Effects of High-Butterfat Diet on Embryo Implantation in Female Rats Exposed to Bisphenol A

Abstract

Bisphenol A (BPA) is an endocrine disruptor associated with poor pregnancy outcomes in human and rodents. The effects of butterfat diets on embryo implantation and whether it modifies BPA's actions are currently unknown. We aimed to determine the effects of butterfat diet on embryo implantation success in female rats exposed to an environmentally relevant dose of BPA. Female Sprague-Dawley rats were exposed to dietary butterfat (10% or 39% kcal/kg body weight [BW]) in the presence or absence of BPA (250 μg/kg BW) or ethinylestradiol (0.1 μg/kg BW) shortly before and during pregnancy to assess embryo implantation potentials by preimplantation development and transport, in vitro blastulation, outgrowth, and implantation. On gestational day (GD) 4.5, rats treated with BPA alone had higher serum total BPA level (2.3–3.7 ng/ml). They had more late-stage preimplantation embryos, whereas those receiving high butterfat (HBF) diet had the most advanced-stage embryos; dams cotreated with HBF and BPA had the most number of advanced embryos. BPA markedly delayed embryo transport to the uterus, but neither amount of butterfat had modifying effects. An in vitro implantation assay showed HBF doubled the outgrowth area, with BPA having no effect. In vivo, BPA reduced the number of implanted embryos on GD8, and cotreatment with HBF eliminated this adverse effect. HBF diet overall resulted in more and larger GD8 embryos. This study reveals the implantation disruptive effects of maternal exposure to an environmentally relevant dose of BPA and identifies HBF diet as a modifier of BPA in promoting early embryonic health.

INTRODUCTION

Infertility affects one of eight couples in the United States, with one-third of these cases without a known etiological cause [1]. One culprit is implantation failure, which may account for 30% of all early pregnancy losses [2, 3]. Evidence is mounting that environmental [4–6] and lifestyle factors such as endocrine disruptors [7, 8] and diet [9] affect embryo implantation and may contribute significantly to infertility. Bisphenol A (BPA), a prototype endocrine-disrupting chemical (EDC), is ubiquitous in our environment, and 95% of the U.S. population has detectable urinary level [10]. Large quantities are synthesized in the manufacture of plastic containers for food and beverages, water pipes, currency notes, and thermal paper receipts [11–14]. Its unintended release from these consumer products contaminates our food, water, and bodies. Because higher urinary BPA in the low-dose range (>1.243 μg/L) is associated with multiple chronic diseases [15–20], its effect on female fertility with regard to early embryo loss is thus of great concern. Indeed, in a cohort of women undergoing in vitro fertilization, higher levels of urinary BPA (3.80–26.48 μg/L) correlated with increased implantation failure [21].

Embryo implantation is orchestrated by a series of synchronized precise events of attachment, invasion, and successful embedding of the embryo into the receptive endometrium [2, 22]. However, we know very little about how BPA affects early embryo implantation in a mechanistic manner. Animal studies afford an opportunity for gaining new knowledge in this area. When mouse embryos were exposed to 1 nM BPA, the development to blastocysts in vitro was accelerated, whereas exposure to a higher dose of 100 μM impeded this process [23]. Pregnant dams administered with 100–1000 mg/kg/day BPA by gavage from Gestational Day (GD) 1 to 20 showed significant increases in embryo loss and pregnancy failure [24]. Collectively, these studies provide preliminary evidence that BPA may affect embryo implantation. However, these animal studies did not have internal exposure doses, and the route of exposure did not mimic the main route of exposure in humans, which is via dietary consumption [25]. In a recent study, 14% of human volunteers that consumed a diet high in BPA showed a serum level of 1.8–5.7 nM (0.4–1.3 ng/ml) [26]. Furthermore, in a study that included pregnant women, serum BPA was detected at around 1.5 ng/ml during early pregnancy [27]. Hence, an animal study using dietary BPA as the exposure route and with internal BPA dose measurement should provide new experimental data with higher human relevance on the impact of this EDC on early embryo loss.

