tent anti-estrogen. The following results were ob- tained. 1) The antibody prevented the decrease in the dissociation rate of the receptor-estradiol complex.
Vol. 260, No. 29,Issue of December 15,pp. 15547-15553,1985 Printed in U.S.A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Societyof Biological Chemists, Inc.
A Monoclonal Antibody to theEstrogen Receptor Inhibitsin Vitro Criteria of Receptor Activation byan Estrogen and anAnti-estrogen* (Received for publication, January 31, 1985)
Joel Fauque, Jean-Louis BorgnaS, and Henri Rochefort From the Unite d’Endocrinologie Celluluireet Mokculaire, Institut National de la Santt? et de la Recherche MedicaleW 148, 60 Rue de Nauacelles, 34100 Montpellier, France
The effect of an IgM class monoclonal antibody (B36) (Greene, G. L., Fitch, F. W.,and Jensen, E. V. (1980) Proc. Natl. Acad. Sei. U. S.A. 77, 157-161) raised against the calf uterine estrogen receptor was tested in vitro on certain parameters of estrogen receptor activation by estradiol or 4-hydroxytamoxifen, a potent anti-estrogen. The following results were obtained. 1) The antibody prevented the decrease in the dissociation rate of the receptor-estradiol complex which results from activation of the complex, whereas it did not affect the dissociation rate of the receptor-4hydroxytamoxifen complex, whichremains unchanged upon activation. 2) The antibody also increased the dissociation rate of the preactivated receptor-estradiol complex. 3)The antibody protected the naked estrogen receptor against heat-inactivation. 4) B36 partially inhibited the binding of the estradiol- and ri-hydroxytamoxifen-receptor complexes to DNA adsorbed onto cellulose, but did not reverse the receptor-DNA binding. This inhibition was not overcome byhigher DNA concentrations and was more pronounced for the $eceptor interacting with estrogen than with anti-estrogen. All these effects were specific since they were related to antibodylantigen recognition and were dosedependent. These results indicate that the binding of the antibody to theestrogen-activated receptor induces a conformational change in the receptor and that the antibody can preventand overcome the effect of activation whatever its mechanism. Theyalso confirm that the conformations of the estrogen receptor differ when bound to estradiol or to 4-hydroxytamoxifen.
Activation of steroid hormone receptors can be defined as the hormone-dependent process which, in target cells, transforms receptors from an inactive to a biologically active state. The current major operational indication of receptor activation is the in vivo transformation of the receptor from a cytosolic form into a “nuclear” form, which is more tightly bound to nuclear structures. Nonactivated cytosolic receptors can also be transformed in vitro by hormones, in a process which is accelerated by heatandsalt,into forms which display properties (DNA
* This work was supported by the InstitutNational de la SantB et dela Recherche MBdicale and the Caisse Nationale #Assurance Maladie des Travailleurs Salariis, the University of Montpellier I, the Fondation pour la Recherche MBdicale Franpise, and theAssociation pour le DBveloppement de la Recherche sur le Cancer. The costs of publication of this article were defrayed in part by the payment of page charges. This articlemust therefore be hereby marked ‘‘advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ To whom correspondence should be addressed,
binding, chromatographic and kinetic behavior) very similar to those of activated in vivo receptors extracted from nuclei of target tissues after hormone administration. Therefore, these changes in receptor properties, elicited in vitro by hormones plus heat or salt, arewidely used as criteria for receptor activation or transformation. In the case of the estrogen receptor, the two major criteria used to assess the in vitro activation of the cytosolic estrogen receptor are the increase in its affinity for DNA,chromatin, or nuclei (Yamamotoand Alberts, 1972; Milgrom, 1983),and the increase in its affinity for estrogens, evidenced bya decrease in thedissociation rate of estrogens from their receptor site (Weichman and Notides, 1977). The transformation of a 4 S form into a 5 S form in high salt-sucrose gradients, has also been used to estimate estrogen receptor activation (Jensen et al., 1971; Notides et al., 1975). The two first criteria seem particularly relevant since anti-estrogens, which usually display little or no estrogen agonistic activity, are less efficient in promoting the binding of the receptor to DNA (Katzenellenbogen et al., 1981;Evans et al., 1982)and do not show any marked decrease in their dissociation rate from the receptor (Rochefort and Borgna, 1981; Ruh et al., 1983). The third criterion appears to be less valid since it is notrequired forreceptor binding to nuclei (Bailly et al., 1980) nor for increasing the affinity of the receptor for estrogen (Sakai and Gorski, 1984). Moreover anti-estrogens are as efficient as estrogens (Rochefort et al., 1983; Geynet et d., 1983) in promoting the 4 to 5 S shift of the receptor in high salt-sucrose gradients. The use of monoclonal antibodies to steroid receptors as probes for studying the mechanism of receptor activation is potentially interesting as long as one can find an antibody which interferes with the activation process without preventing hormone binding.The B36 monoclonal antibody prepared against the calf uterine estrogen receptor (Greene et al., 1980) appears to be such a probe since this antibody is more reactive with the nuclear than with the cytosolic receptor-estrogen complex (Greene et al., 1980). In a previous paper, we quantified the degree of interaction of the B36 monoclonal antibodywith the estrogen receptor, whichwas activated or molybdate-stabilized,and bound to estradiol or 4-hydroxytamoxifen, a high affinity anti-estrogen (Borgna and Rochefort, 1980) formedin vivo from tamoxifen and selectively retained in estrogen target tissues, bound to the estrogen receptor (Borgna and Rochefort, 1981). We foundthat themonoclonal antibody had a higher affinity for the activated receptor than for the nonactivated receptor and could discriminate between the receptor bound to estradiol or to 4-hydroxytamoxifen (Borgna et al., 1984). In an attemptto approach the mechanism of receptor activation, which remains totally unknown in molecular terms, we here characterize the consequences of the binding
15547
15548
Antibody Estrogen and
of the B36 monoclonal antibody to the calf uterine estrogen receptor on two portions of the estrogen receptor which are modified during its activation: (i) the hormone binding domain interacting with estrogen or anti-estrogen and (ii) the DNA binding domain (Andrh and Rochefort, 1973).
