Defining an endocrine disruptor

Over half a century has passed since the first observation that environmental pollution can potentially represent a big threat to public health. Specifically, in 1962, Rachel Carson underlined the adverse effects of pesticides (dichlorodiphenyltrichloroethane -DDT) on sexual development and reproduction, postulating that “as the tide of chemicals born of the Industrial Age has arisen to engulf our environment, a drastic change has come about in the nature of the most serious public health problems” . The “diethylstilbestrol (DES) catastrophe” followed almost a decade later, in which prenatal use of diethylstilbestrol (DES), an estrogen-based drug that was thought to prevent miscarriage, was linked with vaginal adenocarcinoma in a series of patients, establishing the concept of “hormone disruption” as an etiologic parameter for human health perturbations .

However, it was in 1991, when a diverse group of 21 experts assembled at the Wingspread Conference Center in Racine, Wisconsin, USA (July 26–28, 1991) and introduced for the first time in the scientific community the terms “endocrine disruption” and “endocrine disruptor”, opening up a pivotal chapter in the field of contemporary Environmental Endocrinology . As years came by, mounting evidence was accumulating regarding EDCs’ effects on multiple endocrine systems, including thyroid, reproduction, neurodevelopment, obesity, and metabolism. In particular, the coinciding rapid increase in the incidence of various endocrine diseases (obesity, diabetes mellitus, thyroid disorders) during the past decades has strengthened the belief that EDCs are their potential environmental drivers.

EDCs are now recognized as one of the most severe threats to public health, potentially emerging as one of the leading environmental risks globally . Subsequently, various scientific societies, non-governmental organizations, and governmental agencies, such as the Endocrine Society , the International Federation of Gynecology and Obstetrics , World Health Organization (WHO), the United Nations Environment Programme (UNEP), and the American Academy of Pediatrics are acknowledging the role of EDCs and issuing extensive reports concerning their adverse effects in the human body.

According to them, EDCs are defined as exogenous chemicals or mixtures of chemicals that interfere with any aspect of endogenous hormonal signaling, affecting production, release, transport, cellular metabolism, binding action, and elimination of hormones that are present in the body and are responsible for homeostasis, reproduction, and developmental process. They represent a continuously expanding, highly heterogenous group of natural or man-made chemical compounds, originating from various, commonly encountered in everyday life sources, such as industrial solvents, plastics, plasticizers, storage containers, pesticides and fungicides, pharmaceutical agents, highly processed foods, cigarettes smoking, textiles and personal care products ( Fig. 1 ).

Schematic presentation of some of the most commonly encountered in our everyday life categories of EDCs and the main sources they can be found in.

Simultaneously, while the scientific community was broadening its knowledge regarding EDCs, over the last several years, a series of economic evaluations estimated the financial impact of EDCs in the frail healthcare system . Specifically, in a paper by Trasande et al., the total annual cost of all medical conditions probably attributable to EDCs in the European area was estimated to be 191 billion Euros , while when focusing on diabetes and obesity, attributed to EDC exposure associated costs were calculated to be over €18 billion per year . Although the aforementioned disease burden costs are considered by a share of scientists as highly speculative and an arbitrary estimation of the financial impact of EDCs , there is no doubt that EDCs burden global healthcare systems via their detrimental health consequences.

All the above have urged governments globally to restructure their regulations and policies regarding chemical substances and their environmental hazard, not only to improve citizens’ health via reduction or elimination of exposures—but also restrict the associated financial burden. In this context, recently European Commission has embraced a new strategic approach, based on the precautionary principle, aiming at the minimization of overall exposure to EDCs and the development of a thorough research basis for effective decision-making, via promoting dialogue and allowing all stakeholders to be heard . Although this is a positive step, regulatory bodies still have to put a lot of effort towards decreasing human exposure to EDCs and establishing global policies.

