Two-dimensional-quantitative structure-activity relationship studies of a novel series of diaryl furanone derivatives: An approach to design selective and effective cyclooxygenase-2 inhibitors

Abstract

INTRODUCTION: Over a decade, a large number of selective cyclooxygenase-2 (COX-2) inhibitors with diverse chemical characteristics have been designed but only a few have emerged as drugs. MATERIALS AND METHODS: In the present study, two-dimensional-quantitative structure-activity relationship (2D-QSAR) studies were performed on a set of 43 novel derivatives of diaryl furanone using V-Life Molecular Design Suite (MDS 3.5) QSAR plus module using multiple linear regression (MLR) and Partial Least Square (PLS) regression methods against a COX-2 enzyme. RESULTS AND DISCUSSION: PLS method has displayed a very significant prediction results. QSAR model was generated by a training set of 33 molecules with correlation coefficient (r2) of 0.7695, cross-validated correlation coefficient (q2) of 0.5359 and F test of 23.3734. The estate contribution, chi, hydrophobic, and alignment-independent descriptors were major contributors. CONCLUSION: 2D-QSAR model result showed the positive contribution of oxygen atoms and negative contribution of rotatable bonds and distance between double bonds toward biological activity.

Keywords

• Cyclooxygenase
• diaryl furanone
• multiple linear regression
• partial least squares regression
• quantitative structure-activity relationship