Review
Organic Synthetic Environmental Endocrine Disruptors: Structural Classes and Metabolic Fate
Jan Schmidt and Lucija Peterlin Ma{i~*
University of Ljubljana, Faculty of Pharmacy, Aškerčeva 7, 1000 Slovenia * Corresponding author: E-mail: Corresponding author: E-mail: lucija.peterlin@ffa.uni-lj.si
Received: 14-05-2012
Abstract
Endocrine disruption is the modification of the endocrine system causing harmful effects in healthy subjects or their offspring. Physiological endocrine hormones act at very low plasma concentrations, and certain chemicals known as endocrine disrupting compounds (EDCs) are suspected of modifying endocrine function at similarly low concentrations. In our review we focus mainly on the structural classes of organic synthetic environmental endocrine disruptors and their common structural elements that enable them to interact with estrogen signalling. EDCs can affect estrogenic signalling directly through interaction with estrogen receptors (ERs) or indirectly through transcription factors such as the aryl hydrocarbon receptor (AhR) or by modulation of critical metabolic enzymes engaged in estrogen biosynthesis and metabolism. However, some structural elements can also pose a great risk of cytotoxicity and genotoxicity, especially after biotransformation to reactive metabolites.
Key words: endocrine disrupting compounds (EDCs), estrogen receptors (ERs), aryl hydrocarbon receptor (AhR), reactive metabolites
1. Introduction
Most structural classes of synthetic endocrine disrupting compounds (EDCs) affect physiological estrogen signalling and are therefore called estrogenic endocrine disruptors, or xeno-estrogens. EDCs can modify genomic and non-genomic estrogen receptor activity through direct interaction with estrogen receptors (ERs). EDCs can also interact with other targets involved in estrogen signalling. They can act as ligands for transcription factors such as the aryl hydrocarbon receptor (AhR) and by modulation of metabolic enzymes that are essential for normal estrogen synthesis and metabolism.1
Xeno-estrogens belong to a number of chemical classes that display a broad range of structural diversity. Many structure-activity relationships (SARs) have been studied in which activity was measured using a validated rat uterine cytosol ER competitive binding assay and where the focus was on identification of structural commonalities between diverse ER ligands.2 Only after discovering crystal structures of ERa and ERp ligand binding domains could binding properties and selectivity of ER li-gands be properly determined.3,4
Synthetic estrogenic endocrine disruptors can be divided into eight main structural classes: steroids, diethyl-stilbestrol-like compounds, triphenylethylene derivatives,
diphenylmethanes, biphenyls and phenols, and two classes of AhR ligands: polycyclic aromatic hydrocarbons and dioxins (Table 1). For comparison, AhR binds a wide array of structurally different hydrophobic ligands such as polycyclic aromatic hydrocarbons (PAHs) (Figure 9) and halogenated aromatic hydrocarbons (HAHs). Among the high affinity AhR ligands are dioxins, primarily (TCDD), a known EDC.
Involvement of AhR ligands in estrogen signalling is a consequence of intertwined signalling pathways of AhR and ERa (ERa binds to AhR-regulated genes and AhR binds to ERa-regulated genes). AhR ligands 2,3,7,8-te-trachlorodibenzo-p-dioxin (TCDD) and 3-methylcholant-hrene (3-MC) although treated as equivalent compounds, they have distinct effects. TCDD was shown to act solely as antiestrogenic compound, while 3-MC exerts estroge-nic properties depending on the cell type. Genomic study from Swedenborg et al. was examining the transcriptional effects of TCDD and 3-MC with regards to ER ligand diethylstilbestrol (DES). All three ligands regulated separate sets of genes, thereby inducing different signaling pathways, with the exception of CYP1A1 and aldehyde dehydrogenase 3A1 genes that were upregulated by both 3-MC and TCDD. It was showed that 3-MC and TCDD control distinct gene expression and probably also have different biological functions. Additionally, 3-MC had an
Table 1. EDC structural classes and representative compounds
EDC structural class
Representative compound
Steroids
ethynylestradiol
DES-like chemicals
diethylstilbestrol
Triphenylethylene derivatives
tamoxifen
Diphenylmethanes
bisphenol A
Biphenyls
2',3',4',5'-tetrachloro-4-biphenylol
Phenols
ho
nonylphenol
Dioxins
O'
TCDD
PAHs
benzo [a] anthracene
impact on almost 700 genes, hypothesising that many different metabolites were generated in HepG2 cells. The regulation pattern suggests that 3-MC and/or its metabolites have estrogenic effects that are mediated only through ERa.5 Antiestrogenic effect of TCDD were discovered by Wormke et al. implicating nuclear localization of the AhR, AhR-ERa interactions and finally proteasome-depen-
dent degradation of AhR and ERa.6 Study of Palumbo et al. showed that AhR agonists reduced vitellogenin synthesis in fish and therefore have opposite effects than estrogen chemicals.7 However, many EDCs act through multiple mechanisms, as exemplified by compounds that bind to multiple targets, for example ERs and AhRs such as
3-MC.
Some groups of chemicals also influence estrogen synthesis and metabolism, e.g. triazine herbicides and vinclozolin. Triazine herbicides and their metabolites induce the activity and gene expression of the aromatase, also known as the cytochrome oxidase, CYP19A1.8,9 Apart from estrogenic effects, some EDCs are able to form reactive metabolites that can induce oxidative DNA damage and/or covalent binding to DNA. The role of estrogene reactive metabolites in mammal carcinogenesis is described in more detail in many review articles.10,11,12 The main pathway leading to carcinogenesis is formation of the catechol estrogen metabolites, especially 4-hydroxy E2.
4-Hydroxy	E2 is the most carcinogenic causing DNA damage, kidney tumors in Syrian hamster and uterine tumors in CD-1 mice, what was reported by Newbold et al.13 NADPH-dependent metabolic oxidation pathway of E2 is well established and consists of the following steps: 1) 2- or/and 4- hydroxylation of E2, 2) subsequent oxidation to semiquinones and ortho-quinones, 3) redox cycling between the ortho-quinones and their semiquinones generating radicals and 4) covalent binding of ortho-qui-nones, especially of 4-hydroxy E2 to DNA.8,14 However, when it comes to the extent of 4-hydroxy E2 formation with regards to 2-hydroxy E2 formation, studies on this subject are not uniform. For example, Wilson et al. reported that 4-hydroxy E2 was a predominant metabolite in female ACI rats but on the other side Mesia-Vela et al. found that the main metabolite formed from liver microsomes of ACI rats was 2-hydroxy E2. ACI rat model is very useful, since E2 induced tumours appear in a very short period at low E2 concentrations.15,16 Authors suggest that the differences could appear due to the possibility of different diet of the laboratory animals or due to existence of other 4-hydroxy E2 formation pathways, meaning that extra hepatic metabolism could play a more important role that was thought before. It is also possible that, especially if contaminated with almost ubiquitous TCDD could influence the metabolism of E2 in the same animal species. In principle reactive intermediates could be formed directly by bioactivation of ingested EDCs or through EDC induction of certain CYPs. For example, Beedanagari et al. examined the induction of CYP1A1 and CYP1B1 genes in human MCF-7 breast cancer cells by TCDD that are involved in E2 hydroxylation.17 Spink et al. proposed the scheme of hydroxylation reactions for E2 and for equine estrogens, which are commonly used in estrogen replacement therapy. Equine estrogens: equilin, equilenin and 8-dehydroestrogene are structurally B-ring unsaturated estrogens. The slight difference in the metabolic behaviour
to the E2 is due to the aromaticity of the steroid B ring that favours C-4 hydroxilation by TCDD induced CYP1A1 and CYP1B1.18 Wang et al. have developed a sensitive LC-MS/MS method based on the knowledge that 4-hy-droxy equine estrogens form their ortho-quinones, which form stable DNA adducts. This method could provide a good basis for developing a sensitive assay for monitoring estrogen induced DNA damage.19 In human MCF-7 breast cancer cells CYP1B1 predominantly catalyzes hydroxyla-tion of E2 at the C-4 position, whereas CYP1A1 predominantly catalyzes 2-hydoxylation.18 It should be also mentioned that CYP 1A1 and 1B1 are also involved in hy-droxylation of PAHs that are consequently forming carcinogenic metabolites. Buters et al. monitored metabolism of dibenzo-[a,l]pyrene that is a very strong carcinogen in mouse skin and rat mammary glands and found that CYP1B1 is the most critical enzyme for formation of its DNA reactive (-)-(11R,-12S,13S,14R)-dibenzo[a,l]pyre-ne-11,12-dihydrodiol 13,14-epoxide metabolite.20 Cova-lent DNA-adducts of such bulky compounds are removed successfully by cellular repair system, if not DNA adducts can cause mutations of proto-oncogenes. Spencer et al. studied the influence of dibenzo[a,l]pyrene adduct formation on carcinogenesis using nucleotide excision repair +/- cell lines.21
As we can see, majority of metabolic studies were performed on compounds that are endogenous hormones or synthetic compounds used in estrogen replacement therapy. Undoubtedly, chronic, high-dose intake of estrogens for relieving of menopausal symptoms means a highly statistically significant increase in development of breast can-cer.22 However, it should be mentioned that contribution of reactive metabolite formation to estrogen-induced carcino-genesis is still not clear. The general and most widely recognized mechanism of estrogen mediated carcinogenesis is ER-mediated hormonal activity of ER ligands (as growth factors), inducing uterine growth and malignant genital tract changes.23 Mechanistically, hypothesis of estrogen induced carcinogenesis via metabolic activation should be expanded to other classes of "orphan" P450s that have bioactivation potential. Many P450s exist for which functions are not fully explained and as such could have important impact on estrogen bioactivation. Recently, Wang et al. have identified the potential of an largely unknown CYP2W1 to mediate bioactivation to reactive metabolites from fluorinated 2-aryl-benzothiazole analogs.24 Future studies should also focus on other industrial EDCs (eg. bisphenols) and their carcinogenic potential in terms of estrogen activity and their metabolic profile.
