Association of PM2.5 and its components with cognitive dysfunction among rural older adults: Epidemiological evidence with hypothesis-generating network toxicology analysis.

Background

Although fine particulate matter (PM2.5) has been associated with cognitive dysfunction (CD), the roles of specific PM2.5 components and potential mechanisms remain limited. This study aimed to examine the association of long-term PM2.5 and its components with CD, identify the relative importance of components in the mixture, and generate hypotheses regarding possible biological pathways.

Methods

A total of 14,379 participants aged ≥ 60 from the Henan Rural Cohort were recruited. CD was assessed using the Mini-Mental State Examination (MMSE) and the Hasegawa Dementia Scale (HDS). Concentrations of PM2.5, black carbon (BC), organic matter (OM), ammonium (NH4+), sulfate (SO42-), and nitrate (NO3-) were derived from the TAP dataset. Logistic regression and weighted quantile sum (WQS) regression were applied to estimate associations of single pollutants and pollutant mixture with CD. Network toxicology and machine learning analyses were performed to explore biological pathways and prioritize candidate genes.

Results

For each interquartile range (IQR) increase in PM2.5, BC, OM, NH4+, SO42-, and NO3-, the odds of CD increased by 19%-43%. BC consistently ranked as the leading contributor (weights: 86.38% to 54.31%), followed by OM (weights: 7.24% to 34.57%) across the four CD definitions. Network toxicology analysis identified 47 candidate PM2.5-CD genes, and RPL23, RPL36AL, and RPS25 were prioritized as hypothesis-generating candidate genes by the random forest model (RF).

Conclusion

Long-term exposure to PM2.5 and its components is associated with higher odds of CD in rural older adults. BC and OM may represent important contributors within the PM2.5 mixture. Ribosomal and mitochondrial pathways may be dysregulated in PM2.5-related neurotoxicity, but candidate genes require further validation.