The Western diet is typically high in fat. According to the U.S. Department of Agriculture, Agricultural Research Services, a diet is considered high fat if >35% of calories are from fat, whereas a low-fat diet is one in which <10% of calories are from fat. Early-life exposure of rodents to a high-fat diet was found to exacerbate the adverse effect of BPA on metabolism [28, 29] and testicular function [30] in their adult life. These studies illustrate that high-fat diets can modify BPA action in models of developmental origin of adult diseases. However, it has not been established whether a high-fat diet affects embryo implantation per se and modifies this disruptive effect of BPA in adult females. In recent years, Americans have increased their consumption of butterfat over other fats [31] because of the consumers' perception that it is a more natural, real fat [32]. As such, we believe it is timely to elucidate whether a high-butterfat (HBF) diet affects early embryonic development and implantation and if it modifies the impact of BPA, a known reproductive disruptor, in a rat model. Our data examined whether the consumption of HBF shortly before and during early pregnancy affects embryonic development and implantation, and modulates the BPA effects on embryo implantation. Thus, it attempted to unveil whether an acute HBF diet might be protective against female infertility caused by reproductive toxicants.

MATERIALS AND METHODS

Animal Ethics, Husbandry, and Treatment

Seven-week-old Sprague-Dawley female rats from Taconic Farms (Germantown, MD) were used in this study according to the animal protocol approved by the University of Cincinnati Institutional Animal Care and Use Committee. Female rats were kept in a highly controlled BPA-free environment according to the previously described conditions [33] throughout the study: all personnel wore protective garments; rats were housed in single-use, 100% polyethylene terephthalate, BPA-free cages (Innovive) and kept under positive-pressure filtered air; housing accessories were autoclaved in BPA-free containers; rats were fed ad libitum with certified BPA- and phytoestrogen-free food (Research Diets); and charcoal-filtered ultrapure water (Millipore) free of EDCs was supplied in polyethylene terephthalate, BPA-free bottles (Innovive). Upon arrival, female rats underwent a 2-wk acclimation period on experimental diets: (1) low butterfat, modifiedAIN-93G-based diet (LBF) (10% kcal butterfat), (2) LBF+250 μg/kg body weight (BW) BPA, (3) HBF (39% kcal butterfat), (4) HBF+BPA, and (5) LBF+ 0.1 μg/kg BW ethinylestradiol (EE₂). In addition, 7-wk-old adult males were acclimated similarly on a LBF diet. Body weight of the animals was monitored by weighing the animals after 2-wk acclimation and before being euthanized. Food consumption was monitored by supplying the animals with a standard amount of food while weighing the amount of remaining food every 2 wk. Our single BPA dose was selected based on the data from our preliminary data in which female rats were exposed to four log doses of BPA (0, 25, 250, and 2500 μg/kg BW); these rats plus the LBF-negative and EE₂ (0.1 μg/kg BW)-positive controls (N = 11 per group) were mated with unexposed males for 14 days. We found that only females exposed to 250 and 2500 μg/kg BW BPA had lower pregnant rate and with a significant decrease in the number of live pups on Postnatal Day 1 when compared to control LBF. The lower dose of 250 μg/kg BW BPA was thus chosen for this study. The single EE₂ dose was selected based on a previous publication [34] and our unpublished data. The 39% kcal fat was selected with reference to the fat content in a normal Western diet [35].

After the animals were acclimated, vaginal smears from the females were sampled twice daily (0700 h and 1400 h) to monitor their estrous cycles. At 1500 h, proestrous females were set mating individually with a fertile male and both animals were under fasting conditions to minimize paternal exposure. On the following morning, females with spermatozoa in the smear were denoted as GD1 (0.5 Day Postcoital).

