A Possible Novel Mechanism of Action of Genistein and Daidzein for Activating Thyroid Hormone Receptor-Mediated Transcription

A Possible Novel Mechanism of Action of Genistein and Daidzein for Activating Thyroid Hormone... Abstract Thyroid hormone receptors (TRs) are members of the nuclear receptor superfamily that regulate their target genes for controlling organ development and functional maintenance. Soybean isoflavones, especially genistein and daidzein, modulate various hormone-mediated pathways. However, their effects on TRs have not yet been extensively studied. In this study, the effects of these isoflavones on TR action were evaluated using transient transfection-based reporter gene assays and molecular docking studies. Genistein and daidzein augmented T3-liganded TR-mediated transcription in a concentration-dependent manner. In the mammalian 2-hybrid study, these isoflavones augmented the recruitment of steroid receptor coactivator-1 and nuclear corepressor to liganded or unliganded TRs. Using a series of mutant TRs, we also showed that the activation function-2 domain of TRs was responsible for the augmentation by these isoflavones. CV-1 cells had expressed TRα, TRβ1, and ERα mRNAs. However, neither the overexpression nor the knocking down of ERα altered the augmentation of TR action by isoflavones, indicating that the effects of isoflavones are exerted through their direct action on TRs. In silico molecular docking studies showed that genistein and daidzein can directly bind to the TR-ligand-binding domain. These findings indicate that the augmentation of the TR-mediated transcription by genistein and daidzein is due to their direct binding to TR-ligand-binding domain to induce the recruitment of steroid receptor coactivator-1. Our study reports a novel mode of action of soybean isoflavones on TR function. The biological effects and the relevance of these isoflavones to human health may be partially attributable to the activation of thyroid hormone signaling. isoflavone, thyroid hormone receptor, estrogen receptor, T3, molecular docking Thyroid hormones (THs) (3, 5, 3’-tri-iodo-L-thyronine or T3; 3, 5, 3’, 5’-tetra-iodo-L-thyronine or thyroxine or T4) are essential for the development and functional maintenance of various organs. The actions of THs are mainly mediated by the nuclear TH receptor (TR), which binds to a specific DNA sequence called the TH response element (TRE), as a homodimer or as a TR-retinoid X receptor (RXR) heterodimer with the RXR. In the absence of a ligand, corepressor complexes, such as nuclear receptor (NR) corepressor (N-CoR) and silencing mediator for retinoid and thyroid receptors (SMRT) bind to TR-RXR heterodimers (Cheng et al., 2010; Yen, 2001; Brent, 2012). In the presence of T3, the corepressors are replaced by coactivator complexes including the steroid receptor coactivators (SRC)-1, SRC-2, and SRC-3, to activate transcription (Koibuchi and Chin, 2000; Koibuchi, 2008, 2013; Wong et al., 2014). Isoflavones are a natural class of isoflavonoids. They are produced exclusively by the legume family (Leuner et al., 2013). They exert various effects at molecular, cellular, and organ levels (Węgrzyn et al., 2010). Among the isoflavones, soybean isoflavones such as genistein and daidzein have been intensively studied. They inhibit proliferation and induce apoptosis in prostate cancer cell lines (LNCaP and PCa cell lines) (Dong et al., 2013), enhance the retention of bone calcium, down-regulate energy metabolism in human adipose tissues in postmenopausal women (Pawlowski et al., 2015; Velpen et al., 2014), and improve peroxisome proliferator-activated receptor (PPAR)α-mediated fatty acid oxidation in mouse liver (Qiu and Chen, 2015). Such a wide variety of actions indicate that these compounds probably act through several different signaling pathways. Both genistein and daidzein are well-known phytoestrogens that modulate the action of NRs including the estrogen receptor (ER) by binding to the ligand-binding domain (LBD) of ERα and ERβ (Leclercq and Jacquot, 2014). Soybean isoflavone, especially genistein, showed higher binding affinity to ERβ compare with ERα by radioligand solid-phase-binding assay and transactivation assays on human embryonal kidney 293 cells (Kuiper et al., 1998). Genistein at physiological range (0.5–10 μmol/l) also have been reported induces the reduction of ERβ promoter methylation with corresponding increases in ERβ expression and induces phosphorylation of ERβ (pS105 and pS87), nuclear translocation, and ERβ transcriptional activity in prostate cancer cell lines (LNCaP, LAPC-4, and PC-3 cells) (Mahmoud et al., 2013). In addition, genistein partially antagonizes the activity of the androgen receptor (AR) in a tissue-specific and AR target gene-specific manner in male mice (effective in prostate, testes, and brain, but not in skeletal muscles and lung) (Pihlajamaa et al., 2011). It may also act as a weak agonist in the brain and prostate tissues of mice (Pihlajamaa et al., 2011). Genistein also upregulates pregnane X receptor (PXR)-mediated transcription and PXR-mediated CYP3A4 mRNA expression in mouse hepatocytes (Li et al., 2009). Genistein and daidzein also modulate the protein and mRNA levels of NRs; for example, genistein increases the mRNA levels of ERβ in the rat hypothalamus (Patisaul et al., 2002), decreases the mRNA levels of PPARγ during adipogenesis in human primary bone marrow stromal cells (Heim et al., 2004), and decreases AR protein levels in LNCaP cell lines (Basak et al., 2008). These findings indicate that genistein and daidzein may exert their actions partly by modulating the action and/or expression of NRs. The effect of dietary soybean isoflavones on the TH system is rather controversial. Although the goitrogenic effects of soy formula in infants have been well documented, the effects can be reversed by switching the formula to cow milk (Chorazy, 1995). Although dietary genistein treatment in young adult rat models significantly suppressed the levels of thyroid peroxidase (TPO), the levels of T3, T4, and TH-stimulating hormone (TSH) were not altered (Chang and Doerge, 2000). In orchidectomized middle-aged rats, dietary genistein and daidzein weakly suppressed the levels of T3 and T4, and increased both the TSH level and the cellular volume of TSH cells in the pituitary, while decreasing the volume of colloid in the thyroid gland (Šošić-Jurjević, 2010) and suppressing the mRNA levels of thyroglobulin and TPO, which indicate the impairment of TH synthesis. On the other hand, the mRNA levels of spot 14 and type 1 iodothyronine deodinase, which are positively regulated by TH, increased in the liver of the same experimental animal (Šošić-Jurjević et al., 2014). These results indicate that soybean isoflavones may have bimodal effects on the TH system. Although the effect of soybean isoflavones on TR-mediated transcription has not yet been extensively studied, a weak induction of TRα-mediated transcription by genistein using a conventional reporter assay system has been reported (Hofmann et al., 2009). However, whether these isoflavones bind to TRs have not been studied. In addition to the possibility that genistein and daidzein act directly on the TRs, these isoflavones may alter TR action through ERs. Crosstalk between NRs is important for the conversion of external and internal stimuli, which is necessary for eliciting cellular physiological responses (Vasudevan et al., 2001). The crosstalk and cross-interference between TRs and ERs have been reported in several gene promoters and cell lines (Vasudevan et al., 2001 2008), and is one of the reasons for the binding of TRs and ERs to the common DNA half-site sequence, 5’AGGTCA 3’. These findings indicate that genistein and daidzein may alter the interaction between ERs and TRs, or alter the interaction between TRs and their cofactors due to cofactor squelching by the ERs. However, their effects on the interaction between TRs and ERs are not clear. In this study, to clarify the effects of the isoflavones on TR activity, we employed reporter gene assays using TRα and TRβ expression vectors together with their mutants, in addition to in silico molecular docking. Since a recent study has identified a second TH-binding site located among H9, H10, and H11 of the TR-LBD (Souza et al., 2014), we also examined the possible binding of the isoflavones to this second binding site. Moreover, we also examined the crosstalk of TRs with ERs by using knock-down and overexpression strategies. MATERIALS AND METHODS Chemicals T3, genistein [4’, 5, 7-trihydoxyisoflavone or 5, 7-dihydroxy-3-(4-hydoxyphenyl)-4H-1-benzopyran-4-one] (Figure 1A), and daidzein [4′, 7-dihydroxyisoflavone or 7-hydroxy-3-(4-hydroxyphenyl) chromen-4-one] (Figure 1B) were purchased from Sigma (St Louis, Missouri). The purity of all chemicals was above 98%. Figure 1. View largeDownload slide Genistein and daidzein augmented TR-mediated transcription. A and B, Molecular structures of genistein and daidzein. C and D, Expression plasmids encoding 4 ng TRβ1 (C) and TRα1 (D) were co-transfected with 40 ng DR4-TK-LUC into CV-1 cells. The cells were cultured in the absence or presence of 10−7 T3 along with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05 indicates statistical significance by Bonferroni’s test compared with TRα1 or TRβ1 (+), T3 (+), and genistein or daidzein (–). Figure 1. View largeDownload slide Genistein and daidzein augmented TR-mediated transcription. A and B, Molecular structures of genistein and daidzein. C and D, Expression plasmids encoding 4 ng TRβ1 (C) and TRα1 (D) were co-transfected with 40 ng DR4-TK-LUC into CV-1 cells. The cells were cultured in the absence or presence of 10−7 T3 along with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05 indicates statistical significance by Bonferroni’s test compared with TRα1 or TRβ1 (+), T3 (+), and genistein or daidzein (–). Plasmids The TRα1 and TRβ1 expression vectors, luciferase (LUC) reporter constructs, the artificial direct repeat TRE, DR4-TK-LUC (DR4-TRE), and chick lysozyme (F2)-thymidine kinase (TK)-LUC (F2-TRE) have been described in previous studies (Iwasaki et al., 2001; Koibuchi et al., 1999). The LBD of VP16-TRβ1- has been previously described in Miyazaki et al. (2004). Gal4-blank, the expression vector for the Gal4-DNA-binding domain (DBD)-fused SRC-1-NR-binding domain (NBD)-1 (amino acids 595-780), Gal4-N-CoR-nuclear receptor-interacting domain (RID) (amino acids 1579–2454) and Gal4-SMRT- RID (amino acids 1669–2507) have been described in a previous study (Takeshita et al., 2002). The mutated TR plasmid ΔN.hTRβ1 that has an AF-1-binding domain with a truncated N-terminal; E457A, a full-length hTRβ1 containing a point mutation (glutamate to alanine) in codon 457 at the AF-2 domain; and ΔN.E457A, that has both the truncated N-terminal and the E457A mutation have been described in a previous study (Iwasaki et al., 2006). Clonal cell culture Monkey kidney fibroblast-derived clonal cells, CV-1, were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and antibiotics (100 U/ml penicillin, and 100 µg/ml streptomycin), at 37°C with 5% CO2. The serum was stripped of hormones by constantly mixing with 5% (w/v) AGX1-8 resin (Bio-Rad, Hercules, California) and powdered charcoal, prior to ultrafiltration (Iwasaki et al., 2002). Transient transfection-based reporter gene assay Cells were plated at a density of 1 × 104/0.1 ml in 96-well plates and incubated for 24 h, followed by transfection of the expression vectors and a reporter plasmid using HilyMax transfection reagents (Dojindo Molecular Technologies, Inc.), according to the protocol described in the technical manual. Expression vectors encoding TRα1 or TRβ1 (4 ng) were cotransfected with the reporter plasmid (F2-TRE-LUC) (40 ng) into CV-1 cells. The internal control was a cytomegalovirus-β-galactosidase plasmid (4 ng). The cells were incubated after 16–18 h of transfection with fresh medium containing indicated concentrations of the ligand (10−7 M T3) and either of genistein or daidzein for 24 h. The cells were then harvested to measure LUC activity as described in a previous study (Iwasaki et al., 2002). The total amounts of DNA per well were balanced by adding pcDNA3 plasmids (Invitrogen, San Diego, California). The LUC activity was normalized to β-galactosidase activity and represented as relative LUC activity. All transfection experiments were carried out in triplicate. The data are represented as the mean ± SEM of one representative experiment performed in triplicate. RNA interference and cell transfection Short interfering RNAs (siRNAs) for ERα (ESR1) and control RNAs were purchased from Thermo Fisher Scientific (Massachusetts). The siRNAs were transfected into CV-1 cells using lipofectamine RNAiMAX reagent (Thermo Fisher), according to the manufacturer’s protocol. Briefly, the siRNA-lipid complexes (25 nM control siRNA (scrambled RNA) or 25 nM ESR1 siRNA) were added to 96-well plates, and incubated for 20 min. CV-1 cells at a density of 1 × 104 cells/well were seeded in 96-well plates using media free from phenol red and antibiotics. After 16–24 h, the cells were subjected to analyses with reporter gene assays. The efficacy of the siRNA knockdown of ERα was