The fundamentals of endocrine disruptors’ mechanisms of action

Until today, endocrine disrupting properties have been recognized in over 1000 natural or man-made chemical compounds. Naturally occurring EDCs can be found in plants and can be introduced into the food chain through consumption by animals/humans (such as phytoestrogens). Synthetic EDCs have been widely utilized as drugs [diethylstilbestrol (DES)], pesticides [dichlorodiphenyltrichloroethane (DDT)], plastics [Bisphenol A (BPA)], plasticizers/dispersants [phthalates], or industrial solvents [polychlorinated biphenyls (PCBs), polybrominated biphenyls (PBBs)]. As the list of EDCs is continuously growing and their uses are expanding, the human body is becoming the recipient of a variety of environmental contaminants, via multiple routes, including air, water, food, and everyday consumer products. Subsequently, EDCs have been detected in various biological fluids in humans, such as sera, urine, amniotic fluid, and breast milk.

Although the concept of endocrine disruption is currently considered strong and validated, the underlying mechanisms whereby EDCs can induce their adverse effects in humans or the environment are only partially understood and remain a challenging scientific enigma. However, multidisciplinary research in endocrinology, ecotoxicology, toxicology, epidemiology, clinical research, epigenetics, and environmental sciences have established the fundamentals of their mode of action. In a recent consensus of an international group of experts, an effort was made to recognize the key characteristics of EDCs and reflect on how these can be used to identify, organize and utilize mechanistic data when evaluating chemicals as EDCs. Apart from identifying their intrinsic endocrine-disrupting hazard, this approach can serve as a universal framework not only for the scientific community but also for jurisdictions and regulatory agencies .

Following the principles of endocrine physiology, EDCs can exert their disrupting effects multifariously. More specifically, EDCs can directly perturb signaling pathways regulating steroid biosynthesis and/or metabolism of the human body, by interacting with, activating, or antagonizing nuclear and transmembrane hormone receptors, as well as via altering hormone receptor expression and signal transduction (including changes in protein or RNA expression, post-translational modifications and/or ion flux) in hormone-responsive cells. Simultaneously, EDCs can modulate hormone synthesis, hormone distribution or circulating hormone levels, hormone metabolism/clearance, or the fate of hormone-producing or hormone-responsive cells . Among their cellular effects, literature has shown that EDCs can also target mitochondria, perturbing mitochondrial bioenergetics, biogenesis, and dynamics and promoting excessive ROS production and activation of the mitochondrial pathway of apoptosis. This mitochondrial dysfunction, except for being particularly crucial for insulin-responsive tissues and metabolic homeostasis, is also emerging as a contributing factor in the pathogenesis of PCOS . Finally, apart from all the above modes of action, another major mechanism in which EDCs interfere with hormone action is by inducing epigenetic changes, especially during development and differentiation, by modifying epigenetic processes, such as DNA and histone modifications and non-coding RNA expression .

All the above features should not be used as a “checklist.” The innate disrupting hazard of an EDC is not necessarily increasing when more than one from the above-mentioned mechanisms is observed. Even one pathogenetic mechanism can be sufficient to disrupt an entire system, as long as it targets/interferes with a key event in hormone action/signaling. For example, BPA and DES are two EDCs exerting their effects via multiple ways, such as interacting or antagonizing with different receptors, altering signal transduction in various pathways, or inducing epigenetic changes. On the other hand, perchlorate, an inorganic ion that is widely manufactured for use in rocket propellant, matches, fireworks, and other explosives, inhibits thyroid hormone synthesis only by acting as a potent competitive inhibitor of iodide uptake through the sodium–iodide symporter from humans, rodents and other vertebrates .

Finally, it should be mentioned that EDCs can be potent even at very low doses and a long latency period between exposure and disease can be observed. Some of them have low accumulation in the human body (BPA, phthalates), while others are very lipophilic, accumulating easily in the food chain and the adipose tissue (persistent organic pollutants – POPs). Furthermore, they are usually characterized by a much lower affinity for hormone receptors, compared to natural ligands, as well as low-dose effects and non-monotonic dose- responses, which means that they can promote disrupting effects even in very low levels of exposure . Added to all the above particularities, EDCs are being added on top of the endogenous hormonal milieu, such that complex mixtures, dose additivity, and synergism between and among hormones and chemicals are the norms, further complicating their understanding .

