- Email: [email protected]
A novel pesticide-induced conformational state of the oestrogen receptor ligand-binding domain, detected by conformation-specific peptide binding
FEBS 29176
FEBS Letters 579 (2005) 541–548
A novel pesticide-induced conformational state of the oestrogen receptor ligand-binding domain, detected by conformation-specific peptide binding Vadim V. Sumbayeva,*, Eva C. Bonefeld-Jørgensenb, Troels Winda, Peter A. Andreasena a
Laboratory of Cellular Protein Science, Department of Molecular Biology and The Interdisciplinary Nanoscience Centre, University of Aarhus, Aarhus, Denmark b Unit of Environmental Biotechnology, Department of Environmental Medicine, Aarhus University, Aarhus, Denmark Received 25 November 2004; revised 13 December 2004; accepted 13 December 2004 Available online 22 December 2004 Edited by Barry Halliwell
Abstract The diverse effects of different natural and synthetic oestrogen receptor ligands depend on induction of different receptor conformations, allowing differential interactions with other transcription factors. Different conformations of the oestrogen receptor ligand binding domains can be monitored by conformation-specific binding to peptides selected from phage-displayed peptide libraries. We now report that a group of chlorinated pesticides, including 2,4-dichlorodiphenyl-dichloroethylene, induces a peptide recognition pattern different from those induced by any one of the classical oestrogen receptor ligands. The pesticide-complexed oestrogen receptors recognized peptides reacting with the receptors complexed both with the natural oestrogen 17b-oestradiol and with the synthetic partial antagonist 4-hydroxy-tamoxifen, respectively, indicating that the pesticide-induced conformation shares features with both the 17b-oestradiol- and the 4-hydroxy-tamoxifen-induced conformations. The substitution H524A in the ligand binding domain conferred the pesticide-specific peptide recognition pattern and transactivation activity to the oestradiol- and the 4-hydroxy-tamoxifen-complexed receptors, indicating that one important determinant of the pesticide-induced conformation is a lack of stabilisation of any one particular receptor conformation by ligand interaction with H524, which is known to interact with both oestradiol and 4-hydroxy-tamoxifen. Thus, peptide binding analyses of oestrogen receptor conformations induced by environmental endocrine disruptors can give novel information about molecular mechanisms of oestrogen action in general. 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Oestrogen receptor; Oestrogen; Endocrine disruptor; Phage-displayed peptide library
1. Introduction The physiological effects of oestrogens are mediated by the oestrogen receptors (ERs), ERa and ERb, members of the nuclear receptor family of ligand-regulated transcription factors. Hormone binding to the ERs ligand-binding domain (LBD) initiates a series of molecular events culminating in the activation or repression of target genes [1,2]. When oestradiol (E2), * Corresponding author. Fax: +45 8612 3178. E-mail address: [email protected] (V.V. Sumbayev).
the natural oestrogen, binds to ERs, the phenolic hydroxyl of its A-ring nestles between a-helix (H) 3 and H6 of the receptor and makes direct hydrogen bonds to the carboxylate of Glu353 and the guanidinium group of Arg394 (using the ERa residue numbering system). The 17b hydroxyl of oestradiolÕs D-ring makes a hydrogen bond with His524 in H11. Transcriptional regulation arises from direct interaction of ligand-occupied ERs with its cognate DNA target site (oestrogen responsive elements, EREs) in the regulatory regions of oestrogen-sensitive genes [3–5]. Moreover, conformational changes of the LBD induced by ligand binding confer further specificity upon the hormonal response and is a key event permitting discrimination between agonists (E2, oestriol (E3)), partial antagonists (4-hydroxytamoxifen (OHT)), and pure antagonists (ICI 182 780). The most prominent conformational change is a ligand-mediated repositioning of H12. In the agonist-liganded receptor, H12 lies over the ligand-binding cavity and forms, together with H3 and H5, a binding surface for transcriptional co-regulators. In receptor liganded by partial antagonists, H12 occupies the groove between H3 and H5 and prevents co-regulator binding. With pure antagonists, H12 is pushed away and is invisible in X-ray crystal structure analysis. Different positionings of H12 and associated changes in other helices thus result in different co-regulator binding surfaces and hence underlie the different physiological effects of agonists and antagonists [6–9]. In recent decades, there has been an increasing concern about environmental chemicals that may interfere with oestrogenic signalling and adversely affect normal reproduction of humans and wildlife. These compounds, known as environmental endocrine disruptors, encompass a wide range of substances including natural products, pesticides, pharmaceuticals and industrial chemicals [10–12]. Some of those chemicals have shown either ER agonistic or antagonistic properties in cell proliferation and transactivation assays with the breast carcinoma cell line MCF-7 [13]. Many of them, especially a number of widely distributed pesticides, do not share any obvious structural similarity to natural oestrogens, making their identification, based solely on molecular structure, impossible [10,14,15]. In the present study, we have investigated oestrogen receptor conformational changes mediated by 2,4-dichlorodiphenyldichloroethylene (DDE), a metabolite of 2,4-dichlorodiphenyl-trichloroethane (DDT), and other pesticides. For the study, we have utilized the ability of peptides isolated from
0014-5793/$22.00 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2004.12.025
542
phage-displayed peptide libraries to recognise specific ER conformations [16]. We have found that a group of chlorinated pesticides induces a ER peptide recognition pattern not described before and hence an ER conformational state, which is not shared with any of the classical ER ligands. Moreover, we demonstrate by site-directed mutagenesis in phage displayed peptide binding and transactivation assays that His524 is one important determinant for the ligand-induced conformational changes. Our studies give novel important information about ER conformational changes and the mechanisms by which endocrine disruptors may interfere with oestrogen action.
2. Materials and methods 2.1. Materials The following materials were purchased from the indicated suppliers: streptavidin (Calbiochem, San Diego, CA, USA); ERa and ERb (Panvera, Madison, WI, USA); DNA cassettes, primers, and biotinylated oligonucleotides corresponding to vitellogenin ERE (biotin-GATCTAGGTCACAGTGACCTGCG – forward and biotin-GATCCGC AGGTCACTGTGACCTA – reverse) (DNA Technology, Aarhus, Denmark); E2, E3, and OHT (Sigma, St. Louis, MO, USA); DDE and other pesticides (Dr. Ehrenstorfer, Augsburg, Germany); mouse anti-M13 monoclonal antibody conjugated with horseradish peroxidase (HRP) and anti-glutathione-S-transferase (GST) goat antibody (Amersham Biosciences, Hillerød, Denmark); glutathione-(GSH) agarose (Molecular Probes, Eugene, OR, USA); biotinylated GSH (Biopeptide, San-Diego, CA, USA); MaxisorpTM microtitre plates (Nunc, Roskilde, Denmark); E. coli BL21 (Stratagene, La Jolla, CA, USA); and FuGene6 (Roche Diagnostics Corp., Indianapolis, IN, USA). All other chemicals were of the best grade commercially available. 2.2. Fusion phage M13 phage particles displaying ER-binding peptides were prepared by ligating oligonucleotide cassettes coding for each peptide into SfiIdigested fUSE5 vector followed by production of fusion phage in E. coli DH-5a cells [17]. The created fusion phage was validated by DNA sequencing of PCR products generated with the primers M13for (5 0 -TTTCGACACAATTTATCAGG-3 0 ) and M13back (5 0 TGAATTTTCTGTATGAGGTTTTG-3 0 ) and Taq DNA polymerase (Invitrogen, Denmark). The PCR-products were purified (Qiaquick PCR Purification Kit, Qiagen, Germany) and sequenced with the DYEnamic ET Terminator Cycle Sequencing Kit (Amersham Biosciences, Hillerød, Denmark) and an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, CA, USA). 2.3. Site-directed mutagenesis and purification of the GST-ERa-LBD fusion protein The H524A mutation was introduced into the GST-ERa-LBD expression vector, kindly provided by Dr. M. Brown (Boston, MA, USA) [18], using the QuickChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA) with the primers 5 0 -GGCATGGAGGCGCTGTACAGCATGAAGTGCAAGAACGTGGTGCCC3 0 and 5 0 -GGGCACCACGTTCTTGCACTTCATGCTGTACAGCG CCTCCATGCC-3 0 . The ERa-LBD-H524A coding region and its flanking cloning sites were validated by DNA sequencing as described above. GST-ERa-LBD and GST-ERa-LBD-H524A fusion proteins were purified from E. coli BL21 cells harbouring the corresponding expression vectors. Briefly, gene expression was induced by adding 1 mM isopropyl-b-D -thiogalactopyranoside (IPTG) to log-phase cultures followed by incubation for 2 h at 37 C. Cells were harvested by centrifugation and lysed by sonication on ice in a buffer containing 20 mM Tris–HCl (pH 8.0), 100 mM NaCl, 10 lM ZnCl2, 0.5% NP-40, 0.5 mM dithiothreitol, 1 mM phenyl-methyl-sulfonyl-fluoride and 0.3 lg/ml lysozyme. Cell debris was removed by centrifugation and the GST-ERa-LBD fusion proteins were purified by chromatography on a GSH-agarose column as described [10].