verified by quantitative real-time PCR (qRT-PCR). RNA isolation and quantitative real-time RT-PCR The total RNA was isolated using QIAzol Lysis reagent (QIAGEN) and reverse transcribed using ReverTra Ace qPCR RT master mix (TOYOBO Bio-Technology, Japan) based on the instruction manual provided by the supplier. RT-PCR was performed using THUNDERBIRD SYBR qPCR mix (TOYOBO Bio-Technology, Japan) as described in the instruction manual and the StepOne RT-PCR System (Applied Biosystems). The list of primers used in this study is listed in Table 1. The RT-PCR protocol for all genes involved denaturation at 95°C for 20 s, followed by amplification at 95°C for 3 s and at 60°C for 30 s (40 cycles). All experiments were repeated 3 times, using independent RNA preparations to confirm the consistency of the results. The mRNA levels were normalized by the mRNA level of GAPDH. Table 1. PCR Primers Used in the Study     Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA      Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA  Table 1. PCR Primers Used in the Study     Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA      Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA  In silico ligand-receptor-binding calculations All in silico calculations were performed using an Asus N43s notebook with an Intel Core i3 -2310 M dual-core processor, 3 M cache, 2.10 GHz CPU, 1333 MHz DDR3 SDRAM, and 8 GHz RAM, running on a Windows 10 professional operating system. The molecular structure of genistein (PubChem CID 5280961), daidzein (PubChem CID 5281708), and T3 (PubChem CID 5920) were downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/; last accessed May 17, 2017) in the structure data file format. The crystal structures of the TR-LBD were downloaded from the Protein Data Bank (PDB) (http://www.rcsb.org/pdb/home/home.do; last accessed May 17, 2017) in PDB format (PDB IDs: 4LNX, 4LNW, 3JZB, and 3HZF). The 3D structure files of the TR-LBDs and ligands were opened and modified with Discovery Studio structure-based design software, version 4.0 (BIOVIA/Accelrys Inc., San Diego, California). The water molecules and other substructures (bound molecules or ligand molecules) were removed from the coordinate file before docking. The unliganded TR-LBD was used for the individual dockings of T3, genistein, and daidzein. Polar hydrogen atoms were added to the 3D structure of the TR-LBD and the input file was generated in the PDBQT format, which contained the structure of the TR-LBD, using AutoDockTools of MGLTools (http://autodock.scripps.edu/resources/adt; last accessed May 17, 2017). The coordinates for docking were determined through a grid box using the PyRx - Python Prescription 0.8 Virtual Screening software for Computer-Aided Drug Design (http://pyrx.sourceforge.net/; last accessed May 17, 2017), using AutoDock 4 and AutoDock Vina as docking software (Trott and Olson, 2010). A blind docking strategy was utilized in order to include the entire possible binding site for ligands. For more reliable results, refinement docking experiments with repetitions of 30 runs were performed with complexes which had high affinity scores (lower than −9 kcal/mol). LigPlot+ v.1.4 (http://www.ebi.ac.uk/thornton-srv/software/LigPlus/; last accessed May 17, 2017) was used to determine the interactions between the TR-LBDs and the ligands in complexes with the best affinity scores. The binding affinity was expressed as the binding free energy (kcal/mol). Statistical analysis All the data are expressed as the mean ± SEM of 3 individual experiments performed in triplicate and analyzed using ANOVA. Post hoc comparisons were made using Bonferroni’s test. A p value < .05 was considered to be significant. RESULTS Genistein and Daidzein Augmented TR-Mediated Transcription by THs in CV-1 Cells To investigate the effects of genistein and daidzein on TR-mediated transcription, we first performed transient transfection-based reporter gene assays in CV-1 cells. Both genistein and daidzein augmented the transcription mediated by TRβ1 and TRα1 with 10−7 M T3 through DR4-TRE in a concentration-dependent manner (Figs. 1C and 1D). At 10−5 M concentration, genistein augmented TRβ1-mediated transcription by 3-fold, and activated TRα1-mediated transcription by 2-fold compared with the group that received only T3. Daidzein at a concentration of 10−5 M, also upregulated the transcription mediated by TRβ1 and TRα1 by 2-fold, respectively. Genistein and daidzein also augmented TRβ1- and TRα1-mediated transcription through F2-TRE (Supplementary Figs. 1A and 1B). On the other hand, although a significant increase in transcription was observed at higher doses of the isoflavones in the absence of T3, (TRβ1, 10−6 and 10−5 M genistein or 10−5 M daidzein; TRα, 10−5 M genistein or daidzein), the increase was rather weak compared with the transcription in the presence of T3 (Figs. 1C and 1D;Supplementary Figs. 1A and 1B). Genistein and Daidzein Augmented the Interaction of TRs with SRC-1 and N-CoR We hypothesized that the increase in transcription induced by the isoflavones may be caused due to the increased recruitment of coactivators, such as SRC-1 or dissociation corepressors such as N-CoR, to TRs. To examine the binding of such cofactors to TRs in the presence of the isoflavones, we carried out mammalian 2-hybrid assays. The transcription by VP16-TRβ1-LBD and Gal4-SRC-1-NBD (Figure 2A) through 5xUAS was activated by 10−7 M T3 (Figure 2B, lane 4). This activation was further augmented by genistein and daidzein in a concentration-dependent manner (Figure 2B, lanes 6–8 and 11–12). On the other hand, in the absence of 10−7 M T3, only a high concentration of genistein and daidzein (10−5 M) could activate the transcription by VP16-TRβ1-LBD and Gal4-N-CoR (Figure 2C) through 5xUAS (Figure 2D, lanes 8 and 12). We also observed the activation of transcription by VP16-TRβ1-LBD and Gal4-SMRT (Supplementary Figure 2A) in the absence of 10−7 M T3 (Supplementary Figure 2B, lanes 8 and 12). These results indicate that an increase in the recruitment of coactivators to TRs by the isoflavones might be the reason underlying the TR-mediated transcriptional augmentation. Figure 2. View largeDownload slide The effects of genistein and daidzein on the interaction of TRs with cofactors. Schematic diagrams of Gal4-SRC-1-NBD-1 (A) and Gal4-N-CoR-RID (C). Expression plasmids encoding Gal4-DBD-fused SRC-1-NBD-1 (B) (4 ng), or Gal4-N-CoR- RID (D) (4 ng) were co-transfected with VP16-constructs (4 ng) and 5× UAS-TK-LUC-reporter plasmids (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05, indicates statistical significance by Bonferroni’s test compared with Gal4-SRC-1 (+) or Gal4-N-CoR (+), VP16-TRβ1 (+), and T3 (10−7 M). Figure 2. View largeDownload slide The effects of genistein and daidzein on the interaction of TRs with cofactors. Schematic diagrams of Gal4-SRC-1-NBD-1 (A) and Gal4-N-CoR-RID (C). Expression plasmids encoding Gal4-DBD-fused SRC-1-NBD-1 (B) (4 ng), or Gal4-N-CoR- RID (D) (4 ng) were co-transfected with VP16-constructs (4 ng) and 5× UAS-TK-LUC-reporter plasmids (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05, indicates statistical significance by Bonferroni’s test compared with Gal4-SRC-1 (+) or Gal4-N-CoR (+), VP16-TRβ1 (+), and T3 (10−7 M). The AF-2 Domain of TRs Is Essential for the Augmentation of TR-Mediated Transcription by the Isoflavones We further performed transient transfection-based reporter gene assays in CV-1 cells using a series of truncation and/or point mutants of TRβ1 to identify the domain of TR responsible for the transcriptional augmentation by isoflavones. The transcription through ΔN.hTRβ1, which is an N-terminus truncated mutant, was augmented by genistein and daidzein, and the transcription levels were similar to the wild type TRβ1 (Figure 3). On the other hand, the transcriptions through the hTRβ1-AF2 (E457A) and ΔN.hTRβ1-AF2 mutants were not augmented by the isoflavones (Figure 3). These results indicate that the AF-2 domain is responsible for the transcriptional augmentation by genistein and daidzein. Figure 3. View largeDownload slide Genistein and daidzein augmented the transcription mediated by ΔN.TRβ1 but not in AF-2 mutants. A and B, Expression plasmids encoding TRβ1 (4 ng) or ΔN.hTRβ1(4 ng) or E457A (4 ng) or ΔN.E457A (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein (A) or daidzein (B). The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test. Figure 3. View largeDownload slide Genistein and daidzein augmented the transcription mediated by ΔN.TRβ1 but not in AF-2 mutants. A and B, Expression plasmids encoding TRβ1 (4 ng) or ΔN.hTRβ1(4 ng) or E457A (4 ng) or ΔN.E457A (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein (A) or daidzein (B). The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test. The Augmentation of TR-Mediated Transcription by Isoflavones Is Not Altered by ERα Overexpression or Knockdown We measured the mRNA expression levels of TRs and ERs in CV-1 cells prior to the modulations in ER expression. The expression of TRs and ESR1 (ERα) was confirmed, whereas ESR2 (ERβ) expression could not be detected (Supplementary Figure 3). Based on the expression levels of ERs in CV-1 cells, we cotransfected ERα to CV-1 cells to further examine the involvement of ER, especially the effects of ERα overexpression. The overexpression of ERα did not further augment the transcription mediated by TRβ (Figure 4A) and TRα (Supplementary Figure 4A) in the presence of the isoflavones. These results indicate that ERα may not be involved in isoflavone-induced augmentation of TR-mediated transcription. To further examine whether the effects of genistein and daidzein were indeed induced through ERα, the ERα mRNA was knocked down using RNA interference. The CV-1 cells were transiently transfected with ERα siRNA. The decrease in the expression of ERα mRNA was confirmed by qRT-PCR. The siRNA treatment induced a 98.6% reduction in ERα mRNA levels compared with the control (Supplementary Figure 4B). After knocking down the ERα mRNA in CV-1 cells, we performed reporter gene assays. The magnitude of TRβ transcription induced by 10−5 M genistein or daidzein (Figure 4B) and TRα-mediated transcription in DR4-TRE (Supplementary Figure 4C) did not decrease after ERα knockdown. These results indicated that both genistein and daidzein might directly bind to TRs and augment TR-mediated transcription, although there might be differences in genistein and daidzein action on the interaction of TRs. Taken together with the study of ERα overexpression, the results of this study indicate that ERα is not involved in isoflavone-augmented TR transcription. Figure 4. View largeDownload slide The augmentation of TR-mediated transcription by genistein and daidzein was not altered after ERα overexpression or knockdown. A,. Expression plasmids encoding TRβ1 (4 ng) and ERα (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) and genistein (10−5 M) or daidzein (10−5 M). B, 25 nM scrambled RNA or 25 nM ESR1 mRNA was transfected into the CV-1 cells; 16–24 h after transfection, expression plasmids encoding TRβ1 (4 ng) were cotransfected with DR4-TK-LUC (40 ng). The cells were cultured in the absence or presence of 10−7 M T3 together with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), and genistein or daidzein (−). ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), scRNA (+) and the TRβ1 group (+), T3 (+), scrambled siRNA (+) or genistein or daidzein (−). ###p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), ESR1 siRNA (+) and genistein or daidzein (−). Figure 4. View largeDownload slide The augmentation of TR-mediated transcription by genistein and daidzein was not altered after ERα overexpression or knockdown. A,. Expression plasmids encoding TRβ1 (4 ng) and ERα (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) and genistein (10−5 M) or daidzein (10−5 M). B, 25 nM scrambled RNA or 25 nM ESR1 mRNA was transfected into the CV-1 cells; 16–24 h after transfection, expression plasmids encoding TRβ1 (4 ng) were cotransfected with DR4-TK-LUC (40 ng). The cells were cultured in the absence or presence of 10−7 M T3 together with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), and genistein or daidzein (−). ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), scRNA (+) and the TRβ1 group (+), T3 (+), scrambled siRNA (+) or genistein or daidzein (−). ###p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), ESR1 siRNA (+) and genistein or daidzein (−). Genistein and Daidzein Can Directly Bind to the First and Second Binding Pockets of TR-LBD To investigate the binding modes of genistein and daidzein to the TR-LBD, we generated in silico binding models by molecular docking using AutoDocks Vina (Trott and Olson, 2010). We used several TRα-LBD crystal structures, including 4LNX (crystal structure of TRα bound to T4 in the second binding site), 4LNW (crystal structure of TRα bound to T3 in the second binding site), 3JZB (crystal structure of TRα bound to the selective thyromimetic TRIAC (3,3′,5-Triiodothyroacetic acid)), and 3HZF (crystal structure of TRα bou[nd to selective thyromimetic GC-1 in C2 space group). We used the TRα-LBD crystal structure with PDB ID 4LNX and performed molecular docking study with T3 (Figure 5A). We then performed the molecular docking study with genistein and daidzein. Genistein and daidzein could directly bind to the binding pockets of the TRα-LBD (Figs. 5B and 5C and Supplementary Figs. 5B and 5C) with a binding affinity −9.9 and −9.8 kcal/mol, respectively. The molecular docking studies of genistein and daidzein showed that the same amino acid residues in the crystal structure that interacted with T3, also interacted with genistein and daidzein, which included residues ile 222, ala 225, ser 277, leu 276, and leu 292 of the first binding site (Figure 5D) and arg 375 of the second binding site (Figure 5E). In the model obtained by molecular docking, genistein was located at 2.76 Å from ser 277 and 2.90 Å from phe 218 in the first binding site, and had the possibility to form hydrogen bonds. In the second binding site, genistein was located at 3.32 Å from arg 375, 2.49 Å from gln 342, and 2.52 Å from thr 327. The residues shown in red circles in Figures 5D and 5E represent the amino acid residues of TR-LBD that are common to T3 and genistein binding. We also performed in silico calculations for daidzein (Supplementary Figure 5) and determined the binding affinity of T3, genistein, and daidzein with other crystal structures of the TRα-LBD (PDB IDs 4LNW, 3JZB, and 3HZF) in both the binding pockets (Table 2). Table 2. Results of the Refinement Docking Experiments With AutoDock Vina   Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5    Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5  Table 2. Results of the Refinement Docking Experiments With AutoDock Vina   Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5    Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5  Figure 5. View largeDownload slide Genistein can bind to the first or second binding site of TR. The 3D structure of the TRα-LBD (PDB ID: 4LNX) with T3 bound to both binding sites (A), T3 with genistein at the second binding site (B), and genistein with T3 at the second binding site (C). The interaction plots between the ligand and the TRα-LBD (PDB ID: 4LNX) were generated by LigPlot+ v.1.4, with T3 and genistein at the first binding site (D) and the second binding site (E), with each subsequent plot being automatically fitted. The red circles and ellipses in each plot indicate protein residues that have equivalent 3D positions with respect to the residues in the first plot. Hydrogen bonds are shown as green dotted lines, while the spooked arcs represent residues making nonbonded contacts with the ligand. Figure 5. View largeDownload slide Genistein can bind to the first or second binding site of TR. The 3D structure of the TRα-LBD (PDB ID: 4LNX) with T3 bound to both binding sites (A), T3 with genistein at the second binding site (B), and genistein with T3 at the second binding site (C). The interaction plots between the ligand and the TRα-LBD (PDB ID: 4LNX) were generated by LigPlot+ v.1.4, with T3 and genistein at the first binding site (D) and the second binding site (E), with each subsequent plot being automatically fitted. The red circles and ellipses in each plot indicate protein residues that have equivalent 3D positions with respect to the residues in the first plot. Hydrogen bonds are shown as green dotted lines, while the spooked arcs represent residues making nonbonded contacts with the ligand. DISCUSSION In this study, we examined the effect of soybean isoflavones (genistein and daidzein) on TR-mediated transcription. We found that these isoflavones augmented the transcription mediated by liganded TRs in a concentration-dependent manner. ERs may not be involved in such augmentation as shown by the ER overexpression and siRNA studies. Using in silico analyses, we also found that these isoflavones may directly interact with the TR-LBD. These results describe a novel mechanism of action of genistein and daidzein on the augmentation of TR-mediated transcription by the direct binding of the isoflavones to TRs and TH-mediated signal transduction. Soybean isoflavones activate various NR-mediated transcriptions including those of ERs (Leclercq and Jacquot, 2014; Patisaul et al., 2002), and PXR (Li et al., 2009). Dietary soy protein isolates have been reported to upregulate hepatic TRβ1 expression levels in rats (Xiao et al., 2004). LUC reporter gene assay with ERs in human hepatoma cells (HepG2 cells) showed that both genistein and daidzein were complete agonist at both ER, genistein being more potent than daidzein. In addition, both genistein and daidzein were more potent to ERβ compare with ERα (Casanova et al., 1999). However, their effects on TR-mediated transcription have not been extensively studied. This study clearly shows that both genistein and daidzein augmented the transcription mediated by liganded TRs in a concentration-dependent manner. In contrast, in the absence of T3, only high doses of genistein and daidzein could weakly augment the transcription mediated by TRβ and TRα. This weak activation appears to be in agreement with a previous study which reported a weak transactivation of TR activity by genistein in the absence of T3 (Hofmann et al., 2009). In that study, genistein at a concentration of 10−6 M induced a 2.3-folds increase in the level of transcription in comparison to the basal level (without T3). Although the 2.3-fold increase with respect to the basal level seemed significant, the level of transcription was still rather low, because the basal transcription level in the absence of a ligand is repressed by interactions with corepressors. In this study, the TR-mediated transcription through F2-TRE with 10−7 M T3 was further augmented to 3-fold by genistein and daidzein at concentrations of 10−5 M. Thus, the effect of isoflavones on the expression of TRs may be greater in the presence of T3. The next question that may arise would concern the effect of isoflavones on modulating TH action in vivo. In this regard, additional studies using animal models are necessary. The plasma levels of genistein and daidzein in prostate cancer patients reached 1–10 µM, 2–8 h after the intake of 2 slices of standard soy or soy-almond bread (Ahn-jarvis et al., 2015), and reached 1–10 µM at 4–7 h after the intake of 50 g of kinako (roasted soybean powder) in human subjects (Hosoda et al., 2008). In this study, TR-mediated transcription was augmented by genistein and daidzein at doses comparable to those in the aforementioned studies (Figs. 1C and 1D). Previously, we had performed the MTS cell proliferation assay and showed that both genistein and daidzein concentrations that were used in this study did not affect the viability of CV-1 cells (Supplementary Figure 6). Unless large amounts are administered, their adverse effects cannot be reported. Since soybeans have been consumed throughout human evolution, isoflavones can be effectively metabolized in humans. The half-lives of plasma genistein and daidzein are 8.36 and 5.79 h, respectively (Watanabe et al., 1998). Thus, the endocrine-modulating action of the soy isoflavones under normal physiological conditions may be beneficial to our health, although further studies are required to confirm this, especially at acute and massive doses. Such a possibility indicates that isoflavones can be used as a supplement for patients with low T3 syndromes or to relieve hypothyroidism-induced clinical symptoms. The interaction of TRs with the cofactor SRC-1 induced by T3, was augmented by genistein and daidzein. However, only high doses of genistein and daidzein could augment the interactions with TR-N-CoR and TR-SMRT. Studies have also reported the augmentations in the interactions of SRC-1 with other NRs by genistein and daidzein. According to previous studies, genistein augmented the interaction of SRC-1 with ERα (Schwartz et al., 1998), and ERβ (Mueller et al., 2004). Additionally, modulations in N-CoR action by genistein have been reported. Genistein upregulated the expression of N-CoR protein and reversed promyelocytic leukemia (PML)-retinoic acid receptor (RAR)-induced misfolding of N-CoR protein possibly by inhibiting the selective phosphorylation-dependent binding of N-CoR to PML-RAR in acute PML cells (Ng et al., 2007). In this study, we showed that both genistein and daidzein augmented the interactions of a TR-coactivator (TR-SRC-1) and TR-corepressors (TR-N-CoR and TR-SMRT) at high doses (Figure 2). Moreover, using a series of truncated and/or point mutants of TRβ1, we found that the TR-LBD region and especially its AF-2 domain, is responsible for the augmentation in TR-mediated transcription induced by genistein and daidzein (Figure 3). The augmentation was observed with the N-terminus-truncated mutant, but not with AF-2 mutants. These results indicate that the TR-mediated transcription augmented by genistein and daidzein might have altered the conformation of AF-2, which modulated the binding of TRs to cofactors. Together with the results of mammalian 2 hybrid assays, it is likely that the isoflavone-induced increase in the interactions of SRC-1 to the AF-2 domain modified the chromatin structure which augmented transcription. On the other hand, although these isoflavones also augmented the recruitment of corepressors, transcription was not further repressed. The reason for this discrepancy cannot be clarified. Further studies are necessary for clarifying the mechanisms of increase in coactivator/corepressor recruitment by isoflavones. Soybean isoflavones are well-known phytoestrogens that act through ERs (Mueller et al., 2004; Patisaul et al., 2002). Since a previous study reported the possibility of crosstalk between TRs and ERs (Vasudevan et al., 2001), we speculated whether the effect of isoflavones on TRs is a result of the indirect binding of isoflavones to ERs. To further investigate the crosstalk between ERs and TRs, we separately overexpressed and knocked-down ERα, and observed that both genistein and daidzein augmented TR-mediated transcription in both the states of ERα overexpression and knockdown. These results indicate that ERα is not involved in the augmentations of both TR-mediated transcription and TR-cofactor interactions, induced by genistein and daidzein. We also confirmed that the transactivation mediated by TRs in the presence of T3 was increased by ERα knockdown (Figure 4B and Supplementary Figure 4C). However, genistein and daidzein might acts through both of ERs and TRs in the in vivo study. Transcriptomic analysis showed that genistein regulated some ER-responsive genes in the cell proliferation, adhesion, motility and inflammatory response proses (Gong et al., 2014). Some of these genes were also included in TR responsive gene. The increase of body weight in ovariectomized rats was reduced by co-administration of genistein and ERβ-specific agonist 8β-VE2, and suggesting that the activation of ERβ may modulate ERα-mediated physiological effects in vivo (Hertrampf et al., 2009). A previous study showed that there is a cross-interference (squelching) between TRs and ERs with regard to the binding of coactivators, including SRC-1 (Lopez et al., 1999). In fact, we confirmed the expression of SRC-1 mRNA in CV-1 cells (Supplementary Figure 3). These reasons explain the increased transactivation induced by ER knockdown. Nevertheless, the action of isoflavones on TRs may be directly induced and exerted through the TH signal transduction pathway since THs regulate various metabolic pathways. In this study, we also showed a possibility using in silico docking studies that these isoflavones may directly bind to the TR-LBD. Like other NRs, the TR LBDs has a core-binding site for T3 where the ligand is buried. The binding of THs change the conformation of TRs (Souza et al., 2014; Moras et al., 2015). The binding of the isoflavones to the TR-LBD may also induce conformational changes in the structure of TRs by inducing the recruitment of cofactors. In addition, a recent study identified a new hormone-binding site referred to as the second binding site, on the surface of the TR-LBD (Souza et al., 2014). We have shown that isoflavones can also bind to this second binding site, although the affinity is lower than the affinity with the first binding site. Importantly, the augmentations induced by the isoflavones were mainly observed in the presence of T3, indicating that both T3 and the isoflavones should be simultaneously bound in the TR complex to cause augmentations. Although the exact binding mode of T3 and the isoflavones could not be verified in this study, several possibilities have been considered, which need to be investigated in order to identify the roles and interaction between the first and second binding sites in TR-mediated transactivation. Moreover, we also need to examine whether the binding mode of TRs to the TRE may be changed by isoflavone-liganded TRs, because TRs can form homodimers and TR-RXR heterodimers for binding to TRE. In summary, isoflavones such as genistein and daidzein augment TR-mediated transcription at least in part by increasing the recruitment of SRC-1. Such augmentation may be caused by the direct binding of genistein and daidzein to the TR-LBD. However, further studies are required to clarify the mechanism; the results of this study provide a new insight into the understanding of the actions of genistein and daidzein on TRs. