help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2006-0316
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/8/3952    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Adler, M.
Right arrow Articles by Adler, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Adler, M.
Right arrow Articles by Adler, S.
Endocrinology Vol. 147, No. 8 3952-3960
Copyright © 2006 by The Endocrine Society

Derivatives of Z-Bisdehydrodoisynolic Acid Provide a New Description of the Binding-Activity Paradox and Selective Estrogen Receptor Modulator Activity

Molly Adler, Yuqing Hou, Paul Sandrock, Cal Y. Meyers, Todd A. Winters, William J. Banz and Stuart Adler

Departments of Physiology (M.A., T.A.W., W.J.B., S.A.), Internal Medicine/Endocrinology (S.A.), Chemistry and Biochemistry (Y.H., P.S., C.Y.M.), and Animal Science, Food, and Nutrition (T.A.W., W.J.B.) and the Meyers Institute for Interdisciplinary Research in Organic and Medicinal Chemistry (Y.H., P.S., C.Y.M.), Southern Illinois University, Carbondale, Illinois 62901

Address all correspondence and requests for reprints to: Stuart Adler, Southern Illinois University, Department of Physiology - Mailcode 6523, Carbondale, Illinois 62901. E-mail: sadler{at}siumed.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Z-Bisdehydrodoisynolic acid [(±)-Z-BDDA], an estrogenic carboxylic acid, is highly active in vivo yet binds poorly to estrogen receptors (ERs). Studies of Z-BDDA and its enantiomers demonstrate therapeutic potential as selective ER modulators; however, the activity vs. binding paradox has remained. One possible explanation is that the carboxylic acid group of Z-BDDA may be modified in vivo to an ester or amide. Synthesis of these derivatives showed the relative binding affinity (RBA) of the methyl ester for ER{alpha} and ERß was increased approximately 14- and 20-fold, respectively, relative to the parent compound. Yet, this increased affinity did not result in increased reporter gene expression. In contrast, the amide showed an unexpected approximately 4-fold decrease in RBA to both ERs compared with the parent. The relationship among the RBAs of ester, acid, and amide is consistent with their predicted polarity, suggesting the carboxylic acid, and not the carboxylate of BDDA, binds to ERs. Studies at pH 6.5, 7.4, and 8.0 were consistent with a simple acid-base equilibrium model, with BDDA binding as the undissociated acid and with affinity equal to or exceeding that of estradiol, consistent with high in vivo potency. Furthermore, the alcohol BDD-OH also demonstrated high affinity and increased activity in gene expression assays. In addition to suggesting a resolution to the decades-old binding/activity paradox, these studies may provide a direction for definitive in vivo metabolic and pharmacokinetic studies and provide additional insight into the chemical and metabolic determinants of BBDA’s unique tissue selectivity and selective ER modulator activities.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE DOISYNOLIC ACIDS and their derivatives stand out as unusual estrogenic compounds, even among the multitude of diverse steroidal and nonsteroidal molecules that comprise this important group of bioactive chemicals (1). The doisynolic acids, and the similar two-ring allenolic acids, were some of the earliest estrogenic compounds available for clinical use. Nonetheless, and despite their long history, their unusual properties continue to present unresolved questions with regard to estrogenicity and mechanism of action. In particular (±)-Z-bisdehydrodoisynolic acid [(±)-Z-BDDA] (Fig. 1Go, compound II) and its 3-methyl ether, (±)-Z-BDDA-3-OMe (Fig. 1Go, compound I), are among the most potent estrogenic compounds ever tested in animal studies in vivo (2, 3, 4, 5). Yet, in vitro binding studies (5, 6) and cell culture assays of estrogenicity, performed using a variety of cell types and reporter genes (6), show low activity and create an activity/binding paradox. Specifically, whereas other potent estrogens demonstrate a high affinity for preparations of estrogen receptors (ERs) in in vitro competitive binding assays, (±)-Z-BDDA competes only weakly with estradiol (E2) (5, 6, 7, 8, 9). Recent studies of this activity/binding paradox extended these classical observations to include ERß as well as ER{alpha} (10). In addition, in vivo analyses of (±)-Z-BDDA, and its separated enantiomers, (+)- and (–)-Z- BDDA, revealed selective ER modulator (SERM) activity (11). This unique tissue-specific spectrum of activities is distinct from that of E2 and includes effects on weight regulation and, in males, a potential separation of antiprostate effects from other activities associated with E2-mediated feminization (8, 11). The unusual SERM activities of Z-BDDA, together with the activity/binding paradox, previously raised several alternative hypotheses. One possibility is that there are other, as yet unidentified, high-affinity interactions through which Z-BDDA and its derivatives mediate some, or all, of their biological activities. These interactions would potentially include alternative ERs as well as mechanisms distinct from those exhibited by E2, including completely distinct receptors, pathways, and actions (12). Another possibility is that Z-BDDA and its derivatives are biologically metabolized to become high-affinity estrogens and mediate their distinct SERM activities via potent metabolites but through interactions with conventional pathways involving ER{alpha} and ERß.


Figure 1
View larger version (21K):
[in this window]
[in a new window]
  		FIG. 1. Structures of the compounds used in these studies. The structures of (±)-Z-BDDA and its derivatives are shown along with a summary of their synthesis schemes.