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548 2.4. Phage ELISA with full-length ERs and with GST-ERa-LBD The ELISAs were performed at 37 C in TBST buffer (10 mM Tris– HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween 20) [16,19]. Two ELISA formats were used. In ELISA-format I, wells of a microtitre plate coated with streptavidin [19] were incubated with biotinylated vitellogenin ERE (2 pmol per well) for 1 h followed by monomeric, fulllength ERa or ERb (2 pmol per well) for 1 h. In ELISA-format II, the streptavidin-coated wells [20] were incubated with biotinylated GSH (100 pmol per well) for 1 h followed by GST-ERa-LBD (3 pmol per well) for 1 h. In both ELISA formats, fusion phage (1010 colony forming units per well) was allowed to bind to the immobilized ER or GST-ERa-LBD, respectively, in the presence of the appropriate ER ligand as indicated for each experiment. After extensive washing, the bound fusion phage was detected with a HRP-conjugated anti-M13 monoclonal antibody and a peroxidase reaction (ortho-phenylenediamine/H2O2, Kem-En-Tek Diagnostics, Copenhagen, Denmark). The reactions were quenched after 10 min with an equal volume of 1 M H2SO4 and the colour development was measured in a microplate reader as the absorbance at 492 nm. The ligand concentrations giving a half-maximal response in the phage ELISAs, i.e., the EC50 values, were calculated by plotting the relative signals semilogarithmically versus the ligand concentrations and fitting a curve to the data on the basis of the equation: Relative signal = EC50/([ligand] + EC50), using SigmaPlot. 2.5. Cell culture and transient transfections Chinese hamster ovary cells (CHO-K1) were grown in the DMEM/F12 supplemented with 10% heat inactivated foetal calf serum, 2 mM L -glutamine, and 0.64 lg/ml garamycine in a humidified 5% CO2 atmosphere at 37 C. 7 · 103 cells were seeded in each well of 96 well plates 1 day before co-transfection with the plasmids encoding wild type full-length human ERa (generously provided by S. Green and P. Chambon, Strasbourg, France) or a mutated version of the same plasmid, in which His524 was substituted to Ala (H524A), together with the reporter plasmid pERE-sv luc, based on the pGL3-promoter vector (Promega) carrying the Xenopus laevis vitellogenin ERE upstream of the SV40 promotor and the coding region of the firefly luciferase gene [13] in a ratio of 1:10 000. Cotransfection was performed by a total of 150 ng of DNA during 5 h using FuGene6 reagent according to the manufacturerÕs protocol. The cells were harvested after 19 h and the luciferase activity analysed as described [13]. The measured luciferase activities were corrected for variations in cell density by normalisation against the protein content of each sample, determined by adding 50 lL fluorescamine (0.5 g/L) diluted in acetonitrile to each well and fluorometric measurements in the WALLAC VICTOR2 (PerkinElmer, USA) at 355/460 nm [21]. 2.6. Statistical analysis Each experiment was performed three independent times (4–6 wells for each transfection). Unless otherwise indicated, statistical analysis was done using the two-tailed StudentÕs t test with a significance level of P < 0.01.
3. Results A number of peptides, each with a specific pattern of recognition of different ER conformations, were previously isolated by Paige et al. [16] from a phage displayed peptide library. Among those, we chose 5 peptides, which allow the specific recognition of ERa and ERb conformations induced by E2, E3, OHT and ICI182 780, respectively. To investigate the interactions of peptides, which we considered optimal – a/bI (SSNHQSSRLIELLSR), a/bII (SAPRATISHYLMGG), a/bIII (SSWDMHQFFWEGVSR), aII (SSLTSRDFGSWYASR), and bII (SSLDLSQFPMTASFLRESR) – we used an ELISA with streptavidin coated onto the solid phase, followed by a layer of biotinylated vitellogenin ERE; a layer of full-length human ERa or ERb in combination with 1 lM of various ligands; a layer of fusion phage displaying one of the
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
peptides on its surface; and finally a layer of anti-phage antibody. a/bII recognized non-liganded receptors, a/bI recognized E2- and E3-complexed ERa, bII recognized E2- and E3-liganded ERb, a/bIII recognized OHT-complexed receptors, and aII recognized ERa complexed with any ligand. ICI 182 780 suppressed the a/bII binding but E3 did not do so completely (Fig. 1). These peptide recognition patterns were consistent with the data reported previously [16,19]. We were thus able to reproduce the original finding [16] that different conformations induced by different ligands can be distinguished by their peptide recognition patterns. We then screened DDE and 24 other widely distributed pesticides with the same approach to examine whether they induce ER conformational changes. ERa incubated with DDE, endosulfan, dieldrin, vinclozolin, iprodione and paclobutrazol bound a/bI, a/bIII and aII. ERb incubated with these com-
543
pounds bound bII and a/bIII. The pesticide-complexed ERs had a lower affinity to the peptides than the ERs complexed with the other ligands (Figs. 1 and 2). This peptide recognition pattern differs from the patterns induced by any other tested ligand. We will refer to the peptide recognition pattern induced by the chlorinated pesticides as the DDE-type peptide recognition pattern. Fenarimol was found to give the same peptide recognition pattern as OHT. Prochloraz was found to give the same pattern as ICI 182 780 (Fig. 2). Testing the sensitivity of the assay to different concentrations of the ligands, we found that the response increased with the ligand concentration, to reach saturation below 1 lM in all cases. A representative experiment, with DDE, is shown in Fig. 3. This conclusion is also evident from the lack of binding of the a/ bII peptide with the 1 lM pesticide concentration used routinely (Figs. 1 and 2). Thus, the variation described above
Fig. 1. Peptide recognition pattern of ERa and ERb complexed with established ligands. The binding of indicated peptides to ERa and ERb in the presence of 1 lM of the indicated ligands was estimated by ELISA format I with the peptides displayed on M13 phage and full-length ERa and ERb immobilized on the solid phase. The signals obtained with peptides a/bI and ERa in the presence of E2 were set to 100%, and the signals obtained with other peptide/ligand combinations expressed as a fraction thereof. Data are shown as means ± S.D. of at least five individual experiments.
544
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
Fig. 3. Interactions between ERa-LBD and conformation-specific peptides in the presence of various concentrations of DDE. The ERaLBD was incubated with 1–10 000 nM DDE. Binding of the conformation specific peptides was detected. The signals obtained were expressed relative to the signal obtained with the peptide a/bI and ERa LBD in the presence of E2. Data are shown as means ± S.D. of three individual experiments.
Fig. 2. Peptide recognition patterns of ERa and ERb complexed with chlorinated pesticides. For details, see legend to Fig. 1.
was not due to differential receptor saturation, but to differential affinity of the saturated receptors to the peptides. The EC50 value for each compound tested was estimated as the concen-
tration of the compound halving the signal obtained with the a/bII peptide specifically recognising the non-liganded receptor (Table 1). Seventeen other pesticides (pirimicarb, propamocarb, fenpropathrin, deltamethrin, chlorpyrifos, methiocarb, methiocarb sulfoxide, dimethoate, dichlorvos, tolchlofos-methyl, methomyl, chlorothalonil, phosetyl-aluminium, tribenuron methyl, diaminozide, chlormequat and ethephon) were found not to lead to any changes compared to the non liganded ERs in the peptide recognition pattern when tested in a concentration of 1 lM (data not shown). Taken together, our data suggest that DDE and several other pesticides induce a previously unrecognized ER conformational state. Considering the molecular structures of these ligands, we hypothesized that they interact directly with the part of the ligand binding cavity which binds the A-ring of E2. However, we would expect them to be unable to interact with His524, which interacts with the D-ring of E2, and this
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548 Table 1 The EC50 values for induction of conformational changes of ERa by various ligands Compound
EC50 (nM)
E2 OHT ICI 182 780 DDE Endosulfan Dieldrin Vinclozolin Iprodione Paclobutrazol Fenarimol Prochloraz
3.7 ± 0.4 2.8 ± 0.3 9.0 ± 2 16 ± 4 85 ± 30 21 ± 6 11 ± 4 13 ± 5 12 ± 4 30 ± 14 84 ± 32
The EC50 values were determined as the ligand concentrations halving the signal obtained with a/bII, the peptide recognising specifically the unliganded ERa. The assay was run in the ELISA set-up with ERaLBD fused to GST. Means and S.D. for three determinations are indicated.
