Placenta disrupted: endocrine disrupting chemicals and pregnancy

Abstract

Endocrine disrupting chemicals (EDCs) are chemicals that can interfere with normal endocrine signals. Human exposure to EDCs is particularly concerning during vulnerable periods of life, such as pregnancy. However, often overlooked is the effect that EDCs may pose to the placenta. The abundance of hormone receptors makes the placenta highly sensitive to EDCs. We have reviewed the most recent advances in our understanding of EDC exposures on the development and function of the placenta such as steroidogenesis, spiral artery remodeling, drug-transporter expression, implantation and cellular invasion, fusion, and proliferation. EDCs reviewed include those ubiquitous in the environment with available human biomonitoring data. This review also identifies critical gaps in knowledge to drive future research in the field.

1. Introduction

1.a. Environmental chemicals and pregnancy outcomes

Over 86,000 chemicals are registered with the EPA through the Toxic Substances Control Act [1], and many of which are considered endocrine disrupting chemicals (EDCs) (see Glossary) as their exposure can alter normal endocrine function. Growing evidence supports the notion that these chemicals pose a risk to human health. Particularly concerning are exposures that occur during pregnancy whose effects on the developing fetus lead to long-term postnatal pathologies [2]. Discrepancies between the volume of chemical produced annually and its detection in humans through biomonitoring studies (Figure 1A) [3–11] highlights the fact that higher production does not necessarily translate to a higher human exposure. Chemicals used in everyday personal care products and plastics, such as phthalates, parabens, and bisphenols are detected in human circulation at the highest levels (Figure 1B) [12–20] compared to other high-volume production chemicals, like organophosphates, that are not as prevalent in human circulation.

Figure 1.

(A) Graphical representation of worldwide endocrine disrupting chemicals (EDC) production. (B) Paired maternal (solid bars) blood or urine and umbilical cord blood (hatched bars) EDCs concentrations (B). EDCs included here have been demonstrated to alter placental function (see text for details and Figures 2 and 3 for a graphical summary). BPA: bisphenol A, Σtri-hexaBDE: sum of tri - hexa brominated diphenyl ethers, DDT: dichlorodiphenyltrichloroethane, PCB-105: polychlorinated biphenyl congener #105, PFOS: perfluorooctane sulfonate. Blue asterisk means that the size of the dot is the smallest visible size and thus larger than the total production volume, † means that concentration was taken from placental tissue due to lack of available maternal-fetal paired-matched data.

During pregnancy, women can be exposed to over 50 different chemicals in combination [21, 22], stressing the need to understand the combined effects of the total chemical body burden during pregnancy. Epidemiological studies have begun to associate environmental chemical exposure to pathological pregnancy outcomes on birth weight, placental weight, and more recently pregnancy complications [23–25]. Whether this increased prevalence is related to the ever-increasing exposure to environmental chemicals [26] is a highly active area of research. Epidemiological evidence linking chemical exposures to hypertensive pregnancy disorders prevalence [27] and pregnancy outcomes [28] has been reviewed elsewhere. However, the evidence of chemical exposure outcomes in the context of placental dysfunction is at its infancy, which is the focus of this review.

1.b. The placenta, a transient, vulnerable organ

Pregnancy is a vulnerable period for fetal and maternal health due to the dynamic nature of the developmental and tissue remodeling processes. Pregnancy complications occur in ~19% of pregnancies [29], and include disorders like gestational diabetes, gestational hypertension, preeclampsia, eclampsia, preterm birth, and placenta percreta spectrum disorders. The prevalence of pregnancy complications such as hypertension and postpartum hemorrhaging have steadily increased over the past few decades [30], pointing to environmental exposures as one of the potential contributors to this increasing prevalence [31].

The placenta is a transient multifunctional organ necessary for fetal development that facilitates cholesterol and steroid biosynthesis, and chemical metabolism and transport. Multiple factors including nutrition, stress, and maternal diseases can result in inadequate placental development [32], causing harmful long-lasting effects to the fetus, including cardiovascular and metabolic diseases [33, 34]. Of fetal origin, the placenta develops from the trophectoderm layer of the blastocyst, comprised of stem cells known as cytotrophoblast cells (CTBs). In humans, around day 10 of pregnancy, CTBs begin to differentiate into two functionally different paths, invasion or syncytialization [35]. What determines CTB differentiation to either pathway is still not fully understood, but may be due to gene regulation, epigenetic changes [36], secretory peptides [37], or distinct stem cell populations [38]. Invasive CTBs, known as extravillous trophoblasts (EVTs), migrate away from the primary trophectoderm bundle, forming an anchoring villus (Figure 2). EVTs embed themselves in the maternal uterine lining and upon invasion, act by widening the spiral arteries to increase blood flow to the endometrial space where implantation occurred. As gestation progresses, endometrial vessel remodeling is required to perfuse the main body of the placenta and the developing fetus with maternal blood (Figure 2). Dysregulation in EVT invasion can result in placental defects such as placenta accreta, increta, or percreta, which may result in miscarriage and postpartum hemorrhaging [39]. Concurrently, CTBs begin to terminally fuse, forming a multinucleated syncytium of cells known as syncytiotrophoblasts (STBs) located above the CTB layer, which continue to proliferate and fuse. STBs line the placental villi and act as a barrier for direct maternal blood exposure to the fetus. Functionally, STBs secrete progesterone, human chorionic gonadotropin (hCG) and other proteins [40–42]. Abnormal or poor syncytialization can impair pregnancy through the loss of progesterone production, and has been implicated in abnormal birth pathologies such as preeclampsia and intrauterine growth restriction [43]. Importantly, the syncytium facilitates gas and nutrient exchange between mother and fetus, serving as a semi-permeable barrier for fetal chemical exposures [32].