Activation Receptor
la
R-E2
b
R-OHJ
I
EXPERIMENTAL PROCEDURES' RESULTS
The B36 Antibody Prevents theDecrease in the Dissociation Rate of the Receptor-Estradiol Complex Induced by Activation-We have shown that an anti-estrogen is unable to activate the estrogen receptor with respect to thedecrease in the dissociation rate of the complex (Rochefort and Borgna, 1981). These results were obtained using crude estrogen receptor preparations. In order to exclude a possible indirect action of an unidentified cytosol component, the dissociation of [3H]estradiol and [3H]4-hydroxytamoxifen from the calf uterine estrogen receptor was studied after preincubation at 25 "C (activation step) of each complex in the presence or absence of an equal volume of cytosol incubated similarly with the other nonradioactive ligand. As shown in Fig. 1the dissociation rates of 3H-labeled complexes were not modified by the unlabeled cytosols. Therefore a nonspecific indirect effect of one of the two ligands involving a cytosol component inaddition to the receptor is unlikely and the difference observed between the dissociation kinetics of the two ligands probably reflects intrinsicproperties of the two resulting complexes. The effect of the B36 antibody on the dissociation rate of the calf uterine receptor-estradiol complex was then studied. When added before the activation step, the antibody prevented the decrease in the dissociation rate of the receptorestradiol complex. As shown in Fig. 2a, the dissociation rate of the receptor-estradiol complex preheated in the presence of B36 was practically equivalent to thatof the nonactivated (molybdate-stabilized) complex. The half-dissociation time of the complex was reduced by a factor of 6 by molybdate or the B36 antibody. Table I shows that theB36 antibody consistently decreased the half-dissociation time of the receptor-estradiol complex from 3.2 to 6.5-fold. Fig. 3 shows that this inhibitory effect of B36 was concentration-dependent7reaching a nearly maximal effect at 100 pg/ml and a half-maximal effect at 10 pg/ml (DE5,,).The effect of molybdate reported by others (Shyamala and Leonard, 1980) was found to be similar to that found in this study and was constant from 4 to 60 mM (not shown). When both the B36 antibody and molybdate were added together at optimal concentrations, an additive effect was observed (Fig. 2a; Table I). However, the effect of B36 on the dissociation rate of the receptor-estradiol complex persisted when the complex was formed in cytosol containing 0.4 M KCl, whereas molybdate in agreement with results reported by Mauck et al. (1982), no longer prevented the dissociation Portions of this paper (including "Experimental Procedures," Tables I and 11, and Figs. 1, 3, 5, 6, 8) are presented in miniprint a t the end of this paper. The abbreviations used are: T buffer, 10 mM Tris-HC1, pH 7.4; T E buffer, T buffer containing 1.5 mM EDTA; TED buffer, T buffer containing 1.5 mM EDTA and 2 mM dithiothreitol; dextran-coated charcoal, 0.5% charcoal Norit A, 0.05% dextran T70 in T E buffer. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 85M-305, cite the authors, and include a check or money order for $8.80 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal thatis available from Waverly Press.
J
0
2
2
4
Dissociation
time
L
(hl
FIG. 2. Effects of the B36 antibody andof molybdate on the of estradiol- and 4-hydroxytamoxifen-etrodissociation rate gen receptor complexes. Calf uterine cytosol with or without 10 m M Na2Mo04was incubated for 2 h at 0 "C with 6 nM [3H]estradiol or [3H]4-hydroxytamoxifenin the presence or absence of 1~ L radioM inert estradiol. Aliquots were treated by charcoal for 0.5 h at 0 "C. Then a portion of the samples was supplemented with 10%T buffer (10 mM Tris-HC1, pH 7.4) and the other with B36 in T buffer (final B36 concentration: 200 pglml). After incubation for 1.5 h a t 0 "C then for 0.5 h a t 25 "C, the stability and dissociation of the complexes a t 20 "C were assayed as described in the legend to Fig. 1after 1h of charcoal treatment at0 "C.The saturable binding of [3H]estradiol ( a ) and of [3H]4-hydroxytamoxifen(6) are plottedas a percentage of the corresponding value at time 0. Open symbols, saturable binding in the absence of radioinert estradiol; closed symbols, saturable binding in the presence of 1p~ radioinert estradiol. 0 and 0,control cytosol; A and A, molybdate-treated cytosol; 0 and M, B36-treated cytosol; V and V, cytosol treated with molybdate and B36.
