Wildfire-related PM2.5 and DNA methylation: An Australian twin and family study

作者：Rongbin Xu, Shanshan Li, Yao Wu, Xu Yue, Ee Ming Wong, Melissa C. Southey, John L. Hopper, Michael J. Abramson, Shuai Li, Yuming Guo · 发表于：Environment International · 年份：2022 · DOI：10.1016/j.envint.2022.107704 · 被引用次数：36 · 研究领域：Air Quality and Health Impacts、Epigenetics and DNA Methylation、Energy and Environment Impacts

Wildfire-related fine particulate matter (PM2.5) has many adverse health impacts, but its impacts on human epigenome are unknown. We aimed to evaluate the associations between long-term exposure to wildfire-related PM2.5 and blood DNA methylation, and whether the associations differ from those with non-wildfire-related PM2.5. We studied 479 Australian women comprising 132 twin pairs and 215 of their sisters. Blood-derived DNA methylation was measured using the HumanMethylation450 BeadChip array. Data on 3-year (year of blood collection and previous two years) average wildfire-related and non-wildfire-related PM2.5 at 0.01°×0.01° spatial resolution were created by combining information from satellite observations, chemical transport models, and ground-based observations. Exposure data were linked to each participant’s home address, assuming the address did not change during the exposure window. For DNA methylation of each cytosine-guanine dinucleotide (CpG), and for global DNA methylation represented by the average of all measured CpGs or CpGs in repetitive elements, we evaluated their associations with wildfire- or non-wildfire-related PM2.5 using a within-sibship analysis controlling for factors shared between siblings and other important covariates. Differentially methylated regions (DMRs) were defined by comb-p and DMRcate. The 3-year average wildfire-related PM2.5 (range: 0.3 to 7.6 µg/m3, mean: 1.6 µg/m3) was negatively, but not significantly (p-values greater than 0.05)...