Figure 2.

Overview of developmental, anatomical, and histopathological sites in the placenta affected by endocrine disrupting chemical (EDC) exposure. In the immature placenta (left), cytotrophoblasts (CTB; grey) may differentiate into two distinct lineages: invasive extravillous trophoblasts (EVT; yellow), and barrier syncytiotrophoblasts (STB; green). EVTs invade into maternal tissues and allow increased maternal blood perfusion (arrows) towards the placenta through spiral artery (SA) remodeling. In the mature placenta (right), CTBs replenish the STB population, and EVT-remodeled SAs bathe the fetal villi in maternal blood. EDC-induced alterations include: placental gross mass/wet weight (grey box 1), CTB fusion/syncytialization (grey box 2), and EVT invasion and SA remodeling (grey box 3). PBDEs: polybrominated diphenyl ethers, PCBs: polychlorinated biphenyls, and PFCs: perfluorinated compounds.

2. Endocrine disrupting chemicals and the placenta

The high abundance of steroid hormone receptor expression in the placenta [32] make it especially vulnerable to endocrine disruption. Research focused to understand the effect of EDCs on placental development has steadily increased over the past 5 years and is presented here by chemical class. Chemical classes included in this review have 1) available biomonitoring data, 2) a reported chemical-induced placental defect and 3) reported congeners that are able to cross from maternal circulation - through the placenta - into fetal circulation [16, 44–52]. As data on heavy metal exposures, like with cadmium, on placenta-specific outcomes is extensive and has been previously summarized [53, 54], those compounds were excluded from this review. World-wide human exposure levels to these chemicals have been previously summarized [13]. Overviews of anatomical sites on the placenta affected by EDC exposure, and their subsequent mechanisms and outcomes are summarized in Figures 2, 3, and 4, respectively.

Figure 3.

Summary of mechanisms of action of endocrine disrupting chemicals (EDCs) on the placenta. EDCs reviewed are listed in left and right-side boxes. Potential effectors are shown in center boxes (transporters, signaling pathways, transcriptions factors, enzymes, endoplasmic reticulum stress, and microRNA). PFCs, PCBs, and parabens have no identified mechanisms of action in the context of placental outcomes. See text for additional details. ABCB1: ATP-binding cassette sub-family B member 1, COX: cyclooxygenase, CY-P450: cytochromes P450, DDT: dichlorodiphenyltrichloroethane, ER: endoplasmic reticulum, HCaBP: calcium-binding protein, HSD3β1: hydroxy-Δ-5-steroid dehydrogenase 3-β and steroid Δ-isomerase 1, JNK: c-Jun N terminal kinase, MAPK: p38/mitogen-activated protein kinase, miRNA: microRNA, NFκB: nuclear factor κ-light-chain-enhancer of activated B cells, Nrf2: nuclear factor erythroid 2-related factor 2, PBDEs: polybrominated diphenyl ethers, PCBs: polychlorinated biphenyl, PFCs: perfluorinated compounds, PPARγ: peroxisome proliferator-activated receptor γ.

Figure 4.

Summary of functional placental disruptions linked to endocrine disrupting chemical (EDC) exposures on the placenta. EDCs reviewed are listed in left and right-side boxes Potential outcomes are shown in center boxes (hormones production, oxidative stress, invasion/migration, viability, placental weight, cell proliferation, lipid accumulation, apoptosis, and fusion). See text for additional details. DDT: dichlorodiphenyltrichloroethane, PBDEs: polybrominated diphenyl ethers, PCBs: polychlorinated biphenyl, PFCs: perfluorinated compounds, CRHs: corticotropin-releasing hormone, E2: estradiol, P4: progesterone, PGE2: prostaglandin E2.

2.a. Bisphenols

Bisphenols are man-made chemicals widely used in the production of polycarbonate plastics, epoxy resins, and thermal receipt paper [55, 56]. With the exception of bisphenol A (BPA), the effect of bisphenol exposure on the development of the placenta is not well established. Although BPA biomonitoring exposure levels may have been underestimated to date [57], epidemiological data demonstrate a positive association between total BPA concentration in the placenta and placental global methylation [58]. Lower birth weight has also been linked to a higher ratio of BPA concentration in the amniotic fluid vs. maternal plasma in pair-matched samples [44], suggesting that an individual’s placental permeability to bisphenols may be one of the defining factors driving exposure levels and subsequent outcomes.

In vivo exposure to BPA during pregnancy has been studied in doses ranging from 0.002 to 200 mg/kg/day across pregnancy. Much of these data have been already reviewed [59, 60], and highlights BPA’s impact on inducing placental cell apoptosis, labyrinth layer loss, and altered expression of nuclear hormone receptors. Importantly, both intrauterine growth restriction (IUGR)-like [61] and pre-eclampsia-like phenotypes [62] have been reported, and were hypothesized to result from aberrant spiral artery remodeling (Figure 2). At higher doses (BPA: 0.5 mg/kg/day), an IUGR-like phenotype accompanied by placental inflammatory changes has been recently reported in sheep [63]. Studies regarding placenta-specific outcomes following exposure to BPA-analogues, or “replacement” chemicals remain scarce. A single study has reported a placental defect following exposure to bisphenol S (BPS: 0.5 mg/kg/day) [64] with a reduction in binucleate cells, the sheep homologue of human STBs and hypothesized to occur through a cell fusion defect (downregulation of e-cadherin) [64]. Despite lower world-wide exposures to BPS than BPA [65], a recent toxicokinetic study reported a prolonged half-life in fetal circulation [66].