rate of the complex in these conditions (not shown). The effect of B36 on the dissociation of the receptor-estradiol complex was specific and related to antibody/receptor recognition since: (i) the B36 antibody, which does not interact with the rat uterine estrogen receptor, did not modify the dissociation rate of the ratuterine receptor-estradiol complex (Table I); (ii) nonspecific rat (IgG) antibodies did not affect noticeably the dissociation rate of the calf uterine receptorestradiol complex (Table I). Moreover, the dissociation rate of the calf uterine receptor-4-hydroxytamoxifen complex, which is unchanged after activation of the complex (Rochefort and Borgna, 1981), was not affected by the antibody (Fig. 2b) even though under these conditions, the estrogen receptor bound to 4-hydroxytamoxifen reacts with this antibody (Borgna et al., 1984).In all cases, the same pseudo-first order process was observed. These results indicate that the effect was specific for the estradiol-induced modification of the estrogen receptor and was not observed following partial activation of the receptor by an anti-estrogen. The Decrease in the Dissociation Rate of the ReceptorEstradiol Complex Is Reversed by the B36 Antibody-We then compared the effects of the B36 antibody and of molybdate on the dissociation rate of the calf uterine receptor-estradiol complex when the inhibitors were added before or after receptor activation. B36 was able to increase the dissociation rate of the preactivated receptor-estradiol complex in theabsence or presence of 0.4 M KC1. Fig. 4bshows that itsefficiency was practically the same when it was added before or after the preheating step indicating that the effect of receptor activation on the dissociation rate of the complex was reversed by the antibody. Conversely molybdate, when added after the preheating step, no longer noticeably decreased the dissociation rate of the complex (Fig. 4a). Moreover, the inhibition observed with molybdate was transitory and progressively
Antibody and Estrogen Receptor Activation
I
Ib
0
0
2
4
I 0
Dissociation
15549
1 2
4
time ( h l
FIG. 4. Reversibility of receptor-estradiolactivation by the B36 antibody and molybdate.Calf uterine cytosol with or without 10 mM NazMoOl wasincubated for 2 h at 0 "C with 6 nM [3H]estradiol in the presence or absence of 1 p~ radioinert estradiol. It was then treated by charcoal for 0.5 h at 0 "C. Aliquots of cytosol without molybdate were supplemented or not with B36 (final concentration 200 pg/ml). Samples were incubated for 1.5 h at 0 "C and for 0.5 h at 25 "C. B36 or molybdate were also added to portions of the untreated cytosol following the 25 "C incubation. All samples were then further incubated for 1.5 h at 0 "C. The stability and dissociation of the re~eptor-[~H]estradiol complex were assayed at 20 "C as described in the legends to Figs. 1 and 2. a, effect of molybdate; b, effect of the B36 antibody. Open symbols, stability of the re~eptor-[~H]estradiol complex; closed symbols, dissociation of the complex. 0 and 0, untreated complex; 0 and m, complex treated before the 25 "C activation step; A and A, complex treated following the activation step. The results shown in a and b are from two different experiments.
decreased when the duration of the preactivation step increased (Fig. 5). The Antibody Prevents Heat-inactivation of the Estrogen Receptor-Estradiol (Rochefort and Baulieu, 1971) and molybdate (Shyamala and Leonard, 1980) protect the estrogen binding site of the receptor against heat-inactivation. We tested theability of the B36 antibody to prevent heat-inactivation of the naked estrogen receptor. At saturating concentrations, theantibody (200 pg/ml) clearly inhibited the 25 "C inactivation of the naked receptor (Fig. 6). This effect was specific since it was not observed with an equivalent concentration of nonspecific rat IgG antibodies (not shown). For short incubation times, the B36 efficiency was lower than that of molybdate. However, for increasing timesof inactivation, the antibody appeared to become more efficient than molybdate whose efficiency decreased with time (not shown). There was no significant additive effect of the two protecting agents. The B36 Antibody Inhibits DNA Binding of the Estrogen Receptor-Preliminary experiments showed that theB36 antibody prevented DNA binding of the estrogen receptor bound to estradiol or 4-hydroxytamoxifen (Borgna et al., 1984). We have more extensively studied this inhibition under conditions, more favorable for DNA binding, in which the receptor was activated for 18 h at 0 "C. As shown in Fig. 7, the DNA binding of both complexes was markedly inhibited when the complexes were preincubated with the B36 antibody. Increasing concentrations of B36, progressively decreased the DNA binding of both complexes. The DNA binding of the receptoranti-estrogen complex appeared more sensitive to B36 than that of the receptor-estradiol complex, since the half-maximal inhibition of its binding was obtained at lower concentrations of B36 (DE5o-3 Kg/ml and -5 pg/ml for 4-hydroxytamoxifenand estradiol-receptor complexes, respectively). However, the
5
IO 836
15
0
concentration
so
100
I )rg/rnL)
FIG. 7. Partial inhibition by the B36 antibody of estrogen receptor-DNA binding. Calf uterine cytosol in TED buffer (T buffer containing 1.5 mM EDTA and 2 mM dithiothreitol) was labeled with 2 nM [3H]estradiol or [3H]4-hydroxytamoxifenfor 4 h at 0 "C, then treated with charcoal suspension for 1 h at 0 "C. Aliquots (300 p l ) were incubated with or without varying amounts of B36 antibody (1 to 125 pg/ml) for 2.5 h at 0 "C, then added to a 2 0 0 4 suspension of cellulose or DNA-cellulose (240 pg of DNA/ml). The DNA binding of receptor complexes was determined as described under "Experimental Procedures." The results of two different experiments (a and b) are represented as percentages of receptor binding measured in the absence of the B36 antibody. 0,receptor-estradiol complex; 0,receptor-4-hydroxytamoxifen complex.
