A Genetically Encoded Fluorescent Indicator Capable of

(A) Principle of the SCCoR indicator, based on intramolecular FRET, for visualizing the ligand-dependent interaction between the ERα LBD and LXXLL mo...
0 downloads 0 Views 214KB Size
Download PDF
Anal. Chem. 2004, 76, 2181-2186

A Genetically Encoded Fluorescent Indicator Capable of Discriminating Estrogen Agonists from Antagonists in Living Cells Muhammad Awais, Moritoshi Sato, Kazuki Sasaki, and Yoshio Umezawa*

Department of Chemistry, School of Science, The University of Tokyo and Japan Science and Technology Corporation (JST), Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

A genetically encoded fluorescent indicator was developed for the detection and characterization of estrogen agonists and antagonists. Two different color mutants of green fluorescent protein were joined by a tandem fusion domain composed of LXXLL (L ) leucine, X ) any amino acid) motif from the nuclear receptor-box II of steroid receptor coactivator 1, a flexible linker sequence, and the estrogen receptor r ligand binding domain (ERr LBD). Monitoring real-time ligand-induced conformational change in the ERr LBD to recruit the LXXLL motif allowed screening of natural and synthetic estrogens in single living cells using fluorescence resonance energy-transfer technique. The indicator was named SCCoR (single cellcoactivator recruitment). The high sensitivity of the present indicator made it possible to distinguish between estrogen strong and weak agonists in a dose-dependent fashion, immediately after adding ligand to live cells. Discrimination of agonists from antagonists was efficiently achieved using the present study. The approach described here can be applied to develop biosensors for other hormone receptors as well. Estrogens are responsible for the growth, development, and maintenance of many reproductive tissues.1 They exert their physiological and pharmacological effects by binding to estrogen receptor (ER).2,3 Steroid receptor coactivator 1 (SRC-1) belongs to a family of p160 nuclear hormone receptor coactivators, which interacts with the ligand binding domain (LBD) of ER to start the transcriptional activity of ER when it is complexed with an agonist but not when it is unoccupied or bound with antagonist.4 Agonists induce conformational change in ER that promotes and stabilizes ER-coactivator binding. At the same time ER-coactivator interaction stabilizes ER-ligand binding reciprocally and markedly slows the rate of dissociation of bound agonist from * To whom correspondence should be addressed. Tel: 81-3-5841-4351. Fax: 81-3-5841-8349. E-mail: [email protected]. (1) Ciocca, D. R.; Roig, L. M. V. Endocr. Rev. 1995, 16, 35-62. (2) Tsai, M. J.; O’Malley, B. W. Annu. Rev. Biochem. 1994, 63, 451-486. (3) Nilsson, S.; Ma¨kela, S.; Treuter, E.; Tujague, M.; Thomsen, J.; Andersson, G.; Enmark, E.; Pettersson, K.; Warner, M.; Gustafsson, J. Physiol. Rev. 2001, 81, 1535-1565. (4) McKenna, N. J.; Lanz, R. B.; O’Malley, B. W. Endocr. Rev. 1999, 20, 321344. 10.1021/ac030410g CCC: $27.50 Published on Web 03/13/2004