In vitro, BPA exposure at 1,000 μM can either reduce [67] or increase [68] cell proliferation in the choriocarcinoma cell line BeWo - an in vitro model of syncytialization. Functionally, BPA exposure in human metastatic choriocarcinoma-derived JEG-3 cell line (doses: 0.1 – 50 μM) reduced estrogen synthesis [69, 70] and altered cytochrome P450 (CYP) enzymatic activity [71] and protein expression (CYP11A1 [72]; CYP19 [73]; and CYP1A1 [70]) (Figure 3). These exposure conditions disrupted hormone signaling via a reduction in corticotropin gene expression [74]. Of the emerging BPA-analogue compounds [75], only BPS exposure has been studied for placenta-specific outcomes. BPS reduces the activity of the transport protein ATP-binding cassette transporter (ABC) B1 in CRL-1584 cells, a transformed placental epithelial cell line (0.5 nM) [76]. In contrast, BPA can directly stimulate ABCB1 expression in the choriocarcinoma cell line BeWo (10 μM) [77], leading to an increase in drug efflux. In vitro BPA’s effect on EVTs invasiveness and apoptosis has been reviewed [78]. However, the mechanism of action of BPA on EVT invasiveness has not been fully elucidated [78]. BPA has also been shown to have epigenetic effects, increasing microRNA expression at relatively high doses (25 ng/μl BPA) in the EVT cell-line HTR8-SVneo [79]. Despite this breadth of in vitro data demonstrating that bisphenols can have placenta-specific effects, epidemiological evidence supportive of such bisphenol-induced placental dysfunctions in humans remains lacking.

2.b. Phthalates

Phthalates are ubiquitous chemicals present in a myriad of consumer and personal care products, pesticides, and solvents. Found in ~100% of humans tested [45, 80, 81] phthalates tend to be higher in females compared to males [82]. Total maternal urinary metabolites for di-2-ethylhexyl phthalate (ΣDEHP) have been inversely associated with placental weight at term in U.S. and E.U. cohorts, suggestive of placental insufficiency [83, 84]. ΣDEHP urine concentrations are also higher in IUGR pregnancies [85], and associated with lower expression of trophoblast differentiation genes [86] and various long non-coding RNAs (lncRNAs) with unknown placental function [87]. Phthalate exposure during the first trimester has also been negatively associated with the expression and methylation of the epidermal growth factor receptor (EGFR) in placental tissue [25]. Given that EGFR is most abundant in placental tissue compared to any other tissue and the role of EGFR in human placental development [88], the implications of these epigenetic modifications on placental pathology should be investigated.

In vivo data on the effects of phthalate exposure on placental function is limited to rodents and oral exposure to the most historically produced phthalate [89], di-2-ethylhexyl phthalate (DEHP), although other phthalates such as dihexyl phthalate (DHP) and dicyclohexyl phthalate (DCHP) have also been investigated [90]. Phthalate exposure regimens span gestation, in most cases starting at vaginal plug detection, and use doses ranging from 20 to 1,500 mg DEHP/kg/day. Data from these studies has been previously reviewed [59, 60]. In brief, all histopathological observations reported a reduction in the placental labyrinth layer which is analogous to the human syncytium, containing spongiotrophoblast cells similar in function to human STBs [91]. Reduction in the STB population in human placentas has been linked to the pregnancy complication preeclampsia, and is a noted a defect in the placentas from IUGR pregnancies [43]. Gene expression changes in the placenta following DEHP exposure reflect altered fatty acid homeostasis, apoptosis, and angiogenesis which is accompanied by irregular vessel formation in the labyrinth resulting in an IUGR-like phenotype [92] (Figure 2). Importantly, dosing regimens in the above-mentioned studies exceed not only the estimated population daily intake [93], but both the U.S. Environmental Protection Agency’s (EPA) reference dose (RfD) and the E.U. Scientific Committee for Toxicity, Ecotoxicity and the Environment’s the tolerable daily intake for DEHP by, in some cases, over three orders of magnitude [93]. This highlights the need to conduct studies that better reflect environmentally relevant exposure levels in human pregnancies.

Although non-cytotoxic at doses up to 500 μM [94], in vitro studies for the DEHP metabolite monoethylhexyl phthalic acid (MEHP) can induce apoptosis, and increase reactive oxygen species production and DNA damage in HTR-8/SVneo cells [59] (Figure 4). Of note, H₂O₂-induced oxidative stress alters the expression levels of miRNAs and mRNA expression of genes involved in placental development [95], an effect also observed with miR-16 [96], which plays an important role in MEHP-induced trophoblast cell apoptosis by decreasing B-cell lymphoma 2 (BCL-2) expression [96]. Additionally, in human CTBs, MEHP exposure inhibited hCG production [97], but enhanced mRNA expression of corticotrophin releasing hormone (CRH) [98]. Marked accumulation of glycerolipids and glycerophospholipids in the rat trophoblast cell line HRP1 [99] coupled with altered lipid metabolism in JEG-3 cells [100] also points to a potential phthalate-induced placental lipid imbalance. Although the effect of an altered placental lipidome is not yet understood [100], glycerolipids and glycerophospholipids can inhibit receptor binding of progesterone and estrogen [101]. An enhanced inflammatory response following MEHP exposure was also noted in primary isolated placental macrophages (180 μM) [102] and human CTBs [98] through an increase in cyclooxygenase 2 (COX2) mRNA expression and protein abundance (Figure 3). Even though the mechanisms underlying most of these phenotypes remain elusive, MEHP-induced inflammatory responses appear to be driven through peroxisome proliferator-activated receptor γ (PPARγ) activation [86], of which MEHP has been shown to be a high affinity ligand [103]. Furthermore, it has been demonstrated that HTR-8/SVneo cell invasion is reduced upon MEHP exposure [59].