2. Interaction of EDCs With ERs
2. 1. Structures and Features of ERs
ERa ligand-binding characteristics show little differences among vertebrates.25-27 ER is a ligand-activated
transcriptional regulator and a member of the nuclear receptor superfamily. Its endogen ligand is an ovarian steroid hormone E2. Two subtypes of ER are known, ERa and ERp, which are coded by separate genes.28 Both possess three major functional domains: the N terminal A/B domain, activation function-1 (AF-1) domain and the DNA binding domain (DBD), which binds the receptor to a DNA helix.28-31
ERa and ERP differ in two features: distribution in the tissues and ligand binding characteristics. This could explain the tissue selective actions of estrogens. ER con-formational changes are induced after ligand binding, followed by homodimerization. We should also mention the fact that many cells express both ERa and ERp, which can form either homo- or heterodimers (heterodimeriza-tion).32 The homodimer binds to a specific estrogen responsive element (ERE), located in the promoter region of the target gene. The role of heterodimerization and ERa/p heterodimers still remains open but it is believed that they regulate ERE, whereas ERp counteracts the stimulatory effects of ERa through heterodimerization (reduced ERa-mediated cell proliferation).33 ER agonists induce transcriptional activation through interaction with coactivators. ER antagonists interact preferentially with corepressor complexes.
Selective ER modulators (SERMs) are synthetic ER ligands that act as ER agonists in certain tissues (bone, liver and cardiovascular system), while in other tissues (brain and breast) as ER antagonists.34-36 In uterus they have mixed agonistic/antagonistic activity. SERMs are chemically categorized into three structural classes: 1) triphenylethylene 2) benzothiophene and 3) benzopyran compounds. The prototype SERM compound from trip-henylethylene class is tamoxifen and is still used for the treatment of hormone-dependent breast cancer. The most commonly used SERM today is raloxifene, chemically a benzothiophene, which is indicated for prevention and treatment of postmenopausal osteoporosis.37 Due to significant presystemic clearance of raloxifene in vivo and potential of forming electrophilic diquinone methides and orho-quinones a lot of effort is made to chemically improve the basic structure. As result Liu et al. found that fluo-ro-substituted desmethyl arzoxifene (4'F-DMA) has improved metabolic profile and represent a safer alternative of SERM.38 It is worth of noting that tamoxifen bioactiva-tion remains controversial due to carbocation metabolic formation and detection of its DNA adducts in endome-trium, what coincides with increased incidence of prema-lignant and malignant endometrial changes.39
Numerous steroidal and non-steroidal compounds can bind to ERs. These compounds have a common phenol group, which is responsible for ER binding, but they contain different core structures. Available ER-ligand 3D structures allow establishment of the structure-activity relationship. ERs can also indirectly alter the expression of genes without directly binding to DNA. This is possible
when interaction with other promoter proteins occurs or by the inhibition of certain transcription factors.40'41
Rapid effects of E2 on a time scale of seconds to minutes could not be explained by genomic effects previously described. It is suggested that they are rather conducted through different signalling pathways and are therefore termed non-genomic effects.42 The presence of membrane impermeable estradiol-protein conjugates may support these effects since membrane initiated estrogen signalling has been reported' implicating the existence of special ERs. In SK-BR-3 breast cancer cell lines that are deficient in nuclear ERs, an orphan receptor, G-protein-coupled ER (GPER; formerly known as GPR30), was discovered.43-45 GPER was also found in MCF-7 cells, macrophages, keratinocytes, brain cells and vascular cells.46-48
Non-genomic effects of E2 were studied on many different cell models. Alyea et al. studied non-genomic estrogen effects on dopamine efflux via dopamine transporter on PC12 cell culture that contains three types of membrane ERs (mERs): mERa, mERp and GPER. Assay was conducted through measuring the 3H-dopamine efflux in the presence of E2, estrone and estriol. Additionally, they showed that for E2-mediated dopamine efflux the protein kinase C and mitogen-activated protein kinases are involved and that the presence of intracellular Ca2+ is required.49 After discovery that platelets contain ERs, Moro et al. studied E2-dependent signaling pathway in platelets. Exposure to E2 leads to rapid phosphorylation of tyrosine kinase Src in platelet membrane and thereby playing an important role in regulating blood aggregation.50-52 Yu et al. showed that non-genomic effects of E2 help preventing the intestinal injury after traumatic bleeding due to restoring the PI-3K activity, which may prevent neutrophil infiltration and consequent harmful intestinal inflamma-tion.53
We should add to this the report on a novel, plasma membrane associated ER with high affinity binding sites named ERX.5455 Walsh et al. have demonstrated rapid non-genomic effects of environmental estrogenic compounds at nanomolar concentrations on intracellular concentrations of Ca2+ which suggests the existence of an alternative pathway.56
2. 2. Concept of EDC Binding to ER
Responses to estrogens and other ligands depend on four main features of ERs: affinity, saturability, ligand specificity and receptor distribution that is tissue specific. Several intracellular pathways are regulated by liganded or non-liganded ERs. Non-liganded ERs may be trans-criptionally activated, for example, by selective phosp-horylation of certain serine residues of ERa.28 Structurally different natural and synthetic ER ligands trigger various conformations that interact with other transcription factors in a different way. Sumbayev et al. screened 2,4-
dichlorodiphenyldichloroethylene (DDE) and 24 other widely distributed pesticides in order to examine whether they induce ER conformational changes. DDE and several other pesticides (endosulfan, dieldrin, vinclozolin, ipro-dione, paclobutrazol, fenarimol, prochloraz) were shown to induce a previously unrecognized ER conformational state, which has properties of both E2 (agonist) and 4-hy-droxy-tamoxifen (partial antagonist) induced confirmation. This means that pesticide liganded ERs are in equilibrium between two states, one favouring co-regulator binding (agonist liganded receptor) and one preventing co-regulator binding (partial antagonist liganded receptor).57 High concentrations of hormones that are above the physiological range may also bind to other hormone receptors. For example, E2 present at very high concentrations can bind to androgen receptors.28,58
In vivo physiological concentrations of endogenous hormones are usually below the dissociation constant (Kd) which makes such a system sensitive and effective with respect to hormone concentration fluctuations.28,59-62 Different responses also require different E2 doses in the same cell systems (e.g. rat pituitary GH3 cells and MCF-7 cells).63,64 In malignant conditions (e.g. MCF-7 cells) estrogen responsiveness is maintained for a sustained period compared with physiological conditions.65-68 In addition, chemicals that are inducing growth of MCF-7 cells at lower concentrations can slow or completely stop their growth because of acute toxicity (toxic concentrations). These dual effects can have a wide interval: bisphenol A and octylphenol ranging 1,000 to 100,000-fold and can result in excess of 100 million-fold for DES and E2.286269-71
3. Aryl Hydrocarbon Receptor (AhR)
AhR is a ligand-dependent transcription factor. Li-gands such as TCDD (Table 1) and 3-MC bind to AhR, and the following sequence of events occurs: translocation into the cell nucleus, dimerization with the AhR nuclear translocator (ARNT) and the formation of AhR/ ARNT heterodimer complex, leading to gene transcription. The AhR/ARNT heterodimer complex binds to the following specific DNA recognition elements: the AhR responsive element (AHRE), the xenobiotic response element (XRE) or dioxin response element (DRE).72 Numbers of signalling pathways are cross-connected with the AhR signalling. Observed anti-estrogenic effects after AhR ligand binding are due to AhR/ERa crosstalk. There is certain evidence that activated AhR induces binding of ERa to AhR-regulated genes and that AhR binds to ERa-regulated genes. Genes such as CYP1A1, CYP1B1 and TiPARP (poly-ADP ribose polymerase) are expressed as a consequence of TCDD binding to AhR. The same genes are also estrogen responsive (human breast cancer cells), suggesting that AhR and ERs regulate the expression of
common genes.73 One of the consequences of AhR/ERa crosstalk is impaired DNA binding of inhibitory dioxin response elements (iDREs) in promoter regions of some E2 responsive genes. AhR agonists also mediate degradation of ERs through activation of the proteasome complex, contributing to anti-estrogenic effects.74 AhR involvement in ubiquitination was confirmed by the inhibition of ER degradation with proteasome inhibitor MG-132.75 Polyubiquitination of ERa in the presence of AhR ligands occurred when ubiquitin E1 and E2 ligases were supplemented in vitro J5'16
Beside antiestrogenic effects, AhR ligands can also have proestrogenic effects. AhR agonist, 3-MC, can activate ERs transcriptional activity (especially ERa) in Hep-G2 or CV-1 cells. It is suggested that these cells are capable of metabolizing parent 3-MC into compounds that act as ER ligands. However, 3-MC itself is not an ER ligand.77
4. Properties of Synthetic Estrogenic EDCs
EDCs are usually polar and hydrophilic with a low octanol/water partition coefficient (KOW). They also have lower binding affinity to organic fractions of sludge or suspended sediments than other persistent organic compounds with higher KOW that are capable of bioaccumulation (PAHs, PCBs, or organochlorine pesticides).78-81
