Homology modeling and virtual screening for inhibitors of lipid kinase PI (4) K from Plasmodium

Authors: Ren JX, Gao NN, Cao XS, Hu QA, Xie Y
Publication: Biomed Pharmacother
Software: ADMET Predictor®

Abstract

Malaria parasite strains have emerged to tolerate the therapeutic effects of the prophylactics and drugs presently available. Recent studies have shown that KAI715 and its analogs inhibit malaria parasites growth by binding to lipid kinase PI(4)K (phosphatidylinositol-4-OH kinase) of the parasites. Therefore, targeting PI(4)K may open up new avenues of target-based drug discovery to identify novel anti-malaria drugs. In this investigation, we describe the discovery of novel potent PfPI(4)K (PI(4)K from P. falciparum) inhibitors by employing a proposed hybrid virtual screening (VS) method, including pharmacophore model, drug-likeness prediction and molecular docking approach. 3D structure of PfPI(4)K has been established by homology modeling. Pharmacophore model HypoA of PfPI(4)K inhibitors has been developed based on the ligand complexed with its corresponding receptor. 174 compounds with good ADMET properties were carefully selected by a hybrid virtual screening method. Finally, the 174 hits were further validated by using a new pharmacophore model HypoB built based on the docking pose of BQR685, and 95 compounds passed the last filter. These compounds would be further evaluated by biological activity assays. The molecular interactions of the top two potential inhibitors with the active site residues are discussed in detail. These identified hits can be further used for designing the more potent inhibitors against PfPI(4)K by scaffold hopping, and deserve consideration for further structure-activity relationship (SAR) studies.