Environmental Endocrine Disruptors and Endometriosis

Abstract

As a consequence of industrialization, thousands of man-made chemicals have been developed with few undergoing rigorous safety assessment prior to commercial use. Ubiquitous exposure to these compounds, many of which act as endocrine-disrupting chemicals (EDCs), has been suggested to be one factor in the increasing incidence of numerous diseases, including endometriosis. Endometriosis, the presence of endometrial glands and stroma outside the uterus, is a common disorder of reproductive-age women. Although a number of population-based studies have suggested that exposure to environmental EDCs may affect a woman’s risk of developing this disease, results of epidemiology assessments are often equivocal. The development of endometriosis is, however, a process occurring over time; thus, a single assessment of toxicant body burden cannot definitively be linked to causation of disease. For this reason, numerous investigators have utilized a variety of rodent models to examine the impact of specific EDCs on the development of experimental endometriosis. These studies identified multiple chemicals capable of influencing physiologic processes necessary for the establishment and/or survival of ectopic tissues in rodents, suggesting that these compounds may also be of concern for women. Importantly, these models serve as useful tools to explore strategies that may prevent adverse outcomes following EDC exposure.

1. Introduction

Endometriosis is one of the most poorly understood conditions affecting the health of reproductive-age women. It is characterized by the presence of endometrial glands and stroma in areas outside of the uterus, most commonly within the lower abdomen and pelvis (rev by (Bruner-Tran et al. 2018; Bulun et al. 2019)). Endometriosis is considered a benign condition and can be asymptomatic; however, the majority of women with this disease experience chronic pelvic pain, dyspareunia, dysmenorrhea, and/or subfertility. Furthermore, endometriosis patients often present with one or more comorbidities including adenomyosis, adhesive disease, and inflammatory conditions such as interstitial cystitis and inflammatory bowel disease (Bruner-Tran et al. 2018). Multiple theories have attempted to explain the development of endometriosis; however, none can explain all incidences of the disease (Bulun et al. 2019). Furthermore, several lines of evidence suggest that endometriosis may be an autoimmune disorder since the disease is associated with both chronic inflammation and the expression of autoantibodies (Zhang et al. 2018). Our lack of understanding of the etiology of endometriosis has limited the development of effective treatment options for these women, often leading to minimally effective hormonal manipulations and/or surgical removal of diseased tissue. Unfortunately, disease recurrence after treatment, including surgical resection, is common.

Despite our limited understanding of the early mechanisms associated with the initiation of endometriosis, it is well-established that ectopic lesion growth is both an estrogen-dependent process (Bulun et al. 2019) and an inflammatory disorder (Patel et al. 2018). Furthermore, the majority of women with endometriosis exhibit some degree of endometrial progesterone resistance, which contributes to subfertility and pregnancy loss (Bruner-Tran et al. 2002, 2010; Yilmaz and Bulun 2019). Equally important is the fact that hormones influence inflammatory response (Straub 2014); thus, alterations in response to these steroids would likely have far-reaching effects in multiple systems. For these reasons, it is appropriate to explore a potential role of exposure to environmental endocrine disruptors (EDCs) in the development of this common and debilitating disease.

2. Environmental Endocrine-Disrupting Chemicals

The endocrine system is responsible for the regulation of the numerous hormones that control essential life processes. For instance, insulin regulates the metabolism of carbohydrates, fats, and proteins and ensures the storage of excess fuels not immediately required. The sex hormones (estradiol, progesterone, and testosterone) regulate reproduction and the development of primary and secondary sex characteristics. Endocrine-disrupting chemicals (EDCs) are exogenous chemicals that interfere with the endocrine system by mimicking or inhibiting the action of one or more of these endogenous hormones. More specifically, endocrine disruptors interfere with the synthesis, metabolism, and/or action of hormones, leading to the dysregulation of normal physiological processes and potentially promoting the development of disease (Schug et al. 2011). Importantly, many EDCs also dysregulate the immune system (rev by (Mokarizadeh et al. 2015; Rogers et al. 2013); thus, compounds with both endocrine and immune disrupting properties are of particular concern with regard to the development of endometriosis.

Linking any human disease to a specific chemical exposure is difficult; however, EDCs are ubiquitous in the environment and include both organic and synthetic compounds. Many EDCs are also classified as persistent organic pollutants (POPs), a broader class of environmental contaminants that do not degrade within the environment quickly. Most EDCs are lipophilic, facilitating their ability to penetrate the lipid bilayer of the plasma membrane and bioaccumulate in adipose tissue. Importantly, following an initial exposure, adipose tissue can slowly release these EDCs into the bloodstream, which can lead to long-term adverse health effects. The lipophilic nature of EDCs also allows these chemicals to bioaccumulate and bio-magnify within the food chain; thus, the primary source of human and animal EDC exposure is ingestion of contaminated foods.