Solubility. Hydrophilic molecules (e.g. diphenyl-methanes) have a tendency to accumulate in aqueous phases and are not distributed in the sediment, body fat or adsorbed to particulate matter in waste water. Compounds with low solubility in water (e.g. p,p'-DDT) are mainly characterized by lipophilic structural fragments. They are largely adsorbed to sediment or particulate matter and tend to accumulate in biological systems, especially in animal fat tissues (Table 2).82
Degradation. EDCs found in the environment are prone to chemical and microbial degradation (e.g. hydrolysis and photolysis). EDCs with half-lives longer than 2 weeks, some even longer than six months, are considered to be the most persistent (e.g. DDT and PCBs in natural waters).82 For comparison, BPA has a relatively short biodegradation half-life, ranging from 2-6 days in rivers and is therefore considered less persistent.84
Bioaccumulation. With the exception of DDT and certain organometallic compounds, EDCs do not accumulate in biological systems significantly.82 Several active ingredients of oral contraceptives can be traced in aquatic species causing infertility. Active ingredient in birth control pills, ethynylestradiol, has been detected in wild fishes, levonorgestrel, active ingredient in post-coital contraception was found in fishes and mussels.85 Interestingly, plasma levels of levonorgestrel in fish even exceeded human therapeutic levels.86 Bioaccumulation in marine animals could present a serious treat for human health because they represent an important source of human nutrition. Relations between ingestion of EDCs and their effects on human cells are very complex and should be studied in more detail (eg. different cell lines with EDCs mixtures). Lasserre at al. studied effects of selected EDCs atrazine and 2, 2', 4, 4', 5, 5'-hexachlorobiphenyl on human MCF-7 cells. Expression of many proteins involved in oxidative stress, DNA repair, cell shape, spermatogenesis were affected.87
5. Structural Classes of Estrogenic EDCs
Among the ER ligands there are several major categories classified according to their chemical structural characteristics (Table 1). Steroids can be divided in two groups. One group has a phenolic A ring (e.g. E2, ethynylestradiol), while the other one lacks a phenolic A ring (e.g. norethynodrel) (Figure 1). Diethylstilbestrol (DES)-like compounds feature two benzene rings separated by two carbons that are connected by a double bond in DES derivatives (Figure 4), however hexestrol contains only a single bond. Triphenylethylene derivatives constitute a group of compounds with an additional phenyl group attached to the ethylene bridge, a common property of synthetic anti-estrogens. The Diphenylmet-hane class includes diphenolalkanes (e.g. bisphenol A and its analogs). Biphenyls are chemicals with two connected phenyl rings that may or may not contain halogens (chlorine or bromine). All compounds in the class of phenols contain a single phenolic ring, most of them also contain a long alkyl chain substituted at the para position (e.g. p-nonylphenol). In addition to ER ligands two structural classes of AhR ligands are considered: PAHs and dioxins (Figure 9).
Table 2. Comparison of water solubility and log KOW of selected EDCs
Chemical class	Example of compounds	Solubility in water	log KOW
Diphenylmethanes	Bisphenol A	120 mg/La 83	3,3283
Steroids	Ethynylestradiol	11,3 mg/Lb 83	3,6783
PAHs	Benzo[a]pyrene	1,62 x 10-3 mg/La 83	6,1483
Diphenylmethanes	p.p'-DDT	5,5 x 10-3 mg/La 83	6,9183
a - at 25 °C b - at 27 °C
5. 1. Steroids
The most active natural estrogen is E2 (Figure 1), a typical prototype for a steroidal compound that acts on ERs and contains a hydrophobic core with two OH groups. EDCs that bind to ERs are actually structural mi-metics of endogenous E2. Crystal studies of E2 bound to ER showed that the 3-OH and 17P-OH groups act as H-bond donors and acceptors, respectively. Elimination or modification of these two OH groups significantly reduces ER binding affinity.2,3,88 The impact of elimination and modification is more drastic at the 3-position than at the 17P-position (Figure 1). Other steroidal hormones like testosterone or progesterone do not bind to ER under physiological conditions because they lack an aromatic hydroxyl group. The precise distance between the 3- and 17P-OH groups of steroids and their spatial orientation significantly impacts binding affinity. The optimal distance between the 3- and 17P-OH groups ranges from 9.7 to 12.3 A. For comparison the oxygen-oxygen distance in E2 is 10.9 A.89 Additionally, steric hindrance at the 7a- and 1ip-position of the steroid backbone is also an important feature. The volume of the ER binding pocket is about
twice that of E2. Large unoccupied cavities at 7a- and 11P- positions of E2 allow bulky groups to fit in ERs. This observation is of great importance for xenoestrogens such as DES-like chemicals (Figure 4), diphenylmethanes (Figure 5), and biphenyls (Figure 7). For example, the binding affinities of DES, dimethylstilbestrol (DMS) and 4,4-dihydroxystilbene decrease with shortening of the alkyl side chains that mimic the 7a- and 11P- positioned bulky groups. The addition of small substituents such as 11 P-chloromethyl, ethyl and vinyl groups to the 11 P-po-sition usually increases ER binding affinity.2,88,89
Metabolism. The nature of E2 metabolism is entirely oxidative (Figure 2). The most frequent hydroxyla-tion site is C-2, which is followed by C-16 and C-4.90-92 The least frequent sites are at C-6, C-7 and C-15. All these hydroxylations are catalyzed by the specific cytochrome P450 with great stereospecificity, since hydroxylations can occur a (below) or P (above) the plane of the steroid backbone. Among the most common P450s that take part in hydroxylations are CYP1A1 for C-2, CYP1B1 for C-4 and CYP3A4/5 for C-16.93 P450 polymorphic forms differ significantly in reaction rates from that of the major isoform found in populations.94,95
a)
b)
Figure 1. Steroids (a) with and (b) without phenolic structure (ring A); In each case the ratio of its binding affinity to that of E2 is expressed as a percentage given in parentheses.2
Figure 2. The most common P450s that take part in C-2, C-4 and C-16 hydroxylations of E2 in vitro96-98
Enzyme catechol-O-methyltransferase (COMT) is responsible for O-methylation and therefore inactivation of catecholamine neurotransmitters with S-adenosyl-L-methionine (SAM) as a cofactor. However, any compound with a catechol structure is a potential substrate for COMT. In vivo methylation of E2 at C-2, C-4 or less frequently at C-3, results in metabolites with various activities. For example, 2-methoxyestradiol has minimal estro-genic activity but very promising antitumor effects.99,100 Alternatively, the catechol structures in estrogenic compounds are able to form electrophilic semiquinones and quinones, which can react with glutathione (GSH) and other nucleophilic compounds. While the formation of GSH conjugates in vivo may act as a detoxification mechanism, the reaction of semiquinones with DNA leads to stable ad-ducts in the case of the 2,3-catechols, but in the case of the 3,4-catechols depurination adducts are formed, increasing
mutation rate.101-103 16a-OH-Estrone has ability to react irreversibly to histones and ERs through the so called Heyns rearrangement, contributing to continuous activation of estrogen signaling.104-106
5. 2. Synthetic Non-steroidal Compounds
Synthetic non-steroidal estrogens retain phenolic function, but have a remarkable range of structural motifs in core regions that encompass the following possible fragments: simple acyclic, variety of alicyclic systems and he-terocycles. Efforts have been made to optimize ER ligand structures to improve tissue selectivity, especially for therapeutic purposes. For example bazedoxifene has improved tissue selectivity with little or no agonistic activity in the uterus in comparison with other SERMs.107 Synthetic non-steroidal ER agonists have a common structural model and composition: (1) a core structure, (2) an essential
phenolic unit that should not be chemically altered to retain optimal ER binding affinity,108 (3) an optional second aromatic group and (4) one or two optional substituents
- one of them could be aromatic (Figure 3). It should be mentioned that ER antagonists and partial agonists contain a polar group on the aromatic ring. To achieve high ERbinding affinity, the peripheral groups (2-4) must display a certain geometric arrangement.
Fink et al. have incorporated certain sub-structural motifs into 1,2- and 1,3-azole systems that are outlined in Figure 3. All heterocycles binding affinities to ER were determined with competitive radiometric binding assay using tritium [3H] labelled estradiol.109 They discovered that the homodibenzyl structure is a common structural element in non-steroidal ligands such as benzestrol and raloxifene and can be substituted with sterically very similar 3,5-diaryl-1,2pyrazoles or isoxazoles and 2,4-di-aryl-1,3-imidazoles, thiazoles or oxazoles. The dibenzyl
Figure 3. Common structures found in ER ligands and proposed experimental analogues109
structure is a common structural element found in hydroxytamoxifen that can be mimicked with various 4,5-diaryl-1,3-azoles. The diazole N,N-systems (namely pyra-zoles and imidazoles) can accommodate up to four peripheral substituents, whereas the N,O- and N,S-heterocycles (oxazoles, isoxazoles and thiazoles) can have a maximum of three attached substituents.109
Studies from Fink et al. are important in two ways. First, they have confirmed the hypothesis that bulky core structures contribute to ER binding affinity and second, that simple heterocycles can effectively replace the complex core structures of steroids retaining ER binding.
