Associations of Heavy Metals with Cognitive Function: An Epigenome‐Wide View of DNA Methylation and Mediation Analysis

• Published in: Annals of neurology Vol. 96; no. 1; pp. 87 - 98
• Main Authors: Wei, Yue, Zhou, Yan‐Feng, Xiao, Lili, Qin, Jian, Cheng, Hong, Cai, Haiqing, Chen, Xing, Zou, Yunfeng, Yang, Li, Zhang, Haiying, Zhang, Zhiyong, Yang, Xiaobo
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.07.2024; Wiley Subscription Services, Inc
• Subjects: Adults; Arsenic; Bayesian analysis; Biomarkers; Blood; Blood levels; Cadmium; Cognitive ability; Deoxyribonucleic acid; DNA; DNA methylation; Heavy metals; Manganese; Mathematical models; Older people; Peripheral blood; Statistical analysis
• ISSN: 0364-5134; 1531-8249; 1531-8249
• DOI: 10.1002/ana.26942

Objective Exposure to heavy metals has been reported to be associated with impaired cognitive function, but the underlying mechanisms remain unclear. This pilot study aimed to identify key heavy metal elements associated with cognitive function and further explore the potential mediating role of metal‐related DNA methylation. Methods Blood levels of arsenic, cadmium, lead, copper, manganese, and zinc and genome‐wide DNA methylations were separately detected in peripheral blood in 155 older adults. Cognitive function was evaluated using the Mini‐Mental State Examination (MMSE). Least absolute shrinkage and selection operator penalized regression and Bayesian kernel machine regression were used to identify metals associated with cognitive function. An epigenome‐wide association study examined the DNA methylation profile of the identified metal, and mediation analysis investigated its mediating role. Results The MMSE scores showed a significant decrease of 1.61 (95% confidence interval [CI]: −2.64, −0.59) with each 1 standard deviation increase in ln‐transformed arsenic level; this association was significant in multiple‐metal models and dominated the overall negative effect of 6 heavy metal mixture on cognitive function. Seventy‐three differentially methylated positions were associated with blood arsenic (p < 1.0 × 10−5). The methylation levels at cg05226051 (annotated to TDRD3) and cg18886932 (annotated to GAL3ST3) mediated 24.8% and 25.5% of the association between blood arsenic and cognitive function, respectively (all p < 0.05). Interpretation Blood arsenic levels displayed a negative association with the cognitive function of older adults. This finding shows that arsenic‐related DNA methylation alterations are critical partial mediators that may serve as potential biomarkers for further mechanism‐related studies. ANN NEUROL 2024;96:87–98