Comprehensive bioinformatics and in vitro studies reveal the carcinogenic role and molecular basis of endocrine disruptors in prostate cancer.

Background: In recent years, growing attention has been paid to the carcinogenicity of endocrine disruptors (EDs). However, their relationship with prostate cancer (PCa) remains unclear. This study investigates the association between EDs and PCa to identify key genes that may bridge this relationship. Methods: The ADME properties and carcinogenicity of the selected endocrine-disrupting chemicals EDs were predicted using the ADMETlab 3.0 and ProTox 3.0 platforms, respectively. Potential target genes related to EDs and PCa were obtained by integrating multiple public databases. A protein-protein interaction (PPI) network of the overlapping genes was constructed and visualized, followed by GO and KEGG enrichment analyses to explore their potential biological mechanisms. From 101 machine learning algorithm combinations, the most relevant key genes for PCa progression were screened. Molecular docking analysis was used to evaluate the binding properties between ED compounds and key targets. Pan-cancer analysis was employed to examine the general role of key genes across multiple cancer types. The Comparative Toxicogenomics Database (CTD) was used to identify natural active products potentially targeting the core genes. Finally, in vitro cell experiments were conducted to validate the effects of EDs on PCa cells and the intervention effects of related natural products. Results: Initially, predictions from the ADMETlab 3.0 and ProTox 3.0 platforms indicated significant in vivo accumulation, endocrine-disrupting effects, and carcinogenicity for the 12 common EDs. Subsequently, the integration of multiple databases identified 233 overlapping targets associated with PCa. GO and KEGG enrichment analyses revealed that these targets are primarily involved in regulating cell proliferation, inflammatory responses, and cancer cell metabolism. Among the evaluated machine learning algorithms, the CoxBoost + SuperPC hybrid model demonstrated superior predictive performance and robustness. Subsequent analysis pinpointed three key regulatory genes: CD38, MMP11, and PLK1. Molecular docking simulations confirmed potential interactions between EDs compounds and the core target, PLK1. Furthermore, five natural active products were identified as potential agents to mitigate the adverse effects induced by EDs exposure. Finally, in vitro cell experiments demonstrated that Benzo[a]pyrene promotes PLK1 expression and PCa progression, whereas Cryptotanshinone effectively counteracts these effects. Conclusion: This multidisciplinary study unveils PLK1 as a pivotal molecular target through which EDs drive PCa progression. Furthermore, we identify five natural compounds, notably Cryptotanshinone, that counteract the carcinogenic effects of EDs by targeting PLK1. These findings provide crucial molecular insights into ED-induced carcinogenesis and reveal promising targets for the prevention and intervention of PCa.