to be one of the reasons for the different conformation and hence the different peptide recognition pattern. To test our hypothesis, we substituted His524 to Ala by site-directed mutagenesis. According to the literature, this mutation decreases the affinity of both E2 and OHT to the receptor [22,23]. In order to simplify the protein purification and to verify that the peptides recognized exactly the LBD, we used the human ERa-LBD fused to the C-terminus of GST. The construct was immobilized on biotinylated GSH bound to streptavidin coated onto the solid phase of microtitre plates, followed by incubation with 1 lM of various ligands in the presence of fusion phage displaying a/bI, a/bII, a/bIII, or aII. The peptide recognition patterns of GST-ERa-LBD wild type (wt) were identical to those of the full-length ERa (Fig. 4). However, in contrast to the wt, E2-, OHT-, and fenarimol-liganded GST-ERa-LBD, the H524A was recognized by a/bI, a/bIII and aII (Fig. 4), although the interactions were weaker than in the case of wt. Thus, with the mutant, all these compounds gave the DDE-type peptide recognition pattern. No changes in peptide recognition pattern and affinity were observed for DDE, endosulfan, dieldrin, vinclozolin, iprodion and paclobutrazol. Also, the peptide recognition patterns given by ICI 182 780 and prochloraz were not changed by the substitution although the peptide affinity was reduced (Fig. 4). Thus, it seems that the H524A mutation equalizes agonists, partial antagonists and chlorinated pesticides to all give the DDE-type peptide recognition pattern, while the pure antagonists maintain theirs. This conclusion is in agreement with the hypothesis that the lack of interaction of the respective chlorinated pesticides with His524 is critical for the DDE-type peptide recognition pattern. To support our finding with the peptides, we investigated the ability of the various pesticides to induce transactivation of vitellogenin ERE by ERa wild type and ERa H524A. CHOK1 cells were co-transfected with wild type or H524A fulllength ERa-encoding plasmids together with a plasmid with the firefly luciferase gene under transcriptional control of the vitellogenin ERE. Cells were incubated in the presence of 0, 10, 100 and 1000 nM E2, OHT, DDE and fenarimol followed by the luminometric detection of the luciferase activity. E2 and DDE, but not OHT or fenarimol, activated wild type ERadependent transcription from the vitellogenin ERE (Fig. 5). Substitution of His524 to Ala decreased the E2, but not the
545
DDE-mediated transactivation. Also, OHT and, with less affinity, fenarimol were found to induce ERa H524A transcriptional activity in DDE-like manner (Fig. 5).
4. Discussion The peptide recognition pattern of ERa and ERb in complex with DDE and the other chlorinated pesticides is the sum of the peptide recognition patterns of ERa and ERb in complex with E2 and OHT, respectively, although the pesticide-complexed receptors gave a weaker interaction with a/bI, a/bIII and aII than the E2- and OHT-complexed receptors. On the other hand, the pesticide-induced peptide recognition pattern was clearly different from the patterns induced by ICI 182 780 and E3 and the patterns of the unliganded receptors. We therefore propose that the DDE-type peptide recognition pattern reflects a conformation, which possesses features of both the E2- and the OHT-induced conformations. It was recently proposed that the ER LBD is in a state of ‘‘ligand-sensitive conformational equilibrium’’ between different conformations [9]. Applying this concept on our present data would imply that pesticidecomplexed ERs are in equilibrium between states resembling the more stable conformations induced by E2 and OHT, respectively. Since a/bI binds in the co-regulator-binding groove [16], which is occupied by H12 in the OHT-complexed receptors [6,7], one aspect of the conformational equilibrium could be H12 switching between the position over the ligandbinding pocket and the co-regulator binding groove position. However, as the binding site for a/bIII is unknown, smaller changes in the other helices may be equally important. The strategy of defining ER conformations by their peptide recognition pattern was also recently utilized by Iannone et al. [24]. In contrast to the static picture provided by a crystal structure, this analysis may provide a sense of the dynamic state of the receptor in balance between conformations with different functional properties. For example, the three-dimensional structures of ERa in complex with E2 or the phyto-oestrogen genistein, determined by X-ray crystal structure analysis, are very similar [7], but E2- and genestein-complexed ERa gave different peptide recognition patterns in our ELISA assays, where genistein reacted like E3 (data not shown). We showed that an H524A mutation in ERa LBD caused the ERa-E2 complex and the ERa-OHT complex to assume the DDE-type peptide recognition pattern. According to the above proposal of a ligand-sensitive conformational equilibrium, this observation has to be interpreted as His524 being important for the stability of E2- and OHT- induced conformations. Previous evidence has suggested that improper contacts of the D-ring of E2 and similar ligands may induce an antagonistic conformation, including positioning of H12 over the co-regulator groove [9,25]. The H524A mutation did not cause the ERa-ICI 182 780 complex to assume the same peptide recognition pattern, in agreement with the long side chain of ICI 182 780 severely restricting the conformational flexibility. Interestingly, the H524A mutation did also not change the ICI-like ERa-prochloraz peptide recognition pattern, in accordance with the observed ERa antagonistic capacity of prochloraz cell-based assays [13,26,27]. In order to support our data with peptide binding, we performed co-transfection experiments with ERa wt or ERa H524A and a reporter gene under the control of a X. laevis
546
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
Fig. 4. Peptide recognition patterns of ERa-LBD wt and ERa-LBD-H524A complexed with established ligands and chlorinated pesticides. The binding of the indicated peptides to GST-ERa-LBD wild type (h) and GST-ERa-LBD-H524A (j) in the presence of 1 lM of the indicated ligands was estimated by ELISA format II with peptides displayed on M13 phage or GST-ERa-LBD immobilized on the solid phase via biotinylated GSH and streptavidin. The signal obtained with the a/bI peptide, E2, and GST-ERa-LBD wt was set to 100%, and the signals obtained with other peptide/ ligand/GST-ERa-LBD variant combinations expressed as a fraction thereof. Data are shown as means ± S.D. of at least five individual experiments.
vitellogenin ERE. The ligand-induced trans-activation activity varied in parallel with the affinity of the receptor to a/bI, in good agreement with the observation that oestrogen stimulation of transcription from the X. laevis vitellogenin ERE in CHO-K1 cells depends on interaction with the steroid receptor co-activator protein-1 (SRC-1) [28], having the same ERa
binding motif (LXXLL) as a/bI. The co-transfection data thus confirm our in vitro peptide binding data. Due to the proposed intrinsic properties of the pesticide-complexed ERs, being in an equilibrium between agonistic and antagonistic conformations, the exact state of the equilibrium varying slightly between individual pesticides and depending on the complexes
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
547
[2] [3] [4] [5]
Fig. 5. Wild type and H524A ERa transactivation upon stimulation with different ligands. CHO-K1 cells were co-transfected with wild type or H524A full-length ERa encoding plasmid together with the pEREsv luc luciferase encoding reporter plasmids. Cells were stimulated with 10, 100 and 1000 nM E2, OHT, DDE and fenarimol, followed by the luminometric detection of the luciferase activity. All combinations of ligands and concentrations were examined with 4–6 dations in each one of 3–4 independent experiments. A BartlettÕs test showed that the variances and the means did not vary significantly from one experiment to the next. Data are therefore shown as means ± S.D. of all determinations. \: indicates that the data are significantly different (P < 0.01) from the respective control values (0 wt or 0 H524A); \\: indicates that the data obtained by the H524A plasmid are significantly different (P < 0.01) from the corresponding data obtained by wt ER plasmid, as well as from the respective control values.
[6]
[7]
[8]
[9]
being stabilized in one or another conformation by interaction with cellular co-regulators, we would expect the transcriptional response to be different in other cell types with other complements with co-regulators. In fact, another transactivation assay with site-specifically mutated ERa suggested different molecular mechanisms of the oestrogenic effects of E2 and some non-planar compounds related to DDE [29]. The agonistic and antagonistic effects of the pesticides tested here were previously studied by cell-based assays, including transactivation assays with the X. laevis vitellogenin ERE and MCF-7 proliferation assays [13]. However, those data are not directly comparable to the present ones, as a different cell line was used, which had an endogenous expression of ERs and probably a different set of co-factors exist and as higher pesticide concentrations were used in the previous study than in our study. In general, the interpretation of data obtained by in vitro and cell-based assays is complicated by several factors, including restricted penetration of the compounds into the cells; the compounds being activated or inactivated by cellular metabolism; binding to serum proteins present in the cultures. The combination of animal and cell culture-based assays with the presently described conformational mapping may resolve many of the ambiguities associated with observations with the more complex biological systems. Acknowledgements: We are grateful to Dr. A. Pike, Structural Biology Laboratory, University of York, York, UK for a critical reading of the manuscript. We thank Dr. M. Brown as well as Dr. S. Green and Dr. P. Chambon for the gift of reagents. Excellent technical assistance of T. Kru¨ger in the cell culture experiments is greatly appreciated. This work was supported by Grant No. 2055-03-0004 from the Danish Research Agency to the iNANO Centre, University of Aarhus.
