Objective: We aim to provide new insight and scientific evidence for rational design and
discovery of phosphodiesterase-9A (PDE9A) inhibitors with remarkable potency and weak side effect
to treat CNS diseases such as Alzheimer’s disease.
Methods: Comparative molecular field analysis (CoMFA) and comparative molecular similarity
indices analysis (CoMSIA) were performed on a series of PDE9A inhibitors. Moreover, two different
alignment methods, docking-based structural alignment (DCBA) and local lowest energy structure
based alignment (LESBA), were employed to scrutinize their effects on the robustness and predictive
capability of 3D-QSAR models.
Results: The models generated by CoMFA had a cross-validated coefficient (q2) of 0.771 and a
regression coefficient (r2
) of 0.983. The CoMSIA models had a (q2
) of 0.776 and (r2
) of 0.960. The
external predictive capability of the built models was evaluated by using the test set of nine
compounds. From obtained results, the CoMSIA models were found to have highly predictive
capability in comparison with CoMFA models. Contour maps of CoMSIA models provided many
helpful structural insights, including N1-bulkier hydrophobic group such as cyclopentyl group better
filling the metal binding pocket in the PDE9A to show stronger inhibitory activity.
Conclusions: Docking-based 3D-QSAR studies is helpful to improve the design of pyrazolopyrimidinone
derivatives as PDE9A inhibitors to develop new chemical entities with higher