Translating endocrine disruption into endocrine disease

Accumulating experimental data and epidemiological observations have proven numerous probable exposure-outcome associations, establishing EDCs as serious and urgent threats to public health . As a result, EDCs have been linked with an array of human diseases and adverse health outcomes, such as neurobehavioral disorders (e.g., attention-deficit disorder, autism spectrum disorder, and cognitive and behavioral dysfunction), pulmonary diseases (e.g., asthma), autoimmune diseases, various types of cancer (e.g., breast, ovarian, prostate), cardiovascular diseases (e.g., hypertension, ischemic heart disease, peripheral vascular disease) and obesity, as well as adverse birth outcomes (e.g., preterm birth and low birth weight) . More interestingly, rushed by the ongoing pandemic by severe acute respiratory syndrome (SARS) coronavirus-2 (SARS-CoV-2), preliminary data has postulated the hypothesis that exposure to EDCs can possibly be associated with increased incidence and increased severity of COVID-19 infection, just like other documented comorbidities, such as diabetes, hypertension, obesity, and cardiovascular disease. Except for the fact that EDC exposure can promote cardio-metabolic diseases, endocrine-related cancers, and immune system dysregulation that are also linked to a higher risk of severe COVID-19, preliminary molecular data indicates that potentially BPA can target key SARS-CoV-2 infection mediators, such as angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2). There is no doubt that this is still an explorative topic, which warrants further in vitro, computational, preclinical, and in vivo studies. However, it becomes schematically clear how determining is our living environment for disease susceptibility, even for viral infections .

Focusing on the endocrine system, literature has demonstrated that every endocrine axis can be a potential target for EDCs. A great number of EDCs seem to be able to interfere with hypothalamic and pituitary function, peripheral endocrine glands, as well as energy homeostasis and metabolism . Table 1 represents a synopsis of the endocrine systems targeted by EDCs, as well as the underlying pathophysiological effects that have been highlighted by experimental studies and how these are translated into endocrine diseases.

Vulnerable windows of susceptibility: Altering genomics and maximizing disrupting effects

Based on the “developmental origins of health and disease” (DOHaD) concept that has been introduced by David Barker over 50 years ago, we currently know that environmental stressors, such as malnutrition, drugs, infections, and EDCs, during critical periods of development, are capable of promoting subtle but vital changes in gene expression that lead to permanent alterations in an organ or tissue. These functional changes are translated into increased susceptibility to disease/dysfunction that can occur over the whole lifespan of an individual . Indeed, the risk of lifelong adverse health effects is enhanced when periods of EDC exposure coincide with the formation and differentiation of organ systems in early development, including early fetal development, infancy childhood, and puberty . Furthermore, apart from developmentally programming a single individual, evidence has shown that environmental stressors can exert multigenerational and transgenerational effects, mainly via promoting epigenetic changes. Epigenetics are mitotically and meiotically heritable changes in gene function, able to alter the phenotype of an organism without changes in the genotype (DNA sequence). There are three main mechanistic groups, in which they are categorized: DNA methylation leading to transcriptional repression, histone and chromatin modifications that alter gene expression, and non-coding RNAs that modulate gene expression post-transcriptionally. All of them can broadly influence and control cell and tissue development by directly controlling gene expression . To obtain a transgenerational phenomenon, epigenetic modifications must be transmitted through the affected germline to the unexposed generation. Prenatal exposure represents a classic paradigm of transgenerational inheritance, in which exposure to an environmental stressor of a female (F0 generation) during pregnancy does not only concomitantly affect her fetus (F1), but also its germline cells (F2). Consequently, the F3 generation will be the first generation that has not been directly exposed to the EDC. Therefore, if the effects of the EDC persist in the F3 generation, they are considered to be transgenerational .

