Effects of particulate matter on genomic DNA methylation content and iNOS promoter methylation

Background: Exposure to Particulate Matter (PM) from ambient and occupational sources has been associated with adverse health outcomes, including cardiovascular and respiratory disease, but the mechanisms linking PM inhalation to adverse health outcomes have not been completely clarified. Inhaled PM has been shown to produce systemic changes in gene expression, which can be detected in peripheral blood of exposed individuals. DNA methylation is a primary epigenetic mechanism regulating the expression of human genes, and initial observations in in-vitro and animal models have shown that air particles, or air particle components such as toxic metals, can induce changes in DNA methylation. Whether DNA methylation changes occur in human subjects exposed to PM has never been determined. Objectives: To identify short- and long-term effects on blood DNA methylation in foundry workers with well-characterized exposure to PM with aerodynamic diameters < 10 μm (PM10), we measured promoter DNA methylation of inducible nitric oxide synthase iNOS, a gene activated by PM exposure, and global genomic methylation content, estimated in Alu and LINE-1 repeated elements. Lower DNA iNOS promoter methylation is known to be associated with increased iNOS expression. Reduced genomic methylation content in blood DNA has been observed in subjects with cardiovascular disease, as well in cancer subjects, and can be produced in vitro by reactive oxygen species. Methods: High-precision quantitative DNA methylation analysis was performed through PCR-Pyrosequencing of bisulfite-treated blood DNA from 63 male foundry workers (mean age = 44 years, SD = 7.6) who were examined before (baseline) and after (post-exposure) three days of work. Individual PM10 exposure during the three days of work was estimated based on PM10 area measurements and time spent by the study subjects in each area. Individual PM10 level was between 73.4-1220 μg/m3. PM10 within each work area showed very high correlations (R > 0.90) in measurements taken over one year. Because all study subjects reported to have performed their standard work routines during the three days of the study, PM10 level represented both the exposure during the week of the study, as well as the usual exposure at work of the study subjects. Results: iNOS promoter DNA methylation exhibited a significant decrease in post-exposure blood samples compared to the baseline (difference = -0.61 %5 mC; P = 0.02). PM10 level showed a non-linear effect on iNOS promoter methylation measured on post-exposure samples, with the strongest decrease in the second tertile of PM10 exposure (difference = -2.93 %5 mC; P = 0.005, compared to the first PM10 tertile). Global methylation content estimated in Alu and LINE-1 repeated elements did not show changes in post-exposure measures compared to the baseline. PM10 exposure levels were significantly associated with decreased global methylation content, regardless of whether DNA methylation was measured in baseline or post-exposure blood samples. When baseline and post-exposure samples were included in the same statistical models, PM10 levels were negatively correlated with decreased methylation in both Alu (Beta = -0.19; P = 0.04) and LINE-1 (Beta = -0.34; P = 0.04), likely reflecting long-term effects of PM10 on DNA methylation. Conclusions: Our findings indicate DNA methylation as a novel molecular mechanism which may mediate PM effects on human health.