TY - JOUR
AU - Ferguson, Patrick Lee
AB - Abstract Brominated phenolic compounds (BPCs) are found in the environment, and in human and wildlife tissues, and some are considered to have endocrine disrupting activities. The goal of this study was to determine how structural differences of 3 BPC classes impact binding affinities for the thyroid receptor beta (TRβ) in humans and zebrafish. BPC classes included halogenated bisphenol A derivatives, halogenated oxidative transformation products of 2,2′,4,4′-tetrabromodiphenyl ether (BDE-47), and brominated phenols. Affinities were assessed using recombinant TRβ protein in competitive binding assays with 125I-triiodothyronine (125I-T3) as the radioligand. Zebrafish and human TRβ displayed similar binding affinities for T3 (Ki = 0.40 and 0.49 nM) and thyroxine (T4, Ki = 6.7 and 6.8 nM). TRβ affinity increased with increasing halogen mass and atomic radius for both species, with the iodinated compounds having the highest affinity within their compound classes. Increasing halogen mass and radius increases the molecular weight, volume, and hydrophobicity of a compound, which are all highly correlated with increasing affinity. TRβ affinity also increased with the degree of halogenation for both species. Human TRβ displayed higher binding affinities for the halogenate bisphenol A compounds, whereas zebrafish TRβ displayed higher affinities for 2,4,6-trichlorophenol and 2,4,6-trifluorophenol. Observed species differences may be related to amino acid differences within the ligand binding domains. Overall, structural variations impact TRβ affinities in a similar manner, supporting the use of zebrafish as a model for TRβ disruption. Further studies are necessary to investigate how the identified structural modifications impact downstream receptor activities and potential in vivo effects. endocrine disrupting chemicals, tetrabromobisphenol A, BDE-47, 2,4,6-tribromophenol, competitive binding assay, thyroid receptor Thyroid hormones (THs) are essential regulators of a range of developmental processes and physiological functions in vertebrates. Processes include fetal and postnatal growth and development, neurogenesis, metabolism, and key developmental transitions including metamorphosis and smoltification (Holzer and Laudet, 2015; Mullur et al., 2014; Pascual and Aranda, 2013; Préau et al., 2015). The biological effects of THs are mediated through interactions with thyroid hormone receptors (TRs, TRα, and TRβ). TRs are members of the nuclear receptor superfamily of ligand-activated transcription factors, and serve as a link between thyroid hormones and target gene expression (Aranda et al., 2013). The potential for environmental contaminates to mimic native hormones and disrupt nuclear receptor (NR) signaling pathways is well established (reviewed in Zoeller et al., 2012). Hormone binding is the molecular initiating event of many NR signaling cascades. The significance of this event is well recognized, and binding assays for steroid receptors are included in the EPA’s Endocrine Disruptor Screening Program, and the OECD’s conceptual framework for assessing endocrine disrupting chemicals (OECD, 2015; USEPA, 2017). These assays are used as a screening tool to identify contaminants that directly interact with receptors, quantify the strength of that interaction, and aid in prioritizing chemicals for further analysis. Receptors such as the estrogen receptor (ER) and androgen receptor (AR) have been shown to bind a wide range of structurally diverse compounds, indicating they are vulnerable targets of many endocrine disrupting chemicals (Blair et al., 2000; Fang et al., 2003; Laws et al., 2006). In contrast, TRα and TRβ are highly specific and selective for thyroid hormones. The size, geometry, and chemical properties of the ligand-binding pocket (LBP) closely match the hormones. Previous binding studies that tested a structurally diverse range of common endocrine disrupting chemicals revealed that compounds that deviate from the molecular attributes and physicochemical properties of thyroid hormones are excluded from binding (Ishihara et al., 2003a,b). Despite the high receptor selectivity, halogenated phenolic contaminants that share features and physicochemical properties with thyroid hormones may bind and disrupt TR signaling. However, the current data regarding the influence of chemical structure on TR affinity are unresolved and suggest the possibility of species differences, particularly between mammalian and nonmammalian vertebrates. Studies examining the affinities of halogenated contaminants to mammalian TRs (human and rat) suggest a strong preference for a hydroxylated diphenyl core structure similar to that of thyroid hormones. For example, OH-PBDEs bound TRs with affinities ranging from 10−7 to 10−4 M, but the parent PBDEs lacked affinity (Kitamura et al., 2005a, 2008; Ren et al., 2013). The binding affinities were additionally influenced by the position of the hydroxyl group and halogenation pattern. Specifically, at least 1 aromatic ring of the compounds with the highest affinity for TR were substituted similar to the outer ring of thyroxine (T4): a para hydroxyl group (4-OH) with adjacent halogens on each side (Kitamura et al., 2005a, 2008). Single-ring structures such as 2,4,6-tribromophenol (TBP), tetrabromohydroquinone (TBHQ), and the herbicide acetochlor also lacked TR affinity (Kitamura et al., 2008). Binding assays with halogenated bisphenol A (BPA) compounds and TRs isolated from a rat pituitary cell line found that halogen species influenced affinities, as tetrabromobisphenol A bound with higher affinity than tetrachlorobisphenol A (Kitamura et al., 2005b), possibly due to bromine being more chemically similar to iodine than chlorine. A study by Ren et al. (2013) demonstrated that the increasing the degree of bromination on the aromatic rings of OH-PBDEs increased their affinity for human thyroid receptor beta (TRβ), likely due to increased ligand-receptor hydrophobic interactions. Compared with mammalian TRs, binding studies using nonmammalian TRs are limited. To date, nonmammalian studies have included TRβ orthologs from 2 amphibians and 2 birds: the African clawed frog (Xenopus laevis), bullfrog (Rana catesbeiana), chicken (Gallus gallus), and Japanese quail (Coturnix japonica) (Ishihara et al., 2003a,b; Kitamura et al., 2005b; Kudo and Yamauchi, 2005; Kudo et al., 2006; Yamauchi et al., 2003). Similar to human and rat, TRs from nonmammalian vertebrates display a preference for a hydroxylated diphenyl core, whereas compounds that diverge from this core structure lack affinity (Ishihara et al., 2003a,b). This includes compounds such as PAHs, alkylphenols, phthalates, and biocides such as organophosphates and chloroacetanilides. Studies with brominated and chlorinated BPAs and Xenopus TRβ imply nonmammalian TRs prefer bromine over chlorine (Kudo and Yamauchi, 2005; Kudo et al., 2006). However, unlike the study by Ren et al. (2013) with rat TRs, increasing halogenation did not increase binding affinity for Xenopus TRβ. 3,3′-dibromobisphenol A had the highest affinity for Xenopus TRβ of the brominated BPA compounds, followed by 3,3′,5-tribromobisphenol A and 3,3′,5,5′-tetrabromobisphenol A (TBBPA) (Kudo et al., 2006). Of the chlorinated bisphenol A compounds, 3,3′,5-trichlorobisphenol A had the highest affinity for Xenopus, bullfrog and chicken TRs, however, the pattern between species varied for additional chlorinated BPAs (Kudo and Yamauchi, 2005; Yamauchi et al., 2003). These results suggest there may be differences in ligand preferences between mammalian and nonmammalian TRs. As exposure to many halogenated contaminants is widespread and a concern for both human and ecosystem health, species differences in ligand preferences may have important implications for hazard assessments. The goal of this study was to examine how structural modifications of 3 brominated phenolic compound classes (BPCs) impact the potential of these compounds to bind TRβ using competitive binding assays. We focused primarily on structural variations of 3,3′,5,5′-tetrabromobisphenol A (TBBPA), 2,4,6-tribromophenol (TBP), and the oxidative transformation products of BDE-47 (OH-BDE-47). TBBPA is the highest production volume brominated flame retardant (BFR) currently in use, with almost 120 million pounds produced in 2011 (USEPA, 2012). TBBPA is a tetrabrominated analog of bisphenol A that is used as a flame retardant in electronic circuit boards and plastics, but also has applications in art supplies, textiles, and children’s toys (Covaci et al., 2009; USEPA, 2015). Polybrominated diphenyl ether (PBDE) mixtures were used as flame retardants for decades in products containing polyurethane foam, such as furniture, mattresses, and carpeting. BDE-47 is the most abundant PBDE congener found in human tissues (Leonetti et al., 2016; Stapleton et al., 2011, 2012), and serum concentrations of the oxidative transformation products of BDE-47 (OH-BDE-47) are found at higher concentrations than other OH-BDEs in humans, marine mammals, and fish (De la Torre et al., 2013; Houde et al., 2009; Nomiyama et al., 2011a,b; Qiu et al., 2009; Valters et al., 2005). PBDE use has been phased out or banned in many countries, however, exposure is expected to continue due to long-term product use and recycling. 