- Email: [email protected]
Effect of progesterone on the activity of occupied nuclear estrogen receptor in vitro
111
Molecular and Cellular Endocrinology, 64 (1989) 111-117 Elsevier Scientific Publishers Ireland, Ltd.
MCE 02078
Effect of progesterone on the activity of occupied nuclear estrogen receptor in vitro Elizabeth
J. Smanik, Juan-Jose Department
Calderon,
Thomas
G. Muldoon
and Virendra
of Physiology and Endocrinology, Medical College of Georgia, Augwta,
B. Mahesh
GA 30912, U.S.A.
(Received 16 May 1988; accepted 21 February 1989)
Key worak
Progesterone; Estrogen; Estrogen receptor; (Rat uterus)
In previous studies, we have demonstrated that progesterone administration in vivo can selectively alter estrogen receptor levels and distribution in the rat anterior pituitary. The present study represents an attempt to extend these observations to an in vitro model. Cytosolic and nuclear preparations of uterine homogenates from ovariectomized adult rats were shown to be capable of temperature-dependent estrogen-mediated receptor activation and translocation from cytosol to nuclei upon recombination. Addition of progesterone to isolated cytosol did not diminish estrogen receptor binding capacity over at least a 2 h period at 22°C. Preincubation of the subcellular fractions with progesterone, followed by removal of free progesterone prior to cytosol-nuclear recombination, resulted in dramatic reduction in nuclear estrogen receptor activity. This action was equally apparent whether progesterone was introduced to the cytosolic or nuclear fraction, and was confined to the steroid-occupied subpopulation of nuclear receptors. The ability of this in vitro system to mimic the estrogen receptor-suppressive effect of progesterone provides a good model in which to analyze the biochemical basis for a direct estrogen-inhibitory effect of progesterone on estrogen action.
Introduction
Progesterone exerts facilitative and inhibitory effects on gonadotropin secretion, depending on the hormonal milieu, the time of progesterone administration in relationship to the projected preovulatory gonadotropin surge, and the dose of progesterone used (Everett, 1948; Ode11 and Swerdloff, 1968; McPherson et al., 1975; McPherson and Mahesh, 1979). Critical analysis of the rat
Address for correspondence: Virendra B. Mahesh, Department of Physiology and Endocrinology, Medical College of Georgia, Augusta, GA 30912, U.S.A. 0303-7207/89/$03.50
(Kalra and Kalra, 1974; Rao and Mahesh, 1986), monkey (Helmond et al., 1980) and human (Liu and Yen, 1983) ovulatory cycles suggests that progesterone may be necessary for the normal magnitude of the preovulatory surge. Thus, the mechanism involved in this action of progesterone becomes very important. Progesterone has been shown to engender a release of LHRH from the hypothalamus (Kim and Ramirez, 1982; Peduto and Mahesh, 1985) and also a tissue-specific decrease in the nuclear estrogen receptors of the anterior pituitary, but not the hypothalamus (Smanik et al., 1983; Calderon et al., 1987). Progesterone has also been shown to regulate the uterine estrogen receptor in the hamster during
0 1989 Elsevier Scientific Publishers Ireland, Ltd.
