DNA hypomethylation in association with internal and external markers of traffic exposure in a panel of healthy adults

Abstract

Traffic-related air pollution (TRAP) exposure is associated with negative health outcomes. Changes in DNA methylation level may be an important mechanism through which air pollution can induce its effects. The objective of this study was to evaluate the association between DNA methylation in blood and personally measured TRAP exposure. Global DNA methylation in whole blood was analyzed with HPLC in a population of 55 healthy adults (average age 41 years). TRAP was assessed for each participant with a portable aethalometer measuring black carbon (BC). Exposure measurements were collected during one typical working week. These data were used in combination with ambient levels measured at a reference site to derive subchronic BC exposure. Urinary trans,trans-muconic acid (t,t-MA), a metabolite of benzene, was used as an internal proxy of traffic exposure. DNA methylation levels were associated with short- and subchronic BC exposure. An IQR increase in BC exposure on lag 24 h (477 ng/m3), lag 48 h (491 ng/m3), lag 1 week (314 ng/m3) and subchronic exposure (618 ng/m3) were associated with a decrease in DNA methylation levels of respectively 0.0020% (− 0.0040 to − 0.0001, p = 0.047), 0.0028% (− 0.0054 to − 0.0001, p = 0.043), 0.0024% (− 0.0043 to − 0.0005, p = 0.019), and 0.025% (− 0.048 to − 0.0015, p = 0.044). In addition, an IQR increase in t,t-MA (0.135 mg/l) was associated with a 0.0021% (− 0.0033 to − 0.0008, p = 0.0019) decrease in global DNA methylation levels. Analysis of a panel of cytokines in blood samples failed to demonstrate an association between inflammatory and oxidative stress biomarkers and TRAP or DNA methylation. In a panel of healthy adults, we found negative associations between total DNA methylation and markers of TRAP exposure. Considering that change in DNA methylation concentration is a biological marker connecting environmental and lifestyle exposures and disease development trajectories, our results warrant further study.