Binding Assays Using a Benzofurazan-Labeled Fluorescent Probe for Estrogen Receptor–Ligand Interactions

Abstract

Binding assays are widely used to study the estrogenic activity of compounds targeting the estrogen receptor (ER). The fluorescence properties of benzofurazan (BD), an environmentally sensitive fluorophore, are affected by solvent polarity. In this study, we synthesized BD-labeled estradiol (E2) derivatives hoping to develop a fluorescent ligand to be used in ER binding assays, without the separation of free- from bound-ligand. Three fluorescent ligands with a BD skeleton were obtained and their fluorescence properties were investigated. Analysis of the fluorescent ligands and human recombinant ERα (hr-ERα) interactions revealed that the fluorescence intensity increased in hydrophobic environments, such as the receptor-binding site. In saturation binding assays, ABD-E2 derivative 2c showed positive cooperative binding, and its dissociation constant (K_d) and Hill coefficient were 23.4 nM and 1.34, respectively. The estrogenic compounds affinity, assessed by competitive binding assays was well correlated with the results obtained by conventional studies, using the fluorescence polarization method. Overall, the developed assay using BD-labeled ligands was a simple, rapid, and reliable method for the evaluation of ER binding affinity.

Introduction

Estrogen receptor (ER), a member of the nuclear receptor superfamily, is a ligand-dependent transcription factor involved in cell growth and differentiation processes, consisting of two different subtypes, ERα (66 kDa) and ERβ (59 kDa).^1,2) When a hormone binds to ER, a series of events are initiated, including conformational changes, dimerization, binding to estrogen response elements (ERE) in the promoter region of estrogen-regulated genes, and interaction with coactivators.^3–5) Detection and evaluation of estrogenic activity are important in drug discovery and pathological analysis of estrogen-linked diseases, such as breast cancer, osteoporosis, cardiovascular disease, type II diabetes, and Alzheimer disease.⁶⁾

In the last few decades, various synthetic chemicals have been produced and released into the environment. Some of these chemicals can interfere with normal biological processes and adversely affect development and/or reproductive function in wildlife and humans; these are called endocrine-disrupting chemicals (EDCs).^7–9) Environmental estrogenic chemicals bind to ER and can disrupt steroid signaling in organisms through agonistic or antagonistic effects.

Many in vitro and cell-based screening methods have been designed to identify estrogenic compounds, each of them with recognized advantages and disadvantages. ER transcriptional assays using reporter genes’ knock-in cell lines, e.g. expressing luciferase or β-galactosidase downstream of an ERE, can evaluate if ER-interacting compounds are agonists or antagonists.¹⁰⁾ However, these assays require the mastering of cell culture protocols and the consequent maintenance of cultured cell lines. Additionally, binding assays are useful to study the binding affinity of test compounds to ER binding sites. Traditional radioisotope-labeled estradiol (E2) derivatives have been used to characterize the ER-binding affinities of various environmental compounds.^11–14) As an alternative method suitable for the screening of many EDCs without the use of radioactive reagents, fluorescence polarization (FP)-based binding assays have also been developed.^15–18) The FP method is based on the principle that a fluorescent molecule, when excited with polarized light, will emit fluorescence with a degree of polarization inversely proportional to its rate of rotation. Small fluorescent molecules (e.g., probe) will rotate faster in solution than larger molecules, (e.g., ER-probe complex); hence, when the probe is bound to ER, its rotation is slowed, and the polarization value is increased.¹⁶⁾ Therefore, the bound to free ligand ratio (bound/free) can be easily quantified according to the FP value without the need to evaluate separately bound and free ligand conditions. Although FP-based methods are well suited for high-throughput screening (HTS), they still require special instruments.

Commonly used fluorophores (such as fluorescein and rhodamine) are associated with several limitations, including the unstable fluorescence intensities and the alteration of excitation/emission spectra, depending on the molecular environment.¹⁹⁾ Benzofurazan (2,1,3-benzoxadiazole [BD]) fluorophores are small molecules with large Stokes shifts. Additionally, BD derivatives have long excitation and emission wavelengths, ideal to prevent biomatrixes-derived interference.²⁰⁾ The fluorescence intensity of BD derivatives usually increases sharply as the solvent polarity decreases²¹⁾: although these compounds do not fluoresce in aqueous solution, their fluorescence intensity increases sharply in nonpolar solvents or hydrophobic environments, such as receptor binding sites. However, no studies have reported the use of BD derivatives in the context of ER binding affinity assays.

To overcome the existing methods/approaches’ disadvantages mentioned above, we attempted to design a fluorescent ligand that would be an effective and reliable reporter in ER binding assays. We synthesized a series of BD-labeled E2 derivatives and evaluated their binding affinities to ER via fluorescence intensity measurements. We further performed ER competitive binding assays with BD-labeled ligands and unlabeled estrogenic compounds.

