Evaluating maternal exposure to an environmental per and polyfluoroalkyl substances (PFAS) mixture during pregnancy: Adverse maternal and fetoplacental effects in a New Zealand White (NZW) rabbit model

Abstract

Mixtures of per- and polyfluoroalkyl substances (PFAS) are often found in drinking water, and serum PFAS are detected in up to 99% of the population. However, very little is known about how exposure to mixtures of PFAS affects maternal and fetal health. The aim of this study was to investigate maternal, fetal, and placental outcomes after preconceptional and gestational exposure to an environmentally relevant PFAS mixture in a New Zealand White (NZW) rabbit model. Dams were exposed via drinking water to control (no detectable PFAS) or a PFAS mixture for 32 days. This mixture was formulated with PFAS to resemble levels measured in tap water from Pittsboro, NC (10 PFAS compounds; total PFAS load = 758.6 ng/L). Maternal, fetal, and placental outcomes were evaluated at necropsy. Thyroid hormones were measured in maternal serum and kit blood. Placental gene expression was evaluated by RNAseq and qPCR. PFAS exposure resulted in higher body weight (p = 0.01), liver (p = 0.01) and kidney (p = 0.01) weights, blood pressure (p = 0.05), and BUN:CRE ratio (p = 0.04) in dams, along with microscopic changes in renal cortices. Fetal weight, measures, and histopathology were unchanged, but a significant interaction between dose and sex was detected in the fetal: placental weight ratio (p = 0.036). Placental macroscopic changes were present in PFAS-exposed dams. Dam serum showed lower T4 and a higher T3:T4 ratio, although not statistically significant. RNAseq revealed that 11 of the 14 differentially expressed genes (adj. p < 0.1) are involved in placentation or pregnancy complications. In summary, exposure elicited maternal weight gain and signs of hypertension, renal injury, sex-specific changes in placental response, and differential expression of genes involved in placentation and preeclampsia. Importantly, these are the first results to show adverse maternal and placental effects of an environmentally-relevant PFAS mixture in vivo.

Keywords: PFAS mixture; Per- and polyfluoroalkyl substances (PFAS); Placental and birth outcomes; Rabbit.

Figure 1:

(A) New Zealand White Rabbits were exposed to control or PFAS-mixture drinking water on day 0, bred on day 7, and necropsied on day 32 (or gestational day 25). At baseline, initial blood pressure, dam weight, and blood collection was determined/collected. Timepoint 1, 2, and 3 consisted of a blood draw and dam weight check, and final blood pressure was determined at timepoint 3 as well. (B) The PFAS-mixture drinking water was created to mimic an environmental drinking water sample collected in Pittsboro, NC. It contains 10 PFAS for a total load of 760ng/L.

Figure 2:

Maternal health outcomes in PFAS exposed dams include increased: (A) blood pressure (although it did not meet statistical significance) (p=0.052), (B) weight gain with gravid uterus weight correction (p=0.0125), (C) liver weight with BW correction (p=0.0066), and (D) left kidney weight with BW correction (p=0.0125). All graphs are plotted mean values with standard error bars. P-values were derived from an unpaired Welch’s t-test or a Kolmogorov-Smirnov unpaired t-test for parametric or nonparametric data, respectively. Sample size for each test: Control, n=10; PFAS-mix, n=11.

Figure 3:

(A) Representative images of maternal kidney histopathology results under light microscopy. Kidney from a PFAS-mixture exposed dam shows mild cases of nephropathy as compared to control. Arrows designate basophilic tubules. (B) Livers in exposed dams showed minimal changes in cytoplasmic vacuolation (arrow) change at low incidence.

Figure 4.

(A) Fetoplacental outcomes measured at necropsy (GD 25). All measurements are displayed as mean of means (standard deviation) and were analyzed via nested t-test, and no measures were significant (p>0.05). (B,C) Mixed models showing beta estimates and 95% confidence intervals for fixed effects of PFAS mixture exposure, kit sex, litter size, and the interaction effect of PFAS mixture treatment and kit sex on PW and BW:PW ratio. All beta estimates and 95% CI were determined using the Wald method and are centered around the control group, represented by the vertical line at zero (*p<0.05, **p<0.01, ***p<0.001). All p-values are listed to the right of each comparison. (D) Interaction plot demonstrating the interaction effect of PFAS mixture exposure and kit sex on kit BW:PW ratio. Abbreviations: RES, resorptions; BW, body weight; CRL, crown-rump length; BrW, brain weight; PW, placenta weight.

Figure 5.

(A) Incidence of dams with the presence of at least one abnormal placenta is displayed as percent of total dam number. (B) Incidence of kits with normal and abnormal placentas stratified by sex. Incidence was analyzed via contingency table and Fisher’s exact test, significant p-values are recorded on each comparison. (C) Representative photos of placenta morphology. Normal placentas are discoid and lobes are equivalent sizes and fully intact, whereas abnormal placenta morphology was noted when lobes were disconnected or uneven sizes.

Figure 6.

Placental RNA sequencing gene expression results visualized as an MA plot. Log10 transformed base mean expression of all genes is plotted on the x-axis, and the log2 fold change is plotted on the y-axis. Genes that were found to be significantly differentially expressed were labelled with gene names.