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. FUNDING This work was supported in part by Grants-in-Aid for Scientific Research (Nos. 26340036 and 25281024) to T.I and N.K from the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT). ACKNOWLEDGMENTS We thank Dr Yusuke Takatsuru, Dr Izuki Amano, and Dr Asahi Haijima of the Department of Integrative Physiology, for technical assistance and advice. REFERENCES Ahn-Jarvis J. H., Clinton S. K., Grainger E. M., Riedl K. M., Schwartz S. J., Lee M.-L. T., Cruz-Cano R., Young G. S., Lesinski G. B., Vodovotz Y. ( 2015). Isoflavone pharmacokinetics and metabolism after consumption of a standardized soy and soy-almond bread in men with asymptomatic prostate cancer. Cancer Prevent. Res . 8, 1045– 1054. Google Scholar CrossRef Search ADS   Basak S., Pookot D., Noonan E.J., Dahiya R. ( 2008) Genistein down-regulates androgen receptor by modulating HDAC6-Hsp90 chaperone function. Mol. Cancer Ther ., 7, 3195– 3202. Google Scholar CrossRef Search ADS PubMed  Brent G. A. ( 2012). Mechanisms of thyroid hormone action. Sci. Med . 122, 3035– 3043. Casanova M., You L., Gaido K. W., Archibeque-Engle S., Janszen D. B., Heck H. D. A. ( 1999). Developmental effects of dietary phytoestrogens in Sprague-Dawley rats and interactions of genistein and daidzein with rat estrogen receptors α and β in vitro. Toxicol. Sci . 51, 236– 244. Google Scholar CrossRef Search ADS PubMed  Chang H.C., Doerge D.R. ( 2000) Dietary genistein inactivates rat thyroid peroxidase in vivo without an apparent hypothyroid effect. Toxicol. Appl. Pharmacol ., 168, 244– 252. Google Scholar CrossRef Search ADS PubMed  Cheng S. Y., Leonard J. L., Davis P. J. ( 2010). Molecular aspects of thyroid hormone actions. Endocr. Rev . 31, 139– 170. Google Scholar CrossRef Search ADS PubMed  Chorazy P.A., Himelhoch S., Hopwood N.J., Greger N.G., Postellon D.C. ( 1995) Persistent Hypothyroidism in an Infant Receiving a Soy Formula: Case Report and Review of the Literature. Pediatrics , 96, 148– 150. Google Scholar PubMed  Dong X., Xu W., Sikes R. A., Wu C. ( 2013). Combination of low dose of genistein and daidzein has synergistic preventive effects on isogenic human prostate cancer cells when compared with individual soy isoflavone. Food Chem . 141, 1923– 1933. Google Scholar CrossRef Search ADS PubMed  Gong P., Madak-erdogan Z., Li J., Cheng J., Greenlief C. M., Helferich W., Katzenellenbogen J. A., Katzenellenbogen B. S. ( 2014). Transcriptomic analysis identifies gene networks regulated by estrogen receptor α (ERα) and ERβ that control distinct effects of different botanical estrogens. Nucl. Recept. Signal  12, 1– 13. Heim M., Frank O., Kampmann G., Sochocky N., Pennimpede T., Fuchs P., Hunziker W., Weber P., Martin I., Bendik I. ( 2004). The phytoestrogen genistein enhances osteogenesis and represses adipogenic differentiation of human primary bone marrow stromal cells. Endocrinology  145, 848– 859. Google Scholar CrossRef Search ADS PubMed  Hertrampf T., Seibel J., Laudenbach U., Fritzemeier K. H., Diel P. ( 2009). Analysis of the effects of oestrogen receptor α (ERα)- and ERβ-selective ligands given in combination to ovariectomized rats. Br. J. Pharmacol . 153, 1432– 1437. Google Scholar CrossRef Search ADS   Hofmann P. J., Schomburg L., Ko J. ( 2009). Interference of endocrine disrupters with thyroid hormone receptor-dependent transactivation. Toxicol. Sci . 110, 125– 137. Google Scholar CrossRef Search ADS PubMed  Hosoda K., Furuta T., Yokokawa A., Ogura K., Hiratsuka A., Ishii K. ( 2008). Plasma profiling of intact isoflavone metabolites by high-performance liquid chromatography and mass spectrometric identification of flavone glycosides daidzin and genistin in human plasma after administration of kinako. Drug Metab. Dispos . 36, 1485– 1495. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Chin W. W., Ko L. ( 2001). Identification and characterization of RRM-containing coactivator activator (CoAA) as TRBP-interacting protein, and its splice variant as a coactivator modulator (CoAM). J. Biol. Chem . 276, 33375– 33383. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Miyazaki W., Takeshita A., Kuroda Y., Koibuchi N. ( 2002). Polychlorinated biphenyls suppress thyroid hormone-induced transactivation. Biochem. Biophys. Res. Commun . 299, 384– 388. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Takeshita A., Miyazaki W., Chin W. W., Koibuchi N. ( 2006). The interaction of TRβ1-N terminus with steroid receptor coactivator-1 (SRC-1) serves a full transcriptional activation function of SRC-1. Endocrinology  147, 1452– 1457. Google Scholar CrossRef Search ADS PubMed  Koibuchi N. ( 2008). The role of thyroid hormone on cerebellar development. Cerebellum  7, 530– 533. Google Scholar CrossRef Search ADS PubMed  Koibuchi N. ( 2013). The role of thyroid hormone on functional organization in the cerebellum. Cerebellum  12, 304– 306. Google Scholar CrossRef Search ADS PubMed  Koibuchi N., Chin W. W. ( 2000). Thyroid hormone action and brain development. Trends Endocrinol. Metab . 11, 123– 128. Google Scholar CrossRef Search ADS PubMed  Koibuchi N., Liu Y., Fukuda H., Takeshita A., Yen P. M., Chin W. W. ( 1999). RORα augments thyroid hormone receptor-mediated transcriptional activation. Endocrinology  140, 1356– 1364. Google Scholar CrossRef Search ADS PubMed  Kuiper G. G., Lemmen J. G., Carlsson B., Corton J. C., Safe S. H., van der Saag P. T., van der Burg B., Gustafsson J. A. ( 1998). Interaction of estrogenic chemicals and pytoestrogens with estrogen receptor beta. Endocrinology  139, 4252– 4263. Google Scholar CrossRef Search ADS PubMed  Leclercq G., Jacquot Y. ( 2014). Interactions of isoflavones and other plant derived estrogens with estrogen receptors for prevention and treatment of breast cancer—Considerations concerning related efficacy and safety. J. Steroid Biochem. Mol. Biol . 139, 237– 244. Google Scholar CrossRef Search ADS PubMed  Leuner O., Havlik J., Hummelova J., Prokudina E., Novy P., Kokoska L. ( 2013). Distribution of isoflavones and coumestrol in neglected tropical and subtropical legumes. J. Sci. Food Agric . 93, 575– 579. Google Scholar CrossRef Search ADS PubMed  Li Y., Ross-Viola J. S., Shay N. F., Moore D. D., Ricketts M.-L. ( 2009). Human CYP3A4 and murine Cyp3A11 are regulated by equol and genistein via the pregnane X receptor in a species-specific manner. J. Nutr . 139, 898– 904. Google Scholar CrossRef Search ADS PubMed  Lopez G. N., Webb P., Shinsako J. H., Baxter J. D., Greene G. L., Kushner P. J. ( 1999). Titration by estrogen receptor activation function-2 of targets that are downstream from coactivators. Mol. Endocrinol . 13, 897– 909. Google Scholar CrossRef Search ADS PubMed  Mahmoud A. M., Zhu T., Parray A., Siddique H. R., Yang W., Saleem M., Bosland M. C. ( 2013). Differential effects of genistein on prostate cancer cells depend on mutational status of the androgen receptor. PLoS One  8, e78479– e78421. Google Scholar CrossRef Search ADS PubMed  Miyazaki W., Iwasaki T., Takeshita A., Kuroda Y., Koibuchi N. ( 2004). Polychlorinated biphenyls suppress thyroid hormone receptor-mediated transcription through a novel mechanism. J. Biol. Chem . 279, 18195– 18202. Google Scholar CrossRef Search ADS PubMed  Moras D., Billas I. M. L., Rochel N., Klaholz B. P. ( 2015). Structure-function relationships in nuclear receptors: The facts. Trends Biochem. Sci . 40, 287– 290. Google Scholar CrossRef Search ADS PubMed  Mueller S. O., Simon S., Chae K., Metzler M., Korach K. S. ( 2004). Phytoestrogens and their human metabolites show distinct agonistic and antagonistic properties on estrogen receptor α (ERα) and ERβ in human cells. Toxicol. Sci . 80, 14– 25. Google Scholar CrossRef Search ADS PubMed  Ng A. P. P., Nin D. S., Fong J. H., Venkataraman D., Chen C.-S., Khan M. ( 2007). Therapeutic targeting of nuclear receptor corepressor misfolding in acute promyelocytic leukemia cells with genistein. Mol. Cancer Ther . 6, 2240– 2248. Google Scholar CrossRef Search ADS PubMed  Patisaul H. B., Melby M., Whitten P. L., Young L. J. ( 2002). Genistein affects ERβ- but not ERα-dependent gene expression in the hypothalamus. Endocrinology  143, 2189– 2197. Google Scholar CrossRef Search ADS PubMed  Pawlowski J. W., Martin B. R., Mccabe G. P., Mccabe L., Jackson G. S., Peacock M., Barnes S., Weaver C. M. ( 2015). Impact of equol-producing capacity and soy-isoflavone profiles of supplements on bone calcium retention in postmenopausal women : A randomized crossover trial. Am. J. Clin. Nutr . 102, 695– 703. Google Scholar CrossRef Search ADS PubMed  Pihlajamaa P., Zhang F.P., Saarinen L., Mikkonen L., Hautaniemi S., Jänne O.A. ( 2011) The phytoestrogen genistein is a tissue-specific androgen receptor modulator. Endocrinology , 152, 4395– 4405. Google Scholar CrossRef Search ADS PubMed  Qiu L., Chen T. ( 2015). Novel insights into the mechanisms whereby isoflavones protect against fatty liver disease. World J. Gastroentrol . 21, 1099– 1107. Google Scholar CrossRef Search ADS   Schwartz J. A., Liu G., Brooks S. C. ( 1998). Genistein-mediated attenuation of tamoxifen-induced antagonism from estrogen receptor-regulated genes. Biochem. Biophys. Res. Commun . 253, 38– 43. Google Scholar CrossRef Search ADS PubMed  Šošić-Jurjević B., Filipović B., Ajdžanović V., Savin S., Nestorović N., Milošević V., Sekulić M. ( 2010) Suppressive effects of genistein and daidzein on pituitary–thyroid axis in orchidectomized middle-aged rats. Exp. Biol. Med ., 235, 590– 598. Google Scholar CrossRef Search ADS   Šošić-Jurjević B., Filipović B., Wirth E. K., Živanović J., Radulović N., Janković S., Milošević V., Köhrle J. ( 2014). Soy isoflavones interfere with thyroid hormone homeostasis in orchidectomized middle-aged rats. Toxicol. Appl. Pharmacol . 278, 124– 134. Google Scholar CrossRef Search ADS PubMed  Souza P. C. T., Puhl A. C., Martínez L., Aparício R., Nascimento A. S., Figueira A. C. M., Nguyen P., Webb P., Skaf M., Polikarpov I. ( 2014). Identification of a new hormone-binding site on the surface of thyroid hormone receptor. Mol. Endocrinol . 28, 534– 545. Google Scholar CrossRef Search ADS PubMed  Takeshita A., Taguchi M., Koibuchi N., Ozawa Y. ( 2002). Putative role of the orphan nuclear receptor SXR (steroid and xenobiotic receptor) in the mechanism of CYP3A4 inhibition by xenobiotics. J. Biol. Chem . 277, 32453– 32458. Google Scholar CrossRef Search ADS PubMed  Trott O., Olson A. ( 2010). AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading. J. Comput. Chem . 31, 455– 461. Google Scholar PubMed  Vasudevan N., Koibuchi N., Chin W. W., Pfaff D. W. ( 2001). Differential crosstalk between estrogen receptor (ER)α and ERβ and the thyroid hormone receptor isoforms results in flexible regulation of the consensus ERE. Mol. Brain Res . 95, 9– 17. Google Scholar CrossRef Search ADS   Vasudevan N., Zhu Y. S., Daniel S., Koibuchi N., Chin W. W., Pfaff D. ( 2008). Crosstalk between oestrogen receptors and thyroid hormone receptor isoforms results in differential regulation of the preproenkephalin gene. J. Neuroendocrinol . 13, 779– 790. Google Scholar CrossRef Search ADS   Velpen V., Geelen A., Hollman P. C. H., Schouten E. G., Veer P. V., Afman L. A. ( 2014). Isoflavone supplement composition and equol producer status affect gene expression in adipose tissue: A double-blind, randomized, placebo-controlled crossover trial in postmenopausal women. Am. J. Clin. Nutr . 100, 1269– 1277. Google Scholar CrossRef Search ADS PubMed  Watanabe S., Yamaguchi M., Sobue T., Takahashi T., Miura T., Arai Y., Mazur W., Wähälä K., Adlercreutz H., ( 1998). Pharmacokinetics of soybean isoflavones in plasma, urine and feces of men after ingestion of 60 g baked soybean powder (kinako). J. Nutr . 128, 1710– 1715. Google Scholar CrossRef Search ADS PubMed  Węgrzyn G., Jakóbkiewicz-Banecka J., Gabig-Cimińska M., Piotrowska E., Narajczyk M., Kloska A., Malinowska M., Dziedzic D., Gołębiewska I., Moskot M.et al.  , ( 2010). Genistein: A natural isoflavone with a potential for treatment of genetic diseases. Biochem. Soc. Trans . 38, 695– 701. Google Scholar CrossRef Search ADS PubMed  Wong M. M., Guo C., Zhang J. ( 2014). Nuclear receptor corepressor complexes in cancer: Mechanism, function and regulation. Am. J. Clin. Exp. Urol . 2, 169– 187. Google Scholar PubMed  Xiao C. W., L’Abbé M. R., Gilani G. S., Cooke G. M., Curran I. H., Papademetriou S. A. ( 2004). Dietary soy protein isolate and isoflavones modulate hepatic thyroid hormone receptors in rats. J. Nutr . 134, 743– 749. Google Scholar CrossRef Search ADS PubMed  Yen P. M. ( 2001). Physiological and molecular basis of thyroid hormone action. Physiol. Rev . 81, 1097– 1142. 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) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Toxicological Sciences Oxford University Press

A Possible Novel Mechanism of Action of Genistein and Daidzein for Activating Thyroid Hormone Receptor-Mediated Transcription

Loading next page...
 
/lp/ou_press/a-possible-novel-mechanism-of-action-of-genistein-and-daidzein-for-neS4b33XPD
Publisher
Oxford University Press
Copyright
© 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
ISSN
1096-6080
eISSN
1096-0929
D.O.I.
