Genetic Evolution and Molecular Selection of the HE Gene of Influenza C Virus

Abstract

Influenza C virus (ICV) was first identified in humans and swine, but recently also in cattle, indicating a wider host range and potential threat to both the livestock industry and public health than was originally anticipated. The ICV hemagglutinin-esterase (HE) glycoprotein has multiple functions in the viral replication cycle and is the major determinant of antigenicity. Here, we developed a comparative approach integrating genetics, molecular selection analysis, and structural biology to identify the codon usage and adaptive evolution of ICV. We show that ICV can be classified into six lineages, consistent with previous studies. The HE gene has a low codon usage bias, which may facilitate ICV replication by reducing competition during evolution. Natural selection, dinucleotide composition, and mutation pressure shape the codon usage patterns of the ICV HE gene, with natural selection being the most important factor. Codon adaptation index (CAI) and relative codon deoptimization index (RCDI) analysis revealed that the greatest adaption of ICV was to humans, followed by cattle and swine. Additionally, similarity index (SiD) analysis revealed that swine exerted a stronger evolutionary pressure on ICV than humans, which is considered the primary reservoir. Furthermore, a similar tendency was also observed in the M gene. Of note, we found HE residues 176, 194, and 198 to be under positive selection, which may be the result of escape from antibody responses. Our study provides useful information on the genetic evolution of ICV from a new perspective that can help devise prevention and control strategies.

Keywords: Influenza C virus; codon usage bias; hemagglutinin-esterase fusion glycoprotein (HE); natural selection; selection pressure.

Figure 1

(A) ML phylogenetic tree of the ICV HE gene reconstructed in RAxML (_v8.2.10) using the GTR + GAMMA + I model and supported by 1000 bootstraps. The observed six individual lineages: C/Kanagawa, C/Yamagata, C/Aichi, C/Sao Paulo, C/Taylor, and C/Mississippi lineages are represented in dark blue, orange, light green, yellow, light blue and cyan, respectively. Coloured boxes along the right of the tree represent the hosts and countries of isolation. The host species coloured in dark green, light brown, and dark grey represent human, swine, and bovine, respectively. The individual countries are indicated with different colours. The no-coloured regions denote no related information recoded in NCBI. PCA analysis explained with the first two axes. (B) PCA according to the six evolutionary lineages (colours codes same as A). (C) PCA according to the isolated host species (colours same as A).

Figure 2

(A). The mean and standard deviation of effective number of codons (ENC). The respective evolutionary lineages: C/Kanagawa, C/Yamagata, C/Aichi, C/Sao Paulo, C/Taylor lineage, and C/Mississippi are represented in dark blue, orange, light green, yellow, light blue, and cyan, respectively. (B) ENC-plot analysis of the ICV HE gene. The curve represents the standard expected values. The different colour circles represent the observed ENC-GC_3s values of the individual lineages (colours same as A). (C) PR2 bias plot of the HE gene.

Figure 3

Neutrality analysis of the ICV HE gene. GC_3s are plotted against GC_12s. (A) C/Aichi lineage, C/Kanagawa lineage, and C/Yamagata lineage. (B) C/Mississippi lineage, C/Taylor lineage, and C/Sao Paulo lineage.

Figure 4

Effective number of codons (ENC) (A) and the neutrality analysis (B) of the ICV M gene.

Figure 5

(A) Codon adaption index (CAI) and relative codon deoptimization index (RCDI) of the ICV HE gene. The abscissa denotes the evolutionary lineages and the ordinate denotes the host species. The upper panel corresponds to RCDI and the bottom panel corresponds to CAI analysis. The dark green, light brown, and dark grey indicate human, swine, and bovine, respectively. (B) Similarity index (SiD) analysis of the ICV HE gene, and the evolutionary lineages in relation to the host species.

Figure 6

The codon adaption index (CAI), the relative codon deoptimization index (RCDI) (A), and similarity index (SiD) analysis of the ICV M gene in relation to the host species. Dark green, pink, and black indicate human, swine, and bovine, respectively.

Figure 7

Localisation of amino acids under positive selection in the crystal structure of HE. (A) Complete ectodomain of an HE monomer from the 1/JHB/1/66 strain shown as a cartoon. The HE2 subunit is drawn in green and the HE1 in blue. Amino acids involved in receptor binding and esterase activity are shown as green and red sticks, respectively. Amino acids under positive selection (N176, E194 and K198) are shown as red spheres. (B) Magnification of the top part of HE containing the relevant amino acids. Note that N176 is part of a used N-glycosylation site (N175, N176, S177). N175 is shown as an orange stick. Figures were created with PyMol from pdb file 1FLC.