From viral genome to specific peptide epitopes: methods for identifying porcine T cell epitopes based on in silico predictions, in vitro identification and ex vivo verification

Abstract The affinity with which major histocompatibility complex (MHC) class I molecules bind peptides is instrumental to presentation of viral epitopes to cytotoxic T lymphocytes (CTLs). We analyzed three swine leukocyte antigen (SLA) molecules for complete nonamer peptide-based binding matrices in order to predict likely candidates for peptide-SLA binding. These results were combined with binding predictions generated by the algorithm, NetMHCpan (http://www.cbs.dtu.dk/services/NetMHCpan/) in order to select peptide candidates for in vitro analysis. The correlation between high affinity and high stability was determined using luminescence oxygen channeling (LOCI) and scintillation proximity assay (SPA) for peptides bound by the commonly occurring SLA molecules, SLA-1*0401, SLA-2*0401, and SLA-3*0401. Using these tools, peptides bound by SLA with high affinity and stability were identified from a library of 10.000 peptides. T cell epitopes were identified using peptide-SLA complexes assembled into fluorescent tetramers to stain swine influenza specific CTLs derived from immunized animals and MHC-defined pigs vaccinated against foot-and-mouth disease virus. These results demonstrate the broad applicability of methods originally developed for analysis of human leukocyte antigen (HLA) presentation of peptides. The methods presented provide a timely and cost-effective approach to CTL epitope discovery that can be applied to diseases of swine and of other mammalian species of interest.