Structure-based cheminformatics and molecular dynamics profiling of potential SIRT6 inhibitors

Sirtuin-6 (SIRT6) is a NAD+-dependent deacetylase that maintains genome stability, metabolic regulation, and cellular stress responses, making it an attractive target for therapeutic intervention in metabolic and age-related diseases. Despite its biological importance, the identification of potent SIRT6 modulators remains limited. In this study, we applied an integrative computational approach combining cheminformatics, network pharmacology, molecular docking, and molecular dynamics simulations to explore new inhibitory candidates targeting SIRT6. A curated dataset of 78 CHEMBL compounds was used to develop robust multi-fingerprint QSAR models using Random Forest algorithms, validated through Y-randomization, external testing, and applicability domain analysis. Network pharmacology analysis revealed functional associations between SIRT6 and key regulatory proteins such as NAMPT, CD38, and HIF1A, highlighting its involvement in NAD⁺ biosynthesis and cellular stress pathways. Molecular docking identified CHEMBL50 (Quercetin) and CHEMBL4217987 as top candidates with favorable interactions at the SIRT6 catalytic site. These complexes were further evaluated through 200 ns MD simulations. Binding stability was confirmed using MM-GBSA free energy calculations, dynamic cross-correlation matrix (DCCM), and principal component analysis (PCA), demonstrating energetically favorable and stable protein-ligand interactions. Overall, this study offers a predictive and mechanistic framework for SIRT6 inhibitor discovery and provides lead scaffolds for further optimization and experimental validation.