 
These studies were undertaken to address the possibility that relatively simple derivatives of (±)-Z-BDDA might exhibit high-affinity binding to ER{alpha} and ERß. Our approach targeted the carboxylic acid group, because it is a group that is unusual among potent estrogens. We reasoned that this moiety would be unlikely to interact favorably with the ERs’ hydrophobic ligand-binding pockets optimized for binding to uncharged and nonpolar steroid hormones. Three distinct series of experiments are presented. The first approach was to synthesize and determine the in vivo and cell culture activities of two simple derivatives of the BDDA carboxylic acid group that are neither bulky nor significantly ionized at physiologically relevant pH, the methyl ester (Fig. 1Go, compound III) and the amide (Fig. 1Go, compound IV). Next we determined whether the observed in vitro binding affinity of BDDA to ERs varied with pH. Finally, we prepared and evaluated the activities of the derivative of BDDA in which the carboxylic acid is reduced to the corresponding alcohol, BDD-OH (Fig. 1Go, compound V). Taken together, these results provide new insight into the unique activities of BDDA and its derivatives.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Compounds and hormones
(+)-17ß-Estradiol (E2) was purchased from Sigma Chemical Co. (St. Louis, MO). (±)-Z-BDDA-3-OMe (Fenocylin) (Fig. 1Go, compound I) was obtained from Ciba-Geigy, Inc. (Ardsley, NY). The syntheses of the derivatives of (±)-Z-BDDA are outlined in Fig. 1Go. (±)-Z-BDDA (Fig. 1Go, compound II) was prepared from the 3-methyl ether as previously described (5).

Preparation of the methyl ester of (±)-Z-BDDA (3-OH) [(±)-Z-BDDA methyl ester]
The methyl ester of (±)-Z-BDDA (Fig. 1Go, compound III) was also prepared as previously described (5). A solution of (±)-Z-BDDA-3-OMe (109 mg) in a mixture of 10 ml of acetic acid and 6 ml of 48% HBr aqueous solution was refluxed for 2 h. Water (20 ml) was added, and the mixture was cooled to room temperature. The slightly pink solid was filtered through a Büchner funnel and dried overnight on the lab bench. The solid was dissolved in acetone and the solution transferred to a 50-ml round-bottomed flask. The acetone was removed in vacuo, and the solid residue was dissolved in 6 ml dry dimethylsulfoxide. To this solution was added NaHCO3 (1.78 g). The suspension was stirred for a few minutes before methyl iodide (0.85 ml) was added to the mixture. The mixture was stirred at room temperature for 15 min, and when thin-layer chromatography (TLC) showed completion of the reaction with no (±)-BDDA-3-OH remaining, water was added to quench the reaction and the product was extracted with ether. TLC of the ether extract showed only one spot. The ether extract was dried over anhydrous Na2SO4 and evaporated to dryness to give a pale yellow thick oil. The oil was dissolved in hot isooctane and then left on the lab bench to cool to room temperature, when some oil came out from the solution. The mixture was stored in the freezer overnight, and the oil solidified. The clear solution was decanted, and the solid was dried under high vacuum for 5 h. 1H nuclear magnetic resonance (NMR) of the solid showed it was the desired product, contaminated with a minor amount of isooctane, as evidenced by the evolution of bubbles when a fast melting point (60–90 C, crude) was taken. 1H NMR (CDCl3, 300 MHz): {delta} 0.93 (t, 3 H, J = 7.5 Hz), 1.16 (s, 3 H), 1.38 (m, 1 H), 1.66 (m, 1 H), 2.04 (dd, 1 H, J = 5.1, 14.4 Hz), 2.35 (m, 1 H), 2.81 (m, 1 H), 2.99 (m, 1 H), 3.31 (dd, 1 H, J = 7.5, 18.3 Hz), 3.78 (s, 3 H), 5.28 (br s, 1 H), 7.14 (m, 3 H), 7.47 (d, 1 H, J = 8.4 Hz), 7.88 (d, 2 H, J = 8.7 Hz).

Preparation of (±)-Z-BDDA amide
The amide of (±)-Z-BDDA (Fig. 1Go, compound IV) was prepared by first preparing the acid chloride and then reacting this compound with ammonia. A mixture of (±)-Z-BDDA (100 mg), oxalyl chloride (3 ml), and benzene (5 ml) was refluxed for 30 min. All solvent and extra oxalyl chloride were removed in vacuo. The residue was dissolved in CH2Cl2. Ammonia was bubbled into the solution for 10 min, during which time a white precipitate formed. The solvent and excess ammonia were removed in vacuo. The crude product was purified with silica gel column chromatography (hexanes/ether, 1:3) to provide a yellow solid, which was recrystallized from hexanes/CH2Cl2 to provide an off-white solid (10 mg). 1H NMR of the solid showed it was the desired product. 1H NMR (CDCl3, 300 MHz): {delta} 0.93 (t, 3 H, J = 7.5 Hz), 1.17 (s, 3 H), 1.36 (m, 1 H), 1.75 (m, 1 H), 1.98 (m, 1 H), 2.37 (m, 1 H), 2.75 (m, 1 H), 3.06 (m, 1 H), 3.36 (dd, 1 H, J = 7.5, 18.3 Hz), 5.27 (s, 1 H), 5.46 (br s, 1 H), 5.78 (br s, 1 H), 7.15 (m, 3 H), 7.49 (d, 1 H, J = 8.1 Hz), 7.89 (d, 2 H, J = 8.4 Hz); 13C NMR (CDCl3, 75 MHz): {delta} 13.2, 22.6, 23.1, 24.8, 28.3, 45.0, 49.7, 110.3, 117.3, 124.0, 124.9, 127.5, 128.5, 130.4, 133.6, 133.7, 152.9, 180.3. The correct composition was confirmed by high-resolution mass spectrometry with the calculated m/z of 283.1572 for C18H21NO2 and observed m/z of 283.1566.