Chemical Reagents

BPA (4, 4-isopropylidene-2′-diphenol) (Sigma-Aldrich) was ≥99% purity by weight. The basic modified phytoestrogen-free AIN-93G with 10% kcal butterfat (LBF) or 39% kcal butterfat (HBF) diets with or without BPA (250 μg/kg BW) or EE₂ (0.1 μg/kg BW) were prepared by Research Diets (see Supplemental Table S1; Supplemental Data are available online at www.biolreprod.org). The cages and water bottles were purchased from Innovive. Rat embryo culture medium mR1ECM was from Cosmo Bio; all other embryo culture-related media were purchased from Sigma. All the primary antibodies were from Abcam, and the secondary antibodies and serum were from Vector Laboratories unless otherwise stated.

Analysis of Serum BPA Levels

Sera of three to four GD4.5 rats from each experimental group were collected at the time of embryo retrieval for determining free and total BPA concentrations.

Standards and reagents.

Analytical standard of BPA was purchased from Sigma-Aldrich (≥97%). ¹³C₁₂-BPA (≥99%) was purchased from Cambridge Isotope Laboratories. β-Glucuronidase from Helix pomatia (197114 units/ml β-glucuronidase and 876 units/ml sulfatase) was purchased from Sigma-Aldrich. The stock solutions of target analytes and internal standards were prepared at 1 mg/ml in methanol and stored at −20°C.

Methanol (high-performance liquid chromatography [HPLC] grade) and ethyl acetate (ACS grade) used in the experiments were purchased from Mallinckrodt Baker. Milli-Q water was purified by an ultrapure water system (Barnstead International).

Sample preparation.

Serum samples were extracted with a liquid-liquid extraction method [36]. For the measurement of free BPA, 100 μl of serum was transferred into a 15-ml polypropylene tube. All the blanks and samples were spiked with a known amount of internal standard (20 ng ¹³C₁₂-BPA) prior to extraction. The samples were twice extracted with ethyl acetate, the first time with 6 ml and the second time with 4 ml. The extracts were combined and washed with 1 ml of Milli-Q water and then concentrated to near dryness and reconstituted with 0.1 ml of methanol before injection. For the measurement of total BPA, 150 μl of serum samples were transferred into a 15-ml polypropylene tube. The samples were buffered with 300 μl of 1.0 M ammonium acetate, which contained 41 units of β-glucuronidase, at 37°C for 12 h in an incubator shaker. Thereafter, the samples were extracted twice with ethyl acetate, the first time with 6 ml and second time with 4 ml. For each successive extraction, the mixture was shaken in an oscillator shaker for 60 min and then centrifuged at 5000 rpm for 5 min. The supernatants were combined and washed with 1 ml of Milli-Q water. After centrifuging again at 5000 rpm for 5 min, the supernatant was transferred into a 15-ml glass tube and concentrated to near-dryness under a gentle nitrogen stream. Finally, 0.15 ml of methanol was added and vortexed for analysis by high performance liquid chromatography tandem-mass spectrometry (HPLC-MS/MS).

LC-ESI(-)MS/MS analysis of BPA.

The chromatographic separation was carried out using a SHIMADZU Prominence Modular HPLC system (Shimadzu Corporation) consisting of a system controller, a binary pump, and an automatic sampler. Identification and quantification of target analytes were performed with an Applied Biosystems API 3200 electrospray triple quadruple mass spectrometer (ESI-MS/MS; Applied Biosystems). A Betasil C18 column (2.1 mm × 100 mm, 5 μm; Thermo Electron Corp.) serially connected to a Javelin guard column (Betasil C18, 2.1 mm × 20 mm, 5 μm; Thermo Electron Corp.) was used. The injection volume was 10 μl, and the mobile phase consisted of methanol and Milli-Q water that contained 1% (v/v) ammonium hydroxide. The target compounds were separated by gradient elution at a flow rate of 300 μl/min starting at 15% (v/v) methanol, held for 2 min; increased to 75% methanol within 3 min (5th min), held for 2 min; further increased to 99% methanol within 3 min (10th min), held for 4 min (14th min); and reverted to 15% methanol at the 14th min that was held for 5.5 min (20th min) for a total run time of 20 min. The MS/MS system was operated in multiple reaction monitoring positive ion mode. The compound specific MS/MS parameters for BPA are shown in Supplemental Materials and Methods (see also Supplemental Table S2). Nitrogen was used as both a curtain and a collision gas. The electrospray ionization voltage was set at −4.5 kV. The curtain and collision gas flow rates were set at 25 and 2 pounds per square inch (psi), respectively, and the source heater was set at 650°C. The nebulizer gas (ion source gas 1) was set at 20 psi, and the heater gas (ion source gas 2) was set at 70 psi. Data acquisition was set at 80 msec for scan speed and 0.70 full width at half maximum for resolving power.