Results and Discussion

Preparation of BD-Labeled E2 Derivatives

First, we developed the fluorescent ligands for ER binding assays (Fig. 1). The benzofurazan-labeled estradiol derivatives (BD-E2, 2a–2c) were based on the structure of fluorescein-labeled estradiol (F-E2),¹⁷⁾ a 17α-substituted E2 derivative with a spacer containing an amino group. The fluorescent derivatization reagents were mixed with 17α-(4-aminobutynyl)estradiol, prepared as previously described (9; Chart 1). Following silica column chromatography and C18 solid-phase extraction, the purified BD-E2s were stored in MeOH after quantification via absorption spectra measurements.

Fig. 1. Chemical Structures of 17α-Estradiol (1) and the Benzofurazan-Labeled Fluorescent Ligands 2a–2c

Chart 1. Preparation of the 17α-Substituted E2 Derivative 9

Reagents and conditions: (a) MOMCl, DIEA, CH₂Cl₂, 0 °C to room temperature, 16 h, 79%. (b) 3-butyn-1-ol THP ether, n-BuLi, THF, −78 °C to room temperature, 16 h, 68%. (c) CSA, MeOH, room temperature, 30 min, 94%. (d) NsNHBoc, Ph₃P, DEAD, benzene, room temperature, 16 h, (e) PhSH, Cs₂CO₃, DMF, room temperature, 1 h, 94% for two steps. (f) TFA, room temperature, 20 min.

The fluorescence properties of BD-E2s are described in Table 1. The Stokes shifts were approximately 120 (DBD-E2, ABD-E2) and 70 nm (NBD-E2). The effects of hydrophobicity on their fluorescence intensities was examined using various MeOH in water mixtures. As shown in Fig. 2, the fluorescence values (F_obs) of BD-E2 compounds were increased with increasing MeOH proportions. From these results, we anticipated that the BD-E2 compounds’ fluorescence intensity increase occurred upon binding to ER owing to the more hydrophobic environment. Thus, the binding ability of BD-E2 to ER may probably be extrapolated based on the change of BD-E2 fluorescence intensity upon binding to the ER binding site.

Table 1. Absorption and Emission Data of Benzofurazan-Labeled Compounds in Three Solvents

Solvent	MeCN		EtOH		MeOH
	λex (nm)	λem (nm)	λex (nm)	λem (nm)	λex (nm)	λem (nm)
DBD-E2 (2a)	427	542	430	544	429	552
NBD-E2 (2b)	462	533	467	529	466	537
ABD-E2 (2c)	427	548	428	555	428	555

Fig. 2. The Fluorescence Intensities of BD-E2 Compounds (0.5 µM) Were Measured Using Fixed Excitation and Fluorescence Wavelengths for Various Proportions of Methanol in Water

The excitation wavelengths used were 428 (2a, 2c) or 465 nm (2b) and emission was detected at 546 (2a, 2c) or 533 nm (2b). 2a, DBD-E2; 2b, NBD-E2; and 2c, ABD-E2.

Saturation Binding Assay

To assess the interactions of ER and BD-E2, the F_obs values were determined for different human recombinant (hr)-ERα concentrations and used to calculate the K_d values. Figure 3 shows the results of the saturation binding assays using BD-E2s. The F_obs values for NBD-E2 (2b) showed a sigmoid curve, dependent on the concentration of hr-ERα. The F_obs values for DBD-E2 and ABD-E2 (2a and 2c) were lower than those of NBD-E2. However, F_obs values were almost fully saturated around the hr-ERα concentration of 160 nM. A Scatchard plot analysis of the F_obs values was performed to obtain the K_d value for the BD-E2 compounds-hr-ERα interaction. The Scatchard curves for the three BD-E2 derivatives showed convex forms (Fig. 4A), suggesting a homotropic allosteric effect.

Fig. 3. Saturation Binding Assay Using BD-E2 Compounds

Increasing concentrations of hr-ERα were incubated with 1 nM of each BD-E2 compound for 1 h. Fluorescence was then measured. The binding isotherms of hr-ERα and BD-E2 compounds are represented: 2a, DBD-E2; 2b, NBD-E2; and 2c, ABD-E2. Data represent means ± standard deviations. (n = 3).

Fig. 4. Scatchard (A) and Hill Plot (B) Analyses of Compound 2c-hr-ERα Interactions

Data represent means ± standard deviation (n = 3).

