Differences Between Estrogen- and Antiestrogen-Estrogen Receptor Complexes from Human Breast Tumors Identified with an Antibody Raised Against the Estrogen Receptor1

Home , Antiestrogen, Nonsteroidal estrogen

[CANCER RESEARCH 44,1012-1018, March 1984]

Differences between Estrogen- and Antiestrogen-Estrogen Receptor Complexes from Human Breast Tumors Identified with an Antibody Raised against the Estrogen Receptor1

Anna C. Tate,2 Geoffrey L. Greene,3 Eugene R. DeSombre, Elwood V. Jensen,4 and V. Craig Jordan5

Department of Human Oncology, Wisconsin Clinical Cancer Center, University of Wisconsin, Madison, Wisconsin 53792 [A. C. T., V. C. J.J, and Ben May Laboratory lor Cancer Research, University of Chicago, Chicago, Illinois 60637 [G. L G., E. R. D., E. V. J.]

ABSTRACT INTRODUCTION

Radiolabeled estrogens 17/3-[3H]estradiol and diethylstilbestrol The nonsteroidal antiestrogens are triphenylethylene deriva ([3H]DES) and the antiestrogen [3H]monohydroxytamoxifen tives that inhibit the effect of estrogens. This inhibition occurs in ([3H]MHT) all bind with high affinity to the extranuclear estrogen normal estrogen target tissue as well as in estrogen-dependent receptor (ER) from the MCF-7 human breast tumor cell line (KÂ¿ malignant tissues (17) and has led to the widespread clinical use = 3 x 1(T10, 2 x 10.-10, and 0.63 x 1CT10M, respectively). A of the antiestrogen clomiphene for the treatment of infertility (12) polyclonal antibody raised in a goat to the calf nuclear ER and of the antiestrogen tamoxifen for the treatment of advanced selectively decreased the binding affinity and number of binding breast cancer (18). The subcellular mechanisms of action of the sites for 17j8-[3H]estradiol, but did not appear to affect these antiestrogens are, however, unclear (27). Both antiestrogens and binding parameters for [3H]MHT. In the presence of goat anti estrogens bind to the cytoplasmic ER6 and are translocated to body, the binding of the nonsteroidal estrogen DES was so the nucleus (3, 22), but antiestrogens inhibit or only partially perturbed that it was not possible to quantitate the decreased stimulate the nuclear events associated with estrogen action. number of binding sites or affinity of this compound as assessed Several possible subcellular sites of antiestrogen action have been proposed (17), and differences in the "activation" of the by Scatchard saturation analysis. These results were confirmed in human breast tumor cytosols extranuclear receptor (21, 24) and in the form of the nuclear by sucrose density gradient analysis. The binding of 17/3-[3H]- receptor (4) have been described. The presence of specific estradiol and [3H]DES to the ER was significantly reduced by "antiestrogen-binding sites" in mammalian tissue has also been preincubation with the polyclonal antibody, whereas the binding observed (28). At present, it is difficult to develop a single model of [3H]MHT was reduced only when the tumor cytosol was for antiestrogen action to incorporate all of these observations. preincubated with a very high concentration of antibody. At these [3H]Tamoxifen is available for studies of antiestrogens, but concentrations of antibody, the binding of 17/3-[3H]estradiol and this compound has relatively low binding affinity for the ER which [3H]DES to the receptor was prevented completely. In contrast, complicates the interpretation of biochemical studies. Recently, when the antibody was added to the tumor cytosol after the 3H- MHT, a metabolite of tamoxifen, has been synthesized in radio- ligand had bound to the receptor, the binding properties of all labeled form with high specific activity; this antiestrogen has an 3H-ligands were unaffected. The [3H]MHT-ER antibody complex affinity for the ER that is comparable to that of 170-estradiol (15) consistently sedimented as a higher-molecular-weight complex and has been used for studies both in vivo (14) and in vitro (1, on sucrose density gradients than did the corresponding estro- 29). The interactions of [3H]MHT with the cytosolic ER are genie complexes. qualitatively similar to those of 170-[3H]estradiol, although some The decrease in the affinity of estrogenic ligands can be differences have been observed (21). explained in part by an increase in the dissociation rate at 4Â°of Recently, a series of monoclonal and polyclonal antibodies these compounds from the ER. The dissociation rate of MHT raised to the ER have been developed that recognize different was unaffected by the goat antibody. antigenic determinants on the ER (9). These antibodies are, These results imply that there are important differences in the therefore, ideal probes to investigate steric or conformational binding of antiestrogen and estrogens to the tumor cytosol ER. changes on the ER that might be induced by antiestrogens but A ligand-binding model is proposed that may aid in the under not estrogens. Studies with monoclonal antibodies to the ER standing of antiestrogen action. have demonstrated no differences between 17/3-estradiol and MHT-estrogen receptor complexes from human breast tumors 1This study was funded in part by Grant P30-CA-14520 awarded to Wisconsin (29). Furthermore, a polyclonal antibody raised in the rabbit to Clinical Cancer Center and grants from I.C.I. Pic., Pharmaceuticals Division, Mac- the ER has been reported as able to distinguish between com clesfield, England, and an American Cancer Society Grant BC-86, at the University pounds with "high" and "low" affinity for the ER but not between of Chicago. 2 Recipient of a Fulbright-Hays Scholarship (1980/1981) and a University of estrogens and antiestrogens (5). Wisconsin Graduate School Scholarship in the Graduate Program of the McArdle In the present study, we have used a polyclonal antibody Laboratory for Cancer Research, University of Wisconsin. 3 Recipient of American Cancer Society Grant BC-86. raised in the goat which decreases the affinity and number of 4 Present address: Medical Director, Ludwig Institute for Cancer Research, 8001 binding sites for 17/3-[3H]estradiol in calf uterine cytosol (6). We Zurich, Switzerland. 5 To whom requests for reprints should be addressed, at Department of Human Oncology, Wisconsin Clinical Cancer Center, University of Wisconsin, 600 Highland * The abbreviations used are: ER, estrogen receptor; MHT, 1-[4-<2-dimethylami- Avenue, Madison, Wl 53792. Recipient of a grant from I.C.I., Pte., Macdesfield, noethoxy)phenyl]-1 -(4-hydroxyphenyl)-2-phenylbut-1 -ene; DES. diethylstilbestrol; England. DCC, dextran-coated charcoal; TEM, Tris (10 mM), EDTA (1.5 ITIM),and monothio- Received May 9,1983; accepted December 2, 1983. glycerol (5 mw), pH 7.4.