[10]
[11]
[12] [13]
[14] [15] [16]
[17] [18]
[19]
References [20] [1] Nilsson, S., Makela, S., Treuter, E., Tujague, M., Thomsen, J., Andersson, G., Enmark, E., Pettersson,
K., Warner, M. and Gustafsson, J.A. (2001) Mechanisms of estrogen action. Physiol. Rev. 81, 1535–1565. McKenna, N.J. and OÕMalley, B.W. (2002) Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108, 465–474. Egea, P.F., Klaholz, B.P. and Moras, D. (2000) Ligand-protein interactions in nuclear receptors of hormones. FEBS Lett. 476, 62–67. Smith, C.L., Nawaz, Z. and OÕMalley, B.W. (1997) Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed 4-hydroxytamoxifen. Mol. Endocrinol. 6, 657–666. Lavinsky, R.M., Jepsen, K., Heinzel, T., Torchia, J., Mullen, T.M., Schiff, R., Del-Rio, A.L., Ricote, M., Ngo, S., Gemsch, J., Hilsenbeck, S.G., Osborne, C.K., Glass, C.K., Rosenfeld, M.G. and Rose, D.W. (1998) Diverse signaling pathways modulate nuclear receptor recruitment of N-CoR and SMRT complexes. Proc. Natl. Acad. Sci. USA 95, 2920–2925. Brzozowski, A.M., Pike, A.C., Dauter, Z., Hubbard, R.E., Bonn, T., Engstrom, O., Ohman, L., Greene, G.L., Gustafsson, J.A. and Carlquist, M. (1997) Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389, 753–758. Pike, A.C., Brzozowski, A.M., Hubbard, R.E., Bonn, T., Thorsell, A.G., Engstrom, O., Ljunggren, J., Gustafsson, J.A. and Carlquist, M. (1999) Structure of the ligand-binding domain of oestrogen receptor beta in the presence of a partial agonist and a full antagonist. EMBO J. 18, 4608–4618. Pike, A.C., Brzozowski, A.M., Walton, J., Hubbard, R.E., Thorsell, A.G., Li, Y.L., Gustafsson, J.A. and Carlquist, M. (2001) Structural insights into the mode of action of a pure antiestrogen. Structure 9, 145–153. Shiau, A.K., Barstad, D., Radek, J.T., Meyers, M.J., Nettles, K.W., Katzenellenbogen, B.S., Katzenellenbogen, J.A., Agard, D.A. and Greene, G.L. (2002) Structural characterization of a subtype-selective ligand reveals a novel mode of estrogen receptor antagonism. Nat. Struct. Biol. 9, 359–364. Matthews, J. and Zacharewski, T. (2000) Differential binding affinities of PCBs, HO-PCBs, and aroclors with recombinant human, rainbow trout (Onchorhynkiss mykiss), and green anole (Anolis carolinensis) estrogen receptors, using a semi-high throughput competitive binding assay. Toxicol. Sci. 53, 326–339. Scippo, M.L., Argiris, C., Van De Weerdt, C., Muller, M., Willemsen, P., Martial, J. and Maghuin-Rogister, G. (2004) Recombinant human estrogen, androgen and progesterone receptors for detection of potential endocrine disruptors. Anal. Bioanal. Chem. 378, 664–669. Katzenellenbogen, J.A. (1995) The structural pervasiveness of estrogenic activity. Environ. Health Perspect. 103, 99–101. Andersen, H.R., Vinggaard, A.M., Rasmussen, T.H., Gjermandsen, I.M. and Bonefeld-Jørgensen, E.C. (2002) Effects of currently used pesticides in assays for estrogenicity, androgenicity, and aromatase activity in vitro. Toxicol. Appl. Pharmacol. 179, 1–12. Dees, C., Askari, M., Foster, J.S., Ahamed, S. and Wimalasena, J. (1997) DDT mimicks estradiol stimulation of breast cancer cells to enter the cell cycle. Mol. Carcinog. 18, 107–114. Matthews, J., Celius, T., Halgren, R. and Zacharewski, T. (2000) Differential estrogen receptor binding of estrogenic substances: a species comparison. J. Steroid Biochem. Mol. Biol. 74, 223–234. Paige, L.A., Christensen, D.J., Grøn, H., Norris, J.D., Gottlin, E.B., Padilla, K.M., Chang, C.Y., Ballas, L.M., Hamilton, P.T., McDonnell, D.P. and Fowlkes, D.M. (1999) Estrogen receptor (ER) modulators each induce distinct conformational changes in ER alpha and ER beta. Proc. Natl. Acad. Sci. USA 96, 3999–4004. Scott, J.K. and Smith, G.P. (1990) Searching for peptide ligands with an epitope library. Science 249, 386–390. Halachmi, S., Marden, E., Martin, G., MacKay, H., Abbondanza, C. and Brown, M. (1994) Estrogen receptor-associated proteins: possible mediators of hormone-induced transcription. Science 264, 1455–1458. Norris, J.D., Paige, L.A., Christensen, D.J., Chang, C.Y., Huacani, M.R., Fan, D., Hamilton, P.T., Fowlkes, D.M. and McDonnell, D.P. (1999) Peptide antagonists of the human estrogen receptor. Science 285, 744–746. Valimaa, L., Pettersson, K., Vehniainen, M., Karp, M. and Lovgren, T. (2003) A high-capacity streptavidin-coated microtitration plate. Bioconjug. Chem. 14, 103–111.
548 [21] Bohlen, P., Stein, S., Imai, K. and Udenfriend, S. (1974) A simplified protein assay with fluorescamine in samples containing interfering material. Anal. Biochem. 74, 559–562. [22] Ekena, K., Weis, K.E., Katzenellenbogen, J.A. and Katzenellenbogen, B.S. (1996) Identification of amino acids in the hormone binding domain of the human estrogen receptor important in estrogen binding. J. Biol. Chem. 271, 20053–20059. [23] Aliau, S., Mattras, H., Richard, E., Bonnafous, J.C. and Borgna, J.L. (2002) Differential interactions of estrogens and antiestrogens at the 17 beta-hydroxyl or counterpart hydroxyl with histidine 524 of the human estrogen receptor alpha. Biochemistry 41, 7979– 7988. [24] Iannone, M.A., Simmons, C.A., Kadwell, S.H., Svoboda, D.L., Vanderwall, D.E., Deng, S.J., Consler, T.G., Shearin, J., Gray, J.G. and Pearce, K.H. (2004) Correlation between in vitro peptide binding profiles and cellular activities for estrogen receptormodulating compounds. Mol. Endocrinol. 18, 1064–1081. [25] Gangloff, M., Ruff, M., Eiler, S., Duclaud, S., Wurtz, J.M. and Moras, D. (2001) Crystal structure of a mutant hERalpha ligand-
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
[26]
[27] [28]
[29]
binding domain reveals key structural features for the mechanism of partial agonism. J. Biol. Chem. 276, 15059–15065. Hofmeister, M.V. and Bonefeld-Jørgensen, E.C. (2004) Effects of the pesticides prochloraz and metiocarb on humane Estrogen receptor a and b mRNA levels analyzed by on-line RT-PCR analyses. Toxicology in vitro 18, 427–433. Gru¨nfeld, H.T. and Bonefeld-Jørgensen, E.C. (2004) Effect of in vitro estrogenic pesticides on human estrogen receptor and b mRNA levels. Toxicology Letters 151, 467–480. Metivier, R., Penot, G., Flouriot, G. and Pakdel, F. (2001) Synergism between ERalpha transactivation function 1 (AF-1) and AF-2 mediated by steroid receptor coactivator protein-1: requirement for the AF-1 alpha-helical core and for a direct interaction between the N- and C-terminal domains. Mol. Endocrinol. 15, 1953–1970. Bentrem, D., Fox, J.E., Pearce, S.T., Liu, H., Pappas, S., Kupfer, D., Zapf, J.W. and Jordan, V.C. (2003) Distinct molecular conformations of the estrogen receptor alpha complex exploited by environmental estrogens. Cancer Res. 63, 7490–7496.
