Transgenerational effects of polychlorinated biphenyls: 1. Development and physiology across 3 generations of rats

Abstract

Background: Polychlorinated biphenyls (PCBs) are persistent organic environmental contaminants and known endocrine-disrupting chemicals (EDCs). Previous studies demonstrated that developmental exposure to the weakly estrogenic PCB mixture Aroclor 1221 (A1221) in Sprague-Dawley rats altered sexual development, adult reproductive physiology and body weight. The current study tested the hypothesis that prenatal A1221 exposure not only disrupts these endpoints within an exposed individual's (F₁ generation) lifespan, but may also affect subsequent generations (F₂-F₃).

Methods: We treated pregnant female rats on embryonic days (E) 16 and E18 with A1221 (1 mg/kg), estradiol benzoate (50 μg/kg, positive estrogenic control), or vehicle (3% DMSO in sesame oil, negative control). Endpoints related to sexually dimorphic developmental trajectories of reproductive and developmental physiology were measured, and as adults, reproductive endocrine status was assessed, in the F₁, F₂, and F₃ generations.

Results: Significant effects of transgenerational EDCs were found for body weight and serum hormones. The A1221 descendants had significantly higher body weight in the F₂-maternal lineage throughout postnatal development, and in F₃-maternal lineage animals after weaning. In females, generation- and lineage-specific effects of exposure were found for serum progesterone and estradiol. Specifically, serum progesterone concentrations were lower in F₂-A1221 females, and higher in F₃-A1221 females, compared to their respective F₂- and F₃-vehicle counterparts. Serum estradiol concentrations were higher in F₃-A1221 than F₃-vehicle females. Reproductive and adrenal organ weights, birth outcomes, sex ratio, and estrous cycles, were unaffected. It is notable that effects of A1221 were only sometimes mirrored by the estrogenic control, EB, indicating that the mechanism of action of A1221 was likely via non-estrogenic pathways.

Conclusions: PCBs caused body weight and hormonal effects in rats that were not observed in the directly exposed F₁ offspring, but emerged in F₂ and F₃ generations. Furthermore, most effects were in the maternal lineage; this may relate to the timing of exposure of the F₁ fetuses at E16 and 18, when germline (the future F₂ generation) epigenetic changes diverge in the sexes. These results showing transgenerational effects of EDCs have implications for humans, as we are now in the 3rd generation since the Chemical Revolution of the mid-twentieth century, and even banned chemicals such as PCBs have a persistent imprint on the health of our descendants.

Keywords: Body weight; Corticosterone; Endocrine-disrupting chemical (EDC); Estradiol; Polychlorinated biphenyl (PCB); Progesterone; Reproduction; Sex difference; Transgenerational.

Fig. 1

a The transgenerational experimental design and breeding strategy of the study. b The developmental timeline and endpoints for individual animals. Abbreviations: E: embryonic day of age

Fig. 2

Developmental profiles of female body weight are shown for pre-weaning rats from P1–21 (a), and for adolescents from P28–56 (b), across the three generations. Data were analyzed by repeated measure ANOVAs, and significant overall treatment effects between groups (p < 0.05) are indicated within the panels below the graphs. In cases of significantly different treatment, lineage, and time interactions, specific differences for each time point were resolved by Sidak-adjusted post-hoc comparisons, indicated above individual time points. For the latter, the treatment groups are abbreviated as D, DMSO; E, EB; A, A1221

Fig. 3

Developmental profiles of male body weight are shown for pre-weaning rats from P1–21 (a), and for adolescents from P28–56 (b), across the three generations. Data were analyzed, and abbreviations are the same, as in Fig. 2

Fig. 4

Developmental milestones of age at eye opening (a) and puberty (b) are shown for female and male rats of the three generations. Puberty in females was assessed by the day of vaginal opening, and in males, the day of preputial separation. Here and in subsequent figures that include data from both sexes, data from female are shaded yellow, and males in blue, to improve visualization. Significant (p < 0.05) treatment effects are indicated. Data were analyzed using the Kruskal-Wallis test. Abbreviations: MAT, maternal; PAT, paternal

Fig. 5

Anogenital index (AGI), calculated as [anogenital distance / ³√ body weight] is shown for females (a) and males (b) of the three generations. Data were analyzed using repeated measures ANOVAs. Significant overall treatment effects between groups (p < 0.05) are indicated within the panels below the graphs. In cases of significantly different main effects of treatment between groups, specific differences for each time point were resolved by Sidak-adjusted comparisons, indicated above individual time points

Fig. 6

Endocrine tissue indices (organ weight normalized to body weight) are shown for the ovary (a), testis (b), uterus (c), and adrenal (d). Data were analyzed by univariate one-way ANOVAs. A significant sexual dimorphism in adrenal index was found across the 3 generations. No treatment effects were observed. Abbreviations: MAT, maternal; PAT, paternal

Fig. 7

Serum hormone concentrations are shown in adult (~P60) female and male rats across the three generations for estradiol (a), progesterone (b), testosterone (c), and corticosterone (d). Note that serum progesterone concentrations are shown on different y-axes for the sexes; estradiol and corticosterone are plotted on the same y-axis scale for both sexes. To more easily visualize data, yellow shading shows results from females, and blue from males. Data were analyzed by univariate one-way ANOVAs. In cases of significantly different main effects of treatment between groups (p < 0.05), specific differences for each time point have been resolved by Tukey’s post-hoc comparisons. Abbreviations: MAT, maternal; PAT, paternal