3D QSAR analyses of novel tyrosine kinase inhibitors based on pharmacophore alignment

In an effort to develop a quantitative ligand-binding model for the recepto r tyrosine kinases, a pharmacophore search was first used to identify struc tural features that are common in two novel sets of 12 molecules of the 3-s ubstituted indolin-2-ones and 19 compounds of the benzylidene malononitrile s with low-to-high affinity for HER2, a kind of receptor tyrosine kinase. T he common pharmacophore model based on these 31 compounds was used as a tem plate to obtain the aligned molecular aggregate, which provided a good star ting point for 3D-QSAR analysis of only the 19 benzylidene malononitriles. Two molecular field analysis (MFA) techniques, including CoMFA and CoMSIA, were used to derive the quantitative structure-activity relationships of th e studied molecules. From the studied results, it was obvious that the 3D-Q SAR models based on the pharmacophore alignment were superior to those base d on the simple atom-by-atom fits. Considering the flexibility of the studi ed molecules and the difference between the active conformers and the energ y-lowest conformers, the pharmacophore model can usually provide the common features for the flexible regions. Moreover, the best CoMSIA model based o n the pharmacophore hypothesis gave good statistical measure from partial l east-squares analysis (PLS) (q(2) = 0.71), which was slightly better than t he CoMFA one. Our study demonstrated that pharmacophore modeling and CoMSIA research could be effectively combined. Results obtained from both methods helped with understanding the specific activity of some compounds and desi gning new specific HER2 inhibitors.