Preparation of the alcohol of (±)-Z-BDDA [(S)-8-ethyl-(R,S)-7-hydroxymethyl-7-methyl-5,6,7,8-tetrahydro-phenanthren-2-ol]
The alcohol of (±)-Z-BDDA, BDD-OH (Fig. 1Go, compound V), was prepared directly from the carboxylic acid using reduction with LiAlH4, rather than by the classical stepwise reduction through the aldehyde, as had been previously described (13). To a flame-dried 50-ml round-bottomed flask was added 0.23 g (0.8 mmol) of (±)-Z-BDDA and 20 ml dry tetrahydrofuran. The solution was placed under argon and cooled to –70 C in an acetone/liquid nitrogen bath. To this was injected 3.0 ml of 1.0 M (3 mmol) lithium aluminum hydride in tetrahydrofuran. The cooling bath was removed, and the reaction was stirred at room temperature for 24 h. The mixture was carefully quenched with water, acidified with 1.0 N HCl, and extracted with ether. The ether extract was dried over anhydrous MgSO4 and evaporated to yield 0.21 g of the crude desired (±)-BDD-OH (95% yield). TLC (10:1 hexane/ethyl acetate) showed one major spot and a very minor, more-polar impurity. The material was then recrystallized from chloroform, which yielded 68 mg of pure white powder (melting point = 102–103 C). 1H NMR (CDCl3, 300 MHz): {delta} 7.89 (d, 1 H; J = 8.4 Hz), 7.45 (d, 1 H; J = 8.4 Hz), 7.16–7.09 (m, 3 H), 3.77 (d, 1 H, J = 10.8 Hz), 3.53 (d, 1 H, J = 10.8 Hz), 3.26 (dd, 1 H), 3.06–2.94 (m, 1 H), 2.53 (dd, 1 H), 2.19 (s, 1 H), 1.91–1.81 (m, 3 H), 1.59–1.52 (m, 1 H), 1.41–1.31 (m, 1 H), 1.25 (s, 3 H), 0.93 (t, 3 H); 13C NMR (CDCl3, 75 MHz): {delta} 152.6, 134.9, 133.5, 130.8, 129.9, 127.7, 125.1, 123.6, 110.3, 110.1, 70.6, 47.4, 37.2, 26.1, 25.6, 22.4, 21.7, 13.3. The correct composition was confirmed by high-resolution mass spectrometry with the calculated m/z of 270.1620 for C18H22O2 and observed m/z of 270.1620.

Fluorescence polarization-based competitive binding inhibition assays
Competitive binding inhibition studies were performed using a Beacon 2000 instrument and software (Pan Vera Corp., Madison, WI) to determine fluorescence polarization. Reactions used purified recombinant human ER{alpha} or recombinant human ERß and a fluorescent estrogen, Fluormone ES2 (Pan Vera), and were performed according to the manufacturer’s protocols. Binding studies were typically performed at pH 7.4 using the ES2 screening buffer, which contains 100 mM potassium phosphate. For binding studies performed at pH 6.5 and pH 8.0, ES2 screening buffer was modified by the addition of concentrated HCl or 10 M KOH, respectively. Polarization was determined as millianisotropy units for various concentrations of each compound or E2. Relative binding affinities were determined using nonlinear least-squares analysis (Prism; Graphpad Software, Inc., San Diego, CA). Theoretical binding of the free acid form was based on an estimated pKa of 4.8 for the BDDA carboxyl group and the Henderson-Hasselbalch equation.

Reporter genes and expression vectors
Plasmids for model gene activation were as previously described (10). Rous sarcoma virus (RSV) promoter-based-ß-galactosidase expression plasmid was used for standardization of transfection efficiency (14). Assays for gene activation used the Vit2-P36L luciferase reporter plasmid, which contains two copies of a 26-bp estrogen response element from the Xenopus vitellogenin A2 gene linked to a minimal 36-bp promoter derived from the rat prolactin gene. The full-length wild-type human ER{alpha} cDNA (Gly 400) was obtained from P. Chambon (15) and subcloned for expression into a vector containing the RSV promoter (16). The full-length, 530-amino-acid human ERß cDNA (17) in a cytomegalovirus promoter-based expression plasmid was the generous gift of Dennis Lubahn (University of Missouri, Columbia, MO).

Cell lines and transfections
Hela cells were obtained from American Type Culture Collection (Rockville, MD). All cells are routinely surveyed for mycoplasma using a PCR method from Stratagene (La Jolla, CA). Cells are maintained and grown under estrogen-free conditions using media without phenol red and serum preparations that are treated with activated charcoal to remove endogenous steroid compounds (10, 18). Hela cells were grown in 10% CO2 in DMEM with 10% charcoal-stripped newborn calf serum (Equitech-Bio, Inc., Kerrville, TX). Transient transfections used a calcium-phosphate method modified for six-well plates (19). One day after hormone treatment, cells were harvested in a Triton lysis buffer containing 50 mM Tris, 50 mM 2 (N-morpholino)ethanesulfonic acid (pH 7.8), 1 mM dithiothreitol, and 1% Triton X-100. The lysate was assayed for luciferase activity as previously described (20), using a Monolight 2010 luminometer (Analytical Luminescence Laboratories, San Diego, CA). ß-Galactosidase assays were performed using chlorophenol red ß-galactopyranoside (Boehringer Mannheim, Indianapolis, IN) as substrate (21) and read on a Bio-Tek Plate Reader (Bio-Tek Instruments, Inc., Winooski, VT). EC50 values were determined using nonlinear least-squares analysis (Prism) relative to the maximal activity achieved for each individual compound.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Binding activities of ester and amide derivatives of (±)-Z-BDDA
Competitive binding studies using (±)-Z-BDDA methyl ester (III) and (±)-Z-BDDA amide (IV) were performed using purified recombinant ER{alpha} and ERß preparations and compared with the results obtained with E2 and the parent compound (±)-Z-BDDA (II). The methyl ester shows a marked increase in binding affinity relative to the (±)-Z-BDDA parent compound with both ER{alpha} (14-fold higher) and ERß (20-fold higher) (see Fig. 2Go and Table 1Go). In fact, the relative binding affinities of the methyl ester derivative approach that of E2 for both ER{alpha} (0.12) and ERß (0.06). Surprisingly, even though both amide and ester are derivatives of the carboxylic acid moiety, competitive binding affinities are markedly different. In contrast to the improved binding results seen with the methyl ester, the amide displays even weaker binding than the parent carboxylic acid, with relative binding affinities approximately 4-fold lower than the (±)-Z-BDDA parent compound for both ER{alpha} (0.002 vs. 0.008) and ERß (0.0007 vs. 0.003).