Quality assurance/quality control.

Quantification of all analytes was performed by an isotope-dilution method based on the responses of ¹³C₁₂-BPA. Calibration standards injected at various concentrations ranging from 0.1 to 10 ng/ml showed a regression coefficient of ≥0.99. The limits of quantitation (0.2 ng/ml) were determined based on the lowest point of the calibration standard and volume of sample taken for analysis and the concentration factor. As a check for instrumental drift in response factors, a midpoint calibration standard was injected after every five samples. To prevent carryover of target analytes from sample to sample, a pure solvent (methanol) was injected after every five samples.

For the measurement of both free and total BPA, five procedural blanks were analyzed to determine the concentration arising from laboratory materials and solvents. Average levels of BPA detected in procedural blanks were 0.20 and 0.26 ng/ml, respectively, and these concentrations were subtracted from the measured values of samples.

Embryo Retrieval and Culture

Preimplantation embryos were retrieved from the oviductal and uterine flushings of 39 GD4.5 rats (n = 6–9 rats per group) in M2 medium (Sigma) and cultured until blastocyst formation in 20 μl mR1ECM droplets (Cosmo Bio) according to the protocol described [37]. The location, developmental stage, and blastulation of the embryos from the different treatment groups were compared.

Blastocyst Outgrowth Assay

The blastocyst outgrowth assay was performed as described previously [38]. In brief, the zona pellucida of the rat blastocysts was dissolved in acid tyrode (Sigma). The zona pellucida-free embryos were then cultured on fibronectin matrix for 4 days in Dulbecco modified Eagle medium (Sigma) supplemented with 10% fetal bovine serum (Sigma). Images of the outgrowing embryos were captured daily with an inverted microscope (Carl Zeiss Microscopy GmH) using a 20× objective lens and analyzed by AxioVision Microscopy Software (Carl Zeiss). The area covered by the embryo was recorded as the outgrowth area. The outgrown embryos were fixed with 4% paraformaldehyde (v/v) (Sigma) and immunostained for the actin cytoskeleton using Alexa Fluor-488 labeled antibody against F-actins (Invitrogen). Fluorescent images were captured using the LSM 710 confocal microscope (Carl Zeiss) with laser lines 488nm, 561nm, and 633nm.

Assessment of GD8 Embryo Sizes

The reproductive tract of the 17 GD8 rat (n = 3–4 rats per group) was harvested and digitally photographed for analyses. Embryo size, referred to as the perimeter of each implantation site, was estimated for all embryos (a total of 215 embryos; 9–18 embryos per rat) using an AxioVision microscope and Image J analysis software.

Histology and Immunohistochemistry

GD8 rat uterine tissues were fixed in 10% formalin, dehydrated, and embedded in paraffin. Serial sections at 5 μm were stained with hematoxylin/eosin (Thermo Fisher Scientific) for histologic analysis. Immunohistochemical (IHC) analysis was performed on the sections as described previously [39] to determine protein expression of cytokeratin-7 (CK7) and epidermal growth factor receptor (EGFR). Normal serum (Vector Laboratories) was used as the immunoglobulin G-negative control. Primary antibodies for CK7 (Abcam) and EGFR (Abcam) were diluted 1:100 in blocking solution. Secondary goat anti-rabbit IgG (Vector Laboratories) was diluted 1:200 in blocking solution.