2,4,6-tribromophenol has multiple applications, including as a wood preservative, intermediate in synthesis of other BFRs, and BFR metabolite or degradation by-product (Qiu et al., 2009; Suzuki et al., 2008). The structural analogs included in this study were chosen primarily based on specific structural differences between the analogs and the 3 BPC compounds, and their environmental relevance. Differences include halogen species, degree of halogenation, presence/absence of halogens and/or hydroxyl groups, chemical backbones, and number of aromatic rings. For this study, a comparative approach was applied by including TRβ orthologs from human and zebrafish to further our understanding of the influence of chemical structure on TRβ affinities in mammalian and nonmammalian species. This initial study focused on TRβ for multiple reasons. Like most vertebrates, humans have a single gene for both TRβ and TRα. Zebrafish also have a single TRβ gene, however, zebrafish and other teleost fish species maintain 2 TRα genes as a result of a whole genome duplication event that took place early in the evolution of teleost fish (Bertrand et al., 2007; Meyer and Van de Peer, 2005). The evolutionary fate of duplicate genes can vary: 1 copy can evolve a novel function, the functions of the original gene can be divided between the 2 paralogs, or 1 paralog may become nonfunctional while the other retains the original function (reviewed in Hahn, 2009). As the potential functional differences of the TRα paralogs in fish are not well characterized, we chose to focus on TRβ to better enable a comparison between species. Furthermore, previous work from our lab has suggested a potential role for TRβ in thyroid endocrine system disruption in zebrafish (Dong et al., 2014; Macaulay et al., 2015). Finally, we chose TRβ to facilitate comparisons with previous studies in nonmammalian vertebrates that focused on the TRβ ortholog from their respective species (Ishihara et al., 2003a,b; Kudo and Yamauchi, 2005; Kudo et al., 2006; Yamauchi et al., 2003). Zebrafish are a popular model species for vertebrate development due to a high degree of conservation of many endocrine signaling pathways between zebrafish and mammals, despite the fact that the 2 lineages diverged roughly 450 Ma (Kumar and Hedges, 1998). The hypothalamus-pituitary-thyroid (HPT) axis is highly conserved, and the major components of the thyroid system have been identified in zebrafish, including thyroid hormone receptors (Blanton and Specker, 2007; Heijlen et al., 2013; Liu et al., 2000; Porazzi et al., 2009; Walpita et al., 2007). For these reasons zebrafish and other small fish have become increasingly popular models of thyroid related disease and disruption (Macaulay et al., 2015; Noyes et al., 2013; Raldúa et al., 2012; Thienpont et al., 2011). However, studies examining TR-mediated effects of these compounds in zebrafish are limited. A clearer understanding of how chemical structure impacts TRβ function will better enable scientists to predict receptor-mediated effects of potential endocrine disruptors and aid in understanding potential human and ecosystem health effects. MATERIALS AND METHODS Chemicals Bisphenol S (BPS, 98%), tetrabromobisphenol S (TBBPS, 98.8%), dimethylbisphenol A (DMBPA, 98.9%), 3-hydroxy-2,2′,4,4′-tetrabromodiphenyl ether (3-OH-BDE-47, 97%), 6-hydroxy-2,2′,4,4′-tetrabromodiphenyl ether (6-OH-BDE-47, 98.6%), 2,2′,4,4′-tetrabromodiphenyl ether (BDE 47, 100%), were purchased from AccuStandard (New Haven, Connecticut). 2,4,6-tribromophenol (TBP, 99%), 2,4,6-trichlorophenol (TCP, 98%), 2,4,6-trifluorophenol (TFP, 99%), bisphenol A (BPA, 99%), bisphenol AF (BPAF, 97%), tetrabromobisphenol A (TBBPA, 97%), thyroxine (T4, 98%), triclosan (97%), and dimethyl sulfoxide (DMSO, > 99%) were purchased from Sigma-Aldrich (St. Louis, Missouri). 2,4,6-triiodophenol (TIP, 97%) was purchased from Alfa Aesar (Ward Hill, Massachusetts). Nonradioactive triiodothyronine (T3, 98.7%) was purchased from Calbiochem (Billerica, MA). 3-monobromobisphenol A (3-BrBPA), 3,3′-dibromobisphenol A (3,3′-BrBPA), and 3,3′,5-tribromobisphenol A (3,3′,5-BrBPA) were a gift from Dr Goran Marsh (Department of Materials and Environmental Chemistry, Stockholm University, Sweden). Tetraiodobisphenol A (TIBPA, 98%) was purchased from Spectra Group Limited, Inc (Millbury, Ohio). Tetrachlorobisphenol A (TCBPA, 98%), tetramethylbisphenol A (TMBPA, > 98%), and 4,4′-dihydroxydiphenyl ether (4,4′-diOH-DE, > 98%) were purchased from TCI America (Portland, Oregon). Pentabromophenol (PBP, 96%) was purchased from Matrix Scientific (Columbia, South Carolina). L-3,5,3′-[125I]-Triiodothyronine ([125I]-T3, original specific activity 779 Ci/mmol) was purchased from Perkin Elmer (Boston, Massachusetts). All hormone and competitor stocks were made in DMSO. See Supplementary Figure 1 for chemical structures and Supplementary Table 1 for CAS numbers and select chemical properties of all the test compounds. Protein Expression Constructs Zebrafish TRβ The ligand binding domain (LBD) of zebrafish (Danio rerio) TRβ (GenBank ID AF109732.1) corresponding to amino acids No. 99-386 was cloned from an expression construct containing full-length zebrafish TRβ that was a gift from Dr Chan Woon Khiong (National University of Singapore) (Liu et al., 2000). The sequence was PCR amplified using Titanium Taq DNA polymerase (Clontech, Mountain View, California) following the manufacturer’s protocol. The primer sequences with incorporated restriction sites are listed in Supplementary Table 2. The PCR product was gel-purified and ligated into the pGEM-T Easy Vector (Promega, Madison, Wisconsin) following the manufacturer’s protocol. The ligated construct was transformed into Subcloning Efficiency DH5α Competent Escherichia coli (Invitrogen, Carlsbad, California), and positive clones were identified via a blue/white screen. DNA constructs were isolated and sequenced to confirm gene identity. The TRβ sequence was subcloned into the pET32a(+) protein expression vector (Merck Millipore, Darmstadt, Germany) via the NspV (5′) and XhoI (3′) restriction sites. The construct was transformed into Rosetta2(DE3)pLysS chemically competent Escherichia coli (Merck Millipore, Darmstadt, Germany) for protein expression. Sequence integrity and orientation was verified by sequencing in both directions. The protein size was 49.4 kDa including the pET32a vector tags. Human TRβ The ligand binding domain of the human TRβ (GenBank ID HQ692825.1) corresponding to amino acids No. 202-462, was codon-optimized for expression in Escherichia coli, and assembled from synthetic oligonucleotides and/or PCR products. The synthetically prepared gene was created and inserted into the pET32b(+) protein expression vector via the NcoI (5′) and BamHI (3′) restriction sites by GeneArt (ThermoFisher Scientific, Waltham, Massachusetts). DNA sequencing verified construct integrity, and 5 µg of the plasmid preparation were lyophilized prior to shipping from GeneArt. Upon receipt, the DNA was reconstituted in sterile water and transformed into TOP10 and BL21(DE3)pLysS chemically competent Escherichia coli bacteria (Merck Millipore, Darmstadt, Germany). The protein size was 46.6 kDa including vector tags. Sequence Analysis Vertebrate TRβ sequences were obtained through the Basic Local Alignment Search Tool (BLAST) of the National Center for Biotechnology Information (URL: www.ncbi.nih.gov/Blast.cgi; last accessed August 28, 2017). The ligand binding domains within the TRβ amino acid sequences were identified and aligned using ClustalW (Thompson et al., 1994) via the Biology Workbench web-based tool (Unwin et al., 2002) (version 3.2, URL: http://workbench.sdsc.edu; last accessed August 28, 2017). Key features of the TRβ LBDs were identified based on previous work (Wagner et al., 1995) and human TRβ structures deposited in the RCSB Protein Data Bank (URL: www.rcsb.org; last accessed August 28, 2017, Berman et al., 2000; PDB IDs: 3GWS, Nascimento et al., 2006; and 1XZX, Sandler et al., 2004). Protein Expression For zebrafish TRβ protein expression, a single bacteria colony from a freshly streaked plated was used to inoculate 10 ml of lysogeny broth (LB/carb: 1% wt/vol tryptone, 0.5% wt/vol yeast extract, 1% wt/vol NaCl, 100 µg/ml carbenicillin, 34 µg/ml chloramphenicol, pH 7.4), and grown overnight at 37˚C with vigorous shaking (200 rpm). The following morning, 500 ml of fresh LB/carb (500 ml) was inoculated with the starter culture, and incubated at 37˚C with shaking (200 rpm) until the OD600 ≈0.6. The temperature was decreased to 18˚C, and protein expression was induced with 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG, Apex Biosciences, Chapel Hill, North Carolina). The culture was incubated for an additional 3 h at 200 rpm. For human TRβ protein expression, a 20-ml starter culture was used to inoculate 1 l of autoinduction media (1% wt/vol tryptone, 0.5% wt/vol yeast extract, 2 mM MgSO4, 6 mM lactose, 3 mM glucose, 0.5% vol/vol glycerol, 25 mM (NH4)2SO4, 50 mM KH2PO4, 50 mM Na2HPO4, 25 µg/ml carbenicillin). The autoinduction culture was incubated at 28˚C overnight with shaking at 250 rpm. Following expression, both cultures were pelleted by centrifuging at 4000 rpm at 4˚C for 20 min. The supernatants were discarded, and the pellets were stored at −20˚C until protein purification. Protein Purification Zebrafish TRβ Frozen bacteria pellets were thawed on ice, and resuspended in 50 ml of lysis buffer (40 mM Na2HPO4, 10 mM NaH2PO4, 100 mM NaCl, 10% glycerol, 0.2% Tween 20, 15 mM β-mercaptoethanol, 1 Pierce protease inhibitor tablet without EDTA [ThermoFisher Scientific, Waltham, MA], 1 mM phenylmethylsulfonyl fluoride [PMSF], 0.2 mg/ml lysozyme, 20 U/ml DNaseI, 1 mM MgCl2, pH 7.4). The suspension was gently rotated at 4˚C for 2 h, then sonicated on ice for 10 5-s bursts at 30% amplitude (Branson Digital Sonifier 450 with 1/8″ microtip: Branson Ultrasonics, Danbury, CT). The cell debris was pelleted by centrifuging the lysate at 12 000×g at 4˚C for 10 min. The pellet was discarded and the supernatant containing the soluble zebrafish TRβ protein was purified using immobilized metal affinity chromatography (IMAC) on a BioLogic HR FPLC system kept at 4˚C (Bio-Rad Laboratories, Hercules, California). The concentrations of NaCl and imidazole were increased to 500 mM and 25 mM prior to purification. The lysate was loaded onto a Ni-NTA column (HisTrap FF 1.0 ml, GE Healthcare Life Sciences, Pittsburgh, Pennsylvania) pre-equilibrated with binding buffer (40 mM Na2HPO4, 10 mM NaH2PO4, 500 mM NaCl, 10% glycerol, 0.2% Tween-20, 15 mM β-mercaptoethanol, 1 mM PMSF, 25 mM imidazole, pH 7.4). The column was washed with binding buffer, and the polyhistadine-tagged TRβ protein was eluted by gradually increasing the concentration of imidazole from 25 mM to 400 mM. Protein elution was monitored by UV-Vis chromatography (280 nm). The fractions containing zebrafish TRβ were combined and concentrated using a Microsep Advance Centrifugal Device (Pall Corporation, Ann Arbor, Michigan) following the manufacturers protocol. Concentrated protein was divided into single-use aliquots (20 µl) in PCR tubes and flash-frozen in liquid nitrogen (Deng et al., 2004), and stored at −80˚C. Protein purity and integrity was assessed via SDS-PAGE (Supplementary Figure 2), and protein concentration was quantified via a Bradford assay (Pierce Protein Biology, ThermoFisher Scientific, Waltham, MA). Human TRβ Recombinant human TRβ protein was purified using IMAC followed by anion exchange chromatography on an ÄKTA avant FPLC system (GE Healthcare Bio-Sciences, Uppsala, Sweden) in a cold room kept at 4˚C. Frozen bacteria pellets were thawed on ice and resuspended in 100 ml lysis buffer (50 mM NaH2PO4, 500 mM NaCl, 10 mM imidazole, cOmplete EDTA-free protease inhibitor cocktail tablets [Roche Diagnostics, Mannheim, Germany], 10 U/ml DNaseI, pH 7.4). The suspension was rotated gently