Receptor site topographies for phencyclidine-like and sigma drugs: predictions from quantitative conformational, electrostatic potential, and radioreceptor analyses.

Abstract

Computer-assisted molecular modelling techniques and electrostatic analyses of a wide range of phenycyclidine (PCP) and sigma ligands, in conjunction with radioreceptor studies, were used to determine the topographies of the PCP and sigma receptors. The PCP receptor model was defined using key molecules from the arylcyclohexylamine, benzomorphan, bridged benz[f]isoquinoline, and dibenzocycloalkenimine drug classes. Hypothetical receptor points (R1, R2) were constructed onto the aromatic ring of each compound to represent hydrophobic interactions with the receptor, along with an additional receptor point (R3) representing a hydrogen bond between the nitrogen atom and the receptor. The superimposition of these key molecules gave the coordinates of the receptor points and nitrogen defining the primary PCP pharmacophore as follows: R1 (0.00, 3.50, 0.00), R2 (0.00, -3.50, 0.00), R3 (6.66, -1.13, 0.00), and N (3.90, -1.46, -0.32). Additional analyses were used to describe secondary binding sites for an additional hydrogen bonding site and two lipophilic clefts. Similarly, the sigma receptor model was constructed from ligands of the benzomorphan, octahydrobenzo[f]quinoline, phenylpiperidine, and diphenylguanidine drug classes. Coordinates for the primary sigma pharmacophore are as follows: R1 (0.00, 3.50, 0.00), R2 (0.00, -3.50, 0.00), R3 (6.09, 2.09, 0.00), and N (4.9, -0.12, -1.25). Secondary binding sites for sigma ligands were proposed for the interaction of aromatic ring substituents and large N-substituted lipophilic groups with the receptor. The sigma receptor model differs from the PCP model in the position of nitrogen atom, direction of the nitrogen lone pair vector, and secondary sigma binding sites. This study has thus demonstrated that the differing quantitative structure-activity relationships of PCP and sigma ligands allow the definition of discrete receptors. These models may be used in conjunction with rational drug design techniques to design novel PCP and sigma ligands of high selectivity and potency.