FEBS Letters 579 (2005) 541–548
A novel pesticide-induced conformational state of the oestrogen receptor ligand-binding domain, detected by conformation-specific peptide binding Vadim V. Sumbayeva,*, Eva C. Bonefeld-Jørgensenb, Troels Winda, Peter A. Andreasena a
Laboratory of Cellular Protein Science, Department of Molecular Biology and The Interdisciplinary Nanoscience Centre, University of Aarhus, Aarhus, Denmark b Unit of Environmental Biotechnology, Department of Environmental Medicine, Aarhus University, Aarhus, Denmark Received 25 November 2004; revised 13 December 2004; accepted 13 December 2004 Available online 22 December 2004 Edited by Barry Halliwell
Abstract The diverse effects of different natural and synthetic oestrogen receptor ligands depend on induction of different receptor conformations, allowing differential interactions with other transcription factors. Different conformations of the oestrogen receptor ligand binding domains can be monitored by conformation-specific binding to peptides selected from phage-displayed peptide libraries. We now report that a group of chlorinated pesticides, including 2,4-dichlorodiphenyl-dichloroethylene, induces a peptide recognition pattern different from those induced by any one of the classical oestrogen receptor ligands. The pesticide-complexed oestrogen receptors recognized peptides reacting with the receptors complexed both with the natural oestrogen 17b-oestradiol and with the synthetic partial antagonist 4-hydroxy-tamoxifen, respectively, indicating that the pesticide-induced conformation shares features with both the 17b-oestradiol- and the 4-hydroxy-tamoxifen-induced conformations. The substitution H524A in the ligand binding domain conferred the pesticide-specific peptide recognition pattern and transactivation activity to the oestradiol- and the 4-hydroxy-tamoxifen-complexed receptors, indicating that one important determinant of the pesticide-induced conformation is a lack of stabilisation of any one particular receptor conformation by ligand interaction with H524, which is known to interact with both oestradiol and 4-hydroxy-tamoxifen. Thus, peptide binding analyses of oestrogen receptor conformations induced by environmental endocrine disruptors can give novel information about molecular mechanisms of oestrogen action in general. 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Oestrogen receptor; Oestrogen; Endocrine disruptor; Phage-displayed peptide library
1. Introduction The physiological effects of oestrogens are mediated by the oestrogen receptors (ERs), ERa and ERb, members of the nuclear receptor family of ligand-regulated transcription factors. Hormone binding to the ERs ligand-binding domain (LBD) initiates a series of molecular events culminating in the activation or repression of target genes [1,2]. When oestradiol (E2), * Corresponding author. Fax: +45 8612 3178. E-mail address: [email protected] (V.V. Sumbayev).
the natural oestrogen, binds to ERs, the phenolic hydroxyl of its A-ring nestles between a-helix (H) 3 and H6 of the receptor and makes direct hydrogen bonds to the carboxylate of Glu353 and the guanidinium group of Arg394 (using the ERa residue numbering system). The 17b hydroxyl of oestradiolÕs D-ring makes a hydrogen bond with His524 in H11. Transcriptional regulation arises from direct interaction of ligand-occupied ERs with its cognate DNA target site (oestrogen responsive elements, EREs) in the regulatory regions of oestrogen-sensitive genes [3–5]. Moreover, conformational changes of the LBD induced by ligand binding confer further specificity upon the hormonal response and is a key event permitting discrimination between agonists (E2, oestriol (E3)), partial antagonists (4-hydroxytamoxifen (OHT)), and pure antagonists (ICI 182 780). The most prominent conformational change is a ligand-mediated repositioning of H12. In the agonist-liganded receptor, H12 lies over the ligand-binding cavity and forms, together with H3 and H5, a binding surface for transcriptional co-regulators. In receptor liganded by partial antagonists, H12 occupies the groove between H3 and H5 and prevents co-regulator binding. With pure antagonists, H12 is pushed away and is invisible in X-ray crystal structure analysis. Different positionings of H12 and associated changes in other helices thus result in different co-regulator binding surfaces and hence underlie the different physiological effects of agonists and antagonists [6–9]. In recent decades, there has been an increasing concern about environmental chemicals that may interfere with oestrogenic signalling and adversely affect normal reproduction of humans and wildlife. These compounds, known as environmental endocrine disruptors, encompass a wide range of substances including natural products, pesticides, pharmaceuticals and industrial chemicals [10–12]. Some of those chemicals have shown either ER agonistic or antagonistic properties in cell proliferation and transactivation assays with the breast carcinoma cell line MCF-7 [13]. Many of them, especially a number of widely distributed pesticides, do not share any obvious structural similarity to natural oestrogens, making their identification, based solely on molecular structure, impossible [10,14,15]. In the present study, we have investigated oestrogen receptor conformational changes mediated by 2,4-dichlorodiphenyldichloroethylene (DDE), a metabolite of 2,4-dichlorodiphenyl-trichloroethane (DDT), and other pesticides. For the study, we have utilized the ability of peptides isolated from
0014-5793/$22.00 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2004.12.025
542
phage-displayed peptide libraries to recognise specific ER conformations [16]. We have found that a group of chlorinated pesticides induces a ER peptide recognition pattern not described before and hence an ER conformational state, which is not shared with any of the classical ER ligands. Moreover, we demonstrate by site-directed mutagenesis in phage displayed peptide binding and transactivation assays that His524 is one important determinant for the ligand-induced conformational changes. Our studies give novel important information about ER conformational changes and the mechanisms by which endocrine disruptors may interfere with oestrogen action.
2. Materials and methods 2.1. Materials The following materials were purchased from the indicated suppliers: streptavidin (Calbiochem, San Diego, CA, USA); ERa and ERb (Panvera, Madison, WI, USA); DNA cassettes, primers, and biotinylated oligonucleotides corresponding to vitellogenin ERE (biotin-GATCTAGGTCACAGTGACCTGCG – forward and biotin-GATCCGC AGGTCACTGTGACCTA – reverse) (DNA Technology, Aarhus, Denmark); E2, E3, and OHT (Sigma, St. Louis, MO, USA); DDE and other pesticides (Dr. Ehrenstorfer, Augsburg, Germany); mouse anti-M13 monoclonal antibody conjugated with horseradish peroxidase (HRP) and anti-glutathione-S-transferase (GST) goat antibody (Amersham Biosciences, Hillerød, Denmark); glutathione-(GSH) agarose (Molecular Probes, Eugene, OR, USA); biotinylated GSH (Biopeptide, San-Diego, CA, USA); MaxisorpTM microtitre plates (Nunc, Roskilde, Denmark); E. coli BL21 (Stratagene, La Jolla, CA, USA); and FuGene6 (Roche Diagnostics Corp., Indianapolis, IN, USA). All other chemicals were of the best grade commercially available. 2.2. Fusion phage M13 phage particles displaying ER-binding peptides were prepared by ligating oligonucleotide cassettes coding for each peptide into SfiIdigested fUSE5 vector followed by production of fusion phage in E. coli DH-5a cells [17]. The created fusion phage was validated by DNA sequencing of PCR products generated with the primers M13for (5 0 -TTTCGACACAATTTATCAGG-3 0 ) and M13back (5 0 TGAATTTTCTGTATGAGGTTTTG-3 0 ) and Taq DNA polymerase (Invitrogen, Denmark). The PCR-products were purified (Qiaquick PCR Purification Kit, Qiagen, Germany) and sequenced with the DYEnamic ET Terminator Cycle Sequencing Kit (Amersham Biosciences, Hillerød, Denmark) and an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, CA, USA). 2.3. Site-directed mutagenesis and purification of the GST-ERa-LBD fusion protein The H524A mutation was introduced into the GST-ERa-LBD expression vector, kindly provided by Dr. M. Brown (Boston, MA, USA) [18], using the QuickChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA) with the primers 5 0 -GGCATGGAGGCGCTGTACAGCATGAAGTGCAAGAACGTGGTGCCC3 0 and 5 0 -GGGCACCACGTTCTTGCACTTCATGCTGTACAGCG CCTCCATGCC-3 0 . The ERa-LBD-H524A coding region and its flanking cloning sites were validated by DNA sequencing as described above. GST-ERa-LBD and GST-ERa-LBD-H524A fusion proteins were purified from E. coli BL21 cells harbouring the corresponding expression vectors. Briefly, gene expression was induced by adding 1 mM isopropyl-b-D -thiogalactopyranoside (IPTG) to log-phase cultures followed by incubation for 2 h at 37 C. Cells were harvested by centrifugation and lysed by sonication on ice in a buffer containing 20 mM Tris–HCl (pH 8.0), 100 mM NaCl, 10 lM ZnCl2, 0.5% NP-40, 0.5 mM dithiothreitol, 1 mM phenyl-methyl-sulfonyl-fluoride and 0.3 lg/ml lysozyme. Cell debris was removed by centrifugation and the GST-ERa-LBD fusion proteins were purified by chromatography on a GSH-agarose column as described [10].