1012 CANCER RESEARCH VOL. 44

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1984 American Association for Cancer Research. Human Breast Tumor ERs have confirmed that this antibody produces a similar effect on fluid (Amersham) to determine the level of radioactivity. Specific binding the binding of 17/3-[3H]estradiolin the MCF-7 human breast was obtained by subtracting nonspecific binding from total bound radio tumor cytosol that is known to contain ER (11). We have ex activity. tended these studies by comparing and contrasting the effect of Determination of Dissociation Rate in the Presence of Goat Anti the goat antibody on the binding of [3H]DES and [3H]MHT to body. Cytosols from MCF-7 cells were preincubated with goat antibody (at approximately 50 pg antibody/pmol ER) for 2 hr at 4Â°.The cytosol human breast tumor ER. The effect of the antibody appears to was then transferred onto 10 *il 17/J-[3H]estradiol (5 nM), [3H]DES (5 nM), be selective for estrogenic compounds and indicates that there or [3H]MHT (10 nM) in ethanol and incubated for 2 hr at 4Â°;double the are different steric or conformational changes in the cytosol concentration of [3H]MHT as 17/3-[3H]estradiol was used, since only the receptor when complexed to antiestrogens compared with estro frans isomer (50% mixture) binds under these conditions (13). Nonspe gens. These findings form the basis for a proposed model of cific binding was determined in parallel incubations containing a 100-fold antiestrogen and estrogen binding to the ER. excess of cold ligand. The cytosol was then transferred onto DCC pellets formed from 1 ml of DCC suspension, and the tubes were agitated and incubated for 20 min at 4Â°and then centrifuged at 1000 x g for 10 min. MATERIALS AND METHODS The cytosols were transferred onto 10 n\ of ethanol or 10 n\ of a 1000- fold excess of the appropriate ligand to promote dissociation of bound 17/Ã-[6,7-3H]Estradiol (60 Ci/mmol in tolueneiethanol, 9:1) was ob ligand (23). At the time points indicated, bound and free ligand were tained from the Radiochemical Centre, Amersham, England, and was separated by removing 200-^1 aliquots of cytosol in duplicate onto 500 repurified on an LH-20 column by elution with 85% toluene: 15% meth- n\ of DCC suspension. The tubes were agitated for 10 min at 4Â°and anol. [3H]DES (112 Ci/mmol) in toluene was obtained from Amersham, then centrifuged at 1000 x g for 10 min to remove charcoal. Aliquots and its purity was checked by thin-layer chromatography developed with (100 fi\) were removed from each tube and transferred into 10 ml aqueous chloroform:ethyl acetate (85:15). Approximately 25% of the [3H]DES was the c/'s geometric isomer, and this percentage did not change appreciably counting scintillant scintillation fluid, and the radioactivity was determined in a Tracor analytical liquid scintillation counter. Specific binding was over a 2-week period. No attempt was made to separate the c/s and obtained by subtracting nonspecific binding from total bound radioactiv frans geometric isomers on this thin-layer chromatography system, since ity. the eluant was found to have significantly diminished binding capacity The inactivation of 17/3-[3H]estradiol-ER complex, during the course for the receptor. The [3H]DES was used as an approximately 75:25 of these assays, was assayed without the addition of a 1000-fold excess mixture of frans and c/s geometric isomers. [3H]MHT (42 Ci/mmol in of radioinert estradiol and was found to be negligible. The first order ethanol) was supplied as a 1:1 mixture of c/s and frans geometric isomers dissociation rate constant was determined from the slope of the plot of and was obtained from I.C.I. Pic., Macclesfield, Cheshire, England. The log percentage of specific 3H ligand bound versus time of incubation; this [3H]MHT was repurified with thin-layer chromatography by elution with value was not corrected for the slower dissociating activated species toluene:triethylamine:ethanol (90:10:0.1). [I4C]Ovalbumin (9.94 pCi/pg) which may have been present at 4Â°. was obtained from New England Nuclear (Boston, MA). Sucrose Density Gradient Analysis of