Structural patterns in class 1 major histocompatibility complex‐restricted nonamer peptide binding to T‐cell receptors

Abstract

The startling diversity in αβ T-cell receptor (TCR) sequences and structures complicates molecular-level analyses of the specificity and sensitivity determining T-cell immunogenicity. A number of three-dimensional (3D) structures are now available of ternary complexes between TCRs and peptides: major histocompatibility complexes (pMHC). Here, to glean molecular-level insights we analyze structures of TCRs bound to human class I nonamer peptide-MHC complexes. Residues at peptide positions 4-8 are found to be particularly important for TCR binding. About 90% of the TCRs hydrogen bond with one or both of the peptide residues at positions 4 and 8 presented by MHC allele HLA-A2, and this number is still ~79% for peptides presented by other MHC alleles. Residue 8, which lies outside the previously-identified central peptide region, is crucial for TCR recognition of class I MHC-presented nonamer peptides. The statistics of the interactions also sheds light on the MHC residues important for TCR binding. The present analysis will aid in the structural modeling of TCR: pMHC complexes and has implications for the rational design of peptide-based vaccines and T-cell-based immunotherapies.

TCR-H: Machine Learning Prediction of T-cell Receptor Epitope Binding on Unseen Datasets

“…In recent work we demonstrated how 3D structure can be incorporated into computational methods to improve prediction of peptide binding to MHC(Aranha et al, 2020), and how 3D MHC pocket similarity determines peptide binding strength. Further, we and others have shown that certain 3D interactions are commonly found in epitope binding to TCRs(Rajitha Rajeshwar & Smith, 2022).…”

“…In recent work we demonstrated how 3D structure can be incorporated into computational methods to improve prediction of peptide binding to MHC (Aranha et al, 2020), and how 3D MHC pocket similarity determines peptide binding strength (Yue et al, 2023). Furthermore, we and others have shown that certain 3D interactions are commonly found in epitope binding to TCRs (Rajeshwar & Smith, 2022). We therefore postulated that TCR:epitope recognition algorithms would benefit from further consideration of physicochemical properties determining recognition.…”

“…Binding primarily occurs through three complementarity-determining regions (CDRs) found on each of these chains. Although in 3D structures of TCR:epitope:MHC ternary complexes significant contacts are made outside the CDR3 region, this region most directly interacts with the peptide epitope, with CDR3b, in particular, forming direct contacts with the epitopes and MHC and displaying particularly high sequence diversity (Dai et al, 2008;Garcia et al, 2009;Morris & Allen, 2012;Szeto et al, 2021, Rajeshwar & Smith, 2022). Hence, in computational approaches CDR3b has commonly been taken as an approximation to the specificity-determining TCR sequence.…”