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548 2.4. Phage ELISA with full-length ERs and with GST-ERa-LBD The ELISAs were performed at 37 C in TBST buffer (10 mM Tris– HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween 20) [16,19]. Two ELISA formats were used. In ELISA-format I, wells of a microtitre plate coated with streptavidin [19] were incubated with biotinylated vitellogenin ERE (2 pmol per well) for 1 h followed by monomeric, fulllength ERa or ERb (2 pmol per well) for 1 h. In ELISA-format II, the streptavidin-coated wells [20] were incubated with biotinylated GSH (100 pmol per well) for 1 h followed by GST-ERa-LBD (3 pmol per well) for 1 h. In both ELISA formats, fusion phage (1010 colony forming units per well) was allowed to bind to the immobilized ER or GST-ERa-LBD, respectively, in the presence of the appropriate ER ligand as indicated for each experiment. After extensive washing, the bound fusion phage was detected with a HRP-conjugated anti-M13 monoclonal antibody and a peroxidase reaction (ortho-phenylenediamine/H2O2, Kem-En-Tek Diagnostics, Copenhagen, Denmark). The reactions were quenched after 10 min with an equal volume of 1 M H2SO4 and the colour development was measured in a microplate reader as the absorbance at 492 nm. The ligand concentrations giving a half-maximal response in the phage ELISAs, i.e., the EC50 values, were calculated by plotting the relative signals semilogarithmically versus the ligand concentrations and fitting a curve to the data on the basis of the equation: Relative signal = EC50/([ligand] + EC50), using SigmaPlot. 2.5. Cell culture and transient transfections Chinese hamster ovary cells (CHO-K1) were grown in the DMEM/F12 supplemented with 10% heat inactivated foetal calf serum, 2 mM L -glutamine, and 0.64 lg/ml garamycine in a humidified 5% CO2 atmosphere at 37 C. 7 · 103 cells were seeded in each well of 96 well plates 1 day before co-transfection with the plasmids encoding wild type full-length human ERa (generously provided by S. Green and P. Chambon, Strasbourg, France) or a mutated version of the same plasmid, in which His524 was substituted to Ala (H524A), together with the reporter plasmid pERE-sv luc, based on the pGL3-promoter vector (Promega) carrying the Xenopus laevis vitellogenin ERE upstream of the SV40 promotor and the coding region of the firefly luciferase gene [13] in a ratio of 1:10 000. Cotransfection was performed by a total of 150 ng of DNA during 5 h using FuGene6 reagent according to the manufacturerÕs protocol. The cells were harvested after 19 h and the luciferase activity analysed as described [13]. The measured luciferase activities were corrected for variations in cell density by normalisation against the protein content of each sample, determined by adding 50 lL fluorescamine (0.5 g/L) diluted in acetonitrile to each well and fluorometric measurements in the WALLAC VICTOR2 (PerkinElmer, USA) at 355/460 nm [21]. 2.6. Statistical analysis Each experiment was performed three independent times (4–6 wells for each transfection). Unless otherwise indicated, statistical analysis was done using the two-tailed StudentÕs t test with a significance level of P < 0.01.
3. Results A number of peptides, each with a specific pattern of recognition of different ER conformations, were previously isolated by Paige et al. [16] from a phage displayed peptide library. Among those, we chose 5 peptides, which allow the specific recognition of ERa and ERb conformations induced by E2, E3, OHT and ICI182 780, respectively. To investigate the interactions of peptides, which we considered optimal – a/bI (SSNHQSSRLIELLSR), a/bII (SAPRATISHYLMGG), a/bIII (SSWDMHQFFWEGVSR), aII (SSLTSRDFGSWYASR), and bII (SSLDLSQFPMTASFLRESR) – we used an ELISA with streptavidin coated onto the solid phase, followed by a layer of biotinylated vitellogenin ERE; a layer of full-length human ERa or ERb in combination with 1 lM of various ligands; a layer of fusion phage displaying one of the
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
peptides on its surface; and finally a layer of anti-phage antibody. a/bII recognized non-liganded receptors, a/bI recognized E2- and E3-complexed ERa, bII recognized E2- and E3-liganded ERb, a/bIII recognized OHT-complexed receptors, and aII recognized ERa complexed with any ligand. ICI 182 780 suppressed the a/bII binding but E3 did not do so completely (Fig. 1). These peptide recognition patterns were consistent with the data reported previously [16,19]. We were thus able to reproduce the original finding [16] that different conformations induced by different ligands can be distinguished by their peptide recognition patterns. We then screened DDE and 24 other widely distributed pesticides with the same approach to examine whether they induce ER conformational changes. ERa incubated with DDE, endosulfan, dieldrin, vinclozolin, iprodione and paclobutrazol bound a/bI, a/bIII and aII. ERb incubated with these com-
543
pounds bound bII and a/bIII. The pesticide-complexed ERs had a lower affinity to the peptides than the ERs complexed with the other ligands (Figs. 1 and 2). This peptide recognition pattern differs from the patterns induced by any other tested ligand. We will refer to the peptide recognition pattern induced by the chlorinated pesticides as the DDE-type peptide recognition pattern. Fenarimol was found to give the same peptide recognition pattern as OHT. Prochloraz was found to give the same pattern as ICI 182 780 (Fig. 2). Testing the sensitivity of the assay to different concentrations of the ligands, we found that the response increased with the ligand concentration, to reach saturation below 1 lM in all cases. A representative experiment, with DDE, is shown in Fig. 3. This conclusion is also evident from the lack of binding of the a/ bII peptide with the 1 lM pesticide concentration used routinely (Figs. 1 and 2). Thus, the variation described above
Fig. 1. Peptide recognition pattern of ERa and ERb complexed with established ligands. The binding of indicated peptides to ERa and ERb in the presence of 1 lM of the indicated ligands was estimated by ELISA format I with the peptides displayed on M13 phage and full-length ERa and ERb immobilized on the solid phase. The signals obtained with peptides a/bI and ERa in the presence of E2 were set to 100%, and the signals obtained with other peptide/ligand combinations expressed as a fraction thereof. Data are shown as means ± S.D. of at least five individual experiments.
544
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
Fig. 3. Interactions between ERa-LBD and conformation-specific peptides in the presence of various concentrations of DDE. The ERaLBD was incubated with 1–10 000 nM DDE. Binding of the conformation specific peptides was detected. The signals obtained were expressed relative to the signal obtained with the peptide a/bI and ERa LBD in the presence of E2. Data are shown as means ± S.D. of three individual experiments.
Fig. 2. Peptide recognition patterns of ERa and ERb complexed with chlorinated pesticides. For details, see legend to Fig. 1.
was not due to differential receptor saturation, but to differential affinity of the saturated receptors to the peptides. The EC50 value for each compound tested was estimated as the concen-
tration of the compound halving the signal obtained with the a/bII peptide specifically recognising the non-liganded receptor (Table 1). Seventeen other pesticides (pirimicarb, propamocarb, fenpropathrin, deltamethrin, chlorpyrifos, methiocarb, methiocarb sulfoxide, dimethoate, dichlorvos, tolchlofos-methyl, methomyl, chlorothalonil, phosetyl-aluminium, tribenuron methyl, diaminozide, chlormequat and ethephon) were found not to lead to any changes compared to the non liganded ERs in the peptide recognition pattern when tested in a concentration of 1 lM (data not shown). Taken together, our data suggest that DDE and several other pesticides induce a previously unrecognized ER conformational state. Considering the molecular structures of these ligands, we hypothesized that they interact directly with the part of the ligand binding cavity which binds the A-ring of E2. However, we would expect them to be unable to interact with His524, which interacts with the D-ring of E2, and this
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548 Table 1 The EC50 values for induction of conformational changes of ERa by various ligands Compound
EC50 (nM)
E2 OHT ICI 182 780 DDE Endosulfan Dieldrin Vinclozolin Iprodione Paclobutrazol Fenarimol Prochloraz
3.7 ± 0.4 2.8 ± 0.3 9.0 ± 2 16 ± 4 85 ± 30 21 ± 6 11 ± 4 13 ± 5 12 ± 4 30 ± 14 84 ± 32
The EC50 values were determined as the ligand concentrations halving the signal obtained with a/bII, the peptide recognising specifically the unliganded ERa. The assay was run in the ELISA set-up with ERaLBD fused to GST. Means and S.D. for three determinations are indicated.