at 4˚C for 15 min before passing through a M-110 P microfluidizer (Microfluidics Corp, Westwood, Massachusetts) at 15 000 psi to lyse the cells. The lysate was collected and centrifuged at 15 000 rpm for 20 min at 4˚C. The insoluble fraction was discarded and the supernatant containing the soluble human TRβ protein was loaded onto a 5-ml HisTrap FF column pre-equilibrated with binding buffer (50 mM NaH2PO4, 500 mM NaCl, 10 mM imidazole, pH 7.4). The column was washed with binding buffer, and the polyhistadine-tagged human TRβ was eluted by gradually increasing the concentration of imidazole from 10 mM to 500 mM. Protein elution was monitored by UV-Vis chromatography (280 nm). The eluted fractions containing human TRβ were combined and diluted 15-fold into anion exchange binding buffer (20 mM Tris, pH 7.4) containing cOmplete EDTA-free protease inhibitor cocktail tablets. The diluted sample was loaded onto a HiTrap Q HP 5 ml anion exchange column (GE Healthcare Bio-Sciences, Uppsala, Sweden) pre-equilibrated with binding buffer. The column was washed with binding buffer, and human TRβ protein was eluted by gradually increasing the concentration of NaCl from 0 to 1 M. The protein elution was monitored by UV-Vis chromatography (280 nm). The eluted fractions containing the human TRβ protein were combined and dialyzed into 50 mM Tris, 50 mM NaCl, pH 7.4 for 2 h at 4˚C using a 10 MWCO Slide-a-Lyzer cassette (ThermoFisher Scientific, Waltham, MA). Glycerol was added to the dialyzed product at a final concentration of 10% vol/vol prior to flash-freezing single-use aliquots in liquid nitrogen for storage at −80˚C. The protein purity and integrity was assessed by SDS-PAGE (Supplementary Figure 2) and the protein concentration was quantified by the bicinchoninic acid (BCA) assay (Pierce Protein Biology, ThermoFisher Scientific, Waltham, MA). Saturation Binding Assays Optimal receptor concentrations for the binding assays were determined with protein titration curves (Supplementary Figure 3). For the saturating binding assays, each 100 µl reaction contained 10 nM purified zebrafish TRβ or 40 nM human TRβ in binding buffer (40 mM Na2HPO4, 10 mM NaH2PO4, 150 mM NaCl, 10% glycerol, 15 mM β-mercaptoethanol, pH 7.4) and 0–10 nM of [125I]-T3 for a total of 10 concentrations per curve. The [125I]-T3 dilutions were made in binding buffer. Nonspecific binding (NSB) reactions additionally included 1 µM of unlabeled T3, and served as the positive control. Reactions were incubated at 4˚C overnight (approximately 18 h). Bound ligand was separated from free using the charcoal adsorption method (Strange, 1992): 100 µl of a 1.6% dextran: activated charcoal (1:10 wt/vol) solution in separation buffer (40 mM Na2HPO4, 10 mM NaH2PO4, 150 mM NaCl, 0.1% gelatin, 0.02% NaN3, pH 7.4) was added to each tube and incubated on ice for 15 min with gentle shaking every 5 min. The reaction tubes were centrifuged at 4000 rpm for 15 min at 4˚C to pellet the charcoal suspension. A 100-µl aliquot of the supernatant containing the receptor-bound ligands was removed for scintillation counting. Data were globally analyzed in GraphPad Prism 7 (GraphPad Software Inc, La Jolla, California) using a one-site binding model for total and nonspecific binding. Each curve was repeated 3 times with duplicate tubes per concentration. The reported affinity (Kd) is the average of the 3 curves ± SEM. Between species comparisons were made using Welch’s t-test in GraphPad Prism 7 (GraphPad Software Inc, La Jolla, CA). Competitive Binding Assays Competitive binding assays included at least 10 competitor concentrations for individual chemicals. As described for the saturation curves, each 100 µl reaction included 10 nM recombinant zebrafish TRβ or 40 nM recombinant human TRβ in binding buffer. Additionally, each tube contained 3 nM of [125I]-T3 and a range of competitor concentrations (0–1 mM, depending on competitor solubility). Competitor stocks were made in 100% DMSO, and assay dilutions were made in binding buffer. Nonspecific binding reactions included 1 µM of unlabeled T3. Reactions were incubated overnight, and receptor-bound ligand was separated from free and counted as described earlier. Specific binding was determined by subtracting the average of the nonspecific binding counts from the total binding counts. Data were analyzed in GraphPad Prism 7 (GraphPad Software Inc, La Jolla, CA). Sigmoidal curves were initially fit to the raw CPM values to identify the top and bottom plateaus. Data were then normalized to the top plateau (100% bound) and bottom plateau (0% bound) in GraphPad. In the absence of a bottom plateau, data were normalized to the NSB control. The affinity for the competitor (Ki) was determined using a one-site binding model in GraphPad Prism 7 (GraphPad Software Inc, La Jolla, CA). Each competition curve was repeated 3 times with 3 tubes per dose. The reported affinity for each chemical is the average of the 3 curves. To more accurately represent the true variance in the combined data sets, the SD was calculated from the grand variance of all 3 replicates for each dose. The grand variance was calculated from the total sum of squares of the individual values from the 3 reps divided by the degrees of freedom for each dose. The calculated SD is the square root of this number. Between species comparisons were made using Welch’s t-test in GraphPad Prism 7. Correlations Physicochemical properties were taken from SciFinder (scifinder.cas.org; last accessed March 21, 2017). Spearman rank correlation coefficients were calculated in GraphPad Prism 7 (GraphPad Software Inc, La Jolla, CA) using log-transformed Ki values. Statistical significance was set at α=.5. RESULTS Thyroid Receptors The multiple sequence alignment revealed that the shared sequence identities of the TRβ LBDs ranged between 86% and 99%, indicating a high degree of conservation across species and vertebrate evolution (Figure 1 and Supplementary Table 3). The zebrafish TRβ LBD was 90% identical to human TRβ LBD. Zebrafish TRβ was most like human TRβ compared with the other 4 fish, whereas medaka was the least similar to human (86%). Sequence identities shared between zebrafish TRβ and the other fish species were slightly higher compared with human TRβ, and ranged from 91% to 95%. The shared identity between zebrafish and rat (90%), chicken (92%), and frog (93%) were within the ranges observed with human and the other fish species. Figure 1. View largeDownload slide Sequence alignment of TRβ ligand binding domains. The alignment includes amino acid sequences from sea bream (Sparus aurata, GenBank ID AAO86517.1), flounder (Paralichthys olivaceus, GenBank ID BAA08201.1), Japanese medaka (Oryzias latipes, GenBank ID BAD11773.1), salmon (Salmo salar, GenBank ID NP_001117172.1), zebrafish (Danio rerio, GenBank ID AAF14239.1), human (Homo sapien, GenBank ID ADZ17336.1), rat (Rattus norvegicus, GenBank ID AAA40916), chicken (Gallus gallus, GenBank ID CAA35544.1), and frog (Xenopus tropicalis, GenBank ID BAE93231.1). Sequences were aligned via the CLUSTALW program in the SDSC Biology Workbench (workbench.sdsc.edu). The location of the 12 α-helices (H1–H12) and the 4 β-sheets (S1–S4) are indicated above the sequences. Amino acid residues that directly interact with T3 are in bold type. The gray box immediately following S1 depicts the location of the teleost-specific amino acid insert. Degree of amino acid conservation between species is indicated by the symbols below the sequence. The symbols are defined in the above consensus key. Figure 1. View largeDownload slide Sequence alignment of TRβ ligand binding domains. The alignment includes amino acid sequences from sea bream (Sparus aurata, GenBank ID AAO86517.1), flounder (Paralichthys olivaceus, GenBank ID BAA08201.1), Japanese medaka (Oryzias latipes, GenBank ID BAD11773.1), salmon (Salmo salar, GenBank ID NP_001117172.1), zebrafish (Danio rerio, GenBank ID AAF14239.1), human (Homo sapien, GenBank ID ADZ17336.1), rat (Rattus norvegicus, GenBank ID AAA40916), chicken (Gallus gallus, GenBank ID CAA35544.1), and frog (Xenopus tropicalis, GenBank ID BAE93231.1). Sequences were aligned via the CLUSTALW program in the SDSC Biology Workbench (workbench.sdsc.edu). The location of the 12 α-helices (H1–H12) and the 4 β-sheets (S1–S4) are indicated above the sequences. Amino acid residues that directly interact with T3 are in bold type. The gray box immediately following S1 depicts the location of the teleost-specific amino acid insert. Degree of amino acid conservation between species is indicated by the symbols below the sequence. The symbols are defined in the above consensus key. The 9 TRβ LBD sequences were compared with previously annotated human TRβ sequences, and essential features were identified (Nascimento et al., 2006; Sandler et al., 2004; Wagner et al., 1995). Amino acid residues involved in TRβ-T3 interactions were conserved across species (Figure 1). The 12 α-helices (H1–H12) and 4 β-strands (S1–S4) were identified, and are highly conserved. Of the 246 amino acids of the LBDs, only 23 varied between human and zebrafish. Most of the amino acid variation is located between H1 to the beginning of H3, and H9 to before the start of H11. As noted previously, zebrafish TRβ lack the 9 amino acid insert that is present in other fish species, but absent in humans and other vertebrates (Liu et al., 2000; Nowell et al., 2001). Assay Optimization and Performance: Thyroid Hormones Protein titration experiments indicated that the optimal receptor concentrations for the binding assays were 10 nM for zebrafish TRβ and 40 nM for human TRβ (Supplementary Figure 3). Receptor concentrations were assigned based on the following criteria: 1) nonspecific binding was ≤ 20% of total binding, 2) 10% or less of the total amount of [125I]-T3 added was bound at receptor saturation to avoid issues with ligand depletion, and 3) a Z’-factor ≥ 0.6, which indicates the receptor concentration yields a good dynamic range between NSB and total binding. The Z’-factors were 0.8 for human TRβ and 0.9 for zebrafish TRβ. Saturation binding curves were first conducted to determine the affinity of the receptors for the native ligand (T3) and to identify the saturating concentration range of [125I]-T3 for use in the competitive binding assays (Figure 2A). An unpaired t-test with Welch’s correction indicated that the affinities of [125I]-T3 for zebrafish TRβ (Kd = 0.60 ± 0.11 nM) and human TRβ (Kd = 0.71 ± 0.13 nM) were not significantly different between species (p = .58). The affinities were comparable with previously reported affinities for multiple nonmammalian