Abstract
The extensive use of human-made chemicals in our daily lives results in chronic exposure to complex mixtures of potentially harmful substances. We investigated chemical exposures in pregnant women in New York City by applying a non-targeted analysis (NTA) workflow to 95 paired prenatal urine and serum samples (35 pairs of preterm birth) collected as part of the New York University Children’s Health and Environment Study. The goal was to i) study chemical exposures in this population, ii) explore differences in the chemical profiles comparing urine vs. serum samples, and comparing preterm vs. term birth samples, and iii) investigate potential associations between exogenous chemicals and endogenous metabolites. We analyzed all samples using liquid chromatography coupled with Orbitrap high-resolution mass spectrometry (LC-Orbitrap HRMS) in both positive and negative electrospray ionization modes (ESI+ and ESI-), employing full scan and data-dependent MS/MS fragmentation (ddMS2) scans. We detected a total of 1,524 chemical features for annotation, with 12 chemicals confirmed by authentic standards. Two confirmed chemicals dodecyltrimethylammonium and n,n-dimethyldecylamine n-oxide appear to not have been previously reported in human blood samples. We observed a statistically significant differential enrichment between urine and serum samples, as well as between preterm and term birth (p < 0.0001) in serum samples. When comparing between preterm and term births, an exogenous contaminant, 1,4-cyclohexanedicarboxylic acid (tentative), showed a statistical significance difference (p = 0.003) with more abundance in preterm birth in serum. An example of chemical associations (12 associations in total) observed was between surfactants (tertiary amines) and endogenous metabolites (e.g., bioactive lipid mediators and fatty acid amides).
Supplementary materials
Title
Supporting Spreadsheets
Description
detailing individual compounds' information (e.g., annotation and detection frequency) and statistical results (e.g., fold-change and p-values).
Actions