Novel tumor necrosis factor-α (TNF-α) inhibitors from small molecule library screening for their therapeutic activity profiles against rheumatoid arthritis using target-driven approaches and binary QSAR models.

Tumor necrosis factor alpha (TNF-α) is a multifunctional cytokine that acts as a central biological mediator for critical immune functions, including inflammation, infection, and antitumor responses. It plays pivotal role in autoimmune diseases like rheumatoid arthritis (RA). The synthetic antibodies etanercept, infliximab, and adalimumab are approved drugs for the treatment of inflammatory diseases bind to TNF-α directly, preventing its association with the tumor necrosis factor receptor (TNFR). These biologics causes serious side effects such as triggering an autoimmune anti-antibody response or the weakening of the body's immune defenses. Therefore, alternative small-molecule based therapies for TNF-α inhibition is a hot topic both in academia and industry. Most of small-molecule inhibitors reported in the literature target TNF-α, indirectly. In this study, combined in silico approaches have been applied to better understand the important direct interactions between TNF-α and small inhibitors. Our effort executed with the extensive literature review to select the compounds that inhibit TNF-α. High-throughput structure-based and ligand-based virtual screening methods are applied to identify TNF-α inhibitors from 3 different small molecule databases (∼256.000 molecules from Otava drug-like green chemical collection, ∼ 500.000 molecules from Otava Tangible database, ∼2.500.000 Enamine small molecule database) and ∼240.000 molecules from ZINC natural products libraries. Moreover, therapeutic activity prediction, as well as pharmacokinetic and toxicity profiles are also investigated using MetaCore/MetaDrug platform which is based on a manually curated database of molecular interactions, molecular pathways, gene-disease associations, chemical metabolism and toxicity information, uses binary QSAR models. Particular therapeutic activity and toxic effect predictions are based on the ChemTree ability to correlate structural descriptors to that property using recursive partitioning algorithm. Molecular Dynamics (MD) simulations were also performed for selected hits to investigate their detailed structural and dynamical analysis beyond docking studies. As a result, at least one hit from each database were identified as novel TNF-α inhibitors after comprehensive virtual screening, multiple docking, e-Pharmacophore modeling (structure-based pharmacophore modeling), MD simulations, and MetaCore/MetaDrug analysis. Identified hits show predicted promising anti-arthritic activity and no toxicity. Communicated by Ramaswamy H. Sarma.