Computational Identification and Anti-Inflammatory Evaluation of T19093 as a TLR4/MD2 Inhibitor.

Abstract

Background: The TLR4 (Toll-like receptor 4)/MD2 (Myeloid differentiation protein-2) is a crucial target for developing novel anti-inflammatory drugs. Nevertheless, current inhibitors often have significant adverse effects, necessitating the discovery of safer alternatives.

Objective: The investigation aims to identify novel TLR4/MD2 inhibitors with potential antiinflammatory activity using machine learning and virtual screening technology.

Methods: A machine-learning model was created using the MACCS (Molecular ACCess Systems) key fingerprint. Subsequently, virtual screening and molecular docking were used to evaluate candidate compounds' binding free energy to the TLR4/MD2 complex. Furthermore, ADMET (absorption, distribution, metabolism, excretion, and toxicity) prediction was used to assess the druggable properties of compounds. The most promising compound, T19093, was considered for molecular dynamic simulation. Finally, the anti-inflammatory efficacy of T19093 was further validated using LPS-treated THP-1 cells.

Results: T19093, a polyphenolic compound isolated from the Gnaphalium plant genus, showed strong binding to key residues of the TLR4/MD2 complex, with a docking score of -11.29 kcal/mol. Furthermore, ADMET predicted that T19093 has good pharmacokinetic properties and balanced physicochemical properties. Moreover, molecular dynamics simulation confirmed stable binding between T19093 and TLR4/MD2 complex. Finally, it was found that T19093 alleviated LPSinduced inflammatory response by inhibiting the activation of TLR4/MD2 downstream signaling pathways and disrupting the TLR4/MD2 interaction.

Conclusion: T19093 was discovered as a potential novel TLR4/MD2 inhibitor using machine learning and virtual screening techniques and showed potent anti-inflammatory activity, which could provide a new therapeutic alternative for the treatment of inflammation-related diseases.