112
the estrous cycle and upon administration in vivo (Evans et al., 1980; Evans and Leavitt, 1980). In the present study, the effect of progesterone treatment on the nuclear accumulation of estrogen receptors was studied in vitro in the rat uterus, in order to establish a readily manipulable system for analysis of the molecular basis of this action. Materials and methods Animals Adult female Sprague-Dawley rats (Holtzman, Madison, WI, U.S.A.) were obtained at 65-70 days of age. The rats were housed in groups of 3-4 per cage in air-conditioned rooms with a 14 h light : 10 h dark cycle and were given water and rat chow ad libitum. After 2 days of acclimatization, they were bilaterally ovariectomized under light ether anesthesia. Two weeks after ovariectomy, the animals were killed by decapitation and the uteri were collected into cold TEDG buffer (10 mM Tris, 1.5 mM Na,EDTA, 1 mM dithiothreitol and 10% glycerol (v/v), pH 8.0 at 22 o C). Preparation of cytosol and nuclear fractions All procedures were performed at 4 o C unless otherwise specified. Uteri were minced in a chilled Petri dish and homogenized with a Tekmar stainless steel homogenizer (Model 9098) in fresh TEDG buffer (one uterus/3 ml) using three bursts of 10 s each separated by 1 min cooling intervals. Crude nuclear pellets were obtained by centrifugation at 800 x g for 20 min and the supematant was saved. The nuclear pellet was resuspended in buffer and recentrifuged 3 times to remove cytosol contaminants. The supematant from the first centrifugation was centrifuged at 105 000 x g for 45 min, yielding a cytosol supematant fraction. The nuclear and cytosolic fractions were diluted with buffer to yield respective concentrations of 0.08 mg DNA/ml as determined by the procedure of Burton (1956) and 1.5-2.0 mg protein/ml, measured by the method of Lowry et al. (1951). Experimental protocols The first experiment established the temperature to be used in cytosol and nuclear recombina-
tion experiments in vitro. The cytosol fraction was incubated with a dextran-charcoal pellet obtained by centrifuging an equal volume of dextran-charcoal suspension (0.5 g dextran; 5 g charcoal/liter in TE buffer) for 15 min at 4°C. It was then centrifuged at 1000 x g for 10 min to remove free steroid. The supematant was recombined with the nuclear pellet and incubated at either 4O C for 1 h in the absence of any added estradiol, or at 4 o C and 22 o C in the presence of 3 nM estradiol. The cytosol and nuclear fractions were then reseparated by centrifugation at 800 X g for 10 min. The cytosol fractions were poured directly over dextran-charcoal pellets while the nuclear pellet was washed 3 times using TEDG buffer to remove free steroid. Estrogen receptor assays were then carried out in the cytosol and nuclear fractions. In a second series of experiments, the cytosol and resuspended nuclei were divided into four groups. Groups 1 and 2 cytosol and nuclear fractions did not receive any treatment. Group 3 cytosol received no treatment, but the nuclear suspension was incubated with 0.25 nM progesterone. Group 4 cytosol was incubated with 0.25 nM progesterone while the nuclear suspension was untreated. All cytosol and nuclear fractions were incubated separately at 4“ C for 30 min. The cytosols were then treated with dextrancharcoal and the nuclei were washed 3 times to remove free steroid. The next step involved combination of the four cytosols with the four nuclear pellets for temperature-mediated estrogen receptor translocation in vitro. Group 1 consisted of untreated cytosol and nuclei. Group 2 contained untreated cytosol and nuclei, to which 3 nM estradiol was added at the time of the recombination. Group 3 consisted of progesterone-pretreated nuclei and untreated cytosol, to which estradiol was added for the translocation step. Group 4 consisted of untreated nuclei and progesteronepretreated cytosol, to which estradiol was also added for the recombination step. In vitro receptor translocation was induced by a 1 h incubation at 22O C. This incubation was terminated by centrifugation at 800 x g for 10 min at 4O C. Cytosol supernatants were poured directly onto charcoal pellets to remove free steroid. Charcoalstripped cytosol was then assayed for estrogen receptor binding capacity. Nuclear pellets were
113
washed 3 times, resuspended in TEDG, and assayed for estrogen receptor. To determine whether incubation of uterine cytosol with progesterone altered the stability of the estrogen receptor, uterine cytosol was incubated with 3 nM estradiol at 4 o C for 30 min. After removal of the excess estradiol with dextran-charcoal, the cytosol was incubated with or without 0.25 nM progesterone for 30 min, 1 h, and 2 h at 4O C and 22O C, and estrogen receptor concentration was determined after the removal of free steroid. In the final set of experiments, an attempt was made to determine whether occupied or unoccupied estrogen receptor species (or both) were altered by progesterone treatment. Groups 1, 2 and 4 above were used in the measurement of total and unoccupied cytosol and nuclear receptors.