However, since the K_d values could not be obtained from the convex Scatchard curves, the data were transformed into Hill plots, which normally give linear curves (Fig. 4B). The obtained K_d values of BD-E2 compounds were 32.0 (2a), 47.0 (2b), and 23.4 nM (2c), respectively. Compared with F-E2 (K_d value for ERα of 10.4 nM), each BD-E2 compound showed a moderate affinity for hr-ERα. The Hill coefficients obtained from the Hill plots’ slopes were 0.73 (2a), 1.39 (2b), and 1.34 (2c), respectively. These results suggest that the Hill coefficients of BD-E2 compounds are related to the electron-accepting ability of the benzofurazan skeleton substituent group (R), as was reported elsewhere²²⁾ (NO₂ > SO₂NH₂ > SO₂N(CH₃)₂). A Hill coefficient greater than 1 indicates positive cooperativity: binding of the first ligand induces structural changes to the protein, increasing its affinity for additional ligands. Previous studies using calf uterine-^23,24) and hr-ERα¹⁷⁾ have shown that ER formed a dimer after the conformational changes induced by ligand binding, with a Hill coefficient of approximately 1.6. The Hill coefficients of compounds 2b and 2c were comparable with that of tritium labeled E2 or F-E2.

Competitive Binding Assay

Ideally, the optimal fluorescent ligand should retain the binding affinity for the receptor and detectable fluorescence intensity. NBD-E2 and ABD-E2 (2b and 2c) fluorescence intensities were determined in the context of a fixed concentration of hr-ERα (16 nM). The differences in F_obs values for the ER-ABD-E2 complex (0% inhibition) and free ABD-E2 (100% inhibition) were higher than those for NBD-E2 (data not shown). Thus, we selected ABD-E2 as the fluorescent ligand to use in competitive binding assays in order to confirm whether decreases in the fluorescence intensity were based on changes in environmental polarity, namely, the displacement from the ER binding site by the unlabeled compound ligand.

The ER affinity of test compounds was determined using ABD-E2 (2c) and hr-ERα. The competition binding curves are shown in Fig. 5A. The IC₅₀ of each compound was calculated from the fluorescence intensity values and converted into the inhibition constant (K_i) value using the Kenakin’s correlation²⁵⁾ with the K_d value obtained for compound 2c. The calculated K_i value of 0.10 nM was similar to that previously reported using a radioactive ligand.¹¹⁾ This result suggested that compound 2c was obviously displaced from the hr-ERα binding site by the unlabeled E2, which was the appropriate ligand to afford the K_i value. After the data were converted using pseudo-Hill plot analysis, the Hill coefficient was obtained (Fig. 6A). The Hill coefficient of 1.61 obtained for the unlabeled E2 binding to hr-ERα was similar to that previously reported,^17,18) suggested that the positive cooperativity of compound 2c induces the same hr-ERα conformational changes.

Fig. 5. Inhibition Curves of Estrogenic Compounds by the Proposed Competitive Binding Assay Using Compound 2c

(A) Physiological estrogens, (B) agonists, and (C) antagonists. The abbreviations used in graphs are described in Table 2. Data represent means ± standard deviation (n = 3).

Fig. 6. Pseudo-Hill Plots of Estrogenic Compounds by the Proposed Competitive Binding Assay Using Compound 2c

(A) Physiological estrogens, (B) agonists, and (C) antagonists. Data represent means ± standard deviation (n = 3).

Next, we examined the affinities of 10 compounds, including physiological estrogens, pharmaceuticals, and industrial chemicals, to hr-ERα using competitive binding assays. The obtained inhibition curves of the tested compound are shown in Figs. 5A–5C. Most of the tested compounds showed sigmoidal curves, suggesting that they displaced ABD-E2 from the hr-ERα. The inhibition parameters of the tested compounds are summarized in Table 2. The relative binding affinities (RBAs) of these compounds, which were calculated on the basis of the IC₅₀ value of E2 were easily compared with the values reported from other sources. For physiological estrogens, the order of competition was as follows: E2 > estriol (E3) > estrone (E1) > 17α-estradiol (17α-E2). The current IC₅₀ values of E1 and E3 (2.53 and 2.43 nM, respectively) were lower than those reported in a previous study (9.9 and 5.0 nM, respectively) using F-E2.^17,18) The order of competition for pharmaceuticals and industrial chemicals was as follows: diethylstilbestrol (DES) > ethynylestradiol (EE2) > 4-hydroxytamoxifen (4-OH-TAM) > tamoxifen (TAM) > 4-n-nonylphenol (4-NP) > bisphenol A (BPA). Overall, the order of affinity of these compounds (obtained by competition with ABD-E2) was well correlated with the results obtained by the conventional method using FP (Fig. 7A).