10.1093/toxsci/kfy097
Publisher site
See Article on Publisher Site

Abstract

Abstract Thyroid hormone receptors (TRs) are members of the nuclear receptor superfamily that regulate their target genes for controlling organ development and functional maintenance. Soybean isoflavones, especially genistein and daidzein, modulate various hormone-mediated pathways. However, their effects on TRs have not yet been extensively studied. In this study, the effects of these isoflavones on TR action were evaluated using transient transfection-based reporter gene assays and molecular docking studies. Genistein and daidzein augmented T3-liganded TR-mediated transcription in a concentration-dependent manner. In the mammalian 2-hybrid study, these isoflavones augmented the recruitment of steroid receptor coactivator-1 and nuclear corepressor to liganded or unliganded TRs. Using a series of mutant TRs, we also showed that the activation function-2 domain of TRs was responsible for the augmentation by these isoflavones. CV-1 cells had expressed TRα, TRβ1, and ERα mRNAs. However, neither the overexpression nor the knocking down of ERα altered the augmentation of TR action by isoflavones, indicating that the effects of isoflavones are exerted through their direct action on TRs. In silico molecular docking studies showed that genistein and daidzein can directly bind to the TR-ligand-binding domain. These findings indicate that the augmentation of the TR-mediated transcription by genistein and daidzein is due to their direct binding to TR-ligand-binding domain to induce the recruitment of steroid receptor coactivator-1. Our study reports a novel mode of action of soybean isoflavones on TR function. The biological effects and the relevance of these isoflavones to human health may be partially attributable to the activation of thyroid hormone signaling. isoflavone, thyroid hormone receptor, estrogen receptor, T3, molecular docking Thyroid hormones (THs) (3, 5, 3’-tri-iodo-L-thyronine or T3; 3, 5, 3’, 5’-tetra-iodo-L-thyronine or thyroxine or T4) are essential for the development and functional maintenance of various organs. The actions of THs are mainly mediated by the nuclear TH receptor (TR), which binds to a specific DNA sequence called the TH response element (TRE), as a homodimer or as a TR-retinoid X receptor (RXR) heterodimer with the RXR. In the absence of a ligand, corepressor complexes, such as nuclear receptor (NR) corepressor (N-CoR) and silencing mediator for retinoid and thyroid receptors (SMRT) bind to TR-RXR heterodimers (Cheng et al., 2010; Yen, 2001; Brent, 2012). In the presence of T3, the corepressors are replaced by coactivator complexes including the steroid receptor coactivators (SRC)-1, SRC-2, and SRC-3, to activate transcription (Koibuchi and Chin, 2000; Koibuchi, 2008, 2013; Wong et al., 2014). Isoflavones are a natural class of isoflavonoids. They are produced exclusively by the legume family (Leuner et al., 2013). They exert various effects at molecular, cellular, and organ levels (Węgrzyn et al., 2010). Among the isoflavones, soybean isoflavones such as genistein and daidzein have been intensively studied. They inhibit proliferation and induce apoptosis in prostate cancer cell lines (LNCaP and PCa cell lines) (Dong et al., 2013), enhance the retention of bone calcium, down-regulate energy metabolism in human adipose tissues in postmenopausal women (Pawlowski et al., 2015; Velpen et al., 2014), and improve peroxisome proliferator-activated receptor (PPAR)α-mediated fatty acid oxidation in mouse liver (Qiu and Chen, 2015). Such a wide variety of actions indicate that these compounds probably act through several different signaling pathways. Both genistein and daidzein are well-known phytoestrogens that modulate the action of NRs including the estrogen receptor (ER) by binding to the ligand-binding domain (LBD) of ERα and ERβ (Leclercq and Jacquot, 2014). Soybean isoflavone, especially genistein, showed higher binding affinity to ERβ compare with ERα by radioligand solid-phase-binding assay and transactivation assays on human embryonal kidney 293 cells (Kuiper et al., 1998). Genistein at physiological range (0.5–10 μmol/l) also have been reported induces the reduction of ERβ promoter methylation with corresponding increases in ERβ expression and induces phosphorylation of ERβ (pS105 and pS87), nuclear translocation, and ERβ transcriptional activity in prostate cancer cell lines (LNCaP, LAPC-4, and PC-3 cells) (Mahmoud et al., 2013). In addition, genistein partially antagonizes the activity of the androgen receptor (AR) in a tissue-specific and AR target gene-specific manner in male mice (effective in prostate, testes, and brain, but not in skeletal muscles and lung) (Pihlajamaa et al., 2011). It may also act as a weak agonist in the brain and prostate tissues of mice (Pihlajamaa et al., 2011). Genistein also upregulates pregnane X receptor (PXR)-mediated transcription and PXR-mediated CYP3A4 mRNA expression in mouse hepatocytes (Li et al., 2009). Genistein and daidzein also modulate the protein and mRNA levels of NRs; for example, genistein increases the mRNA levels of ERβ in the rat hypothalamus (Patisaul et al., 2002), decreases the mRNA levels of PPARγ during adipogenesis in human primary bone marrow stromal cells (Heim et al., 2004), and decreases AR protein levels in LNCaP cell lines (Basak et al., 2008). These findings indicate that genistein and daidzein may exert their actions partly by modulating the action and/or expression of NRs. The effect of dietary soybean isoflavones on the TH system is rather controversial. Although the goitrogenic effects of soy formula in infants have been well documented, the effects can be reversed by switching the formula to cow milk (Chorazy, 1995). Although dietary genistein treatment in young adult rat models significantly suppressed the levels of thyroid peroxidase (TPO), the levels of T3, T4, and TH-stimulating hormone (TSH) were not altered (Chang and Doerge, 2000). In orchidectomized middle-aged rats, dietary genistein and daidzein weakly suppressed the levels of T3 and T4, and increased both the TSH level and the cellular volume of TSH cells in the pituitary, while decreasing the volume of colloid in the thyroid gland (Šošić-Jurjević, 2010) and suppressing the mRNA levels of thyroglobulin and TPO, which indicate the impairment of TH synthesis. On the other hand, the mRNA levels of spot 14 and type 1 iodothyronine deodinase, which are positively regulated by TH, increased in the liver of the same experimental animal (Šošić-Jurjević et al., 2014). These results indicate that soybean isoflavones may have bimodal effects on the TH system. Although the effect of soybean isoflavones on TR-mediated transcription has not yet been extensively studied, a weak induction of TRα-mediated transcription by genistein using a conventional reporter assay system has been reported (Hofmann et al., 2009). However, whether these isoflavones bind to TRs have not been studied. In addition to the possibility that genistein and daidzein act directly on the TRs, these isoflavones may alter TR action through ERs. Crosstalk between NRs is important for the conversion of external and internal stimuli, which is necessary for eliciting cellular physiological responses (Vasudevan et al., 2001). The crosstalk and cross-interference between TRs and ERs have been reported in several gene promoters and cell lines (Vasudevan et al., 2001 2008), and is one of the reasons for the binding of TRs and ERs to the common DNA half-site sequence, 5’AGGTCA 3’. These findings indicate that genistein and daidzein may alter the interaction between ERs and TRs, or alter the interaction between TRs and their cofactors due to cofactor squelching by the ERs. However, their effects on the interaction between TRs and ERs are not clear. In this study, to clarify the effects of the isoflavones on TR activity, we employed reporter gene assays using TRα and TRβ expression vectors together with their mutants, in addition to in silico molecular docking. Since a recent study has identified a second TH-binding site located among H9, H10, and H11 of the TR-LBD (Souza et al., 2014), we also examined the possible binding of the isoflavones to this second binding site. Moreover, we also examined the crosstalk of TRs with ERs by using knock-down and overexpression strategies. MATERIALS AND METHODS Chemicals T3, genistein [4’, 5, 7-trihydoxyisoflavone or 5, 7-dihydroxy-3-(4-hydoxyphenyl)-4H-1-benzopyran-4-one] (Figure 1A), and daidzein [4′, 7-dihydroxyisoflavone or 7-hydroxy-3-(4-hydroxyphenyl) chromen-4-one] (Figure 1B) were purchased from Sigma (St Louis, Missouri). The purity of all chemicals was above 98%. Figure 1. View largeDownload slide Genistein and daidzein augmented TR-mediated transcription. A and B, Molecular structures of genistein and daidzein. C and D, Expression plasmids encoding 4 ng TRβ1 (C) and TRα1 (D) were co-transfected with 40 ng DR4-TK-LUC into CV-1 cells. The cells were cultured in the absence or presence of 10−7 T3 along with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05 indicates statistical significance by Bonferroni’s test compared with TRα1 or TRβ1 (+), T3 (+), and genistein or daidzein (–). Figure 1. View largeDownload slide Genistein and daidzein augmented TR-mediated transcription. A and B, Molecular structures of genistein and daidzein. C and D, Expression plasmids encoding 4 ng TRβ1 (C) and TRα1 (D) were co-transfected with 40 ng DR4-TK-LUC into CV-1 cells. The cells were cultured in the absence or presence of 10−7 T3 along with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05 indicates statistical significance by Bonferroni’s test compared with TRα1 or TRβ1 (+), T3 (+), and genistein or daidzein (–). Plasmids The TRα1 and TRβ1 expression vectors, luciferase (LUC) reporter constructs, the artificial direct repeat TRE, DR4-TK-LUC (DR4-TRE), and chick lysozyme (F2)-thymidine kinase (TK)-LUC (F2-TRE) have been described in previous studies (Iwasaki et al., 2001; Koibuchi et al., 1999). The LBD of VP16-TRβ1- has been previously described in Miyazaki et al. (2004). Gal4-blank, the expression vector for the Gal4-DNA-binding domain (DBD)-fused SRC-1-NR-binding domain (NBD)-1 (amino acids 595-780), Gal4-N-CoR-nuclear receptor-interacting domain (RID) (amino acids 1579–2454) and Gal4-SMRT- RID (amino acids 1669–2507) have been described in a previous study (Takeshita et al., 2002). The mutated TR plasmid ΔN.hTRβ1 that has an AF-1-binding domain with a truncated N-terminal; E457A, a full-length hTRβ1 containing a point mutation (glutamate to alanine) in codon 457 at the AF-2 domain; and ΔN.E457A, that has both the truncated N-terminal and the E457A mutation have been described in a previous study (Iwasaki et al., 2006). Clonal cell culture Monkey kidney fibroblast-derived clonal cells, CV-1, were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and antibiotics (100 U/ml penicillin, and 100 µg/ml streptomycin), at 37°C with 5% CO2. The serum was stripped of hormones by constantly mixing with 5% (w/v) AGX1-8 resin (Bio-Rad, Hercules, California) and powdered charcoal, prior to ultrafiltration (Iwasaki et al., 2002). Transient transfection-based reporter gene assay Cells were plated at a density of 1 × 104/0.1 ml in 96-well plates and incubated for 24 h, followed by transfection of the expression vectors and a reporter plasmid using HilyMax transfection reagents (Dojindo Molecular Technologies, Inc.), according to the protocol described in the technical manual. Expression vectors encoding TRα1 or TRβ1 (4 ng) were cotransfected with the reporter plasmid (F2-TRE-LUC) (40 ng) into CV-1 cells. The internal control was a cytomegalovirus-β-galactosidase plasmid (4 ng). The cells were incubated after 16–18 h of transfection with fresh medium containing indicated concentrations of the ligand (10−7 M T3) and either of genistein or daidzein for 24 h. The cells were then harvested to measure LUC activity as described in a previous study (Iwasaki et al., 2002). The total amounts of DNA per well were balanced by adding pcDNA3 plasmids (Invitrogen, San Diego, California). The LUC activity was normalized to β-galactosidase activity and represented as relative LUC activity. All transfection experiments were carried out in triplicate. The data are represented as the mean ± SEM of one representative experiment performed in triplicate. RNA interference and cell transfection Short interfering RNAs (siRNAs) for ERα (ESR1) and control RNAs were purchased from Thermo Fisher Scientific (Massachusetts). The siRNAs were transfected into CV-1 cells using lipofectamine RNAiMAX reagent (Thermo Fisher), according to the manufacturer’s protocol. Briefly, the siRNA-lipid complexes (25 nM control siRNA (scrambled RNA) or 25 nM ESR1 siRNA) were added to 96-well plates, and incubated for 20 min. CV-1 cells at a density of 1 × 104 cells/well were seeded in 96-well plates using media free from phenol red and antibiotics. After 16–24 h, the cells were subjected to analyses with reporter gene assays. The efficacy of the siRNA knockdown of ERα was verified by quantitative real-time PCR (qRT-PCR). RNA isolation and quantitative real-time RT-PCR The total RNA was isolated using QIAzol Lysis reagent (QIAGEN) and reverse transcribed using ReverTra Ace qPCR RT master mix (TOYOBO Bio-Technology, Japan) based on the instruction manual provided by