In Silico-Identified Peptides of Five Borrelia burgdorferi Proteins Binding with High Affinity to Human Leukocyte Antigen (HLA) Class II Alleles

To date, Lyme vaccine development has largely overlooked the vaccinee’s human leukocyte antigen (HLA) genetic makeup on which antibody production critically depends. Here, we evaluated in silico the predicted binding affinities of 192 HLA-II alleles with all 15-mer peptide sequences of five Borrelia burgdorferi proteins to identify peptides with strong binding affinity, as they would be the best candidates for antibody production in response to vaccination. We found the following: (a) 226 of the 1067 peptides tested (21.2%) were found to bind strongly to HLA-II molecules; (b) decorin-binding protein A had the greatest number of strongly binding peptides; and (c) 69 HLA-II alleles (primarily of the DRB1 gene) bound with strong affinity to peptides from Borrelia burgdorferi proteins. Finally, we tested for possible susceptibility to autoimmunity by any one of the 226 peptides above by searching for their occurrence in ~84,000 proteins of the human proteome and found overlap with only two 8-mer peptide sequences (embedded within the 226 15-mer peptides), neither of which was characterized by strong binding to HLA-I, suggesting a reduced likelihood of autoimmunity. These findings emphasize the importance of a personalized vaccine approach based on the vaccinee’s human leukocyte antigen genetic makeup and offer specific vaccine-candidate peptides that are predicted to maximize vaccine effectiveness and safety. The results of this computational study provide novel directions for future development of Lyme vaccines.