Placental transcriptomic signatures of prenatal exposure to Hydroxy-Polycyclic aromatic hydrocarbons

Abstract

Background: Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous pollutants originating from petrogenic and pyrogenic sources. PAH compounds can cross the placenta, and prenatal PAH exposure is linked to adverse infant and childhood health outcomes.

Objective: In this first human transcriptomic assessment of PAHs in the placenta, we examined associations between prenatal PAH exposure and placental gene expression to gain insight into mechanisms by which PAHs may disrupt placental function.

Methods: The ECHO PATHWAYS Consortium quantified prenatal PAH exposure and the placental transcriptome from 629 pregnant participants enrolled in the CANDLE study. Concentrations of 12 monohydroxy-PAH (OH-PAH) metabolites were measured in mid-pregnancy urine using high performance liquid chromatography tandem mass spectrometry. Placental transcriptomic data were obtained using paired-end RNA sequencing. Linear models were fitted to estimate covariate-adjusted associations between maternal urinary OH-PAHs and placental gene expression. We performed sex-stratified analyses to evaluate whether associations varied by fetal sex. Selected PAH/gene expression analyses were validated by treating HTR-8/SVneo cells with phenanthrene, and quantifying expression via qPCR.

Results: Urinary concentrations of 6 OH-PAHs were associated with placental expression of 8 genes. Three biological pathways were associated with 4 OH-PAHs. Placental expression of SGF29 and TRIP13 as well as the vitamin digestion and absorption pathway were positively associated with multiple metabolites. HTR-8/SVneo cells treated with phenanthrene also exhibited 23 % increased TRIP13 expression compared to vehicle controls (p = 0.04). Fetal sex may modify the relationship between prenatal OH-PAHs and placental gene expression, as more associations were identified in females than males (45 vs 28 associations).

Discussion: Our study highlights novel genes whose placental expression may be disrupted by OH-PAHs. Increased expression of DNA damage repair gene TRIP13 may represent a response to double-stranded DNA breaks. Increased expression of genes involved in vitamin digestion and metabolism may reflect dietary exposures or represent a compensatory mechanism to combat damage related to OH-PAH toxicity. Further work is needed to study the role of these genes in placental function and their links to perinatal outcomes and lifelong health.

Keywords: Developmental origins of health and disease; Placenta; Polycyclic aromatic hydrocarbons; TRIP13; Transcriptomics.

Fig. 1.

(A) Boxplot depicting concentrations of PAH metabolites in CANDLE participants detectable in urine. Red asterisk indicates the log LOD for each metabolite, and samples that were below the LOD were reported as $\text{LOD}∕\sqrt{2}$. (B) Correlogram depicting correlations between OH-PAH metabolites within CANDLE participants, based on Pearson correlation coefficients. right upper quadrant depicts correlation coefficient (R), and left bottom quadrant visually depicts these correlations. Only correlations that are statistically significant (P < 0.05) are shown.

Fig. 2.

Summary of the number of DEGs associated with OH-PAH Metabolites in primary model (N = 629 participants, black) and male stratified model (N = 298 participants, blue) and female stratified (N = 331, pink). Genes were considered statistically significant at FDR adjusted P < 0.05.

Fig. 3.

Effect estimates (log CPM/[ng/ml]) of fully adjusted model for all analyses (overall and sex-specific) that were statistically significant, which are presented in Table 2 and Supplemental Table 6. As noted in the figure legend, the top 2 bars of the heatmap depict OH-PAH metabolite and the model type (primary = black, female stratified = pink, male stratified = blue). Within the heatmap, the color of the box represents the magnitude of the effect estimate, with positive associations between OH-PAH metabolite concentration and placental gene expression depicted in red, and negative associations depicted in blue. Only genes/OH-PAH relationships that are statistically significant (FDR adjusted P < 0.05) are depicted. Relationships that were not statistically significant are depicted in grey.

Fig. 4.

Relative TRIP13 expression after treatment with 5 ng/ml and 50 ng/ml of phenanthrene vs DMSO vehicle control. Expression was calculated using the ΔΔCt method, and all experiments were run in triplicate and qCPR reactions were also run with technical triplicates.