Endocrine Disruption in the Omics Era: New Views, New Hazards, New Approaches

The genome revolution has brought about a complete change on our view of biological systems. The quantitative determination of changes in all the major molecular components of the living cells, the "omics" approach, opened whole new fields for all health sciences, including toxicology. Endocrine disruption, i.e., the capacity of anthropogenic pollutants to alter the hormonal balance of the organisms, is one of the fields of Ecotoxicology in which omics has a relevant role. In the first place, the discovery of scores of potential targets in the genome of almost any Metazoan species studied so far, each of them being a putative candidate for interaction with endocrine disruptors. In addition, the understanding that ligands, receptors, and their physiological functions suffered fundamental variations during animal evolution makes it necessary to assess disruption effects separately for each major taxon. Fortunately, the same deal of knowledge on genes and genomes powered the development of new high-throughput techniques and holistic approaches. Genomics, transcriptomics, proteomics, metabolomics, and others, together with appropriate prediction and modeling tools, will mark the future of endocrine disruption assessment both for wildlife and humans. Introduction The "'omics" perspective The completion of the Human Genome Project in 2001 represented a complete turning point in Biology. Together with the complete genomes of the yeast Saccharomyces cerevisiae (1996), the nematode Caenorhabditis elegans (1998), the fruitfly Drosophila melanogaster (2000), the cress Arabidopsis thaliana, and the fish Takifugu rubripes (2002), it allowed for the first time the analysis of the genetic makeup of Eukaryotes in its complete extension and to establish new functional, evolutive and physiological correlations between taxa, individuals, organs and cell types. The development of new highly efficient analytical techniques has allowed similar holistic approaches for essentially all components of the live cell. For example, following the biological flow of information, high throughput techniques of specific RNA quantitation (microarrays) allowed the description of the mRNA complement of a given cell or tissue, i.e., the description of its transcriptome. A similar approach, but using completely different methods (2D electrophoresis, advanced mass spectrometry techniques), has been applied to elucidate the protein composition (the "proteome") with unprecedented precision, whereas several analytical methodologies (gas and liquid chromatography coupled to mass spectrometry, capillary electrophoresis, and high resolution nuclear magnetic resonance (NMR) allow the description of the chemical composition of the cell, that is, the nature and composition of its metabolites (the "metabolome"). Figure 1 summarizes the nature and challenges of each one of these "omic" analyses. Figure (1). Overview on the generic processes that characterize the differences “omics” technologies. On the left, schematic representation of the Biological information flow from the genome to cellular phenotypes. From top to bottom, the information of DNA (genome) is first transcribed to mRNA (transcriptome), which is afterwards translated into proteins (proteome). A subset of proteins, the enzymes catalyse reactions that both consume and produce the different metabolites (metabolome). The main table summarizes different characteristics of each omic technique. The application of “omics” tools/technologies is now widely employed in many research areas, from medicine to environmental sciences, as they give information on some key regulators of various processes in living organisms. They also represent a tremendous opportunity to improve human and wildlife health by the characterization of the environmental elements that impact public and wildlife health [1]. The recognition of these challenges and opportunities, along with the fact that many of the most prevalent diseases are associated with the endocrine system, has led to a focus on chemical exposures and especially endocrine disruptors. The term ecotoxicogenomics refers to the integration of genomic-based science into ecotoxicology using DNA array-based technologies, proteome