extent of inhibition, which varied slightly, was always more marked in the case of the estradiol-receptor complex (50 to 80%) than in the case of the 4-hydroxytamoxifen-receptor complex (35 to 60%). When B36 was added after completion of the interaction between the receptor-estradiol complex and DNA, only a small decrease in thebinding was observed (Table 11).Molybdate was also unable to displace markedly the receptor from DNA. Saturation analyses were then performed, using constant concentrations of receptor and B36 (sufficient concentration to bind the totality of the receptor in sucrose gradients) and increasing concentrations of DNA, to see if the decrease in DNA binding of the receptor induced by B36 resulted from a decrease in the number of binding sites or a decrease in receptor/DNA affinity. As shown in Fig. 8, preincubation with B36 clearly decreased the maximal concentration of the receptor-estradiol complex able to bind to DNA. The antibody similarly decreased the concentration of the receptor-Chydroxytamoxifen complex able to bind to DNA (not shown). However, this decrease was partial and greater for the estrogen complex than for the anti-estrogen complex. DISCUSSION
We report on several effects of the €336 monoclonal antibody on the properties of hormone and DNA binding sites on the calf uterine estrogen receptor. Some of these effects may be related to the inhibition of estrogen receptor activation. Whereas the prevention of the decrease in the dissociation rate of estradiol from receptor appeared to be complete, the inhibition of the binding of the receptor-estradiol complex to DNA was only -75%. Moreover, the antibody reversed the effect of activation on the dissociation rate of the complex, whereas it was unable to reverse receptor-DNA binding. B36 also inhibited, although to a lesser extent (-50%), the DNA binding of the receptor-4-hydroxytamoxifencomplex, but did not affect the dissociation rate of the anti-estrogen-receptor complex, which is not decreased upon activation (Rochefort and Borgna, 1981). The heat-inactivation of the naked estrogen receptor was also inhibited by the antibody. These effects were specific, since they were related to antigen/antibody
15550
Antibody and Estrogen Receptor Activation
recognition; moreover, the dissociation effect was restricted extend previous hypothesesstating that theconformations of to theestrogen ligand. the two types of complex are different. B36 action may corTo our knowledge this is the first example of monoclonal respond to an inactivation process of the receptor. antibodies to steroid receptors inducing such effects. All the results obtained by various approaches converge on Polyclonal antibodies raised against the calf estrogen recep- the same conclusionthat the conformations of the two comtor have been shown to decrease the affinity of estradiol (Tate plexes are different and support the biological relevance of et al., 1984) or weak affinity ligands (Garcia et al., 1982) for the observed modificationsof the estrogen receptorby in vitro the estrogenreceptor. This decreased affinity wasaccomactivation. (i) The decrease in the dissociation rate of the activation is only observed with panied by a decreased association rate (Garcia et al., 1982) or receptor-ligand complex upon increased dissociationrate (Tateet al., 1984),but these anti- the estrogen ligand (Rochefort and Borgna, 1981), and B36 bodies were found to have no effect on the preformed receptor- (this study) and molybdate(Rochefort and Borgna,1981) affect only the dissociation rate of the estrogen ligand. (ii) estradiol complex. Attempts to evidence the interference of antibodies to ste- The two types of complex displaydifferent reactivities for the roid receptors with receptor-DNA interaction have been neg- antibody (Borgna et al.,1984).(iii) They also displaydifferent ative both for the estrogen receptor (Jensen et al., 1983; abilities to bind DNA (Evans et al., 1982). (iv) The inhibition Moncharmont et al., 1982) and the glucocorticoid receptor by B36 of receptor-DNA binding is more pronounced in the (Westphal et al., 1982; Carlstedt-Dukeet al., 1982; Gustafsson more DNA-reactive estradiol-receptor complex than in the less DNA-reactive4-hydroxytamoxifen-receptorcomplex, but et al., 1983). The increase in the dissociation rate of the preactivated occurs at lower concentrations of B36 in the case of the more receptor-estradiol complex caused by B36 indicates that the B36-reactive, 4-hydroxytamoxifen-receptor complex (this binding of the antibody to the complex induces a conforma- study). Using other radiolabeled high affinity anti-estrogens, (CI tional change in theactivated receptor. However,this change was not evident in thereceptor-4-hydroxytamoxifencomplex, 628 M and H 12851, differences in the properties of estrogenreported whose dissociation rate was not increased by antibody bind- and anti-estrogen-receptor complexeshavebeen ing. The two other effects of the antibody, relative to the (Katzenellenbogen et al., 1981; Ruh et al., 1983; Keeneet al., protection of the ligand binding site of the receptor against 1984). In our view, these results suggest that