Table 2. Binding Parameters of the Tested Compounds

Compound	IC50 (nM)	RBA (%)	Log RBA	Ki (nM)	Hill coefficient
17β-Estradiol (E2)	2.24	100	2.00	0.11	1.61
Estrone (E1)	2.53	88.71	1.95	0.13	0.93
Estriol (E3)	2.43	92.33	1.97	0.12	1.23
17α-Estradiol (17α-E2)	2.70	82.99	1.92	0.13	1.46
Ethynylestradiol (EE2)	2.23	100.54	2.00	0.11	1.34
Diethylstilbestrol (DES)	1.64	136.64	2.14	0.08	1.80
Bisphenol A (BPA)	377.56	0.59	−0.23	18.71	0.95
4-n-Nonylphenol (4-NP)	347.96	0.64	−0.19	17.24	1.02
Tamoxifen (TAM)	51.04	4.39	0.64	2.53	3.42
4-Hydroxytamoxifen (4-OH-TAM)	16.70	13.43	1.13	0.83	3.60

Fig. 7. Correlations between the Binding Affinity (A) or Hill Coefficient (B) Obtained for hr-ERα and Compound 2c Interactions, and the FP Method Using F-E2

The log RBA and Hill coefficient obtained with compound 2c and those obtained with F-E2 are plotted on the X and Y axes, respectively.

In a previous report, the relationship between estrogenic compounds’ chemical structures and their estrogenic activities were examined, and the molecules were classified into three categories (agonists, partial agonists, and antagonists) based on their Hill coefficients.¹⁸⁾ Because we found that all tested compound competed with compound 2c, we assessed the Hill coefficients obtained from pseudo-Hill plots (Figs. 6A–6C). DES and EE2 are strong estrogenic compounds, as indicated by their Hill coefficients of 1.80 and 1.34, respectively. The magnitude of Hill coefficient values opposed that obtained using F-E2 (1.59 and 1.73, respectively). However, we consider that these results were attributed to differences in ligand structures. In contrast, TAM and 4-OH-TAM are anti-estrogens that are used for the treatment of estrogen-dependent breast cancer; the Hill coefficient of TAM was found to be 3.42. This value increased by 40% compared with that obtained in the context of F-E2 competition, but met the criteria for the antagonist category according to the Hill coefficient proposed by Ohno et al. Therefore, and overall we considered that our results of competitive binding assays using compound 2c, reproduced those published previously.¹⁸⁾ The correlations of Hill coefficients obtained with compounds 2c and F-E2 are shown in Fig. 7B. Although we observed a clear correlation between the values obtained by fluorescent measurements and those obtained by FP, further studies of the binding mechanisms of BD-E2 to ER are needed to evaluate the transcriptional activities, namely, agonism/antagonism of ER.

Conclusion

In this study, we exploited the structural and binding properties of known fluorescein-labeled E2 compounds and developed novel fluorescent ligands (BD-E2) destined to ER binding assays. Although BD-E2 compounds yield little fluorescence in aqueous solution, they exhibit spectroscopic properties with high fluorescence intensities and large Stokes shifts in a hydrophobic environment. By evaluation of the bound ligands based on changes in the fluorescence intensity, we established a simple, rapid, reliable ER binding assay. The binding properties obtained through this fluorescence-based ER binding assay indicated good correlations with the results obtained by FP analysis and conventional methods using radioactive ligands. On the other hands, it is anticipated that the enhancement of the benzofurazan skeleton electron-accepting ability will contribute to the increase of BD-E2 compounds’ binding affinity to ER. We expect that this approach will be a valuable tool not only for HTS of estrogenic compounds but also for studies of drug discovery or cell imaging.

Experimental

Materials

Commercial materials and solvents were used without further purification. 17β-Estradiol, estrone, estriol, 4-nitro-7-fluoro-2,1,3-benzoxadiazole (NBD-F), 4-(N,N-dimethylaminosulfonyl)-7-fluoro-2,1,3-benzoxadiazole (DBD-F), DES, and 2,2-bis(4-hydroxyphenyl)propane (BPA) were all purchased from Tokyo Chemical Industry (Tokyo, Japan). 4-Fluoro-7-sulfamoylbenzofurazan (ABD-F) and 4-n-nonylphenol were obtained from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). 17α-Estradiol was purchased from Nacalai Tesque (Kyoto, Japan), and 17α-ethynylestradiol was obtained from Sigma-Aldrich (St. Louis, MO, U.S.A.). TLC was performed on TLC silica gel 60 F254 plates. Purification by flash chromatography was carried out using silica gel 60 (0.063–0.200 mm) purchased from Merck (Darmstadt, Germany).

Apparatus

¹H-NMR spectra were obtained on a JEOL JMN-ECZ400S/L1 spectrometer (Tokyo, Japan). Chemical shifts are expressed in δ ppm in relation to tetramethylsilane as an internal standard. The following abbreviations were used: s, singlet; d, doublet; dd, double doublet; t, triplet; and m, multiplet. MS and high-resolution MS (HR-MS) were performed using a JEOL JMS-700 V spectrometer for FAB-MS with glycerol as the matrix. UV-Vis spectra were measured on a Shimadzu MPS-2450 UV-Vis spectrophotometer (Kyoto, Japan). Fluorescence spectra were measured using a Shimadzu RF-5300PC fluorescence spectrophotometer. Binding assays, performed in microplates, were evaluated using a Tecan infinite M1000 (Männedorf, Switzerland).

