Abstract
Purpose :
Mutationally constrained epitopes within infectious pathogens represent promising targets for vaccine design, but sequence-based methods have not been successful. It has been shown that constraints on viral evolution may be due to interdependencies in 3-dimensional structure. However, systematic means to directly evaluate viral protein structure and quantitate mutational constraint in viruses like HIV and HSV have not been defined.
Methods :
To address this, we applied structure-based network analysis, which utilizes protein structure data to quantify the topological importance of each amino acid residue to a protein’s structure. We used atomic-level coordinate data from the Protein Data Bank to build networks of amino acid residues by defining van der Waals interactions, hydrogen bonds, salt bridges, disulfide bonds, pi-pi interactions, pi-cation interactions, metal coordinated bonds and local hydrophobic packing. Using this network-based representation, we then calculated network centrality metrics to measure topological importance of individual amino acid residues. Integration of these metrics into a single value generated a final network score.
Results :
We validated this approach on a set of thirteen bacterial and viral proteins from comprehensive, high-throughput mutagenesis experiments. We found strong inverse correlations between computationally derived network scores and experimentally derived mutational tolerance values across all experimental datasets (Spearman’s r = -0.46 to -0.71, p = < 0.0006). We then applied the method to HIV, and residues that occupied important network positions disproportionately impaired viral replication when mutated and occurred with high frequency in T cell epitopes presented by protective HLA class I alleles (p<0.0001) and even distinguished individuals who spontaneously control HIV without antiretroviral therapy even in the absence of protective HLA alleles.
Conclusions :
Structure-based network analysis has broad potential applications. Here we describe our progress in benchmarking experiments in published bacterial and viral data as well as directed experiments in HIV. In ophthalmic disease, the application to HSV will be highly informative, as this approach provides a means to identify T cell epitopes of topological importance within the proteome of infectious pathogens and provides guidance for the design of rational immunogens including a T cell-based vaccines.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.