Figure 2
View larger version (39K):
[in this window]
[in a new window]
  		FIG. 2. Competitive binding inhibition assays using ER{alpha} and ERß. Results from fluorescence polarization competition binding assays vs. a fluorescent ligand, Fluormone ES2, are shown as millianisotropy units (mA). Data are shown for each compound as means ± SEM from multiple experiments as indicated. A, Binding to human ER{alpha} for E2 (n = 2), the (±)-Z-BDDA methyl ester (n = 2), and (±)-Z-BDDA (n = 2); B, binding to human ERß for E2 (n = 2), the (±)-Z-BDDA methyl ester (n = 2), and (±)-Z-BDDA (n = 2); C, binding to human ER{alpha} for E2 (n = 6), (±)-Z-BDD-OH (n = 2), (±)-Z-BDDA (n = 3), and the (±)-Z-BDDA amide (n = 3); D, binding to human ERß for E2 (n = 8), (±)-Z-BDD-OH (n = 3), (±)-Z-BDDA (n = 5), and the (±)-Z-BDDA amide (n = 5).

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. Relative binding and activity of Z-BDDA compounds

 
Gene activation using ester and amide derivatives of (±)-Z-BDDA
Cotransfection assays were used to compare the effects of (±)-Z-BDDA methyl ester and (±)-Z-BDDA amide on gene activation by ER{alpha} and ERß and compared with the results obtained with E2 and the parent compound (±)-Z-BDDA. Figure 3Go shows the results of experiments in Hela cells. All three compounds show dose-dependent activation of an estrogen-dependent luciferase reporter gene, although the concentration of compound required to achieve gene activations varied tremendously among the compounds (see Fig. 3Go and Table 1Go). In contrast to the increased affinity seen with the methyl ester derivative in receptor binding, there was no improvement in gene activation compared with the parent compound with both ER{alpha} and ERß. This may reflect the instability of esters to hydrolysis by esterases either in the sera-containing culture media or intracellularly. The amide derivative, which was chosen to be chemically more stable than the ester derivative, showed poorer binding than the parent compound and also showed poorer activity in the gene activation assays for both ER{alpha} and ERß. It was unexpected that these experimental results regarding the binding and activity of the amide, ester, and carboxylic acid forms would be so different in these assays. These data are not compatible with a very simple view of carboxylic acid chemistry and prompted more detailed binding studies.


Figure 3
View larger version (36K):
[in this window]
[in a new window]
  		FIG. 3. Estrogenic gene activation via ER{alpha} and ERß. Hela cells were cotransfected with the estrogen-responsive luciferase reporter plasmid Vit2-P36L, RSV-ß-galactosidase for internal standardization, plus an expression vector for either ER{alpha} or ERß. Concentrations of each compound were added as indicated. Light units are shown as relative expression with the maximal activity observed with E2 set to 100%. Data are shown for each compound as means ± SEM from multiple experiments as indicated. A, Activity with human ER{alpha} for E2 (n = 3), (±)-Z-BDDA (n = 3), (±)-Z-BDDA methyl ester (n = 3), and the (±)-Z-BDDA amide (n = 3); B, activity with human ERß for E2 (n = 3), (±)-Z-BDDA (n = 3), (±)-Z-BDDA methyl ester (n = 3), and the (±)-Z-BDDA amide (n = 3); C, activity with human ER{alpha} for E2 (n = 4), (±)-Z-BDD-OH (n = 4), and (±)-Z-BDDA (n = 4); D, activity with human ERß for E2 (n = 3), (±)-Z-BDD-OH (n = 3), and (±)-Z-BDDA (n = 3).

 
Evaluation of the effects of pH on ER binding of (±)-Z-BDDA
The results evaluating the ester and amide derivatives of BDDA suggested that the ionization state of the carboxylic acid moiety might also affect receptor binding. This was evaluated by performing competitive binding reactions for BDDA with ER{alpha} and ERß at pH 6.5 and pH 8.0 to compare with results obtained at the standard pH of 7.4. The binding reactions performed at the lower pH showed markedly increased relative binding affinities to both receptors vs. E2, whereas those at elevated pH showed a corresponding decrease in relative binding affinity to both ERs (Fig. 4Go). This pH-dependent decrease in relative binding affinity of (±)-Z-BDDA with both ER subtypes is observed despite a slight increase in the affinity between E2 and the receptors. That is, although increased pH increases the affinity between E2 and the receptors, it decreases the affinity between (±)-Z-BDDA and the receptors. The affinity changes observed for E2 suggest that there are effects of pH on the receptors themselves that would increase ligand binding. Yet, for (±)-Z-BDDA, rather than similar increases in binding affinity with pH, there are instead decreases in binding affinity. The magnitude of the relative binding affinity changes of (±)-Z-BDDA with pH corresponded closely to the fraction of (±)-Z-BDDA that is predicted by simple acid-base equilibria to remain undissociated at each pH (Table 2Go). Indeed, the relative binding affinity of the free acid fraction of (±)-Z-BDDA, based on an estimated pKa of 4.8 for the BDDA carboxyl group, is equivalent to or higher than E2 itself.