© 2004 American Chemical Society

ER LBD.5 SRC-1 contains a central nuclear receptor interaction domain that consists of three equally spaced conserved copies of leucine-rich signature motif referred to as the LXXLL motif (L ) leucine, X ) any amino acid), also called the NR box. This LXXLL motif is important in mediating the interaction of SRC-1 with the receptor.5,6,28,29 (5) Gee, A. C.; Carlson, K. E.; Martini, P. G.; Katzenellenbogen, B. S.; Katzenellenbong, J. A. Mol. Endocrinol. 1999, 13, 1912-1923. (6) Herry, D. M.; Kalkhoven, E.; Hoare, S.; Parker, M. G. Nature 1997, 387, 733-736. (7) Korach, K. S. Endocrinology 1993, 132, 2277-2278. (8) Shelby, M. D., Newbold, R. R., Tully, D. B., Chae, K., Davis, V. L. Environ. Health Perspect. 1996, 104, 1296-1300. (9) Oosterkamp, A. J.; Hock, B.; Seifert, M.; Irth, H. Trends Anal. Chem. 1997, 16, 545-553. (10) Sonnenschein, C.; Soto, A. M. J. Steroid Biochem. Mol. Biol. 1998, 65, 143150. (11) Colborn T. Environ. Health Perspect. 1995, 103, 135-136. (12) Neubert, D., Regul. Toxicol. Pharmacol. 1997, 26, 9-29. (13) Daston, G. P.; Gooch, J. W.; Berslin, W. J.; Shuey, D. L.; Nikiforov, A. I.; Fico, T. A.; Gorsuch, J. W. Reprod. Toxicol. 1997, 11, 465-481. (14) Salomonsson, M.; Carlsson, B.; Ha¨ggblad, J. J. Steroid Biochem. Mol. Biol. 1994, 50, 313-318. (15) Kuiper, G. G. J. M.; Carlsson, B.; Grandien, K.; Enmark, E.; Haggblad, J.; Nilsson, S.; Gustafsson, J. A. Endocrinology 1997, 138, 863-870. (16) Tabira, T.; Nakai, M.; Asai, D.; Yakabe, Y.; Tahara, Y.; Shinmyozu, T.; Naguchi, M.; Takatsuki, M.; Shimohigashi, Y. Eur. J. Biochem. 1999, 262, 240-245. (17) Blair, R. M.; Fang, H.; Branham, W. S.; Hass, B. S.; Dial, S. L.; Moland, C. L.; Tong, W. D.; Shi, L. M.; Perkins, R.; Sheehan, D. M. Toxicol. Sci. 2000, 54, 138-155. (18) Bolger, R.; Wiese, T. E.; Ervin, K.; Nestich, S.; Checovich, W. Environ. Health Perspect. 1998, 106, 551-557. (19) Kuramitz, H.; Natsui, J.; Sugawara, K.; Itoh, S.; Tanaka, S. Anal. Chem. 2002, 74, 533-538. (20) Usami, M.; Mitsunaga, K.; Ohno, Y. J. Steroid Biochem. Mol. Biol. 2002, 81, 47-55. (21) Soto, A. M.; Sonnenschein, C.; Chung, K. L.; Fernandez, M. F.; Olea, N.; Serrano, F. O. Environ. Health Perspect. 1995, 103 (Suppl. 7), 113-122. (22) Bronstein, I.; Fortin, J.; Stanely, P. E.; Stewart, G. S.; Kricka, L. J. Anal. Biochem. 1994, 219, 169-181. (23) Gaido, K. W.; Leonard, L. S.; Lovell, S.; Gould, J. C.; Babia, D.; Portier, C. J.; McDonnell, D. P. Toxicol. Appl. Pharmcol. 1997, 143, 205-213. (24) Pollock, B. A.; Heim, R. Trends Cell Biol. 1999, 9, 57-60. (25) Mochizuki, N.; Yamashita, S.; Kurokawa, K.; Ohba, Y.; Nagai, T.; Miyawaki, A.; Matsuda, M. Nature 2001, 411, 1065-1068. (26) Sato M.; Ozawa, T.; Inukai, K.; Asano, T.; Umezawa, Y. Nat. Biotechnol. 2002, 20, 287-294. (27) Sasaki, K.; Sato, M.; Umezawa, Y. J. Biol. Chem. 2003, 278, 30945-30951. (28) Mak, H. Y.; Hoare, S.; Henttu, P. M.; Parker, M. G. Mol. Cell. Biol. 1999, 19, 3895-3903. (29) Weatherman, R. V.; Chang, C. Y.; Clegg, N. J.; Carroll, D. C.; Day, R. N.; Baxter, J. D.; McDonnell, D. P.; Scanlan, T. S.; Schaufele, F. Mol. Endocrinol. 2002, 16, 487-496.