vertebrates, including fish (sea bream: 0.625 nM, and tilapia: 0.2 nM), quail (0.31 nM), chicken (0.63 nM), and bullfrog (0.76 nM), as well as previously reported affinities for mammalian TRβ including rat (0.49 nM) and human (0.14 nM) (Ishihara et al., 2003a,b; Mendoza et al., 2013; Nowell et al., 2001; Togashi et al., 2005; Williams, 2000). Figure 2. View largeDownload slide Thyroid hormone binding curves. A, Saturation binding curve for human TRβ (black circles) and zebrafish TRβ (white circles). Purified receptors were incubated with 0–10 nM [125I]-T3 overnight at 4˚C. Unbound ligand was removed with the charcoal adsorption method. Data were globally analyzed using a one-site binding model. Each curve was repeated 3 times with duplicate tubes per concentration. The reported affinity (Kd) is the average of the 3 curves ± SEM. B, Competitive binding curves for T3 and T4. Purified zebrafish TRβ (white shapes) and human TRβ (black shapes) were incubated with 3 nM [125I]-T3 and 0–10 µM unlabeled T3 (circles) or T4 (triangles). Bound and free were separated by the charcoal adsorption method. Data were normalized to the top and bottom plateaus, and the affinity of each competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki is the average of 3 curves ± SEM. Figure 2. View largeDownload slide Thyroid hormone binding curves. A, Saturation binding curve for human TRβ (black circles) and zebrafish TRβ (white circles). Purified receptors were incubated with 0–10 nM [125I]-T3 overnight at 4˚C. Unbound ligand was removed with the charcoal adsorption method. Data were globally analyzed using a one-site binding model. Each curve was repeated 3 times with duplicate tubes per concentration. The reported affinity (Kd) is the average of the 3 curves ± SEM. B, Competitive binding curves for T3 and T4. Purified zebrafish TRβ (white shapes) and human TRβ (black shapes) were incubated with 3 nM [125I]-T3 and 0–10 µM unlabeled T3 (circles) or T4 (triangles). Bound and free were separated by the charcoal adsorption method. Data were normalized to the top and bottom plateaus, and the affinity of each competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki is the average of 3 curves ± SEM. The competitive binding assays were optimized and validated using unlabeled thyroid hormones as competitors (Figure 2B). The affinities (Ki) and 95% confidence intervals of the unlabeled hormones for zebrafish TRβ were 0.40 nM (0.33–0.48 nM) for T3, and 6.66 nM (5.25–8.44 nM) for T4. The values for human TRβ were 0.49 nM (0.41–0.57 nM) for T3 and 6.77 nM (5.30–8.64 nM) for T4. The Ki values between species were not significantly different for T3 (p = .10) or T4 (p = .93). Tetrabromobisphenol A and Related Compounds All tested bisphenols competed with T3 and bound TRβ in a concentration-response manner, however, the affinities of each compound for TRβ ranged between 10−8 and 10−4 M. The competitive binding curves are depicted in Figures 3A–C, and Table 1 lists the binding affinity (Ki), 95% confidence interval, and statistical results for affinity comparisons between species. Table 1. Ki Values, 95% Confidence Intervals (95% CI) and t-Test Results for the Tested Compounds     Zebrafish TRβ (µM)   Human TRβ (µM)   Welch’s t-Test   Category  Compound  Ki  95% CI  Ki  95% CI  t Value  p Value  TBBPA and related bisphenols  TBBPA  0.53  0.37–0.78  0.22  0.19–0.26  4.444  .026*  TBBPS  5.70  3.36–9.66  7.33  5.91–9.10  0.897  .444  TCBPA  1.96  1.11–3.46  0.47  0.40–0.55  4.985  .028*  TIBPA  0.08  0.05–0.13  0.09  0.07–0.10  0.162  .883  TMBPA  6.50  3.57–11.82  1.33  1.07–1.63  5.111  .024*  DMBPA  13.65  8.14–22.88  4.58  3.78–5.54  4.023  .038*  3,3′,5-BrBPA  1.42  0.94–2.15  0.28  0.24–0.33  7.556  .008**  3,3′-BrBPA  4.42  2.93–6.66  0.82  0.70–0.98  7.716  .007**  3-BrBPA  16.86  10.7–26.58  3.34  2.68–4.16  6.548  .008**  BPA  NB  NB  33.29  27.14–40.84  —  —  BPAF  8.79  5.84–13.24  1.21  1.04–1.41  9.277  .005**  BPS  51.85  31.41–58.6  71.72  56.70–90.71  1.192  .323  Halogenated diphenyl ethers  BDE-47  NB  NB  NB  NB  —  —  3-OH-BDE-47  0.21  0.14–0.31  0.27  0.23–0.33  1.217  .316  6-OH-BDE-47  0.30  0.20–0.46  0.54  0.44–0.67  2.488  .090  Triclosan  4.66  2.97–7.29  4.06  2.92–5.64  0.498  .647  4,4′-OH-DE  IC  IC  IC  IC  —  —  Halogenated phenols  2,4,6-TFP  58.26  29.26–116  NB  NB  —  —  2,4,6-TCP  13.68  8.35–22.42  56  42.58–73.65  5.082  .013*  2,4,6-TBP  31.28  21.87–44.74  30.59  26.63–39.6  0.096  .929  2,4,6-TIP  2.68  1.91–3.76  1.58  1.33–1.89  2.501  .090  PBP  1.19  0.83–1.71  2.04  1.59–2.60  2.499  .075      Zebrafish TRβ (µM)   Human TRβ (µM)   Welch’s t-Test   Category  Compound  Ki  95% CI  Ki  95% CI  t Value  p Value  TBBPA and related bisphenols  TBBPA  0.53  0.37–0.78  0.22  0.19–0.26  4.444  .026*  TBBPS  5.70  3.36–9.66  7.33  5.91–9.10  0.897  .444  TCBPA  1.96  1.11–3.46  0.47  0.40–0.55  4.985  .028*  TIBPA  0.08  0.05–0.13  0.09  0.07–0.10  0.162  .883  TMBPA  6.50  3.57–11.82  1.33  1.07–1.63  5.111  .024*  DMBPA  13.65  8.14–22.88  4.58  3.78–5.54  4.023  .038*  3,3′,5-BrBPA  1.42  0.94–2.15  0.28  0.24–0.33  7.556  .008**  3,3′-BrBPA  4.42  2.93–6.66  0.82  0.70–0.98  7.716  .007**  3-BrBPA  16.86  10.7–26.58  3.34  2.68–4.16  6.548  .008**  BPA  NB  NB  33.29  27.14–40.84  —  —  BPAF  8.79  5.84–13.24  1.21  1.04–1.41  9.277  .005**  BPS  51.85  31.41–58.6  71.72  56.70–90.71  1.192  .323  Halogenated diphenyl ethers  BDE-47  NB  NB  NB  NB  —  —  3-OH-BDE-47  0.21  0.14–0.31  0.27  0.23–0.33  1.217  .316  6-OH-BDE-47  0.30  0.20–0.46  0.54  0.44–0.67  2.488  .090  Triclosan  4.66  2.97–7.29  4.06  2.92–5.64  0.498  .647  4,4′-OH-DE  IC  IC  IC  IC  —  —  Halogenated phenols  2,4,6-TFP  58.26  29.26–116  NB  NB  —  —  2,4,6-TCP  13.68  8.35–22.42  56  42.58–73.65  5.082  .013*  2,4,6-TBP  31.28  21.87–44.74  30.59  26.63–39.6  0.096  .929  2,4,6-TIP  2.68  1.91–3.76  1.58  1.33–1.89  2.501  .090  PBP  1.19  0.83–1.71  2.04  1.59–2.60  2.499  .075  A Welch’s t-test was conducted to identify significant differences in affinity between species (α = .05). Abbreviations: IC, inconclusive due to incomplete curve; NB, no observed binding. Significance is indicated by the following symbols: *p < .05, **p < .01. Table 1. Ki Values, 95% Confidence Intervals (95% CI) and t-Test Results for the Tested Compounds     Zebrafish TRβ (µM)   Human TRβ (µM)   Welch’s t-Test   Category  Compound  Ki  95% CI  Ki  95% CI  t Value  p Value  TBBPA and related bisphenols  TBBPA  0.53  0.37–0.78  0.22  0.19–0.26  4.444  .026*  TBBPS  5.70  3.36–9.66  7.33  5.91–9.10  0.897  .444  TCBPA  1.96  1.11–3.46  0.47  0.40–0.55  4.985  .028*  TIBPA  0.08  0.05–0.13  0.09  0.07–0.10  0.162  .883  TMBPA  6.50  3.57–11.82  1.33  1.07–1.63  5.111  .024*  DMBPA  13.65  8.14–22.88  4.58  3.78–5.54  4.023  .038*  3,3′,5-BrBPA  1.42  0.94–2.15  0.28  0.24–0.33  7.556  .008**  3,3′-BrBPA  4.42  2.93–6.66  0.82  0.70–0.98  7.716  .007**  3-BrBPA  16.86  10.7–26.58  3.34  2.68–4.16  6.548  .008**  BPA  NB  NB  33.29  27.14–40.84  —  —  BPAF  8.79  5.84–13.24  1.21  1.04–1.41  9.277  .005**  BPS  51.85  31.41–58.6  71.72  56.70–90.71  1.192  .323  Halogenated diphenyl ethers  BDE-47  NB  NB  NB  NB  —  —  3-OH-BDE-47  0.21  0.14–0.31  0.27  0.23–0.33  1.217  .316  6-OH-BDE-47  0.30  0.20–0.46  0.54  0.44–0.67  2.488  .090  Triclosan  4.66  2.97–7.29  4.06  2.92–5.64  0.498  .647  4,4′-OH-DE  IC  IC  IC  IC  —  —  Halogenated phenols  2,4,6-TFP  58.26  29.26–116  NB  NB  —  —  2,4,6-TCP  13.68  8.35–22.42  56  42.58–73.65  5.082  .013*  2,4,6-TBP  31.28  21.87–44.74  30.59  26.63–39.6  0.096  .929  2,4,6-TIP  2.68  1.91–3.76  1.58  1.33–1.89  2.501  .090  PBP  1.19  0.83–1.71  2.04  1.59–2.60  2.499  .075      Zebrafish TRβ (µM)   Human TRβ (µM)   Welch’s t-Test   Category  Compound  Ki  95% CI  Ki  95% CI  t Value  p Value  TBBPA and related bisphenols  TBBPA  0.53  0.37–0.78  0.22  0.19–0.26  4.444  .026*  TBBPS  5.70  3.36–9.66  7.33  5.91–9.10  0.897  .444  TCBPA  1.96  1.11–3.46  0.47  0.40–0.55  4.985  .028*  TIBPA  0.08  0.05–0.13  0.09  0.07–0.10  0.162  .883  TMBPA  6.50  3.57–11.82  1.33  1.07–1.63  5.111  .024*  DMBPA  13.65  8.14–22.88  4.58  3.78–5.54  4.023  .038*  3,3′,5-BrBPA  1.42  0.94–2.15  0.28  0.24–0.33  7.556  .008**  3,3′-BrBPA  4.42  2.93–6.66  0.82  0.70–0.98  7.716  .007**  3-BrBPA  16.86  10.7–26.58  3.34  2.68–4.16  6.548  .008**  BPA  NB  NB  33.29  27.14–40.84  —  —  BPAF  8.79  5.84–13.24  1.21  1.04–1.41  9.277  .005**  BPS  51.85  31.41–58.6  71.72  56.70–90.71  1.192  .323  Halogenated diphenyl ethers  BDE-47  NB  NB  NB  NB  —  —  3-OH-BDE-47  0.21  0.14–0.31  0.27  0.23–0.33  1.217  .316  6-OH-BDE-47  0.30  0.20–0.46  0.54  0.44–0.67  2.488  .090  Triclosan  4.66  2.97–7.29  4.06  2.92–5.64  0.498  .647  4,4′-OH-DE  IC  IC  IC  IC  —  —  Halogenated phenols  2,4,6-TFP  58.26  29.26–116  NB  NB  —  —  2,4,6-TCP  13.68  8.35–22.42  56  42.58–73.65  5.082  .013*  2,4,6-TBP  31.28  21.87–44.74  30.59  26.63–39.6  0.096  .929  2,4,6-TIP  2.68  1.91–3.76  1.58  1.33–1.89  2.501  .090  PBP  1.19  0.83–1.71  2.04  1.59–2.60  2.499  .075  A Welch’s t-test was conducted to identify significant differences in affinity between species (α = .05). Abbreviations: IC, inconclusive due to incomplete curve; NB, no observed binding. Significance is indicated by the following symbols: *p < .05, **p < .01. Figure 3. View largeDownload slide Competitive binding curves of tetrabromobisphenol A and related bisphenols. A, 3,3′,5,5′-tetrahalogenated and methylated bisphenol A compounds (TIBPA, TBBPA, TCBPA, TMBPA) and tetrabromobisphenol S (TBBPS). (B) Debromination products of tetrabromobisphenol A, (C) Bisphenol AF (BPAF), bisphenol S (BPS), and dimethylbisphenol A (DMBPA), as well as BPA and TMBPA for comparison. Human TRβ graphs are on the left side, and zebrafish TRβ are on the right. Purified receptors were incubated with 3 nM [125I]-T3 and increasing concentrations of the bisphenol competitor. Bound and free ligands were separated by the charcoal adsorption method. Data were normalized to the top (100% bound) and bottom (0% bound) plateaus. In the absence of a bottom plateau, data were normalized to the nonradioactive T3 control. The affinity of the bisphenol competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki and 95% confidence intervals were calculated from the average of 3 curves. Figure 3. View largeDownload slide Competitive binding curves of tetrabromobisphenol A and related bisphenols. A, 3,3′,5,5′-tetrahalogenated and methylated bisphenol A compounds (TIBPA, TBBPA, TCBPA, TMBPA) and tetrabromobisphenol S (TBBPS). (B) Debromination products of tetrabromobisphenol A, (C) Bisphenol AF (BPAF), bisphenol