Preparation of 17α-Substituted E2 Derivative

The overview is represented in Chart 1. The solution of estrone (3, 998.0 mg, 3.69 mmol) in dry CH₂Cl₂ (10 mL) was added to methoxymethyl chloride (MOMCl, 806 µL) and N,N-diisopropyl ethylamine (DIEA, 973 µL) at 0 °C. Then, the reaction mixture was stirred and warmed to room temperature. After 16 h, the mixture was diluted with diethyl ether. The organic layer was washed with water, dried over Na₂SO₄, and filtered. The filtrate was evaporated and purified by flash chromatography (hexane : EtOAc, 6 : 1) to originate 4 (916.8 mg, 79%) as a colorless solid. ¹H-NMR (CDCl₃): δ (ppm) 0.91 (3H, s), 3.48 (3H, s), 5.15 (2H, s), 6.79 (1H, d, J = 2.7 Hz), 6.85 (1H, dd, J = 8.5, 2.7 Hz), 7.21 (1H, d, J = 8.3 Hz).

3-Butyn-1-ol tetrahydropyranyl (THP) ether²⁶⁾ (972.0 mg, 6.31 mmol) in dry tetrahydrofuran (THF; 4 mL) was added to n-BuLi (3.8 mL at 1.6 M in n-hexane) at −78 °C under nitrogen gas stream. After stirring for 1 h, the solution of 4 in dry THF (15 mL) was added to the reaction mixture and stirred for 16 h in order to slowly warm the solution to room temperature. The reaction was quenched by the addition of water, and the mixture was extracted with diethyl ether. The organic layer was washed with water, dried over Na₂SO₄, filtered, and evaporated. The crude product was purified by flash chromatography (hexane : EtOAc, 6 : 1) to obtain 5 (928.3 mg, 68%) as a pale-yellow oil. ¹H-NMR (CDCl₃): δ (ppm) 0.86 (3H, s), 2.56 (2H, t, J = 6.9 Hz), 3.48 (3H, s), 3.54 (2H, m), 3.85 (2H, m), 4.66 (1H, br s), 5.15 (2H, s), 6.77 (1H, d, J = 2.7 Hz), 6.83 (1H, dd, J = 8.5, 2.7 Hz), 7.21 (1H, d, J = 8.5 Hz).

(+)-10-Camphorsulfonic acid (CSA, 92.3 mg) was added to the solution of 5 (928.3 mg, 1.98 mmol) in MeOH (10 mL). The reaction mixture was stirred at room temperature for 30 min and then diluted with EtOAc. The organic layer was washed with water, dried over Na₂SO₄, filtered, and evaporated. The residue was purified by recrystallization from acetone to originate 6 (712.1 mg, 94%) as a colorless solid. ¹H-NMR (CDCl₃): δ (ppm) 0.87 (3H, s), 2.54 (2H, t, J = 6.2 Hz), 3.48 (3H, s), 3.75 (2H, t, J = 6.3 Hz), 5.15 (2H, s), 6.77 (1H, d, J = 2.7 Hz), 6.84 (1H, dd, J = 8.5, 2.7 Hz), 7.21 (1H, d, J = 8.3 Hz).

The mixture of 6 (100.8 mg, 0.26 mmol), N-(tert-butoxycarbonyl)-2-nitrobenzenesulfonamide^27,28) (NsNHBoc, 700.2 mg), and triphenylphosphine (Ph₃P, 417.1 mg) in dry benzene (5 mL) was added to diethyl azodicarboxylate (DEAD, 685 µL at 40% in toluene), and the reaction mixture was then stirred at room temperature for 16 h. The reaction was quenched by the addition of water, and the mixture was diluted with EtOAc. The organic layer was washed with water, dried over Na₂SO₄, filtered, and evaporated. The crude sample was removed by flash chromatography (hexane : EtOAc, 2 : 1) to originate 7 (561.5 mg) as a colorless solid. Next, benzene thiol (PhSH, 105 µL) and cesium carbonate (Cs₂CO₃, 826.2 mg) were added to the solution of 7 in N,N-dimethylformamide (DMF, 10 mL), and the mixture was stirred at room temperature for 1 h. The reaction was quenched by the addition of water, and the mixture was extracted with EtOAc. The organic layer was washed with 5% HCl, sat. NaHCO₃ solution, and water; dried over Na₂SO₄; filtered; and evaporated. The residue was purified by flash chromatography (hexane : EtOAc, 3 : 1) to obtain 8 (119.3 mg, 94%) as a pale-yellow oil. ¹H-NMR (CDCl₃): δ (ppm) 0.87 (3H, s), 1.42 (9H, s), 2.45 (2H, t, J = 6.6 Hz), 3.29 (2H, t, J = 5.9 Hz), 4.79 (1H, br s), 6.56 (1H, d, J = 2.9 Hz), 6.63 (1H, dd, J = 8.4, 2.9 Hz), 7.16 (1H, d, J = 8.4 Hz).