Analytical Chemistry, Vol. 76, No. 8, April 15, 2004 2181

There is mounting evidence that many synthetic chemicals having little structural resemblance to natural estrogens mimic or block the natural estrogen activities in the living body by binding with estrogen receptor.7-10 These xenoestrogens are also called endocrine disrupting chemicals (EDCs). Early developmental exposure to estrogenic chemicals is known to cause reproductive tract abnormalities, decrease in reproductive organ weights, and low sperm counts and quality in wildlife and humans.11-13 Several methods are employed to follow the ligandER binding interactions.8,14-17 These binding assays are based on competitive reaction in which a test compound displaces a labeled ligand, usually radioactive 17β-estradiol (E2), which is bound to ER. Large-scale screening of EDCs is possible by competitive receptor binding assays, but a number of shortcomings are there; for example, it is not possible to distinguish between agonistic and antagonistic effects, as binding to the receptor may not necessarily result in transcriptional activation (as in the case of an antagonist). These assays are not suitable to find the binding affinities of the compounds having low aqueous solubility. Moreover, incubation is necessary at subphysiological temperature; several washes are required to separate the free radiolabeled probe molecule from the molecules bound to ER before measurement, which may disturb the reaction equilibrium between ER and a ligand. With time, some new in vitro binding assays for EDC screening without the use of radioisotope emerged, such as the fluorescence polarization binding assay,18 electrochemical binding assay,19 and surface plasmon resonance biosensor technique,20 which are however unable to discriminate estrogen agonists from antagonists. All receptor binding assays need a large amount of purified receptor protein. A cell proliferation assay, also called E-Screening,21 an analysis of the ability of a chemical to stimulate growth of estrogen-sensitive cells, is another class of in vitro test system for EDC screening using MCF-7 or T47D cells. Reporter gene assays22,23 in yeast or mammalian cells, an analysis of the ability of a chemical to stimulate the transcription of a reporter gene construct in a cell culture, are very useful and powerful tools to identify substances of estrogenic potency. Although the E-Screening and reporter gene assays have the ability to distinguish between agonists and antagonists, one has to wait for almost one day to get the final results after a ligand is added in the culture media containing mammalian cells or yeast. In the present study, we developed a high-throughput single living cell-based EDC screening assay that is able to distinguish estrogen agonists from antagonists using the fluorescence resonance energy-transfer (FRET) technique.24-27 The principle of the present method is shown schematically in Figure 1A. An intramolecular FRET-based indicator was constructed to visualize the ligand-dependent recruitment of a coactivator peptide (687HKILHRLLQEG697) of SRC-16,28,29 to the estrogen receptor R ligand binding domain (ERR LBD) connected by a short flexible linker (GGNGG). This fusion protein was sandwiched between two different colored fluorescent proteins, cyan fluorescent protein (CFP, a donor) and yellow fluorescent protein (YFP, an acceptor), in such a way that the excitation and emission spectra for these green fluorescent protein (GFP) mutants are suitable for FRET in single living cells. We named this indicator SCCoR (single cell-coactivator recruitment) (Figure 1B), and the assay can be called the SCCoR assay. Upon agonist 2182