Significantly, meta-analyses have described a link between exposure to a variety of EDCs and altered reproductive potential in humans (Jeseta et al. 2019; Levine et al. 2017). Numerous human epidemiology studies have explored the relationship between the adult body burden of a variety of EDCs with a concurrent diagnosis of endometriosis. While multiple studies suggest a link between EDC exposure and the risk of a woman developing endometriosis, other studies have failed to identify such a link (rev by (Bruner-Tran et al. 2010; Bruner-Tran and Osteen 2010; Smarr et al. 2016)). Clearly, despite their unquestionable value, human epidemiology assessments can only provide population associations and are not designed to identify specific mechanisms of disease development. In order to address this issue, we must first identify which EDCs are capable of disrupting key physiologic processes necessary for maintaining reproductive health. EDCs can be found in many classes of compounds, such as pharmaceutical agents, plasticizers, pesticides, and herbicides. Herein we will review evidence linking specific EDCs to reproductive tract dysfunction, suggesting that they also have the potential to promote the development of endometriosis (Table 1).

Table 1.

Human health effects of selected endocrine disruptors

Endocrine-disrupting chemical	Common exposure sources	Endometriosis risk assessment	Other health effects	References
Diethylstilbestrol (DES)	Medication prescribed to pregnant women between 1940 and 1970 to reduce risk of adverse outcomes	Some studies indicate a slight increased risk of endometriosis in women exposed to DES in utero	Increased incidence of vaginal adenocarcinoma, infertility, and early menopause in daughters and granddaughters of DES-treated women Sons and grandsons at risk of hypospadias, infertility, and certain cancers	Al Jishi and Sergi (2017)
Bisphenol A (BPA)/bisphenol S (BPS)	Plastics (e.g., water bottles and canned food liners)	Association of endometrioma with increased urinary BPA concentration	Alterations in fetal development (most commonly linked to birth weight, behavior disorders in children, reduced fertility in adults) (males and females)	Matsunawa et al. (2009) and Rashidi et al. (2017)
Phthalates	Plastics, cosmetics (e.g., nail polish, hair spray, and shampoo), food packaging materials	Numerous studies report an association between phthalate exposure and endometriosis	Associated with hormone-regulated diseases and infertility (males and females)	Upson et al. (2013), Buck Louis et al. (2013), FDA US (2019), and Nazir et al. (2018)
Polychlorinated biphenyls (PCB)	Production banned in the USA in 1979, but environmental contamination persists. Waste incineration	Epidemiology studies inconsistently link exposure to dioxin-like PCBs and endometriosis	Probable human carcinogen, endocrine and immune disruption	Rev in Bruner-Tran and Osteen (2010)
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD or dioxin)	Fossil fuel burning, paper production, forest fires, waste incineration	Epidemiology studies inconsistently link exposure to TCDD to endometriosis	Chloracne, probable human carcinogen, teratogen, possible increased risk of adverse birth outcomes (stillbirth, miscarriage, prematurity)	Rev in Bruner-Tran and Osteen (2010)

Perhaps the most notorious pharmaceutical EDC is diethylstilbestrol (DES). DES is a synthetic, highly potent estrogen that was first manufactured in 1938 (Darbre 2017; Harris and Waring 2012). Soon after, pregnant women at risk of miscarriage or preterm birth were prescribed DES, but DES was soon recommended to all pregnant women to prevent miscarriage. The FDA banned the use of DES during pregnancy in 1971 because the drug was ineffective in preventing pregnancy loss and increased the incidence of a rare vaginal cancer (Harris and Waring 2012). Furthermore, at least two studies reported an association between maternal DES exposure and the development of endometriosis in their daughters (Upson et al. 2015a; Giuseppe Benagiano and LIppi 2015; Brosens et al. 2014).

Although the medical use of DES has been discontinued, commercial use of plasticizers, many of which act as EDCs, persists. Of significant concern are the estrogen-like bisphenol A (BPA), bisphenol S (BPS), and the phthalates (Darbre 2017). BPA, BPS, and phthalates are used in the processing of polycarbonate plastics and epoxy resins due to their cross-linking properties (Darbre 2017). The broad commercial utility of these chemical resulted in their ubiquitous expression in plastic products such as water bottles, dental sealants, the lining of water pipes, canned food liners, and many children’s toys (Quitmeyer and Roberts 2007). Interestingly, a number of studies suggest exposure to phthalates is linked to the onset of hormone-regulated diseases, including endometriosis, in reproductive-age women (Upson et al. 2013, 2015b). BPA was found to affect developing fetuses and infants (Quitmeyer and Roberts 2007) and is associated with alterations in adolescent and adult behavior (Inadera 2015). Although human data linking BPA/BPS exposure to endometriosis is limited, reproductive tract alterations following developmental exposures are well-established (Minguez-Alarcon et al. 2016; Ziv-Gal and Flaws 2016) and deserve further consideration.