S (BPS), and dimethylbisphenol A (DMBPA), as well as BPA and TMBPA for comparison. Human TRβ graphs are on the left side, and zebrafish TRβ are on the right. Purified receptors were incubated with 3 nM [125I]-T3 and increasing concentrations of the bisphenol competitor. Bound and free ligands were separated by the charcoal adsorption method. Data were normalized to the top (100% bound) and bottom (0% bound) plateaus. In the absence of a bottom plateau, data were normalized to the nonradioactive T3 control. The affinity of the bisphenol competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki and 95% confidence intervals were calculated from the average of 3 curves. The binding curves for the 3,3′,5,5′-tetrasubstituted bisphenols in Figure 3A illustrate the role of halogen species on TRβ affinities. Of the 3 tetrahalogenated bisphenol A compounds (TCBPA, TBBPA, TIBPA), the binding curve shifts left with increasing halogen size and weight, indicating the affinity for TRβ is increasing. TIBPA had the highest affinity of all the compounds tested, and the affinity was only approximately 12× lower than T4 for both species. The TIBPA affinities were not significantly different between species (p = .883), however, the affinity for zebrafish TRβ was significantly lower than human TRβ for TBBPA (p = .026) and TCBPA (p = .028). Replacing the dimethylmethylene bridge of TBBPA with a sulfone (TBBPS) decreased TRβ affinity by over 10× for zebrafish and 30× for human. TBBPA and TMBPA have identical molecular volumes (Supplementary Table 1), however, replacing the bromines with methyl groups decreased TRβ affinity 12× for zebrafish and 6× for human TRβ. Results in Figure 3B demonstrate that TRβ affinity increased with increasing levels of bromination. TBBPA had the highest affinity for human TRβ (Ki=0.22 µM), followed by 3,3′,5-BrBPA (0.28 µM), 3,3′-BrBPA (0.82 µM), 3-BrBPA (3.34 µM), and BPA (33.3 µM). Zebrafish TRβ affinities followed the same pattern, however, the affinities were lower than for human TRβ: TBBPA (Ki=0.53 µM) > 3,3′,5-BrBPA (1.42 µM) > 3,3′-BrBPA (4.42 µM) > 3-BrBPA (16.9 µM). BPA was a nonbinder for zebrafish TRβ. A similar pattern was also observed for the methylated bisphenol A compounds (Figure 3C). TMBPA had the highest affinity for both species (Ki=6.5 µM for zebrafish TRβ and 1.33 µM for human TRβ), followed by DMBPA (Ki=13.7 and 4.6 µM, respectively). Replacing the dimethylmethylene bridge of TBBPA with a sulfone (TBBPS) decreased binding affinities for both species, however, species differences were observed for the nonbrominated core structures BPA and BPS (Figure 3C). Similar to TBBPA and TBBPS, human TRβ bound BPS with a 2.2× lower affinity than BPA. Zebrafish TRβ maintained a similar affinity for BPS as human TRβ, however, BPS has a higher affinity for zebrafish TRβ than BPA, as the nonsigmoidal shape of the curve indicates that BPA was a nonbinder for zebrafish. Replacing the methyl groups on the methylene bridge of BPA with trifluoromethyl groups (BPAF) increased affinity 28× for human, and resulted in binding affinity for zebrafish (Figure 3C). Hydroxylated BDE-47 and Related Compounds 3-OH-BDE-47 and 6-OH-BDE-47 bound both receptors with submicromolar affinity (Figure 4A and Table 1). The Ki for 3-OH-BDE-47 was 0.21 µM for zebrafish TRβ and 0.27 for human TRβ. For 6-OH-BDE-47, the Ki values were 0.31 µM for zebrafish and 0.54 µM for human TRβ. The affinity of triclosan, a hydroxylated trichlorodiphenyl ether, was an order of magnitude lower compared with the brominated compounds (Ki = 4.06 µM for human TRβ, and 4.65 µM for zebrafish TRβ). An unpaired t-test with Welch’s correction indicated that the Ki values were not significantly different between species for 3-OH-BDE-47 (p = .32), 6-OH-BDE-47 (p = .09), and triclosan (p = .65). The binding affinities of 4,4′-diOH-DE were not calculated, as binding was not observed except at the highest doses (Figure 4A). BDE-47, the nonhydroxylated PBDE congener, was unable to compete with [125I]-T3 and bind TRβ at any tested concentration. Figure 4. View largeDownload slide Competitive binding curves of (A) Oxidative transformation products of BDE-47 and related compounds, and (B) 2,4,6-tribromophenol and related halogenated phenols. Human TRβ graphs are on the left side, and zebrafish TRβ are on the right. Purified receptors were incubated with 3 nM [125I]-T3 and increasing concentrations of the competitor. Bound and free ligands were separated by the charcoal adsorption method. Data were normalized to the top (100% bound) and bottom (0% bound) plateaus. In the absence of a bottom plateau, data were normalized to the nonradioactive T3 control. The affinity of each competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki and 95% confidence intervals were calculated from the average of 3 curves. Figure 4. View largeDownload slide Competitive binding curves of (A) Oxidative transformation products of BDE-47 and related compounds, and (B) 2,4,6-tribromophenol and related halogenated phenols. Human TRβ graphs are on the left side, and zebrafish TRβ are on the right. Purified receptors were incubated with 3 nM [125I]-T3 and increasing concentrations of the competitor. Bound and free ligands were separated by the charcoal adsorption method. Data were normalized to the top (100% bound) and bottom (0% bound) plateaus. In the absence of a bottom plateau, data were normalized to the nonradioactive T3 control. The affinity of each competitor (Ki) was determined via nonlinear regression using a one-site binding model. The reported Ki and 95% confidence intervals were calculated from the average of 3 curves. 2,4,6-Tribromophenol and Related Compounds Unlike the halogenated bisphenols, the halogenated phenols displayed different affinity patterns between species with regards to halogen size. The affinity of the trihalogenated phenols for human TRβ increased with halogen molecular weight and atomic radius (Figure 4B): TFP (no affinity) < TCP (Ki = 56 µM) < TBP (30.6 µM) < TIP (1.69 µM). However, the affinity for zebrafish TRβ did not follow the same pattern: TFP (Ki = 58.26 µM) < TBP (31.28 µM) < TCP (13.68 µM) < TIP (2.68 µM). The affinity between species was not significantly different for TIP and TBP, however, the affinity of TCP for zebrafish TRβ was over 4× higher than human TRβ (p = .013). Similar to the brominated bisphenols, TRβ affinity increased with increasing degree of bromination. The affinity of pentabromophenol for TRβ was 26× (zebrafish) and 15× (human) higher than TBP. Correlations with Physicochemical Properties Spearman correlations were conducted to compare binding affinities to select physicochemical properties of the test compounds (Figs. 5A–C). Affinity for zebrafish TRβ was negatively correlated with molecular weight (r=−0.84, p < .001), molar volume (r=−0.21, p = .023), and log KOW (r=−0.85, p < .001). Human TRβ affinity was also negatively correlated with molecular weight (r=−0.71, p < .001), molar volume (r=−0.68, p = .001), and log KOW (r=−0.85, p < .001). The affinities between the 2 receptors were also correlated (r = 0.86, p < .001) (Figure 5D). Figure 5. View largeDownload slide Spearman rank correlations. A–C, Correlations between chemical properties of the tested compounds and the Ki values for human TRβ (black circles, dashed line) and zebrafish TRβ (white circles, solid line). Properties include (A) molecular weight (g/mol), (B) molar volume (cm3/mol), and (C) log KOW. Chemical properties were obtained from SciFinder (scifinder.cas.org). Statistical significance was set at α=.5. D, Correlation between the Ki values (µM) of zebrafish TRβ and human TRβ for all tested compounds. Statistical significance was set at α=.5. Figure 5. View largeDownload slide Spearman rank correlations. A–C, Correlations between chemical properties of the tested compounds and the Ki values for human TRβ (black circles, dashed line) and zebrafish TRβ (white circles, solid line). Properties include (A) molecular weight (g/mol), (B) molar volume (cm3/mol), and (C) log KOW. Chemical properties were obtained from SciFinder (scifinder.cas.org). Statistical significance was set at α=.5. D, Correlation between the Ki values (µM) of zebrafish TRβ and human TRβ for all tested compounds. Statistical significance was set at α=.5. DISCUSSION Brominated phenolic contaminants have long been known to disrupt the thyroid endocrine system, however, the potential of these compounds to disrupt TR-mediated gene expression is currently unresolved. This study assessed how structural variations of 3 common BPCs impact binding affinity for TRβ. As BPC-mediated thyroid disruption is a concern for both humans and wildlife, a comparative approach was applied by selecting receptors from 2 distantly related species: human and zebrafish. Zebrafish are a popular model for vertebrate development and thyroid disruption, with relevance to both human and ecosystem health. Overall, both receptors displayed similar ligand selectivity and specificity for the test compounds, and the ligand affinities for each receptor were highly correlated (r = 0.86, p < .001). However, some species-specific differences in ligand affinity were observed. The Ki values for 9 of the 22 test compounds were significantly different between species. Eight of these compounds had significantly higher affinities for human TRβ compared with zebrafish, and all 8 were part of the TBBPA group. The affinity of the ninth compound, TCP, was significantly higher for zebrafish TRβ over human. In addition, TFP has a higher affinity for zebrafish TRβ as it did not bind human, whereas the opposite was observed for BPA, which bound human TRβ but not zebrafish. The fact that the 2 receptors display divergent affinities for a select group of compounds yet highly conserved affinities for thyroid hormones is intriguing. The molecular interactions driving the differential affinities is currently unknown, however, there are several potential possibilities involving species-specific differences within the LBD. The LBDs of zebrafish and human differ by 23 amino acids. Most of these differences occur between H1 to the beginning of H3, and H9 to between H10 and H11 (Figure 1). The physicochemical properties of the species-specific amino acids may influence the tertiary protein structure of the receptor and the chemical nature of the LBP. The resulting effects on receptor structure would likely be minor or localized to avoid disrupting interactions with thyroid hormones. However, minor structural changes may be enough to enable human TRβ to better accommodate the larger bisphenols, whereas zebrafish TRβ may