Statistical Analysis

All the data were expressed as mean ± SEM from at least three independent experiments. The number of embryos, trophectoderm (TE), and inner cell mass (ICM) cells when compared among groups were analyzed by the Fisher exact test with SigmaPlot 11.0 software (Jandel Scientific), and the multiple-group comparison was analyzed by one-way ANOVA and the post hoc Tukey test with Prism 5 (Graphpad Software). Serum BPA levels were analyzed using ANOVA and Kruskal-Wallis test. P < 0.05 was considered statistically significant.

RESULTS

Exposure of Females to Butterfat Did Not Affect Serum BPA Levels

We measured serum BPA levels of four representative rats from each experimental group to determine the effects of butterfat and BPA on serum BPA levels. The free and total BPA levels detected in our LBF (control) rats were below LOQ (Table 1). Exposure of rats to HFB and/or BPA and EE₂ did not affect free BPA levels in serum. While serum total BPA levels did not change significantly in rats exposed to HBF and EE₂ when compared to that of rats exposed to LBF, serum total BPA level was significantly increased in rats exposed to BPA (2.575 ± 1.12 ng/ml) and HFB+BPA(2.96 ± 2.22 ng/ml).

TABLE 1.

The effect of butterfat and BPA exposure on serum BPA levels.

^aData were expressed as mean ± SEM.

^*P < 0.05 when compared among groups.

Exposure of Females to Butterfat and BPA Did Not Affect Food Consumption

Following exposure to different experimental diets for 2 wk, we did not observe any significant changes in food consumption rate of the females (Supplemental Fig. S1A). Although females exposed to EE₂ were lighter than their counterparts from other exposure groups, overall, there was no difference in body weight of the animals among groups (Supplemental Fig. S1B).

Exposure of Females to Butterfat and BPA Advanced Preimplantation Embryo Development

We recorded the developmental stages of embryos retrieved on GD4.5 to determine the effect of BPA and of butterfat singularly or in conjunction with BPA on preimplantation embryo development (Fig. 1). Among embryos retrieved from dams fed with a LBF diet (the control group), 75% were 5- to 8-cell embryos and 25% were 10-cell embryos (Fig. 1A). Exposure of females to BPA in conjunction with a LBF diet promoted embryo development, resulting in a significant increase in the percentage of 10-cell embryos (59% vs. 25%, P < 0.05) and morula embryos (3% vs. 0%) when compared with the percentage from dams fed only a LBF diet.

FIG. 1.

Exposure to BPA and HBF stimulated preimplantation embryo development, but BPA delayed embryo transport. A) Distribution of developmental stages of preimplantation embryos retrieved from GD4.5 females exposed to LBF and HBF either with or without BPA (250 μg/kg BW), with exposure to LBF + 0.1 μg/kg BW EE₂ as positive control for the disruptive effect on embryo implantation. The number for embryos assessed in each group is shown in parentheses within each bar, with N referring to the total number of rats used per group. B) Effect of butterfat and BPA on the incidence of embryo retrieval in the uterine flushing. *P < 0.05 when compared to treatment with LBF alone by the Fisher exact test.

When compared with all other treatment groups, dams fed a HBF diet alone had the most advanced embryos: >30% were 10-cell embryos, 17% were morulas, and 3% reached the blastocyst stage. Of significance, exposure of females to HBF+BPA resulted in a greater percentage of late-stage embryos compared with that resulting from LBF+BPA treatment, with >65% of embryos developing beyond the 5- to 8-cell stage, with 40% of those reaching the 10-cell stage and 25% reaching the morula stage (compared with 3% reaching the morula stage in the LBF+BPA group; P < 0.05). Finally, EE₂ in the LBF diet did not affect preimplantation embryo development.

Exposure of Females to BPA Delayed Embryo Transport in the Reproductive Tract

The effect of BPA and butterfat on embryo transport can be determined by the location (oviduct vs. uterus) of the embryos during retrieval on GD4.5. BPA was the only exposure that significantly delayed the transport of embryos from the oviducts to the uterus by >50% (P < 0.05) under both dietary backgrounds (Fig. 1B). In contrast, different butterfat backgrounds and EE₂ did not affect embryo transport.