5. 2. 1. DES-like Chemicals
Diethylstilbestrol (DES, Figure 4) is one of the highest-affinity synthetic estrogens, with a relative binding
affinity (RBA) of 400, compared to E2 with an RBA of 100. DES molecule seems to have an optimal spatial arrangement for hydrophobic and H-bond interactions.2 In the case where the OH group is being methylated, ER binding affinity decreases due to loss of H-bonding capability similar to E2. When both OH groups of DES are methylated, the reduction ER binding affinity is even more significant. The two ethyl groups (bulky core groups) increase ER binding activity of DES as significantly as its phenolic OH groups. The two ethyl substituents increase hydrophobicity, maintain rigidity and help to occupy the ER binding pocket. Binding affinity decreases from DES to dimethylstilbestrol and 4,4'-dihydroxystilbene. The ethyl groups may contribute to ER binding. Both ethyl groups resemble the lip- and 7a-substituents of E2.2,109
DES was synthesized in the 1930s and was used as an estrogen supplement, but it was banned in the USA af-
HO
SG
GSH conjugate of DES Figure 4. In vitro and in vivo metabolism of DES by cytochrome P450 or peroxidase110114
ter identification of its adverse carcinogenic effects. Moreover, in utero exposure to DES leads to increased incidence of cervical cancer in adult females that are exposed to DES before birth.110 The mechanism of the carcinogenesis is still not known. One hypothesis suggest that reactive metabolites of DES might play a role, since it is metabolized to a number of reactive metabolites, including its oxidative metabolite, diethylstilbestrol-4',4"-quinone (DQ), catalysed by P450 or peroxidase. DQ is further metabolized to a non-carcinogenic metabolite Z,Z-dienestrol (ZZ-DIEN) (Figure 4).110-113 DES is also metabolized to its catechol, 3'-OH-DES that intercalates into DNA and is then enzymatically oxidized to its quinone which reacts with DNA. The resulting compounds are depurinating ad-ducts 3'-OH-DES-6'-N3Ade and 3'-OH-DES-6'-N7Gua, analogous to those formed by the natural estrogens.114
5. 2. 2. Triphenylethylene Derivatives
Triphenylethylenes act as anti-estrogens (ER antagonists) and are chemically similar to DES but have an additional phenyl group attached to the ethylene moiety. In the case of ER antagonists and mixed agonist/antagonists, one of the substituents generally contains a basic or polar function. The prototype of this group is tamoxifen, which is used in breast cancer therapy.108,109
Metabolism. Tamoxifen is a model substance for the triphenylethylene derivatives and is extensively metabolized by human liver enzymes to several primary and secondary metabolites. CYP3A4/5 is the major P450 isoform responsible for the formation of N-desmethyltamo-xifen. Formation of 4-hydroxytamoxifen or endoxifen is mainly catalyzed by CYP2D6. Other P450 isoforms play a less important role in tamoxifen metabolism. SULT1A1 has been proposed to form an sulphate of 4-hydroxytamo-xifen thereby contributing to endoxifen clearance.115117
5. 2. 3. Diphenylmethanes
Diphenylmethanes are chemicals with two benzene rings separated by one carbon atom that contain a 4-OH substituent that is critical for binding to ERs. There are
Bisphenol A (R^CHg and R2=CH3) Bisphenol F (R^H and R2-H) Bisphenol AF (R^CFa and R2=CF3) Bisphenol B (R^CHj and R2=CH3CH2)
Figure 5. General formula of diphenolalkanes
three groups of diphenylmethanes: diphenolalkanes, ben-zophenones and DDTs. Diphenolalkanes (bisphenols) contain two phenol rings separated by one carbon atom; their generic formula is depicted in Figure 5.
Bisphenols. The most widespread diphenolalkane is bisphenol A (BPA), first synthesized in 1891 by A.P. Dianin. BPA binds to ERp with an approximately 10-fold higher affinity than to ERa. BPA is used as monomer in epoxy-phenol resin synthesis (canned food) and polycarbonate plastic (medical equipment) or as antioxidant in PVC plastic. Two structural elements enable BPA to bind to ERs: a) 4-OH group on the A-phenyl ring and b) a hydrophobic moiety at the central C-atom. For example, bisphenol F has lower affinity to ERs than BPA and bisphenol B. BPA has approximately 10,000 times weaker affinity for ERs than E2.118119
Metabolism. Glucuronidation of BPA is catalyzed with UDP-glucuronosyltransferases (e.g. UGT2B1). The resulting BPA glucuronide is completely without estrogenic activity. Part of BPA is also sulfated by phenol sulphotransferases present in the liver (e.g. ST1A3). BPA sulphate has diminished estrogenic activity.120123 Besides metabolic phase II reactions, there is strong evidence to support the hypothesis that the metabolism of BPA in the human liver involves an oxidative pathway and formation of reactive metabolites (e.g. quinones, semiquino-nes), catalysed mainly by P450s. Jaeg et al. studied BPA metabolism using liver microsomes. Several metabolites have since been discovered: isopropyl-hydroxyphenol, BPA glutathione conjugate, glutathionyl-phenol, glutat-hionyl-4-isopropylphenol and BPA dimers.124 The BPA metabolite bisphenol-o-quinone was also identified and could bind DNA in vitro and in vivo. From these results it was suggested that covalent modification of DNA by in vivo exposure to BPA may be a factor in the induction of hepatotoxicity.125 Moreover, it was suggested that the es-trogenicity of BPA is increased (two to five times) through its biodegradation by rat liver S9 microsomal and cytosolic fractions. The active estrogenic metabolite was confirmed to be 4-methyl-2,4-bis(p-hydroxyp-henyl)pent-1-ene (MBP).126
Dichlorodiphenyltrichloroethanes (DDTs). Dich-lorodiphenyltrichloroethane (DDT), dichlorodiphenyl-dichloroethane (DDD) and dichlorodiphenyldichloroethy-lene (DDE) isomers have structural frameworks similar to that of BPA. The o,p'- isomers are active in binding, while p,p'-isomers are not. The ortho--chlorine of o,p'-iso-mers mimics the steric 11P-substituent of E2 and increases structural rigidity, which favours ER binding. p,p'-DDT is metabolized by rat liver microsomes to p,p'-DDD and p,p'-DDE (Figure 6A). p,p'-DDE was also formed from p,p'-DDD. 1-chloro-2,2-bis(4'-chlorophenyl)ethyle-ne (p,p'-DDMU) metabolite is formed via dehydrochlori-nation of p,p'-DDD.127 Kow values for DDT (o',p'-DDT: 5.65; p,p'-DDT: 5.50), DDD (o,p'-DDD: 4.87; p,p'-DDD: 4.82) and DDE (o,p'-DDE: 5.43; p,p'-DDE: 5.78) impli-
Figure 6a. Reductive metabolism of p,p'-DDT in rat liver microsomes by Kitamura et al.127
Methoxychlor	HPTE
Figure 6b. Kupfer et al. proposed a two step demethylation of methoxychlor by using liver microsomes.130
cate that the higher KOW values for DDT and DDE increase bioaccumulation in the body fats, whereas DDD and DDMU are mainly eliminated via the kidney.128 129 The pesticide methoxychlor represents a "safer" version of DDT and also exerts estrogenic effects but is known to induce the enzyme aromatase. It needs metabolic activation for endocrine activity. The resulting active metabolite 2,2-bis(p-hydroxyphenyl)-1,1,1-trichloroethane (HPTE) with bisphenol structure can act as an ERa agonist. Although it has many toxic endocrine effects (e.g. induction of utero-trophic response, reduced mass of ovaries), it is still used as a pesticide because of its faster metabolism and significantly reduced likelihood of bioaccumulation (Figure 6B).130
5. 2. 4. Biphenyls
Biphenyls are divided into two types of compounds: polychlorinated biphenyls (PCBs) and phenylphenols.5 PCBs hydroxylated on position 4 (4-OH-PCBs) are better ER binders than non-hydroxylated PCBs, which is consistent with observations in other chemical classes. With an increasing number of chloro substitutions at the non-phenolic ring, more electron withdrawal is found in the phenolic ring. This results in higher pKa values and better H-bond formation capabilities. 2',3',4',5'-Tetrachloro-4-biphenylol and 2',5'-dichloro-4-biphenylol are the strongest binders in the group (Figure 7).131132
CI CI
CI
2',3',4\5'-tetrachloro-4-biphenylol CI
CI
2\5'-dichloro-4-biphenylol
Figure 7. Structures of 4-OH-PCBs.