Receptor determinations The modified nuclear exchange assay of Anderson et al. (1972,1973) was used for the determination of total nuclear estrogen receptor content. This entailed a 37 o C, 30 min incubation period to effect exchange with endogenous steroid. The nuclear binding assay involved a concentration range for radiolabeled estradiol of 0.25-2.0 nM. Levels of cytosol estrogen receptor were determined by construction of 7-point saturation binding curves, according to the procedure of Korach and Muldoon (1974). The concentration range was 0.2-5.0 nM for [2,4,6,7-3H]estradiol (New England Nuclear, 100 Ci/mmol). The steroid was purified before use, as described previously (Muldoon and Cidlowski, 1980), using successive separation by descending paper chromatography in two solvent systems, as established by Mahesh (1964). Companion assay tubes containing a lOOfold molar excess of unlabeled estradiol were used to correct for nonspecific binding. Protamine sulfate precipitation methodology (Steggles and King, 1970; Korach and Muldoon, 1974) was used for separation of bound and unbound steroid. Values for specific binding have been normalized for presentation on the basis of DNA levels in nuclei per tissue and protein levels in cytosol. Each reported value for cytosol or nuclei repre-
sents the level determined from graphical analysis according to the method of Scatchard (1949). For unoccupied and total cytosol estradiol receptor measurements, the determinations were carried out at 4 o C for 18 h for unoccupied receptors (nonexchange conditions) and at 22” C for 17 h followed by 4°C for 1 h for total receptors (exchange conditions). Occupied receptor concentration represented the difference between the two values. Unoccupied receptors in nuclei were measured following incubation at 4” C for 16 h (nonexchange), whereas total nuclear receptors were measured after 30 min at 37” C (exchange conditions). Analytical and statistical procedures Sample radioactivity was measured in a Beckman LS-9000 spectrometer (Beckman Instruments, Palo Alto, CA, U.S.A.) within a 2% error limit, with conversion to disintegrations per min by the H gate method. The scintillation medium was composed of 5 g of Permablend II (Packard, Downers Grove, IL, U.S.A.) per liter toluene. Counting efficiencies were 48% for nuclear and 55% for cytosol samples. Data were analyzed by analysis of variance (ANOVA); the Newman-Keul’s multirange test was used to determine significance at the 95% confidence limit. Results The first set of experiments was designed to determine a suitable temperature at which nuclear accumulation of estrogen receptors could be reproducibly effected in our in vitro system. Using the results of Jensen et al. (1968) as a guideline, uterine cytosol and nuclei were recombined at 4 o C or at 22 o C in the presence of 3 nM estradiol. As expected, no significant nuclear uptake of receptor occurred at 4O C, relative to uncombined control levels (Fig. 1). However, recombination at 22” C for 1 h in the presence of 3 nM estradiol resulted in significant decreases in cytosol receptor binding, accompanied by increases in the nuclear compartment. An increase in total receptor binding activity measurable in the cells was also occasioned by the elevated temperature. The standard errors are quite large in such experiments because
114 * ? p siz
c2 %!z g$ $2 eg '58 cn
1600
1200 800 400
Cytosol
Nuclei
Fig. 1. Effect of temperature on the accumulation of nuclear estrogen receptors in vitro. Cytosol and nuclei were recombined and incubated at 4OC in the absence (control) or presence of 3 nM estradiol, and at 22OC in the presence of estradiol. The cytosol and nuclei were separated, the free steroid was removed from cytosol by dextran-charcoal adsorption and from the nuclei by washing, and estrogen receptor content was determined. * P c 0.05 by Neuman-Keul multirange test as compared to control and 4O C incubation; n = 6. In this and other figures, the receptor levels referred to represent specific [ 3H]estradiol binding capacity.