Ligand-Receptor Com Radioinert 17/i-estradiol and DES, Tris base, EDTA, monothioglycerol, plexes. Breast tumor cytosol (400 /Â¿I)wasincubated with 10 n\ of "high" sucrose, and goat immunoglobulin were obtained from Sigma Chemical or "low" concentrations of goat antibody (30 ^g of antibody and 5 pg of Co., St. Louis, MO. MHT was a gift from I.C.I. Pic., Pharmaceuticals antibody, respectively) for 2 hr at 4Â°.The cytosol was then transferred Division, Macclesfield, England. All stock solutions in ethanol were stored onto 170-[3H]estradiol (5 nM), [3H]DES (5 nM), or [3H]MHT (10 nw) and at -20Â°. DCC was obtained from Wien Laboratories, Sucassuna, NJ. incubated for a further 2 hr at 4Â°.In a separate experiment, the antibody Lyophilized cytosols prepared from MCF-7 breast cancer cells were was added 2 hr after the ligands had been incubated with the cytosols. obtained from Dr. W. L. McGuire, San Antonio, TX. The cytosols were Nonspecific binding was determined in parallel incubations by adding 50 prepared with a receptor concentration of approximately 50 fmol/mg P\ of a 100-fold excess of DES in TEM buffer. The incubation mixtures protein. Protein concentration was increased in the cytosol by adding were transferred to preformed DCC pellets (produced from 1 ml of DCC bovine serum albumin. suspension), and the tubes were shaken intermittently by hand for 10 A purified goat antibody raised against the calf extranuclear ER was min at 4Â°and spun at 2300 x g for 10 min. A 200-^1 aliquot of the prepared as described previously (17). resulting supernatant was layered onto 10 to 30% or 10 to 50% hand- Tumor Cytosol Preparation. Breast tumor cytosols were prepared as layered sucrose gradients in TEM, and the gradients were centrifuged described previously (29). The tumor was finely minced and homogenized for 16 hr at 4Â°at 230,000 x g in a swinging bucket rotor AH650 in a in TEM buffer at 4Â° (1 g tumor/5 ml buffer). The homogenate was Sorvall OTD-65 centrifuge. Successive lOO-pl fractions were collected centrifuged at 100,000 x g for 1 hr at 4Â°,and the supernatant (cytosol) from the bottom with a Beckman fraction recovery system, and the was used immediately. radioactivity was determined in 3 ml of aqueous counting scintillant. Scatchard Equilibrium Binding Analysis. The influence of the goat [14C]Ovalbumin (3.6S) was used as an internal sedimentation marker in antibody on the ER from MCF-7 cytosol was determined by equilibrium the gradients. saturation analysis with increasing amounts (0.1 to 10 nw) of 17/3-[3H]- estradiol, [3H]DES, or [3H]MHT (cis:trans mixture; i.e., concentrations of the active frans isomer from 0.05 to 5 nw) (25). MCF-7 cytosols were RESULTS incubated with the goat antibody (-54 pg/pmol ER) or an equivalent concentration of control goat immunoglobulin for 2 hr, at 4Â°,and aliquots Equilibrium Binding Analysis. [3H]MHT, 17,i-|:)H|estradiol, of cytosol (400 n\) were pipeted onto 10 pi of 17/3-[3H]estradiol, [3H]DES, and [3H]DES all bound with high affinity to the ER in MCF-7 or [3H]MHT in ethanol to give the appropriate final concentration and cytosols, as demonstrated by Scatchard equilibrium saturation were then incubated for 20 hr at 4Â°.Specific binding was determined by analysis (Chart 1); where the kd values are 0.63 Â±0.13 x 10~10 adding 10 pi of a 100-fold excess of the appropriate cold competitor in (S.E.);3.0Â±0.17 x IO'10; and2.0 Â±0.45x IO'10 M,respectively; parallel incubations. In the case of [3H]MHT, a 100-fold excess of the these values are the means of 5 determinations. 170-[3H]Estra- frans isomer concentration was used. Bound and free 17/3-estradiol, diol (Chart 1/4)and [3H]MHT(Chart 1B)bound to a similar number DES, or MHT was separated by adding 500 Â¡Aof DCC suspension (0.25% Norit charcoal: 0.025% dextran) in TEM for 10 min. The tubes of sites in these cytosols (43 Â±5 and 46 Â±3 fmol/mg protein, were then centrifuged at 1000 x g for 10 min, and each supernatant respectively; however, [3H]DES (Chart 1C) appeared to bind to was decanted into 10 ml of aqueous counting scintillant liquid scintillation fewer sites (25 Â±9 fmol/mg protein). The preincubation of