Analytical Chemistry, Vol. 76, No. 8, April 15, 2004

binding, a conformational change in the ERR LBD creates a coactivator (LXXLL motif) recruitment surface to interact with.30-32 This interaction brings CFP closer to YFP, thus inducing FRET between the two fluorescent units. The binding of an antagonist ligand to ERR LBD does not allow the LXXLL motif to interact, resulting in no FRET increase. The phenomenon of LBD/LXXLL motif interaction is of prime importance to discriminate between estrogen agonist and antagonist ligands. We describe herein a novel screening method for structurally diverse xenoestrogens ranging from natural to synthetic compounds such as genistein (Gen, a phytoestrogen), diethylstilbestrol (DES, a stilbene), bisphenol A (Bis-A, a biphenolic compound), and nonylphenol (NP, an alkylphenol) and antagonists such as 4-hydroxytamoxifen (OHT) and ICI 182,780. Our devised method allows the screening and characterization of EDCs immediately in the physiological environment by expressing the fluorescent indicator in living cells without purification of any protein and without using any artificial buffers, which are required in the case of conventional binding assays. EXPERIMENTAL SECTION Materials. Ham’s F-12 medium, fetal calf serum (FCS), Hank’s balanced salt solution (HBSS), and LipofectAMINE 2000 reagents were purchased from Life Technologies (Rockville, MD). Dulbecco’s modified Eagle medium (DMEM), trypsin-EDTA, E2, DES, OHT, and Gen were purchased from Sigma Chemicals Co. (St. Louis, MO). Anti-GFP antibody was obtained from Clontech (Palo Alto, CA). All cloning enzymes were from Takara Biomedical (Tokyo, Japan). A hERR cDNA plasmid was purchased from American Type Culture Collection. A mammalian expression vector pcDNA3.1 (+) was from Invitrogen Co.(Carlbad, CA). BisA, NP, and ICI 182,780 were from WAKO Pure Chemicals Industries Ltd. (Osaka, Japan). All other chemicals used were of analytical reagent grade. Plasmid Construction. To prepare the cDNAs for the constructs shown in Figure 1B, fragment cDNAs of ECFP (1238 aa), EYFP (1-238 aa), human estrogen receptor R LBD (305550 aa), a flexible linker (GGNGG), and SRC-1 NR box II (687697 aa) were generated by standard polymerase chain reaction (PCR) to attach a Kozak sequence and restriction sites shown in constructs. All PCR fragments were sequenced with an ABI310 genetic analyzer. This cDNA was inserted at HindIII and XhoI sites of mammalian expression vector pcDNA3.1 (+). Cell Culture and Transfection. Chinese hamster ovary (CHO-K1) cells were cultured in Ham’s F-12 medium supplemented with 10% FCS, with 1% penicillin/streptomycin, 1.0 mM sodium pyruvate, and 0.1 mM nonessential amino acids at 37 °C in 5% CO2. Cells were transfected with an expression vector pcDNA3.1 (+) containing SCCoR in the presence of LipofectAMINE 2000 reagent in glass-bottom dishes. In 12-24 h after transfection, cells were ready for imaging. Immunoblot Analysis for Protein Expression. The cell lysate of CHO-K1 transfected with expression vector pcDNA3.1 (30) Brzozowski, A M.; Pike, A. C.; Dauter, Z.; Hubbard, R.; Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G.; Gustafsson, J.; Carlquist, M. Nature 1997, 389, 753-758. (31) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P. J.; Agard, D. A.; Greene, G. L. Cell 1998, 95, 927-937. (32) Moras, D.; Gronemeyer, H. Curr. Opin. Cell Biol. 1998, 10, 384-391.

Figure 1. Fluorescent indicator for a SCCoR assay. (A) Principle of the SCCoR indicator, based on intramolecular FRET, for visualizing the ligand-dependent interaction between the ERR LBD and LXXLL motif of SRC-1. Upon agonist binding, a conformational change occurs in ERR LBD that facilitates its interaction with coactivator peptide (LXXLL motif). In the case of antagonist binding, the conformational change produced in ERR LBD prohibits the LXXLL motif interaction. The magnitude of FRET strongly depends on the relative orientation and distance between the donor (CFP) and acceptor (YFP) fluorophore. (B) Construct of indicators for expression and imaging in mammalian cells. Shown at top of each bar are restriction sites. LXXLL motif is a part of coactivator peptide, which is from NR box II of SRC-1 (687-697 aa). GGNGG is a flexible linker. ERR LBD contains residues from 305 to 550aa. CFP and YFP are different-colored mutants of GFP derived from Aequorea victoria with mammalian codons and the following additional mutations: CFP, F64L/S65T/Y66W/N146I/M153T/V163A/N212K, and YFP, S65G/V68L/Q69K/ S72A/T203Y. Kz is the Kozak sequence, which allows optimal translation initiation in mammalian cells. (C) Immunoblot analysis was performed as described in Experimental Section using whole cell extract of CHO-K1 cells transfected with an expression vector encoding the SCCoR construct, and the expression of expected 84-kDa protein was confirmed by western blotting. (D) Emission spectrum of the SCCoR indicator at 440-nm excitation.