and metabolomic analyses [2]. The approach has been fundamental in several aspects of our current comprehension of the effects of endocrine disruption: The elucidation of potential effects on different animal taxa, thanks to comparative genomics, the comprehension of the multiple targets for exogenous ligands in the cell, and the development of new and more precise methodologies to monitor endocrine disruption both in humans and in wildlife. Hormones and physiological processes Hormone are endogenous molecules secreted by endocrine glands that travel through the blood stream and induce physiological effects on distant cells and tissues [3]. The mechanism by which hormones exert their actions on target cells remained obscure for decades, although the concept of the existence of an endogenous receptor responsible for the recognition of the hormone (the "ligand") and the triggering of the physiological changes derived from its presence dates from the early 70's [4]. The isolation and cloning of the first hormone receptors in the 80's opened the Molecular Endocrinology era [5, 6]. However, the recognition of the amazing complexity and implication of the hormonal signaling did not appear until the advancement of whole genome sequencing techniques and its extension to evolutionary distant species [6, 7]. Analysis of complete genomes of different species revealed the presence of dozens, if not hundreds, of evolutionary related proteins for which no ligand or function was known. These so-called "orphan" receptors are widespread across Metazoa and their functional characterization implicated a major progress on the understanding of cell regulation, development and even evolutionary links between animal taxa [6-8]. An overview of the different receptor families, their physiological ligands and their distribution among the major Metazoan taxa is shown in Table 1. Table 1. Nuclear receptors nomenclature, known ligands and distribution among animal taxa. Common Name Abbreviation Unified Name Ligands (mammalian) Taxonomic distribution Dosage-sensitive sex reversaladrenal hypoplasia congenital critical region on the X chromosome, gene 1 DAX-1 NR0B1 V, Ar Short heterodimeric partner SHP NR0B2 V, Ar Thyroid hormone receptor α TRα NR1A1 thyroid hormones V, UC, Mol, Ann Thyroid hormone receptor b TRb NR1A2 thyroid hormones V, UC, Mol, Ann Retinoic acid receptor α RARα NR1B1 retinoic acids V, UC Retinoic acid receptor β RARβ NR1B2 retinoic acids V, UC Retinoic acid receptor γ RARγ NR1B3 retinoic acids V, UC Peroxisome proliferatoractivated receptor α PPARα NR1C1 fatty acids V, UC Peroxisome proliferatoractivated receptor β /d PPARβ /δ NR1C2 fatty acids V, UC Peroxisome proliferatoractivated receptor γ PPARγ NR1C3 fatty acids V, UC Reverse-Erb α REV-ERBα NR1D1 [heme] V, UC, Ar Reverse-Erb β REV-ERBβ NR1D2 [heme] V, UC, Ar RAR-related orphan receptor α RORα NR1F1 [sterols] V, UC, Ar RAR-related orphan receptor β RORβ NR1F2 [sterols] V, UC, Ar RAR-related orphan receptor γ RORγ NR1F3 [sterols] V, UC, Ar Liver X receptor β LXRβ NR1H2 oxysterols V, UC, Ar Liver X receptor α LXRα NR1H3 oxysterols V, UC, Ar Farnesoid X receptor α FXRα NR1H4 bile acids V, UC, Ar Farnesoid X receptor βa FXRβ NR1H5 V, UC, Ar Ecdisone receptor EcR NR1H Ecdysone Ar, Nem** Vitamin D receptor VDR NR1I1 1a,25dihydroxyvitamin D3 and lithocholic acid V, UC, Ar, Nem Pregnane X receptor PXR NR1I2 endobiotics and xenobiotics V, UC, Ar, Nem Constitutive androstane receptor CAR NR1I3 xenobiotics V, UC, Ar, Nem Hepatocyte nuclear factor 4 α HNF4α NR2A1 [fatty acids] V, UC, Ar, Nem Hepatocyte nuclear factor 4 γ HNF4γ NR2A2 [fatty acids] V, UC, Ar, Nem Retinoid X receptor α RXRα NR2B1 9-cis retinoic acid and docosahexanoic acid V, UC, Ar, Mol Retinoid X receptor β RXRβ NR2B2 9-cis retinoic acid and docosahexanoic acid V, UC, Ar, Mol Retinoid X receptor γ RXRγ NR2B3 9-cis retinoic acid and docosahexanoic acid V, UC, Ar, Mol Table 1 (Cont) Common Name Abbrevia tion Unifi ed Nam e Ligands (mammalian) Taxono mic distribut ion Testicular orphan receptor 2 TR2 NR2 C1 V, UC, Ar, Nem Tailless homolog orphan receptor TLX NR2 E1 V, UC, Ar, Nem Photoreceptor-cell-specific nuclear receptor PNR NR2 E3 V, UC, Ar, Nem Chicken ovalbumin upstream promotertranscription factor α COUPTFα NR2 F1 V, UC, Ar, Nem