Finally, a number of commonly used pesticides, insecticides, and herbicides contain polyhalogenated aromatic hydrocarbons (PHAH). PHAHs are a family of natural and man-made planar polycyclic organic compounds containing a variable number of halogen atoms (e.g., chlorine). This family includes (i) polychlorinated dibenzodioxins (PCDD), (ii) dibenzofurans (PCDF), and (iii) the biphenyls (PCB). Many members of these classes of compounds have been found to disrupt normal reproductive function in a variety of animal species, including humans (Birnbaum and Tuomisto 2000). Among the family of PHAHs, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), also referred to as dioxin, is perhaps the most well studied with regard to endometriosis. TCDD is lipophilic in nature and is a by-product of various industrial processes including paper bleaching and the burning of fossil fuels. Interestingly, TCDD is also a by-product of natural processes, such as forest fires and volcanic eruptions, making the chemical both a natural toxin and a man-made toxicant depending on its source.

A potential role of TCDD exposure in the development of endometriosis was first reported 1993 following the identification of this disease in a colony of toxicant-exposed rhesus macaques (Rier et al. 1993). Since that initial report, multiple human epidemiology studies have attempted to link exposure to TCDD and/or other EDCs to the presence of endometriosis (Heilier et al. 2005; Eskenazi et al. 2002; Bois and Eskenazi 1994). Unfortunately, these studies failed to reach a consensus, most likely due the variable toxicity of individual congeners at different stages of life and the inherent complexity of definitively determining the absence of endometriosis within a control population.

3. EDC Mechanism of Action

A number of EDCs, including TCDD and the dioxin-like PCBs, mediate their effects via binding to the aryl hydrocarbon receptor (AhR), an orphan nuclear receptor that is a member of the basic helix-loop-helix/Per-Arnt-Sim family of dimeric transcription factors (Bock and Kohle 2006; Burbach et al. 1992; Humphrey-Johnson et al. 2015). The AhR is expressed throughout the male and female reproductive tract (Hernandez-Ochoa et al. 2009; Jiang et al. 2010; Vezina et al. 2009) and contributes to reproductive function as exemplified by significant reduction in fertility in AhR knockout mice (Peters et al. 1999). The AhR is also broadly expressed by cells of the immune system, and its activation by endogenous ligands is critical to maintaining immune homeostasis as well as defending against infection (Stockinger et al. 2014). In its inactive form, AhR localizes to the cytoplasm with nuclear translocalization prevented by an inhibitory complex formed by the chaperone proteins HSP90 and XAP2 (Noakes 2015). Following ligand binding and activation, the AhR dissociates from the chaperone proteins, revealing the receptor’s nuclear localization signal (NLS). Once the NLS is exposed, AhR is translocated to the nucleus where it forms a heterodimer with the aryl hydrocarbon receptor nuclear transporter (ARNT) (Humphrey-Johnson et al. 2015).

Within the nucleus, the AhR/ARNT heterodimer acts as an active transcription factor that interacts with the xenobiotic response element within the promoter sequence of numerous genes (Humphrey-Johnson et al. 2015; Noakes 2015). Importantly, the impact of AhR activation is dependent upon the ligand to which it binds (Larigot et al. 2018). Ligand binding has been suggested to influence the conformational structure of the bound receptor, thereby altering its transcriptional activities (Fujii-Kuriyama and Kawajiri 2010; Denison and Faber 2017). AhR/ARNT dimers interact with xenobiotic response elements (XRE) and induce transcription of P450 genes, such as CYP1A1 (Fujii-Kuriyama and Kawajiri 2010; Matsunawa et al. 2009; Mimura and Fujii-Kuriyama 2003). Depending on the ligand, AhR also transcribes other members of the cytochrome P450 superfamily such as CYP2A1 and CYP24A1 (Matsunawa et al. 2009). These cytochrome P450 enzymes act to mediate the metabolism of xenobiotics, such as pharmaceutical drugs and environmental contaminants (Androutsopoulos et al. 2009). It should be noted that TCDD is largely resistant to metabolic degradation, which likely contributes to its high level of toxicity (Gasiewicz et al. 2008). In addition to contributing to the detoxification and inactivation of EDCs, the AhR also regulates the expression of enzymes necessary for the metabolism of endogenous substances, including some hormones. In this respect, the AhR participates in the modulation of a wide range of cellular processes, including proliferation, differentiation, and apoptosis (De Coster and van Larebeke 2012). Not surprisingly, inappropriate activation of the AhR by exogenous ligands can have wide-ranging detrimental effects that may promote diseases (Bock and Kohle 2006) such as endometriosis (Bulun et al. 2000).

