Analysis of potential molecular targets and mechanisms of brominated flame retardants in causing osteoarthritis using network toxicology, machine learning, SHAP analysis, and molecular dynamics simulation

Abstract

Abstract Background The commonly used brominated flame retardant (2,2’,4,4’-Tetrabromodiphenyl Ether, BDE-47) is a persistent organic pollutant that is widely distributed in the environment and is associated with adverse health effects, including an increased risk of osteoarthritis. However, the specific molecular mechanisms by which BDE-47 contributes to the development of osteoarthritis (OA) remain largely unclear. This study aimed to investigate the potential relationship between BDE-47 exposure and osteoarthritis pathogenesis. Methods To achieve this, we employed an integrative approach combining network toxicology, machine learning, SHAP analysis, molecular docking, and molecular dynamics simulations. Results Initially, potential target genes associated with BDE-47 and OA were retrieved from multiple public databases, including PharmMapper, ChEMBL, and GEO. Subsequently, a comprehensive machine learning workflow involving 113 different algorithms was used to identify 10 core target genes as potential mediators of BDE-47-induced osteoarthritis. Finally, SHAP analysis identified FKBP5 as the most critical toxic target. Molecular docking and molecular dynamics (MD) simulations were then performed to evaluate the binding interactions and stability between BDE-47 and FKBP5. Conclusion The research results suggest that BDE-47 may be involved in the pathogenesis of osteoarthritis by targeting and regulating specific toxic targets, with FKBP5 being particularly prominent as the most crucial potential therapeutic target. These insights provide a theoretical foundation for developing preventive and therapeutic strategies against BDE-47-induced OA and underscore the importance of raising public awareness regarding the health risks of environmental pollution.

Keywords

• Osteoarthritis
• BDE-47
• Network toxicology
• Machine learning
• Molecular docking