Proposing novel TNFα direct inhibitor Scaffolds using fragment-docking based e-pharmacophore modeling and binary QSAR-based virtual screening protocols pipeline

Abstract

Tumor necrosis factor alpha (TNFα) is a homotrimer protein that plays a pivotal role for critical immune functions, including infection, inflammation and antitumor responses. It also plays a primary role in autoimmune diseases like rheumatoid arthritis (RA). So far, only biological therapeutics like infliximab, etanercept, and adalimumab are available as treatment of inflammatory diseases. They directly bind to TNFα and interrupt its binding to its receptor protein tumor necrosis factor receptor (TNFR). However, they may also cause serious side effects such as activating an autoimmune anti-antibody response or the weakening of the body's immune defenses. Thus, small molecule-based therapies can be considered as alternative methods. In this study, a novel method is applied to develop energetically optimized, structure-based pharmacophore models for rapid in silico drug screening. Fragment-based docking results were used in the construction of an universal e-pharmacophore model development. The developed model is then used for screening of small-molecule library Specs-screening compounds (Specs-SC) which includes more than 200.000 drug-like molecules. In another approach, binary QSAR-based models were used to screen Specs-SC, as well as Specs-natural products (NP) which has around 750 compounds, and a library of drugs registered or approved for use in humans NIH's NCGC pharmaceutical collection (NPC) which has around 7500 molecules. The MetaCore/MetaDrug platform was used for binary QSAR models for therapeutic activity prediction as well as pharmacokinetic and toxicity profile predictions of screening molecules. This platform is constructed based on a manually curated database of molecular interactions, molecular pathways, gene-disease associations, chemical metabolism, and toxicity information. Molecular docking and molecular dynamics (MD) simulations were performed for the selected hit molecules. As target protein, both homodimer and homotrimer forms of TNFα were considered. The screening results showed that indinavir and medroxalol from NPC chemical library and a set of compounds (AT-057/43115940, AP-970/42897107, AK-968/41925665, AI-204/31679053, AN-648/41666950, AN-698/42006940) from Specs-SC database were identified as safe and active direct inhibitors of TNFα.

Keywords: Blood brain barrier; MetaCore/MetaDrug analysis; Molecular docking; Molecular dynamics (MD) simulation; Rheumatoid arthritis (RA); TNFα; Toxicity prediction; e-pharmacophore modeling.