Nuclear Receptors: Connecting Human Health to the Environment

STEFANO LORENZETTI AND LAURA NARCISO

Istituto Superiore di Sanità – ISS, Department of Food Safety and Veterinary Public Health, Food and Veterinary Toxicology Unit, viale Regina Elena 299, 00161 Rome, Italy

1.1 Introducing Nuclear Receptors

Nuclear receptors (NRs) are evolutionary conserved proteins whose encoding genes are expressed in the animal kingdom (metazoans); they are also present in animals that do not have any endocrine system. NRs function as transcription factors activated by small (<1000 Da) lipophilic compounds able to cross the plasma membrane and, owing to their discovery as mediators of the sex steroid hormones, they were initially defined as endocrine receptors, although recently some of them have been suggested to act as sensors of their environment interacting with ligands external to the host organism (xenobiotics). Indeed, such ligand-activated transcription factors control levels of both xenobiotics (i.e., man-made chemicals such as pesticides and plasticizers) and endobiotics (i.e., endogenous chemicals such as sex steroid and thyroid hormones, vitamins) since the xenosensing activity of the NRs developed evolutionarily to support the spread of metazoans during the Cambrian age through the setting up of a whole endocrine system. The survival of all organisms (i.e., metazoans) relies on energy maintenance (via dietary intake, storage and utilization) and self-propagation (via reproduction), two physiological activities completely controlled by the central nervous system (CNS) through the signaling to the peripheral effector tissues/organs: NRs allow different, multiple signals to be integrated between central and peripheral organs acting as xenosensors and orchestrating hormone-dependent signaling.

1.2 Linking the Environment to the Human Organism: Nuclear Receptors as Mediators of the Action of Endocrine-active Compounds (EACs)

Ligands of NRs are usually defined as endocrine-active compounds (EACs) or endocrine-disrupting chemicals (EDCs), substances able to interfere with the function of hormonal systems affecting human and wild-life health, for example, contributing to developmental, reproductive and metabolic diseases. Although many EACs are xenobiotics – man-made chemicals manufactured by industry and released into the environment (e.g., pesticides, plasticizers, flame retardants, organotins, alkylphenols dioxins, polychlorinated biphenyls) – many others are naturally occurring (e.g., phyto- and mycoestrogens), being present in plants or fungi as part of their defensive mechanism against physical and biological stresses. Either via the environment or via the food chain, exposure to EACs is usually persistent and common to a wide range of compounds whose role might be potentially both harmful (e.g., xenobiotics) and/or beneficial (e.g., phytoestrogens). Indeed, industrialized and agricultural areas are typically polluted with a wide range of chemicals spread into the air, soil and groundwater, hence people working with (or in some cases living near sources of) pesticides, fungicides and other man-made chemicals are particularly exposed to these toxic compounds and thus have a higher risk of developing reproductive and/or endocrine dysfunction. With dietary and environmental exposure, EACs can affect the endocrine system by (i) mimicking natural hormones, (ii) antagonizing their action or (iii) modifying their synthesis, metabolism and transport. Most of the reported harmful effects of EACs are attributed to their interference with hormone-like, NR-mediated signaling.

1.3 How Nuclear Receptors Work

As mentioned above, human NRs (Table 1.1) constitute a superfamily of 48 ligand-activated transcription factors able to regulate cognate gene networks involved in key physiological functions such as cell growth and differentiation, development, homeostasis or metabolism. NRs have a fairly simple and conserved general structure and, as depicted in Figure 1.1a, is constituted by five distinct domains characterized by subdomains having specific functions. The modulatory A/B domain, at the amino terminus of each NR, contains the transcriptional activation function (AF-1) that, together with the AF-2 domain, takes part in receptor dimerization, nuclear localization and binding to co-activators and co-repressors (Figure 1.1b). The C domain is highly conserved since overlaps with the DNA-binding domain (DBD), a region recognizing NR-specific response elements (NR-RE) in the promoter sequences of targeted genes. The DBD contains two zinc fingers and TA boxes: the first zinc finger has a highly conserved sequence, termed P-box, involved in NR–DNA helix binding, whereas the second zinc finger contains a D-box necessary for protein–protein interactions and also for NR binding to the NR-RE. Finally, the TA boxes within the DBD are essential in NRs acting as monomers. NR binding specificity relies on the orientation of the binding sites (the single consensus binding site being the sequence AGGTCA) that can be formed by AGGTCA inverted, reverted or direct repeats, and also on their separation (from one to eight nucleotides) between the two single consensus binding sites. Indeed, whereas some NRs bind with high stringency to the consensus binding site, others display greater flexibility. The D domain is a short hinge region allowing each NR to undergo conformational changes upon ligand binding. The E–F domain, close to the carboxy terminus, contains the ligand-binding domain (LBD), a region containing the AF-2 domain that, as mentioned above, participates with the AF-1 domain in receptor dimerization, nuclear localization and binding to co-activators and co-repressors. The LBD is structured in α-helices that provide conformational flexibility to this region, allowing the DBD–LBD interaction and thus contributing also to the recruitment of co-regulators essential for the initiation of the transcription of target genes.

