Revealing age-related changes in the intraocular microenvironment and senescence modulators using aqueous humor proteomics and machine learning.

BACKGROUND: In conjunction with age, aqueous humor (AH) proteomics can affect the occurrence and development of age-related eye diseases, which are poorly understood. OBJECTIVE: We characterized the proteomic changes in AH throughout the aging process to better understand the aging mechanisms of the intraocular environment. METHODS: We analyzed the AH proteomes of 33 older and 19 younger individuals using liquid chromatography-tandem mass spectrometry, from which we clustered similar expression trajectories of AH proteomics using local regression analysis. Aging proteins (APs) and their functional enrichment were evaluated using various statistical and bioinformatics methods, while aging modulators were predicted using multiple machine-learning models. RESULTS: AH proteomic expression patterns exhibited various types of linear and nonlinear changes across the age groups. A set of 179 proteins identified as significant APs were enriched in various eye processes, such as detoxification, eye development, negative regulation of hydrolase activity, and humoral immune response. According to AH proteomics, hallmarks of aging include oxidative damage, defective extracellular matrices, and loss of proteostasis. A total of 11 APs were considered senescence signatures for predicting AH age with strong predictive ability. Furthermore, 22 APs were classified as modulators that may affect the aging process in the eye. CONCLUSION: These findings establish a framework for age-related changes in the AH proteome and provide potential senescence biomarkers and therapeutic targets for age-related eye diseases.