Glycosylation site alteration in the evolution of influenza A (H1N1) viruses

Abstract

Influenza virus typically alters protein glycosylation in order to escape immune pressure from hosts and hence to facilitate survival in different host environments. In this study, the patterns and conservation of glycosylation sites on HA and NA of influenza A/H1N1 viruses isolated from various hosts at different time periods were systematically analyzed, by employing a new strategy combining genome-based glycosylation site prediction and 3D modeling of glycoprotein structures, for elucidation of the modes and laws of glycosylation site alteration in the evolution of influenza A/H1N1 viruses. The results showed that influenza H1N1 viruses underwent different alterations of protein glycosylation in different hosts. Two alternative modes of glycosylation site alteration were involved in the evolution of human influenza virus: One was an increase in glycosylation site numbers, which mainly occurred with high frequency in the early stages of evolution. The other was a change in the positional conversion of the glycosylation sites, which was the dominating mode with relatively low frequency in the later evolutionary stages. The mechanisms and possibly biological functions of glycosylation site alteration for the evolution of influenza A/H1N1 viruses were also discussed. Importantly, the significant role of positional alteration of glycosylation sites in the host adaptation of influenza virus was elucidated. Although the results still need to be supported by experimental data, the information here may provide some constructive suggestions for research into the glycosylation of influenza viruses as well as even the design of surveillance and the production of viral vaccines.

Figure 1. Structural overviews of the HA and NA monomers for human influenza H1N1 viruses.

Each represented a unique glycosylation site pattern on HA or NA. (A) The monomers of HA with glycans attached at the potential glycosylation sites for representative human influenza H1N1 viruses from 1918 to 2010. (B) The monomers of NA with glycans attached at the potential glycosylation sites for representative human influenza H1N1 viruses from 1918 to 2010. (C) A monomer of HA with the location of the receptor-binding site (RBS) and the five antigenic sites , . Regions A and B are important regions with frequently altered glycosylation sites. (D) A monomer of NA with the location of the enzyme active site and the seven antigenic sites surrounding the enzyme active , . Region C is an important glycosylation region surrounding the enzyme active site. The amino acid locations are numbered according to the HA of SC 1918 and NA of BM 1918 numbering, respectively.

Figure 2. The alteration processes of glycosylation sites in some important regions of HA and NA.

(A) The alteration process of glycosylation sites in region A (located at the receptor binding domain) of HA. (B) The alteration process of glycosylation sites in region B (located at the vestigial esterase domain) of HA. (C) The alteration process of glycosylation sites in region C (located around the enzymatic active site) of NA. (D) The alteration process of glycosylation sites on the stalk of NA. The dotted lines represented the superficial alterations based on genome-based analysis, while the corresponding full lines illustrated the possibly alteration processes after further analysis by homology modeling and in silico protein glycosylation.