some of the effects heat-inactivation and to inhibition of receptor binding to DNA, could also be explained bya conformational change in of B36 antibody could berelated to the inhibition of receptor the receptor. However, other more trivial reasons, such as a activation. We based this interpretation on a series of condirect steric hindrance of receptor domains by the bulky IgM vergent arguments. (i) The extent of inhibition by the B36 antibody, cannot be excluded.The inability of B36 to reverse antibody of receptor bindingto DNA differs whenthe receptor receptor-DNA binding, suggestseither that theantibody can is activated by an estrogen or by an anti-estrogen, as discussed no longer interact with the receptor bound to DNA-cellulose above. (ii) The decrease in the dissociation rate of estrogen to be one ofthe best currently or that B36 fails to alter receptor conformation in the pre- from the receptor, which seems available criteria for receptor activation, is clearly inhibited formed receptor-DNA complex. However, it interactswith the receptor bound to soluble DNA, since preincubation of the and reversed by the antibody. (iii) The effects of the B36 receptor with saturating amounts of soluble DNA did not antibody parallel those of molybdate.Nevertheless, these prevent the effect of B36, added subsequently, on the disso- results do not enable us to specify the molecular mechanism of receptor activation which could result from a covalent ciation rate of the receptor-estradiol complex? The effects of B36 were similar to those of molybdate, a modification of the receptor (phosphorylation, ADP-ribosyclassical inhibitor of steroid receptor activation (Dahmer et lation, acetylation, etc.) or from a noncovalent modification al., 1984), since molybdate:(i) inhibits denaturation and the of the receptor (transconformational change, dimerization, decrease in the dissociation rate of the receptor-estradiol etc.). The fact that the antibody is able to increase the complex (Shyamala and Leonard, 1980); (ii) does not affect dissociation rate of estradiol from the preactivated receptorthe dissociation rate of the receptor-4-hydroxytamoxifencom- estradiol complex to the level of the nonactivated complex plex (Rochefort and Borgna, 1981);(iii) prevents the binding suggests that it induces a conformational change oppositeto that which would be triggered by activation of the receptor. of the receptor to DNA (Muller et al., 1982). However, the mechanism of action of molybdate and ofB36 on the calf However, we cannot exclude other possibilities for the activation process. A more direct proof of the inhibition of recepestrogen receptorare probably different, since: (i) both inhib- tor activation by B36 would require extensivepurification of itors act synergically to prevent the decrease in the dissocia- the receptor, and the demonstration that thisantibody is able tion rate of the receptor-estradiolcomplex. (ii) B36 is effective to prevent the in vitro action of the estrogen receptor either in the presence of 0.4 M KC1, whereas molybdate is not. (iii) in a cell-free system or following its transfer into estrogenB36 not only prevents, but also reverses the decrease in the responsive cells. dissociation rate of the receptor-estradiol,whereas molybdate does.not prevent, but onlydelays this decrease. (iv) B36 Acknowledgments-We are grateful to Drs. G. L. Greene and E. V. prevents the binding of the preactivated receptor to DNA Jensen for providingus with the B36 monoclonal antibody. We thank (Borgna et al., 1984), whereas molybdate does not (Gschwendt J. Scali for her skillful technical help, andE. Barrib and M. Eg6a for and Kittstein, 1983). The last result suggests that the inter- their excellent secretarial assistance. acting site of molybdate on the receptor may be masked after REFERENCES activation, whereas the site of interaction with the B36 anti& Herrick, G . (1971) Methods Enzymol. 21,198-217 Alberts, B. body remains accessible. Andrb, J. & Rochefort, H. (1973) Febs Lett. 32,330-334 The effects of B36 on the properties of the receptor-ligand Bailly, A., Le Fevre, B., Savouret,J.-F. & Milgrom, E. (1980) J. Biol. complex, which vary accordingto the ligand (either estradiol Chem. 255,2729-2734 or 4-hydroxytamoxifen) bound to the receptor, confirm and Borgna, J.-L. & Rochefort, H. (1980) Mol. Cell. Endocr. 20,71-85 J. Fauque,J.-L. Borgna, andH. Rochefort, unpublished results.
Borgna, J.-L. & Rochefort, H. (1981) J. Biol. Chem. 256,859-868 Borgna, J.-L. & Ladrech, S. (1982) Mol. Cell. Endocr. 27,l-15
