Identification of benzo(a)pyrene-related toxicological targets and their role in chronic obstructive pulmonary disease pathogenesis: a comprehensive bioinformatics and machine learning approach

作者：Jiehua Deng, Lixia Wei, Yongyu Chen, Xiaofeng Li, Hui Zhang, Xuan Wei, Xin Feng, Xue Qiu, Bin Liang, Jianquan Zhang · 发表于：BMC Pharmacology and Toxicology · 年份：2025 · DOI：10.1186/s40360-025-00842-1 · 被引用次数：12 · 研究领域：Chronic Obstructive Pulmonary Disease (COPD) Research、Mangiferin and Mango Extracts、Computational Drug Discovery Methods

BACKGROUND: Chronic obstructive pulmonary disease (COPD) pathogenesis is influenced by environmental factors, including Benzo(a)pyrene (BaP) exposure. This study aims to identify BaP-related toxicological targets and elucidate their roles in COPD development. METHODS: A comprehensive bioinformatics approach was employed, including the retrieval of BaP-related targets from the Comparative Toxicogenomics Database (CTD) and Super-PRED database, identification of differentially expressed genes (DEGs) from the GSE76925 dataset, and protein-protein interaction (PPI) network analysis. Functional enrichment and immune infiltration analyses were conducted using GO, KEGG, and ssGSEA algorithms. Feature genes related to BaP exposure were identified using SVM-RFE, Lasso, and RF machine learning methods. A nomogram was constructed and validated for COPD risk prediction. Molecular docking was performed to evaluate the binding affinity of BaP with proteins encoded by the feature genes. RESULTS: We identified 72 differentially expressed BaP-related toxicological targets in COPD. Functional enrichment analysis highlighted pathways related to oxidative stress and inflammation. Immune infiltration analysis revealed significant increases in B cells, DC, iDC, macrophages, T cells, T helper cells, Tcm, and TFH in COPD patients compared to controls. Correlation analysis showed strong links between oxidative stress, inflammation pathway scores, and the infiltration of immune cells, including aDC, ma...