Study of Benzodiazepines Receptor Sites Using a Combined QSAR‐CoMFA Approach

Abstract

A quantitative structure-activity relationship study on a set of anxiolytic benzodiazepines was carried out using the Hansch approach (QSAR) and the Comparative Molecular Field Analysis (CoMFA).The aim of the study was to investigate the nature of some of the main factors influencing the binding of the benzodiazepines to the CNS receptors. Since our preliminary QSAR results provided evidence of an involvement of a specific ligand-receptor hydrophobic interaction, the methodological problem of how to properly include the scalar descriptor x in the development of the CoMFA model was extensively examined. The following explanatory variables were considered in the regression analyses: the HOMO and LUMO energies, the total dipole moment, the substituent constants x, . ' 7 , BI , B5, L, MR, the steric and the electrostatic CoMFA fields. The QSAR and the CoMFA were performed with Multilinear Regression Analysis and Partial Least Squares (PLS) Analysis, respectively. In each of the two methods the cross-validation procedure was systematically applied to evaluate the significance of the correlation and to establish the most appropriate parametric dimensionality of the calibration models. A deeper insight into the meaning of the CoMFA electrostatic field as explanatory variable of the binding affinity was obtained by correlating the HOMO and LUMO energies with the electrostatic field itself. The results of this study suggest the existence of a specific interaction between the substituents which are in the "C7-position", formally corresponding to the 7-chlorine atom in the structure of diazepam, and a receptor hydrophobic pocket. These substituents can additionally promote, through an electron-withdrawing effect, a charge-transfer interaction between the ligand and an electronegative receptor site.

Comparative molecular field analysis of non-steroidal aromatase inhibitors related to fadrozole

“…The possible contribution to the 3D QSAR of log P was explored by calculating the CLOGP values (not reported) of the compounds and inserting this new column in the CoMFA table. No reduction of the residual variance was found after considering the CLOGP column, even when the column was scaled 10, 100 or 1000 times as suggested by Greco et al [16], in order to appropriately weight the hydrophobic parameter in the CoMFA table.…”

Application of neural networks in structure-activity relationships

“…Unlike CoMFA described, 2 in which a hypothetical fixed spatial arrangement of the phenyl ring was assumed for all ligands, we always used the minimum energy conformers for the neural network calculations. In the cited article, 2 for activity values there were given no experimental errors. In this study, there were included more compounds with biological data, as in the case of antiarrhythmics.…”

“…This parameter has now been recognized to be possibly related to proarrhythmic properties of Class 1 antiarrhythmics. 1 In the example of anxiolytic benzodiazepines, 2 the performances of neural network, CoMFA, and classical QSAR methods are examined in predicting binding data to benzodiazepine receptor.…”

Classification and Regression Tree Analysis for Molecular Descriptor Selection and Binding Affinities Prediction of Imidazobenzodiazepines in Quantitative Structure-Activity Relationship Studies

“…1 These drugs have been subjected to extensive QSAR studies. [2][3][4][5][6][7][8] GABAA receptors are the major inhibitory neurotransmitter receptors in the brain, in the site of action of many clinically important drugs, and are important drug targets representing the sites of action of benzodiazepines, barbiturates, and neurosteroids. These receptors are ligand-gated chloride channels composed of five subunits that can belong to eight different subunit classes.…”