to be one of the reasons for the different conformation and hence the different peptide recognition pattern. To test our hypothesis, we substituted His524 to Ala by site-directed mutagenesis. According to the literature, this mutation decreases the affinity of both E2 and OHT to the receptor [22,23]. In order to simplify the protein purification and to verify that the peptides recognized exactly the LBD, we used the human ERa-LBD fused to the C-terminus of GST. The construct was immobilized on biotinylated GSH bound to streptavidin coated onto the solid phase of microtitre plates, followed by incubation with 1 lM of various ligands in the presence of fusion phage displaying a/bI, a/bII, a/bIII, or aII. The peptide recognition patterns of GST-ERa-LBD wild type (wt) were identical to those of the full-length ERa (Fig. 4). However, in contrast to the wt, E2-, OHT-, and fenarimol-liganded GST-ERa-LBD, the H524A was recognized by a/bI, a/bIII and aII (Fig. 4), although the interactions were weaker than in the case of wt. Thus, with the mutant, all these compounds gave the DDE-type peptide recognition pattern. No changes in peptide recognition pattern and affinity were observed for DDE, endosulfan, dieldrin, vinclozolin, iprodion and paclobutrazol. Also, the peptide recognition patterns given by ICI 182 780 and prochloraz were not changed by the substitution although the peptide affinity was reduced (Fig. 4). Thus, it seems that the H524A mutation equalizes agonists, partial antagonists and chlorinated pesticides to all give the DDE-type peptide recognition pattern, while the pure antagonists maintain theirs. This conclusion is in agreement with the hypothesis that the lack of interaction of the respective chlorinated pesticides with His524 is critical for the DDE-type peptide recognition pattern. To support our finding with the peptides, we investigated the ability of the various pesticides to induce transactivation of vitellogenin ERE by ERa wild type and ERa H524A. CHOK1 cells were co-transfected with wild type or H524A fulllength ERa-encoding plasmids together with a plasmid with the firefly luciferase gene under transcriptional control of the vitellogenin ERE. Cells were incubated in the presence of 0, 10, 100 and 1000 nM E2, OHT, DDE and fenarimol followed by the luminometric detection of the luciferase activity. E2 and DDE, but not OHT or fenarimol, activated wild type ERadependent transcription from the vitellogenin ERE (Fig. 5). Substitution of His524 to Ala decreased the E2, but not the
545
DDE-mediated transactivation. Also, OHT and, with less affinity, fenarimol were found to induce ERa H524A transcriptional activity in DDE-like manner (Fig. 5).
4. Discussion The peptide recognition pattern of ERa and ERb in complex with DDE and the other chlorinated pesticides is the sum of the peptide recognition patterns of ERa and ERb in complex with E2 and OHT, respectively, although the pesticide-complexed receptors gave a weaker interaction with a/bI, a/bIII and aII than the E2- and OHT-complexed receptors. On the other hand, the pesticide-induced peptide recognition pattern was clearly different from the patterns induced by ICI 182 780 and E3 and the patterns of the unliganded receptors. We therefore propose that the DDE-type peptide recognition pattern reflects a conformation, which possesses features of both the E2- and the OHT-induced conformations. It was recently proposed that the ER LBD is in a state of ‘‘ligand-sensitive conformational equilibrium’’ between different conformations [9]. Applying this concept on our present data would imply that pesticidecomplexed ERs are in equilibrium between states resembling the more stable conformations induced by E2 and OHT, respectively. Since a/bI binds in the co-regulator-binding groove [16], which is occupied by H12 in the OHT-complexed receptors [6,7], one aspect of the conformational equilibrium could be H12 switching between the position over the ligandbinding pocket and the co-regulator binding groove position. However, as the binding site for a/bIII is unknown, smaller changes in the other helices may be equally important. The strategy of defining ER conformations by their peptide recognition pattern was also recently utilized by Iannone et al. [24]. In contrast to the static picture provided by a crystal structure, this analysis may provide a sense of the dynamic state of the receptor in balance between conformations with different functional properties. For example, the three-dimensional structures of ERa in complex with E2 or the phyto-oestrogen genistein, determined by X-ray crystal structure analysis, are very similar [7], but E2- and genestein-complexed ERa gave different peptide recognition patterns in our ELISA assays, where genistein reacted like E3 (data not shown). We showed that an H524A mutation in ERa LBD caused the ERa-E2 complex and the ERa-OHT complex to assume the DDE-type peptide recognition pattern. According to the above proposal of a ligand-sensitive conformational equilibrium, this observation has to be interpreted as His524 being important for the stability of E2- and OHT- induced conformations. Previous evidence has suggested that improper contacts of the D-ring of E2 and similar ligands may induce an antagonistic conformation, including positioning of H12 over the co-regulator groove [9,25]. The H524A mutation did not cause the ERa-ICI 182 780 complex to assume the same peptide recognition pattern, in agreement with the long side chain of ICI 182 780 severely restricting the conformational flexibility. Interestingly, the H524A mutation did also not change the ICI-like ERa-prochloraz peptide recognition pattern, in accordance with the observed ERa antagonistic capacity of prochloraz cell-based assays [13,26,27]. In order to support our data with peptide binding, we performed co-transfection experiments with ERa wt or ERa H524A and a reporter gene under the control of a X. laevis
546
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
Fig. 4. Peptide recognition patterns of ERa-LBD wt and ERa-LBD-H524A complexed with established ligands and chlorinated pesticides. The binding of the indicated peptides to GST-ERa-LBD wild type (h) and GST-ERa-LBD-H524A (j) in the presence of 1 lM of the indicated ligands was estimated by ELISA format II with peptides displayed on M13 phage or GST-ERa-LBD immobilized on the solid phase via biotinylated GSH and streptavidin. The signal obtained with the a/bI peptide, E2, and GST-ERa-LBD wt was set to 100%, and the signals obtained with other peptide/ ligand/GST-ERa-LBD variant combinations expressed as a fraction thereof. Data are shown as means ± S.D. of at least five individual experiments.
vitellogenin ERE. The ligand-induced trans-activation activity varied in parallel with the affinity of the receptor to a/bI, in good agreement with the observation that oestrogen stimulation of transcription from the X. laevis vitellogenin ERE in CHO-K1 cells depends on interaction with the steroid receptor co-activator protein-1 (SRC-1) [28], having the same ERa
binding motif (LXXLL) as a/bI. The co-transfection data thus confirm our in vitro peptide binding data. Due to the proposed intrinsic properties of the pesticide-complexed ERs, being in an equilibrium between agonistic and antagonistic conformations, the exact state of the equilibrium varying slightly between individual pesticides and depending on the complexes
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
547
[2] [3] [4] [5]
Fig. 5. Wild type and H524A ERa transactivation upon stimulation with different ligands. CHO-K1 cells were co-transfected with wild type or H524A full-length ERa encoding plasmid together with the pEREsv luc luciferase encoding reporter plasmids. Cells were stimulated with 10, 100 and 1000 nM E2, OHT, DDE and fenarimol, followed by the luminometric detection of the luciferase activity. All combinations of ligands and concentrations were examined with 4–6 dations in each one of 3–4 independent experiments. A BartlettÕs test showed that the variances and the means did not vary significantly from one experiment to the next. Data are therefore shown as means ± S.D. of all determinations. \: indicates that the data are significantly different (P < 0.01) from the respective control values (0 wt or 0 H524A); \\: indicates that the data obtained by the H524A plasmid are significantly different (P < 0.01) from the corresponding data obtained by wt ER plasmid, as well as from the respective control values.
[6]
[7]
[8]
[9]
being stabilized in one or another conformation by interaction with cellular co-regulators, we would expect the transcriptional response to be different in other cell types with other complements with co-regulators. In fact, another transactivation assay with site-specifically mutated ERa suggested different molecular mechanisms of the oestrogenic effects of E2 and some non-planar compounds related to DDE [29]. The agonistic and antagonistic effects of the pesticides tested here were previously studied by cell-based assays, including transactivation assays with the X. laevis vitellogenin ERE and MCF-7 proliferation assays [13]. However, those data are not directly comparable to the present ones, as a different cell line was used, which had an endogenous expression of ERs and probably a different set of co-factors exist and as higher pesticide concentrations were used in the previous study than in our study. In general, the interpretation of data obtained by in vitro and cell-based assays is complicated by several factors, including restricted penetration of the compounds into the cells; the compounds being activated or inactivated by cellular metabolism; binding to serum proteins present in the cultures. The combination of animal and cell culture-based assays with the presently described conformational mapping may resolve many of the ambiguities associated with observations with the more complex biological systems. Acknowledgements: We are grateful to Dr. A. Pike, Structural Biology Laboratory, University of York, York, UK for a critical reading of the manuscript. We thank Dr. M. Brown as well as Dr. S. Green and Dr. P. Chambon for the gift of reagents. Excellent technical assistance of T. Kru¨ger in the cell culture experiments is greatly appreciated. This work was supported by Grant No. 2055-03-0004 from the Danish Research Agency to the iNANO Centre, University of Aarhus.