Similar to other EDCs, the occurrence of phthalates in humans is in a mixture [104]. Phthalates in mixture have been shown to act through PPARγ in a sex-specific manner, and are hypothesized to be PPARγ agonists in females and antagonists in males using primary isolated CTBs [104]. Although phthalates are among the best studied chemicals in the context of placental function, emerging chemicals in DEHP-free plasticizers, such as di(isononyl) cyclohexane 1,2-dicarboxylate (DINCH), di(2-propylheptyl) phthalate (DPHP), di(ethylhexyl) adipate (DEHA), and O-acetyl tributyl citrate (ATBC) [105] have yet to be evaluated.

2.c. Parabens

Generally recognized as safe by the U.S. Food and Drug Administration (FDA), parabens are used as antimicrobial agents in personal care products [106]. To our knowledge, a single epidemiologic study has evaluated placenta-specific outcomes, detecting a placental growth defect via a positive association between total maternal urinary paraben levels and placental weight [83] (Figure 2). In vivo pregnancy exposure studies are restricted to a pharmacokinetic study in pregnant rats where a 3-fold higher concentration of ethylparaben was observed in the placenta vs. the fetal liver [107], suggestive of placental accumulation. Additionally, in humans, the negative association observed between cord blood ethylparaben and testosterone concentrations points to a potential risk for prenatal development [46]. A recent in vitro study using HTR-8/SVneo cells reported that butylparaben exposure inhibits cell proliferation and induces both apoptosis and endoplasmic reticulum stress (200 μM) [108] (Figure 4). However, the specific molecular mechanism(s) of how paraben exposure results in these outcomes remains unexplored. As world-wide exposure to parabens during pregnancy is second most only to phthalates (see Figure 1B), the large discrepancy between known gestational exposure outcomes between EDC classes is likely driven by the fact that parabens are generally recognized as safe. This provides a gap in knowledge worth evaluating, especially in combination with other common chemical exposures.

2.d. Polychlorinated biphenyls

Polychlorinated biphenyls (PCBs) are man - made organic chemicals used in the production of electrical equipment and building materials, and contain over 200 known congeners [109]. Despite being banned in the U.S. since 1979, many products manufactured earlier still contain PCBs and thus, exposures continue to occur to this day [110]. Because PCBs were used in the form of chemical mixtures (trademark examples include: Arochlor, Clorphen, or Phenochlor) that included several PCB congeners, PBCs were among the first chemicals evaluated as EDC mixtures [111]. The 10–15 years half-life estimated for PCBs results in long-term exposure in humans. Given that PCBs can cross the placental barrier [47] the developing fetus is at risk of PCB exposure. Epidemiologically, PCBs concentrations in the placentas from the Japan Environment and Children’s Study cohort have been associated with a decrease in syncytiotrophoblast volume in the placenta and elevated placental growth factor (PIGF) expression, which stimulates placental vessel branching and spiral artery remodeling [112]. Importantly, birth weight has been inversely correlated with placental PCB concentrations in a Chinese cohort [113], an effect that could be attributed to placental disruption.

Gestational exposure to over two magnitudes of the EPA’s RfD for PCBs ((20 ng/kg BW/day; [114]) disrupts the placental labyrinth layer in rats (20 μg/kg BW/day, PCB-126 [115]) and minks (0.65 mg/day in feed, Clophen A50 [116]). In minks, this effect was combined with altered spiral artery remodeling resulting in fetal growth retardation or demise [116]. However, despite the estrogenic, antiestrogenic, or androgenic effects of PCBs [117, 118], human studies have reported no association between PCB exposure and the risk of spontaneous abortion and/or stillbirth [119]. In vitro models using BeWo cells have demonstrated placental transfer of PCBs with transfer speeds differing across PCB congeners (i.e. PCB-180 transfers more rapidly than PCB-52) [47]. PCB mixtures can induce trophoblast cell apoptosis through upregulation of the adaptive immune response (PCB mixtures #77, #126, and #169: 40–120 μmol/l [120]), disrupt invasion in HTR-8/SVneo cells (10 μg/ml Aroclor 1254 [121]), and induce anti-angiogenic effects at the maternal-fetal interface (Aroclor 1254 [121]) (Figure 4). Bovine placental explants exposed to different doses of a PCB mixture (PCB-153, PCB-126, and PCB-77; 1–100 ng/ml) report increased connexin 43 (Cx43) and 32 (Cx32) expression [122]; of which Cx43 is involved in the intercellular communication required for placental cell fusion [123]. Most PCB congeners are formulated for use as a mixture. This, coupled with the fact that PCBs have an accumulative environmental persistence, identifying any placenta-specific mechanisms responsible for exposure outcomes remain amongst the most challenging.

2.e. Perfluorinated compounds

Produced since the 1950s, perfluorinated compounds (PFCs) are used in the production of antifouling paints, non-stick cookware, and waterproof clothing [124]. While perfluorooctonoic acid (PFOA) and perfluorooctane sulfonate (PFOS) are the most commonly studied PFCs, others persist in the environment [125]. Using pair-matched maternal blood, cord blood, breast milk, and placental samples, epidemiological studies have demonstrated that all PFCs examined (PFOS, PFOA, PFDA, and PFTrDA) are able to enter fetal circulation both prenatally through placental transport, and postnatally during lactation [48, 49]. Recently, umbilical cord blood levels of PFCs perfluorobutane sulfonate (PFBS), perfluorohexane sulfonate (PFHxS) and perfluoroundecanoic acid (PFUA) from a Chinese birth cohort have all been positively associated with preeclampsia [126]. Because of PFCs cumulative nature and long half-lives (~3.4 and ~2.7 years for PFOS and PFOA, respectively [127]), pregnancy exposures are of particular concern. Limited animal studies include mice [128] and rat [129] PFOS exposure through oral gavage (8 – 20 mg/kg/day) during mid-to-late gestation resulting in reduced fetal and placental weights accompanied with placental necrosis [128] (Figure 2), an increase in fetal serum corticosterone [129], and an inhibition of placental 11-β-hydroxysteroid dehydrogenase (HSD) activity [129]. However, doses used were 6 magnitudes higher than EPA’s RfD (20 ng/kg BW/day) for both PFOS and PFOA [130, 131].