Antibody and Estrogen Receptor Activation Borgna, J.-L., Fauque, J. & Rochefort, H. (1984) Biochemistry 2 3 , 2162-2168 Bradford, M. M. (1976) Anal. Biochem. 72,248-254 Burton, K. (1956) Biochem. J. 62,315-323 Carlstedt-Duke, J., Okret, S., Wrange, 0. & Gustafsson, J.-A. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,4260-4264 Dahmer, M.K., Housley, P. R. & Pratt, W. B. (1984) Annu. Reu. Physiol. 46,67-81 Evans, E., Baskevitch, P. P. & Rochefort, H. (1982) Eur. J. Biochem. 1 2 8 , 185-191 Garcia, M., Greene, G. L., Rochefort, H. & Jensen, E. V. (1982) Endocrinology 1 1 0 , 1355-1361 Geynet, C., Shyamala, G. & Baulieu, E. E. (1983) Biochim. Biophys. Acta 756,349-353 Greene, G. L.,Fitch, F. W. & Jensen, E. V. (1980) Proc. Natl. Acad. Sci. U. S. A. 7 7 , 157-161 Gschwendt, M. & Kittstein, W. (1983) J. Receptor. Res. 3, 377-392 Gustafsson, J.-A. & Okret, S. (1983) in Steroid Hormone Receptors: Structure und Function (Eriksson, H. & Gustafsson, J.-A., e&) pp. 355-388, Elsevier Scientific Publishing Co., Amsterdam Jensen, E. V., Numata, M., Brecher, P. I. & De Sombre, E. R. (1971) in The Biochemistry of Steroid Hormone Action (Smellie, R. M. S. ed) pp. 133-159, Academic Press, New York Jensen, E. V., Sobel, N. B., King, W. J. & Greene, G.L. (1983) in Steroid Hormone Receptors: Structure and Function (Eriksson H. & Gustafsson, J.-A., eds) pp. 337-353, Elsevier Scientific Publishing Co., Amsterdam Katzenellenbogen, B. S., Pavlik, E. J., Robertson, D. W. & Katzenellenbogen, J. A. (1981) J. Biol. Chem. 2 5 6 , 2908-2915
15551
Keene, J. L., Sweet, F., Ruh, M. F. & Ruh, T.S. (1984) Biochem. J. 217,819-826 Layne, E. (1957) Methods Enzymol. 3,444-454 Mauck, L. A., Day, R. N. & Notides, A. C. (1982) Bi0che.rnist.q 21, 1788-1793 Milgrom, E. (1983) in Steroid HormoneReceptors: Structure and Function (Eriksson, H. & Gustafsson, J.-A., eds) pp. 195-200, Elsevier Scientific Publishing Co., Amsterdam Moncharmont, B., Su, J.-L. & Parikh, I. (1982) Biochemistry 2 1 , 6916-6921 Miiller, R. E., Traish, A. M., Beebe, D. A. & Wotiz, H. H. (1982) J. Biol. Chem. 257,1295-1300 Notides, A. C., Hamilton, D. E. & Auer, H. E. (1975) J. Biol. Chem. 250,3945-3950 Rochefort, H. & Baulieu, E. E. (1971) Biochimie (Paris) 52,893-907 Rochefort, H. & Borgna, J.-L. (1981) Nature 292,257-259 Rochefort, H., Borgna, J.-L. & Evans, E. (1983) J. Steroid Biochem. 19,69-74 Ruh, M. F., Brzyski, R. G., Strange, L. & Ruh, T.S. (1983) Endocrinology 112,2203-2205 Sakai, D. & Gorski, J. (1984) Biochemistry 2 3 , 3541-3547 Shyamala, G. & Leonard, L. (1980) J. Biol. Chem. 255,602&6031 Tate, A. C., Greene, G. L., De Sombre, E. R., Jensen, E. V. &Jordan, V. C. (1984) Cancer Res. 44, 1012-1018 Weichman, B. M. & Notides, A. C . (1977) J. Biol. Chem. 252,88568862 Westphal, H. M., Moldenhauer, G. & Beato, M. (1982) EMBO J. 1 , 1467-1471 Yamamoto, K. R. & Alberts, B. (1972) Proc. Natl. Acad. Sci. U. S. A. 69,2105-2109
SUPPLEMENT TO A IIONOCIANAL ANTIBODY TO THE ESTROGEN RECEPTOR INHIBITS IN VITRO CRITERIA OF BBCEPMR ACTIVATIOII BY AN ESTROG& lvlD AN AUTI-ESTROGEN
by JoEl Fauque, Jean-Louis Bargna, and Henri Rochefort EXPERIJIENTAL PROCEDURES
Haterids - (z)-13Hl-4-hydroxytamoxifen (specificactivity 44 Ci/mmol i radiochemioal purity >95 %) was Obtained from the Radiochemical Centre 50 Cilmmol ; (Amersham, England). 16,7-3H/Estradiol(specificactivity radiochemical purity >95 %) was obtained from the "Commissariat P l'Energie Atomique" (Gif-sur-Yvette. France). The836 monoclonal antibody(40 X ammonium sulfate precipitate from the culture medium of a secreting hybridoma cell line) raised against the calf uterine estrogen r-eoeptor was kindly provided by Dr. G.L. Greene (Chicago University).
-
-
Cytosol estrogen receptor Pz-eparation and assay Cytosol (1.5 to 4 mg ( - B O T ) immature calf or rat Uteri was prepared in TE proteinlml) from frozen buffer. For DNA binding Studies, Cytosol was added With dithiothreitol to get a final concentration Of2 Cytosol was diluted with 5 % T buffer or with 200 mM Na2Mo04 in T buffer, and then 2 to 6 nM 13HI-4-hydroxytamarifen w a s totaland added alone or with 1 UM nonradioactive estradiol to measure nonspecifio binding respectively. Stock Solutions Of ligands were in ethanol : 1 % ) . Aliquots were incubated for 2 to 4 h at (final ethanol concentration 0.5 h at 25'12. Aliquots were then treated With O°C. then Sometimes for dextran-ooated charcoal at ODC for 0.5 to 2 h as described previously (Borgna and Rochefort, 1980). The radioactivity in the charcoal-treated samples was measuredandSpecificbinding of 'H-labeled ligands was taken as the difference between the total and the nonspecific binding.
m M .
-
Dissociation kinetics These were performed as described previously (Borgna and Ladrech, 1982). Briefly, after completion of the various facultative treatments,includingtheadditionofmolybdate,treatmentbycharcoal, addition of nonspecific rat IgG or B36 antibody, and incubation at 25OC, as specified for each experiment, the labeled Cytosolwas incubated at 2 0 W in the absence (to measure the Stability of complexes) OP in the presence (to of complexes) of 1 IIM radioinert estradiol. At measure the dissociation rate various times, 300 uL aliquots were removed and treated for 1 to 2 h at O'C with an equal VOllune Of charcoal suspension. Each value was Corrected for nonspecific binding and expressed as a percentage of undissociated complex.
-
Inactivation 01 the estrogen receptor Unlabeled Cytosol, treated or not with 10 m M Na2MoO4. 836, OF nonspecific IgG antibodies (150 to 300 pglmt), was 25'C. It was then incubated for 15 h at ODC incubated for increasing times at with 5 nM I3Hlestradiol in the presence or absence of 1 UM radioinert estradiol. The binding 0f1~Hlebtradiolwas measuped after 1.5 h of charcoal treatment.
R-OH?