Exposure of Females to HBF Increased the Implantation Potentials of Embryos Whereas Exposure to BPA Impeded Embryo Outgrowth

The effect of butterfat and BPA on blastulation and implantation (blastocyst outgrowth) was assessed in vitro to minimize the maternal effect by determining the number of blastocyst developed per total number of retrieved embryos. The blastocysts were then subjected to assisted hatching to assess blastocyst outgrowth potential. The incidence of outgrowth initiation (outgrowth rate) and the area of embryo outgrowth on fibronectin matrices were used for estimating the implantation potential of an individual blastocyst. Exposure of butterfat and BPA singularly or in combination did not affect the blastulation rate (Supplemental Fig. S2). However, exposure of females to BPA reduced the incidence of blastocyst outgrowth (42.0%), an effect comparable to that resulting from exposure to EE₂ (33.3%) (Fig. 2A). In contrast, exposure of dams to HBF promoted embryo outgrowth (81.8% vs. LBF 61.1%; Fig. 2A), and its cotreatment with BPA partially promoted the outgrowth of the embryos when compared to that of embryos from the BPA group (50.0% vs. 42% for BPA alone) (Fig. 2B). Exposure of females to BPA and EE₂ did not affect the blastocyst outgrowth area (Fig. 2A). In contrast, exposure of females to HBF increased the blastocyst outgrowth area (Fig. 2A) regardless of BPA exposure. These changes in blastocyst outgrowth were associated with the capacity of outward branching of F-actins, which signifies the presence of the actin cytoskeleton (Fig. 2B). To determine whether the effect of butterfat and BPA on embryo outgrowth was a consequence of changes in the ratio of TE to ICM of the blastocyst, we performed differential staining to assess the number of TE and ICM cells per embryo. Although HBF and BPA under a LBF background seemed to increase the number of TE cells (see Supplemental Fig. S3), the TE:ICM ratio overall did not correlate with embryo outgrowth.

FIG. 2.

HBF induced blastocyst outgrowth on a fibronectin matrix. GD4.5 embryos retrieved from rats exposed to diets as described in Figure 1 were grown to the blastocyst stage in vitro and then cultured on fibronectin matrix for 4 days. The area covered by each embryo was denoted as the outgrowth area. Effect of exposure to butterfat and BPA on the initiation rate of blastocyst outgrowth (A) and blastocyst outgrowth (B). ***P < 0.001 compared among groups by one-way ANOVA and post hoc Tukey test. C) Representative images of the actin cytoskeletal staining of embryo outgrowth. The outgrowth area of the embryo was outlined in white dotted line. Original magnification ×200.

Exposure of Females to BPA Reduced the Number of Embryos, but Exposure to HBF Increased GD8 Embryo Number and Size Regardless of BPA

Histologic analysis on GD8 embryos revealed that their number and size were not affected by exposure of females to BPA as compared with those of females with LBF exposure (Fig. 3A). However, exposure of females to HBF showed a trend of increasing the number (Fig. 3A) and size (Fig. 3C) of GD8 embryos. Regardless of BPA exposures, HBF overall significantly increased GD8 embryo number (Fig. 3B) and size (Fig. 3D) when compared to exposure of females to LBF. It also appeared to increase the number (Fig. 3A) and size (Fig. 3B) of GD8 embryos of females cotreated with BPA, although the difference was insignificant (P > 0.05).

FIG. 3.

Exposure of females to HBF increased the number and size of GD8 embryos and ameliorated the adverse effects of BPA on these parameters. Effect of butterfat and BPA singularly or in combination on the number (A) and size (C) of GD8 embryos The overall effects of exposure to LBF and HBF on the number (B) and size (D) of GD8 embryos. Each circle represents the total number of embryos per GD8 rat. The total number of rats is shown in parentheses. Data expressed as mean ± SEM. *P < 0.05 versus LBF by Fisher exact test and **P < 0.01 versus LBF by t-test and Mann-Whitney test.

