2920

Prenatal exposures to environmental chemicals have been associated with preterm birth and low birth weight. However, limited biomonitoring data exist on the majority of the chemicals that are in commerce and are actively used in the United States. Our aim was twofold: i) to characterize exposure profiles of a diverse group of pregnant women to a broad spectrum of chemicals and identify compounds that are currently not biomonitored; ii) to identify chemicals that showed significant correlations with clinical features such as gestational age, birth weight and levels of human metabolites in serum. We analyzed 300 maternal and 300 matched cord serum samples with liquid-chromatography quadrupole time-of-flight mass spectrometry (LC-QTOF/MS) in both positive and negative electrospray ionization modes (ESI+ and ESI-). The data was processed through a non-targeted analysis workflow, which included feature extraction, formula assignment, database screening, feature annotation and finally structure elucidation. Furthermore, for selected features, MS/MS fragmentation was used to confirm structures of certain assigned chemicals. For the statistical analyses, we used the peak areas of the detected features after correcting for batch effect. For the correlations we used the Pearson’s r and p-value and we corrected for multiple hypothesis testing using the Benjamini-Hochberg method. After data cleaning and manual curation, we tentatively identified 731 chemical features (mass and retention time) in ESI+ and 824 chemical features in ESI-, for which we were able to assign molecular formulas to. Among them, 22 (3 after multiple hypothesis correction) chemicals showed significant negative correlations with gestational age and 249 (192 after multiple hypothesis correction) chemicals showed positive correlations between measured levels (peak areas) and levels of chenodeoxycholic acid (bile acid) in blood, which has been previously associated with preterm birth and cholestasis. Non-targeted analysis of serum samples elucidates novel chemical exposures and response biomarkers in pregnant women and their newborns.