C
20
"
0
2
d
R-OHT"/R-E;
"
4 0
Dissociation
2
4
time Ih )
Fig. 1. Dissociation rate Of estradiol- and 4-hydraxytamoxifensstrogen receptor CoPplexeB : Effect of molybdate in the presence or absence of the other complex. Calf uterine Cytosol prepared in TE buffer and containing 01. not 10 mM Na2Mo04 w a s incubated for3 h at ODC with 5 nM '€I-labeled estradiol 01. 4-hydroxytamoxifen in the absence OF pPeSence of 1 U M estradiol. Aliquots of Cytosol were also incubatedwith5 nM nonradioactiveestradiol or a portion of the labeled aliquot9 Was supplemented 4-hydroxytamoxifen. Then with an equal volume Of the cormsponding Cytosol containing the other nonradioactive ligand. and the remaining Portion was undiluted. All samples were incubated for0.5 h at 25-C. Samples were mixed at O'C with 1 % volume of or not 0.1 mM radioinertestradiol.They were then ethanol containing incubated at 2O'C and the bindingOf the labeled ligandwas assayed at various a 1.5 h Charcoal treatment at OQC. The saturable binding of times by I3Hlestradiol incubated in the absence (a) or pf-esence (b) of the 4-hydroxytamoxifen-receptor complex, and Of 13HI-4-hydr~xytamOXifenincubated in the absence ( c l or Presence (d) Of the estPadio1-receptor complex is plotted as a peroentage of initial bindingof the 'H-labeled Peceptor complex ZO'C. CloseSymbols : CytosolWithout againsttheincubationtimeat molybdate i Open symbols : molybdate-treated cytosol. Dissociation ( - 0 ) and stability of the PecePtor-13HleStrBdiol complex ; dissociation ( A n ) and Stability ( 1 V ) of the receptor-13HI-4-hydroxytamoxifen complex.
Antibody and Estrogen Receptor Activation
15552
DNAsellulose binding - Cytosol in TED buffer was labeled with 13Hlestradioi or 13H1-4-hydroxytamoxifen at O'C. After charcoal treatmentfor 0.5 to 1 hat uL) O'C, it was incubated withor without 036for 2.5 h at OOC. Aliquot5 (300 were then agitated with 200 OP 300 VL cellulose 01. DNA-Cellulose (prepared according to Albert5 and Herrick ( 1 9 7 1 ) ) for 16 to 18 h at O W . ?he Composition of the incubation medium was 1 mM dithiothreitol, 50 mM KC1, 5 X glycerol in TE buffer. After centrifugation, the supernatants wepe discarded and the peilets were washed twice 3 With mL ?E. The radioactivity in the final pellets was extracted with2 x 1.5 mL ethanol and counted. Specific binding to DNA w a s calculated as thedlfferencebetweenradioactivityboundto DNA-cellulose and that bound to cellulose (Evans et 31.. 1982).
-
Miscellaiimus Cytosol prDtein concentcation was measured spectrophotometrically (Layne, 1957). 836 concentration was measuPed according to DNA concentration was evaluated as described by Burton Bradford (1976). (1956).
Fig. 3. Dose-related effect of the 836 antibody on the disso~iationPate of the receptor-estradiol complex. Calf uterine Cytosolwas incubated for3.5 h at O'C with 5 nM 1% estradiol in the presence or absence of 1!AM radioinert AliqUOtS were estradiol. It was then treated by,chaPCOal for 0.5 h at O'C. 1 h at O'C, then for0.5h at 25'C withincreasing incubatedfor Concentrations Of the 836 antibody. The stability and dissociation of the ~=ceptor-l~Hle~t~adiol complex w e ~ eassayed at ZO'C as in Fig. 1. a.The decrease in the saturable binding of 13Hle6t~adi01 in the presence of radioinert estr-adiol as a function of tlme is expressed as a percentage of the corresponding undissociated complex which is as in Figs. 1 and 2. ( 0 ) COntPol i (A) 12.5 ug not shown but was Stable 836fmL ; ( 0 25 ug/mL ; ( 0 ) 50 ug/mL ; ( A ) 100 uglmL ; ( r ) 200 ugfmL ; (a) 300 ug/mL. b.The half-dissociation time of the 1.eceptor-l~Hle~tradio1 complex of determined from Fig. 3a is plotted against the corresponding ooncentration E36.
I , ,\-
g 0
Q u)
$
30 0
6
4
2
H o u r s at
0
25OC
Fig. 6. Protection by the836 antibody of Unmcupied estrogen receptor sites against heat-inactivation. Aliquot= of the calf uterine cytosol contaming OF not B36 (200 eg/mL) Na2Mo04 (10 mM) or both were incubated at 25'C for to bind eStPadio1 increasing periodsO f time. The residual capacity of samples was then measuped after 15h of incubationat O D C with 5 nJ4 I3Hlestradiol in the Presence 01. absence of 1 u M radioinert estradiol followed by 1.5 h of charcoal treatment. The Satupable bindingof 13Hl estradiol is expressed as a percentage of . . the Saturable bindinr of nonincubated complexes. (.,C) untreated complex ; (m) 836- and
2(
Dissociation time ( h )
Pre-incub. at 25°C ( h )
Fig. 5. Effect of molybdate on the dissociation PateO f the receptop-estradiol canplex according to pre-incubation times at 25-C. Calf uterine CytoSol containing or not 10 mM Na2Mo04 was pre-incubated with 6 nM )3Hlestradiol for at O ' C then for decreasing periods increasing periodsof time (1 h to 4.5 h) of the The Stability and dissociation of time ( 4 h to 0 . 5 h) at 25'C. receptor-l3HleSt=adiol complex at 20%w e ~ esubsequently assayed a6 described a.The decrease in the saturable binding of estradiol in Figs. 1 and 2. as a percentage of the corresponding undissociated complex, is expressed complex in complex. Open symbols : dissociation Of the ~leceptor-I~Hle~t~adiol the absence of molybdate. For clarity in the figure, Only the dissociation after ( A ) 0.5 h and ( a ) 4 h of pre-activation at 25OC are represented. Close symbols : dissociation O f thecomplex in thepresenceofmolybdate pre-incubated at 250c for (A) 0.5 h ; (V) 1 h : (a) 2 h ; (e) 4 h. The complexes were stable during the dissociation incubation as shown for CytOSOl pre-incubated for 4 h at 25'C without molybdate ( x ) . b.The half-dissociation time of the ~ e ~ e p t ~ ~ - I ~ H l e s t ~complex. a d i o l determined from the kinetics shown in a. is plotted against the dUPation of the 25PC-p~e-inc~bationstep. ( 0 ) control ; (.) molybdate-treated Cytosol.