of the difficulty in accurately measuring nuclear receptors in vitro, but the changes were large enough to permit evaluation of factors that would modulate the response. To determine if progesterone added to the cytosol or nuclear fraction in vitro could alter estrogen receptor dynamics in a manner similar to that we had observed in vivo, the next series of experiments were carried out. Two control groups were used (groups 1 and 2) in which recombination was done in the absence or presence of estradiol. Two experimental groups consisted of: group 3 in which the nuclei were preincubated with progesterone; and group 4 in which the cytosol was preincubated with progesterone. It is important to note that, following progesterone exposure of either isolated fraction, measures were taken to remove free progesterone prior to recombination. Thus, the effects of progesterone during recombination should result from an action initiated prior to this step, in the isolated subcellular fraction. Recombination of the cytosol and nuclei in the presence of estradiol (group 2) brought about the expected decrease in cytosol estrogen receptor binding with an increase in nuclear retention (Fig. 2). The total receptor concentration (cytosol + nuclei) was similar in these two groups (1426 &-250 fmol/uterus in group 1; 1676 + 222 in group 2). In the presence of progesterone, regardless of
whether this steroid was introduced into the cytosol or nuclei prior to recombination, the recombination of the subfractions in the presence of estradiol resulted in a significant decrease in the nuclear accumulation of the estrogen receptor (groups 3 and 4) as compared to the estrogen control not treated with progesterone (group 2) (Fig. 2). No differences were found in specific cytosol estrogen binding levels among groups 2, 3 and 4. It is remarkable that this response of progesterone, previously demonstrated to require mediation by the progesterone receptor in vivo (Smanik et al., 1983), can be elicited by the low progesterone level of 0.25 nM. This is far below the saturation level of progesterone receptors and, although it is difficult to compare in vivo administration doses with in vitro addition of steroid, it appears that the decrease in estrogen receptor activity by progesterone is an extremely sensitive response. The possibility needed addressing that the effect of progesterone on estrogen receptor dynamics involved a decrease in the stability of the estrogen receptor during the preincubation process rather than an active progesterone-mediated event. To test this, uterine cytosol was incubated in the absence or presence of 0.25 nM pro-
Ep at recombination Progesterone to cytosol Progesterone to nuclei
-
+ -
+ -
+ + -
-
+ -
+ +
+ -
Fig. 2. Effect of 0.25 nM progesterone preincubation with either nuclei (group 3) or cytosol (group 4) on estradiol receptors, following recombination at 22 o C for 1 h in the presence of 3 nM estradiol. Groups 1 and 2 served as controls without (group 1) or with (group 2) the addition of estradiol at the time of recombination. Number of observations = 5. * P < 0.05 relative to all cytosol groups; * * P < 0.05 relative to all nuclear groups. As specified by our receptor assays, cytosol receptor levels represent unoccupied forms, whereas nuclear receptor levels are total.
115 TABLE
1
THE STABILITY OF CYTOSOLIC UTERINE ESTROGEN RECEPTOR IN THE PRESENCE OF PROGESTERONE Rat uterine cytosol was treated with 3 nM estradiol for 30 min at 4O C. After removal of excess estradiol by dextran-charcoal adsorption, 0.25 nM progesterone was added for different periods of time at 4O C and 22O C. After removal of free progesterone with dextran-charcoal, estrogen receptor level was measured. Exposure time
30 mill lh 2h
Cytosol receptor (fmol/mg protein) Without progesterone
With progesterone
4°C
22OC
4OC
22OC
275 + 14 256 f 36 299 f 32
245 f 73 273 f 34 257 rt 20
272 f 19 238k19 259+ 18
222* 8 204k55 300*7s
gesterone at 4O C and 22O C for 30 min, 1 h and 2 h, and estrogen receptor concentrations were determined. The results in Table 1 show no detri-
mental effects of progesterone or the incubation conditions on the estrogen receptor binding capability. Experiments were done to determine whether the progesterone effect of decreasing the nuclear level of estrogen receptor binding was directed toward a specific nuclear receptor subpopulation. For this purpose, total and unoccupied estrogen receptors were assayed in uterine cytosol and nuclei in groups 1, 2 and 4. The preincubation of uterine cytosol with progesterone resulted in a decrease in the total and occupied estrogen receptor binding activity of the nuclear fraction (group 4, Fig. 3) as compared to nonprogesterone-treated controls (group 4, Fig. 3). The effect of progesterone appeared to be exclusively on the occupied estrogen receptors in the nucleus since occupied and unoccupied cytosol fractions, and unoccupied nuclear fractions, were unchanged by preincubation with progesterone, Discussion
CYTOSOL ESTROGEN RECEPTORS
1500,
Control bJrouP 1)
Estmdid kJ~P 2)
Estmdid and Progesterone kJ~P 41
NUCLEAR ESTROGEN RECEPTORS
1500, 1250 loo0 750
;i 250 1
Estmdiol and Progesterone IclrOUo 4)
Fig. 3. Uterine cytosol and nuclear total and unoccupied receptor content after preincubation of cytosol with 0.25 nM progesterone (group 4) prior to recombination at 22O C for 1 h with 3 nM estradiol. Groups 1 and 2 are the respective recombination controls without (group 1) or with (group 2) addition of estradiol. Unoccupied receptors were measured under nonexchange conditions; occupied receptors, under exchange conditions. Occupied receptors were the calculated difference between total and unoccupied. n = 3. * P < 0.05 compared to other unoccupied cytosol receptor levels; * * P < 0.05 compared with values obtained after preincubation with progesterone.
