Multivariate Free‐Wilson Analysis of Some N‐Alkylmorphinan‐6‐one Opioids Using PLS

Abstract

MethodsThe PLS method is used to derive an improved, unified and statistically significant Free;Wilson QSAR model for the analgesic potency of some N-alkylmorphinan-6-one opioids by the inclusion of a large number of first order cross-terms between the substituents.To obtain high potency the R I / R~ substituent-pair should be a cyclo-butyl and a hydrogen (fenol), respectively, while the R3 substituent can be either an oxygen or a methylene bridge.The R4-and R = , substituents can both be a hydrogen or a methyl group to retain high potency of the opioids. Contrary to this, the R I / R~ pairs cyclo-propylhydrogen and cyclo-butyllmethyl as well as a n-propyl substituent in the R5 position are quite unfavorable in this respect. Free-Wilson Description of the StructuresThe FW descriptor matrix consisted of thirteen columns for the R1-R5 substituents (see scheme). An additional sixtysix columns representing all possible first order cross-terms between substituents were also used (i.e. var # 14 = 1*4,15 = 1*5 . . . 24 = 2*4 . . . 79 = 9*13). CHZ-R,I Abbreviations: PLS-Partial least squares projections to latent 0 structures; CoMFA-Comparative molecular field analysis; QSAR-Quantitative structure-activity relationship; MLR-Multiple linear regression; FB-Fujita-Ban modification; FW-Free-Wilson