Developmental programming and transgenerational inheritance had also been evaluated as a potential pathogenetic mechanism of PCOS . Among the most extensively studied, increased exposure to androgens in utero has been highlighted as a potential developmental commonality preceding PCOS. Various animal and human studies have described that in utero androgen exposure leads to a PCOS-like phenotype during adulthood, including disruption of pancreatic β-cells’ function and altered insulin sensitivity . Furthermore, nutrition can also have a central role in fetal programming, as both fetal growth restriction and overnutrition in utero have been linked with insulin resistance in young individuals and increased susceptibility to PCOS manifestation in adolescence . More recently, the transgenerational effects of anti-müllerian Hormone (AMH) have been also investigated. Specifically, elevated prenatal levels of AMH resulted in aberrant GnRH receptor signaling in the offspring, which can possibly serve as the necessary dysfunctional neuroendocrine basis to induce PCOS .

In the same context, EDCs can also perturb the developmental trajectory of the female reproductive system, when acting during vulnerable windows of susceptibility. Specifically, EDCs can target the hypothalamus-pituitary-ovarian axis at multiple stages and levels and perturb all physiological processes, such as gonadal development, follicular assembly, and folliculogenesis, uterus development and proliferation, steroidogenesis, hormonal production, and secretion. The total of these alterations induced by EDCs has been established as the “ovarian dysgenesis syndrome” and has been correlated with the manifestation of various female reproductive disorders, including PCOS . For instance, one of the first experimental models justifying the above observations was conducted by Fernandez et al. In this rat model, it was shown that neonatal exposure to BPA was accompanied by the development of a PCOS-like syndrome characterized by biochemical hyperandrogenemia, anovulation, infertility, polycystic ovarian morphology, and increased GnRH pulse frequency during adulthood .

Apart from fetal life, puberty is another well-established window of susceptibility, as well as a critical period in the evolutionary process of PCOS, during which environmental insults can decisively interfere with the hormonal, metabolic, and neuroendocrine milieu of a woman and/or promote dysfunctional adaptations that eventually favor PCOS manifestation . This causation is further supported by a recent human study, in which it was shown that adolescents with PCOS displayed significantly higher BPA levels, compared with age-matched healthy controls .

Targeting the reproductive axis

As mentioned above, the conceptual paradigm of “ovarian dysgenesis syndrome” has taught us that EDCs can affect pleiotropically the female reproductive axis, at multiple levels and functions. Starting from the hypothalamic-pituitary unit, EDCs can alter the neuroendocrine synchronization of reproduction, by altering centrally the signaling of female reproductive processes. The majority of the available data may derive mainly from animal studies. However, they are strongly supporting the disrupting properties of various chemicals and need to be further elucidated. Specifically, in animal models, gestational or neonatal exposure to BPA was linked with various derangements in the gonadotropin secretory capability, such as increased mRNA (messenger RNA) levels of the gonadotropins, lowered basal and gonadotropin-releasing hormone (GnRH), induced luteinizing hormone (LH) levels, increased GnRH pulsatility and altered pituitary GnRH signaling . Likewise, Di(2-ethylhexyl) phthalate (DEHP), one of the most commonly used agents from the group of phthalates, was found to augment the ability of pituitary cells to produce and secrete LH in response to GnRH in primary cultures, a finding that was afterward confirmed in prepubertal rats . Gonadotropin secretory derangements were also observed after atrazine exposure, a common pesticide, which was shown to promote pituitary hormone release and inhibit LH, through alterations in adrenal hormone secretion .

Moving to the ovary, numerous experimental studies have highlighted that EDCs can interfere with several physiological processes of the ovary, deregulating hormonal signaling by attacking at multiple stages. Gonadal development, follicular assembly and folliculogenesis, uterus development and proliferation, steroidogenesis, hormonal production, and secretion can be perturbed, via modifying the expression and/or activity of regulatory enzymes or by altering hormone receptor binding and action, leading inevitably to an array of reproductive disorders, including PCOS .