MARCH 1984 1013

.20

B/F .6 -

.20

.15 vÃ©-C3H]DES -Ab B/F B/F .10

.05

io 20 30 5 10 15 20 25 IO 20 30 40 fmol /mg Protein fmol /mg Protein fmol /mg Protein Chart 1. Influence of the goat antibody on the binding of (A) 17/3-t3H]estradiol <[3H]Â£2).(B)[3H]DES, and (C) [3H]MHT to the ER from MCF-7 cytosols, determined by equilibrium saturation analysis with increasing amounts of radiolabeted Ikjand (19). MCF-7 cytosol (400 pi) was incubated with the antibody (Ao) (=54 ^g/pmol ER) (A, O) or an equivalent concentration of control goat immunoglobulin (â€¢)for2 hr at 4Â°.Aliquots of cytosol (400 pi) were pipetted onto 10 Â¡A3H-ligand in ethanol to give the appropriate final concentration and incubated for 20 hr at 4Â°.Specific binding was determined in parallel incubations by adding a 100-fold excess of the appropriate radioinert competitor. Bound and free ligands (B/F) were separated by adding 500 pi of a DCC suspension for 10 min. Tubes were centrifugea at 1000 x g for 10 min, and each supernatant was decanted into 10 ml aqueous counting scintillant liquid scintillation fluid (Amersham). Specific binding was obtained by subtracting nonspecific binding from total bound radioactivity. The range of nonspecific binding as percentage of total was [3H]DES, 35 to 70%; [3H]MHT, 35 to 75%; and 17/3-[3H]estradiol, 15 to 30%. cytosols with the goat antibody caused a decrease in the affinity binding sites or affinity. Nevertheless, the large decrease in these of 17j8-[3H]estradiol (K, = 5 x 10~10M) and number of binding parameters for [3H]DES in the presence of the goat antibody sites (25 Â±4 fmol/mg protein) present in the MCF-7 cytosol, but suggests that the antibody effect is selective for high-affinity had no effect on the binding of [3H]MHT. Preincubation of the estrogenic compounds and, since DES is a nonsteroidal estro cytosol with the goat antibody obliterated the linear Scatchard gen, the lack of effect of the antibody on [3H]MHT cannot be plot obtained with [3H]DES and produced scattered points from ascribed to its nonsteroidal nature. which it was impossible to determine the decreased kÂ«or the Sucrose Density Gradient Analysis. The interaction of the number of binding sites present. There was no difference in any goat antibody with the human breast tumor ER was determined of the Scatchard plots in the presence or absence of normal on sucrose density gradients. Preincubation of ER with a low (5 goat immunoglobulin. However, the affinity of [3H]DES is lower Â¿Â¿g)concentrationof antibody decreased the binding of 170- than that previously determined in mouse mammary tissue (20). [3H]estradiol to approximately 48% of that obtained in the pres A major problem with the use of [3H]DES is the rapid conversion ence of control immunoglobulin (Chart 2A). However, the inter to an equilibrium mixture of c/s and frans geometric isomers (31) action of the antibody with the [3H]MHT-ER complex in this so that it is difficult to assess the absolute amount of active frans cytosol decreased the binding only to 85% of that obtained in isomers present. Furthermore, [3H]DES has high nonspecific the presence of control immunoglobulin (Chart 2B). This finding binding in these cytosols, similar to that observed with [3H]- was reproducible for 5 different breast tumor cytosols. Interest dimethylstilbestrol (2), which leads to an underestimate of the ingly, the [3H]MHT-ER goat antibody complex appears to sedi binding affinity and number of binding sites of these compounds ment consistently as a higher molecular weight complex than and is reflected in the larger S.E. of the results when [3H]DES is the [3H]estrogen-antibody complexes. used. This problem has been described previously for [3H]DMS The [3H]MHT-ER-antibody complex also sedimented nearer (2). Thus, the Scatchard plot is, at best, only an estimate of the bottom of the gradient when the breast tumor cytosols were

