Computational determination of side chain specificity for pockets in class I MHC molecules

Area of Science:

• Computational immunology
• Structural biology
• Bioinformatics

Background:

• Major histocompatibility complex (MHC) class I molecules present antigenic peptides to T cells, a critical process in adaptive immunity.
• Understanding the specificity of peptide binding to MHC molecules is crucial for designing vaccines and immunotherapies.
• Experimental methods for determining MHC binding motifs are often laborious and time-consuming.

Purpose of the Study:

• To develop and validate a rapid computational procedure for predicting antigenic peptide side chain specificity in MHC class I binding.
• To establish a predictive model for understanding the interactions between peptide side chains and the MHC molecule's binding pocket.

Main Methods:

• A computational procedure combining conformational search and selection was employed.
• The method identifies joint conformations of peptide and MHC pocket side chains.
• Target functions based on solvation and modified electrostatic energies guide conformational selection.

Main Results:

• The computational method successfully predicted the anchor residue specificities for five diverse MHC class I molecules (HLA-A*0201, HLA-B*2705, HLA-A*6801, HLA-B*4001, H-2K(d)).
• Predictions showed high accuracy, even for MHC molecules with undetermined experimental structures (HLA-B40 and H-2K(d)).
• The procedure correctly identified specific anchor residue types (hydrophobic, basic, acidic, bulky) at peptide position 2 for each tested MHC molecule.

Conclusions:

• The developed computational procedure provides a reliable and rapid basis for predicting peptide binding specificity to MHC class I molecules.
• This approach offers a computationally efficient alternative to experimental methods for determining MHC binding motifs.
• The findings open possibilities for routine computational determination of MHC binding motifs, accelerating immunological research and drug development.