the supplier. RT-PCR was performed using THUNDERBIRD SYBR qPCR mix (TOYOBO Bio-Technology, Japan) as described in the instruction manual and the StepOne RT-PCR System (Applied Biosystems). The list of primers used in this study is listed in Table 1. The RT-PCR protocol for all genes involved denaturation at 95°C for 20 s, followed by amplification at 95°C for 3 s and at 60°C for 30 s (40 cycles). All experiments were repeated 3 times, using independent RNA preparations to confirm the consistency of the results. The mRNA levels were normalized by the mRNA level of GAPDH. Table 1. PCR Primers Used in the Study     Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA      Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA  Table 1. PCR Primers Used in the Study     Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA      Sequence   Gene    Forward (5’ → 3’)  Reverse (5’ → 3’)  THRA  Human  AAGACGAGCAGTGTGTCGTG  GGAATAGGTGGGATGGAGGT  THRB  Human  GGTGGAAAGGTTGACTTGGA  CATGGCAGCTCACAAAACAT  ESR1  Human  GATGAATCTGCAGGGAGAGG  TCCAGAGACTTCAGGGTGCT  ESR2  Human  TCAGGCATGCGAGTAACAAG  TCCAGCAGCAGGTCATACAC  SRC1  Human  CACACAGGCCTCTACTGCAA  GACGTCAGCAAACACCTGAA  GAPDH  Human  GATCATCAGCAATGCCTCCT  TGAGTCCTTCCACGATACCA  In silico ligand-receptor-binding calculations All in silico calculations were performed using an Asus N43s notebook with an Intel Core i3 -2310 M dual-core processor, 3 M cache, 2.10 GHz CPU, 1333 MHz DDR3 SDRAM, and 8 GHz RAM, running on a Windows 10 professional operating system. The molecular structure of genistein (PubChem CID 5280961), daidzein (PubChem CID 5281708), and T3 (PubChem CID 5920) were downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/; last accessed May 17, 2017) in the structure data file format. The crystal structures of the TR-LBD were downloaded from the Protein Data Bank (PDB) (http://www.rcsb.org/pdb/home/home.do; last accessed May 17, 2017) in PDB format (PDB IDs: 4LNX, 4LNW, 3JZB, and 3HZF). The 3D structure files of the TR-LBDs and ligands were opened and modified with Discovery Studio structure-based design software, version 4.0 (BIOVIA/Accelrys Inc., San Diego, California). The water molecules and other substructures (bound molecules or ligand molecules) were removed from the coordinate file before docking. The unliganded TR-LBD was used for the individual dockings of T3, genistein, and daidzein. Polar hydrogen atoms were added to the 3D structure of the TR-LBD and the input file was generated in the PDBQT format, which contained the structure of the TR-LBD, using AutoDockTools of MGLTools (http://autodock.scripps.edu/resources/adt; last accessed May 17, 2017). The coordinates for docking were determined through a grid box using the PyRx - Python Prescription 0.8 Virtual Screening software for Computer-Aided Drug Design (http://pyrx.sourceforge.net/; last accessed May 17, 2017), using AutoDock 4 and AutoDock Vina as docking software (Trott and Olson, 2010). A blind docking strategy was utilized in order to include the entire possible binding site for ligands. For more reliable results, refinement docking experiments with repetitions of 30 runs were performed with complexes which had high affinity scores (lower than −9 kcal/mol). LigPlot+ v.1.4 (http://www.ebi.ac.uk/thornton-srv/software/LigPlus/; last accessed May 17, 2017) was used to determine the interactions between the TR-LBDs and the ligands in complexes with the best affinity scores. The binding affinity was expressed as the binding free energy (kcal/mol). Statistical analysis All the data are expressed as the mean ± SEM of 3 individual experiments performed in triplicate and analyzed using ANOVA. Post hoc comparisons were made using Bonferroni’s test. A p value < .05 was considered to be significant. RESULTS Genistein and Daidzein Augmented TR-Mediated Transcription by THs in CV-1 Cells To investigate the effects of genistein and daidzein on TR-mediated transcription, we first performed transient transfection-based reporter gene assays in CV-1 cells. Both genistein and daidzein augmented the transcription mediated by TRβ1 and TRα1 with 10−7 M T3 through DR4-TRE in a concentration-dependent manner (Figs. 1C and 1D). At 10−5 M concentration, genistein augmented TRβ1-mediated transcription by 3-fold, and activated TRα1-mediated transcription by 2-fold compared with the group that received only T3. Daidzein at a concentration of 10−5 M, also upregulated the transcription mediated by TRβ1 and TRα1 by 2-fold, respectively. Genistein and daidzein also augmented TRβ1- and TRα1-mediated transcription through F2-TRE (Supplementary Figs. 1A and 1B). On the other hand, although a significant increase in transcription was observed at higher doses of the isoflavones in the absence of T3, (TRβ1, 10−6 and 10−5 M genistein or 10−5 M daidzein; TRα, 10−5 M genistein or daidzein), the increase was rather weak compared with the transcription in the presence of T3 (Figs. 1C and 1D;Supplementary Figs. 1A and 1B). Genistein and Daidzein Augmented the Interaction of TRs with SRC-1 and N-CoR We hypothesized that the increase in transcription induced by the isoflavones may be caused due to the increased recruitment of coactivators, such as SRC-1 or dissociation corepressors such as N-CoR, to TRs. To examine the binding of such cofactors to TRs in the presence of the isoflavones, we carried out mammalian 2-hybrid assays. The transcription by VP16-TRβ1-LBD and Gal4-SRC-1-NBD (Figure 2A) through 5xUAS was activated by 10−7 M T3 (Figure 2B, lane 4). This activation was further augmented by genistein and daidzein in a concentration-dependent manner (Figure 2B, lanes 6–8 and 11–12). On the other hand, in the absence of 10−7 M T3, only a high concentration of genistein and daidzein (10−5 M) could activate the transcription by VP16-TRβ1-LBD and Gal4-N-CoR (Figure 2C) through 5xUAS (Figure 2D, lanes 8 and 12). We also observed the activation of transcription by VP16-TRβ1-LBD and Gal4-SMRT (Supplementary Figure 2A) in the absence of 10−7 M T3 (Supplementary Figure 2B, lanes 8 and 12). These results indicate that an increase in the recruitment of coactivators to TRs by the isoflavones might be the reason underlying the TR-mediated transcriptional augmentation. Figure 2. View largeDownload slide The effects of genistein and daidzein on the interaction of TRs with cofactors. Schematic diagrams of Gal4-SRC-1-NBD-1 (A) and Gal4-N-CoR-RID (C). Expression plasmids encoding Gal4-DBD-fused SRC-1-NBD-1 (B) (4 ng), or Gal4-N-CoR- RID (D) (4 ng) were co-transfected with VP16-constructs (4 ng) and 5× UAS-TK-LUC-reporter plasmids (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05, indicates statistical significance by Bonferroni’s test compared with Gal4-SRC-1 (+) or Gal4-N-CoR (+), VP16-TRβ1 (+), and T3 (10−7 M). Figure 2. View largeDownload slide The effects of genistein and daidzein on the interaction of TRs with cofactors. Schematic diagrams of Gal4-SRC-1-NBD-1 (A) and Gal4-N-CoR-RID (C). Expression plasmids encoding Gal4-DBD-fused SRC-1-NBD-1 (B) (4 ng), or Gal4-N-CoR- RID (D) (4 ng) were co-transfected with VP16-constructs (4 ng) and 5× UAS-TK-LUC-reporter plasmids (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. The data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, **p < .01, *p < .05, indicates statistical significance by Bonferroni’s test compared with Gal4-SRC-1 (+) or Gal4-N-CoR (+), VP16-TRβ1 (+), and T3 (10−7 M). The AF-2 Domain of TRs Is Essential for the Augmentation of TR-Mediated Transcription by the Isoflavones We further performed transient transfection-based reporter gene assays in CV-1 cells using a series of truncation and/or point mutants of TRβ1 to identify the domain of TR responsible for the transcriptional augmentation by isoflavones. The transcription through ΔN.hTRβ1, which is an N-terminus truncated mutant, was augmented by genistein and daidzein, and the transcription levels were similar to the wild type TRβ1 (Figure 3). On the other hand, the transcriptions through the hTRβ1-AF2 (E457A) and ΔN.hTRβ1-AF2 mutants were not augmented by the isoflavones (Figure 3). These results indicate that the AF-2 domain is responsible for the transcriptional augmentation by genistein and daidzein. Figure 3. View largeDownload slide Genistein and daidzein augmented the transcription mediated by ΔN.TRβ1 but not in AF-2 mutants. A and B, Expression plasmids encoding TRβ1 (4 ng) or ΔN.hTRβ1(4 ng) or E457A (4 ng) or ΔN.E457A (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein (A) or daidzein (B). The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test. Figure 3. View largeDownload slide Genistein and daidzein augmented the transcription mediated by ΔN.TRβ1 but not in AF-2 mutants. A and B, Expression plasmids encoding TRβ1 (4 ng) or ΔN.hTRβ1(4 ng) or E457A (4 ng) or ΔN.E457A (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) along with the indicated concentrations of genistein (A) or daidzein (B). The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test. The Augmentation of TR-Mediated Transcription by Isoflavones Is Not Altered by ERα Overexpression or Knockdown We measured the mRNA expression levels of TRs and ERs in CV-1 cells prior to the modulations in ER expression. The expression of TRs and ESR1 (ERα) was confirmed, whereas ESR2 (ERβ) expression could not be detected (Supplementary Figure 3). Based on the expression levels of ERs in CV-1 cells, we cotransfected ERα to CV-1 cells to further examine the involvement of ER, especially the effects of ERα overexpression. The overexpression of ERα did not further augment the transcription mediated by TRβ (Figure 4A) and TRα (Supplementary Figure 4A) in the presence of the isoflavones. These results indicate that ERα may not be involved in isoflavone-induced augmentation of TR-mediated transcription. To further examine whether the effects of genistein and daidzein were indeed induced through ERα, the ERα mRNA was knocked down using RNA interference. The CV-1 cells were transiently transfected with ERα siRNA. The decrease in the expression of ERα mRNA was confirmed by qRT-PCR. The siRNA treatment induced a 98.6% reduction in ERα mRNA levels compared with the control (Supplementary Figure 4B). After knocking down the ERα mRNA in CV-1 cells, we performed reporter gene assays. The magnitude of TRβ transcription induced by 10−5 M genistein or daidzein (Figure 4B) and TRα-mediated transcription in DR4-TRE (Supplementary Figure 4C) did not decrease after ERα knockdown. These results indicated that both genistein and daidzein might directly bind to TRs and augment TR-mediated transcription, although there might be differences in genistein and daidzein action on the interaction of TRs. Taken together with the study of ERα overexpression, the results of this study indicate that ERα is not involved in isoflavone-augmented TR transcription. Figure 4. View largeDownload slide The augmentation of TR-mediated transcription by genistein and daidzein was not altered after ERα overexpression or knockdown. A,. Expression plasmids encoding TRβ1 (4 ng) and ERα (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) and genistein (10−5 M) or daidzein (10−5 M). B, 25 nM scrambled RNA or 25 nM ESR1 mRNA was transfected into the CV-1 cells; 16–24 h after transfection, expression plasmids encoding TRβ1 (4 ng) were cotransfected with DR4-TK-LUC (40 ng). The cells were cultured in the absence or presence of 10−7 M T3 together with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), and genistein or daidzein (−). ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), scRNA (+) and the TRβ1 group (+), T3 (+), scrambled siRNA (+) or genistein or daidzein (−). ###p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), ESR1 siRNA (+) and genistein or daidzein (−). Figure 4. View largeDownload slide The augmentation of TR-mediated transcription by genistein and daidzein was not altered after ERα overexpression or knockdown. A,. Expression plasmids encoding TRβ1 (4 ng) and ERα (4 ng) were co-transfected with DR4-TK-LUC (40 ng) into CV-1 cells. The cells were cultured with or without T3 (10−7 M) and genistein (10−5 M) or daidzein (10−5 M). B, 25 nM scrambled RNA or 25 nM ESR1 mRNA was transfected into the CV-1 cells; 16–24 h after transfection, expression plasmids encoding TRβ1 (4 ng) were cotransfected with DR4-TK-LUC (40 ng). The cells were cultured in the absence or presence of 10−7 M T3 together with the indicated amounts of genistein or daidzein. The total amounts of DNA in all the wells were balanced by adding the pcDNA3 vector. All the data are represented as the mean ± SEM of 3 individual experiments performed in triplicate. ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), and genistein or daidzein (−). ***p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), scRNA (+) and the TRβ1 group (+), T3 (+), scrambled siRNA (+) or genistein or daidzein (−). ###p < .001, indicates statistical significance by Bonferroni’s test compared with TRβ1 (+), T3 (+), ESR1 siRNA (+) and genistein or daidzein (−). Genistein and Daidzein Can Directly Bind to the First and Second Binding Pockets of TR-LBD To investigate the binding modes of genistein and daidzein to the TR-LBD, we generated in silico binding models by molecular docking using AutoDocks Vina (Trott and Olson, 2010). We used several TRα-LBD crystal structures, including 4LNX (crystal structure of TRα bound to T4 in the second binding site), 4LNW (crystal structure of TRα bound to T3 in the second binding site), 3JZB (crystal structure of TRα bound to the selective thyromimetic TRIAC (3,3′,5-Triiodothyroacetic acid)), and 3HZF (crystal structure of TRα bou[nd to selective thyromimetic GC-1 in C2 space group). We used the TRα-LBD crystal structure with PDB ID 4LNX and performed molecular docking study with T3 (Figure 5A). We then performed the molecular docking study with genistein and daidzein. Genistein and daidzein could directly bind to the binding pockets of the TRα-LBD (Figs. 5B and 5C and Supplementary Figs. 5B and 5C) with a binding affinity −9.9 and −9.8 kcal/mol, respectively. The molecular docking studies of genistein and daidzein showed that the same amino acid residues in the crystal structure that interacted with T3, also interacted with genistein and daidzein, which included residues ile 222, ala 225, ser 277, leu 276, and leu 292 of the first binding site (Figure 5D) and arg 375 of the second binding site (Figure 5E). In the model obtained by molecular docking, genistein was located at 2.76 Å from ser 277 and 2.90 Å from phe 218 in the first binding site, and had the possibility to form hydrogen bonds. In the second binding site, genistein was located at 3.32 Å from arg 375, 2.49 Å from gln 342, and 2.52 Å from thr 327. The residues shown in red circles in Figures 5D and 5E represent the amino acid residues of TR-LBD that are common to T3 and genistein binding. We also performed in silico calculations for daidzein (Supplementary Figure 5) and determined the binding affinity of T3, genistein, and daidzein with other crystal structures of the TRα-LBD (PDB IDs 4LNW, 3JZB, and 3HZF) in both the binding pockets (Table 2). Table 2. Results of the Refinement Docking Experiments With AutoDock Vina   Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5    Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5  Table 2. Results of the Refinement Docking Experiments With AutoDock Vina   Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5    Binding Affinity (kcal/mol)   Genistein   Daidzein   T3   PDB ID of the TRα-LBD  First Site  Second Site  First Site  Second Site  First Site  Second Site  4LNX  –9.9  –8.8  –9.8  –6.6  –9.4  –7.4  4LNW  –10.2  –8.5  –9.9  –6.7  –9.1  –7.8  3JZB  –10.0  –8.0  –10.0  –7.8  –9.2  –7.2  3HZF  –10.0  –7.8  –10.0  –7.9  –9.3  –7.5  Figure 5. View largeDownload slide Genistein can bind to the first or second binding site of TR. The 3D structure of the TRα-LBD (PDB ID: 4LNX) with T3 bound to both binding sites (A), T3 with genistein at the second binding site (B), and genistein with T3 at the second binding site (C). The interaction plots between the ligand and the TRα-LBD (PDB ID: 4LNX) were generated by LigPlot+ v.1.4, with T3 and genistein at the first binding site (D) and the second binding site (E), with each subsequent plot being automatically fitted. The red circles and ellipses in each plot indicate protein residues that have equivalent 3D positions with respect to the residues in the first plot. Hydrogen bonds are shown as green dotted lines, while the spooked arcs represent residues making nonbonded contacts with the ligand. Figure 5. View largeDownload slide Genistein can bind to the first or second binding site of TR. The 3D structure of the TRα-LBD (PDB ID: 4LNX) with T3 bound to both binding sites (A), T3 with genistein at the second binding site (B), and genistein with T3 at the second binding site (C). The interaction plots between the ligand and the TRα-LBD (PDB ID: 4LNX) were generated by LigPlot+ v.1.4, with T3 and genistein at the first binding site (D) and the second binding site (E), with each subsequent plot being automatically fitted. The red circles and ellipses in each plot indicate protein residues that have equivalent 3D positions with respect to the residues in the first plot. Hydrogen bonds are shown as green dotted lines, while the spooked arcs represent residues making nonbonded contacts with the ligand. DISCUSSION In this study, we examined the effect of soybean isoflavones (genistein and daidzein) on TR-mediated transcription. We found that these isoflavones augmented the transcription mediated by liganded TRs in a concentration-dependent manner. ERs may not be involved in such augmentation as shown by the ER overexpression and siRNA studies. Using in silico analyses, we also found that these isoflavones may directly interact with the TR-LBD. These results describe a novel mechanism of action of genistein and daidzein on the augmentation of TR-mediated transcription by the direct binding of the isoflavones to TRs and TH-mediated signal transduction. Soybean isoflavones activate various NR-mediated transcriptions including those of ERs (Leclercq and Jacquot, 2014; Patisaul et al., 2002), and PXR (Li et al., 2009). Dietary soy protein isolates have been reported to upregulate hepatic TRβ1 expression levels in rats (Xiao et al., 2004). LUC reporter gene assay with ERs in human hepatoma cells (HepG2 cells) showed that both genistein and daidzein were complete agonist at both ER, genistein being more potent than daidzein. In addition, both genistein and daidzein were more potent to ERβ compare with ERα (Casanova et al., 1999). However, their effects on TR-mediated transcription have not been extensively studied. This study clearly shows that both genistein and daidzein augmented the transcription mediated by liganded TRs in a concentration-dependent manner. In contrast, in the absence of T3, only high doses of genistein and daidzein could weakly augment the transcription mediated by TRβ and TRα. This weak activation appears to be in agreement with a previous study which reported a weak transactivation of TR activity by genistein in the absence of T3 (Hofmann et al., 2009). In that study, genistein at a concentration of 10−6 M induced a 2.3-folds increase in the level of transcription in comparison to the basal level (without T3). Although the 2.3-fold increase with respect to the basal level seemed significant, the level of transcription was still rather low, because the basal transcription level in the absence of a ligand is repressed by interactions with corepressors. In this study, the TR-mediated transcription through F2-TRE with 10−7 M T3 was further augmented to 3-fold by genistein and daidzein at concentrations of 10−5 M. Thus, the effect of isoflavones on the expression of TRs may be greater in the presence of T3. The next question that may arise would concern the effect of isoflavones on modulating TH action in vivo. In this regard, additional studies using animal models are necessary. The plasma levels of genistein and daidzein in prostate cancer patients reached 1–10 µM, 2–8 h after the intake of 2 slices of standard soy or soy-almond bread (Ahn-jarvis et al., 2015), and reached 1–10 µM at 4–7 h after the intake of 50 g of kinako (roasted soybean powder) in human subjects (Hosoda et al., 2008). In this study, TR-mediated transcription was augmented by genistein and daidzein at doses comparable to those in the aforementioned studies (Figs. 1C and 1D). Previously, we had performed the MTS cell proliferation assay and showed that both genistein and daidzein concentrations that were used in this study did not affect the viability of CV-1 cells (Supplementary Figure 6). Unless large amounts are administered, their adverse effects cannot be reported. Since soybeans have been consumed throughout human evolution, isoflavones can be effectively metabolized in humans. The half-lives of plasma genistein and daidzein are 8.36 and 5.79 h, respectively (Watanabe et al., 1998). Thus, the endocrine-modulating action of the soy isoflavones under normal physiological conditions may be beneficial to our health, although further studies are required to confirm this, especially at acute and massive doses. Such a possibility indicates that isoflavones can be used as a supplement for patients with low T3 syndromes or to relieve hypothyroidism-induced clinical symptoms. The interaction of TRs with the cofactor SRC-1 induced by T3, was augmented by genistein and daidzein. However, only high doses of genistein and daidzein could augment the interactions with TR-N-CoR and TR-SMRT. Studies have also reported the augmentations in the interactions of SRC-1 with other NRs by genistein and daidzein. According to previous studies, genistein augmented the interaction of SRC-1 with ERα (Schwartz et al., 1998), and ERβ (Mueller et al., 2004). Additionally, modulations in N-CoR action by genistein have been reported. Genistein upregulated the expression of N-CoR protein and reversed promyelocytic leukemia (PML)-retinoic acid receptor (RAR)-induced misfolding of N-CoR protein possibly by inhibiting the selective phosphorylation-dependent binding of N-CoR to PML-RAR in acute PML cells (Ng et al., 2007). In this study, we showed that both genistein and daidzein augmented the interactions of a TR-coactivator (TR-SRC-1) and TR-corepressors (TR-N-CoR and TR-SMRT) at high doses (Figure 2). Moreover, using a series of truncated and/or point mutants of TRβ1, we found that the TR-LBD region and especially its AF-2 domain, is responsible for the augmentation in TR-mediated transcription induced by genistein and daidzein (Figure 3). The augmentation was observed with the N-terminus-truncated mutant, but not with AF-2 mutants. These results indicate that the TR-mediated transcription augmented by genistein and daidzein might have altered the conformation of AF-2, which modulated the binding of TRs to cofactors. Together with the results of mammalian 2 hybrid assays, it is likely that the isoflavone-induced increase in the interactions of SRC-1 to the AF-2 domain modified the chromatin structure which augmented transcription. On the other hand, although these isoflavones also augmented the recruitment of corepressors, transcription was not further repressed. The reason for this discrepancy cannot be clarified. Further studies are necessary for clarifying the mechanisms of increase in coactivator/corepressor recruitment by isoflavones. Soybean isoflavones are well-known phytoestrogens that act through ERs (Mueller et al., 2004; Patisaul et al., 2002). Since a previous study reported the possibility of crosstalk between TRs and ERs (Vasudevan et al., 2001), we speculated whether the effect of isoflavones on TRs is a result of the indirect binding of isoflavones to ERs. To further investigate the crosstalk between ERs and TRs, we separately overexpressed and knocked-down ERα, and observed that both genistein and daidzein augmented TR-mediated transcription in both the states of ERα overexpression and knockdown. These results indicate that ERα is not involved in the augmentations of both TR-mediated transcription and TR-cofactor interactions, induced by genistein and daidzein. We also confirmed that the transactivation mediated by TRs in the presence of T3 was increased by ERα knockdown (Figure 4B and Supplementary Figure 4C). However, genistein and daidzein might acts through both of ERs and TRs in the in vivo study. Transcriptomic analysis showed that genistein regulated some ER-responsive genes in the cell proliferation, adhesion, motility and inflammatory response proses (Gong et al., 2014). Some of these genes were also included in TR responsive gene. The increase of body weight in ovariectomized rats was reduced by co-administration of genistein and ERβ-specific agonist 8β-VE2, and suggesting that the activation of ERβ may modulate ERα-mediated physiological effects in vivo (Hertrampf et al., 2009). A previous study showed that there is a cross-interference (squelching) between TRs and ERs with regard to the binding of coactivators, including SRC-1 (Lopez et al., 1999). In fact, we confirmed the expression of SRC-1 mRNA in CV-1 cells (Supplementary Figure 3). These reasons explain the increased transactivation induced by ER knockdown. Nevertheless, the action of isoflavones on TRs may be directly induced and exerted through the TH signal transduction pathway since THs regulate various metabolic pathways. In this study, we also showed a possibility using in silico docking studies that these isoflavones may directly bind to the TR-LBD. Like other NRs, the TR LBDs has a core-binding site for T3 where the ligand is buried. The binding of THs change the conformation of TRs (Souza et al., 2014; Moras et al., 2015). The binding of the isoflavones to the TR-LBD may also induce conformational changes in the structure of TRs by inducing the recruitment of cofactors. In addition, a recent study identified a new hormone-binding site referred to as the second binding site, on the surface of the TR-LBD (Souza et al., 2014). We have shown that isoflavones can also bind to this second binding site, although the affinity is lower than the affinity with the first binding site. Importantly, the augmentations induced by the isoflavones were mainly observed in the presence of T3, indicating that both T3 and the isoflavones should be simultaneously bound in the TR complex to cause augmentations. Although the exact binding mode of T3 and the isoflavones could not be verified in this study, several possibilities have been considered, which need to be investigated in order to identify the roles and interaction between the first and second binding sites in TR-mediated transactivation. Moreover, we also need to examine whether the binding mode of TRs to the TRE may be changed by isoflavone-liganded TRs, because TRs can form homodimers and TR-RXR heterodimers for binding to TRE. In summary, isoflavones such as genistein and daidzein augment TR-mediated transcription at least in part by increasing the recruitment of SRC-1. Such augmentation may be caused by the direct binding of genistein and daidzein to the TR-LBD. However, further studies are required to clarify the mechanism; the results of this study provide a new insight into the understanding of the actions of genistein and daidzein on TRs. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. FUNDING This work was supported in part by Grants-in-Aid for Scientific Research (Nos. 26340036 and 25281024) to T.I and N.K from the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT). ACKNOWLEDGMENTS We thank Dr Yusuke Takatsuru, Dr Izuki Amano, and Dr Asahi Haijima of the Department of Integrative Physiology, for technical assistance and advice. REFERENCES Ahn-Jarvis J. H., Clinton S. K., Grainger E. M., Riedl K. M., Schwartz S. J., Lee M.-L. T., Cruz-Cano R., Young G. S., Lesinski G. B., Vodovotz Y. ( 2015). Isoflavone pharmacokinetics and metabolism after consumption of a standardized soy and soy-almond bread in men with asymptomatic prostate cancer. Cancer Prevent. Res . 8, 1045– 1054. Google Scholar CrossRef Search ADS   Basak S., Pookot D., Noonan E.J., Dahiya R. ( 2008) Genistein down-regulates androgen receptor by modulating HDAC6-Hsp90 chaperone function. Mol. Cancer Ther ., 7, 3195– 3202. Google Scholar CrossRef Search ADS PubMed  Brent G. A. ( 2012). Mechanisms of thyroid hormone action. Sci. Med . 122, 3035– 3043. Casanova M., You L., Gaido K. W., Archibeque-Engle S., Janszen D. B., Heck H. D. A. ( 1999). Developmental effects of dietary phytoestrogens in Sprague-Dawley rats and interactions of genistein and daidzein with rat estrogen receptors α and β in vitro. Toxicol. Sci . 51, 236– 244. Google Scholar CrossRef Search ADS PubMed  Chang H.C., Doerge D.R. ( 2000) Dietary genistein inactivates rat thyroid peroxidase in vivo without an apparent hypothyroid effect. Toxicol. Appl. Pharmacol ., 168, 244– 252. Google Scholar CrossRef Search ADS PubMed  Cheng S. Y., Leonard J. L., Davis P. J. ( 2010). Molecular aspects of thyroid hormone actions. Endocr. Rev . 31, 139– 170. Google Scholar CrossRef Search ADS PubMed  Chorazy P.A., Himelhoch S., Hopwood N.J., Greger N.G., Postellon D.C. ( 1995) Persistent Hypothyroidism in an Infant Receiving a Soy Formula: Case Report and Review of the Literature. Pediatrics , 96, 148– 150. Google Scholar PubMed  Dong X., Xu W., Sikes R. A., Wu C. ( 2013). Combination of low dose of genistein and daidzein has synergistic preventive effects on isogenic human prostate cancer cells when compared with individual soy isoflavone. Food Chem . 141, 1923– 1933. Google Scholar CrossRef Search ADS PubMed  Gong P., Madak-erdogan Z., Li J., Cheng J., Greenlief C. M., Helferich W., Katzenellenbogen J. A., Katzenellenbogen B. S. ( 2014). Transcriptomic analysis identifies gene networks regulated by estrogen receptor α (ERα) and ERβ that control distinct effects of different botanical estrogens. Nucl. Recept. Signal  12, 1– 13. Heim M., Frank O., Kampmann G., Sochocky N., Pennimpede T., Fuchs P., Hunziker W., Weber P., Martin I., Bendik I. ( 2004). The phytoestrogen genistein enhances osteogenesis and represses adipogenic differentiation of human primary bone marrow stromal cells. Endocrinology  145, 848– 859. Google Scholar CrossRef Search ADS PubMed  Hertrampf T., Seibel J., Laudenbach U., Fritzemeier K. H., Diel P. ( 2009). Analysis of the effects of oestrogen receptor α (ERα)- and ERβ-selective ligands given in combination to ovariectomized rats. Br. J. Pharmacol . 153, 1432– 1437. Google Scholar CrossRef Search ADS   Hofmann P. J., Schomburg L., Ko J. ( 2009). Interference of endocrine disrupters with thyroid hormone receptor-dependent transactivation. Toxicol. Sci . 110, 125– 137. Google Scholar CrossRef Search ADS PubMed  Hosoda K., Furuta T., Yokokawa A., Ogura K., Hiratsuka A., Ishii K. ( 2008). Plasma profiling of intact isoflavone metabolites by high-performance liquid chromatography and mass spectrometric identification of flavone glycosides daidzin and genistin in human plasma after administration of kinako. Drug Metab. Dispos . 36, 1485– 1495. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Chin W. W., Ko L. ( 2001). Identification and characterization of RRM-containing coactivator activator (CoAA) as TRBP-interacting protein, and its splice variant as a coactivator modulator (CoAM). J. Biol. Chem . 276, 33375– 33383. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Miyazaki W., Takeshita A., Kuroda Y., Koibuchi N. ( 2002). Polychlorinated biphenyls suppress thyroid hormone-induced transactivation. Biochem. Biophys. Res. Commun . 299, 384– 388. Google Scholar CrossRef Search ADS PubMed  Iwasaki T., Takeshita A., Miyazaki W., Chin W. W., Koibuchi N. ( 2006). The interaction of TRβ1-N terminus with steroid receptor coactivator-1 (SRC-1) serves a full transcriptional activation function of SRC-1. Endocrinology  147, 1452– 1457. Google Scholar CrossRef Search ADS PubMed  Koibuchi N. ( 2008). The role of thyroid hormone on cerebellar development. Cerebellum  7, 530– 533. Google Scholar CrossRef Search ADS PubMed  Koibuchi N. ( 2013). The role of thyroid hormone on functional organization in the cerebellum. Cerebellum  12, 304– 306. Google Scholar CrossRef Search ADS PubMed  Koibuchi N., Chin W. W. ( 2000). Thyroid hormone action and brain development. Trends Endocrinol. Metab . 11, 123– 128. Google Scholar CrossRef Search ADS PubMed  Koibuchi N., Liu Y., Fukuda H., Takeshita A., Yen P. M., Chin W. W. ( 1999). RORα augments thyroid hormone receptor-mediated transcriptional activation. Endocrinology  140, 1356– 1364. Google Scholar CrossRef Search ADS PubMed  Kuiper G. G., Lemmen J. G., Carlsson B., Corton J. C., Safe S. H., van der Saag P. T., van der Burg B., Gustafsson J. A. ( 1998). Interaction of estrogenic chemicals and pytoestrogens with estrogen receptor beta. Endocrinology  139, 4252– 4263. Google Scholar CrossRef Search ADS PubMed  Leclercq G., Jacquot Y. ( 2014). Interactions of isoflavones and other plant derived estrogens with estrogen receptors for prevention and treatment of breast cancer—Considerations concerning related efficacy and safety. J. Steroid Biochem. Mol. Biol . 139, 237– 244. Google Scholar CrossRef Search ADS PubMed  Leuner O., Havlik J., Hummelova J., Prokudina E., Novy P., Kokoska L. ( 2013). Distribution of isoflavones and coumestrol in neglected tropical and subtropical legumes. J. Sci. Food Agric . 93, 575– 579. Google Scholar CrossRef Search ADS PubMed  Li Y., Ross-Viola J. S., Shay N. F., Moore D. D., Ricketts M.-L. ( 2009). Human CYP3A4 and murine Cyp3A11 are regulated by equol and genistein via the pregnane X receptor in a species-specific manner. J. Nutr . 139, 898– 904. Google Scholar CrossRef Search ADS PubMed  Lopez G. N., Webb P., Shinsako J. H., Baxter J. D., Greene G. L., Kushner P. J. ( 1999). Titration by estrogen receptor activation function-2 of targets that are downstream from coactivators. Mol. Endocrinol . 13, 897– 909. Google Scholar CrossRef Search ADS PubMed  Mahmoud A. M., Zhu T., Parray A., Siddique H. R., Yang W., Saleem M., Bosland M. C. ( 2013). Differential effects of genistein on prostate cancer cells depend on mutational status of the androgen receptor. PLoS One  8, e78479– e78421. Google Scholar CrossRef Search ADS PubMed  Miyazaki W., Iwasaki T., Takeshita A., Kuroda Y., Koibuchi N. ( 2004). Polychlorinated biphenyls suppress thyroid hormone receptor-mediated transcription through a novel mechanism. J. Biol. Chem . 279, 18195– 18202. Google Scholar CrossRef Search ADS PubMed  Moras D., Billas I. M. L., Rochel N., Klaholz B. P. ( 2015). Structure-function relationships in nuclear receptors: The facts. Trends Biochem. Sci . 40, 287– 290. Google Scholar CrossRef Search ADS PubMed  Mueller S. O., Simon S., Chae K., Metzler M., Korach K. S. ( 2004). Phytoestrogens and their human metabolites show distinct agonistic and antagonistic properties on estrogen receptor α (ERα) and ERβ in human cells. Toxicol. Sci . 80, 14– 25. Google Scholar CrossRef Search ADS PubMed  Ng A. P. P., Nin D. S., Fong J. H., Venkataraman D., Chen C.-S., Khan M. ( 2007). Therapeutic targeting of nuclear receptor corepressor misfolding in acute promyelocytic leukemia cells with genistein. Mol. Cancer Ther . 6, 2240– 2248. Google Scholar CrossRef Search ADS PubMed  Patisaul H. B., Melby M., Whitten P. L., Young L. J. ( 2002). Genistein affects ERβ- but not ERα-dependent gene expression in the hypothalamus. Endocrinology  143, 2189– 2197. Google Scholar CrossRef Search ADS PubMed  Pawlowski J. W., Martin B. R., Mccabe G. P., Mccabe L., Jackson G. S., Peacock M., Barnes S., Weaver C. M. ( 2015). Impact of equol-producing capacity and soy-isoflavone profiles of supplements on bone calcium retention in postmenopausal women : A randomized crossover trial. Am. J. Clin. Nutr . 102, 695– 703. Google Scholar CrossRef Search ADS PubMed  Pihlajamaa P., Zhang F.P., Saarinen L., Mikkonen L., Hautaniemi S., Jänne O.A. ( 2011) The phytoestrogen genistein is a tissue-specific androgen receptor modulator. Endocrinology , 152, 4395– 4405. Google Scholar CrossRef Search ADS PubMed  Qiu L., Chen T. ( 2015). Novel insights into the mechanisms whereby isoflavones protect against fatty liver disease. World J. Gastroentrol . 21, 1099– 1107. Google Scholar CrossRef Search ADS   Schwartz J. A., Liu G., Brooks S. C. ( 1998). Genistein-mediated attenuation of tamoxifen-induced antagonism from estrogen receptor-regulated genes. Biochem. Biophys. Res. Commun . 253, 38– 43. Google Scholar CrossRef Search ADS PubMed  Šošić-Jurjević B., Filipović B., Ajdžanović V., Savin S., Nestorović N., Milošević V., Sekulić M. ( 2010) Suppressive effects of genistein and daidzein on pituitary–thyroid axis in orchidectomized middle-aged rats. Exp. Biol. Med ., 235, 590– 598. Google Scholar CrossRef Search ADS   Šošić-Jurjević B., Filipović B., Wirth E. K., Živanović J., Radulović N., Janković S., Milošević V., Köhrle J. ( 2014). Soy isoflavones interfere with thyroid hormone homeostasis in orchidectomized middle-aged rats. Toxicol. Appl. Pharmacol . 278, 124– 134. Google Scholar CrossRef Search ADS PubMed  Souza P. C. T., Puhl A. C., Martínez L., Aparício R., Nascimento A. S., Figueira A. C. M., Nguyen P., Webb P., Skaf M., Polikarpov I. ( 2014). Identification of a new hormone-binding site on the surface of thyroid hormone receptor. Mol. Endocrinol . 28, 534– 545. Google Scholar CrossRef Search ADS PubMed  Takeshita A., Taguchi M., Koibuchi N., Ozawa Y. ( 2002). Putative role of the orphan nuclear receptor SXR (steroid and xenobiotic receptor) in the mechanism of CYP3A4 inhibition by xenobiotics. J. Biol. Chem . 277, 32453– 32458. Google Scholar CrossRef Search ADS PubMed  Trott O., Olson A. ( 2010). AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading. J. Comput. Chem . 31, 455– 461. Google Scholar PubMed  Vasudevan N., Koibuchi N., Chin W. W., Pfaff D. W. ( 2001). Differential crosstalk between estrogen receptor (ER)α and ERβ and the thyroid hormone receptor isoforms results in flexible regulation of the consensus ERE. Mol. Brain Res . 95, 9– 17. Google Scholar CrossRef Search ADS   Vasudevan N., Zhu Y. S., Daniel S., Koibuchi N., Chin W. W., Pfaff D. ( 2008). Crosstalk between oestrogen receptors and thyroid hormone receptor isoforms results in differential regulation of the preproenkephalin gene. J. Neuroendocrinol . 13, 779– 790. Google Scholar CrossRef Search ADS   Velpen V., Geelen A., Hollman P. C. H., Schouten E. G., Veer P. V., Afman L. A. ( 2014). Isoflavone supplement composition and equol producer status affect gene expression in adipose tissue: A double-blind, randomized, placebo-controlled crossover trial in postmenopausal women. Am. J. Clin. Nutr . 100, 1269– 1277. Google Scholar CrossRef Search ADS PubMed  Watanabe S., Yamaguchi M., Sobue T., Takahashi T., Miura T., Arai Y., Mazur W., Wähälä K., Adlercreutz H., ( 1998). Pharmacokinetics of soybean isoflavones in plasma, urine and feces of men after ingestion of 60 g baked soybean powder (kinako). J. Nutr . 128, 1710– 1715. Google Scholar CrossRef Search ADS PubMed  Węgrzyn G., Jakóbkiewicz-Banecka J., Gabig-Cimińska M., Piotrowska E., Narajczyk M., Kloska A., Malinowska M., Dziedzic D., Gołębiewska I., Moskot M.et al.  , ( 2010). Genistein: A natural isoflavone with a potential for treatment of genetic diseases. Biochem. Soc. Trans . 38, 695– 701. Google Scholar CrossRef Search ADS PubMed  Wong M. M., Guo C., Zhang J. ( 2014). Nuclear receptor corepressor complexes in cancer: Mechanism, function and regulation. Am. J. Clin. Exp. Urol . 2, 169– 187. Google Scholar PubMed  Xiao C. W., L’Abbé M. R., Gilani G. S., Cooke G. M., Curran I. H., Papademetriou S. A. ( 2004). Dietary soy protein isolate and isoflavones modulate hepatic thyroid hormone receptors in rats. J. Nutr . 134, 743– 749. Google Scholar CrossRef Search ADS PubMed  Yen P. M. ( 2001). Physiological and molecular basis of thyroid hormone action. Physiol. Rev . 81, 1097– 1142. 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)

Journal

Toxicological SciencesOxford University Press

Published: Apr 23, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off