Revelation of Influencing Factors in Overall Codon Usage Bias of Equine Influenza Viruses

Abstract

Equine influenza viruses (EIVs) of H3N8 subtype are culprits of severe acute respiratory infections in horses, and are still responsible for significant outbreaks worldwide. Adaptability of influenza viruses to a particular host is significantly influenced by their codon usage preference, due to an absolute dependence on the host cellular machinery for their replication. In the present study, we analyzed genome-wide codon usage patterns in 92 EIV strains, including both H3N8 and H7N7 subtypes by computing several codon usage indices and applying multivariate statistical methods. Relative synonymous codon usage (RSCU) analysis disclosed bias of preferred synonymous codons towards A/U-ended codons. The overall codon usage bias in EIVs was slightly lower, and mainly affected by the nucleotide compositional constraints as inferred from the RSCU and effective number of codon (ENc) analysis. Our data suggested that codon usage pattern in EIVs is governed by the interplay of mutation pressure, natural selection from its hosts and undefined factors. The H7N7 subtype was found less fit to its host (horse) in comparison to H3N8, by possessing higher codon bias, lower mutation pressure and much less adaptation to tRNA pool of equine cells. To the best of our knowledge, this is the first report describing the codon usage analysis of the complete genomes of EIVs. The outcome of our study is likely to enhance our understanding of factors involved in viral adaptation, evolution, and fitness towards their hosts.

Fig 1. Comparative analysis of Relative Synonymous Codon Usage (RSCU) patterns of EIV subtypes (H3N8 and H7N7) with their clinical and accidental hosts.

Fig 2. Relative dinucleotides frequencies among EIV strains.

Fig 3. The selective force on the CpG dinucleotide as a function of time for H3N8 viruses. The force moves becomes larger in magnitude over time.

Fig 4. Comparative analysis of Effective Number of Codon (ENc) values in H3N8 and H7N7 subtypes.

Fig 5. Codon usage preferences of EIV subtypes in relation to the codon usage of potential host species as estimated by Codon Adaptation Index (CAI).

Fig 6. ENc-plot analysis (ENc plotted against GC3s).

ENc denotes the effective number of codons, and GC3s denotes the GC content on the third synonymous codon position. The red dotted line represents the expected curve derived from the positions of strains when the codon usage was only determined by the GC3s composition.

Fig 7. Neutrality plot analysis (GC12 vs GC3) for the entire coding sequences of EIVs.

GC12 stands for the average value of GC contents at the first and second positions of the codons (GC1 and GC2), while GC3 refers to the GC contents at the third codon position. The red dotted line is the linear regression of GC12 against GC3, R² = 0.938, P <0.001.

Fig 8. Contributions of the axes generated by Correspondence analysis (COA).

The relative and cumulative inertia of the first 20 factors from a COA of the relative synonymous codon usage values. (R.Iner.—Relative Inertia, R.Sum—Relative sum or cumulative relative inertia).

Fig 9. Correspondence analysis of the synonymous codon usage towards codons in EIVs.

The analysis was based on the RSCU values of the 59 synonymous codons. The positions of each codon were described in the first two-main-dimensional coordinates. Different base-ended codons were marked in the figure, where the brown star, blue triangle, green triangle, and red circle refer to codons ending with A, U, G, and C, respectively.

Fig 10. Correspondence analysis of the synonymous codon usage in two subtypes of EIVs.

The analysis was based on the RSCU values of the 59 synonymous codons. The positions of each EIV strain were described in the first two-main-dimensional coordinates. Different subtypes of EIVs were marked in the figure, where the blue circle and red triangle refer to H3N8 and H7N7 subtypes, respectively.