Trifluoroacetic acid (TFA, 2.14 mL) was added to 8 (57.8 mg, 0.12 mmol) and held at room temperature for 20 min. The mixture was then evaporated to obtain 9 (60.6 mg) as a brown solid (the trifluoroacetate salt). ¹H-NMR (CDCl₃): δ (ppm) 0.86 (3H, s), 2.40 (2H, t, J = 6.6 Hz), 3.10–3.20 (2H, m), 4.80–4.90 (1H, br s), 6.47 (1H, d, J = 2.4 Hz), 6.54 (1H, dd, J = 8.3, 2.7 Hz), 7.08 (1H, d, J = 8.3 Hz).

Preparation of BD-E2 Derivatives

The fluorescent ligands (BD-E2 derivatives) were synthesized by reaction with 7-fluoro-2,1,3-benzoxadiazole (BD-F) reagent using the two procedures described below.

Procedure A: 9 (0.04 mmol) was dissolved in pyridine (0.5 mL). After the addition of BD-F (5.7 mg) in MeCN (1.1 mL), the reaction mixture was warmed at 60 °C for 4 h. The mixture was cooled, added to 5% HCl, and extracted with EtOAc. The organic layer was washed with sat. NaHCO₃ solution and water, dried over Na₂SO₄, filtered, and evaporated. After purification by preparative TLC (hexane : EtOAc, 1 : 1), a yellow oil was obtained, which was further purified by solid-phase extraction using Agilent Bond Elut C18 (200 mg, 3 mL) with MeCN–water to give BD-E2 compounds.

Procedure B²⁹⁾: 9 (0.07 mmol) was dissolved in DMF containing 0.2% triethylamine (11.1 mL). After the addition of BD-F (14.3 mg), the reaction mixture was stirred at room temperature for 2 h. Then, the reaction mixture was quenched by the addition of water and extracted with EtOAc–CH₂Cl₂ (1 : 1). The organic layer was washed with brine, dried over Na₂SO₄, filtered, and evaporated. After purification by flash chromatography (hexane : acetone, 3 : 1), a yellow solid was obtained, which was further purified by solid-phase extraction using Agilent Bond Elut C18 with MeCN–water to give BD-E2 compounds.

17α-[{4-(N,N-Dimethylaminosulfonyl)-7-nitrobenzo-2-oxa-1,3-diazole-4-yl}amino-1-butynyl]estra-1,3,5(10)-trien-3-ol (DBD-E2, 2a)

Yellow solid (0.4% by procedure A). ¹H-NMR (CDCl₃): δ (ppm) 0.96 (3H, s), 2.76 (2H, t, J = 6.5 Hz), 3.64 (2H, br s), 6.19 (1H, d, J = 8.1 Hz), 6.56 (1H, d, J = 2.7 Hz), 6.63 (1H, dd, J = 8.4, 2.8 Hz), 7.15 (1H, d, J = 8.5 Hz), 7.91 (1H, d, J = 7.8 Hz). FAB-MS m/z: 565 ([M + H]⁺), 587 ([M + Na]⁺). HR-MS m/z: 565.2261 (Calcd for C₃₀H₃₆N₄O₅S: 565.2485).

17α-{4-(7-Nitrobenzo-2-oxa-1,3-diazole-4-yl)amino-1-butynyl}estra-1,3,5(10)-trien-3-ol (NBD-E2, 2b)

Yellow solid (1% by procedure B). ¹H-NMR (CDCl₃): δ (ppm) 0.77 (3H, s), 2.77 (2H, t, J = 6.4 Hz), 3.72 (2H, br s), 6.43 (1H, d, J = 2.6 Hz), 6.45 (1H, d, J = 8.8 Hz), 6.50 (1H, dd, J = 8.4, 2.6 Hz), 6.88 (1H, d, J = 8.8 Hz), 8.48 (1H, d, J = 8.4 Hz). FAB-MS m/z: 503 ([M + H]⁺), 525 ([M + Na]⁺). HR-MS m/z: 503.2301 (Calcd for C₂₈H₃₀N₄O₅: 503.2294).