4. Endometriosis: Adult Toxicant Exposure Models

Epidemiology studies have identified a potential role of EDCs in the development of endometriosis in women, but association studies cannot definitively determine cause and effect. Thus, development of animal models in which specific compounds can be tested under controlled settings has been critically important to further evaluate a possible link between EDCs and reproductive disease (Table 2).

Table 2.

Rodent models of EDC exposure and experimental endometriosis (selected studies)

Model description	EDC exposure	Endometriosis outcome	Other health effects	Reference
Surgical model using autologous tissues (rat vs. mouse)	Animals were treated with TCDD prior to induction of disease and until euthanasia. Dose-response and time-course experiments were conducted	Increased disease burden in both rats and mice, which was greatest in mice	Changes in animal weight, altered ovarian function, and thymic atrophy were observed in some animals	Cummings et al. (1996)
Surgical model using autologous tissues (mouse)	Animals were treated with multiple doses of PCB126, PCB153, 4-PeCDF, 1,3,6,8-TCDD, or 2,4,7,8-TCDD prior to induction of disease and until euthanasia	4-PeCDF and most doses of 2,4,7,8-TCDD were associated with increased disease burden. No effect of PCB126, PCB153, 1,3,6,8-TCDD, or highest dose TCDD on lesion growth was observed	Ovarian toxicity was observed for PCB126 and the highest dose of 2,4,7,8-TCDD	Johnson et al. (1997)
Surgical model using autologous tissues (rat)	After surgical induction of endometriosis, rats were exposed to HCB (1, 10, and 100 mg/kg)	Dose-dependent effects were noted on lesion volume, MMP expression, VEGF, cyclooxygenase, and ESR/PGR expression	Numerous alterations were noted in the eutopic endometrium, consistent with endometriosis (i.e., altered MMP expression and reduced PGR expression	Chiappini et al. (2019)
Developmental TCDD exposure model (nonsurgical)	Dam (F0) exposed to TCDD, and fetus (F1) and germ cells (F2) are exposed concurrently	Endometriosis-like phenotypes such as reduced progesterone response, subfertility, preterm birth in F1-F4.	F1-F3 offspring exhibit increased sensitivity to inflammatory challenge (LPS, GBS) Increased pup/adult mortality and increased risk of adenomyosis	Bruner-Tran et al. (2018), Nayyar et al. (2007), Bruner-Tran and Osteen (2011), Bruner-Tran et al. (2016), and Ding et al. (2011)
Developmental TCDD exposure model (surgical)	In utero TCDD exposure followed by surgically induced endometriosis and secondary dioxin exposure	Prenatal plus adult TCDD exposure was associated with larger lesions compared to adult only exposure	Increased sensitivity to the secondary dioxin exposure	Cummings et al. (1999)

To examine the impact of an adult EDC exposure on the progression of endometriosis, Birnbaum and colleagues (Cummings et al. 1996) conducted a series of studies using a syngeneic experimental model in mice. In the first set of studies, they examined growth and progression of surgically established experimental disease in response to TCDD. Adult Sprague Dawley rats and B6C3F1 mice were treated with TCDD prior to and after surgical induction of endometriosis. In both species, exposure to TCDD was associated with larger lesions compared to untreated animals; however, the effects were greater in mice versus rats (Cummings et al. 1996). A separate study from this same research group similarly compared TCDD to dioxin-like PCBs and non-dioxin-like PCBs. This study found that lower doses of TCDD (1 and 3 ug/kg bw) and 4-PeCDF (100 ug/kg bw) significantly enhanced the growth of experimental endometriosis in mice, while PCB 126 had no impact on lesion growth. Interestingly, the non-dioxin-like compounds tested, PCB 153 and 1,3,6,8-TCDD, also had no impact on the experimental disease, supporting the hypothesis that TCDD and dioxin-like PCBs primarily promote endometriosis via AhR binding (Johnson et al. 1997).

Huang et al. (2017) also examined the dioxin-like PCB126 and the non-dioxin-like PCB153 in a mouse model of endometriosis but with slightly different results than the study described above (Johnson et al. 1997). In the more recent study, endometrial tissues were obtained from control balbc mice; tissues were minced and injected intraperitoneally into recipient mice that had received either PCB153 or PCB126 2 days prior to injection. An additional group of tissue recipients were treated only PBS. Mice were euthanized up to 20 days after tissue injection. In contrast to the study from the Birnbaum laboratory, described above, this study revealed that the dioxin-like PCB126, which binds the AhR, promoted the development of experimental disease. Furthermore, this study revealed that this dioxin-like PCB was found to induce both E2 synthesis and the expression of inflammatory cytokines at ectopic sites, which likely contributed to the progression of experimental disease. The differing outcomes between the two studies may reflect the different mouse strains utilized and/or choice of doses. Nevertheless, both studies found the non-dioxin-like PCB153 to have no effect on the development of experimental endometriosis (Johnson et al. 1997; Huang et al. 2017).