In vitro, PFOS modulates steroid hormone signaling by suppressing aromatase production, estradiol secretion, and progesterone production in a concentration-dependent manner in primary isolated human CTBs with effects noted at doses as low as 0.001 μM [132] (Figure 4). Aromatase inhibition was also observed in JEG-3 cells after PFOS, PFOA and PFBS exposure (IC50: 57 – 80 μM [133]. PFOS additionally led to decreased cell viability [132, 133] and induction of apoptosis [132] in the same cell lines (Figure 4). A perfusion model using human placental explants has reported a negative correlation between the organic anion uptake transporter OAT4 and fetal PFOA transfer [134], demonstrating the protective potential of placental OAT4 against fetal PFC exposure. Most in vivo data use rodents, which have been shown to eliminate PFCs more rapidly than humans [135], making them a less than ideal animal model for gestational PFC exposures. Despite the concerns raised from in vitro experiments using human cell lines, studies that focus on lower, more physiologically relevant dosing strategies are necessary to further the toxicological evaluation of PFCs.

2.f. Organophosphate

Organophosphates are esters of phosphoric acid used as insecticides due to their direct interference with the neurotransmitter acetylcholinesterase (AChE), causing systemic muscle paralysis [136]. To our knowledge, no epidemiologic studies exist reporting placenta-specific outcomes in association with organophosphates exposure. However, the assessment of life-long or past organophosphate exposures appears to be a crucial limitation to epidemiologic studies because organophosphates are not persistent in the human body [137]. Despite the existence of over 40 organophosphates, placenta-specific outcomes evaluated are restricted to chlorpyrifos and methyl parathion. Limited mid-to-late gestation exposures (GD 9 – 21) in rats, high dose exposures (10 – 30 mg/kg) to either chlorpyrifos, methyl parathion, or a mixture of the two, inhibited placental AChE activity [138]. Gestational exposure to only methyl parathion (1 – 2 mg/kg) reduced the trophoblast giant cell population, and increased phagosome vacuoles in the labyrinth layer [139] (Figure 2). To note, these cytotoxic findings occurred at doses within two magnitudes of the EPA established RfD for human exposure to organophosphates (0.025 – 100 μg/kg/day [140, 141]).

Chlorpyrifos exposure, even at micromolar concentrations, is also cytotoxic in human placental choriocarcinoma cells (JEG-3 [142]) (Figure 4) and can induce apoptosis in JAR cells through tumor necrosis factor (TNF) modulation [143] (Figure 3). However, not all studies have reported cytotoxicity, even with the same cell line (JEG-3) [144]. Chlorpyrifos also altered expression of pregnancy maintenance markers such as the ABC transporter ABCG2, the transcription factor GCM1 (glial cells missing transcription factor 1) and hormone subunit β-hCG [144], but not progesterone or estradiol production [145]. Additionally, enhanced reactive oxygen species (ROS) production [146] and upregulation of endoplasmic reticulum (ER) stress-related proteins [147] occurred after chlorpyrifos exposure in JEG-3 cells. Attenuation of chlorpyrifos-induced oxidative stress [146] and ER stress [147] occurs through adaptive activation of the Nrf2-antioxidant response element signaling pathway.

One major limitation in the study of the effects of organophosphates exposures is that they are commonly found in mixtures, such as the flame-retardant mixture fire-master (FM) 550, for which the exact composition and RfD are not available. FM 550 accumulates in the placenta to a greater extent in males than females [148], but no association with developmental outcomes has yet been reported. Dosing regimens used in these studies (300 or 1,000 μg/kg/day FM 550 [148, 149]) are both within the RfD range for organophosphates. Due to the sex-specific accumulative nature of FM 550 in the placenta, outcomes following exposure to other organophosphates chemical mixtures as well as long-term fetal exposure outcomes should be further evaluated.

2.g. Dichlorodiphenyltrichloroethane

Dichlorodiphenyltrichloroethane (DDT) is an insecticide whose use has been banned in the U.S. since 1973 due to its environmental persistence (biological half-life: ~7 years). Despite concerns over its estrogenic properties, accumulation in adipose tissue, potential carcinogenicity, and developmental neurotoxicity [150], DDT continues to be used for malaria outbreaks in developing countries [151, 152]. DDT crosses the placenta and enters fetal circulation [51, 153]. Epidemiologically, exposure to DDT has been associated with maternal hypertensive disorders [154]. Associations of DDT and birth weight are cohort dependent with an inverse correlation in a Saudi Arabian cohort [155], and a positive [156] or no correlation [157] in two U.S. cohorts. Despite a breadth of knowledge on human health effects following DDT exposure [158, 159], to our knowledge, no available in vivo studies exploring gestational DDT exposure focus on placenta-specific outcomes. In vitro, DDT exposure results in decreased placenta cell viability at doses higher than 25 μM (HTR-8/SVneo [160]) but did not alter proliferation at lower doses (1 nM) [161] (Figure 4). In bovine placental explants, DDT increased the explant’s expression of prostaglandin E2 (PGE2) synthase, 3β-HSD, and CYP11A1 (doses: 1, 10 or 100 ng/ml [162]) (Figure 3). Alterations in enzyme expressions were accompanied by an increase in oxytocin, estradiol, and progesterone secretion [162]. These effects were also observed in human placental explants [163, 164] along with inhibition of aromatase activity [163]. Given the role of intercellular communication in placental syncytium formation [123], DDT exposure on connexin protein expression was tested, but no effect was observed [122]. Overall, DDT exposure reduces the secretory activity of the placenta, and given its carcinogenic effect [165] and continued commercial use [151], further understanding of its effect upon gestational exposure - specifically on the placenta - is warranted.