Chicken ovalbumin upstream promotertranscription factor β COUPTFβ NR2 F2 V, UC, Ar, Nem Chicken ovalbumin upstream promotertranscription factor γ COUPTFγ NR2 F6 V, UC, Ar, Nem Estrogen receptor α ERα NR3 A1 estrogens V, UC**, Mol*, Ann Estrogen receptor β ERβ NR3 A2 estrogens V, UC**, Mol*, Ann Estrogen related receptor α ERRα NR3 B1 V, UC, Ar Estrogen related receptor β ERRβ NR3 B2 V, UC, Ar Estrogen related receptor γ ERRγ NR3 B3 V, UC, Ar Glucocorticoid receptor GR NR3 C1 glucocorticoi ds V Mineralocorticoid receptor MR NR3 C2 mineralocorti coids and glucocorticoi ds V Progesterone receptor PR NR3 C3 progesterone V Androgen receptor AR NR3 C4 androgens V Nerve-growth-factor-induced gene B NGF1-B NR4 A1 V, UC, Ar Nur-related factor 1 NURR1 NR4 A2 V, UC, Ar Testicular orphan receptor 4 TR4 NR4 A2 V, UC, Ar Neuron-derived orphan receptor 1 NOR-1 NR4 A3 V, UC, Ar Steroidogenic factor 1 SF-1 NR5 A1 [phospholipid s] V, UC, Ar Liver receptor homolog-1 LRH-1 NR5 A2 [phospholipid s] V, UC, Ar Germ cell nuclear factor GCNF NR6 A1 V, UC, Ar, Nem AnnAnnelida, ArArthropoda, MolMollusca, NemNematode, UCUrochordata/Cephalochordata, Vvertebrata * No evidence for ligand binding. ** Hypotetical Ligands in brackets appear to be constitutively bound to their receptors Data from reference [6], with modifications Endocrine Disruption Environmental pollutants in the various ecosystems are a major concern worldwide. Our understanding of the potential adverse effects of anthropogenic contaminants has driven to an increasing public awareness on their influence on health and well-being of both humans and wildlife [3, 9]. The primary observed effects of those pollutants are related to several developmental and reproductive disorders in wildlife species, which have been clearly linked to environmental compounds that act as endocrine disrupting chemicals (EDCs) [3]. By definition “An endocrine disruptor is an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub) populations" [10]. Reports of adverse effects by EDCs in the last decades include reproductive impairment (decrease fertility, hermafroditism and sex reversal, altered sex ratios, hatching success), metabolic defects (Thyroid dysfunction, body weight and fat tissue control alteration), and alteration of immunological and behavioural functions have been observed in wild populations of mammals, birds, reptiles, amphibians, fish and mollusks; epidemiological evidences indicate that at least part of these or similar effects may be already occurring in human populations [3, 10-14]. There are varied sources of environmental contaminants that may disrupt the endocrine system (Figure 2). The human exposure typically occurs with the environmental contamination of the food chain, especially fresh water fish and meat, contact with contaminated household dust, and occupational exposure [9]. Some chemicals were banned or removed from production years ago but persist in the environment. On the other hand other EDCs are high production volume chemicals found in a many household products. Bisphenol A (BPA), for example, is present in polycarbonate plastics, including beverage and food storage containers; epoxy resins that line the interior of metal cans, and in the ink used for thermal paper receipts. Many textiles contain contaminants, such as flame-retardants, including tetrabromobisphenol A and polybrominated diphenyl ethers. Some individuals have also been exposed to contaminants with adverse effects as a result of medical (diethylstilbestrol; DES), dental (diglycidyl methacrylate; GMA) or dietary (phytoestrogens) interventions. Urban wastewaters are important pollutant sources of natural and synthetic estrogens and other hormones that are subsequently found in surface waters [15, 16]. Thus, exposure to EDCs is ubiquitous and inevitable and there is growing concern that living in an EDC contaminated world may be contributing to adverse health trends, such as early puberty and infertility, because of growing evidence that a number of EDCs can produce varied effects [3, 17], (Figure 2). Figure (2). Summary of the common sources and mechanisms of endocrine disruptors in humans and how they may influence some key developmental processes, in particular through their actions during critical periods of development. Evolutionary