In the absence of the cognate ligand, some NRs are located in the nucleus, bind to the DNA response elements of their target genes and recruit co-repressors, whereas others are located in the cytoplasm in an inactive complex with chaperones (Figure 1.1b). Ligand binding induces major structural alterations of the receptor LBDs, leading to (i) destabilization of co-repressor or chaperone interfaces, (ii) exposure of nuclear localization signals to allow nuclear translocation and DNA binding of cytoplasmic receptors and (iii) recruitment of co-activators triggering gene transcription through chromatin remodeling and activation of the general transcription machinery. The crystal structures of most LBDs of the NRs have been determined, revealing a conserved core of 12 α-helices (H1–H12) and a short two-stranded anti-parallel β-sheet arranged into a three-layered sandwich fold; this arrangement generates a mostly hydrophobic cavity in the lower half of the domain which can accommodate the cognate ligand. In all ligand-bound LBD structures, the ligand binding pocket is sealed by the helix H12. This conformation is specifically induced by the binding of cognate ligands (endo- or xenobiotic) and is referred to as the 'active conformation' because it allows the dissociation of co-repressors and favors the recruitment of transcriptional co-activators.

It is noteworthy that this conformational state can also be achieved by some constitutively active orphan receptors for which no natural ligands have been identified. In this active-form, helices H3, H4 and H12 define a hydrophobic binding groove for short LxxLL helical motifs (L represents leucine and x any amino acid) found within co-activators. In contrast to agonist binding, interaction with antagonists prevents the correct positioning of helix H12, thus avoiding association with the LxxLL motifs of co-activators.

Since NRs form a complex set of interacting proteins that allow the body to coordinate responses to fluctuations in chemical levels, it is obvious that they undergo 'cross-talk'. It is becoming increasingly clear that NRs interact together and one of the great challenges is to ascertain how this interaction network fully functions and to be able to predict what the biological response will be for any given stimuli. NR 'cross-talking', indeed, has the twin advantages both of ensuring the most efficient response to a given stimuli and of providing a safety net to guarantee always an active capture system for a stimulus, even in absence of the cognate receptor. Interactions between NRs may occur at the level of sharing ligands and/or co-regulators and/or heterodimer partners and/or DNA binding elements. The best studied of these interactions is probably at the level of the target gene sets activated by NRs: the constitutive androstane receptor (CAR) and the pregnane-X-receptor (PXR), for example, coordinately regulate a battery of genes involved in all aspects of drug metabolism, including oxidative metabolism, conjugation and transport. PXR and CAR, indeed, have been shown to regulate commonly several genes encoding drug-metabolizing enzymes and drug transporters, a common feature of their biology, but also genes involved in apparently completely different processes such as the karyopherin-mediated nuclear import. The meaning of such coordinated interplay and the biological impact on the response to chemical stimuli are a challenge for future investigations on NR 'cross-talking'.

1.4 Environmental Chemicals and Adverse Effects on Human Health

In our environment, in consumer products and in foods, are commonly present chemicals, the above-mentioned EACs or EDCs, that interfere with hormone biosynthesis, metabolism or action resulting in a deviation from normal homeostatic control even during the developmental and reproductive life stages. Indeed, they have been defined as EDCs because they often possess phenolic moieties that structurally mimic natural steroid hormones and enable EDCs to interact with steroid hormone receptors as (anti)antagonists. Most EDC actions and their role in pathophysiological conditions are exerted through NRs (see Table 1.1), including estrogen receptors (ERs), androgen receptor (AR) and thyroid receptors (TRs).