Figure 4
View larger version (29K):
[in this window]
[in a new window]
  		FIG. 4. Relative binding affinity of (±)-Z-BDDA to ER{alpha} and ERß varies with pH. Results from fluorescence polarization competition binding assays conducted at pH 6.5 (top), pH 7.4 (middle), or pH 8.0 (bottom) using ER{alpha} (left panels) or ERß (right panels) are shown as millianisotropy units (mA). Data are shown for E2 (bullet) and (±)-Z-BDDA ({blacksquare}) as means ± SEM from two independent experiments.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. pH Effects on relative binding of Z-BDDA

 
Binding and gene activation activities of the alcohol corresponding to (±)-Z-BDDA
One additional derivative of (±)-Z-BDDA was synthesized as a potential simple conversion of the parent compound that would not have an ionizable functional group, the primary alcohol formed by reduction of the carboxylic acid residue, (±)-Z-BDD-OH (V). Once again, competitive binding studies (Fig. 2Go) were performed followed by gene activation studies in cell culture (Fig. 3Go). For both ER{alpha} and ERß, (±)-Z-BDD-OH showed a marked increase in relative binding affinity to the receptors, and in this case, the activity of (±)-Z-BDD-OH in cell culture also showed a similar increase in biological potency, with effects showing a modest selectivity for ERß over ER{alpha}.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The ERs and the chemical compounds that interact with them have been studied extensively, reflecting the central role estrogens play not only in the physiology of growth, development, and reproduction but also in their influence on health and disease including cancer, heart disease, strokes, and osteoporosis. This diversity of function is matched by the extraordinary number of natural, synthetic, and environmental estrogens, comprised not only of steroids but also a large number of secosteroids and nonsteroidal molecules including diethylstilbestrol, hexestrol, triphenylethylene derivatives, flavones, isoflavones, coumestans, PCBs, DDTs, alkylphenols, and others (22). Most of these compounds share at least some common features, including phenolic rings, rigid structures, and significant hydrophobicity, and for the most part, biological activity is well correlated to in vitro binding affinity to the ERs.

Doisynolic acid derivatives, such as (±)-Z-BDDA, and the similar allenolic acid derivatives, are notable for a mismatch between in vitro binding, which is poor, and in vivo activity, which is high and similar to E2 (5, 7, 10). These compounds are also distinguished by the presence of a carboxylic acid moiety, unique among 230 estrogenic compounds recently reviewed (22). The current study was designed to address the role this carboxylic acid group may play in the estrogenic activity of (±)-Z-BDDA and to determine its potential to resolve the binding/activity paradox.

The structures of the ligand-binding pockets of both ER{alpha} and ERß have been determined and, consistent with their interactions with the natural hormone E2, are highly hydrophobic but also include hydrogen bonding to the 3-phenol and 17ß-hydroxyl groups of the hormone (23, 24). It is not known whether specific favorable interactions with receptor amino acid groups surrounding the pocket might accommodate a carboxylic acid moiety, but structurally, the carboxylic acid group may align well with the 17ß-hydroxyl of E2 in three-dimensional modeling of the ligand-receptor complex (22). Nonetheless, if the accommodation of a carboxylic acid group with a large dipole moment in a hydrophobic ligand pocket is relatively unfavorable energetically, this should result in a correspondingly lower binding affinity. In addition, derivatives of the compound that have greater or lesser polar properties would be expected to have lower or higher binding affinities, respectively, if the overall structures are similar enough that potential steric effects and the bonding interactions between ligand and receptor are similar.

We therefore chose to synthesize and evaluate the (±)-Z-BDDA methyl ester, because methyl esters are in general less polar than their corresponding carboxylic acids, and (±)-Z-BDDA amide, because such amides are in general more polar than the corresponding parent carboxylic acids. In support of the hydrophobic and apolar nature of the receptor ligand pocket, we observed the ester had a markedly higher affinity than the carboxylic acid, whereas the amide had a lower binding affinity than the acid.

The current study, in part, reevaluates and extends previous studies performed using crude uterine cytosolic extracts as the source of ER{alpha} (5). The previous receptor preparations likely contained esterase activity that rapidly hydrolyzes methyl esters to the corresponding carboxylic acid and methanol. The current competitive binding studies, obtained with purified recombinant preparations of ER{alpha} and ERß, readily distinguished (±)-Z-BDDA methyl ester from (±)-Z-BDDA and are in marked contrast to the previous studies that showed competitive binding results indistinguishable from that of the parent (±)-Z-BDDA (5). The improved binding of the (±)-Z-BDDA methyl ester did not result in improved transcriptional activity in cell culture, presumably again because of the lability of the ester group and the presence of ubiquitous esterases. The evaluation of (±)-Z-BDDA amide was considered because amide derivatives are markedly more stable in biological systems than esters, yet (±)-Z-BDDA amide bound even more poorly to the receptors than the parent compound, a result, in this case, also seen in the cell culture activity assay.