Exposure of Females to BPA Increased Implantation Failure, but Cotreatment with HBF Counteracted This Adverse Effect

Histological and IHC analyses on GD8 uterine sections showed that exposure of females to BPA with a LBF diet reduced the incidence of embryo implantation (Fig. 4A) in response to the disruption in trophoblast differentiation, as reflected by the lack of staining for the trophoblast differentiation marker CK7 (Fig. 4B) and not the active blastocyst marker EGFR (Fig. 4C). These adverse effects of BPA were ameliorated when the females were fed a HBF diet.

FIG. 4.

Exposure of females to BPA halted trophoblast differentiation of GD8 embryos, but HBF stimulated the differentiation and ameliorated the adverse effect of BPA. A) Histologic analysis of GD8 uterine sections from rats exposed to diets as described in Figure 1. Arrowhead indicates the embryo. IHC analyses for trophoblast marker cytokeratin-7 (CK7; B) and the active blastocyst marker epidermal growth factor receptor (EGFR; C) on GD8 uterine sections. Cell nuclei were counterstained with hematoxylin, and normal IgG staining was used as the negative control. Sections from the BPA group are highlighted by a black box. Original magnification ×40 (A) and ×200 (B and C).

DISCUSSION

We have established the conditions for studying the effects of environmentally relevant doses of BPA and their modulation by dietary fats in the Sprague-Dawley rat model in a carefully controlled BPA-free environment [33]. In this study, BPA at 250 μg/kg BW, orally administered with diet, advanced the development of preimplantation embryos but delayed embryo transport to the uteri in both the LBF and the HBF dietary backgrounds. Dams fed the LBF diet shortly before and during early pregnancy were found to be more susceptible to the adverse effects of BPA than those given the HBF diet. The former group produced fewer and smaller embryos, and the majority of these embryos lacked the capacity for implantation or postimplantation development when exposed to BPA. In contrast, HBF restored implantation in the BPA-exposed embryos. Dams fed a HBF diet also produced more and larger embryos. Taken together, these findings demonstrate for the first time that dietary fat content may reduce early pregnancy loss either directly or via amelioration of the adverse effects of some EDCs.

BPA, as a weak estrogenic endocrine disruptor, is known to disrupt early embryonic development. Previous studies showed that subcutaneous injection of BPA at 10–100 mg/kg once a day reduced the number of embryos, delayed embryo transport to the uteri, impeded preimplantation embryo development, and adversely affected embryo implantation [40, 41], with the effects being more prominent with the exposure to higher doses. In the present study, a much lower dose of BPA (250 μg/kg BW) was delivered via diet, a route of exposure more relevant to human scenarios. When measuring serum BPA levels, while free BPA level was similar among groups, rats exposed to BPA, either in LBF or HBF background, had significantly higher total BPA level when compared to the no BPA groups. This result was similar to other studies where rats exposed to BPA at 250 μg/kg BW/day (6.13 ± 3.88 ng/ml vs. vehicle-treated <0.9 ng/ml) [42] and 15 mg/kg BW/day (22.93 ± 4.02 ng/ml vs. vehicle-treated 0.1 ± 0.05 ng/ml) (data not shown) administered subcutaneously via osmotic pumps during GD18–21 and Silastic capsules for 32 wk, respectively, had higher BPA levels. However, these total BPA levels detected following chronic BPA exposure was relatively higher than that detected in our BPA exposed rats (2.5–2.9 ng/ml). The fact that our serum total BPA level was close to the range of detectable serum BPA levels in human volunteers with high dietary BPA intake (1.8 ng/ml) [26] and in pregnant woman (1.5 ng/ml) [27] supports our route of exposure and that the respective findings would be relevant to human.

