Abstract
BACKGROUNDParticulate matter (PM) arises from a number of sources, and its composition reflects this diversity. Prior literature on PM2.5 and PM10 components in relation to circulating immune markers is scant, and nonexistent for PM0.1. Because PM toxicity depends on its particle size, source, and chemical/biological constituents, we hypothesized that specific PM components and sources may disproportionately drive immune responses.METHODSWe evaluated associations between source- and component-resolved PM0.1, PM2.5, and PM10 at 3 time lags (1-, 3-, and 12-months prior to blood draw) and 15 circulating immune markers among 1,898 female participants in the California Teachers Study who provided blood samples in 2015. Using interquartile range increases in PM sources/components and immune marker levels in quartiles, we fit multinomial logistic regression models and applied Bonferroni correction.RESULTSHigher PM2.5 elemental carbon at all exposure periods were positively associated with elevated circulating IL-1β, sCD27, and sTNFR2. Associations were strongest for 12-month elemental carbon with increased odds for the highest quartile of IL1β (ORQuartile4=2.91, 95% CI=1.92-4.43), sCD27 (ORQuartile 4=2.04, 95% CI=1.34-3.12) and sTNFR2 (ORQuartile4=2.68, 95% CI=1.72-4.18). Associations for PM10 and PM0.1 paralleled PM2.5, with two exceptions. PM10 off-road diesel was associated with elevated IL-8 at the 12-month (ORQuartile4=1.71, 95% CI=1.25-2.34) and 3-month exposure lags (ORQuartile4=1.74, 95% CI=1.33-2.26) and with CCL2 at the 3-month (ORQuartile4=1.57, 95% CI=1.20-2.04) and 1-month exposure lags (ORQuartile4=1.48, 95% CI=1.14-1.91).CONCLUSIONSThese findings support heterogeneous biological effects of PM sources/components, highlighting the importance of considering this heterogeneity when interpreting PM mass associations across populations.