1014 CANCER RESEARCH VOL. 44

preincubated with the ligands for 2 hr at 4Â°and then incubated 100 with the goat antibody (Chart 3). However, under these condi tions, no reduction in the total binding of 17/3-[3H]estradiol, [3H]- 90 DES, or [3H]MHT was produced by the goat antibody. When high concentrations of antibody (30 ^g) were preincu 80 bated with cytosols, the specific binding of 17/3-[3H]estradiol or C7> C [3H]DES was decreased to virtually nonspecific levels (Chart 4, l70 A and ÃŸ).Although the binding of [3H]MHT was also decreased DÃ» (Chart 4C), a [3H]MHT-ER-antibody complex was still observed E 60 towards the bottom of the gradient. This complex sedimented I at a position similar to those observed in gradients described K 50 o previously (Charts 2 and 3). 5 Analysis of Dissociation Rate. The decrease in affinity of ^ 40 [3H]DES and 17/3-[3H]estradiol in the presence of goat antibody o could be the result of a change in the association or dissociation v 30 rate of the ligands, or both. Preliminary data7 suggested that O. neither the association rate of [3H]MHT nor that of 17/3-[3H]- 20 estradiol was altered at 4Â°in the presence of the goat antibody. However, the association rate was determined at a single con IO centration of ligand and ER, and further studies were limited by supply of antibody. The first-order dissociation rate constant was determined from Bottom IO 20 30 Top Fraction Number the slope of the plot of log percentage bound versus time of incubation (23, 30). When the cytosol was preincubated with the goat antibody (Chart 5), an increase in the first-order dissociation rate constant of (Chart 5C) [3H]DES (k_, 1.3 x 1Q-5 sec~1 compared with 1.86 x 10~5 see"1 in the presence of antibody) 100 and (Chart 5B) 17/3-[3H]estradiol (k_, 1.6 x 1Q-5 sec"1 compared with 1.9 x 10~5sec~1 in the presence of antibody) was observed 90 at 4Â°.The dissociation rate constant of [3H]MHT was essentially [3H]MHT unchanged in the presence of antibody (k-i 1.6 x 10~5 sec"1) 80 (Chart 5/4). These values are not corrected for slower dissociating 70 species which may have been revealed if the time points had been extended. Nevertheless, the data demonstrate that the 60 antibody has an effect on ligand dissociation. C3H]MHT+Ab When the antibody was added after preincubation with 170- [3H]estradiol or [3H]MHT, there was no alteration in the dissocia 50 tion rate (1.6x10~5 sec"1).

40 DISCUSSION 30 The interaction of ER antibodies 170-[3H]estradiol-ER com plexes has been extensively investigated with Scatchard and 20 sucrose density gradient analysis (7, 8). In this study, these techniques were used to investigate the interaction of the goat IO MHT + Ab +DES antibody with ligand-ER complexes from human breast tumor cytosols. 170-[3H]Estradiol and MHT displayed high affinity for the ER from MCF-7 cytosols (K<,= 3.0 x 10~10and 0.6 x 10"1Â° Bottom 10 20 30 Top Fraction Number M, respectively) and bound to a similar number of specific sites in the cytosol. Preincubation of these cytosols with goat antibody Chart 2. Effect of preincubation of goat antibody on the binding of (A) 17/3- [3H]estradiol (Â¡3HJE,)or (B) [3H]MHT to the ER from human breast tumor, demon did not alter either of these parameters for [3H]MHT; conversely, strated on sucrose density gradients. Cytosol (400 /if) was incubated with 5 ^g both the affinity and number of binding sites for 170-[3H]estradiol antibody (Ab) (â€¢,â€¢A) or control antibody (O, A) for 2 hr at 4Â°and was then transferred onto 10 fil ethanol containing 3H-ligands and incubated for a further 2 were decreased by preincubation with the antibody. This is hr at 4Â°.The concentrations were 170-[3H]estradiol, 5 nwi, and [3H]MHT, 10 nw. consistent with the effect the goat antibody has on 17/3-[3H]- Nonspecific binding was determined in parallel incubations by adding 50 n\ of a 100-fold excess of DES in buffer. The incubation mixtures were transferred onto estradiol binding in calf uterine cytosols (6). preformed DCC pellets and shaken for 10 min at 4Â°and spun at 2300 x g for 10 In Chart 6, we have summarized our results to propose a min. A 200-iil aliquot of the resulting supernatant was layered onto 10 to 50% model to explain the selective interaction of this antibody with sucrose gradients in TEM and centrifuged for 16 hr at 4Â°at 230,000 x g in a swinging bucket rotor AH650 in a Servali OTD-65 centrifuge. Successive 100-/il the ER; the model describes events taking place only at the fractions were collected from the bottom of the tube and counted in 3 ml of aqueous ligand binding site, which is probably a small part of the total counting scintillant. ["CJOvalbumin (3.6S) was used as an internal sedimentation marker on the gradient. 7A. C. Tate and V. C. Jordan, unpublished observations.

MARCH 1984 1015

ivi 16 O [SHJDES ['HJMHT o 12 m ['H: DESÂ»AD [5H]MHTÂ»Ab 5 20 I S e O- 8 i

Bottom IO 20 io 20 IO 20 Fraction Number Fraction Number Fraction Number Charts. Influence of postincubation with goat antibody on the binding of (A) 170-[3H]estradiol ([3H]E2), (B) [3H]DES, and (C) [3H]MHT to human breast tumor cytosol ER, demonstrated on sucrose density gradients. Cytosol was incubated with 17/3-[3H]estradiol (5 hM), [3H]DES (5 niw), or [3H]MHT (10 HM)for 2 hr at 4Â°.Goat antibody (Ab) (5 fig) (A, â€¢â€¢)orcontrol immunoglobulin (A, D, O) was then added to the incubates and incubated for 2 hr at 4Â°.The incubation mixtures were transferred onto preformed DCC pellets, shaken for 10 min at 4Â°,and spun at 2300 x g for 10 min. A 200-nl aliquot of the resulting supernatant was layered onto 10 to 50% sucrose gradients. Conditions of centrifugation, radioactivity measurements, and determination of nonspecific binding were as described in the legend to Chart 1. Nonspecific binding, determined with 100-fold excess of DES, produced values of 20% 17/3-[3H]estradiol, 50% [3H]DES, and 50% [3H)MHT total values.

