A viruses isolated in various years, it was found that isolates obtained in different years usually formed distinct clusters (Fitch et al., 1991). The clusters branched successively in the chronological order of isolation years from the trunk, which was defined as a set of branches [trunk (T) branches] connecting the root of the phylogenetic tree and the cluster of the latest isolates. It has been proposed that the trunk corresponds to evolution of the virus circulating continuously in the East and Southeast (E-SE) Asia or the tropics, and each cluster corresponds to annual epidemics in temperate regions, which are seeded from the E-SE Asia or the tropics and do not contribute to long-term evolution of the virus (Rambaut et al., 2008, Russell, 2008).
The antigenicity of influenza A virus evolves mainly by accumulation of amino acid substitutions (antigenic drift). When antigenic distances between HA1s of H3N2 human influenza A viruses isolated in various years were measured by the hemagglutination inhibition (HI) assay, isolates were found to form distinct antigenic clusters (Smith et al., 2004). It was proposed that antigenic evolution was punctuated, where the evolution was mostly small and occurred within antigenic clusters, and the cluster transition occurred only occasionally (de Jong et al., 2007). In contrast, evolution of nucleotide and amino acid sequences for HA1 appeared to be continuous.
To understand the mechanisms of antigenic evolution, it is interesting to detect natural selection operating on epitopes A–E of HA1. Natural selection operating on HA1 of H3N2 human influenza A virus has been examined by population genetic and molecular evolutionary analyses. In a population genetic analysis, a phylogenetic tree was constructed for HA1s of the viruses isolated in various years. For each T branch, all isolates were divided into antecedents and descendants, and the time interval required for extinction (fixation) of antecedents (descendants) was examined (Wolf et al., 2006). Some T branches were associated with rapid extinction of antecedents, whereas others with slow extinction, and it was proposed that evolution of HA1 was predominated by long-intervals of antigenic stasis, where neutral or nearly neutral amino acid substitutions accumulated to provide the basis of antigenic innovations, which were punctuated by short-intervals of antigenic changes, where positive selection operated on amino acid substitutions that caused antigenic changes through epistasis with the substitutions accumulated during the stasis (Koelle et al., 2006). However, frequencies of sequence variants for HA1 were also observed to change in nearly all the years under investigation, suggesting that positive selection was continuous (Shih et al., 2007).
In a molecular evolutionary analysis, the rates of synonymous (rS) and nonsynonymous (rN) substitutions were compared by dividing HA1 into epitopes A–E and other (non-epitope) sites, as well as dividing the phylogenetic tree into the T branches and other (NT) branches. It was observed that rN/rS > 1 for epitopes A–E on the T branches, whereas rN/rS < 1 for non-epitope sites and on the NT branches, suggesting that positive and negative selection operated for the former and the latter, respectively (Fitch et al., 1991, Wolf et al., 2006). However, the relationship between antigenic evolution and positive selection was unclear. The purpose of the present study was to examine this relationship by molecular evolutionary analysis.
Section snippets
Sequence and antigenic data
The sequence and antigenic data of HA1s for 255 H3N2 human influenza A viruses isolated in various years were available in Smith et al. (2004). Since 89% of these isolates were propagated only in mammalian cell cultures for at most 5 passages (usually only 1 or 2 passages), the passage history did not appear to affect detection of natural selection to any large extent (Bush et al., 2000, Zhai et al., 2007). After eliminating the isolates whose sequences contained ambiguous nucleotides, minor
Phylogenetic tree for HA1
The phylogenetic trees for HA1s of 209 H3N2 human influenza A viruses and 1 duck virus constructed by the NJ method assuming TVM + Γ and TVM + Γ + I that were best fitted to the data are shown in Fig. 1, Fig. 2, respectively. In both phylogenetic trees, each isolate is color-coded according to the antigenic cluster. The T branches, which were colored red in Fig. 1, Fig. 2, were easily identified as a set of branches connecting the root of the human sequences and the cluster of the latest isolates.
Positive selection operates continuously on HA1 during evolution of H3N2 human influenza A virus
It has been proposed that antigenic evolution of HA1 for H3N2 human influenza A virus was punctuated (Smith et al., 2004). In the population genetic analysis, however, it was controversial whether positive selection operated on HA1 in a punctuated manner for the C branches (Wolf et al., 2006), or continuously (Shih et al., 2007). In the molecular evolutionary analysis, positive selection was detected for the T branches (Fitch et al., 1991) but the relationship between antigenic evolution and
Acknowledgments
The author thanks two anonymous reviewers for valuable comments. The present study was supported by KAKENHI 17770007 and KAKENHI 20570008 to Y.S.
