Complementarity of δ opioid ligand pharmacophore and receptor models

Abstract

The elaboration of a pharmacophore model for the δ opioid receptor selective ligand JOM-13 (Tyr-c[D-Cys-Phe-D-Pen]OH) and the parallel, independent development of a structural model of the δ receptor are summarized. Although the backbone conformation of JOM-13's tripeptide cycle is well defined, considerable conformational lability is evident in the Tyr' residue and in the Phe3 Side Chain, key pharmacophore elements of the ligand. Replacement of these flexible features of the ligand by more conformationally restricted analogues and subsequent correlation of receptor binding and conformational properties allowed the number of possible binding conformations of JOM-13 to be reduced to two. Of these, one was chosen as more likely, based on its better superposition with other conformationally constrained δ receptor ligands. Our model of the δ opioid receptor, constructed using a general approach that we have developed for all rhodopsin-like G protein-coupled receptors, contains a large cavity within the transmembrane domain that displays excellent complementarity in both shape and polarity to JOM-13 and other δ ligands. This binding pocket, however, cannot accommodate the conformer of JOM-13 preferred from analysis of ligands, alone. Rather, only the 'alternate' allowed conformer, identified from analysis of the ligands but 'disfavored' because it does not permit simultaneous superposition of all pharmacophore elements of JOM-13 with other δ ligands, fits the binding site. These results argue against a simple view of a single, common fit to a receptor binding site and suggest, instead, that at least some binding site interactions of different ligands may differ. (C) 2000 John Wiley and Sons, Inc.