2.h. Polybrominated diphenyl ethers

Similar in chemical structure to PCBs, polybrominated diphenyl ethers (PBDEs) are persistent chemicals used as flame retardants in paints, plastics, electrical equipment, and textiles [166]. Of the 209 known PBDE congeners [167], less brominated congeners such as tetra- and penta-BDEs have a high affinity for lipids and tend to accumulate in animals, suggestive of a greater toxic potential [166]. PBDEs can be found in human breast milk, cord blood, and placental tissue [52] where they tend to bioaccumulate [168] and get transferred to the fetus [169, 170]. Interestingly, PBDE concentrations have been reported to be up to two-fold higher in the placentas of males than females [171]. Cord blood concentrations of PBDEs have been negatively correlated with birth weight [172], and inversely correlated with placental DNA methylation changes in human pair-matched samples [173, 174], with changes specific to PBDE congener and methylation site (tetraBDE-66, LINE1; hexaBDE-153, NR3C1 and IGF2; decaBDE-209, IGF2) [174]. Placenta PBDE concentrations have also been positively associated with changes in microRNA (miR)-188–5p and miR-1537 expression (decaBDE-209, [175]). Both miRs have unknown roles in the placenta, but miR-188–5p is abnormally upregulated in pre-eclamptic placentas [176], providing with a potential biomarker for early detection of pregnancy complications. Despite the fact that PBDE accumulates in the fetal portion of the placenta [177], no animal studies exploring placental effects of PBDEs are available. Gestational PBDE exposure in rats results in reduced weight at birth, and has been linked to a loss in maternal triiodothyronine production [177].

PBDE exposure is cytotoxic in second trimester human CTBs at doses over 10 mM (BDE-47 and BDE-99 [178]), significantly reducing cell viability and leading to apoptosis [178] (Figure 4). BDE-47 at the same dose also reduced the migration and invasion of CTBs, and altered lipid and cholesterol metabolism [178]. BDE-47 is the most studied PBDE in the context of placental function, with reported effects on oxidative stress [179] that result in an increase in PGE2 production in HTR-8/SVneo cells [128]. Exposure to PBDE mixtures (congeners: 47, 99, and 100) also resulted in higher PGE2 production in second trimester placental explants [180]. In JEG-3 cells, from doses as low as 0.5 nM and in a dose dependent manner, BDE-47 increased CRH production [181], which has been associated with premature delivery in humans. This same dose-dependent effect was observed in the mRNA expression of signal transduction proteins like protein kinase C-α subunit (PKCα), c-Jun N terminal kinase (JNK), and p38/mitogen-activated protein kinase (MAPK) phosphorylation [181] (Figure 3). Considering that only a fraction of >200 PBDEs have been tested for placenta-specific outcomes and that less studied hydroxylated metabolites of PBDEs (OH-PBDE) can inhibit CYP19 [182], research into the effect of PBDEs on reproductive and placenta-specific outcomes is merited.

2.i. Organotins

Organotin compounds are chemicals with a central tin (Sn) atom and hydrocarbon substituents that are commonly used as polyvinyl chloride stabilizers and biocides [183]. Organotins can cross the placental barrier, resulting in the accumulation of Sn in the conceptus and decidual mass in rat pregnancies [184]. Organotins have been shown to lead to embryonic lethality in non-human primates [185] and rats [186], but just recently have been shown to lead to fetal mortality, conceptus apoptosis and malformations, lower placental weight, thinner labyrinth and basal placental layers (dibutyltin (DBT) chloride, 20 mg/kg, [184]) (Figure 2). Although these in vivo studies were conducted with higher doses than those observed in human exposures, a prospective Danish cohort on cryptorchidism (58 male placental homogenates) found an inverse association between the sum of placental organotins and reverse 3,3’,5’-triiodothyronine (rT3), the third most common iodothyronine [50]. This association was more pronounced in samples with higher a tributyltin (TBT) concentration. However, these findings have yet to be reproduced in an animal study and other human cohorts. Out of the many organotins, only TBT and TPT have been studied extensively in the context of placental dysfunctions in vitro. TBT has been shown to either decrease in micromolar dosages (JEG-3, [187]) or increase at nanomolar concentrations (placental explants, [188]; JAR, [189]) progesterone production (Figure 4). This alteration in steroid hormone production was accompanied by a reduction in 3β-HSD activity (JEG-3, [187]) or an increase in 3β-HSD expression (JAR; [189]) (Figure 3). Higher hCG production was also observed in JAR and JEG-3 cells [190]. These endocrine changes have been shown to be mediated either by PPARγ [189] or retinoid X receptor (RXR) [191]. TBT exposure also resulted in gene expression changes associated with cytokine signaling in non-human primate trophoblast stem cells [192]. Importantly, TBT, a known obesogenic chemical [13], can also increase di- and tri-acylglycerol in JEG-3 cells [193]. Despite the global ban in TBT use for anti-fouling paints in 2008, the use of organotin chemicals continues to be widespread [183] and therefore, studies investigating the effects of organotins on placental function at environmentally relevant doses are needed.