to 6 b

c 2 9

5 Q_ Q

C'HIMHTÂ«AbÂ«DES

Bottom 10 20 SO To* Bottom 10 20 30 Top 10 20 30 Top Fraction Number Fraction Number Fraction Number Chart 4. Influence of preincubation of goat antibody (Ab) (30 ^g) on the binding of (A) 17/3-[3H]estradiol (Â¡3H]E2),(B)[3HlDES, and (C) [3H]MHT to the human breast tumor ER on sucrose density gradients. The experimental details were as described in the legend to Chart 2.

receptor protein. When estrogens bind to the receptor, there are If the estrogenic ligand has already induced secondary changes in the ER, then this "activated" or "locked" form is minor changes on the tertiary structure of the protein, so that the estrogens become "locked" (Chart 6) into the ligand binding insensitive to conformational changes induced by the antibody. site. This decreases the dissociation rate of estrogens from the This implies that these antigenic determinants are relatively receptor and may be an event similar to the increase in affinity inaccessible after tertiary changes on the ER (Chart 6) or that, of the activation of the 170-estradiol-ER complex described in once "locking" occurs, the antibody cannot reverse the process the calf uterus (30). Some fraction of the polyclonal goat antibody because the receptor conformation has changed. recognizes antigenic determinants on the protein that inhibit High concentrations of the antibody (30 ^9), which almost "locking," and the planar estrogens remain exposed and disso totally dissociate 17,8-estradiol and DES from the ER, also cause ciate from the receptor. Antiestrogens, on the other hand, by a significant decrease in the specific binding of [3H]MHT. How virtue of their unique alkyl-amino-ethoxy-side chain, are ever, a binding complex is still observed on sucrose density "wedged" into the receptor so that the secondary changes that gradient; this indicates that, even at these concentrations of antibody, the antiestrogen "wedging" is sufficient to retain some occur with estrogens cannot occur with antiestrogens because of stehe hindrance; since preincubation with antibody prevents of the ligand on the ER. these changes, antiestrogen binding would not be affected. The The interpretation of our data to formulate this model is shape of the antiestrogen, with its multiple points of interaction consistent with recent data reported by Rochefort and Borgna on the ER, does not permit rapid dissociation, and the antiestro- (21). They have found that potassium molybdate, that apparently prevents receptor activation ("ligand locking"), does not affect genic ligand will remain bound to the receptor. This would not only explain the lack of effect of the antibody on [3H]MHT binding, the slow dissociation of MHT from the ER. In contrast, molybdate but also would be consistent with the requirement for the cor alters the slow dissociation of 17/3-estradiol from activated re rectly positioned alkyl-amino-ethoxy side chain for antiestrogenic ceptors to the rapid dissociation observed for unactivated recep activity in vivo (16) and in vitro (19). tors.

1016 CANCER RESEARCH VOL. 44

Proposed Model

^ESTROGENRECEPTOR development of ['intrinsic activity" I (transformation?)\ 40 C3H]MHT 0> S

20 P.Â», .â€žQinr , Ligand Locking by Changes Biologically Lstrogen umamg in Tertiary Structure Active Complex

100 B Effect of Ab Preincubation 0 80 o> c 60 T3 1 -o E 3 C3H3 Estrodiol .Â§ 20 1.Interaction near ligand x binding site O 2. Interaction to cause a change in ligand bind ing site â€¢*-100 Â° 80 Ligand Wedging Â£ 60 o | 40 Effect of Ab After Ligand Binding C3H] DES 20 Interaction to produce no Ab-s effect on Ligand affinity -*.