Using a chimeric experimental mouse model, in which human endometrial tissues are deposited into the peritoneal cavity of immunocompromised mice, our group previously examined the impact of acute TCDD exposure on ectopic lesion establishment. Specifically, proliferative phase endometrial biopsies from women without endometriosis were established in vitro as organ cultures and treated with estradiol (E2), E2 plus progesterone (EP), and both steroid groups with TCDD (10 ug/mL). Tissues were maintained in culture for 24 h and then washed in PBS and injected into mice. Disease readily developed in mice receiving E2-treated tissues, while EP treatment inhibited disease. In contrast, co-treatment with TCDD led to larger, more highly vascularized lesions regardless of the additional presence of steroids (Bruner-Tran et al. 1999).

More recently, Burns and colleagues examined the impact of BPA and the BPA alternative, BPAF, on the development of experimental endometriosis in a syngeneic mouse model (Jones et al. 2018). Using low-level exposures, the investigators reported that while both compounds enhanced disease development, effects were greatest following exposure to BPAF, the BPA alternative. Interestingly, both compounds were found to similarly disrupt ovarian signaling and lowered progesterone synthesis, which would negatively impact fertility.

5. Developmental EDC Exposure Models and Endometriosis

Studies described above along with numerous others suggest that acute exposure to EDCs such as TCDD is capable of promoting experimental endometriosis in animal models. Nevertheless, compared to adult exposures, it is well-established that the negative effects of EDCs are greater when exposures occur during development (Meeker 2012). Unfortunately, as a consequence of the maternal diet, exposure to EDCs begins in utero for most people, and thus, consideration of the potential role of a developmental EDC exposure on the adult risk of endometriosis is appropriate. Furthermore, exposure of a pregnant woman to EDCs concomitantly exposes the developing fetus and the germ cells present within the fetus, which have the potential to become the next generation (Fig. 1). Effects observed in the mother (F0), fetus (F1), or grandchild (F2) following exposure during pregnancy are considered multigenerational effects. Effects occurring in subsequent generations (F3 and beyond) are considered transgenerational effects since these generations are not directly exposed to the toxicant.

Fig. 1. Maternal exposure affects three generations simultaneously.

Exposure of a pregnant woman, or any other mammal, to substances such as EDCs directly exposes not only the mother but also her fetus. Present within the fetus are germ cells, which have the potential to become her grandchildren. In this context the pregnant woman is considered the “founding generation” or F0. The fetus she carries is the F1 generation, while the germ cells represent the potential F2 generation. For this reason, chemical or toxicant exposure of a pregnant woman is termed “multigenerational” exposure

Although multigenerational effects of EDCs have been described in humans, transgenerational effects are more difficult to demonstrate since prospective studies cannot be conducted. However, numerous experimental studies support the occurrence of transgenerational effects of EDCs in rodents (rev by (Skinner 2014; Nilsson et al. 2018)). As will be discussed below, examining the potential for multi- and transgenerational effects of toxicants is an active area of research (Meeker 2012). To this end, numerous investigators have developed neonatal exposure models in rodents using various toxicants in order to explore the role of early life exposures on the adult development of endometriosis or endometriosis-like phenotypes.

One of the earliest studies examining the potential role of developmental toxicant exposure and endometriosis was conducted by the Birnbaum laboratory. Birnbaum and colleagues exposed both pregnant rats and mice to TCDD on embryonic day 8 (E8) and then surgically induced endometriosis in adult offspring followed by a second exposure to TCDD. This study confirmed their previous report, discussed above, utilizing acute exposures, which demonstrated murine models of experimental endometriosis to be more responsive to TCDD compared to rats. More significantly, their later study also found that the adverse effects of an adult exposure were enhanced if the animal had also been subjected to a developmental exposure (Cummings et al. 1999).

Signorile and colleagues exposed balbc mice to the EDC bisphenol A (BPA) throughout pregnancy and continuing until PND7 in offspring (Signorile et al. 2010). Mice were euthanized at 3 months, and reproductive tissue, the peritoneal cavity, and selected tissues were examined. Significantly, following prenatal exposure to BPA, 35% of mice were found to have endometriosis-like structures in the adipose tissues surrounding the uteri. Histological examination of lesions identified the presence of both endometrial glands and stroma, which were found to express estrogen receptors. The marked similarity of the lesions present in BPA-exposed mice and endometriosis in women provides support for the potential role of activation of embryonic cell rests or stem cells by EDCs in the development of endometriosis.