3. Concluding Remarks

This review provides a holistic overview of the current knowledge on placental outcomes following EDC exposures by reviewing epidemiological, in vivo and in vitro data. Despite the seemingly large body of literature, there are many limitations to these studies which are summarized in Box 1. One of the main limitations is the use of supraphysiological dosing regimens, which are often magnitudes higher than human exposures and report directly cytotoxic outcomes. Human biomonitoring and in vivo pharmacokinetic data should be used to develop predictive physiologically-based toxicokinetic mathematical models to establish environmentally relevant dosing strategies. These doses should be coupled with in vitro models that better resemble the in vivo placental microenvironment, such as the use of 3-dimensional models that recreate cell-to-cell interactions and incorporate tissue shear stress and extracellular matrix components [194, 195] (see summary of Future Directions in Box 2). These alternative in vitro models should be used in combination with in vivo studies that integrate relevant animal models to capture more human-representative toxicokinetic profiles or anatomy. For instance, two well established models of studying the placenta for translatability into humans are guinea pig and sheep whose advantages in human placental translatability have been previously summarized [196]. However, few studies have adopted either as animal models.

Box 1: Limitations to current studies.

• Most studies evaluating placental toxicity tend to use supraphysiological exposure conditions, thus reducing the relevance for human translatability.

• Single chemical toxicity studies are not representative of complex human exposures. In these conditions, there can be both intra- and inter-class mixtures, further complicating data interpretation.

• The fundamental lack of knowledge regarding human trophoblast commitment, differentiation, invasion, and specific trophoblast populations make it difficult to develop standardized placental cell isolation and purification processes for toxicological screening. This lack of procedural standardization can lead to a lack of reproducibility. In addition, contradicting in vitro results, as noted for BPA, make the process of in vitro to in vivo translatability challenging.

• Currently, there are no available EPA guideline studies for placenta-specific outcomes. This creates not only a barrier for studies which may identify only placenta-specific outcomes from being used in regulatory decision processes, but also makes uniformity among studies difficult.

• Placental studies routinely use altricial, litter bearing rodents. Confounding factors like uterine implantation site and fetal position in these species can lead to non-uniform chemical exposures.

• The detection of sex-specific effects on placental development following chemical insult calls for the need of immortalized placental cell populations with known sex. However, the lack of non-cancer derived commercially available human trophoblast cell populations with known sex makes understanding sex-specific outcomes difficult. Despite the identification of sex-specific placental outcomes upon chemical exposures, very limited knowledge of such is available.

Box 2: Future directions for EDC-induced placental dysfunction research.

• Epidemiologic and/or biomonitoring data should be used to identify emerging “replacement” chemicals. This knowledge will aid to drive in vitro and in vivo toxicological studies.

• Use of low dosing strategies more relevant to world-wide human exposures can help strengthen the impact and translatability of placenta-specific toxicological findings.

• The use of animal models more suitable for human translatability should be chosen, such as guinea pig, which are a precocial species and have a more comparable placentation to humans. Importantly, such animal studies should be design with enough power to detect sex-specific effects.

• The use of a large animal model, like the sheep, would allow non-terminal collection of an intact placenta through natural birth. Because pregnancy would not need to be terminated to collect the placenta in this manner, the offspring can remain as part of a transgenerational study.

• While in vivo studies allow for a more holistic and integrative picture of the chemical disruption, in vitro models allow for higher throughput. However, current “standard” in vitro placental cell culture techniques can be limiting.

• The use of 3-dimensional culture systems with human cell lines that incorporate cell-cell interactions and shear stress can better mimic the in vivo environment and thus, have greater translational relevance. The placenta has a sex and therefore, in vitro studies should also begin to take into consideration sex as a biological variable.

• Greater attention should be paid to the study of chemical mixtures, as they have been associated in complex pregnancy disorders like intrauterine growth restriction and preeclampsia. Identifying pharmacokinetic of pharmacodynamic interactions within mixtures will be critical when interpreting their effects. Additionally, determining if there are causal links between mixture exposures and pregnancy disorders should be a research priority.

• Forming interdisciplinary teams to bridge the knowledge gaps between epidemiological and cause-and-effect studies could exponentially advance the already growing body of literature on placental toxicity outcomes.

• The development of systematic analyses and standardized exposure protocols will help increase interstudy reproducibility and overall data confidence.

Additionally, to date, most placental studies focus on single chemical exposures with mixture studies predominately restricted to rodent models investigating organophosphate mixtures. Using relevant, complex mixtures in toxicology studies will improve our understanding of the potential pharmacokinetic and pharmacodynamic interactions between chemicals and their effects on placental development and function. Additional complexity to placental toxicological studies is fostered by the lack of EPA guidelines for placenta-specific outcomes. Another aspect that is often unrecognized is that, derived from the embryonic trophoblast layer, the placenta has a defined sex. However, sex-specific effects are most often not reported. In vitro studies are also limited by sex, as commercially available placental cell lines are derived from male pregnancies, or do not account for sex. Recent studies reporting sex-specific associations showing PBDE, organophosphate, and paraben accumulation in the male placenta, with no reported placenta-specific outcomes, highlight the need to further explore this gap in knowledge. Overall, significant advances in risk assessment of EDCs, particularly on understanding exposure effects on the development and function of the placenta, can be made by addressing these limitations and working towards the proposed future directions.

INTERACTIVE QUESTIONS

QUESTION #1

Question (max 240 characters)

• 1)How many chemicals are women exposed to during pregnancy?

Check correct answer

• a)1
• b)10
• c)30
• d)over 50

Explain why this is the correct answer (max. 2500 chars):

Chemical body burden in pregnant women occurs as a mixture. In studies testing up to 60 different chemicals, over 50 chemicals have been detected in a single pregnant woman at any given time.