perspective of endocrine disruption: EDCs in invertebrates From its very first origins, the concept of endocrine disruption has been deeply associated to vertebrate wildlife (birds, reptiles) and humans. However, environmental and laboratory research indicate that fish species are particularly sensitive to many forms of endocrine disruption, particularly to estrogens [18]. Finally, the discovery that the presence of TBT in coastal waters is linked to imposex in marine gastropods brought the same concept to mollusks and invertebrates in general [19, 20]. The more recent deleterious effects of neonicotinoid insecticides on bee colonies around the world is probably a further example of the importance of recognizing and controlling environmental EDCs than can effect non-vertebrate species [21]. Nevertheless, information on EDCs in invertebrate species is still limited [22-30]. Difficulties in performing invertebrate studies are largely due to the limited knowledge on the endocrine physiology of many invertebrate groups that represent important components of ecosystems. In the first place, the concept "invertebrate" is essentially instrumental, as it covers species from widely different Metazoans. Only among animals with bilateral symmetry (Bilateria), three major taxonomic clades appear: Deuterostomata (which includes Vertebrates and Echinoderms, among others), Ecdisozoa (Nematodes and Arthropods) and Spiralia/Lophotrochozoa (Mollusks and Annelids), the two later forming the group of Protostomata [31, 32] (Figure 3). These major groups of animals, as well as the ancestral groups of Porifera and Ctenophora, are characterized by their distinct embryonic developments. Last advances on biochemistry, genome sequences and regulatory studies showed that they exhibit fundamental differences in their complements of nuclear receptors, in their ability to synthesize and metabolize molecules with ligand activity, and, most probably, in the physiological roles of these molecules (Figure 3, Table 1) [8, 31-34]. Figure (3). Presence of the different types of hormones among animals. Principal taxonomic groups are indicated. Signs ‘‘+’’ indicate that a given class of hormones has been demonstrated to be involved in life history transitions (LHT) for a particular taxon, ‘‘?’’ indicates that such a role has not been demonstrated, and ‘‘+?’’ indicates preliminary evidence for such a roles (modified from [84]). The current model for evolutionary relationships among animal taxa is shown on the left (from [32]) Despite the great invertebrate genetic and taxonomic diversity, ecotoxicological studies are limited to few worms, mollusks and arthropod species that are easy to maintain and rear in the lab. Furthermore, in regard of their endocrine system, invertebrates have been relatively far less studied than vertebrates, with most of the literature published on invertebrate endocrinology referring to mollusks, insects and crustaceans [19, 34, 35]. When dealing with transcriptomic or proteomic analyses the situation is even worse, because of the relatively low number of invertebrate genomes at least partially sequenced and more or less correctly annotated [36, 37]. For these reasons there is a high degree of uncertainty to relate an adverse effect on invertebrate growth or reproduction to specific changes of its endocrine system. For example, despite the abundance of literature dealing with the development, growth and reproductive effects of EDC in invertebrates, only few studies have assessed unambiguously a truly endocrine disrupting effect [38]. Several studies tested the wrong premise that mammalian or vertebrate EDC should act as endocrine disruptors in invertebrates through identical or at least homologous mechanisms of action. This was especially evident for estrogenic/androgenic compounds tested against different arthropod species, which lack functional estrogenic/androgenic receptors (Table 1). In many cases, the detrimental effects of EDC on growth and reproduction reported, were related to egg mortality or feeding inhibition rather than to effects on endocrine disruption. Research conducted in the crustacean species Daphnia provides conclusive evidence that juvenile hormone agonists enhanced male production disrupting the ultraspiracle receptor signaling