In the absence of known levels of EDC exposures, women with endometriosis frequently exhibit reduced endometrial response to progesterone, which likely contributes to infertility, a common comorbidity of this disease. In order to examine whether developmental exposure to TCDD alters adult uterine response to progesterone, we exposed pregnant C57BL/6 to a single toxicant dose on E15.5, thereby exposing not only the dam (F0) and the fetus (F1) but also the germ cells present in the fetus that have the potential to become the F2 generation (Fig. 1). We found that, similar to women with endometriosis, adult female offspring (F1 mice) of pregnant mice exposed to TCDD exhibit reduced uterine responsiveness to progesterone (Nayyar et al. 2007). Additional studies revealed that mice exhibiting reduced progesterone response, which we refer to as an “endometriosis-like phenotype,” are frequently infertile, while nearly half those able to become pregnant deliver preterm (Bruner-Tran and Osteen 2011). Importantly, the endometriosis-like phenotype persisted in both the F2 and F3 generations (Bruner-Tran and Osteen 2011). As stated above, this is a significant finding since it is the F3 generation that is the first that was not directly exposed to TCDD. A follow-up study revealed that adenomyosis was also common in F1–F3 females, which was associated with dysregulated expression of estrogen receptor isoforms (Bruner-Tran et al. 2016). Adenomyosis, the presence of endometrial glands and stroma within the myometrium, is a common comorbidity of women with endometriosis; thus the presence of spontaneous adenomyosis in our TCDD-exposed mice and their progeny supports examination of a potential role of EDC exposure in the development of endometriosis in women.

6. Potential Mechanisms of EDC Exposure and Endometriosis Development

In 1927, Sampson (1927) proposed his “retrograde menstruation” theory of endometriosis development which was widely embraced. Ectopic establishment of refluxed menstrual tissue cannot account for all occurrences of the disease (i.e., endometriosis outside the pelvic cavity or the rare occurrence in men); leading to the development of additional theories. Perhaps the most well regarded of these later theories are the activation of Mullerian cell rests or activation of stem cells (either of uterine or extrauterine origin) (Baranov et al. 2018). It is likely that the development of endometriosis is multifactorial, and a woman’s individual risk of developing endometriosis will be related to a variety of factors, including her genetic and epigenetic background (Bulun et al. 2019) and the robustness of her immune system (Baranov et al. 2018). Experimental studies demonstrate that a developmental exposure to EDCs is associated with dysregulated immune function in adulthood (Post et al. 2019). However, the potential contribution of toxicant exposure to altered immune function related to a woman’s risk for developing endometriosis remains unclear. Nevertheless, pro-inflammatory cytokines are known to be elevated within the peritoneal fluid of patients with endometriosis (Bruner-Tran et al. 2013; Dull et al. 2019), and chronic peritoneal inflammation can contribute to both dysregulation of the eutopic endometrium and establishment of ectopic lesions (rev by (Bruner-Tran et al. 2013)).

Progesterone has profound anti-inflammatory effects within and outside the reproductive tract (Tibbetts et al. 1999); thus, disrupting the ability of the endometrium to respond to this steroid would be expected to lead to a heightened inflammatory response, potentially contributing to disease development (Bruner-Tran et al. 2013). Within the eutopic endometrium of a healthy woman, cell-cell communication is “stromal dominant,” whereby stromal cells direct the behavior of adjacent epithelial cells. In women with endometriosis, however, the normal stromal dominant pattern of communication is disrupted, resulting in an epithelial dominant signaling pathway that mimics a wound-like phenotype. This wound-like pattern of communication promotes local inflammation, which is a well-known driver of reduced progesterone responsiveness (Guo 2007; Patel et al. 2017). Thus, it is perhaps not surprising that environmental exposures that drive inflammatory phenotypes have also been found to reduce progesterone responsiveness (Bruner-Tran and Osteen 2011; Igarashi et al. 2005; Resuehr et al. 2012).

Following retrograde menstruation, which occurs in most (if not all) cycling women, refluxed endometrial fragments are readily recognized and cleared from the peritoneal cavity by a healthy innate immune system. In women with endometriosis, however, alterations within the peritoneal microenvironment, such as aberrant immune cell function and increased production of pro-inflammatory signals, may act to promote the survival of displaced tissue. Following attachment to ectopic sites, ectopic tissue survival requires a rapid acquisition of a vascular supply. Compared to similar endometrial tissues acquired from control donors, endometrial tissues acquired from women with endometriosis are more readily able to establish a vasculature (Bruner-Tran et al. 2008). Since estrogen receptor signaling is critical to regulating endometrial proliferation, angiogenesis, and inflammatory signaling, EDCs acting as either estrogen receptor agonists or antagonists may influence establishment or progression of endometriosis (Patel et al. 2018, 2017).