BPA was one of the first EDCs, in which all the above findings have been documented and were implicated in PCOS pathogenesis. Due to its phenolic structure, BPA can interact with estrogen receptors (ER), with, however, lower affinity than that of 17b-estradiol (E2) . It can also bind to aryl hydrocarbon receptor (AhR), a ligand-dependent transcription factor, which is almost present in every tissue and is involved in the modulation of steroid hormone synthesis and metabolism, while BPA may also have antiestrogenic properties, via inhibiting aromatase activity . As pointed out in the literature so far, its effects are starting from the early gonadal development, during which BPA can interfere with germ cell nest breakdown, meiosis, and follicle formation and viability, altering permanently ovarian reserve. For example, in animal models, BPA exposure led to increased incidence of unenclosed oocytes in macaques, inhibition of germ cell nest breakdown in mice, decreased number of primordial follicles in CD-1 mice , and disruption of meiotic division, a key stage of folliculogenesis, in mice and monkeys, via altered the expression of genes that control meiosis ( Stra8 , Dazl , Nobox ) . Postnatally, animal models have indicated that BPA can attack both folliculogenesis, by inhibiting the growth of antral follicles via the upregulated expression of cell cycle regulators and pro-and antiatretic factors, which promote atresia , and steroidogenesis, by upregulating the expression of several key cytochrome p 450 steroidogenic enzymes, such as 17-a hydroxylase and cholesterol side-chain cleavage enzyme, respectively, leading to the net effect of increased testosterone and progesterone levels . Finally, further accentuating androgen levels, BPA can also inhibit testosterone catabolism.

Collectively, all the above observations have urged scientific society to deduce that BPA may have a pathogenetic role in the PCOS etiology. To validate this assumption, numerous experimental models and human studies have followed. For example, in rodent models, prenatal and neonatal low-dose BPA exposures were accompanied by ovarian cyclicity disruption, increased testosterone production, and ovarian cysts, while high-dose BPA exposure led to ovarian cysts. Furthermore, via upregulating or downregulating critical enzymatic steps involved in ovarian steroidogenesis, such as the Cyp17a1, Cyp11a1, Star, and Cyp19a1, BPA exposure amplifies the testosterone synthesis in the theca-interstitial cells, while decreasing estradiol production in the granulosa cells . In a recent rat model by Yang et al. a PCOS-like syndrome was observed after neonatal exposure to tributyltin and BPA, with irregular estrous cycle, increased serum testosterone and LH levels and reduced sex hormone-binding globulin (SHBG) levels, as well as polycystic ovarian morphology, with more atretic follicles and cysts . Human data, although limited, have also supported the above-mentioned hypothesis. In a large study of 71 PCOS and 100 healthy women by Kandaraki et al., it was shown that BPA levels are higher in PCOS women compared to controls, independently of body weight as both lean and overweight PCOS individuals had elevated BPA levels compared to normal ovulating non-hyperandrogenemia women of matched body weight . Following this, a recent meta-analysis involving 493 PCOS patients and 440 controls confirmed that serum BPA levels are positively associated with women with PCOS, serving as a potential driving force for PCOS manifestation .

Apart from BPA, there is also a plethora of other EDCs that can also perturb the female reproductive system and potentially promote PCOS manifestation. For example, animal model and in vitro studies have shown that phthalates can also affect both folliculogenesis, by inhibiting follicle growth and inducing atresia and steroidogenesis. In rat models, DEHP and benzo[ a ]pyrene exposure led to inhibition of estradiol production, likely due to reduced aromatase, while, simultaneously, lower secretion of progesterone was also observed . Similarly, methoxychlor (MXC) an organochlorine pesticide used in agriculture, was linked with decreased ovarian weights and increased incidence of cystic ovaries, probably via inhibiting antral follicle growth by AhR and ER pathways , as well as with inhibition of estradiol, testosterone, androstenedione, and progesterone production in animal models .

There is no doubt that more evidence from high-quality studies and larger cohorts are needed to further confirm the intercorrelation between EDC exposure and PCOS. However, data are indicatory enough to support its etiologic role.

Targeting metabolism

Apart from their reproductive toxicity, EDCs are also known to perturb energy homeostasis and metabolism, by interfering with multiple regulatory levels and stages of energy metabolism and targeting the majority of the metabolically crucial organs of the human body, towards the manifestation of obesity, metabolic syndrome, and diabetes. These metabolism-disrupting properties can be critical and of great importance, particularly in women with PCOS, as they can define decisively the magnitude of PCOS clinical expression .