Integrative multi-omics identifies CYP1B1 as a candidate molecular link between toxicant exposure and ferroptosis-related epithelial stress in COPD

Chronic obstructive pulmonary disease (COPD) is characterized by progressive airflow limitation, chronic airway inflammation, and epithelial remodeling. Although oxidative stress and lipid peroxidation are increasingly recognized as important contributors to COPD pathobiology, the epithelial biomarkers and regulatory pathways linking cigarette-smoke exposure to ferroptosis-related injury remain incompletely defined. This study aimed to identify COPD-associated biomarkers and pathways by integrating bulk transcriptomics, single-cell RNA sequencing, machine-learning analysis, and experimental validation, with a particular focus on the potential association between CYP1B1 and ferroptosis-related epithelial stress. Public transcriptomic cohorts, including GSE47460, GSE76925, and GSE37768, were analyzed using differential expression analysis, weighted gene co-expression network analysis, functional enrichment analysis, and machine-learning-based feature prioritization. Single-cell RNA-seq datasets, including GSE196341 and GSE135893, were used to explore cell-type localization of candidate genes. A cigarette-smoke-exposed mouse model and cigarette-smoke extract-treated 16HBE cells were used for experimental validation, and small-scale proteomic data were incorporated as orthogonal supporting evidence. Molecular docking was performed as an exploratory analysis to predict putative interactions between CYP1B1 and candidate compounds. Twenty-four candidate genes were identified by intersecting COPD-associated co-expression modules with differentially expressed genes. Machine-learning analysis prioritized BHLHE22, DPP6, DHRS9, and CYP1B1 as candidate diagnostic biomarkers. The combined model showed high discrimination in the training cohort, with reduced but retained performance in an external validation cohort. Single-gene ROC analysis in an independent dataset suggested moderate discriminatory performance, with CYP1B1 showing relatively consistent performance across analyses. Single-cell analysis indicated that CYP1B1 expression was enriched in airway secretory cell-related populations, and CYP1B1-high airway secretory cells were associated with ferroptosis-related pathway signatures. In cigarette-smoke-exposed mice and CSE-treated 16HBE cells, CYP1B1 expression was increased, accompanied by inflammatory and epithelial injury-related changes. Knockdown of CYP1B1 attenuated CSE-induced alterations in ferroptosis-related markers, including GPX4, MDA, 4-HNE, and GSH/GSSG. Exploratory molecular docking identified several compounds with predicted CYP1B1-binding potential, although experimental validation is required. Integrative multi-omics and experimental analyses identified CYP1B1 as a candidate COPD-associated epithelial stress biomarker linked to ferroptosis-related signatures in airway secretory cell populations. The findings suggest a potential association between cigarette-smoke-induced CYP1B1 upregulation, lipid peroxidation-related epithelial injury, and epithelial remodeling in COPD. However, CYP1B1 should be considered an exploratory biomarker and putative mechanistic node rather than a validated diagnostic tool or therapeutic target at this stage. Further studies using larger clinically annotated cohorts, smoking-adjusted analyses, cell-type-specific perturbation, ferroptosis rescue experiments, and pharmacological validation are needed.