A New Structure-Based QSAR Method Affords both Descriptive and Predictive Models for Phosphodiesterase-4 Inhibitors

Abstract

We describe the application of a new QSAR (quantitative structure-activity relationship) formalism to the analysis and modeling of PDE-4 inhibitors. This new method takes advantage of the X-ray structural information of the PDE-4 enzyme to characterize the small molecule inhibitors. It calculates molecular descriptors based on the matching of their pharmacophore feature pairs with those (the reference) of the target binding pocket. Since the reference is derived from the X-ray crystal structures of the target under study, these descriptors are target-specific and easy to interpret. We have analyzed 35 indole derivative-based PDE-4 inhibitors where Partial Least Square (PLS) analysis has been employed to obtain the predictive models. Compared to traditional QSAR methods such as CoMFA and CoMSIA, our models are more robust and predictive measured by statistics for both the training and test sets of molecules. Our method can also identify critical pharmacophore features that are responsible for the inhibitory potency of the small molecules. Thus, this structure-based QSAR method affords both descriptive and predictive models for phosphodiesterase-4 inhibitors. The success of this study has also laid a solid foundation for systematic QSAR modeling of the PDE family of enzymes, which will ultimately contribute to chemical genomics research and drug discovery targeting the PDE enzymes.

Receptor-dependent QSAR methods

“…Alignment-free pharmacophore key QSAR method The alignment-free pharmacophore key QSAR method (also named SBPPK QSAR) has been published by Dong et al [20]. The descriptors are calculated based on the matching of ligand feature pairs with those of the binding pocket.…”

Structure-Based Quantitative Structure–Activity Relationship Modeling of Estrogen Receptor β-Ligands

“…The overall design and the algorithm of SB-PPK �SAR process have been reported in our earlier publication [46]. Unlike most traditional �SAR methods, where the molecular descriptor generation takes into account only the structural information of the ligands, the SB-PPK method uses the x-ray structural data of the target binding site.…”

“…To complement and further improve these ideas, our group has developed several novel techniques that have been published recently. They covered both alignment-free [46,47] and alignment-dependent [48] structure-based �SAR methods.…”

“…Here, we employ the alignment-free struc-structure-based pharmacophore keys (SB-PPK) method [46] to develop predictive �SAR models for ERb ligands. This method has been ini-This method has been initially developed to construct predictive �SAR models for a set of PDE4 inhibitors [46], which are also being investigated as potential drugs for neurodegenerative diseases.…”

“…This method has been ini-This method has been initially developed to construct predictive �SAR models for a set of PDE4 inhibitors [46], which are also being investigated as potential drugs for neurodegenerative diseases. The SB-PPK methodology, however, is directly transferable to the modeling of ERb ligands.…”