(+) encoding SCCoR construct was subjected to SDS-PAGE using 10% polyacrylamide gel electrophoresis and electrophoreti-

cally transferred onto a nitrocellulose membrane. The membrane was probed with anti-GFP antibody and then with alkaline Analytical Chemistry, Vol. 76, No. 8, April 15, 2004

2183

Figure 2. (A) Pseudocolor images of the CFP/YFP emission ratios before (time 0 min) and at 4, 8, 12, 16, and 20 min after the addition of 100 nM E2, obtained from the CHO-K1 cells expressing SCCoR. (B) Time course of the FRET responses of SCCoR upon (I) E2 addition in live CHO-K1 cell, (II) with blank (HBSS in the absence of E2), and (III) response of the SCCoR/3A (mutated SCCoR) upon 100 nM E2 addition. The observed FRET level is represented by the ratio change between donor (CFP) and acceptor (YFP) emission intensities. Results represent a typical experiment from five independent trials.

phosphatase-labeled anti-rabbit antibody. The protein expression (Figure 1C) was analyzed using an image analyzer (LAS-1000 plus, Fujifilm Co., Tokyo, Japan). Imaging of Cells. Before imaging, the culture medium was replaced with HBSS. Within 12-24 h after transfection, the cells were imaged at room temperature on a Carl Zeiss Axiovert 135 microscope with a cooled charge-coupled device camera MicroMAX (Roper Scientific Inc, Tucson, AZ), controlled by MetaFluor (Universal Imaging, West Chester, PA). As shown in Figure 1D, the two peaks were observed at 480 ( 15 and 535 ( 12.5 nm in the emission spectrum of the SCCoR indicator, extracted from the cells, when the donor fluorophore CFP was excited at 440 nm. The weaker peak at 535 ( 12.5 nm was due to the basal FRET from CFP to YFP. Thus, the fluorescence images were obtained using 480 ( 15 and 535 ( 12.5 nm filters in the microscope with a 40× oil immersion objective (Carl Zeiss, Jena, Germany) (Figure 1D). The exposure time at 440 ( 10 nm excitation was 100 ms.

RESULTS AND DISCUSSION Time Course of the Interaction between ERr LBD and LXXLL Motif. To evaluate the response of the SCCoR indicator for the interaction of LXXLL and ERR LBD upon E2 stimulation, CHO-K1 cells expressing the indicator were stimulated by 100 2184 Analytical Chemistry, Vol. 76, No. 8, April 15, 2004

nM E2 and this event was observed by following the time course of the changes in FRET. The cell images were recorded before and at different time intervals after E2 stimulation as shown in Figure 2A, where the 480 nm/535 nm emission ratio was represented by pseudocolor images that depict the efficiency of FRET between the GFP mutants within SCCoR. Upon stimulation with 100 nM E2, a blue shift of the pseudocolor was observed as a decrease in the CFP/YFP emission ratio (increase in the FRET). The decreases in the emission ratio were detectable within several seconds and reached a plateau within 1000 s, but no detectable change was noted in the absence of E2 under otherwise identical experimental conditions (Figure 2B). To confirm that the increase in the FRET was triggered by the ERR LBD/LXXLL motif interaction upon E2 stimulation, we made a SCCoR mutant, SCCoR/3A, by replacing all the hydrophobic leucine (L) residues of LXXLL motif with alanine (A) residues (Figure 1B). The CHOK1 cells expressing the SCCoR/3A did not show any significant change in the emission ratio of CFP/YFP upon E2 stimulation as shown in Figure 2B. Taking all these together, it is concluded that the FRET response was due to the recruitment of LXXLL motif to ERR LBD, which was E2 dependent. In addition, the hydrophobic leucine residues in the LXXLL motif were confirmed to play a significant role for the interaction with ERR LBD as reported.5,6,28,29

Figure 3. Response of SCCoR for various concentrations of E2 ([), DES (9), Gen (2), NP (O), and Bis-A (b). The results are the means ( SD of emission ratios from five different cells/experiments.