l
0
25
DNA
50
75
100
concentration ()lg/mLl
Fig. 8. Saturation analysis of the effect of the B36 antibody on estrogen FeceptorfDNA binding. Estrogen receptor/DNA binding was determined as in Fig. 7 and Experimental Prooedures using a 5 nM 13Hlestradlal concentration,a constant amount of the 836 antibody (50 vg/mL) and increasing Concentrations of calf thymus DNA (0-100 ug/mL) adsorbed Onto cellulose. The different DNA Doncentrations were obtained by dilution of the DNA-cellulose suspension with was constant in all samples. cellulose so that the total amount of cellulose The concentration of estrogen receptor specifically bound to DNA is expressed as the amount O f estrogen PeCeptOr per volumeOf DNA-CellUlo8e suspensionas described by Evans et a1 (1982). ( 0 ) COntPOl sample : ( e ) Sample treated by B36 antibody.
Antibody and Estrogen Receptor Activation Table I Half-dissociation time of the receptor-estradiol Effect of B36 antibody and molybdate
Prevention
complex :
: 1.4to 3.8 mg1mL) calfuterinecytosol(proteinconcentration containing or not 10 mM Na2Mo04 was labeled with 2 to 6 nM I3Hlestradiol at O Q C in the presence or absence of 1 !JM radioinert estradiol. After charcoal treatment, aliquots Of labeled Cytosols were incubated With or Without 836 antibody (100to 200 IlgImL) f o r 0.5 to 1.5 h at O°C, then for 0.5 h at 25'C. of the receptor-I3Hle~t=adi01 complex Were The dissociation Pate and stability assayed at20'C as described in Experimental Procedures and in Fig. 1. a. Half-dissociationtime of thereceptor-estradiolcomplex from untreated Cytosol (control) and from 836- andlor molybdate-treated CYtoSO1. b. The factor of decrease is the ratio of the measured half-dissociation time of the receptor-estradiol complex from the COntPOl Cytosol to thatfrom the treated cytosol. C. Results obtained using nonrelevant rat IgG instead Of the B36 antibody are given in brackets. d. Mean values SEM f o r the factor of decrease in the half-dissocation time of the receptor-estradiol complex elicited by E36 andlor molybdate. Of calf uterine e.:Cantrol experiment using rat uterine Cytosol instead cytosol.
Table I1 andrwersal of DUA binding of the receptor-estradiolcmplex by theB36 antibody end molybdate
Calf uterine oytosol in TED buffer containing 01.not 10 m M Na2MoO4. was labeled with 5nM 13Hlestradiol for 4 h at O'C, then treated with charcoal Suspension for 0.5 h at O'C. AliqUOts (300p L ) were incubated with (pre-DNA then added incubation) or without B36 antibody(50 ng1mL) for 2 . 5 h at O°C, to 300 pL suspension of cellulose or DNA-Cellulose (0.24 mg DNAImL) and A portion of the samples Without 836 antibody OF agitated for 15 h atO'C. molybdate was then Supplemented (post-DNA incubation) with the B36 antibody or molybdate. Incubation was continued f o r 2 h at O'C. The concentrationOf the receptor-estradiol complex bound to DNAwas evaluated as described in ExperimentalProcedures.TheDNA-boundreceptor-estradiolcomplex is expressed as cpm Of I3Hlestradiol(receptor-l3Hlestradiolinput was 15,800 cpm) and also as a percentage of the DNA binding of Complexes 836 nor molybdate. incubated with neither
Stage
Of
addition of inhibitors
pre-DNA incubation post-DNA incubation Factor of decreaseb
Half-dissociation timea(mi") Expt 2-
Control
250
' 1 2 75 3
280
836
Moo4
50(220) 40
275
410
7
360
6
300
9 loe
Control 836 "00:-
6.2
420
95 95 65
50 75 60 55 95 65
150
150
60
4.3 6.8 35 75
35
3.8 3.2 4.9t0.6'
10.2 3.2 4.0 12.0 6.5 5.2t0.8 7. '7i2.4'
1.0
2.5
6.5
cpm
2-
%
cpm
%
M004/B36 Moo4
4.1
215
6
836
5.6(1.3)
310
4 3.7 5
2-
2-
Mo04/B36
6.5
2610
100
7
2610
720
28
5
2340
370
14 ?.1700 1
100
*7 =
90 3 65 i 6
