Industrialization and urbanization exacerbate global environmental health challenges, with pollutants from industrial and vehicular emissions (e.g., PAHs, NOx, PM2.5) posing significant risks. Growing evidence links air pollution to age-related macular degeneration (AMD), a leading cause of elderly blindness; however, the mechanisms involving multi-pollutant interactions and AMD subtype-specific pathways remain poorly understood, as analyses of individual pollutants fail to capture real-world exposure complexity. To address these gaps, we investigated the molecular mechanisms linking complex pollution mixtures to AMD using a defined urban air pollutant mixture. We integrated transcriptomic data from a cohort of 940 samples (encompassing normal, dry AMD, and wet AMD states across peripheral blood and retinal tissues) and independently analyzed source- and subtype-specific molecular signatures. Through iterative computational analyses, molecular docking, and machine learning, key targets were identified and predictive models for dry and wet AMD were developed. Transcriptomic profiling revealed distinct genetic markers: HTRA1, PTEN, IER3, ERCC6, and CP were implicated in wet AMD, whereas FBN2, DDR1, and CFHR1 were critical for dry AMD. Subtype-specific machine learning models demonstrated robust validation in independent datasets. We uncovered potential synergistic interactions between pollutants and genetic susceptibility, identifying unique molecular cascades for each AMD subtype. This study provides new insights into environment-gene crosstalk in AMD pathogenesis by integrating pollution exposure with multi-omics insights, suggesting potential subtype-specific mechanisms and identifying targets for risk stratification, early diagnosis, and personalized therapies.
Meilan Chen (Sat,) studied this question.