These data identify modification of the carboxylic acid moiety as one potential metabolic change that could dramatically alter the biological effects of (±)-Z-BDDA and potentially resolve the activity/binding paradox. Yet the chemistry of (±)-Z-BDDA, its unusual carboxylic acid group, and the ester and amide derivatives led to several additional questions regarding the nature of the observed binding/activity paradox. Among these questions are the effects of the acid-base equilibrium between carboxylic acid and carboxylate conjugate base on the bioavailability of (±)-Z-BDDA, both via binding and/or transport by serum binding proteins, and the effects that the carboxylic acid moiety may have on the ability of (±)-Z-BDDA to diffuse through cellular membranes and gain access to the cytosol or nucleus. In particular, it was not clear whether the carboxylic acid, its carboxylate salt, or both participated in receptor binding in vitro or in vivo. Simple acid-base equilibria suggest that for a presumed pKa of (±)-Z-BDDA of about 4.8, similar to other simple carboxylic acids, the compound is likely to be almost entirely (~99.75%) in the form of a carboxylate salt at the physiologically relevant pH of 7.4. This would mean that only about 0.25% of the compound would be in the form of the acid, and if all of the observed binding is a result of the acid and not the carboxylate, this would result in an approximately 400-fold underestimate of the actual or intrinsic binding affinity of (±)-Z-BDDA in the form of the acid (see Table 2Go) and even 2-fold higher for the isolated estrogenic enantiomer (–)-Z-BDDA (10). Were even a portion of the (±)-Z-BDDA converted in vivo to a derivative retaining this predicted high intrinsic affinity, it would be more than enough to account for the high biological activities observed and would resolve the binding/activity paradox. However, if it is only the carboxylic acid that interacts favorably with ERs, and not both the acid and conjugate base forms, the methyl ester of (±)-Z-BDDA might display higher binding affinity, not only because of a more favorable charge distribution but also because, unlike the parent compound, it cannot dissociate to form the carboxylate anion. Yet, the dominance of an acid-conjugate base equilibrium as determining the binding affinity of (±)-Z-BDDA and its derivatives is not supported by the results with the amide for which similar arguments also apply, and therefore one must also consider the role of dipole or charge distribution and steric effects.

In addition, if (±)-Z-BDDA carboxylate is evaluated in cell culture systems in the absence of possible metabolic conversion that may occur in vivo, this charged compound would be unlikely to freely diffuse across cell and nuclear membranes to gain access to nuclear ERs, possibly also contributing to reduced activity in cell culture systems, a result markedly different from the high activities observed in animals.

It is worth noting that carboxylic acid ester derivatives of E2 have been investigated as potential soft drugs, drugs that act locally but then are rapidly metabolized to avoid systemic effects (25, 26). These efforts sought to exploit ubiquitous esterases for the rapid conversion of higher-affinity, biologically active E2 esters to lower-affinity, relatively inactive E2 carboxylic acids. The investigators found two different types of compounds that failed as soft drugs because their estrogenic activities were high both locally and systemically (26). The first group included E2 7{alpha}-methylformate, a derivative resistant to enzymatic hydrolysis, which therefore is presumed to act systemically as the active ester. However, another compound, E2 11-methylacetate, was readily hydrolyzed by esterases to the corresponding carboxylic acid, yet it nonetheless exhibited high uterotropic systemic activity in vivo but with low relative binding affinity to ERs and a correspondingly low activity in Ishikawa cells, observations reminiscent of the activity/binding paradox exhibited by (±)-Z-BDDA and the findings we report here for its methyl ester. The presence of the carboxylic acid functional group on BDDA may also affect binding to serum proteins and pharmacological half-life in serum. It is interesting to note that the carboxylic acid modification of the E2 C-11 position have been reported to decrease the metabolic clearance and increase the biological potency of the modified compounds (26, 27). Low metabolic clearance and, therefore, a long duration of action have also been seen with doisynolic acids (28), providing one additional similarity between (±)-Z-BDDA and this unusually active E2 ester and further supporting metabolic effects as key to resolving both of these activity/binding paradoxes.

The phenomenon of SERM activity has attributed the observed tissue-specific effects of particular ligands to the relative abundance of coactivators and corepressors in each target organ combined with other features of the gene promoter and cellular environment that together determine coregulator recruitment (29, 30). Our previous studies have shown a unique spectrum of activities for BDDA and its enantiomers, including cardioprotective effects on reducing serum cholesterol, inhibition of weight gain, and decreases in the deposition of visceral fat, in reproductively intact male and female rats (11). In addition, the (+)-enantiomer exhibited these favorable metabolic effects in males, along with prostate-shrinking activity, without apparent feminization (11) or adverse effects on testis, sperm production, and other male accessory structures (31). Although these tissue-specific activities are certainly appropriately designated as SERM effects, our current studies suggest that prostate specificity might, in addition to favorable coregulator recruitment, also reflect the effects of locally acidic pH in the prostate gland. Indeed, rats and humans express H+ ATPase in the male reproductive tract that maintains a low luminal pH appropriate for sperm maturation (32, 33). Prostate lysosomal pH is also notably acidic (34), and studies have shown that the optimal pH for enzymes characteristic of the prostate gland, such as steroid 5-{alpha}-reductase (35) and prostatic acid phosphatase (36) is near a pH of 5.5, a pH that would also be more favorable for the undissociated state of the carboxylic acid group of BDDA for binding to receptors or for transmembrane diffusion based on acid-base equilibrium considerations. Acidic conditions may also favor distribution and activity of BDDA into tumors of the prostate or other organs, because tumors may also be more acidic than healthy tissues (37), an approach that may have applications in the development of other targeted therapeutic agents.