The link between EDC exposure and human health has been accepted worldwide following the suggestion of Sharpe and Skakkebaek that the increasing incidence of reproductive abnormalities in the human male may be related to increased in utero estrogen exposure. This first speculation led subsequently to the definition of the so-called testicular dysgenesis syndrome (TDS), in which reproductive disorders of newborn (cryptorchidism, hypospadias) and young adult males (low sperm counts, testicular germ cell cancer), could be due to in utero exposure to EACs. The incidence of these symptoms has increased in recent decades and may result from an irreversible developmental disorder originating in early fetal life. This hypothesis has been supported by findings in animal models of TDS involving fetal exposure to phthalates.

Overall, to understand the mechanism of action and consequences of exposure to EDCs, some important and general issues have to be taken into consideration, such as the age and latency of exposure, the low dose–response and mixture effects in addition to heritability.

Concerning both the age and latency of exposure, it is always important to keep in mind that the exposure of an adult to an EDC may have different consequences with respect to a developing fetus or infant. As an example, a recent review by Sharpe is noteworthy, in which the impact of environmental and lifestyle factors on spermatogenesis was reviewed considering the exposure to EACs/EDCs that may occur in fetal life and how this might then impact on spermatogenesis in adulthood. Sharpe highlighted how exposure to environmental contaminants in perinatal life causes adverse effects on testis development that are irreversible (probably because fetal germ cells are affected), whereas the effects on the process of spermatogenesis in adulthood are probably reversible. Hence the consequences of developmental exposure may not be immediately apparent early in life but may be manifested in adulthood or during aging.

Concerning the mixture issue, it is important to remember that, although the effects of contaminants are typically studied in single exposures, environmental exposures are rarely from a sole contaminant since organisms are exposed to different chemicals due to occupational, environmental, dietary or pharmaceutical exposure. As a result, mixtures of chemicals may have activities that are greater than that of one chemical alone (additive or synergistic effects) or less than predicted (antagonistic). As an example, the work of Rider et al. on the cumulative effects of antiandrogenic chemicals on the reproductive development of the male rat upon in utero exposure to binary, ternary or more complex (up to seven) mixtures could be considered pivotal. All tested combinations, using compounds that act by disparate mechanisms of toxicity, produced cumulative, dose-additive effects on the androgen-dependent tissues and, importantly, the chemicals were tested at concentrations below their individual "No Observable Adverse Effect Level" (NOAEL).

Concerning the dose–response curve, it should be remembered that monotonic dose–response relationships are not always applicable to toxicants and in particular to EACs/EDCs, which are recognized to give, for instance, inverted-U dose–response relationships where very low doses stimulate growth and very high doses completely inhibit development. Indeed, non-monotonic (or biphasic) dose–response relationships commonly occur in endocrinology for virtually all hormones and for EACs, as witnessed by the conclusion of a US Environmental Protection Agency expert panel that estrogenic chemicals can cause biological effects in laboratory animals at levels below those normally found to be safe, an assessment that runs counter to the conventional wisdom in toxicology. Unexpected unique low-dose effects have been observed, for instance, for bisphenol A (BPA) stimulation of human prostate cell growth and also for the greater effect of pesticides on mice fertility at low but not high doses of the chemicals.

Concerning heritability, one has to consider the so-called transgenerational phenotype, in which the EACs/EDCs can affect not only the exposed individual but also the children and subsequent generations: in this case, the adverse effect can be transmitted epigenetically (i.e.,via altered DNA methylation or histone acetylation or miRNA expression) through the germline even if the subsequent generations are not directly exposed to the environmental factors. A well-known example of germline transmission is a rat model with the fungicide vinclozolin: embryonic exposure to vinclozolin during gonadal sex determination induced a transgenerational effect (from generation F1 to F4) on adult male reproduction and sperm production and also on breast tumors, prostate disease, kidney disease, immune abnormalities and metabolic disorders.

1.5 Reproductive Diseases and Fertility

All over the world, human fertility rates are declining. In many Western countries, the fertility rate is below the replacement level and often this is attributed to socio-economic factors and increasing control of fertility, although the decreasing ability to conceive is appearing more and more influenced by environmental and lifestyle factors. The most common reproductive pathologies associated with the action of EDCs are clinically dimorphic: (i) male sexual differentiation is androgen dependent and the main associated diseases are the testicular dysgenesis syndrome (TDS) and prostate pathologies in which altered NR expression has been shown; (ii) female differentiation, in contrast, occurs largely independently of sex steroids and the main associated disorders are premature thelarche and pubarche, disorders of ovulation and lactation, breast diseases, endometriosis and uterine fibroids.