A PREDICTIVE METHOD FOR THE EVALUATION OF PEPTIDE BINDING IN POCKET-1OF HLA-DRB1 VIA GLOBAL MINIMIZATION OF ENERGY INTERACTIONS

Human leukocyte antigens (HLA) or histocompatibility molecules are gly coproteins that play a pivotal role in the development of an effective immune response, An important function of the HLA molecules is the ab ility to bind and present antigen peptides to T lymphocytes, Presently there is no comprehensive way of predicting and energetically evaluat ing peptide binding on HLA molecules, To quantitatively determine the binding specificity of a class II KLA molecule interacting with peptid es, a novel decomposition approach based on deterministic global optim ization is proposed that takes advantage of the topography of HLA bind ing grove, and examined the interactions of the bound peptide with the five different pockets. In particular, the main focus of this paper i s the study of pocket 1 of HLADR1 (DRB10101 allele), The determinatio n of the minimum energy conformation is based on the ECEPP/3 potential energy model that describes the energetics of the atomic interactions , The minimization of the total potential energy is formulated on the set of peptide dihedral angles, Ruler angles, and translation variable s to describe the relative position. The deterministic global optimiza tion algorithm, alpha BB, which has been shown to be epsilon-convergen t to the global minimum potential energy through the solution of a ser ies of nonlinear convex optimization problems, is utilized. The PACK c onformational energy model that utilizes the ECEPP/3 model but also al lows the consideration of protein chain interactions is interfaced wit h alpha BB. MSEED, a program used to calculate the solvation contribut ion via the area accessible to the solvent, is also interfaced with al pha BB, Results are presented for the entire array of naturally occurr ing amino acids binding to pocket 1 of the HLA DR1 molecule and very g ood agreement with experimental binding assays is obtained. (C) l997 W iley-Liss, Inc.