It is evident that PCB compounds with single or multi-orthochlorine substitution patterns that restrict the conformational flexibility of PCBs are among strongest ER binders. From a chemical point of view, PCBs can form rotational isomers since rotation around the central bond is restricted. It was found that three or four ortho-chlorine substituents are required to prevent racemization. Two ortho-chlorine substitutions at the non-phenolic ring also act as small steric substituents and mimic the 1ip-po-sition of E2, leading to improved binding to ERs.133
Metabolism. Hydroxylated PCBs (OH-PCBs), or polychlorobiphenylols, are the main PCB metabolites. Hydroxy-PCBs are formed through arene oxides by P450. OH-PCBs are polar compounds that are quickly eliminated. Their reported concentrations in humans have been 10% to 20% of the PCB level. OH-PCBs can also be found in the plasma of healthy pregnant women (study in Netherlands). In addition to mentioned hydroxylation reactions, bacterial metabolic reactions of PCBs are also important, since they are responsible for the removal of PCBs from the environment. Anaerobic dehalogenation of highly chlorinated congeners in aquatic sediments is an important process (Dehalococcoides ethenogenes and related organisms).134-136
Figure 8a. P450 catalysed i^so-position reaction of para-substituted phenols139
5. 2. 5. Phenols
This class is divided into three types of compounds: alkylphenols, parabens and alkyloxyphenols. The member of this class having the highest binding affinity is nonylp-henol (NP, Table 1).2 Length of the alkyl side chain at para position has a large impact on binding activity. NP is a mixture of isomers with differently branched nonyl side-chain structures. A number of isomers are generated during the synthesis (an isomeric mixture of p-nonylphenols), the three main NP isomers being 2-(4-hydroxyphenyl)-nona-ne, 3-(4-hydroxyphenyl)-nonane and 4-(4-hydroxyp-henyl)-nonane. NP is most frequently used in the synthesis of nonylphenol ethoxylates, which are components of non-ionic detergents, antioxidants and oil additives.137,138
Metabolism. The ipso-position biotransformation reaction of para-substituted phenols (para-substituent is an electron donating alkyl group) is catalyzed by P450 with a quinol intermediate formed via an ipso-addition reaction (Figure 8A).139 When a para-substituent is an electron withdrawing group: halogen, methoxy, nitro, cyano and substituents with carbonyl groups (e.g. acetyl, carboxyl, or benzoyl), a hydroquinone or benzoquinone formation is followed. Estrone and E2 also contain a para-alkylphenol substituent and form quinols through P450 ipso-addition. Ohe et al. confirmed quinol formation from estrone and E2 by CYP1A1, CYP2B6 and CYP2E1 accompanied by 10P-hydroxylation.140 Estrogens that undergo ipso-addition lose their ER-binding activity. An example of para-substituted phenol is NP, that undergoes an ipso-addition reaction catalyzed by P450, where it forms a quinol intermediate. The benzyl position could also be oxidated by P450, a factor giving rise to ipso-substitution reaction (Figure 8B).139
HO C9H19 IPSO-ADDITION^ II
HO
4-nonylphenol
BENZYL-OXIDATION
4-hydroxy-4-nony[cyclohexa-2,5-dienone O
c8h17
HO
4-noriylphenol
OH

crh
6n17

HO' -	HO'
4-hydroxynonanophenone 1-hydroxynoriylpheriol
IPSO-SUBSTITUTION
OH
Figure 8b. P450 catalysed metabolic pathway of nonylphenol1
il
HO^ hydroquinone
6. AhR Ligands
Studies on the AhR, also known as the dioxin receptor, and its ligands began about 50 years ago when drug-metabolizing enzymes were found to be induced by PAHs such as 3-methylchoranthrene (3MC) and TCDD (Figure 9).74 The AhR was originally identified as a ligand activated transcription factor involved in induction of the xeno-biotic-metabolizing cytochrome CYP1A1. It is known that compounds that bind and activate the AhR are hydrophobic molecules (e.g. synthetic PAHs and HAHs). Many agonists are planar compounds and coplanarity is one of the most important determinants affecting the affinity of the AhR for its ligand.75 Much attention is given to the intrinsic functions of AhR and its natural or endogenous ligands. Nguyen et al. described different structural types of compounds that act like endogenous activators of AhRs such as indigo and indirubin, equilenin (an equin estrogen), arachidonic acid metabolites (e.g. PGG2 and li-poxin A4), heme metabolites (e.g. bilirubin and biliruver-din), tryptophan derivatives (e.g. tryptamine, indole-3-acetic acid (IAA)), UV photoproducts of tryptophan (e.g. 6-formylindolo[3,2-fc]carbazole (FICZ)) and AhR agonists derived from dietary sources (e.g. indolo[3,2-b]car-bazole (ICZ) and 3,3'-diindolylmethane (DIM)).75141 Interestingly, it has been found that harmine and its main metabolite, harmol (p-carboline compounds) significantly inhibit the induction of CYP1A1 by AhRs agonist dioxin. Several AhR antagonists have shown promising results against several carcinogen-activating agents.142
6.1. Chlorinated Aromatic Hydrocarbons
TCDD is a prototype compound of chlorinated aromatic hydrocarbons causing a large number of toxic effects: immunotoxicity (strong lymphoid atrophy), cance-rogenesis, teratogenesis, skin disorders (chloracne) and infertility. Chlorinated aromatics hydrocarbons are chemically stable compounds with a great bioaccumulation and sorption potential.143 In the sections above we already described TCDD as a potent AhR-dependent inhibitor of estrogen activity, whereas PCBs exert pro-estrogenic activity.
6. 1. 1. Dioxins
Dioxins are a common generic name for polychlori-nated dibenzodioxins (PCDDs) and polychlorinated di-benzofuranes (PCDFs), a group of toxic organic contaminants (Figure 9). They are not produced for commercial purposes, but are generated as byproducts in the production of chlorophenols, herbicides (chlorophenoxy acids), and PCBs. Other sources are the paper industry (chlorine bleaching), automobile exhaust (leaded fuel), burning household and industrial waste and the manufacture of PCBs.144 Most research has been carried out on a repre-
sentative compound TCDD. Toxicity of PCBs, PCDFs and PCDDs correlates with AhR binding affinity. 3D QSAR studies revealed that sterically favoured regions are present near the 2, 3, 7 and 8 positions of TCDD (this is also the case for PCDFs). However, bulky groups in the medial position (1,4-dioxane ring) of the second benzene ring are particularly unfavourable for binding. When regions near the 2,3 position in TCDD and PCDF are substituted by electronegative halogen molecules the activity increases, but on the medial position of the ring a positive charge is favoured. The first benzene ring makes a greater contribution to the hydrophobic interaction with AhRs than the second benzene ring.145
6. 1. 2. PAHs
PAHs constitute a major class of environmental organic pollutants. Incomplete combustion of any organic fuel (coal, diesel, gasoline, or biomass) can result in the formation of PAHs. Five PAHs, benz [a] anthracene, ben-zo[a]pyrene, benzo[b]fluoranthene, dibenz[a,h]anthrace-ne and indeno[l,2,3-c,d]pyrene, are known human carcinogens (Figure 9).146
Several QSARs studies have been conducted to examine the influence of molecular structure on biodegradation rates, photo-induced toxicity and mutagenicity of PAHs in human cells. Shape, electronic energy and substitution have been determined as significant factors. PAHs are known to bind to AhRs. However, several 4- and 5-ring polycyclic PAHs, heterocyclic PAHs and their hydroxy derivatives were examined to determine their ability to interact with ERa and ERp. It was determined that only compounds with a hydroxy group were able to compete with 3H-labelled E2 for binding to glutathione-S-transferase, human ERa, the D, E, and F domain fusion protein or to the full-length human ERp. PAH structures are homogeneous, but their carcinogenic effects are diverse, ranging from highly active to inactive. Dipple et al. proposed a classification into four classes according to their power: inactive, slight, moderate and high.147,148 A consistent explanation for the carcinogenic action of many PAHs was given by the so-called K-L-M-"bay region" theory. Schmidt et al. hypothesized that the carcinogenicity of PAHs is related to the distribution of n-electrons.149 For example PAHs with high n-electron density in the L-re-gion are strong carcinogens. Pullman et al. pointed out the coupling of high n-electron density in the K-region and low density in the L-region to be a determining factor for high carcinogenicity.150,151 Jerina et al. have found that stabilization of the "bay region" cation is relevant for carcinogenicity.152 Thus, the diolepoxide formed at the "bay region" was established as the main factor responsible for the toxic action, since it is believed to react directly with a nucleic acid base and give rise to cancer.146,153-155
Metabolism. It is well known that, in vivo, the metabolic processes of PAHs lead to reactive, electrophilic
Figure 9. Structures of PCDDs, PCDFs and five PAHs that are known human carcinogens
intermediates that can initiate cancer. In the commonly accepted chemical carcinogenesis, a PAH is converted in a stepwise process, by three metabolizing enzymes to a reactive diolepoxide which then intercalates rapidly with the deoxyguanosines or deoxyadenosines in DNA. The first and essential step in DNA adduct formation is a non-covalent interaction of PAH with DNA. The study of such interactions is of great importance in understanding the detailed mechanism of PAH carcinogenesis. Benzo-[a]pyrene, a lead compound for PAHs, is transformed to different arenoxides. Studies of carcinogenicity have revealed that the 7,8-epoxid and 7,8-dihydrodiol forms are proximity mutagens and that 7,8-diol-9,l0-epoxid is a powerful, ultimate carcinogen.155,156
7. Conclusions and Future Perspectives
Development in the field of estrogenic compounds has evolved significantly from the first synthetic hormones and their use in contraceptives, through aromatase inhibitors used in cancer therapy, and selective estrogen receptor modulators used in hormone replacement therapy, cancer and osteoporosis treatment. Conversely, toxicolo-
gists are facing the problem of characterizing EDCs in the environment (mainly industrial chemicals) that present health risks for the population in the form of hormone disruption that affects reproduction and growth. There are a multitude of endocrine disrupters that have widely varying effects and are present in a variety of environments. The persistence of EEDCs in the aquatic environment is governed by biodegradation, sunlight photolysis and other abiotic transformations like hydrolysis. Once they reach surface waters, they can be transformed, mainly via bio-or photo-degradation, or they can be adsorbed onto suspended particles in the aquatic environment. Sorption of hydrophobic contaminants at the particle water interface is one of the most important processes that control their distribution in aquatic environments. A major issue, in terms of metabolism, is the uncertain fate of EDCs through ingestion of food. The increased duration of EDC exposure represents a major risk factor for breast cancer. The underlying mechanisms in the susceptibility of breast tissue to the carcinogenic effect of estrogens remain unclear. It has been shown that estrogen metabolism under normal physiological conditions leads to formation of reactive oxygen species (ROS) playing important roles in cell transformation, cell cycle, migration and invasion (breast cancer). The redox cycling of hydroxylated estro-
gens and their ROS production is especially important.157158 However, it may be suggested that unintended exogenous intake of estrogens (EDCs) could lead to excessive formation of ROS, causing oxidative stress and tissue damage. The many examples given here of EDCs forming various oxidized species emphasizes the need for more extensive studies of their oxidative metabolisms.
