Elaborate ligand-based modeling reveal new migration inhibitory factor inhibitors.

ABSTRACT

Recent research suggested the involvement of migration inhibitor factor (MIF) in cancer and inflammatory diseases, which prompted several attempts to develop new MIF inhibitors. Accordingly, we investigated the pharmacophoric space of 79 MIF inhibitors using seven diverse subsets of inhibitors to identify plausible binding hypotheses (pharmacophores). Subsequently, we implemented genetic algorithm and multiple linear regression analysis to select optimal combination of pharmacophores and physicochemical descriptors capable of explaining bioactivity variation within the training compounds (QSAR model, r63=0.62, F=42.8, rLOO(2)=0.721,rPRESS(2) against 16 external test inhibitors=0.58). Two orthogonal pharmacophores appeared in the optimal QSAR model suggestive of at least two binding modes available to ligands inside MIF binding pocket. Subsequent validation using receiver operating characteristic (ROC) curves analysis established the validity of these two pharmacophores. We employed these pharmacophoric models and associated QSAR equation to screen the National Cancer Institute (NCI) list of compounds. Eight compounds gave >50% inhibition at 100µM. Two molecules illustrated >75% inhibition at 10µM.