Unraveling the molecular mechanisms of air pollution-induced lung adenocarcinoma through machine learning, network toxicology, and molecular docking.

OBJECTIVE

Airborne environmental contaminants are established carcinogens. This investigation elucidates the mechanistic contributions to pulmonary adenocarcinoma (LUAD) pathogenesis.

METHODS

Transcriptomic profiling through differential expression analysis and weighted gene co-expression network analysis (WGCNA) identified malignancy-associated molecular targets. Computational screening of environmental toxicology databases yielded airborne contaminant-responsive genes that were subsequently cross-referenced with LUAD-associated targets. An integrative analytical framework incorporating artificial intelligence-driven pattern recognition, systems toxicology, and computational structural biology characterizes contaminant-macromolecule interactions. Expression-based patient stratification enabled pathway enrichment investigation of CA4-related oncogenic processes.

RESULTS

Systematic evaluation of eight atmospheric pollutants (benzene, sulfur dioxide, nitric oxide, carbon monoxide, nitrogen dioxide, toluene, ozone, and benzo[a]pyrene) identified 265 molecular interactors, including 58 putative LUAD mediators. Predictive modeling established five central regulators exhibiting significant alterations in expression (CA4 and CAT suppression; SPP1, CDH1, and TIMP1 elevation; all P < 0.05). In silico structural analysis demonstrated high-affinity binding between the prototypical polycyclic aromatic hydrocarbon benzo[a]pyrene (BaP) and carbonic anhydrase IV protein.

CONCLUSIONS

Air pollution may promote LUAD via specific gene regulation. The identified hub genes and validated CA4-Benzo[a]pyrene interactions provided mechanistic insights for further research.