be better able to tightly pack around the smaller halogenated phenols. Aside from altering receptor structure, the amino acid variations within the LBPs may lead to novel ligand-receptor interactions with the tested BPCs that are absent in interactions with native thyroid hormones. For example, the LBPs of human TRα and TRβ differ by 1 amino acid: serine-277 of TRα is an asparagine (N331) in TRβ. Asparagine increases the stability of the hydrogen bond formed with the carboxylate of the synthetic thyromimetic GC-1, and increased affinity and selectivity of GC-1 for TRβ over TRα (Bleicher et al., 2008). Despite the observed affinity differences, both receptors responded in a similar manner to the structural changes. The structural feature that had the largest impact on binding affinities was halogenation. Thyroid hormones are one of the few biogenic ligands that are halogenated, and it has been hypothesized that the iodines play an active role in ligand-receptor interactions (Valadares et al., 2009). In general, we observed that TRβ affinity increased with increasing halogen mass and atomic radius (F < Cl < Br < I) for both species, with the iodinated compounds (TIBPA and TIP) having the highest affinity within their compound classes. Increasing halogen mass and atomic radius increases the molecular weight, volume, and hydrophobicity of a compound, which are all highly correlated with increasing affinity (Figs. 5A–C). Increasing the degree of halogenation also positively influenced affinity for TRβ from both species, as is evident with the brominated bisphenols. The addition of a single bromine onto C3 of BPA increased affinity for TRβ by approximately 10× for both species, and affinity increased with each subsequent bromine. We originally hypothesized that 3,3′,5-tribromobisphenol A would have the highest affinity of the brominated BPAs, as this compound is most analogous to T3 from a halogen substituent configuration standpoint. However, our results show that TBBPA, which is more analogous to T4, had the highest affinity. The effect of an additional bromine on TBBPA vs. 3,3′,5-tribromobisphenol A may help compensate for the shorter length of the molecule compared with T3. The role of halogen size and number in determining binding affinity was also evident for the halogenated phenols. At first, the halogenated phenols were not expected to have affinity for TRβ due to the fact that TBP has been previously shown to lack affinity for amphibian and rat TRβ (Kitamura et al., 2008; Kudo et al., 2006). In contrast to previous studies, our results demonstrate that TBP and other halogenated phenols bound TRβ, albeit with generally lower affinities compared with larger diphenyl compounds such as the halogenated bisphenols. In addition, affinity increased with increasing halogen mass, volume, and degree of halogenation, and helped compensate for the smaller size of the compound. For example, the affinity of zebrafish TRβ for PBP and TIP was in the same range as TCBPA, whereas human TRβ affinity for both compounds was comparable with BPAF and 3-BrBPA. Although halogenation appears to greatly influence TRβ binding affinities, the exact mechanisms of action is not completely understood. Halogens have long been considered merely electronegative and hydrophobic moieties that did not directly participate in ligand-protein interactions. However, studies in the past decade have yielded an intriguing possibility: that halogens play an active role in ligand affinities by directly interacting with receptors and other proteins through the formation of halogen bonds (X-bonds) (reviewed in Politzer et al., 2010; Scholfield et al., 2013; Wilcken et al., 2013). Halogen bonds are noncovalent, electrostatic, and directional interactions between a region of positive electrostatic potential on the halogen opposite the R-X covalent bond and an electronegative Lewis base (O, N, S) on the receptor. They are highly directional (≥160˚), and their distance is less than the sum of the van der Waals radii for the 2 molecules. Although halogens are considered electronegative substituents, the distribution of electrostatic potential across the halogen’s surface is not uniform. The polarizability and electronegativity of the halogen, as well as the electron-withdrawing ability of nearby functional groups, can result in an area of positive charge on the halogen directly opposite the X-C covalent bond. This positive region (referred to as a σ-hole) may form a noncovalent and electrostatic interaction with a nearby negative site on the receptor, thus contributing to the strength and selectivity of the ligand-receptor interactions. Halogen bonds have been identified in numerous biological systems in recent studies and through retrospective surveys of protein databases, including proteins involved in the thyroid endocrine system. The thyroid hormone transport protein transthyretin (TTR) has been shown to form halogen bonds with the 3′-iodine and 5′-iodine of T4 in numerous species (Eneqvist et al., 2004; Wojtczak et al., 2001). Deiodinases form halogen bonds with the iodines of thyroid hormones to aid in cleaving the carbon-iodine bond when converting T4 to T3 (Manna et al., 2015). Additionally, a recent QSAR study identified a halogen bond between F272 of TRβ and the 3-I of T3 or 3-Br of brominated T3 analogs (PBD codes 1XZX and 2J4A) (Valadares et al., 2009). Most studies thus far have been limited to thyroid hormones, but they suggest the possibility that additional halogenated phenolic compounds may form halogen bonds with thyroid hormone proteins. Although the current data are limited, it is possible that halogen bonds may play a significant role in determining the TRβ affinities of halogenated phenolic compounds, and is worthy of further exploration. The current results with TBBPA and TBBPS, as well as BPA, BPS, and BPAF indicate that the bridge group of the diphenyl compounds have a role in determining TRβ affinities. The ether bridge of thyroid hormones is not directly involved in hormone-receptor interactions, but instead is responsible for maintaining the correct bond angle and perpendicular orientation of the 2 aromatic rings relative to each other, which aids in determining the fit of the ligand in the LBP (de Araujo et al., 2010; Luthe et al., 2008). Within the diphenyl structures tested, the aromatic rings were bridged by either an ether, a dimethylmethylene, or a sulfone. The current results support the notion that ring orientations that diverge from that observed for T3 may negatively impact binding affinities, as illustrated by the results with TBBPA and TBBPS. Both compounds are tetrabrominated bisphenols, however, the aromatic rings of TBBPS are linked by a sulfone instead of a dimethylmethylene group of TBBPA. Like T3, the 2 rings of TBBPA are perpendicular to each other in a ‘twist’ conformation. The sulfone linkage of TBBPS orients the 2 rings ‘face to face’ rather than perpendicular as in T3 and TBBPA (Sakai et al., 2009). The significantly lower affinity of TBBPS compared with TBBPA is potentially due to the suboptimal ring orientation and position of TBBPS within the LBP, thus disrupting essential ligand-receptor interactions. Although BPA was able to compete with [125I]-T3 and bind human TRβ (albeit with very low affinity), the cause of the irregular nonsigmoidal BPA curve for zebrafish TRβ is not clear. Irregular binding curves typically occur due to reasons unrelated to true competitive binding, such as test compound precipitation, interference with charcoal adsorption, or protein denaturation (Freyberger and Ahr, 2004; Hornung et al., 2017; Laws et al., 2006). Nonsigmoidal curves display a variety of characteristics including a U-shape or inverted U-shape, a steep slope, and plateau before full competition takes place. Hormone binding > 100% has been observed previously with other nuclear receptors, and was found to be due to nonreceptor mediated effects that compromised the integrity and stability of the assay (Blair et al., 2000; Hornung et al., 2017). The current results suggest the issue is specific to the zebrafish TRβ protein and BPA, as this curve shape was not observed with human TRβ or any other test compounds. Considering that the only variable between the human and zebrafish assays is the receptor protein, it is unclear why BPA would disrupt the zebrafish curve and not human. Further work is necessary to determine the true cause of the nonsigmoidal curve. One of the greatest strengths of binding assays is also a major limitation: binding assays only determine if a compound interacts with a receptor of interest, and quantifies the strength of that interaction (Pollard, 2010). They do not distinguish between agonists and antagonists, or inform on subsequent downstream events in the activation pathway. Additional studies are necessary to explore how the differential binding affinities impact key intermediate events in the TR activation pathway between human and zebrafish. These assays would also inform on whether the observed species differences translate into differential cellular and organism level effects. The identification of species-specific sensitivities to contaminants will aid our understanding of the risk of adverse effects in more sensitive populations. Another limitation to this study is that the TRα paralogs were excluded. Future studies that include both zebrafish TRα paralogs could significantly increase our understanding of the molecular functions and potential evolutionary significance of the duplicate genes. A standardized binding assay for TRβ is potentially an effective tool to include in the initial steps of endocrine disruptor screening programs. The assay is a simple and rapid method to screen for chemicals that interact with TRs, and could aid in prioritizing chemicals of concern for further testing and characterization. However, additional development and optimization is necessary to fully characterize assay robustness and reproducibility under varying assay conditions, and to identify potential confounders and sources of bias. For example, as multiple compounds have a lower affinity for zebrafish TRβ compared with human, additional work is necessary to determine if zebrafish TRβ may be more sensitive to false negatives. Assay confounders can be negated with assay standardization and validation, including receptors from > 1 species, and including additional assays that assess other downstream receptor-mediated endpoints such as protein interaction or target gene expression. In conclusion, this study assessed how structural variations of 3 common classes of BPCs impact binding affinity for TRβ from 2 species. In general, structural variations impacted binding affinities of both receptors in a similar manner. TRβ affinity increased with increasing halogen size and decreasing electronegativity, as well as increasing the degree of halogenation. Although many of the calculated affinities were comparable, species differences were identified with the halogenated phenols and many of the bisphenols, possibly due to amino acid differences within the ligand binding domains. The bridge group of diphenyl compounds also influences binding affinities, likely due to varying orientations of the phenyl rings. However, further studies are necessary to explore how the identified structural modifications impact downstream receptor activities and resulting in vivo effects. Understanding how chemicals mimic native thyroid hormones and disrupt normal thyroid signaling is essential to predicting the effects of these chemicals on human and ecosystem health. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. FUNDING This work was supported by a grant from the National Institute of Environmental Health Sciences at the National Institutes of Health [P42ES010356]. ACKNOWLEDGMENT The authors would like to thank Dr Wu Dong for his efforts in cloning the zebrafish TRβ construct. REFERENCES Aranda A., Alonso-Merino E., Zambrano A. ( 2013). Receptors of thyroid hormones. Pediatr. Endocr. Rev . 11, 2– 13. Berman H. M., Westbrook J., Feng Z., Gilliland G., Bhat T. N., Weissig H., Shindyalov I. N., Bourne P. E. ( 2000). The Protein Data Bank. Nucleic Acids Res . 28, 235– 242. Google Scholar CrossRef Search ADS PubMed  Bertrand S., Thisse B., Tavares R., Sachs L., Chaumot A., Bardet P. L., Escrivá H., Duffraisse M., Marchand O., Safi R.et al.  , ( 2007). Unexpected novel relational links uncovered by extensive developmental profiling of nuclear receptor expression. PLoS Genet.  3, e188– 2100. Google Scholar CrossRef Search ADS PubMed  Blair R. M., Fang H., Branham W. S., Hass B. S., Dial S. L., Moland C. L., Tong W. D., Shi L. M., Perkins R., Sheehan D. M. ( 2000). The estrogen receptor relative binding affinities of 188 natural and xenochemicals: structural diversity of ligands. Toxicol. Sci . 54, 138– 153. Google Scholar CrossRef Search ADS PubMed  Blanton M. L., Specker J. L. ( 2007). The hypothalamic-pituitary-thyroid (HPT) axis in fish and its role in fish development and reproduction. Crit. Rev. Toxicol . 37, 97– 115. http://dx.doi.org/10.1080/10408440601123529 Google Scholar CrossRef Search ADS PubMed  Bleicher L., Aparicio R., Nunes F. M., Martinez L., Dias S. M. G., Figueira A. C. M., Santos M. A. M., Venturelli W. H., da Silva R., Donate P. M.et al.  , ( 2008). Structural basis of GC-1 selectivity for thyroid hormone receptor isoforms. BMC Struct. Biol . 8, 8. Google Scholar CrossRef Search ADS PubMed  Covaci A., Voorspoels S., Abdallah M. A. E., Geens T., Harrad S., Law R. J. ( 2009). Analytical and environmental aspects of the flame retardant tetrabromobisphenol-A and its derivatives. J. Chromatogr. A  1216, 346– 363. Google Scholar CrossRef Search ADS PubMed  de Araujo A. S., Martínez L., Nicoluci R. D., Skaf M. S., Polikarpov I. ( 2010). Structural modeling of high-affinity thyroid receptor-ligand complexes. Eur. Biophys. J. Biophys . 39, 1523– 1536. Google Scholar CrossRef Search ADS   De la Torre A., Pacepavicius G., Martínez M. A., Darling C., Muir D., Sherry J., McMaster M., Alaee M. ( 2013). Polybrominated diphenyl ethers and their methoxylated and hydroxylated analogs in brown bullhead (Ameiurus nebulosus) plasma from Lake Ontario. Chemosphere  90, 1644– 1651. Google Scholar CrossRef Search ADS PubMed  Deng J., Davies D. R., Wisedchaisri G., Wu M., Hol W. G. J., Mehlin C. ( 2004). An improved protocol for rapid freezing of protein samples for long term storage. Acta. Crystallogr. D. Biol. Crystallogr  60, 203– 204. Google Scholar CrossRef Search ADS PubMed  Dong W., Macaulay L. J., Kwok K. W., Hinton D. E., Ferguson P. L., Stapleton H. M. ( 2014). The PBDE metabolite 6-OH-BDE-47 affects melanin pigmentation and THRβ mRNA expression in the eye of zebrafish embryos. Endocr. Disruptors (Austin)  2, e969072. Google Scholar CrossRef Search ADS PubMed  Eneqvist T., Lundberg E., Karlsson A., Huang S., Santos C. R., Power D. M., Sauer-Eriksson A. E. ( 2004). High resolution crystal structures of piscine transthyretin reveal different binding modes for triiodothyronine and thyroxine. J. Biol. Chem . 279, 26411– 26416. Google Scholar CrossRef Search ADS PubMed  Fang H., Tong W. D., Branham W. S., Moland C. L., Dial S. L., Hong H. X., Xie Q., Perkins R., Owens W., Sheehan D. M. ( 2003). Study of 202 natural, synthetic, and environmental chemicals for binding to the androgen receptor. Chem. Res. Toxicol . 16, 1338– 1358. Google Scholar CrossRef Search ADS PubMed  Freyberger A., Ahr H. J. ( 2004). Development and standardization of a simple binding assay for the detection of compounds with affinity for the androgen receptor. Toxicology  195, 113– 126. http://dx.doi.org/10.1016/j.tox.2003.09.008 Google Scholar CrossRef Search ADS PubMed  Hahn M. W. ( 2009). Distinguishing among evolutionary models for the maintenance of gene duplicates. J. Hered.  100, 605– 617. http://dx.doi.org/10.1093/jhered/esp047 Google Scholar CrossRef Search ADS PubMed  Heijlen M., Houbrechts A. M., Darras V. M. ( 2013). Zebrafish as a model to study peripheral thyroid hormone metabolism in vertebrate development. Gen. Comp. Endocrinol . 188, 289– 296. http://dx.doi.org/10.1016/j.ygcen.2013.04.004 Google Scholar CrossRef Search ADS PubMed  Holzer G., Laudet V. ( 2015). Thyroid hormones: A triple-edged sword for life history transitions. Curr. Biol . 25, R344– R347. Google Scholar CrossRef Search ADS PubMed  Hornung M. W., Tapper M. A., Denny J. S., Sheedy B. R., Erickson R., Sulerud T. J., Kolanczyk R. C., Schmieder P. K. ( 2017). Avoiding false positives and optimizing identification of true negatives in estrogen receptor binding and agonist/antagonist assays. Appl. In Vitro Toxicol . 3, 163– 181. Google Scholar CrossRef Search ADS   Houde M., Pacepavicius G., Darling C., Fair P. A., Alaee M., Bossart G. D., Solomon K. R., Letcher R. J., Bergman A., Marsh G.et al.  , ( 2009). Polybrominated diphenyl ethers and their hydroxylated analogs in plasma of bottlenose dolphins (Tursiops truncatus) from the United States east coast. Environ. Toxicol. Chem . 28, 2061– 2068. Google Scholar CrossRef Search ADS PubMed  Ishihara A., Nishiyama N., Sugiyama S., Yamauchi K. ( 2003a). The effect of endocrine disrupting chemicals on thyroid hormone binding to Japanese quail transthyretin and thyroid hormone receptor. Gen. Comp. Endocrinol . 134, 36– 43. Google Scholar CrossRef Search ADS   Ishihara A., Sawatsubashi S., Yamauchi K. ( 2003b). Endocrine disrupting chemicals: Interference of thyroid hormone binding to transthyretins and to thyroid hormone receptors. Mol. Cell. Endocrinol . 199, 105– 117. Google Scholar CrossRef Search ADS   Kitamura S., Jinno N., Suzuki T., Sugihara K., Ohta S., Kuroki H., Fujimoto N. ( 2005a). Thyroid hormone-like and estrogenic activity of hydroxylated PCBs in cell culture. Toxicology  208, 377– 387. Google Scholar CrossRef Search ADS   Kitamura S., Kato T., Iida M., Jinno N., Suzuki T., Ohta S., Fujimoto N., Hanada H., Kashiwagi K., Kashiwagi A. ( 2005b). Anti-thyroid hormonal activity of tetrabromobisphenol A, a flame retardant, and related compounds: affinity to the mammalian thyroid hormone receptor, and effect on tadpole metamorphosis. Life Sci . 76, 1589– 1601. Google Scholar CrossRef Search ADS   Kitamura S., Shinohara S., Iwase E., Sugihara K., Uramaru N., Shigematsu H., Fujimoto N., Ohta S. ( 2008). Affinity for thyroid hormone and estrogen receptors of hydroxylated polybrominated diphenyl ethers. J. Health Sci . 54, 607– 614. Google Scholar CrossRef Search ADS   Kudo Y., Yamauchi K. ( 2005). In vitro and in vivo analysis of the thyroid disrupting activities of phenolic and phenol compounds in Xenopus laevis. Toxicol. Sci . 84, 29– 37. http://dx.doi.org/10.1093/toxsci/kfi049 Google Scholar CrossRef Search ADS PubMed  Kudo Y., Yamauchi K., Fukazawa H., Terao Y. ( 2006). In vitro and in vivo analysis of the thyroid system-disrupting activities of brominated phenolic and phenol compounds in Xenopus laevis. Toxicol. Sci . 92, 87– 95. http://dx.doi.org/10.1093/toxsci/kfj204 Google Scholar CrossRef Search ADS PubMed  Kumar S., Hedges S. B. ( 1998). A molecular timescale for vertebrate evolution. Nature  392, 917– 920. http://dx.doi.org/10.1038/31927 Google Scholar CrossRef Search ADS PubMed  Laws S. C., Yavanhxay S., Cooper R. L., Eldridge J. C. ( 2006). Nature of the binding interaction for 50 structurally diverse chemicals with rat estrogen receptors. Toxicol. Sci . 94, 46– 56. Google Scholar CrossRef Search ADS PubMed  Leonetti C., Butt C. M., Hoffman K., Miranda M. L., Stapleton H. M. ( 2016). Concentrations of polybrominated diphenyl ethers (PBDEs) and 2, 4, 6-tribromophenol in human placental tissues. Environ. Int . 88, 23– 29. Google Scholar CrossRef Search ADS PubMed  Liu Y. W., Lo L. J., Chan W. K. ( 2000). Temporal expression and T3 induction of thyroid hormone receptors alpha1 and beta1 during early embryonic and larval development in zebrafish, Danio rerio. Mol. Cell. Endocrinol . 159, 187– 195. http://dx.doi.org/10.1016/S0303-7207(99)00193-8 Google Scholar CrossRef Search ADS PubMed  Luthe G., Jacobus J. A., Robertson L. W. ( 2008). Receptor interactions by polybrominated diphenyl ethers versus polychlorinated biphenyls: a theoretical structure-activity assessment. Environ. Toxicol. Phar . 25, 202– 210. http://dx.doi.org/10.1016/j.etap.2007.10.017 Google Scholar CrossRef Search ADS   Macaulay L. J., Chen A., Rock K. D., Dishaw L. V., Dong W., Hinton D. E., Stapleton H. M. ( 2015). Developmental toxicity of the PBDE metabolite 6-OH-BDE-47 in zebrafish and the potential role of thyroid receptor β. Aquat. Toxicol . 168, 38– 47. Google Scholar CrossRef Search ADS PubMed  Manna D., Mondal S., Mugesh G. ( 2015). Halogen bonding controls the regioselectivity of the deiodination of thyroid hormones and their sulfate analogues. Chemistry  21, 2409– 2416. http://dx.doi.org/10.1002/chem.201405442 Google Scholar CrossRef Search ADS PubMed  Mendoza A., Navarrete-Ramírez P., Hernández-Puga G., Villalobos P., Holzer G., Renaud J. P., Laudet V., Orozco A. ( 2013). 3, 5-T2 is an alternative ligand for the thyroid hormone receptor β1. Endocrinology  154, 2948– 2958. Google Scholar CrossRef Search ADS PubMed  Meyer A., Van de Peer Y. ( 2005). From 2R to 3R: Evidence for a fish-specific genome duplication (FSGD). Bioessays  27, 937– 945. http://dx.doi.org/10.1002/bies.20293 Google Scholar CrossRef Search ADS PubMed  Mullur R., Liu Y. Y., Brent G. A. ( 2014). Thyroid hormone regulation of metabolism. Physiol. Rev . 