In silico Identification of New Antiproliferative Compounds, Inhibitors of Matrix Metalloproteinases

Abstract

Tenascin C (TNC) is a glycoprotein of the extracellular matrix and its overexpression induced by matrix metalloproteinases (MMPs) is associated with the amplification of the proliferative response to growth factors, promotion of cancer cell invasion and tumoral growth. This paper looks forward to find better inhibitors of MMPs, based on structure of batimastat (BAT) – a synthetic inhibitor of MMPs and an experimental drug against matrix metalloproteinase-16 (MMP16). To reach this aim, at first it was built a complete virtual library using 12 scaffolds (SCs) from BAT, concatenated via a dummy linker with 40 building blocks (BBs) found in 14 antineoplastic drugs. Second, all the derivatives from the virtual library were filtered, through mass-computation, for “drug-likeness”, undesirable moieties and checked if they are PAINS (Pan Assay Interference Compounds). At last, a virtual screening of “drug-like” virtual derivatives, free of undesirable moieties (except of the thiophene moiety) and BAT against the homologue model of MMP16 was made, using docking software in order to identify the compounds that specifically bind the enzyme. In the same time it was determinate the smallest binding energy required to bind MMP-16 to reveal the best ligands. Results show that 86 virtual derivates (1% of the 6720 derivates from the virtual library) are “drug-like” compounds. However, only 23 of the “drug-like” virtual derivates have better metrics than BAT, are free of undesirable moieties and are not PAINS. In addition, 15 of them are better ligands for MMP16 than BAT.