Once established, the presence of endometriotic lesions likely amplifies the inflammatory peritoneal microenvironment which may subsequently affect the eutopic endometrium by further repressing progesterone responsiveness. Thus, it is readily apparent that multiple mechanisms associated with the development of endometriosis as a consequence of retrograde menstruation are susceptible to modulation by EDCs. It is equally plausible that exposure to EDCs act to promote endometrial differentiation of embryonic cell rests or displaced uterine stem cells (Susheelamma et al. 2018). If we accept the possibility that exposure to EDCs can promote the development of endometriosis, the next step is to identify potential intervention strategies.

7. Modulating the Impact of EDC Exposure

Clearly, it is not possible to prevent past exposures to EDCs that occurred during development or in our preceding generations; however, steps can be taken to reduce ongoing exposures (Table 3). Since the majority of EDC exposure occurs via diet, choosing organic foods, lean meats, or a vegetarian lifestyle can each help to minimize exposure. Additionally, reducing usage of canned foods containing a BPA liner, using BPA/BPS-free products, and avoiding either long-term storage or heating of foods in plastic containers will also reduce incidental EDC exposure. Although taking these precautionary steps after the development of endometriosis will not reverse the disease, limiting ongoing exposures may still be beneficial. Many EDCs bind both the AhR and estrogen receptor α (ESR1); these dual effects strongly suggest that specific EDCs may promote not just the development of endometriosis but also progression of the disease.

Table 3.

Personal choices for reducing EDC exposure

Reducing dietary exposures	Reducing environmental exposures	Additional precautions prior to and during pregnancy
Use BPA/BPS-free products	Do not smoke; limit exposure to second-hand smoke	Reduce EDC exposures of both sexes prior to seeking pregnancy
Avoid storing plastic water bottles at high temperatures (e.g., trunk of car); do not microwave plastic containers	Avoid standing near idle cars; do not jog in congested areas	Avoid wood fire exposure and limit chargrilled meats
Drink filtered water rather than bottled or tap water	Vacuum frequently to reduce dust (which can contain EDCs)	Do not use permanent hair dyes
Reduce usage of nonorganic canned foods and plastics for food storage	Limit use of pesticides and herbicides	Avoid eating predator fish, but do consume fish with high levels of omega-3 fatty acids
Reduce consumption of nonorganic dairy and meat. Opt for wild (not farmed) seafood	Do not use soaps/toothpastes/hand sanitizers containing triclosan	Let someone else paint the nursery (paint fumes can contain EDCs)
Thoroughly rinse produce; purchase organic when possible	Opt for fragrance-free cosmetics and lotions	Limit maternal exposures to chemical cleaners or use natural products

Treatment strategies designed to manipulate the steroid environment are both limited in effectiveness and frequently associated with unacceptable side effects. For this reason, anti-inflammatory agents are an attractive therapeutic option to explore. Furthermore, since many women with endometriosis wish to preserve their fertility or to actively pursue pregnancy during treatment, dietary agents with anti-inflammatory properties are being explored for their utility in treating this disease without jeopardizing pregnancy or fetal development. Resveratrol, for example, is a polyphenolic compound and natural phytoestrogen with anti-inflammatory effects and antiproliferative activity. Resveratrol is found in grapes, red wine, soy, berries, and stilbenes and is available in high doses in over-the-counter supplements.

Due to its anti-inflammatory effects, several studies have examined the potential therapeutic benefits of resveratrol for the treatment of endometriosis using animal models. Using the chimeric model of experimental endometriosis described above, we previously examined the ability of resveratrol to inhibit lesion establishment. Specifically, proliferative phase endometrial tissues were obtained from healthy donors and established as ectopic lesions in nude mice following intraperitoneal injection. Mice were treated with 17β-estradiol alone or with resveratrol (6 mg/mouse) for up to 20 days beginning 1 day after tissue injection. The gross size and number of ectopic lesions was evaluated at necropsy and recovered tissues evaluated for proliferative activity and apoptosis. We found that resveratrol decreased the number of endometrial implants per mouse by 60% (P < 0.001) and the total volume of lesions per mouse by 80% (P < 0.001). Not surprisingly, resveratrol treatment was also associated with decreased proliferation and increased apoptosis of ectopic tissues (Bruner-Tran et al. 2011).

