Relative implication of peptide residues in binding to major histocompatibility complex class I H-2Db: application to the design of high-affinity, allele-specific peptides.

The H-2Db peptide sequence SMIENLEYM was manipulated (N- and C-terminus truncation and alanine substitution) to determine the role of structural elements (peptide ends and residue side chains) in binding to H-2Db. We found that good binding affinity could be obtained by compensating the minimal binding condition for one element by the optimal condition of the other element. In particular, we showed, that although the minimal binding sequence could be as short as a heptamer (deletion of positions 1 and 2), it needed the presence of optimal amino acids at other positions (IENLEYM). Conversely, the structurally minimal peptide would accept multiple alanine residues, but required the optimal nonameric length (AAAENAEAA). Positions 1, 2, 3, 4, 5, 7 and 9, but not 6 and 8, were involved in the H-2Db-peptide interaction. Most residues interacted directly with the MHC molecule via their main chain (amino and carboxyl) atoms (positions 1 and 2), their side chains (positions 3 and 5), or both (position 9). Positions 4 and 7 were found to play an indirect role, probably by influencing the secondary structure. At the C-terminus, the presence of a residue at position 9, but not the hydrophobic nature of its side chain, was mandatory for binding. At the N-terminus, the role of the residue at position 1 was of either minor or critical importance depending on the presence or not of a strong auxiliary anchor at position 3. The indirect contribution of residue side chains at positions 4 and 7 reflected the importance of dynamic components in the binding process. Based on these results, we designed a series of high-affinity, H-2Db selective peptides derived from the sequence X1 AIX4NAEAL, where X1 = Y or K and X4 = E or K. After radioiodination or fluorescent (FITC) labelling, these peptides bound strongly and specifically to the surface of viable H-2Db-expressing cells. Rationally designed synthetic peptides, either alone or in a stable complex with MHC, might be of value for controlling CTL activity.