Prenatal particulate air pollution and newborn telomere length: Effect modification by maternal antioxidant intakes and infant sex

Background: Evidence links gestational exposure to particulate matter with an aerodynamic diameter of less than 2.5 mu m (PM2.5) with changes in leukocyte telomere length in cord blood with some studies showing sex-specific effects. PM(2.5 )exposure in utero increases oxidative stress, which can impact telomere biology. Thus, maternal antioxidant intakes may also modify the particulate air pollution effects. Methods: We examined associations among prenatal PM2.5 exposure and newborn relative leukocyte telomere length (rLTL), and the modifying effects of maternal antioxidant intake and infant sex. We estimated daily PM2.5 exposures over gestation using a validated spatiotemporally resolved satellite-based model. Maternal dietary and supplemental antioxidant intakes over the prior three months were ascertained during the second trimester using the modified Block98 food frequency questionnaire; high and low antioxidant intakes were categorized based on a median split. We employed Bayesian distributed lag interaction models (BDLIMs) to identify both sensitive windows of exposure and cumulative effect estimates for prenatal PM2.5 exposure on newborn rLTL, and to examine effect modification by maternal antioxidant intakes. A 3-way interaction between PM2.5, maternal antioxidant intake and infant sex was also explored. Results: For the main effect of PM2.5 BDLIMs identified a sensitive window at 12-20 weeks gestation for the association between increased prenatal PM2.5 exposure and shorter newborn rLTL and a cumulative effect of PM2.5 over gestation on newborn telomere length [cumulative effect estimate (CEE) = - 0.29 (95% CI -0.49 to -0.10) per 1 mu g/m(3) increase in PM2.5]. In models examining maternal antioxidant intake effects, BDLIM5 found that children born to mothers reporting low antioxidant intakes were most vulnerable [CEE of low maternal antioxidant intake = - 0.31 (95% CI -0.55 to - 0.06) vs high maternal antioxidant intake = - 0.07 (95% CI -0.34 to 0.17) per 1 mu g/m(3) increase in PM2.5]. In exploratory models examining effect modification by both maternal antioxidant intakes and infant sex, the cumulative effect remained significant only in boys whose mothers reported low antioxidant intakes [CEE = - 0.38 (95% CI -0.80 to - 0.004)]; no sensitive windows were identified in any group. Conclusions: Prenatal PM2.5 exposure in mid-gestation was associated with reduced infant telomere length. Higher maternal antioxidant intakes mitigated these effects.