Associations of prenatal exposure to polybrominated diphenyl ethers and polychlorinated biphenyls with long-term gut microbiome structure: a pilot study

Abstract

Background: The gut microbiome is influenced by early-life exposures, but-despite potentially enormous implications for child health-is understudied in environmental epidemiology. This pilot study is one of the first to explore in utero exposures and long-term gut microbiome profiles. We examined the association between exposure to polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) during pregnancy and the mid-childhood gut microbiome.

Methods: We measured levels of PBDE-47, -99, -100, and -153 and PCB-138, -153, and -180 in maternal plasma during early pregnancy (n=18) and at delivery (n=25) in women of European descent who breastfed the child participant of the Gestation and Environment cohort in Sherbrooke, Québec (recruited 2007-2009). Bacteria in the mid-childhood (6-8 years) fecal microbiome were detected with 16S rRNA sequencing. To test for differences at the taxon level, we used the Microbiome Comprehensive Association Mapping algorithm.

Results: Early pregnancy PCB-153, -180, and the sum of PCBs (Σ₃PCB) concentrations were associated with a higher relative abundance of Propionibacteriales and Propionibacteriaceae in mid-childhood. Higher PCB-180 and Σ₃PCB were associated with higher relative abundance of Bacillales Family XI. Higher PBDE-99 exposure was associated with a decrease in uncultured bacteria within the Ruminococcaceae NK4A214 group and PBDE-47 was associated with differences in Ruminococcus 2. These taxon-level changes did not result in differences in within- or between-subject diversity. Exposures at delivery were not associated with differences in taxa.

Conclusions: Prenatal exposure to PCBs and PBDEs is associated with mid-childhood gut microbiome profiles. Larger studies are needed to confirm these results and explore health implications.

Figure 1.

Correlation matrix of lipids and lipid-adjusted polybrominated diphenyl ethers (PBDE) -47, -99, -100, -153 and polychlorinated biphenyls (PCB) -138, -153, -180 during pregnancy (P-) and at delivery (D-) in the GESTE population. Stronger correlations are indicated by darker color and tighter ellipse. Positive correlations are blue and negative correlations are red.

Figure 2.

Relative abundance of phyla in each subject during the early pregnancy and at delivery ordered by lipid-adjusted exposure concentration.

Figure 3.

Associations of exposure to polybrominated diphenyl ethers (PBDE) -47, -99, -100, -153 and polychlorinated biphenyls (PCB) -138, -153, -180, and the sum of each group of compounds with age 6– year gut microbial alpha diversity for early pregnancy and perinatal exposure measured by the Shannon Index (A) and Faith’s Phylogenetic Diversity (PD) Index (B). All analyses are restricted to women who did not smoke or drink during pregnancy, identified as white, and breastfed, and are adjusted for C-section and socioeconomic status. Point estimates are the result of linear regression; whiskers represent 95% confidence intervals.

Figure 4.

The significance of the association of exposure to polybrominated diphenyl ethers (PBDE) -47, -99, -100, -153 and polychlorinated biphenyls (PCB) -138, -153, -180, and the sum of each group of compounds with age 6–8 year gut microbial beta diversity measured using unweighted and weighted UniFrac values as well as the optimal MiRKAT and OMiAT algorithms. All analyses are restricted to women who did not smoke or drink during pregnancy, identify as white, and breastfed, and are adjusted for mode of delivery and socioeconomic status. P-values <0.2 are displayed.