The action of progesterone on gonadotropin secretion has been difficult to describe, primarily because it is linked to the corresponding levels and actions of estrogens (Mahesh and Muldoon, 1987). At the molecular level, the interaction between these hormones is best characterized by the well-documented prerequisite for an estrogenic background for manifestation of progesterone receptor activity in most tissues (O’Malley et al., 1970; VuHai et al., 1977). It is highly significant to the present studies that this relationship is also readily demonstrable in isolated cells having normal estrogen receptor dynamics (Horwitz and McGuire, 1978a, b). In terms of the mechanism whereby progesterone can modify estrogenic feedback control of gonadotropin secretion, particularly apparent as an acute response at the anterior pituitary level (Turgeon, 1979), little is known, but this is a crucial step at which reproductive intervention of either a positive or negative direction might be directed, if clearly understood. We have previously described the ability of progesterone to selectively suppress nuclear estrogen receptor binding in the anterior pituitary in vivo (Smanik et al., 1983; Calderon et al.,
116
1987). The progesterone effect appeared to be manifested at the level of occupied nuclear receptors in the anterior pituitary and the uterus (Fuentes et al., 1988). A similar effect has been observed and studied in some detail in the hamster uterus (Evans et al., 1980; Evans and Leavitt, 1980). Our results showed that this phenomenon did not occur in the hypothalamus and it appeared to be related to the level of endogenous estrogen. Moreover, in the absence of estrogen replacement to the ovariectomized immature rat model used, progesterone had no effect, indicating that progesterone receptors were mediating the response. A decrease in estrogen receptor binding capacity of the anterior pituitary by progesterone has been demonstrated to be of biological significance. Progesterone has been found to attenuate estrogen-induced progesterone receptor synthesis (Calderon et al., 1987) and estrogen-induced prolactin release (Brann et al., 1988). These studies did not, however, advance our understanding of how progesterone, in a purely physical sense, was capable of altering estrogen receptor activity. It has been shown in numerous systems that progesterone has very little binding affinity for estrogen receptors (Sherman et al., 1970; Muldoon, 1980), apparently ruling out the classical action of antiestrogens to tie up the receptor in a nonfunctional state (Katzenellenbogen et al., 1979). On the other hand, unmasking of cryptic, perhaps compartmentalized, nuclear receptors, stimulation of the rate at which nuclear receptors are degraded, or interference with binding of the estrogen-receptor complex to functional effector sites on chromatin, are three of many viable mechanisms whereby progesterone might selectively reduce nuclear estrogen receptor activity. These mechanisms cannot be explored easily in an in vivo system, so we turned in the present study to an in vitro analysis of the phenomenon. Recombination experiments, involving separation, treatment and recombination of cytosol and nuclei, were introduced by Jensen’s group (Jensen et al., 1968) in the early history of receptorology. They were used to describe the heat-sensitive nature of receptor accumulation in the nucleus and thus introduced the vital component of receptor activation as an integral part of the mechanism of estrogen action (Jensen and
DeSombre, 1973). Although recent years have seen a changing pattern of intracellular receptor distribution (King and Greene, 1984; Welshons et al., 1984) the concept of cytosol receptors as loosely associated nuclear components reasonably accommodates these changes, and recombination-type experiments still represent a valid means of assessing certain induced changes in receptor populations, particularly those that we are describing herein. Having established that we could readily demonstrate temperature-dependent nuclear uptake of estrogen receptor from uterine cytosol fractions, we examined the influence of progesterone on this process. The possibility that progesterone could alter the stability of the cytosol receptor such that the nuclear complex would be likewise compromised appears to be untenable. Even following exposure of cytosol to progesterone at 22’ C for as long as 2 h, no decrease in receptor binding activity was seen. Thus, the decrease in cytosol receptor binding following recombination with nuclei under translocationpromoting conditions was not intensified by a direct deleterious effect of progesterone on this receptor species. In this in vitro system, incubation in the presence of progesterone mimics the in vivo finding of selective diminution of nuclear estrogen receptor binding levels. A surprising and revealing observation was that progesterone was equally effective whether it was introduced to the cytosol or nuclear fraction prior to recombination, even under conditions where free progesterone was removed from the respective fractions before recombination. This necessitates either that progesterone directly influences the estrogen receptor in either compartment (unlikely since it does not destroy cytosol receptor binding directly) or that it acts through its own receptor, as it appears to do in vivo (Smanik et al., 1983). Although the animals used in this experiment were ovariectomized, we did find measurable levels of progesterone receptor in the tissue. Receptor mediation of the progesterone effect under our in vitro conditions requires that viable unliganded progesterone receptors be present in both the cytosolic and nuclear compartments, and this is not in violation of any current concepts of steroid hormone action, which encompass nuclear localization of unoccupied receptor.