8. References
1.	E. K. Shanle and W. Xu, Chem Res Toxicol. 2011, 24, 6-19.
2.	H. Fang, W. Tong, L. M. Shi, R. Blair, R. Perkins, W. Bran-ham, B. S. Hass, Q. Xie, S. L. Dial, C. L. Moland, D. M. Sheehan, Chem Res Toxicol. 2001, 14, 280-94.
3.	A. M. Brzozowski, A. C. Pike, Z. Dauter, G. L. Greene, J. A. Gustafsson, M. Carlquist, Nature 1997, 389, 753-758
4.	A. C. W. Pike, A. M. Brzozowski, R. E. Hubbard, T. Bonn, A. G. Thorsell, O. Engstrom, J. Ljunggren, J. A. Gustafsson, M. Carlquist, The EMBO Journal 1999, 18, 4608-4618.
5.	E. Swedenborg, M. Kotka, M. Seifert, J. Kanno, I. Pon-gratz, J. Ruegg, Mol Cell Endocrinol. 2012, 362, 39-47.
6.	M. Wormke, M. Stoner, B. Saville, K. Walker, M. Abdelra-him, R. Burghardt, S. Safe, Mol Cell Biol. 2003, 23, 1843-55.
7.	A. J. Palumbo, M. S. Denison, S. I. Doroshov, R. S. Tjeer-dema, Environ Toxicol Chem. 2009, 28, 1749-55.
8.	J. T. Sanderson, R. J. Letcher, M. Heneweer, J. P. Giesy, M. van den Berg, Environ. Health Perspect. 2001, 109, 1027-1031.
9.	J. T. Sanderson, J. Boerma, G. W. A. Lansbergen, M. van den Berg, Toxicol. Appl. Pharmacol. 2002, 182, 44-54.
10.	J. L. Bolton and G. R. J. Thatcher, Chem. Res. Toxicol. 2008, 21, 93-101.
11.	J. L. Bolton, M. A. Trush, T. M. Penning, G. Dryhurst, T. J. Monks, Chem Res Toxicol. 2000, 13, 135-60.
12.	J. Russo, M. Hasan Lareef, G. Balogh, S. Guo, I. H. Russo, J. Steroid Biochem. Mol. Biol. 2003, 87, 1-25.
13.	R. R. Newbold and J. G. Liehr, Cancer Res. 2000, 60, 235-7.
14.	Z. Wang, E. R. Chandrasena, Y. Yuan, K. W. Peng, R. B. van Breemen, G. R. Thatcher, J. L. Bolton, Chem Res Toxicol. 2010, 23, 1365-73.
15.	A. M. Wilson and G. A. Reed, Carcinogenesis. 2001, 22, 257-63.
16.	S. Mesia-Vela, R. I. Sanchez, J. J. Li, S. A. Li, A. H. Con-ney, F. C. Kaufman, Carcinogenesis. 2002, 23, 1369-72.
17.	S. R. Beedanagari, R. T. Taylor, P. Bui, F. Wang, D. W. Nic-kerson, O. Hankinson, Mol Pharmacol. 2010, 78, 608-16.
18.	D. C. Spink, F. Zhang, M. M. Hussain, B. H. Katz, X. Liu, D. R. Hilker, J. L. Bolton, Chem Res Toxicol. 2001, 14, 572-81.
19.	Z. Wang, P. Edirisinghe, J. Sohn, Z. Qin, N. E. Geacintov, G. R. Thatcher, J. L. Bolton, Chem Res Toxicol. 2009, 22, 1129-36.
20.	J. T. Buters, B. Mahadevan, L. Quintanilla-Martinez, F. J. Gonzalez, H. Greim, W. M. Baird, A. Luch, Chem Res Toxicol. 2002, 15, 1127-35.
21.	W. A. Spencer, J. Singh, D. K. Orren, Chem Res Toxicol. 2009, 22, 81-9.
22.	G. A. Colditz, J Natl Cancer Inst. 1998, 90, 814-23.
23.	B. E. Henderson, R. Ross, L. Bernstein, Cancer Res. 1988, 48, 246-53.
24.	K. Wang and F. P. Guengerich, Chem Res Toxicol. 2012, 25, 1740-51.
25.	B. S. Katzenellenbogen, J. A. Katzenellenbogen, D. Morde-cai, Endocrinology 1979, 105, 33-40.
26.	F. Pakdel, C. Le Guellec, C. Vaillant, M. G. Le Roux, Y. Va-lotaire, Mol Endocrinol 1989, 3, 44-51.
27.	R. White, S. Jobling, S. A. Hoare, J. P. Sumpter, M. G. Parker, Endocrinology 1994, 135, 175-182.
28.	W. V. Welshons, K. A. Thayer, B. M. Judy, J. A. Taylor, E. M. Curran, F. S. vom Saal, Environmental Health Perspectives 2003, 111, 994-1006.
29.	E. Enmark, M. Pelto-Huikko, K. Grandien, S. Lagercrantz, J. Lagercrantz, G. Fried, M. Nordenskjold, J. A. Gustafsson, J. Clin. Endocrinol. Metab. 1997, 82, 4258-4265.
30.	I. Rogatsky, J. M. Trowbridge, M. J. Garabedian, J. Biol. Chem. 1999, 274, 22296-22302.
31.	P. J. Balfe, A. H. McCann, H. M. Welch, M. J. Kerin, EJSO
2004,	30, 1043-1050.
32.	J. M. Hall and D. P. McDonnell, Endocrinology 1999, 140, 5566-5578.
33.	E. Powell and W. Xu, Proc Natl Acad Sci USA 2008, 105, 19012-7.
34.	X. Li, J. Huang, P. Yi, R. A. Bambara, R. Hilf, M. Muyan, Mol Cell Biol. 2004, 24, 7681-94.
35.	D. G. Monroe, F. J. Secreto, M. Subramaniam, B. J. Getz, S. Khosla, T. C. Spelsberg, Mol Endocrinol. 2005, 19, 1555-68.
36.	E. Powell, E. Shanle, A. Brinkman, J. Li, S. Keles, K. B. Wisinski, W. Huang, W. Xu, PLoS One 2012, 7, 1-13.
37.	T. S. Dowers, Z. H. Qin, G. R. Thatcher, J. L. Bolton, Chem Res Toxicol. 2006, 19, 1125-37.
38.	H. Liu, J. L. Bolton, G. R. Thatcher, Chem Res Toxicol. 2006, 19, 779-87.
39.	S. Y. Kim, N. Suzuki, Y. R. Laxmi, B. P. McGarrigle, J. R. Olson, M. Sharma, M. Sharma, S. Shibutani, Chem Res To-xicol. 2005, 18, 889-95.
40.	S. Nilsson and J. A. Gustafsson, Breast Cancer Res. 2000, 2, 360-366.
41.	A. C. Pike, Best Pract Res Clin Endocrinol Metab. 2006, 20, 1-14.
42.	J. Jacob, K. S. Sebastian, S. Devassy, L. Priyadarsini, M. F. Farook, A. Shameem, D. Mathew, S. Sreeja, R. V. Thampan, Mol Cell Endocrinol. 2006, 246, 34-41.
43.	M. H. Wyckoff, K. L. Chambliss, C. Mineo, I. S. Yuhanna, M. E. Mendelsohn, S. M. Mumby, P. W. Shaul, J. Biol. Chem. 2001, 276, 27071-27076.
44.	P. Thomas, Y. Pang, E. J. Filardo, J. Dong, Endocrinology
2005,	146, 624-632.
45.	E. R. Prossnitz, J. B. Arterburn, L. A. Sklar, Molecular and Cellular Endocrinology. 2007, 265-266, 138-142.
46.	M. R. Meyer, E. R. Prossnitz, M. Barton, Vascul Pharmacol. 2011, 55, 17-25.
47.	E. R. Prossnitz, J. B. Arterburn, H. O. Smith, T. I. Oprea, L. A. Sklar, H. J. Hathaway, Annu Rev Physiol. 2008, 70, 165-90.
48.	E. R. Prossnitz and M. Barton, Nat Rev Endocrinol. 2011, 7, 715-26.
49.	R. A. Alyea and C. S. Watson, BMCNeurosci. 2009, 10, 59.
50.	L. Moro, S. Reineri, D. Piranda, D. Pietrapiana, P. Lova, A. Bertoni, A. Graziani, P. Defilippi, I. Canobbio, M. Torti, F. Sinigaglia, Blood. 2005, 105, 115-21.