4 6 10 Time (hr) Charts. Dissociation rate of (A) [3H]MHT, (ÃŸ)17,8-[3H]estradiol, and (C) ["HI Estrogen Antiestrogen DES from the MCF-7 cytosol ER in the presence of goat antibody (/U>)(â€¢,A, â€¢) Ligand Locked Ligand Wedged or control immunoglobulin (O, A, G). MCF-7 cell cytosols were incubated with goat antibody (approximately 50 Â¿igantibody/pmol ER) for 2 hr at 4Â°.The cytosol was then transferred onto 10 nl 17/3-[3H]estradiol, [3H]DES, or [3H]MHT in ethanol to Chart 6. Effect of goat polyclonal antibody (Ab) on the binding of estradici and give final concentrations of 5, 5, and 10 nu, respectively, and incubated for 2 hr at monohydroxytamoxifen to the ligand-binding site on the ER. 4Â°.Nonspecific binding was determined in parallel incubations containing a 100- fold excess of cold ligand. Nonspecific values were 30% 17/3-[3H]estradiol, 65% [3H]MHT, and 65% [3H]DES of total values. Free ligand was removed by treatment with DCC. The cytosols were then transferred onto 10 n\ of a 1000-fold excess of the appropriate ligand in ethanol. At the time points indicated, aliquots of cytosol Although these results can be used to support a receptor were removed and bound, and free ligand and free hormone were separated by model for the binding of estrogens and antiestrogens at a single treatment with DCC. Aliquots (100 *J) were removed and transferred into 10 ml aqueous counting scintillation fluid. Specific binding was obtained by subtracting binding site (19), it is also possible that these data are describing nonspecific binding from total bound radioactivity. The dissociation rate uncorrected an allosteric interaction of estrogens and antiestrogens with the for faster dissociation species of the ligands was determined from the slope of the receptor. In this alternative model, the polyclonal antibody would plot of log percentage of maximum specific binding (at time zero) versus time. prevent estrogen binding, but the antiestrogen site on the ER would remain unoccupied and free to bind antiestrogenic ligands (10). Only further studies can resolve this issue. Throughout these studies, we observed that [3H]MHT-ER- Recently, there have been reports of a selective antiestrogen binding site in the cytosol of tumor and normal tissue (28) that is antibody complex sedimented as a higher molecular weight distinct from the ER which may have a role in the mechanism of complex on sucrose density gradients than the corresponding estrogenic complex. Since the 170-[3H]estradiol-ER and [3H]- antiestrogen action. In contrast, it has been suggested that this MHT-ER complex sediment at similar positions prior to the site may play a role in antiestrogen action by altering the distri bution and pharmacokinetics of antiestrogens in vivo (26). How addition of antibody, the increase in molecular weight of the ever, the data reported in the present study indicate that estro complexes may be due to increased antibody binding to the MHT-ER complex compared with 17/3-estradiol-ER, or to in gens and antiestrogens can differentially alter the ER in breast creased aggregation of the [3H]MHT-ER complex in the presence tumor cytosols; this may help to explain the competitive action of these ligands in modulating the growth of breast cancer cells. of the antibody. This observation is a further indication of differ ences in the form of the receptor when bound to antiestrogens compared with estrogens. It is interesting to note that the nuclear [3H]MHT-ER complex from MCF-7 breast cancer cells has been ACKNOWLEDGMENTS reported (4) to sediment at a higher molecular weight than nuclear 17/3-[3H]estradiol-ER complexes. We are grateful to Dr. A. H. Todd for the regular dispatch of [3H]MHT to Madison, Wl, from I.C.I. Pic., Macclesfield, England.