117
These studies demonstrate the ability to extrapolate the in vivo observation of progesteronesuppressed nuclear estrogen receptor activity to an in vitro system that will be more amenable to manipulation. Progesterone’s selective inhibition of the estrogen-occupied pool of nuclear receptors is highly suggestive of a chromatin-based action, and represents an initial foray into the use of this system to define the nature, specificity and site of this action. Acknowledgement This investigation was supported by research grant HD16688 from the National Institute of Child Health and Human Development, National Institutes of Health. References Anderson, J.N., Clark, J.H. and Peck, Jr., E.J. (1972) B&hem. J. 126, 561-567. Anderson, J.N., Peck, Jr., E.J. and Clark, J.H. (1973) Endocrinology 93, 711-717. Brann, D.W., Rao, I.M. and Mahesh, V.B. (1988) Biol. Reprod. (in press). Burton, K. (1956) B&hem. J. 62, 315-323. Calderon, J.-J., Muldoon, T.G. and Mahesh, V.B. (1987) Endocrinology 120, 2428-2435. Evans, R.W. and Lea&t, W.W. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 5856-5860. Evans, R.W., Chen, T.J., Hendry, W.J. and Leavitt, W.W. (1980) Endocrinology 107, 383-390. Everett, J.W. (1948) Endocrinology 43, 389-405. Fuentes, M.A., Muldoon, T.G. and Mahesh, V.B. (1988) Endocrinology Suppl. 122, Abstract 716. Helmond, F.A., Simons, P.A. and Heim, P.R. (1980) Endocrinology 107, 478-485. Horwitz, K.B. and McGuire, W.L. (1978a) J. Biol. Chem. 253, 2223-2228. Horwitz, K.B. and McGuire, W.L. (1978b) J. Biol. Chem. 253, 8185-8191. Jensen, E.V. and DeSombre, E.R. (1973) Science 182,126-134.
Jensen, E.V., Suzuki, T., Kawashima, T., Stumpf, W.E., Jungblut, P.W. and DeSombre, E.R. (1968) Proc. Natl. Acad. Sci. U.S.A. 59, 632-638. Kalra, S.P. and Kalra, P.S. (1974) Endocrinology 95, 1711-1718. Katzenellenbogen, B.S., Bhakoo, H.S., Ferguson, E.R., Lan, N.C., Tatee, T., Tsai, T.S. and Katzenellenbogen, J.A. (1979) Recent Prog. Harm. Res. 35, 259-300. Kim, K. and Ramierz, V.D. (1982) Endocrinology 111.750-757. King, W.J. and Greene, G.L. (1984) Nature 307, 746747. Korach, KS. and Muldoon, T.G. (1974) Endocrinology 94, 785-793. Liu, J.H. and Yen, S.S.C. (1983) J. Clin. Endocrinol. Metab. 57, 797-802. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. Mahesh, V.B. (1964) Steroids 3, 647-661. Mahesh, V.B. and Muldoon, T.G. (1987) J. Steroid Biochem. 27, 665-675. McPherson, J.C. and Mahesh, V.B. (1979) Biol. Reprod. 20, 763-772. McPherson, J.C., Costoff, A. and Mahesh, V.B. (1975) Endocrinology 97, 771-779. Muldoon, T.G. (1980) Endocr. Rev. 1, 339-364. Muldoon, T.G. and Cidlowski, J.A. (1980) J. Biol. Chem. 255, 3100-3107. Odell, W.D. and Swerdloff, R.S. (1968) Proc. Natl. Acad. Sci. U.S.A. 61, 529-536. G’Malley, B.W., Sherman, M.R. and Toft, D.O. (1970) Proc. Natl. Acad. Sci. U.S.A. 67, 501-508. Peduto, J.C. and Mahesh, V.B. (1985) Neuroendocrinology 40, 238-245. Rao, I.M. and Mahesh, V.B. (1986) Biol. Reprod. 35, 1154-1161. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672. Sherman, M.R., Corvol, P.L. and G’Malley, B.W. (1970) J. Biol. Chem. 245, 6085-6096. Smanik, E.J., Young, H.K., Muldoon, T.G. and Mahesh, V.B. (1983) Endocrinology 113, 15-22. Steggles, A.W. and King, R.J.B. (1970) Biochem. J. 118, 695-701. Turgeon, J.L. (1979) Endocrinology 105, 731-736. VuHai, M.T., Logeat, F., Warembourg, M. and Milgrom, E. (1977) Ann. N.Y. Acad. Sci. 286, 199-209. Welshons, W.V., Lieberman, M.E. and Gorski, J. (1984) Nature 307, 747-749.
