Impact of prenatal phthalate exposure on newborn metabolome and infant neurodevelopment

Abstract

We evaluated associations among exposure to prenatal phthalate metabolites, perturbations of the newborn metabolome, and infant neurobehavioral functioning in mother-newborn pairs enrolled in the Atlanta African American Maternal-Child Cohort during 2016-2018. We quantified eight phthalate metabolites in prenatal urine samples collected between 8- and 14-weeks' (visit 1; n = 216) and 24- and 30-weeks' gestation (visit 2; n = 145) and metabolite features in newborn dried-blood spot samples collected at delivery. Associations between phthalate metabolite concentrations and metabolic feature intensities at both visits were examined using adjusted generalized linear models (MWAS). Then, an exploratory meet-in-the-middle (MITM) analysis was conducted in a subset with NICU Neonatal Neurobehavioral Scale (NNNS) scores (visit 1 n = 81; visit 2 n = 71). In both the MWAS and MITM, many of the confirmed metabolites are involved in tyrosine and tryptophan metabolism, including tryptophan, tyrosine, thyroxine, and serine. This analysis elucidates how prenatal phthalate exposure disrupts the newborn metabolome and infant neurobehavioral outcomes.

Fig. 1. Biological pathways associated (p < 0.05) with at least one phthalate metabolite showing the pathways unique to visit 1 (28) and visit 2 (12) and the pathways shared between the two visits (11).

Visit 1 n = 216 and visit 2 n = 145. Pathway analyses were conducted using the bioinformatics tool mummichog. Source data are provided as a Source Data file. Atlanta African American Maternal-Child cohort, 2016–2018. met metabolism.

Fig. 2. Potential molecular mechanisms underlying the effects of prenatal phthalate exposure on the newborn metabolome from the metabolomic-wide association study (visit 1 n = 216; visit 2 n = 145).

Items in circles denote pathways and items in rectangle boxes denote metabolites. Blue pathways and metabolites were found in visit 1 (V1). Orange pathways and metabolites were found in visit 2 (V2). Black pathways and metabolites were found in both visits. Gray pathways and metabolites are provided for biological context but were not explicitly identified in this research. Solid arrows represent metabolites directly upstream/downstream of each other. Dashed lines represent a biochemical pathway linking metabolites not shown for clarity. Up arrows represent a positive association and down arrows a negative association. Twenty metabolites were not included, as no clear biological link was identified (see Table 3). Those marked with the maternal metabolome symbol were identified in a paper looking at the metabolic perturbations from phthalates in the maternal metabolome in this same cohort. Source data are provided as a Source Data file.

Fig. 3. Sankey plots illustrating associations between phthalate exposure, metabolite perturbations, and NICU Neonatal Neurobehavioral Scale (NNNS) outcomes at two visits.

Sankey plots in a are from visit 1 (n = 81), and b are from visit 2 (n = 71). The first node is the phthalate exposures, leading to the associated metabolites in the second node, and ending at the NNNS outcomes, attention and arousal, in the last node. The yellow/blue colors on the left-hand side indicate different phthalate metabolites. The red colors on the right-hand side differentiate attention (dark red) and arousal (light red) or both (orange) outcomes. Only significant, annotated findings are displayed in this figure; for details on the significance level and beta values, please see Table 4. Source data are provided as a Source Data file. MEP Monoethyl phthalate, MBP Mono-n-butyl phthalate, MiBP Monoisobutyl phthalate, MBzP Monobenzyl phthalate, MEHP Mono-2-ethylhexyl phthalate, MEOHP Mono-2-ethyl-5-oxohexyl phthalate, MEHHP Mono-2-ethyl-5-hydroxyhexyl phthalate, MECPP Mono-2-ethly-5-carboxypentyl phthalate, $\mathrm{\Sigma }$DEHP Sum of Di-2-ethylhexyl phthalate metabolites.