[10]
[11]
[12] [13]
[14] [15] [16]
[17] [18]
[19]
References [20] [1] Nilsson, S., Makela, S., Treuter, E., Tujague, M., Thomsen, J., Andersson, G., Enmark, E., Pettersson,
K., Warner, M. and Gustafsson, J.A. (2001) Mechanisms of estrogen action. Physiol. Rev. 81, 1535–1565. McKenna, N.J. and OÕMalley, B.W. (2002) Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108, 465–474. Egea, P.F., Klaholz, B.P. and Moras, D. (2000) Ligand-protein interactions in nuclear receptors of hormones. FEBS Lett. 476, 62–67. Smith, C.L., Nawaz, Z. and OÕMalley, B.W. (1997) Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed 4-hydroxytamoxifen. Mol. Endocrinol. 6, 657–666. Lavinsky, R.M., Jepsen, K., Heinzel, T., Torchia, J., Mullen, T.M., Schiff, R., Del-Rio, A.L., Ricote, M., Ngo, S., Gemsch, J., Hilsenbeck, S.G., Osborne, C.K., Glass, C.K., Rosenfeld, M.G. and Rose, D.W. (1998) Diverse signaling pathways modulate nuclear receptor recruitment of N-CoR and SMRT complexes. Proc. Natl. Acad. Sci. USA 95, 2920–2925. Brzozowski, A.M., Pike, A.C., Dauter, Z., Hubbard, R.E., Bonn, T., Engstrom, O., Ohman, L., Greene, G.L., Gustafsson, J.A. and Carlquist, M. (1997) Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389, 753–758. Pike, A.C., Brzozowski, A.M., Hubbard, R.E., Bonn, T., Thorsell, A.G., Engstrom, O., Ljunggren, J., Gustafsson, J.A. and Carlquist, M. (1999) Structure of the ligand-binding domain of oestrogen receptor beta in the presence of a partial agonist and a full antagonist. EMBO J. 18, 4608–4618. Pike, A.C., Brzozowski, A.M., Walton, J., Hubbard, R.E., Thorsell, A.G., Li, Y.L., Gustafsson, J.A. and Carlquist, M. (2001) Structural insights into the mode of action of a pure antiestrogen. Structure 9, 145–153. Shiau, A.K., Barstad, D., Radek, J.T., Meyers, M.J., Nettles, K.W., Katzenellenbogen, B.S., Katzenellenbogen, J.A., Agard, D.A. and Greene, G.L. (2002) Structural characterization of a subtype-selective ligand reveals a novel mode of estrogen receptor antagonism. Nat. Struct. Biol. 9, 359–364. Matthews, J. and Zacharewski, T. (2000) Differential binding affinities of PCBs, HO-PCBs, and aroclors with recombinant human, rainbow trout (Onchorhynkiss mykiss), and green anole (Anolis carolinensis) estrogen receptors, using a semi-high throughput competitive binding assay. Toxicol. Sci. 53, 326–339. Scippo, M.L., Argiris, C., Van De Weerdt, C., Muller, M., Willemsen, P., Martial, J. and Maghuin-Rogister, G. (2004) Recombinant human estrogen, androgen and progesterone receptors for detection of potential endocrine disruptors. Anal. Bioanal. Chem. 378, 664–669. Katzenellenbogen, J.A. (1995) The structural pervasiveness of estrogenic activity. Environ. Health Perspect. 103, 99–101. Andersen, H.R., Vinggaard, A.M., Rasmussen, T.H., Gjermandsen, I.M. and Bonefeld-Jørgensen, E.C. (2002) Effects of currently used pesticides in assays for estrogenicity, androgenicity, and aromatase activity in vitro. Toxicol. Appl. Pharmacol. 179, 1–12. Dees, C., Askari, M., Foster, J.S., Ahamed, S. and Wimalasena, J. (1997) DDT mimicks estradiol stimulation of breast cancer cells to enter the cell cycle. Mol. Carcinog. 18, 107–114. Matthews, J., Celius, T., Halgren, R. and Zacharewski, T. (2000) Differential estrogen receptor binding of estrogenic substances: a species comparison. J. Steroid Biochem. Mol. Biol. 74, 223–234. Paige, L.A., Christensen, D.J., Grøn, H., Norris, J.D., Gottlin, E.B., Padilla, K.M., Chang, C.Y., Ballas, L.M., Hamilton, P.T., McDonnell, D.P. and Fowlkes, D.M. (1999) Estrogen receptor (ER) modulators each induce distinct conformational changes in ER alpha and ER beta. Proc. Natl. Acad. Sci. USA 96, 3999–4004. Scott, J.K. and Smith, G.P. (1990) Searching for peptide ligands with an epitope library. Science 249, 386–390. Halachmi, S., Marden, E., Martin, G., MacKay, H., Abbondanza, C. and Brown, M. (1994) Estrogen receptor-associated proteins: possible mediators of hormone-induced transcription. Science 264, 1455–1458. Norris, J.D., Paige, L.A., Christensen, D.J., Chang, C.Y., Huacani, M.R., Fan, D., Hamilton, P.T., Fowlkes, D.M. and McDonnell, D.P. (1999) Peptide antagonists of the human estrogen receptor. Science 285, 744–746. Valimaa, L., Pettersson, K., Vehniainen, M., Karp, M. and Lovgren, T. (2003) A high-capacity streptavidin-coated microtitration plate. Bioconjug. Chem. 14, 103–111.
548 [21] Bohlen, P., Stein, S., Imai, K. and Udenfriend, S. (1974) A simplified protein assay with fluorescamine in samples containing interfering material. Anal. Biochem. 74, 559–562. [22] Ekena, K., Weis, K.E., Katzenellenbogen, J.A. and Katzenellenbogen, B.S. (1996) Identification of amino acids in the hormone binding domain of the human estrogen receptor important in estrogen binding. J. Biol. Chem. 271, 20053–20059. [23] Aliau, S., Mattras, H., Richard, E., Bonnafous, J.C. and Borgna, J.L. (2002) Differential interactions of estrogens and antiestrogens at the 17 beta-hydroxyl or counterpart hydroxyl with histidine 524 of the human estrogen receptor alpha. Biochemistry 41, 7979– 7988. [24] Iannone, M.A., Simmons, C.A., Kadwell, S.H., Svoboda, D.L., Vanderwall, D.E., Deng, S.J., Consler, T.G., Shearin, J., Gray, J.G. and Pearce, K.H. (2004) Correlation between in vitro peptide binding profiles and cellular activities for estrogen receptormodulating compounds. Mol. Endocrinol. 18, 1064–1081. [25] Gangloff, M., Ruff, M., Eiler, S., Duclaud, S., Wurtz, J.M. and Moras, D. (2001) Crystal structure of a mutant hERalpha ligand-
V.V. Sumbayev et al. / FEBS Letters 579 (2005) 541–548
[26]
[27] [28]
[29]
binding domain reveals key structural features for the mechanism of partial agonism. J. Biol. Chem. 276, 15059–15065. Hofmeister, M.V. and Bonefeld-Jørgensen, E.C. (2004) Effects of the pesticides prochloraz and metiocarb on humane Estrogen receptor a and b mRNA levels analyzed by on-line RT-PCR analyses. Toxicology in vitro 18, 427–433. Gru¨nfeld, H.T. and Bonefeld-Jørgensen, E.C. (2004) Effect of in vitro estrogenic pesticides on human estrogen receptor and b mRNA levels. Toxicology Letters 151, 467–480. Metivier, R., Penot, G., Flouriot, G. and Pakdel, F. (2001) Synergism between ERalpha transactivation function 1 (AF-1) and AF-2 mediated by steroid receptor coactivator protein-1: requirement for the AF-1 alpha-helical core and for a direct interaction between the N- and C-terminal domains. Mol. Endocrinol. 15, 1953–1970. Bentrem, D., Fox, J.E., Pearce, S.T., Liu, H., Pappas, S., Kupfer, D., Zapf, J.W. and Jordan, V.C. (2003) Distinct molecular conformations of the estrogen receptor alpha complex exploited by environmental estrogens. Cancer Res. 63, 7490–7496.