51.	G. Khetawat, N. Faraday, M. L. Nealen, K. V. Vijayan, E. Bolton, S. J. Noga, P. F. Bray, Blood. 2000, 95, 2289-96.
52.	I. Anwaar, M. Rendell, A. Gottsäter, F. Lindgärde, U. L. Hulthen, I. Mattiasson, J Intern Med. 2000, 247, 463-70.
53.	H. P. Yu, Y. C. Hsieh, T. Suzuki, M. A. Choudhry, M. G. Schwacha, K. I. Bland, I. H. Chaudry, J Leukoc Biol. 2007, 82, 774-80.
54.	C. D. Toran-Allerand, Exp. Gerontol. 2004, 39, 1579-1586.
55.	C. D. Toran-Allerand, X. Guan, N. J. MacLusky, T. L. Hor-vath, S. Diano, M. Singh, E. S. Connolly Jr., I. S. Nethrapal-li, A. A. Tinnikov, J. Neurosci. 2002, 22, 8391-8401.
56.	D. E. Walsh, P. Dockery, C. M. Doolan, Mol Cell Endocri-nol. 2005, 230, 23-30.
57.	V. V. Sumbayev, E. C. Bonefeld-J0rgensen, T. Wind, P. A. Andreasen, FEBS Lett. 2005, 579, 541-8.
58.	T. O. Fox, Proc Natl Acad Sci USA 1975, 72, 4303-4307.
59.	F. S. vom Saal, JAnim Sci 1989, 67, 1824-1840.
60.	F. S. vom Saal, B. G. Timms, M. M. Montano, P. Palanza, K. A. Thayer, S. C. Nagel, M. D. Dhar, V. K. Ganjam, S. Parmigiani, W. V. Welshons, Proc Natl Acad Sci USA 1997, 94, 2056-2061.
61.	M. M. Montano, W. V. Welshons, F. S. vom Saal, Biol Re-prod 1995, 53, 1198-1207.
62.	W. V. Welshons, S. C. Nagel, K. A. Thayer, B. M. Judy, F. S. vom Saal, Toxicol Ind Health 1999, 15, 12-25.
63.	J. F. Amara and P. S. Dannies, Endocrinology 1983, 112, 1141-1143.
64.	W. V. Welshons and V. C. Jordan, Eur J Cancer Clin Oncol 1987, 23, 1935-1939.
65.	H. D. Soule, J. Vazquez, A. Long, S. Albert, M. Brennan, J Natl Cancer Inst 1973, 51, 1409-1413.
66.	M. E. Lippman and G. Bolan, Nature 1975, 256, 592-593.
67.	M. E. Lippman, G. Bolan, K. Huff, Cancer Res 1976, 36, 4595-4601.
68.	W. V. Welshons, C. S. Murphy, R. Koch, G. Calaf, V. C. Jordan, Breast Cancer Res Treat 1987, 10, 169-175.
69.	W. V. Welshons, K. S. Engler, J. A. Taylor, L. H. Grady, E. M. Curran, J Cell Physiol 1995, 165, 134-144.
70.	J. A. Taylor, L. H. Grady, K. S. Engler, W. V. Welshons, Breast Cancer Res Treat 1995, 34, 265-277.
71.	S. Oesterreich, P. Zhang, R. L. Guler, X. Sun, E. M. Curran, W. V. Welshons, C. K. Osborne, A. V. Lee, Cancer Res 2001, 61, 5771-5777.
72.	Y. Kanno, Y. Takane, Y. Takizawa, Y. Inouye, Mol Cell Endocrinol. 2008, 291, 87-94.
73.	S. Ahmed, E. Valen, A. Sandelin, J. Matthews, Toxicol Sci.
2009,	111, 254-66.
74.	F. Ohtake, Y. Fujii-Kuriyama, K. Kawajiri, S. Kato, J Steroid Biochem Mol Biol. 2011, 127, 102-7.
75.	Y. Fujii-Kuriyama and K. Kawajiri, Proc Jpn Acad Ser B Phys Biol Sci. 2010, 86, 40-53.
76.	U. A. Bussmann and J. L. Barañao, Biol Reprod. 2006, 75, 360-9.
77.	E. Swedenborg, J. Rüegg, A. Hillenweck, S. Rehnmark, M. H. Faulds, D. Zalko, I. Pongratz, K. Pettersson, Mol Pharmacol. 2008, 73, 575-86.
78.	F. A. Valiman and M. Gavrilescu, Clean 2009, 37, 277-303.
79.	E. Means, A. Ghosh, Z. Chowdhury (Eds.), Endocrine Disruptor and Pharmaceuticals Strategic Initiative Expert Workshop Report, Marina del Rey, California, USA, 2007, pp. 1-111.
80.	C. G. Daughton and T. A. Ternes, Environ. Health Perspective 1999, 107, 907-938.
81.	W. L. Sun, J. R. Ni, T. T. Liu, Water, Air, Soil Pollut. 2006, 6, 583-591.
82.	P. D. Anderson (Eds.), Technical Brief: Endocrine Disrupting Compounds and Implications for Wastewater Treatment. Water Environment Research Foundation, London, United Kingdom, 2005, pp. 11-12.
83.	http://toxnet.nlm.nih.gov (assessed: March 16, 2012)
84.	G. M. Klecka, S. J. Gonsior, R. J. West, P. A. Goodwin, D. A. Markham, Environ Toxicol Chem. 2001, 20, 2725-35.
85.	A. M. Al-Ansari, A. Saleem, L. E. Kimpe, J. P. Sherry, M. E. McMaster, V. L. Trudeau, J. M. Blais, Environ Pollut.
2010,	158, 2566-71.
86.	V. Contardo-Jara, C. Lorenz, S. Pflugmacher, G. Nützmann, W. Kloas, C. Wiegand, Environ Pollut. 2011, 159, 38-44.
87.	J. P. Lasserre, F. Fack, D. Revets, S. Planchon, J. Renaut, L. Hoffmann, A. C. Gutleb, C. P. Muller, T. Bohn, J Proteome Res. 2009, 8, 5485-96.
88.	D. P. Ayan, R. Maltais, J. Roy, D. Poirier, EndocrRev 2011, 32, 1-17.
89.	G. M. Anstead, K. E. Carlson, J. A. Katzenellenbogen, Steroids. 1997, 62, 268-303.
90.	P. H. Jellinck and J. Fishman, Steroids. 1984, 43, 559-569.
91.	P. H. Jellinck, B. Norton, J. Fishman, J. Steroid Biochem. 1984, 21, 361-365.
92.	P. H. Jellinck, E. F. Hahn, B. I. Norton, J. Fishman, Endocrinology. 1984, 115, 1850-1856.
93.	P. H. Jellinck and J. Fishman, Biochemistry. 1988, 27, 6111-6116.
94.	E. Taioli, H. L. Bradlow, S. V. Garbers, D. W. Sepkovic, M. P. Osborne, J. Trachman, S. Gangulyl, S. J. Garte, Cancer Prevention Detection. 1999, 23, 232-237.
95.	J. Fishman, H. L. Bradlow, T. F. Gallagher, J. Biol. Chem. 1960, 235, 3104-3107.
96.	A. J. Lee, M. X. Cai, P. E. Thomas, A. H. Conney, B. T. Zhu, Endocrinology. 2003, 144, 3382-3398.
97.	S. Lieberman, J. Steroid Biochem. Mol. Biol. 2008, 109, 197-199.
98.	B. T. Zhu and A. J. Lee, Steroids. 2005, 70, 225-244.
99.	N. J. Lakhani, M. A. Sarkar, J. Venitz, W. D. Figg, Pharmacotherapy. 2003, 23, 165-172.
100.G.	Z. Han, Z. J. Liu, K. Shimoi, B. T. Zhu, Cancer Res. 2005, 65, 387-393.
101.	E. Cavalieri and E. Rogan, Ann. NY Acad. Sci. 2006, 1089, 286-301.
102.	E. Cavalieri, D. Chakravarti, J. Guttenplan, E. Hart, J. Ingle, R. Jankowiak, P. Muti, E. Rogan, J. Russo, R. Santen, T. Sutter, Biochim. Biophys. Acta. 2006, 1766, 63-78.
103.	E. L. Cavalieri, P. Devanesan, M. C. Bosland, A. F. Badawi, E. G. Rogan, Carcinogenesis. 2002, 23, 329-333.
104.	R. H. Lustig, C. V. Mobbs, H. L. Bradlow, B. S. McEwen,
D.	W. Pfaff, Endocrinology. 1989, 125, 2701-2709.
105.	R. Raftogianis, C. Creveling, R. Weinshilboum, J. Weisz, J. Natl. Cancer Inst. Monogr. 2000, 27, 113-124.
106.	G. E. Swaneck and J. Fishman, Proc. Natl. Acad. Sci. USA. 1988, 85; 7831-7835.
107.	D. M. Biskobing, Clin Interv Aging. 2007, 2, 299-303.
108.	R. Tedesco, J. A. Thomas, B. S. Katzenellenbogen, J. A. Katzenellenbogen, Chem Biol. 2001, 8, 277-87.
109.	B. E. Fink, D. S. Mortensen, S. R. Stauffer, Z. D. Aron, J. A. Katzenellenbogen, Chem Biol. 1999, 6, 205-19.
110.	K. Chae, J. Lindzey, J. A. McLachlan, K. S. Korach, Steroids. 1998, 63, 149-57.
111.	B. Hammes and C. J. Laitman, J Midwifery Womens Health. 2003, 48, 19-29.
112.T.	E. Wiese, D. Dukes, S. C. Brooks, Steroids. 1995, 60, 802-8.