These studies demonstrate that relatively simple chemical modification of (±)-Z-BDDA and its distinctive carboxylic acid moiety, through acid-base equilibria, or via formation of the methyl ester, the amide, or reduction to the alcohol, can dramatically affect its binding affinity for ERs and its estrogenicity and gene regulatory potential. These data thereby support the hypothesis that similar biological modifications of (±)-Z-BDDA may explain the unexpectedly high in vivo activity and offer a potential resolution of the binding/activity paradox. In addition, these studies indicate not only future directions for definitive in vivo metabolic and pharmacokinetic studies of (±)-Z-BDDA and other possible estrogenic carboxylic acids but also provide additional insights into the possible chemical and metabolic bases of BDDA’s unique tissue selectivity and SERM activities. Furthermore, understanding how the chemistry of BDDA enables this distinctive spectrum of biological activities may lead to the design of other selective therapeutic agents.


   Acknowledgments
 
We give thanks to members of all our laboratories for helpful and insightful discussions.


   Footnotes
 
This publication was made possible by Grant R03 CA70515 from the National Cancer Institute, National Institutes of Health, under the National Action Plan on Breast Cancer and Grants R01 ES 08301, R01 ES 08301 02S2, and RO1 ES11125 from the National Institute of Environmental Health Sciences, National Institutes of Health. Additional support for the chemical syntheses and compounds used in these studies was provided by The Meyers Institute for Interdisciplinary Research in Organic and Medicinal Chemistry.

Author Disclosure Summary: M.A. and P.S. have nothing to declare. Y.H., C.Y.M., T.A.W., W.J.B., and S.A. are inventors on U.S. Patent No. 6,608,111. C.Y.M. is an inventor on U.S. Patent No. 5,420,161.

First Published Online May 18, 2006

Abbreviations: E2, Estradiol; ER, estrogen receptor; NMR, nuclear magnetic resonance; RSV, Rous sarcoma virus; SERM, selective ER modulator; TLC, thin-layer chromatography; Z-BDDA, Z-bisdehydrodoisynolic acid; Z-BDDA-OMe, 3-methyl ester of Z-BDDA.

Received March 10, 2006.