According to the X-ray crystallographic analysis of E2/ERR LBD complex, E2 is completely enveloped with the hydrophobic ligand binding pocket (450 Å3, about twice the molecular volume of estradiol, 245 Å3) in the LBD in such a way to adopt a lowenergy conformation.30 The helix 12 is positioned over the ligand binding pocket, thereby forming a hydrophobic groove for recruitment and interaction of the hydrophobic LXXLL motif, which is stabilized by forming van der Waals contacts between side chains of the three leucine residues of the LXXLL motif and with the side chains of Met 543, Lys 362, and Gul 542 residues of ER LBD.31 Due to this interaction between the LXXLL motif and ERR LBD in the SCCoR, CFP gets a chance to be in proximity to YFP, resulting in an increase in the FRET response. Screening of Xenoestrogens Using SCCoR Indicator. By expressing the SCCoR indicator in CHO-K1 cells, xenoestrogens with diverse structural features, such as DES, Gen, NP, and BisA, were assessed for their ability to confer estrogenic activity as shown in Figure 3. The response of these estrogens was tested over a concentration range from 1.0 × 10-4 to 1.0 × 10-11 M. All the examined compounds enabled ERR LBD to recruit the LXXLL motif in a dose-dependent manner. The EC50 (the effective concentration of a ligand to induce a 50% response as a result of LXXLL motif recruitment to ERR LBD) values were determined from the response curves, which are 0.8 × 10-8, 1.3 × 10-8, 6.5 × 10-8, 0.26 × 10-6, and 0.42 × 10-6 M for E2, DES, Gen, NP, and Bis-A, respectively. The EC50 values were converted to relative recruitment abilities (RRA)34 to compare the relative abilities of the tested compounds to promote the recruitment of the LXXLL motif to ERR LBD. The RRA were calculated using the following equation.

RRA ) (concentration E2 (50% response)/ concentration EDC (50% response)) × 100 The RRA value for E2 was arbitrarily set at 100. The RRA for E2, DES, Gen, NP, and Bis-A were thereby obtained as 100, 60, 12, 3.0, and 1.9, respectively. Thus, the ability of the tested compounds to induce the recruitment of LXXLL motif to ERR LBD is as follows in decreasing order:

E2 > DES > Gen > NP > Bis-A

The subtle difference in the FRET response levels reflects the potency of estrogen-like chemicals to produce a ligand-specific conformational change in ERR LBD to recruit the LXXLL motif within the SCCoR. The primary purpose of E2 binding to ER is to induce a conformational change in the tertiary structure of the ER, such that the LBD is in a position to mediate the assembly of the basal transcriptional machinery following the recruitment of coactivator.4,5 However, the actual conformational change, which is the repositioning of the H12, in the tertiary structure of the ER induced by xenoestrogens may differ from that of E228,29,35 due to difference in the steric and electrostatic properties of the various ligands.33 In the present study, we showed that the binding of different chemical compounds to ERR LBD alters its ability to a differing extent to interact with the LXXLL motif in the cellular environment of live cells, so that possible structural changes of SCCoR are reflected in the distance and orientation between the two flanking GFP mutants, thereby resulting in different levels of the observed FRET. The ability of the EDCs to promote association between ERR and coactivator proteins using GST pull-down assays, according to Routledge et al.,34 are in the same order as we observed in single living cells using the intramolecular FRET technique. The ligand binding domains of several nuclear receptors function like the ER LBD.32 Thus, the principle of SCCoR can be applied to develop biosensors for other hormone receptors such as androgen, progesterone, thyroid, glucocorticoid, and orphan receptors. Agonist versus Antagonist. Figure 4 represents the results that discriminate an agonist, E2, from antagonists, ICI 182,780 and OHT. CHO-K1 cells expressing the fluorescent SCCoR indicator were stimulated by 1.0 µM E2, ICI 182,780, and OHT each , and changes in their FRET responses were monitored. A high increase in the FRET was observed in the case of E2, while no increase in the FRET was observed in the case of ICI 182,780 and OHT. Next we determined the ability of E2 to displace the ICI 182,780 and OHT from the LBD of ERR. A 1.0 µM concentration of ICI 182,780 was added to three different glass-bottom dishes containing CHO-K1 cells expressing the SCCoR indicator and the resultant mixture incubated for 15 min at room temperature. One, 10, and 100 µM E2 were added to first, second, and third dishes, respectively, without washing the ICI 182,780, and the changes were monitored in the FRET response. No change in the FRET level was observed upon adding 1.0 and 10 µM E2 in the presence of 1.0 µM ICI 182,780, while 100 µM E2 induced a change in the FRET level, which was ∼40% of that induced by 1.0 µM E2 alone. Similarly, addition of 1.0, 10, and 100 µM E2 in the presence of 1.0 µM OHT exhibited increases in the FRET responses of 40, 65, and 90%, respectively, that with 1.0 µM E2 alone (Figure 4). The results demonstrate that E2 promotes interaction between ERR LBD and LXXLL motif of SCCoR in living cells, whereas antagonists, ICI 182,780 and OHT, block this interaction in a manner that was not completely restored even by (33) Fang, H.; Tong, W.; Shi, L. M.; Blair, R.; Perkins, R.; Branham, W.; Hass, B. S.; Xie, Q.; Dial, S. L.; Moland, C. L.; Sheehan, D. M. Chem. Res. Toxicol. 2001, 14, 280-294. (34) Routledge, E. J.; White, R.; Parker, M.; Sumpter, J. J. Biol. Chem. 2000, 275, 35986-35993. (35) Jordan, V. C.; Schafer, J. M.; Levenson, A. S.; Liu, H.; Pease, K. M.; Simons, L. A.; Zapf, J. W. Cancer Res. 2001, 61, 6619-6623.

