Evolutionary history and phylodynamics of influenza A and B neuraminidase (NA) genes inferred from large-scale sequence analyses

Abstract

Background: Influenza neuraminidase (NA) is an important surface glycoprotein and plays a vital role in viral replication and drug development. The NA is found in influenza A and B viruses, with nine subtypes classified in influenza A. The complete knowledge of influenza NA evolutionary history and phylodynamics, although critical for the prevention and control of influenza epidemics and pandemics, remains lacking.

Methodology/principal findings: Evolutionary and phylogenetic analyses of influenza NA sequences using Maximum Likelihood and Bayesian MCMC methods demonstrated that the divergence of influenza viruses into types A and B occurred earlier than the divergence of influenza A NA subtypes. Twenty-three lineages were identified within influenza A, two lineages were classified within influenza B, and most lineages were specific to host, subtype or geographical location. Interestingly, evolutionary rates vary not only among lineages but also among branches within lineages. The estimated tMRCAs of influenza lineages suggest that the viruses of different lineages emerge several months or even years before their initial detection. The d(N)/d(S) ratios ranged from 0.062 to 0.313 for influenza A lineages, and 0.257 to 0.259 for influenza B lineages. Structural analyses revealed that all positively selected sites are at the surface of the NA protein, with a number of sites found to be important for host antibody and drug binding.

Conclusions/significance: The divergence into influenza type A and B from a putative ancestral NA was followed by the divergence of type A into nine NA subtypes, of which 23 lineages subsequently diverged. This study provides a better understanding of influenza NA lineages and their evolutionary dynamics, which may facilitate early detection of newly emerging influenza viruses and thus improve influenza surveillance.

Figure 1. Phylogeny of influenza A and B neuraminidase (NA) genes.

Influenza NA genes form two groups (Group I and Group II), which correspond to influenza A and B, respectively. Influenza A NA is further classified into two subgroups (Subgroup I and Subgroup II). The viral strains are colored for different hosts: human in green, swine in blue, avian in red and equine in purple. The bootstrap support values are indicated at major nodes. The scale bar at the bottom indicates the numbers of nucleotide substitutions per site.

Figure 2. Maximum-likelihood (ML) tree of influenza A NA subtypes.

A: N1; B: N2; C: N5; D: N8. The annotation for each lineage was labeled on the trees. Three lineages in N1 (1A, 1B and 1C), two lineages in N2 (2A and 2B), two lineages in N5 (5A and 5B), and two lineages in N8 (8A and 8B) were classified. The bootstrap values supporting the corresponding lineages are shown to the left of the major nodes. Scale bars indicate the numbers of nucleotide substitutions per site.

Figure 3. Maximum-likelihood (ML) tree of influenza B NA genes.

Two lineages, Yam88 and Vic87, were classified. The bootstrap values supporting the corresponding lineages are shown on the major nodes. The scale bars indicate the numbers of nucleotide substitutions per site.

Figure 4. Bayesian inferences of random local clocks on influenza NA genes.

A: H5N1 (1A.1), B: North American swine N1 (1B), C: Human H1N1 (1C), D: H9N2 (2A.1), E: Equine N7 (7C), F: Yama88 influenza B NA (Yama88). Branch coloring indicates inferred rates of nucleotide substitution from blue (slow) to red (fast). The scale bar indicates the number of years before the present.

Figure 5. The structures and positive selection sites of human influenza neuraminidase.

A: Influenza A human N1 neuraminidase (1C) (A/Brevig Mission/1/18 H1N1, 1918 “Spanish flu”, PDB ID: 3B7E); B: Influenza A human N2 neuraminidase (2B) (A/Tokyo/3/67 H2N2, 1967, PDB ID: 1IVG); C: Influenza B viral neuraminidase for Yam88 (B/Perth/211/2001, PDB ID: 3K36); D: Influenza B viral neuraminidase for Vic87 (B/Perth/211/2001, PDB ID: 3K36). The positive selection sites are denoted as green balls. Structural regions are denoted in different colors: yellow for alpha-helices, red for beta sheets, and blue for loops.

Figure 6. The evolutionary dynamics of influenza neuraminidase (NA) over time.

The lineages from different hosts are colored, with the emergence times of the lineages represented by the horizontal positions of squared boxes and the mean substitution rates depicted by the degree of line thickness. Note that within 2A there are five swine clusters.