The prevalent Boxer MHC class Ia allele Dog Leukocyte Antigen (DLA)-88*034:01 presents 9-mer peptides with a defined binding motif

Abstract Adoptive cell transfer (ACT) for chemoresistant malignancies can induce dramatic remissions and prolong life, but factors that determine T-cell efficacy and off-target/-tumor events associated with therapy remain incompletely understood. Modeling ACT in spontaneous canine cancer models could provide critical insights and propel development; however, relevant peptides from tumor-specific antigens (TSAs) have yet to be reported in dogs. Discovery of TSA epitopes restricted by a high-prevalence MHC allele can be facilitated by in silico analysis using a defined binding motif and a large peptide dataset to train prediction software. We sought to characterize the binding preferences of the DLA-88*034:01 allele, carried by 82% of Boxers, an inbred, popular breed with increased risks for glioma and lymphoma, important ACT targets. We hypothesized that DLA-88*034:01 presents diverse self-peptides with conserved residue preferences at ≥1 position(s) that can be inferred by pattern analysis. To test this prediction, DLA-88*034:01-FLAG was immunoprecipitated from large-scale cultures of a canine cell clone, PN62, and alkaline-eluted peptides were sequenced by liquid chromatography-tandem mass spectrometry. DLA-88*034:01 prefers peptides that are 9 amino acids (AAs) in length. Analysis of 640 9-mers showed a preference for acidic AAs at position(P)1, non-bulky hydrophobic AAs at P2, and basic AAs at P4. At P6 and P9, H and F were highly favored, respectively. Bulky hydrophobic and charged AAs were unfavorable at P2, as were W, M and Y at P8. This dataset will allow more accurate prediction of TSA-derived peptides recognized by CTL in a majority of Boxers with glioma or lymphoma, and should foster development of an ACT model in this breed.