Accepted for publication May 11, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Miescher K 1948 On doisynolic acids, a new class of estrogens. Chem Rev 43:367–384[Medline]
  2. Anner G, Miescher K 1946 Hydrierungs - und umlagerungs-reaktion in der doisynolsäure - reihe. oestrogene carbonsäuren. XII. Helv Chim Acta 29:1889–1895
  3. Anner G, Miescher K 1947 Die totalsyntheses von racemischen doisynolsäuren. XXI. Über oestrogene carbonsäueren. Helv Chim Acta 30:1422–1432
  4. Tschopp E 1946 Wirksamkeit, organconzentration und ausscheidung der 7-methyl-bisdehydro-doisynolsäure. Helv Physiol Pharmacol Acta 4:401–410
  5. Meyers CY, Kolb VM, Gass GH, Rao BR, Roos CF, Dandliker WB 1988 Doisynolic-type acids: uterotropically potent estrogens which compete poorly with estradiol for cytosolic estradiol receptors. J Steroid Biochem 31:393–404[CrossRef][Medline]
  6. Meyers CY, Lutfi H, Adler S 1997 Transcriptional regulation of estrogen-responsive genes by non-steroidal estrogens: doisynolic and allenolic acids. J Steroid Biochem Mol Biol 62:477–489[CrossRef][Medline]
  7. Dandliker WB, Brawn RJ, Hsu ML, Brawn PN, Levin J, Meyers CY, Kolb VM 1978 Investigation of hormone-receptor interactions by means of fluorescence labeling. Cancer Res 38:4212–4224[Medline]
  8. Meyers CY 1995 Doisynolic acid type compounds as weight and appetite suppressing and control agents. U.S. Patent No. 5, 420:161
  9. Herbai G 1971 Separation of growth inhibiting potency from oestrogenicity in different weak oestrogenic drugs of various chemical structures. Acta Endocrinol (Copenh) 68:249–263[Abstract/Free Full Text]
  10. Meyers CY, Hou Y, Winters TA, Banz WJ, Adler S 2002 Activities of a non-classical estrogen, Z-bisdehydrodoisynolic acid, with ER{alpha} and ERß. J Steroid Biochem Mol Biol 82:33–44[CrossRef][Medline]
  11. Banz WJ, Winters TA, Hou Y, Adler S, Meyers CY 1998 Comparative effects of the selective estrogen receptor modulators (–)-, (+)- and (±)-Z-bisdehydrodoisynolic acids on metabolic and reproductive parameters in male and female rats. Horm Metab Res 30:730–736[Medline]
  12. Meyers CY 1984 The role of hormone receptors in estrogen activity is questionable. Trends Biochem Sci 9:473
  13. Heer J, Miescher K 1947 Über Steroide. LX. Weiltere Derivate der Doisynolsäuren: Aldehyde und Alkohole. Über oestrogene Carbonsäuren XVII. Helv Chim Acta 30:777–785
  14. Scatena CD, Adler S 1996 Trans-acting factors dictate the species specific placental expression of corticotropin-releasing factor genes in choricarcinoma cell lines. Endocrinology 137:3000–3008[Abstract]
  15. Tora L, Mullick A, Metzger D, Ponglikitmongkol M, Park I, Chambon P 1989 The cloned human oestrogen receptor contains a mutation which alters its hormone binding properties. EMBO J 8:1981–1986[Medline]
  16. Waterman ML, Adler S, Nelson C, Greene GL, Evans RM, Rosenfeld MG 1988 A single domain of the estrogen receptor confers DNA binding and transcriptional activation of the rat prolactin gene. Mol Endocrinol 2:14–21[CrossRef][Medline]
  17. Ogawa S, Inoue S, Watanabe T, Hiroi H, Orimo A, Hosoi T, Ouchi Y, Muramatsu M 1998 The complete primary structure of human estrogen receptor ß (hERß) and its heterodimerization with ER{alpha} in vivo and in vitro. Biochem Biophys Res Commun 243:122–126[CrossRef][Medline]
  18. Berthois Y, Katzenellenbogen JA, Katzenellenbogen BS 1986 Phenol red in tissue culture media is a weak estrogen: implications concerning the study of estrogen-responsive cells in culture. Proc Natl Acad Sci USA 83:2496–2500[Abstract/Free Full Text]
  19. Chen C, Okayama H 1987 High-efficiency transformation of mammalian cells by plasmid DNA. Mol Cell Biol 7:2745–2752[Abstract/Free Full Text]
  20. Olansky L, Welling C, Giddings S, Adler S, Bourey R, Dowse G, Serjeantson S, Zimmet P, Permutt MA 1992 A variant insulin promoter in non-insulin dependent diabetes (NIDDM). J Clin Invest 89:1596–1602[Medline]
  21. Eustice DC, Feldman PA, Colberg-Poley AM, Buckery RM, Neubauer RH 1991 A sensitive method for the detection of ß-galactosidase in transfected mammalian cells. Biotechniques 11:739–743[Medline]
  22. Fang H, Tong W, Shi LM, Blair R, Perkins R, Branham W, Hass BS, Xie Q, Dial SL, Moland CL, Sheehan DM 2001 Structure-activity relationships for a large diverse set of natural, synthetic, and environmental estrogens. Chem Res Toxicol 14:280–294[CrossRef][Medline]
  23. Brzozowski AM, Pike ACW, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the estrogen receptor. Nature 389:753–758[CrossRef][Medline]
  24. Pike AC, Brzozowski AM, Hubbard RE, Bonn T, Thorsell AG, Engstrom O, LjunggrenJ, Gustafsson JA, Carlquist M 1999 Structure of the ligand-binding domain of oestrogen receptor ß in the presence of a partial agonist and a full antagonist. EMBO J 18:4608–4618[CrossRef][Medline]
  25. Labaree DC, Reynolds TY, Hochberg RB 2001 Estradiol-16{alpha}-carboxylic acid esters as locally active estrogens. J Med Chem 44:1802–1814[CrossRef][Medline]
  26. Labaree DC, Zhang JX, Harris HA, O’Connor C, Reynolds TY, Hochberg RB 2003 Synthesis and evaluation of B-, C-, and D-ring-substituted estradiol carboxylic acid esters as locally active estrogens. J Med Chem 46:1886–1904[CrossRef][Medline]
  27. Zielinski JE, Yabuki H, Pahuja SL, Larner JM, Hochberg RB 1986 16{alpha}-[125I]Iodo-11ß-methoxy-17ß-estradiol: a radiochemical probe for estrogen-sensitive tissues. Endocrinology 119:130–139[Abstract]
  28. Scribner AW, Jonson SD, Welch MJ, Katzenellenbogen JA 1997 Synthesis, estrogen receptor binding, and tissue distribution of [18F]fluorodoisynolic acids. Nucl Med Biol 24:209–224[CrossRef][Medline]
  29. Shang Y, Brown M 2002 Molecular determinants for the tissue specificity of SERMs. Science 295:2465–2468[Abstract/Free Full Text]
  30. Smith CL, O’Malley BW 2004 Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr Rev 25:45–71[Abstract/Free Full Text]
  31. Meyers CY, Banz WJ, Adler SR, Winters TA, HouY, Dandliker WB 2003 Method for treating or preventing prostatic conditions. US Patent No. 6, 608:111
  32. Breton S, Smith PJ, Lui B, Brown D 1996 Acidification of the male reproductive tract by a proton pumping (H+)-ATPase. Nat Med 2:470–472[CrossRef][Medline]
  33. Herak-Kramberger CM, Breton S, Brown D, Kraus O, Sabolic I 2001 Distribution of the vacuolar H+ ATPase along the rat and human male reproductive tract. Biol Reprod 64:1699–1707[Abstract/Free Full Text]
  34. Lyubimov AV, Carr SN, Brown AP, Art JJ, Crowell JA, Levine BS 2005 Evaluation of hydrogen ion concentrations in prostates from rats and dogs using fluorescent confocal microscopy. J Photochem Photobiol B 80:225–234[CrossRef][Medline]
  35. Span PN, Smals AG, Sweep CG, Benraad TJ 1995 Rat steroid 5{alpha}-reductase kinetic characteristics: extreme pH-dependency of the type II isozyme in prostate and epididymis homogenates. J Steroid Biochem Mol Biol 54:185–192[CrossRef][Medline]
  36. Van Etten RL, McTigue JJ 1977 pH dependence and solvent isotope effects in the hydrolysis of phosphomonoesters by human prostatic acid phosphatase. Biochim Biophys Acta 484:386–397[Medline]
  37. Tannock IF, Rotin D 1989 Acid pH in tumors and its potential for therapeutic exploitation. Cancer Res 149:4373–4384




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/8/3952    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Adler, M.
Right arrow Articles by Adler, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Adler, M.
Right arrow Articles by Adler, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals