Positive selection in the Ebola virus proteome

To gain insight into the recent evolution of EBOLA virus (EBOV) proteins and identify evolutionary hotspots, I investigated patterns of amino acid substitutions in the proteomes of 49 members of EBOV genus including EBOV, Bundibugyo virus (BDBV), Reston virus (RESTV), Sudan virus (SUDV) and Tai Forest virus (TAFV) spanning outbreaks from 1974 to 2014 in Democratic Republic of Congo (DRC), Uganda, Gabon, Sudan, Ivory Coast, Philippines, and Guinea among humans, non-human primates, and pigs. Of the seven genes of the Ebola genome, I found significant variations in evolutionary rates of Glycoprotein (GP), RNA/dependent RNA polymerase (L), and Nucleoprotein (NP) across different lineages. GP and NP were found to evolve at 5- and 3-fold higher rate than other EBOV genes. Several residues in GP and NP show significant dN/dS >1 along the internal branches of the phylogenetic tree leading to Philippines and Sudan outbreaks. Most of these residues are located in solvent exposed areas shown previously to be antigenic. We further identified significant changes in specific amino acid properties of GP, NP and L during the recent evolutionary history of EBOV outbreaks and in particular during the 2014 outbreak. The positively selected residues and chemical properties are consistent with changes in epitope interaction and could thus be important in detecting adaptation of EBOV to the host immune defense system.