94, 355– 382. http://dx.doi.org/10.1152/physrev.00030.2013 Google Scholar CrossRef Search ADS PubMed  Nascimento A. S., Dias S. M., Nunes F. M., Aparício R., Ambrosio A. L., Bleicher L., Figueira A. C., Santos M. A., de Oliveira Neto M., Fischer H.et al.  , ( 2006). Structural rearrangements in the thyroid hormone receptor hinge domain and their putative role in the receptor function. J. Mol. Biol . 360, 586– 598. Google Scholar CrossRef Search ADS PubMed  Nomiyama K., Eguchi A., Mizukawa H., Ochiai M., Murata S., Someya M., Isobe T., Yamada T. K., Tanabe S. ( 2011a). Anthropogenic and naturally occurring polybrominated phenolic compounds in the blood of cetaceans stranded along Japanese coastal waters. Environ. Pollut . 159, 3364– 3373. Google Scholar CrossRef Search ADS   Nomiyama K., Uchiyama Y., Horiuchi S., Eguchi A., Mizukawa H., Hirata S. H., Shinohara R., Tanabe S. ( 2011b). Organohalogen compounds and their metabolites in the blood of Japanese amberjack (Seriola quinqueradiata) and scalloped hammerhead shark (Sphyrna lewini) from Japanese coastal waters. Chemosphere  85, 315– 321. Google Scholar CrossRef Search ADS   Nowell M. A., Power D. M., Canario A. V., Llewellyn L., Sweeney G. E. ( 2001). Characterization of a sea bream (Sparus aurata) thyroid hormone receptor-β clone expressed during embryonic and larval development. Gen. Comp. Endocrinol . 123, 80– 89. Google Scholar CrossRef Search ADS PubMed  Noyes P. D., Lema S. C., Macaulay L. J., Douglas N. K., Stapleton H. M. ( 2013). Low level exposure to the flame retardant BDE-209 reduces thyroid hormone levels and disrupts thyroid signaling in fathead minnows. Environ. Sci. Technol . 47, 10012– 10021. Google Scholar CrossRef Search ADS PubMed  OECD ( 2015). Test No. 493: Performance-Based Guideline for Human Recombinant Estrogen Receptor (hrER) In Vitro Assays to Detect Chemicals with ER Binding Affinity. Available at: http://dx.doi.org/10.1787/9789264242623-en. Accessed May 24, 2017. Pascual A., Aranda A. ( 2013). Thyroid hormone receptors, cell growth and differentiation. BBA-Gen. Subjects  1830, 3908– 3916. http://dx.doi.org/10.1016/j.bbagen.2012.03.012 Google Scholar CrossRef Search ADS   Politzer P., Murray J. S., Clark T. ( 2010). Halogen bonding: An electrostatically-driven highly directional noncovalent interaction. Phys. Chem. Chem. Phys . 12, 7748– 7757. http://dx.doi.org/10.1039/c004189k Google Scholar CrossRef Search ADS PubMed  Pollard T. D. ( 2010). A guide to simple and informative binding assays. Mol. Biol. Cell  21, 4061– 4067. http://dx.doi.org/10.1091/mbc.E10-08-0683 Google Scholar CrossRef Search ADS PubMed  Porazzi P., Calebiro D., Benato F., Tiso N., Persani L. ( 2009). Thyroid gland development and function in the zebrafish model. Mol. Cell. Endocrinol . 312, 14– 23. Google Scholar CrossRef Search ADS PubMed  Préau L., Fini J. B., Morvan-Dubois G., Demeneix B. ( 2015). Thyroid hormone signaling during early neurogenesis and its significance as a vulnerable window for endocrine disruption. BBA-Gene Regul. Mech . 1849, 112– 121. Qiu X. H., Bigsby R. M., Hites R. A. ( 2009). Hydroxylated metabolites of polybrominated diphenyl ethers in human blood samples from the United States. Environ. Health. Perspect . 117, 93– 98. http://dx.doi.org/10.1289/ehp.11660 Google Scholar CrossRef Search ADS PubMed  Raldúa D., Thienpont B., Babin P. J. ( 2012). Zebrafish eleutheroembryos as an alternative system for screening chemicals disrupting the mammalian thyroid gland morphogenesis and function. Reprod. Toxicol . 33, 188– 197. Google Scholar CrossRef Search ADS PubMed  Ren X. M., Guo L. H., Gao Y., Zhang B. T., Wan B. ( 2013). Hydroxylated polybrominated diphenyl ethers exhibit different activities on thyroid hormone receptors depending on their degree of bromination. Toxicol. Appl. Pharmacol . 268, 256– 263. http://dx.doi.org/10.1016/j.taap.2013.01.026 Google Scholar CrossRef Search ADS PubMed  Sakai H., Kawashima A., Kashima Y., Yamada-Okabe T. ( 2009). Characterization of the organohalogen compounds which affect gene expressions mediated by thyroid hormone receptors. Interd. Stud. Environ. Chem . 2, 203– 210. Sandler B., Webb P., Apriletti J. W., Huber B. R., Togashi M., Cunha Lima S. T., Juric S., Nilsson S., Wagner R., Fletterick R. J.et al.  , ( 2004). Thyroxine-thyroid hormone receptor interactions. J. Biol. Chem . 279, 55801– 55808. Google Scholar CrossRef Search ADS PubMed  Scholfield M. R., Zanden C. M. V., Carter M., Ho P. S. ( 2013). Halogen bonding (X-bonding): a biological perspective. Protein Sci . 22, 139– 152. Google Scholar CrossRef Search ADS PubMed  Stapleton H. M., Eagle S., Anthopolos R., Wolkin A., Miranda M. L. ( 2011). Associations between polybrominated diphenyl ether (PBDE) flame retardants, phenolic metabolites, and thyroid hormones during pregnancy. Environ. Health Perspect . 119, 1454– 1459. Google Scholar CrossRef Search ADS PubMed  Stapleton H. M., Eagle S., Sjodin A., Webster T. F. ( 2012). Serum PBDEs in a North Carolina toddler cohort: Associations with handwipes, house dust, and socioeconomic variables. Environ. Health Perspect . 120, 1049– 1054. Google Scholar CrossRef Search ADS PubMed  Strange P. G. ( 1992). Charcoal adsorption for separating bound and free radioligand in radioligand binding assays. In Receptor Ligand Interactions: A Practical Approach  ( Hulme E., Ed.), pp. 247– 253. Oxford University Press, New York. Suzuki G., Takigami H., Watanabe M., Takahashi S., Nose K., Asari M., Sakai S. ( 2008). Identification of brominated and chlorinated phenols as potential thyroid-disrupting compounds in indoor dusts. Environ. Sci. Technol . 42, 1794– 1800. Google Scholar CrossRef Search ADS PubMed  Thienpont B., Tingaud-Sequeira A., Prats E., Barata C., Babin P. J., Raldúa D. ( 2011). Zebrafish eleutheroembryos provide a suitable vertebrate model for screening chemicals that impair thyroid hormone synthesis. Environ. Sci. Technol . 45, 7525– 7532. Google Scholar CrossRef Search ADS PubMed  Thompson J. D., Higgins D. G., Gibson T. J. ( 1994). ClustalW – Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res . 22, 4673– 4680. http://dx.doi.org/10.1093/nar/22.22.4673 Google Scholar CrossRef Search ADS PubMed  Togashi M., Nguyen P., Fletterick R., Baxter J. D., Webb P. ( 2005). Rearrangements in thyroid hormone receptor charge clusters that stabilize bound 3, 5′, 5-triiodo-L-thyronine and inhibit homodimer formation. J. Biol. Chem . 280, 25665– 25673. Google Scholar CrossRef Search ADS PubMed  Unwin R., Fenton J., Whitsitt M., Jamison C., Stupar M., Jakobsson E., Subramaniam S. ( 2002). Biology Workbench: A computing and analysis environment for the biological sciences. In Bioinformatics: Databases and Systems  ( Letovsky S., Ed.) doi: https://doi.org/10.1007/0-306-46903-0_20, pp. 233– 244. Springer, Boston, MA. Google Scholar CrossRef Search ADS   USEPA ( 2012). Chemical Data Access Tool (CDAT). Available at: https://java.epa.gov/oppt_chemical_search/. Accessed November 29, 2016. USEPA ( 2015). TSCA Work Plan Chemical Problem Formulation and Initial Assessment: Tetrabromobisphenol A and Related Chemicals Cluster. Available at: https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/tsca-work-plan-chemical-problem-formulation-and-2. Accessed October 14, 2016. USEPA ( 2017). Endocrine Disruptor Screening Program Tier 1 Battery of Assays. Available at: https://www.epa.gov/endocrine-disruption/endocrine-disruptor-screening-program-tier-1-battery-assays. Accessed 7 September, 2017. Valadares N. F., Salum L. B., Polikarpov I., Andricopulo A. D., Garratt R. C. ( 2009). Role of halogen bonds in thyroid hormone receptor selectivity: pharmacophore-based 3D-QSSR studies. J. Chem. Inf. Model . 49, 2606– 2616. Google Scholar CrossRef Search ADS PubMed  Valters K., Li H. X., Alaee M., D’Sa I., Marsh G., Bergman A., Letcher R. J. ( 2005). Polybrominated diphenyl ethers and hydroxylated and methoxylated brominated and chlorinated analogues in the plasma of fish from the Detroit River. Environ. Sci. Technol . 39, 5612– 5619. Google Scholar CrossRef Search ADS PubMed  Wagner R. L., Apriletti J. W., Mcgrath M. E., West B. L., Baxter J. D., Fletterick R. J. ( 1995). A structural role for hormone in the thyroid hormone receptor. Nature  378, 690– 697. Google Scholar CrossRef Search ADS PubMed  Walpita C. N., Van der Geyten S., Rurangwa E., Darras V. M. ( 2007). The effect of 3, 5, 3′-triiodothyronine supplementation on zebrafish (Danio rerio) embryonic development and expression of iodothyronine deiodinases and thyroid hormone receptors. Gen. Comp. Endocrinol . 152, 206– 214. Google Scholar CrossRef Search ADS PubMed  Wilcken R., Zimmermann M. O., Lange A., Joerger A. C., Boeckler F. M. ( 2013). Principles and applications of halogen bonding in medicinal chemistry and chemical biology. J. Med. Chem . 56, 1363– 1388. Google Scholar CrossRef Search ADS PubMed  Williams G. R. ( 2000). Cloning and characterization of two novel thyroid hormone receptor β isoforms. Mol. Cell. Biol . 20, 8329– 8342. Google Scholar CrossRef Search ADS PubMed  Wojtczak A., Cody V., Luft J. R., Pangborn W. ( 2001). Structure of rat transthyretin (rTTR) complex with thyroxine at 2.5 Å resolution: First non-biased insight into thyroxine binding reveals different hormone orientation in two binding sites. Acta. Crystallogr. D. Biol. Crystallogr . 57, 1061– 1070. Google Scholar CrossRef Search ADS PubMed  Yamauchi K., Ishihara A., Fukazawa H., Terao Y. ( 2003). Competitive interactions of chlorinated phenol compounds with 3, 3′, 5-triiodothyronine binding to transthyretin: Detection of possible thyroid-disrupting chemicals in environmental waste water. Toxicol. Appl. Pharmacol . 187, 110– 117. Google Scholar CrossRef Search ADS PubMed  Zoeller R. T., Brown T. R., Doan L. L., Gore A. C., Skakkebaek N. E., Soto A. M., Woodruff T. J., Saal F. S. V. ( 2012). Endocrine-disrupting chemicals and public health protection: A statement of principles from The Endocrine Society. Endocrinology  153, 4097– 4110. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - The Affinity of Brominated Phenolic Compounds for Human and Zebrafish Thyroid Receptor β: Influence of Chemical Structure
JF - Toxicological Sciences
DO - 10.1093/toxsci/kfy028
DA - 2018-05-01
UR - https://www.deepdyve.com/lp/oxford-university-press/the-affinity-of-brominated-phenolic-compounds-for-human-and-zebrafish-3fKnB0s0Y3
SP - 226
EP - 239
VL - 163
IS - 1
DP - DeepDyve
ER -