Applying bioinformatics for antibody epitope prediction using affinity-selected mimotopes – relevance for vaccine design

Denisova et al. Immunome Research 2010, 6(Suppl 2):S6 http://www.immunome-research.com/content/6/S2/S6 IMMUNOME RESEARCH REVIEW Open Access Applying bioinformatics for antibody epitope prediction using affinity-selected mimotopes – relevance for vaccine design Galina F Denisova*, Dimitri A Denisov, Jonathan L Bramson Abstract To properly characterize protective polyclonal antibody responses, it is necessary to examine epitope specificity. Most antibody epitopes are conformational in nature and, thus, cannot be identified using synthetic linear peptides. Cyclic peptides can function as mimetics of conformational epitopes (termed mimotopes), thereby providing targets, which can be selected by immunoaffinity purification. However, the management of large collections of random cyclic peptides is cumbersome. Filamentous bacteriophage provides a useful scaffold for the expression of random peptides (termed phage display) facilitating both the production and manipulation of complex peptide libraries. Immunoaffinity selection of phage displaying random cyclic peptides is an effective strategy for isolating mimotopes with specificity for a given antiserum. Further epitope prediction based on mimotope sequence is not trivial since mimotopes generally display only small homologies with the target protein. Large numbers of unique mimotopes are required to provide sufficient sequence coverage to elucidate the target epitope. We have developed a method based on pattern recognition theory to deal with the complexity of large collections of conformational mimotopes. The analysis consists of two phases: 1) The learning phase where a large collection of epitopespecific mimotopes is analyzed to identify epitope specific “signs” and 2) The identification phase where immunoaffinity-selected mimotopes are interrogated for the presence of the epitope specific “signs” and assigned to specific epitopes. We are currently using computational methods to define epitope “signs” without the need for prior knowledge of specific mimotopes. This technology provides an important tool for characterizing the breadth of antibody specificities within polyclonal antisera. Introduction Antibodies play a central role in immune memory and long-term protective responses. Serum antibodies for specific pathogens are recognized as a primary read-out for vaccination and recent studies have revealed that pathogen-specific antibodies persist for decades following vaccination [1]. It is important, however, to appreciate that not all antibodies can prevent infection. As an example, a collection of monoclonal antibodies have been isolated against the West Nile virus envelope protein (E) which recognize distinct epitopes within the protein. However, only antibodies that bind to specific epitopes can produce virus neutralization and protective * Correspondence: Department of Pathology and Molecular Medicine, Centre for Gene Therapeutics, McMaster University, 1200 Main Street West, Hamilton, Ontario, Canada, L8N 3Z5 Full list of author information is available at the end of the article immunity in vivo [2]. It was shown also that although many anti-HER2 antibodies inhibit the growth of cancer cells, some of them have no effect on cell growth, while others actively stimulate cancer growth. It has been proposed that this wide spectrum of biological effects is related to the epitope specificity of the Abs and to consequent changes in receptor signalling [3][4]. Therefore, to properly characterize a protective humoral response and to use it for vaccine design, it is necessary to characterize the epitope specificity, in addition to antibody titers. A typical strategy for monitoring specific antibodies to known antigens involves the use of ELISAs coated with recombinant protein(s) or the vaccine itself. When combined with serologic analysis of recombinant cDNA expression libraries (SEREX), ELISAs are a powerful tool for monitoring humoral responses in various disease states where the antigens may not be known a priori © 2010 Denisova et al; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Denisova et al. Immunome Research 2010, 6(Suppl 2):S6 http://www.immunome-research.com/content/6/S2/S6 [5-7]. With regard to autoimmune antibodies, the availablity of whole transcriptome sequences and effective protein expression systems has made it possible to screen the complete human proteome. Protein arrays can be used for monitoring responses to a broad range of proteins. Such arrays have been produced using genomic information [6,7] or using information available in the SEREX database [8,9], which collects data for antigenic proteins identified using polyserum from cancer patients. All of these methods are useful tools for measuring the magnitude and breadth of the humoral response, but they reveal little information regarding epitope specificity. While it is technically feasible to engineer recombinant proteins with specific mutations designed to disrupt putative epitopes, due to the complexity of protein folding, epitopes can be disrupted at sites distal to the mutations that will complicate interpretation of the results. Epitope-specificity can be determined for linear epitopes by screening libraries of short synthetic peptides that span the entire target antigen (PEPSCAN) [10,11]. However, the majority of antibody responses [12] are directed at structural epitopes which are difficult to recapitulate with synthetic peptides because they are typically formed by protein folding and, thus, are composed of amino acid residues which are often separated by great distances within the linear protein sequence. Immunoaffinity selection of random peptides offers an alternate strategy to characterize antibody epitopes because the affinity selection will identify peptides with spatially-proximal residues that may be distant from each other according to linear sequence. Indeed, this strategy offers an unbiased method to screen for epitope mimetics (mimotopes) [13] that can define antibody targets and serve directly as immunogens. To effectively apply this strategy, it is necessary to isolate many random peptides with unique sequences because each immuno-selected peptide will carry only partial homology for the original target. Through the use of computational modeling, it is possible to derive a consensus sequence from the selected peptides and identify the target epitope. The use of linear random peptides for this strategy is limited because the conformation space available to linear peptides is great, allowing the linear peptides to assume a large array of conformations. Constraining the peptide by cyclization reduces the field of conformational possibilities for the molecule and results (...truncated)