pathway during the initial phases of embryo development [39]. There is also evidence that juvenile hormone agonists modulate ecdysteroid activity causing embryo arrest or abnormalities throughout the ultraspiracleand ecdysone-receptor complex [40]. Studies on the effects of antidepressants show that molluscan and crustacean reproductive and locomotion systems are affected by antidepressants at environmentally relevant concentrations [35]. In particular, antidepressants affect spawning and larval release in bivalves and disrupt locomotion and reduce fecundity in snails. In crustaceans, antidepressants affect freshwater amphipod activity patterns, marine amphipod photo-and geotactic behavior, crayfish aggression, and daphnid reproduction and development [34, 35, 41]. The above reported effects are likely to be related through out disruption of neuroendocrine signaling pathways. Like in vertebrates, the endocrine control of growth, reproduction and behavior in invertebrates are initiated by neurohormones [34, 35, 41]. Serotonin and antidepressants targeting these neurohormones induce spawning in bivalves, alter locomotion and foraging behavior in gastropods and alter mimetic and predatory behavior and memory in cephalopods [42-49]. Neurohormones control a wide variety of biological systems in crustaceans, including reproduction, growth, maturation, larval development, immune function, metabolism, behavior and colour physiology. For example, both serotonin and dopamine have been found to stimulate the release of multiple other crustacean neuropeptide hormones including hyperglycaemic hormone, red and black pigment dispersing/concentrating hormone, neurodepressing hormone, molt-inhibiting hormone and gonad-stimulating hormone [34, 41, 50, 51] (Figure 4). Although neurohormonal disruption in molluscan and crustacean species have been mostly limited to antidepressants, there are other pharmaceutical drugs targeting neuronal receptors or other enzymes that may also alter neuroendocrinological pathways that regulate key physiological function. One of those are non-steroidal anti-inflammatory drugs (NSAIDs) that interrupt crustacean eicosanoid metabolism, which appears to disrupt signal transduction affecting juvenile hormone metabolism and oogenesis [52]. Figure (4). (De)regulation of haemoglobin-related genes by juvenile hormone (JH) or its analogs (Methoprene, Methylfarnesoate, Fenoxycarb, and Epofenonene) in Daphnia magna. A), Putative pathways of JH disruption (modified from [51]). B) Example of the hemoglobin deregulated phenotype in D. magna by ectopic activation of the JH pathway. The implementation of omics approaches to invertebrates is limited to only few model species and studies that have addressed omics and EDCs are scarce [36, 37, 53-56]. Probably the best characterized and solidly established mode of action (MoA) of "canonical" (i.e., active in vertebrates) EDCs is the disruption of the enzymatic pathways for steroid synthesis, which are very well conserved within Metazoans [19, 57, 58]. However, several studies [53, 54, 56, 59-63] reported disruption on regulatory mechanisms in addition to the effects on the steroidogenic pathway. Transcriptomic patterns of intersex specimens of clams Scrobicularia plana showed a de-regulation of the androgen receptor signaling pathway [53], an effect also described for Mya arenaria males exposed to TBT [64]. Metabolomic and proteomic analyses revealed different cases of neuroendocrine disruption in crustaceans and mollusks. For example, atrazine and its metabolites affected eicosanoids in the isopode Hyalella azteca suggesting possible perturbations in neuropeptide hormonal systems [55]. Similarly, ibuprofen inhibits reproduction in the crustacean Daphnia magna due to the de-regulation of the ecosanoid signaling pathway, and hence of prostaglandins, which in crustaceans control reproduction [62, 65]. A typical vertebrate estrogenic EDC, ethinylestradiol, alters metabolite pathways related to energy reserves, signal transduction, immune response, and neuromodulation in the unionid mussel Lampsilis fasciola. These effects result in physiological changes, as altered siphon and mantle [54], both in male and females. Conversely, the poly-bromo-diphenyl ether congener (BDE 47) affects differentially