PSI Structural Biology Knowledgebase

PSI | Structural Biology Knowledgebase
Header Icons

Related Articles
Predicting Protein Crystal Candidates
October 2014
Drug Discovery: Finding Druggable Targets
October 2013
Membrane Proteome: Unveiling the Human α-helical Membrane Proteome
August 2013
Infectious Diseases: Determining the Essential Structome
May 2013
Infectious Diseases: Targeting Meningitis
May 2013
Microbial Pathogenesis: Computational Epitope Prediction
January 2013
Microbial Pathogenesis: Influenza Inhibitor Screen
January 2013
Targeting Enzyme Function with Structural Genomics
July 2012
Disordered Proteins
February 2012
The cancer kinome
April 2010
Learning from failure
December 2009
Dealing with difficult families
February 2009

Technology Topics Target Selection

Microbial Pathogenesis: Computational Epitope Prediction

SBKB [doi:10.1038/sbkb.2012.120]
Technical Highlight - January 2013
Short description: A computational approach uses structural information to identify epitopes within antigenic proteins.

Epitope identification within antigens. OppA 3D structure (A-B) is the basis for computational predictions (C) of experimentally reactive epitopes (D). Figure courtesy of Claudio Peri.

Despite rigorous efforts to develop vaccines against them, several pathogens remain refractory to vaccination. To prevent the spread of these pathogens, new tools and techniques are required to develop vaccines with improved properties, enhanced efficacy, and engineered cross-reactivity. One of those approaches lies in structural vaccinology, which uses the three-dimensional (3D) structure of an antigenic protein to identify epitopes that can be used for vaccine design or as diagnostic tools.

Using the 3D structure of oligopeptide-binding protein A (OppABp) from the infectious bacterium Burkholderia pseudomallei—the causative agent of melioidosis— Colombo, Bolognesi and colleagues describe a structure-based computational method to identify epitopes within antigenic proteins. OppABp is a known virulence factor and has previously been shown to elicit an immune response in mice; however, it was unable to protect the vaccinated animals against subsequent infection.

The authors used both computational and experimental methods to identify epitopes within the protein. By comparing the results from two computational epitope prediction methods analyzing the surface properties of proteins —matrix of local coupling energies (MLCE), which was specifically designed to identify antigenic epitopes, and electrostatic desolvation profiles (EDPs), which identify protein-protein interfaces—three consensus sequences were identified. Three further sequences were identified via experimental epitope mapping techniques, which use protein proteolysis fragments and can therefore identify epitopes buried within the protein.

To improve the predictive abilities of the computational methods, the authors applied the MLCE and EDP methods to three fragments of OppABp that had been created via a recently developed decomposition algorithm. This method yielded epitopes that overlapped with those identified via experimental mapping.

The immunogenicity of the epitopes was confirmed by testing them against human sera from three types of individuals: uninfected patients, infected patients who were healthy, and infected patients who had recovered. While all the peptides were recognized by antibodies from infected individuals, some reacted more strongly in the healthy infected group, indicating that they may be useful as diagnostic tools.

The results demonstrate the complementarity of experimental and computational methods and how experimental results can be used to improve and refine computational techniques. With regard to vaccine design, these approaches can help identify new epitopes that may be used in vaccine design or as diagnostic tools.

Jennifer Cable


  1. P. Lassaux et al. A structure-based strategy for epitope discovery in Burkholderia pseudomallei OppA antigen.
    Structure. 21, 1-9 (2013). doi:10.1016/j.str.2012.10.005

Structural Biology Knowledgebase ISSN: 1758-1338
Funded by a grant from the National Institute of General Medical Sciences of the National Institutes of Health