17α-{4-(7-Aminosulfonylbenzo-2-oxa-1,3-diazole-4-yl)amino-1-butynyl}estra-1,3,5(10)-trien-3-ol (ABD-E2, 2c)

Yellow solid (4% by procedure B). ¹H-NMR (CDCl₃): δ (ppm) 0.80 (3H, s), 2.71 (2H, t, J = 6.6 Hz), 3.62 (2H, t, J = 6.6 Hz), 6.30 (1H, d, J = 8.1 Hz), 6.45 (1H, d, J = 2.6 Hz), 6.54 (1H, dd, J = 8.4, 2.6 Hz), 7.01 (1H, d, J = 8.4 Hz), 7.89 (1H, d, J = 8.1 Hz). FAB-MS m/z: 537 ([M + H]⁺), 559 ([M + Na]⁺). HR-MS m/z: 537.2164 (Calcd for C₂₈H₃₂N₄O₅S: 537.2172).

Spectroscopic Measurement

The purified BD-E2 derivatives were quantified by measurement of their absorption spectra, compared with model compounds as follows: DBD-NMe₂ in MeOH³⁰⁾ (ε = 1.06 × 10⁴ M⁻¹ cm⁻¹ at 442 nm), NBD-NHMe in MeCN²²⁾ (ε = 2.30 × 10⁴ M⁻¹ cm⁻¹ at 458 nm), and ABD-NHMe in MeCN (ε = 1.49 × 10⁴ M⁻¹ cm⁻¹ at 426 nm). Fluorescence spectra of the derivatives were measured using various solvents (0.1–0.5 µM) at room temperature. The emission spectra were obtained by excitation at the maximum absorption wavelength of the fluorophores.

Estrogen Receptor Binding Assay

To perform the binding assays of BD-E2 compounds and hr-ERα, full-length ERα was purchased from ThermoFisher Scientific (MA, U.S.A.). hr-ERα was stored at −80 °C and was not subjected to vortex mixing during handling. The buffer solution for the fluorescence measurements was 10 mM Tris–HCl buffer (pH 7.4) containing 50 mM KCl, 10% glycerol, 0.1 mM dithiothreitol, 0.02% sodium azide, and 1 µg/mL bovine γ-globulin. All BD-E2s and competing compounds were prepared as standard solutions in MeOH, and the solvent was removed using a dry nitrogen gas stream. The fluorescence intensities were observed with excitation at 428 (2a, 2c) or 465 nm (2b) and emission at 546 (2a, 2c) or 533 nm (2b).

Saturation Binding Assay

Direct binding studies were performed to examine the affinity of BD-E2 with hr-ERα. The final concentration of BD-E2 was fixed to 1 nM, and the final concentration of hr-ERα varied from 0.8 to 160 nM (0.8, 1.6, 4, 6, 8, 16, 40, 60, 80, 120, and 160 nM). Each sample volume was 100 µL. Equal volumes of BD-E2 and hr-ERα at different concentrations were added to 96-well black plates, to a total 100 µL in triplicate. After being allowed to stand at room temperature for 1 h, the fluorescence intensity of each sample was measured. The average of the measured fluorescence values (F_obs) was transformed into the bound ERα (R_b) using the following equation:

  

where F_max, F_min, and L_T are the fluorescence values for BD-E2 completely bound to hr-ERα, BD-E2 without hr-ERα (negative control), and the total ligand concentration, respectively. Free hr-ERα (R_f) was calculated by subtracting R_b from the total receptor concentration. The bound ligand concentration (B) was equal to R_b, and the free ligand concentration (F) was calculated by subtracting B from L_T. On the basis of the above calculations, Scatchard and Hill plots were designed and analyzed.

Competitive Binding Assays

Competitive binding studies were performed to evaluate the ability of the tested compounds to displace BD-E2 compounds from ER. As the concentration of the competitor increased to displace BD-E2 compounds from the complex, a decrease in the fluorescence value was observed. The final concentrations of hr-ERα and BD-E2 compounds were fixed to 16 and 10 nM, respectively. Each sample volume was 100 µL. Initially, hr-ERα and BD-E2 were mixed in glass test tubes to form the ER/BD-E2 complex. Competitor solutions of different concentrations were prepared in 96-well black plates, and ER/BD-E2 solution was then added. After being allowed to stand at room temperature for 1 h, the fluorescence intensity of each sample was measured. The fluorescence values were converted to percent inhibition using the following equation:

  

where F₀ and F₁₀₀ are the fluorescence values for the ER/BD-E2 complex without competitor as 0% inhibition (positive control) and for BD-E2 alone as 100% inhibition (negative control), respectively. The percent inhibition versus competitor concentration curves were analyzed by nonlinear least-squares curve fitting and yielded IC₅₀ values. The RBA for each competitor was calculated by dividing the IC₅₀ value of E2 by the IC₅₀ of the competitor and was expressed as a percent. Pseudo-Hill plots were also analyzed as previously described. The concentrations of bound competitor (B_c) and free competitor (F_c) were converted to B/F ratios as follows:

  

  

  

where B_T and C_T are the bound ligand and total competitor concentrations, respectively.