male and female specimens of the marine mussel Mytilus galloprovincialis: males show effects in energy metabolism, whereas, in females, BDE 47 disrupts both osmotic regulation and energy metabolism [56]. Very likely, we are just beginning to understand the phenomenon of EDC in invertebrates, and many of the chemicals that are contaminating terrestrial, freshwater and marine ecosystem may have endocrine disrupting effects on invertebrate species. Given the vast genetic diversity of Metazoans, the MoA already defined for vertebrates may or may not apply to any particular taxon. Therefore, EDC toxic effects may have unanticipated, and many times unnoticed until they reach global scale, deleterious effects in the ecosystems that depend upon invertebrate populations. Omics’ application on EDCs assessment Transcriptomics and EDCs The availability of studies that report the transcriptomic changes of different EDCs on several organisms, including human cells is reported in Table S1. One of the goals of the transcriptomic analysis is to discover mechanistically based molecular biomarkers with utility for risk assessment and develop modeling approaches for predicting adverse outcomes. The understanding of population-level impacts of EDCs in biological systems is, however, dependent on an enhanced knowledge of their MoA, and development of mechanism-based indicators suitable for application on field work that enable linkage of exposure to adverse effects at both individual and population levels. The elucidation of the signaling pathways and transcription factors (TF) networks affected by EDCs could be successfully addressed by examining the transcriptomic responses in model species such as the zebrafish (Danio rerio). Among the several fish species with a sequenced genome, zebrafish is one of the best model systems for omics, which is addressed by considerable high number of studies, reporting the effects of EDCs in different tissues and in differential stages of development. In addition, it has relatively abundant genomic resources such as genetic maps, mutants, and markers available [66]. The biological responses to external stressors, including toxicants, involve changes in normal patterns of gene expression [67, 68]. Many responses are a direct result of the chemical, such as alterations in gene expression caused by the binding of a steroid hormone (or analogue) to a specific steroid hormone receptor, which acts as a TF and subsequently modulates (activates or represses) the transcription of its target genes. Importantly, however, different mechanisms of toxicity can generate specific patterns of gene expression that can potentially provide us with molecular biomarkers of disruption of a biological process and be reflective of mechanism or mode of action [67]. The search for networks, enabling new hypotheses to be formulated and tested for the mechanisms underlying specific toxic effects are one of the best challenges that transcriptomic is facing. From systems biology perspective, signaling pathways and TF networks are at the center of a complex biological system. As such, signal transducers and TFs provide critical links between chemical exposures and resultant toxic effects manifested at various levels of biological hierarchy, from molecular to organismic [69]. Mechanistically based molecular indicators would also allow for improved extrapolation of effects across species, biological levels of organization, and diverse chemical structures. Finally, given the pleiotropic nature of signal transducers and TFs, organismic endpoints explicitly mapped to specific toxicity mechanisms may be developed by generating gene knockout mutants in targeted pathways. An ensuing greater efficiency and accuracy in the assessment of both EDC exposure and hazard would improve the overall risk assessment process [69]. It should be emphasized, that changes in gene expression are generally rapid and thus potentially provide a capability of a rapid diagnosis of chemical effect. However, the transcription of messenger ribonucleic acid (mRNA) is only an intermediate step in conversion of genetic information into proteins, the biochemical bases of biological function and gene expression and concentration of functional proteins are not necessarily always directly related.