Three‐Dimensional Quantitative Structure Activity Relationship for Cyp2d6 Substrates

Abstract

Without a crystallized structure for a human cytochrome P450, the use of computational molecular modeling is one approach to examine the active site requirements of substrate and inhibitor specificity. CYP2D6 is undoubtedly the most studied polymorphic human CYP and it is therefore desirable for drug companies to limit its role in the metabolism of drug-candidate molecules due to the need for therapeutic drug monitoring. The purpose of this study was to use a three-dimensional quantitative structure activity relationship (3D-QSAR) approach for understanding the CYP2D6 substrate requirements. Using literature Km values (n=24) derived solely from recombinant sources we were able to build and test one such pharmacophore. This was able to significantly rank-order using the (Spearman's rho coefficient 0.55, p=0.0022) predicted against observed literature Km values (n=28) also derived from recombinant sources. The pharmacophore generated in this study was then fitted into the homology model of the human CYP2D6 based on an alignment of bacterial CYPs and the mammalian CYP2C5 to further validate these modeling approaches. Such models as these represent important tools for quantitative prediction of the level of interaction between a molecule and CYP2D6.