An untargeted metabolome-wide association study of maternal perinatal tobacco smoking in newborn blood spots

Introduction Maternal tobacco smoking in the perinatal period increases the risk for adverse outcomes in offspring. Objective To better understand the biological pathways through which maternal tobacco use may have long-term impacts on child metabolism, we performed a high-resolution metabolomics (HRM) analysis in newborns, following an untargeted metabolome-wide association study workflow. Methods The study population included 899 children without cancer diagnosis before age 6 and born between 1983 and 2011 in California. Newborn dried blood spots were collected by the California Genetic Disease Screening Program between 12 and 48 h after birth and stored for later research use. Based on HRM, we considered mothers to be active smokers if they were self- or provider-reported smokers on birth certificates or if we detected any cotinine or high hydroxycotinine intensities in newborn blood. We used partial least squares discriminant analysis and Mummichog pathway analysis to identify metabolites and metabolic pathways associated with maternal tobacco smoking. Results A total of 26,183 features were detected with HRM, including 1003 that were found to be associated with maternal smoking late in pregnancy and early postpartum (Variable Importance in Projection (VIP) scores > = 2). Smoking affected metabolites and metabolic pathways in neonatal blood including vitamin A (retinol) metabolism, the kynurenine pathway, and tryptophan and arachidonic acid metabolism. Conclusion The smoking-associated metabolites and pathway perturbations that we identified suggested inflammatory responses and have also been implicated in chronic diseases of the central nervous system and the lung. Our results suggest that infant metabolism in the early postnatal period reflects smoking specific physiologic responses to maternal smoking with strong biologic plausibility.