References (27)
- et al.
Changing selective pressure during antigenic changes in human influenza H3
PLoS Pathog.
(2008) - et al.
Effects of passage history and sampling bias on phylogenetic reconstruction of human influenza A evolution
Proc. Natl. Acad. Sci. U. S. A.
(2000) Antigenic and genetic evolution of swine influenza A (H3N2) viruses in Europe
J. Virol.
(2007)- et al.
Positive Darwinian evolution in human influenza A viruses
Proc. Natl. Acad. Sci. U. S. A.
(1991) - et al.
Epochal evolution shapes the phylodynamics of interpandemic influenza A (H3N2) in humans
Science
(2006) - et al.
High temperature (30 degrees C) blocks aerosol but not contact transmission of influenza virus
J. Virol.
(2008) Stochastic processes are key determinants of short-term evolution in influenza A virus
PLoS Pathog.
(2006)- et al.
The evolution of epidemic influenza
Nat. Rev. Genet.
(2007) Architecture of ribonucleoprotein complexes in influenza A virus particles
Nature
(2006)- et al.
MODELTEST: testing the model of DNA substitution
Bioinformatics
(1998)
The genomic and epidemiological dynamics of human influenza A virus
Nature
The global circulation of seasonal influenza A (H3N2) viruses
Science
Patterns of nucleotide substitutions in influenza A virus genes
Jpn. J. Genet.
Cited by (23)
Methods for reducing the number of sequences in molecular evolutionary analyses
2020, Meta GeneCitation Excerpt :However, the πN/πS in the SDM method fluctuated to a relatively large extent. Reportedly, positive selection has operated on multiple lineages for GII.4 HuNoV VP1 (Tohma et al., 2019) and on a single trunk lineage for H3N2 HuIAV HA (Fitch et al., 1991; Wolf et al., 2006; Suzuki, 2008). These observations suggested that the difference in the trajectory of πN/πS in the SDM method between GII.4 HuNoV VP1 and H3N2 IAV HA may reflect difference in the pattern of accumulation of advantageous mutations on the phylogenetic tree.
Selecting vaccine strains for H3N2 human influenza A virus
2015, Meta GeneCitation Excerpt :These results suggest that a strain randomly chosen in the pre-target season was expected to be more effective than the actual vaccine. This is probably because H3N2 human influenza A virus has been continuously escaping from the host immune responses through antigenic evolution (Shih et al., 2007; Suzuki, 2008). It is therefore recommended to replace the vaccine strain for every target season by a strain isolated in the pre-target season.
Molecular and genetic analysis of NS gene from high pathogenic strains of the avian influenza (H5N1) virus isolated in Kazakhstan
2011, GeneCitation Excerpt :A distinctive feature of the influenza virus is its unique aptitude for higher mutability as compared to other agents and for frequent mutations in genome (Suzuki, 2010). First of all, it concerns the hemagglutinin gene that carries determinants responsible for formation of the virus-neutralizing immunity (Suzuki, 2008). Specific changes were shown to occur in other influenza virus genes, particularly in genes of non-structural proteins and polymerase.
Comparison of selection pressures on the HA gene of pandemic (2009) and seasonal human and swine influenza A H1 subtype viruses
2010, VirologyCitation Excerpt :Antigenic evolution of the virus was sufficient to recommend eight different strains for the H1N1 component of the influenza virus vaccine from 1977 to 2009 (Hay et al., 2001; Health Protection Agency, 2009). Based on the theory of natural selection, positively selected sites in HA could be found by calculating the selection pressure (Suzuki, 2006, 2008). This may be useful for identifying the epitopes involved in the elimination of viruses from infected patients.
The 2009 H1N1 influenza outbreak in its historical context
2009, Journal of Clinical VirologyCitation Excerpt :Haemagglutinin in particular has a high amino acid substitution rate in its epitope regions. The ratio of non-synonymous substitutions to synonymous substitutions (omega) significantly exceeds 1 at several amino acid positions, indicating a selective advantage for novel amino acid sequences in this region.6,8 This appears to be directly related to evasion of the host immune system.