QUESTION #2

Question (max 240 characters)

• 2)What are the 3 most common mechanisms that can be altered in placenta cells by endocrine disrupting chemicals (EDC)?

Check correct answer

• a)ER stress, transporters and cytokine signaling
• b)Cytokine signaling, miRNA and HSD-3β1
• c)CY-P450, signaling pathways and molecular targets
• d)Transporters, miRNA and ER stress

Explain why this is the correct answer (max. 2500 chars):

Most of the studies report changes in steroidogenesis through altered CY-P450 activity and abundance (bisphenols, organophosphates, PCBs, DDT, phthalates and organotins). Other commonly reporter changes include altered signaling pathways (like JNK and MAPK) by bisphenols, organophosphates, and PCBs, and molecular targets (like COX, NFkB, Nrf2) by PBDEs, organophosphates and phthalates.

QUESTION #3

Question (max 240 characters)

• 3)What are the gaps in knowledge pertaining to the research of endocrine disrupting chemical (EDC) exposures and placenta development and function?

Check correct answer

• a)Studies with mixtures
• b)Environmentally relevant doses
• c)Sex-specific effects
• d)All of the above.

Explain why this is the correct answer (max. 2500 chars):

Chemicals in mixture can have both pharmacokinetic and pharmacodynamic effects on each other. This can include competition for shared receptors (competitive binding), or metabolic substrates. Additional studies are necessary to begin to understand the complexity of mixtures in the placental compartment.

Many studies investigating the effects of EDCs on placental function rely on relatively large or supraphysiological exposure doses to demonstrate a phenotype, but the use of more environmentally relevant exposure doses will better mimic real-world scenarios.

Despite evidence that some EDCs, such as phthalates, organophosphates and PBDEs can have sex-specific effects, this variable is often ignored in placental studies particularly in cell models where placental cell lines used are of unknown or male origin. This is important because sex can drive specific cellular responses.

QUESTION #4

Question (max 240 characters)

• 4)Which of the following are specific placental functions affected by phthalates? Check correct answer

Check correct answer

• a)Invasion
• b)Apoptosis
• c)Fatty acid homeostasis
• d)All of the above

Explain why this is the correct answer (max. 2500 chars):

In vivo studies have demonstrated that gene changes in the placenta following gestational phthalate exposure results in altered fatty acid homeostasis and apoptosis. In vitro studies have shown that phthalates, such as the DEHP metabolite MEHP, can induce apoptosis and inhibit extravillous trophoblast invasion. Additionally, in vitro effects observed from phthalate exposure includes altered lipid metabolism in both human choriocarcinoma (JEG-3) and rat trophoblast (HRP1) cells.

Outstanding questions.

• How can maternal chemical body burden be minimized, especially during vulnerable windows of exposure such as pregnancy?

• What are the potential pharmacokinetic and pharmacodynamic interactions between chemicals in mixture? Do these interactions affect placental health?

• If placental health is disrupted by exposure to an EDC, is the effect carried over transgenerationally?

• What are the underlying mechanisms driving sex-specific effects on the placenta, including sex-specific chemical accumulation, such as for organophosphates?

• What is the best approach to integrate epidemiologic associations and cause-and-effect animal studies on placenta-specific outcomes?

• Are the effects of EDC exposure observed in mice and rats representative of more relevant placental animal models like guinea pigs or non-human primates?

• Do EDC-induced effects on placental development observed following high exposure doses represent the effects observed at environmentally relevant doses?

Highlights.

• This review covers placenta-specific outcomes associated with exposure to nine major classes of known endocrine disrupting chemicals that cross the placental barrier and are present in human biomonitoring studies.

• Despite epidemiologic associations, direct cause and effect links between EDC exposure and the development of complex obstetric disorders, such as preeclampsia, remain a challenge in the field.

• Mice are the predominate animal model used for placenta-specific outcomes. More translatable placental mammal models, like guinea pig and non-human primate are underused.

• Both in vivo and in vitro toxicological studies focusing on placenta-specific outcomes often use supraphysiological doses.

• A limited number of studies dose chemicals in mixture. Additionally, studies that dose with mixtures use intra-class mixtures, not necessarily reflective of the human condition where both intra- and extra-class mixtures occur.

• In vitro studies are often limited by the use of commercially available cell lines, which originate from male or unknown sex placentas.

Glossary

Endocrine Disrupting Chemicals

Natural or man-made chemicals that can interfere with the endocrine system. This includes altered hormone production, secretion, and/or action

Invasion

Defines the ability of cells to become motile and, through enzymatic secretions, infiltrate into the extracellular matrix within a tissue. During placental development, extravillous trophoblasts (EVT) utilize this process to infiltrate into and remodel maternal uterine arteries

Placenta accreta

A pregnancy complication in which the placenta invades deeply into the maternal uterine wall and remains partially attached during parturition. This often results in severe post-partum hemorrhage

Placenta barrier

The placental cell layer in direct contact with the maternal blood. It is semipermeable, which allows transfer of nutrients such as glucose and fatty acids and prevents transfer of larger molecules like insulin. Expressing drug efflux transporters such as the ATP-binding cassette transporter ABCA1, the syncytium serves as a selective membrane to substances passing from maternal to fetal blood and vice versa

Pregnancy complications

Pathological processes associated with pregnancy. These can occur during or after pregnancy and range from minor discomfort to severe diseases that require medical intervention, such as preeclampsia, eclampsia, or placenta percreta

Syncytialization

The process of cellular fusion in cytotrophoblasts. Fused cytotrophoblasts form the outer-most layer of the placental villi known as the syncytium. The syncytium is hormonally active and represents the site of maternal and fetal gases and nutrients exchange