Although BPA advanced the development of preimplantation embryos, it significantly delayed embryo transport irrespective of the dietary butterfat content, implying that fast developing embryos may reach the uteri too late to gain any advantages for implantation. Moreover, dams fed a LBF diet are more susceptible than those fed a HBF diet to the adverse effects of BPA because the number and size of embryos diminished and trophoblast differentiation and migration came to a halt. It is currently unknown how BPA affects embryo transport from the oviduct to the uterus as well as trophoblast differentiation. A possible explanation is that BPA alters the oviductal environment that supports preimplantation embryo development. Exposure of mice on GD9–16 to BPA (0.1, 1, 10, 100, 1000 μg/kg BW/day) via injection has been shown to increase progressive proliferative lesion in the oviduct of F1 offspring during adulthood [43]. This implies that BPA can induce aberrant gene expression that would adversely affect the embryotrophicity of the oviduct microenvironment and delay preimplantation development and transport to the uterus. Moreover, pregnant mice administered with 50 mg/kg BW BPA from GD1–12 exhibited significant degeneration and necrosis of trophoblast giant cells, with reduction of the spongiotrophoblast layer [44]. Yet, the mechanism of the adverse effect is yet not fully understood. While Wnt proteins are expressed in mouse blastocysts [45, 46], with Wnt1 predominantly expressed in ICM while Wnt9a predominantly expressed in the mural trophoblast and ICM of the blastocoels [46], activation of Wnt signaling has been shown to increase differentiation of invasive trophoblast cells [47]. We propose that BPA may suppress trophoblast differentiation by inhibiting Wnt signaling activation, a mechanism worthy of future investigation.

In stark contrast, a HBF diet readily restored trophoblast differentiation and migration in maternal BPA-exposed embryos. In addition, the diet exerted multiple beneficial effects, including the promotion of preimplantation embryo development, an increase in embryo outgrowth potential, and an augmentation of embryo size and number on GD8. This is, however, contradictory to the reported findings that high fat diet-induced obesity impairs embryo implantation, causing poor pregnancy outcomes [48]. However, female rats in our study were exposed to a HBF diet only shortly before (2 wk) and during pregnancy. They had regular food intake and did not experience any weight gains or show signs of hyperphagia (Supplemental Fig. S1), indicating that the HBF diet was not obesogenic. Taken together, these findings shed light on the need to review whether consumption of a HBF diet during early pregnancy may protect against the idiopathic infertility due to embryo loss.

In contrast, under the LBF background, many of the characteristics of embryo development and implantation of BPA-treated dams were similar to those observed in females exposed to EE₂. Early embryonic lethality has been reported in mice exposed to EE₂ at 50 and 100 μg/kg BW [49]. This finding is consistent with our unpublished data indicating that dams fed a LBF diet had a significantly lower pregnancy rate and smaller litter size than those fed a HBF diet. It is thus crucial to recognize the human relevance of the butterfat content used in this animal study in relation to that in the normal Western diet. Of note, the 39% kcal butterfat in the HBF diet is equivalent to 30%–40% of the calories coming from total dietary fatty acids in a normal Western diet [35], whereas the 10% kcal butterfat content in the LBF diet is comparable to some vegetarian diets [50]. Hence, these findings suggest that a LBF diet may offer little protection against the adverse effects of xenoestrogens during early pregnancy, especially for women with low fertility. While HBF modifies the action of BPA during embryo implantation, the mechanism is yet unknown. We previously showed that, in rats, BPA exposure increased the susceptibility to prostate cancer through epigenetic reprogramming of the phosphodiesterase type 4 variant 4 [51] and high mobility group nucleosome binding domain 5 (previously known as nucleosome binding protein 1) [52]. Moreover, de Assis et al. [53] demonstrated that high-fat diets upregulated DNA methyltransferase 1 (Dnmt1) and promoted histone modifications [54], affecting breast cancer and metabolic disease risk. We thus surmise that HBF may vary the effect of BPA through epigenetic regulation, a topic currently under investigation. In summary, HBF exposure shortly before and during pregnancy was found to favorably modify the implantation potentials and reduce some but not all of the unfavorable effects of BPA on the implantation of embryos. Our findings may have important ramifications for women at risk of low fertility; thus, the mechanism underlying the interaction of diet and xenoestrogenic toxicants warrants future investigation.