Analytical Chemistry, Vol. 76, No. 8, April 15, 2004

2185

Figure 4. Response of the SCCoR indicator upon stimulation with the agonist (E2), and the antagonists ICI 182,782 and OHT. The FRET responses for different concentrations of E2 in the presence of 1.0 µΜ ICI 182,780 and OHT are also shown. The results are the means ( SD of emission ratios from five different cells/experiments.

using a 100-fold higher concentration of E2. The presence of the antagonist may interfere with the signal generated by the agonist. These results can be discussed in the light of X-ray crystallographic studies of the ER LBD in the presence of an agonist and an antagonist.30,31 The agonist binding induces a conformational change in the LBD, exposing a coactivator-docking site on the LBD surface due to the alignment of H12 over the ligand binding cavity. On the other hand, the antagonists are able to prevent the proper alignment of helix12 through direct steric effects between their characteristic basic bulky side chains and helix12. Consequently, the coactivator recruitment site is not properly formed and LBD is unable to interact with the LXXLL motif. Therefore, the transfer of energy from CFP to YFP in the

2186 Analytical Chemistry, Vol. 76, No. 8, April 15, 2004

SCCoR was possible in the case of agonist, E2, but blocked in the case of antagonists, ICI 182,780 and OHT. Thus, coactivator (LXXLL motif) recruitment ability of ER in response to different EDC candidates enabled us to discriminate estrogen agonists from antagonists. The reporter gene assay is another tool to characterize the estrogens through the expression of reporter gene, but one has to wait for almost one day after a ligand is added to mammalian cells or yeast until the expression of a reporter gene for final measurements. In contrast, the present approach, the SCCoR assay, can characterize estrogens immediately after a ligand is added to live cells. In conclusion, dose-dependent screening of estrogen agonists and antagonists can be possible in the physiologically relevant environment (in situ live CHO-K1 cells without purification of any protein) immediately upon adding a ligand to live cells using the intramolecular FRET technique. By using the LXXLL motif of coactivator protein (SRC-1, a protein of significant importance for cellular signaling) besides the E2 receptor, the present approach allows sensitive characterization of a specific event in the activation of ER, which is the recruitment of LXXLL motif to the LBD in response to a ligand (agonist/antagonist). The phenomenon of ER LBD/LXXLL motif interaction is of prime importance to discriminate between estrogen agonist and antagonist ligands. High-throughput screening of a large number of estrogenic compounds including medicinal drugs and environmental and industrial chemicals is possible by the present fluorescent SCCoR indicator. The approach described in this study can be applied to develop biosensors for other hormone receptors such as androgen, progesterone, thyroid, glucocorticoid, and orphan receptors. ACKNOWLEDGMENT This work has been supported by CREST (Core Research for Evolutional Science and Technology) of JST (Japan Science and Technology) and by grants for Scientific Research by the Ministry of Education, Science and Culture.

Received for review December 9, 2003. Accepted January 28, 2004. AC030410G