113.	N. Kalfa, F. Paris, M. O. Soyer-Gobillard, J. P. Daures, C. Sultan, Fertil Steril. 2011, 95, 2574-7.
114.B.	Hinrichs, M. Zahid, M. Saeed, M. F. Ali, E. L. Cavalieri,
E.	G. Rogan, J Steroid Biochem Mol Biol. 2011, 127, 276-81.
115.	Y. Jin, Z. Desta, V. Stearns, B. Ward, H. Ho, K. H. Lee, T. Skaar, A. M. Storniolo, L. Li, A. Araba, R. Blanchard, A. Nguyen, L. Ullmer, J. Hayden, S. Lemler, R. M. Weinshil-boum, J. M. Rae, D. F. Hayes, D. A. Flockhart, J Natl Cancer Inst. 2005, 97, 30-9.
116.Z.	Desta, B. A. Ward, N. V. Soukhova, D. A. Flockhart, J Pharmacol Exp Ther. 2004, 310, 1062-75.
117.	M. Hiratsuka, Y. Agatsuma, F. Omori, K. Narahara, T. Ino-ue, Y. Kishikawa, Biol Pharm Bull 2000, 23, 1131-5.
118.	L. N. Vandenberg, M. V. Maffini, C. Sonnenschein, B. S. Rubin, A. M. Soto, EndocrRev. 2009, 30, 75-95.
119.	J. G. Hengstler, H. Foth, T. Gebel, P. J. Kramer, W. Lilienblum, H. Schweinfurth, W. Völkel, K. M. Wollin, U. Gun-dert-Remy, Crit Rev Toxicol. 2011, 41, 263-91.
120.	A. Atkinson and D. Roy, Biochem. Biophys. Res. Commun. 1995, 210, 424-433.
121.	R. Elsby, J. L. Maggs, J. Ashby, D. Paton, J. P. Sumpter, B. K. Park, J. Pharmacol. Exp. Ther. 2001, 296, 329-337.
122.	Y. Nakagawa and T. Suzuki, Xenobiotica 2001, 31, 113-123.
123.	M. Shimizu, K. Ohta, Y. Matsumoto, M. Fukuoka, Y. Ohno, S. Ozawa, Toxicol. In Vitro 2002, 16, 549-556.
124.	J. P. Jaeg, E. Perdu, L. Dolo, L. Debrauwer, J. P. Gravedi, D. Zalko, J. Agric. Food Chem. 2004, 52, 4935-4942.
125.	S. Yoshihara, M. Makishima, N. Suzuki, S. Ohta, Toxicol. Sci. 2001, 62, 221-227.
126.	S. Yoshihara, T. Mizutare, M. Makishima, N. Suzuki, N. Fu-jimoto, K. Ifarashi, S. Ohta, Toxicol. Sci. 2004, 78, 50-59.
127.	S. Kitamura, Y. Shimizu, Y. Shiraga, M. Yoshida, K. Sugiha-ra, S. Ohta, Drug Metab Dispos. 2002, 30, 113-8.
128.	Y. N. Xing, Y. Guo, M. Xie, R. L. Shen, E. Y. Zeng, Environ Pollut. 2009, 157, 1382-7.
129.	H. Marquardt and S. G. Schäfer (Ed.): Lehrbuch der Toxikologie, Wissenschaftliche Verlagsgesellschaft, Stuttgart, Germany, 2008, pp. 684-85.
130.	D. Kupfer, W. H. Bulger, A. D. Theoharides, Chem Res To-xicol. 1990, 3, 8-16.
131.	K. Haraguchi, N. Koga, Y. Kato, Drug Metab Dispos. 2005, 33, 373-80.
132.	V. V. Sumbayev, J. K. Jensen, J. A. Hansen, P. A. Andreasen, Mol Cell Endocrinol. 2008, 287, 30-9.
133.I. Kania-Korwel, K. C. Hornbuckle, L. W. Robertson, H. J. Lehmler, Food Chem Toxicol. 2008, 46, 637-44.
134.	S. D. Soechitram, M. Athanasiadou, L. Hovander, A. Bergman, P. J. Sauer, Environ Health Perspect. 2004, 112, 1208-12.
135.	K. Furukawa and N. Kimura, Environ Health Perspect. 1995, 103, 21-3.
136.	D. H. Pieper and M. Seeger, J Mol Microbiol Biotechnol.
2008,	15, 121-38.
137.	D. R. Oros and N. David (Eds.), San Francisco Estuary Regional Monitoring Program for Trace Substances, Identification and Evaluation of Unidentified Organic Contaminants in the San Francisco Estuary, Oakland, California, USA, 2002, pp. 19-22.
138.	B. Balakrishnan, E. Thorstensen, A. Ponnampalam, M. D. Mitchell, Placenta. 2011, 32, 788-92.
139.	Y. Tezuka, K. Takahashi, T. Suzuki, S. Kitamura, S. Ohta, S. Nakamura, T. Mashino, Journal of Health Science. 2007, 53, 552-561.
140.	T. Ohe, M. Hirobe, T. Mashino, Drug Metab Dispos. 2000, 28, 110-2.
141.	L. P. Nguyen and C. A. Bradfield, Chem Res Toxicol. 2008, 21, 102-16.
142.M.	A. Gendy, A. A. Soshilov, M. S. Denison, A. O. El-Kadi, Toxicol Lett. 2012, 208, 51-61.
143.N.	S. Fracchiolla, K. Todoerti, P. A. Bertazzi, F. Servida, P. Corradini, C. Carniti, A. Colombi, A. Cecilia Pesatori, A. Neri, G. L. Deliliers, Toxicology. 2011, 283, 18-23.
144.	J. M. Czuczwa and R. A. Hites, Environ Sci Technol. 1984, 18, 444-50.
145.	A. Ashek, C. Lee, H. Park, S. J. Cho, Chemosphere. 2006, 65, 521-9.
146.	L. R. Wang , Y. Wang, J. W. Chen, L. H. Guo, Toxicology.
2009,	262, 250-7.
147.	A. Dipple, in: C. E. Searle (Ed.): Chemical Carcinogens, ACS Monograph 173, American Chemical Society, Washington, DC, 1976, pp. 245-314.
148. A. Dipple, R. C. Moschel, C. A. H. Bigger, in: C. E. Searle (Ed.): Chemical Carcinogens, ACS Monograph 182, American Chemical Society, Washington, DC, USA, 1984, pp. 41-163.
149.0. Schmidt, Z. Physik. Chem., 1939, 42B, 83-110.
150.	A. Pullman and B. Pullman, Rev. Sci. 1946, 84, 145-158.
151.	A. Pullman and B. Pullman, Adv. Cancer Res. 1955, 3, 117-69.
152.	D. M. Jerina, R. E. Lehr, R. M. Schaefer, H. Yagi, J. M. Karle, D. R. Thakker, A. W. Wood, A. H. Conney, in: H. Hiatt, J. D. Watson, I. Winstin (Ed.): Origins of Human Cancer, Cold Spring Harbor Laboratories, New York, USA, 1977, pp. 639-658.
153.	S. L. Simonich, O. Motorykin, N. Jariyasopit, Chem Biol Interact. 2011, 192, 26-9.
154.	K. C. Fertuck, S. Kumar, H. C. Sikka, J. B. Matthews, T. R. Zacharewski, Toxicol Lett. 2001, 121, 167-77.
155.	A. Gallegos, D. Robert, X. Girones, R. Carbo-Dorca, J Comput Aided Mol Des. 2001, 15, 67-80.
156.	J. H. Lin and A. Y. Lu, Pharmacol Rev. 1997, 49, 403-49.
157.	D. R. Thakker, W. Levin, A. W. Wood, A. H. Conney, H. Yagi, D. M. Jerina, in: I. W. Waisser, D. E. Drayer (Ed.): Drug Stereochemistry: Analytical Methods and Pharmacology, Marcel Dakker, New York, USA, 1988, pp. 271-296.
158.	V. Okoh, A. Deoraj, D. Roy, Biochim Biophys Acta. 2011, 1815, 115-33.
Povzetek
Motnje endokrinega sistema pomenijo tiste spremembe v delovanju endokrinega sistema, ki povzročijo škodljive učinki pri zdravih ljudeh ali njihovih potomcih. Fiziološki endokrini hormoni delujejo v zelo nizkih plazemskih koncentracijah, medtem ko se za določene kemijske spojine, poznane kot endokrini motilci (EDCs), prav tako predvideva sprememba endokrine funkcije pri podobno nizkih koncentracijah. V preglednem članku smo se osredotočili predvsem na pregled strukturnih razredov organskih sinteznih okoljskih endokrinih motilcev ter na njihove skupne strukturne elemente, ki jim omogočajo interakcijo z estrogenim signaliziranjem. Endokrini motilci lahko vplivajo na estrogeno signaliziranje neposredno preko interakcije z estrogenimi receptorji (ERs) ali posredno preko transkripcijskih faktorjev kot je receptor za aromatske ogljikovodike (AhR) ali preko modulacije kritičnih encimov, ki sodelujejo pri biosintezi in presnovi estrogenov. Nekateri strukturni elementi hormonskih motilcev predstavljajo tveganje za citotoksično in genotok-sično delovanje, predvsem preko bioaktivacije v reaktivne metabolite.