Study on phylogenetic relationships, variability, and correlated mutations in M2 proteins of influenza virus A

Abstract

M2 channel, an influenza virus transmembrane protein, serves as an important target for antiviral drug design. There are still discordances concerning the role of some residues involved in proton transfer as well as the mechanism of inhibition by commercial drugs. The viral M2 proteins show high conservativity; about 3/4 of the positions are occupied by one residue in over 95%. Nine M2 proteins from the H3N2 strain and possibly two proteins from H2N2 strains make a phylogenic cluster closely related to 2RLF. The variability range is limited to 4 residues/position with one exception. The 2RLF protein stands out by the presence of 2 serines at the positions 19 and 50, which are in most other M2 proteins occupied by cysteines. The study of correlated mutations shows that there are several positions with significant mutational correlation that have not been described so far as functionally important. That there are 5 more residues potentially involved in the M2 mechanism of action. The original software used in this work (Consensus Constructor, SSSSg, Corm, Talana) is freely accessible as stand-alone offline applications upon request to the authors. The other software used in this work is freely available online for noncommercial purposes at public services on bioinformatics such as ExPASy or NCBI. The study on mutational variability, evolutionary relationship, and correlated mutation presented in this paper is a potential way to explain more completely the role of significant factors in proton channel action and to clarify the inhibition mechanism by specific drugs.

Figure 1. Multiple sequence alignment (part) of the influenza virus M2 proton channel proteins.

The consensus sequence for aligned 43 amino acid fragments was obtained for the parameters: identity 96.65% (88), significance 29.35% (27), gaps 50% (46).

Figure 2. The cladograms of 2RLF cluster obtained with the different approaches.

The results were obtained with the aid of the following software: (A) ClustalX, (B) SSSSg, (C) Phylip (maximum parsimony), (D) Phylip (maximum likelihood), (E) ConSurf.

Figure 3. ConSurf color-coded multiple sequence alignment of M2 family.

Only a part of the multiple alignment is shown. The different colors indicate the variability range of each position. See the text for details.

Figure 4. The mutational variability color scale plotted on the chain A of 2RLF protein structure.

The results obtained with the aid of ConSurf (A) and Talana (B).

Figure 5. Mutationally correlated positions plotted on 2RLF chain A presented in the mutational variability mode.

The variability scale is shown below the structures. The sequences indicate the location of positions (cyan) correlated to the reference position (purple) for the structures A, B, C, and D respectively. In B, the reference position is not indicated due to the mutual correlation for all 3 positions (using each of those positions as the reference gives the same cluster of correlation).