MARCH 1984 1017

REFERENCES 16. Jordan, V. C., and Gosden, B. Importance of the alkylaminoethoxy side-chain for the estrogenic and antiestrogenic actions of tamoxifen and trioxifene in the immature rat uterus. Mol. Cell. Endocr., 27: 291-306,1982. 1. Borgna, J. L., and Rochefort, H. High affinity binding to the estrogen receptor 17. Katzenellenbogen, B. S., Bhakoo, H. S., Ferguson, E. R., Lan, N. C., Talee, Of |JH|-4-hydroxytamoxifen. an active antiestrogemc metabolite. Mol. Cell. T., Ten-Ã¼n,S. T., and Katzenellenbogen, J. A. Estrogen and antiestrogen Endocr., 20: 71-85,1980. action in reproductive tissues and tumors. Recent Prog. Homi. Res., 35: 259- 2. Capony, F., and Rochefort, H. In vitro and in vivo interactions of [3H]dimethyl- 300,1979. stilbestrol with the estrogen receptor. Mol. Cell. Endocr., 8: 47-64,1977. 18. Legha, S. S., and Carter, S. K. Antiestrogens in the treatment of breast cancer. 3. Clark, J. H., Anderson, J. N., and Peck, E. J. Estrogen receptor-antiestrogen Cancer Treat. Rev., 3: 205-216,1976. complex: atypical binding by uterine nuclei and effects on uterine growth. 19. Lieberman, M. E., Gorski, J., and Jordan, V. C. An estrogen receptor model Steroids, 22:707-717,1973. to describe the regulation of prolactin synthesis by antiestrogens in vitro. J. 4. Eckert, R. L., and Katzenellenbogen, B. S. Physical properties of estrogen Biol. Chem., 258: 4741-4745,1983. receptor complexes in MCF-7 human breast cancer cells. J. Biol. Chem., 257: 20. Okey, A. B., and Bondy, G. P. Diethylstilbestrol binding to estrogen receptor 8840-8846,1982. in mammary tissue and uteri of C3H/HeJ mice. J. Nati. Cancer Inst., 67; 719- 5. Garcia, M., Greene, G., Rochefort, H., and Jensen, E. V. Effect of antibodies 724, 1978. to estrogen receptor on the binding of 3H-labeled antiestrogens and andro- 21. Rochefort, H. M., and Borgna, J. L. Differences between oestrogen receptor stanediol in the uterus. Endocrinology, 110:1355-1361,1982. activation by oestrogen and antioestrogen. Nature (Lond.), 292: 257-259, 6. Greene, G. L., Ctoss, L. E., DeSombre, E. R., and Jensen, E. V. Antibodies to 1981. 22. Rochefort, H., and Capony, F. Etude comparÃ©edu compartement d'un antioes estrophilin: comparison between rabbit and goat antisera. J. Steroid Biochem., 11: 333-341,1979. trogÃ¨ne et de l'oestradiol dans les cellules utÃ©rines.C. R. Acad. Sci. (Paris), 7. Greene, G. L., Closs, L. E., Fleming, H., DeSombre, E. R., and Jensen, E. V. 276:2321-2324,1973. Antibodies to estrogen receptor: immunochemical similarity of estrophilin from 23. Sanbom, B. M., Rao, R. B., and Korenman, S. G. Interaction of 17/3-estradicd various mammalian species. Proc. Nati. Acad. Sei. U. S. A., 74: 3681-3685, and its specific uterine receptor. Evidence for complex kinetic and equilibrium 1977. behavior. Biochemistry 70: 4955-4961,1971. 24. Sasson, S., and Notides, A. C. The inhibition of the estrogen receptor's positive 8. Greene, G. L., Fitch, F. W., and Jensen, E. V. Monoclonal antibodies to estrophilin: probes for the study of estrogen receptors. Proc. Nati. Acad. Sci. cooperative [3H]estradiol binding by the antagonist, clomiphene. J. Bid. Chem., U. S. A., 77:157-161,1980. 257: 11540-11545, 1982. 9. Greene, G. L., and Jensen, E. V. Monoclonal antibodies as probes for estrogen 25. Scatchard, G. The attraction of proteins for small molecules and ions. Ann. N. receptor detection and characterization. J. Steroid Biochem., 16: 353-359, Y. Acad. Sci. 57. 660-672,1949. 1982. 26. Sudo, K., Monsma, F. J., and Katzenellenbogen, B. S. Antiestrogen binding 10. Hahnel, R., Twaddle, E., and Ratajczak, T. The influence of synthetic anties sites distinct from the estrogen receptor: subcellular localization, ligand spec trogens on the binding of tritiated estradici-17.Â¡ by cytosols of human uterus ificity and distribution in tissues of the rat. Endocrinology, 772:425-434,1983. and human breast carcinoma. J. Steroid Biochem., 4: 687-695,1973. 27. Sutherland, R. L., and Jordan, V. C. Non-Steroidal Antioestrogens. Sydney: 11. Horwitz, K. B., Costlow, M. E., and McGuire, W. L. MCF-7: a human breast Academic Press, 1981. cancer cell line with estrogen, androgen, progesterone, and glucocorticoid 28. Sutherland, R. L., Murphy, L. C., Foo, M. S., Green, M. D., Whyboume, A. M., receptors. Steroids, 26: 785-794,1975. and Krozowski, 2. S. High affinity antiestrogen binding sites distinct from the 12. Huppert, L. C. Induction of ovulation with clomiphene citrate. FÃ©rtil.Steril., 31, oestrogen receptor. Nature (Lond.), 288: 273-275,1980. 1-8,1979. 29. Tate, A. C., DeSombre, E. R., Greene, G. L., Jensen, E. V., and Jordan, V. C. 13. Jordan, V. C. Hydroxylated antioestrogens: new pharmacological probes to Interaction of [3H]estradiol- and [3H]monohydroxytamoxifen-estrogen receptor investigate oestrogen and antioestrogen action. In: M. K. Agarwal, (ed.), complexes with a monoclonal antibody. Breast Cancer Res. Treat., 3: 267- Antihormones, pp. 109-128. Berlin: Walter DeGruyter, 1982. 277, 1983. 14. Jordan, V. C., and Bowser-Finn, R. A. Binding of [3H]monohydroxytamoxifen 30. Weichman, B. M., and Notides, A. C. Estradiol-binding kinetics of the activated by immature rat tissues in vivo. Endocrinology, 770:1281-1291,1982. and nonactivated estrogen receptor. J. Biol. Chem., 252: 8856-8862,1977. 15. Jordan, V. C., Collins, M. M., Rowsby, L., and Prestwich, G. A monohydroxy- 31. Winkler, V. W., Nyman, M. A., and Egan, R. S. Diethylstilbestrol cis-trans lated metabolite of tamoxifen with potent anttestrogenic activity. J. Endocrino!., isomerization and estrogen activity of diethylstilbestrol isomers. Steroids. 77: 75:305-316,1977. 197-207,1971.

1018 CANCER RESEARCH VOL. 44

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1984 American Association for Cancer Research. Differences between Estrogen- and Antiestrogen-Estrogen Receptor Complexes from Human Breast Tumors Identified with an Antibody Raised against the Estrogen Receptor

Anna C. Tate, Geoffrey L. Greene, Eugene R. DeSombre, et al.

Cancer Res 1984;44:1012-1018.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/44/3/1012

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/44/3/1012. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.