Database mining of ZINC15 natural compounds reveals potential thyroid receptor β agonists for NAFLD management: an in silico study

Abstract

Non-alcoholic fatty liver disease (NAFLD) is a prevalent metabolic disorder with limited therapeutic options. Thyroid receptor β (THR-β) agonists have been showing promise for controlling NAFLD via improving hepatic lipid metabolism. This study utilized different in silico tools to screen 47 199 natural compounds from the ZINC15 database to identify potential THR-β agonists. Molecular docking, molecular dynamics simulations, and advanced analyses such as PCA, TICA-FES, and MSM revealed that 4-O-caffeoylquinic acid (compounds 2) and dihydroxydehydrodiconiferyl alcohol (compound 18) are the most promising hits. Both demonstrated high binding affinity and stable agonist interactions with key THR-β residues such as Arg316 and Arg320 that stabilize the ligand binding pocket and support the agonist potential, comparable to the reference agonists resmetirom and {3,5-dichloro-4-[4-hydroxy-3-(propan-2-yl)phenoxy]phenyl}acetic acid. Long-term MD simulations confirmed their stability, and MM/GBSA calculations supported robust thermodynamic profiles. Moreover, the two hits displayed superior selectivity for THR-β over THR-α and favorable pharmacokinetic profiles with minimal toxicity alerts. These findings support compounds 2 and 18 as strong candidates for NAFLD therapy, warranting further experimental validation.