Structural characterization of plasmodial aminopeptidase: a combined molecular docking and QSAR-based in silico approaches

Aminopeptidase M1 (PfAM1) is one of the key enzymes involved in the development of new antimalarials. To accelerate the discovery of inhibitors with selective activity against PfAM1 and microsomal neutral aminopeptidase (pAPN), in the present work, the optimum comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) models were built based on PfAM1 and pAPN inhibitors. The results of the developed 3D-QSAR models were as follows: PfAM1/CoMFA: R-cv(2) = 0.740, R-pred(2) = 0.7781; PfAM1/CoMSIA: R-cv(2) = 0.740, R-pred(2) = 0.7354; pAPN/CoMFA: R-cv(2) = 0.612, R-pred(2) = 0.7318; pAPN/CoMSIA: R-cv(2) = 0.609, R-pred(2) = 0.7480, and the models derived from MLR, PLSR and SVR methods provided high R-2 values of 0.6960, 0.6965, 0.7971 for PfAM1, 0.7700, 0.7697, 0.8228 for pAPN and Q(2) of 0.7004, 0.7004, 0.5632 for PfAM1, 0.7551, 0.7566 and 0.8394 for pAPN, respectively, indicating that the developed 3D-QSAR and 2D-QSAR models possess good ability for prediction of the relative compound activities. Furthermore, all inhibitors were docked into the active site of the PfAM1 and pAPN receptors, the hydrogen-bond interactions between the compound 33 with Glu497, Glu463 and Arg489 of the PfAM1, and the compound 4 with Ala348, Glu384 and Phe467 of the receptor pAPN are able to help to stabilize the conformation. The above results would provide helpful clues to predicting the binding activity of novel inhibitors and the foundation for understanding the interaction mechanism between the inhibitors and the receptors. [GRAPHICS] .