Conflict of Interest

This work was partially supported by a research Grant from ASKA Pharmamedical Co., Ltd. to K. Ohno. No other authors have any potential conflicts of interest to declare.

References

• 1) Brown M., Sharp P. A., J. Biol. Chem., 265, 11238–11243 (1990).
• 2) Wang C., Li C., Zhou H., Huang J., J. Biomol. Screen., 19, 253–258 (2014).
• 3) Gruber C. J., Tschugguel W., Schneeberger C., Huber J. C., N. Engl. J. Med., 346, 340–352 (2002).
• 4) Klinge C. M., Jernigan S. C., Mattingly K. A., Risinger K. E., Zhang J., J. Mol. Endocrinol., 33, 387–410 (2004).
• 5) Yasar P., Ayaz G., User S. D., Gupur G., Muyan M., Reprod. Med. Biol., 16, 4–20 (2017).
• 6) Paterni I., Granchi C., Katzenellenbogen J. A., Minutolo F., Steroids, 90, 13–29 (2014).
• 7) Colborn T., vom Saal F. S., Soto A. M., Environ. Health Perspect., 101, 378–384 (1993).
• 8) Danzo B. J., Cell. Mol. Life Sci., 54, 1249–1264 (1998).
• 9) Bolger R., Wiese T. E., Ervin K., Nestich S., Checovich W., Environ. Health Perspect., 106, 551–557 (1998).
• 10) Shanle E. K., Xu W., Chem. Res. Toxicol., 24, 6–19 (2011).
• 11) Kuiper G. G. J. M., Carlsson B., Grandien K., Enmark E., Haggblad J., Nilsson S., Gustafsson J. A., Endocrinology, 138, 863–870 (1997).
• 12) Nakai M., Tabira Y., Asai D., Yakabe Y., Shimyozu T., Noguchi M., Takatsuki M., Shimohigashi Y., Biochem. Biophys. Res. Commun., 254, 311–314 (1999).
• 13) Blair R. M., Fang H., Branham W. S., Hass B. S., Dial S. L., Moland C. L., Tong W., Shi L., Perkins R., Sheehan D. M., Toxicol. Sci., 54, 138–153 (2000).
• 14) Harris H. A., Bapat A. R., Gonder D. S., Frail D. E., Steroids, 67, 379–384 (2002).
• 15) Parker G. J., Law T. L., Lenoch F. J., Bolger R. E., J. Biomol. Screen., 5, 77–88 (2000).
• 16) Adamczyk M., Reddy R. E., Yu Z., Bioorg. Med. Chem. Lett., 12, 1283–1285 (2002).
• 17) Ohno K., Fukushima T., Santa T., Waizumi N., Tokuyama H., Maeda M., Imai K., Anal. Chem., 74, 4391–4396 (2002).
• 18) Ohno K., Suzuki S., Fukushima T., Maeda M., Santa T., Imai K., Analyst, 128, 1091–1096 (2003).
• 19) Leopoldo M., Lacivita E., Berardi F., Perrone R., Drug Discov. Today, 14, 706–712 (2009).
• 20) Uchiyama S., Santa T., Okiyama N., Fukushima T., Imai K., Biomed. Chromatogr., 15, 295–318 (2001).
• 21) Uchiyama S., Santa T., Fukushima T., Homma H., Imai K., J. Chem. Soc. Perkin Trans. 2, 1998, 2165–2173 (1998).
• 22) Uchiyama S., Santa T., Imai K., J. Chem. Soc. Perkin Trans. 2, 1999, 2525–2532 (1999).
• 23) Sasson S., Notides A. C., J. Biol. Chem., 258, 8118–8122 (1983).
• 24) Brooks S. C., Skafar D. F., Steroids, 69, 401–418 (2004).
• 25) Kenakin T., “Pharmacologic Analysis of Drug/Receptor Interaction,” 2nd ed., Raven, New York, 1993, pp. 385–410.
• 26) Bernady K. F., Floyd M. B., Poletto J. F., Weiss M. J., J. Org. Chem., 44, 1438–1447 (1979).
• 27) Fukuyama T., Cheung M., Kan T., Synlett, 1999, 1301–1303 (1999).
• 28) Kan T., Fukuyama T., Chem. Commun., 2004, 353–359 (2004).
• 29) Fevig T. L., Mao M. K., Katzenellenbogen J. A., Steroids, 51, 471–497 (1988).
• 30) Uchiyama S., Matsumura Y., de Silva A. P., Iwai K., Anal. Chem., 75, 5926–5935 (2003).