In a similar study, Amaya et al. (2014) utilized rag2g(c) double knockout mice (a more severely immunocompromised strain compared to nude mice) as recipients of human endometrial tissue fragments. Tissues were placed subcutaneously in mice treated with estradiol with and without resveratrol for 30 days. In contrast to our study, this study found no difference in lesion weight; however, they did observe reduced cell proliferation in lesions from the mice receiving the highest dose of resveratrol. The subcutaneously placed tissues do not have the same intimate contact with the peritoneal environment nor vascular interaction as peritoneal lesions which may have led their reduced responsiveness to treatment. These studies utilizing human endometrial tissues growing as ectopic lesions in immunocompromised mice nonetheless demonstrate promise for resveratrol to also inhibit lesion establishment and growth in women.

Importantly, data from our laboratory and that of others suggest that an inflammatory environment may precede development of disease (rev by (Bruner-Tran et al. 2013; Burney and Giudice 2012; Maia Jr. et al. 2012). Thus, anti-inflammatory therapies such as resveratrol are a logical avenue to pursue for both prevention and treatment. Furthermore, in addition to its anti-inflammatory effects, resveratrol binds the AhR and, depending on dose and other factors, can exhibit both agonist and antagonist effects (Revel et al. 2003). Indeed, recent clinical trials reported improvement in pain scores in women with endometriosis treated with resveratrol compared to placebo (rev by (Dull et al. 2019)). Since it is difficult to determine if the development of endometriosis in women is related to prior EDC exposure, the use of well-controlled animal models can be valuable.

To this end, in order to explore the impact of resveratrol therapy on the uterine phenotype of adult mice with a developmental history of TCDD exposure, we recently treated adult F1 females with resveratrol (6 mg daily by gavage) or vehicle only for 4 weeks prior to mating. As shown in Table 4, resveratrol treatment prior to pregnancy not only improved fertility but eliminated the risk of preterm birth. Although we have not yet assessed the molecular mechanisms associated with this improvement (i.e., uterine progesterone receptor expression), these data suggest that nutritional anti-inflammatory agents can improve uterine function, which may in turn lead to a reduction in development of disease.

Table 4.

Resveratrol and the prevention of endometriosis-associated phenotypes

Maternal history	Pregnancy	p	Pregnancy outcome		p
			Term (%)	Preterm (%)
Control	15/15 (100%)		100	0
Control+RES	5/5 (100%)		100	0
F1	12/30 (40%)	<0.0001a	58	42	0.004a
F1 + RES	7/10 (70%)	NS	100	0	NS

^acompared to control

NS = not significant; compared to Control+RES

8. Discussion

Although endometriosis has historically been considered an idiopathic disease, recent data suggest that the risk of developing this disease, at least for some women, may be initiated within the fetal environment. The concept of “fetal origins of adult disease” (now more commonly termed “developmental origins of health and disease (DOHaD)” was initially described by Dr. David Barker following the development of metabolic syndrome in adults who experienced maternal malnutrition in utero (Roseboom et al. 2001; Barker 2004). Barker suggested that the fetus exhibits “plasticity” and, taking cues from the fetal environment, develops in accordance to those cues. If the external environment matches the internal environment, the baby is likely to develop into a healthy adult. However, if the environments differ greatly, it is hypothesized that the risk of developing disease in adulthood would be enhanced.

The concept of DOHaD was initially described with regard to maternal malnutrition and adult-onset metabolic syndrome; however, this concept has expanded exponentially and now considers that maternal stress, toxicant exposures, and other factors (both positive and negative) also contribute to fetal programming (Hsu and Tain 2019; Codagnone et al. 2019; Heindel 2006). For example, increased maternal stress is associated with the increased release of stress hormones/glucocorticoids which lead to increased fetal glucocorticoid loads. The increase in fetal hormone load resulting from maternal stress can result in alterations in fetal growth, organ structure, metabolism, and epigenetics and potentially impact the risk of disease in adulthood (Mandy and Nyirenda 2018). Animal studies have clearly demonstrated that developmental EDC exposure can influence adult physiology and promote the development of disease; thus, it is likely that humans are also sensitive to such exposures.

Indeed, accumulating evidence over the last decade strongly suggests a link between EDC exposure and the subsequent development of endometriosis and multiple comorbidities (subfertility, preterm birth, adenomyosis). Experimental evidence demonstrates that developmental exposure to EDCs can lead to transgenerational adverse health consequences, many of which have been linked to endometriosis. Although such a concept is difficult to prove in humans, evidence from accidentally exposed populations suggests the occurrence of transgenerational effects is likely (rev by Bruner-Tran et al. 2019). Moving forward, as we continue our efforts to pursue development of better therapies for women with endometriosis, it is necessary to additionally consider strategies focused on prevention. Furthermore, we must develop a better understanding of the possible contribution of EDC exposure on the incidence of endometriosis and related inflammatory comorbidities that may impact reproductive health across generations.