Molecular and Cellular Endocrinology, 64 (1989) 111-117 Elsevier Scientific Publishers Ireland, Ltd.
MCE 02078
Effect of progesterone on the activity of occupied nuclear estrogen receptor in vitro Elizabeth
J. Smanik, Juan-Jose Department
Calderon,
Thomas
G. Muldoon
and Virendra
of Physiology and Endocrinology, Medical College of Georgia, Augwta,
B. Mahesh
GA 30912, U.S.A.
(Received 16 May 1988; accepted 21 February 1989)
Key worak
Progesterone; Estrogen; Estrogen receptor; (Rat uterus)
In previous studies, we have demonstrated that progesterone administration in vivo can selectively alter estrogen receptor levels and distribution in the rat anterior pituitary. The present study represents an attempt to extend these observations to an in vitro model. Cytosolic and nuclear preparations of uterine homogenates from ovariectomized adult rats were shown to be capable of temperature-dependent estrogen-mediated receptor activation and translocation from cytosol to nuclei upon recombination. Addition of progesterone to isolated cytosol did not diminish estrogen receptor binding capacity over at least a 2 h period at 22°C. Preincubation of the subcellular fractions with progesterone, followed by removal of free progesterone prior to cytosol-nuclear recombination, resulted in dramatic reduction in nuclear estrogen receptor activity. This action was equally apparent whether progesterone was introduced to the cytosolic or nuclear fraction, and was confined to the steroid-occupied subpopulation of nuclear receptors. The ability of this in vitro system to mimic the estrogen receptor-suppressive effect of progesterone provides a good model in which to analyze the biochemical basis for a direct estrogen-inhibitory effect of progesterone on estrogen action.
Introduction
Progesterone exerts facilitative and inhibitory effects on gonadotropin secretion, depending on the hormonal milieu, the time of progesterone administration in relationship to the projected preovulatory gonadotropin surge, and the dose of progesterone used (Everett, 1948; Ode11 and Swerdloff, 1968; McPherson et al., 1975; McPherson and Mahesh, 1979). Critical analysis of the rat
Address for correspondence: Virendra B. Mahesh, Department of Physiology and Endocrinology, Medical College of Georgia, Augusta, GA 30912, U.S.A. 0303-7207/89/$03.50
(Kalra and Kalra, 1974; Rao and Mahesh, 1986), monkey (Helmond et al., 1980) and human (Liu and Yen, 1983) ovulatory cycles suggests that progesterone may be necessary for the normal magnitude of the preovulatory surge. Thus, the mechanism involved in this action of progesterone becomes very important. Progesterone has been shown to engender a release of LHRH from the hypothalamus (Kim and Ramirez, 1982; Peduto and Mahesh, 1985) and also a tissue-specific decrease in the nuclear estrogen receptors of the anterior pituitary, but not the hypothalamus (Smanik et al., 1983; Calderon et al., 1987). Progesterone has also been shown to regulate the uterine estrogen receptor in the hamster during
0 1989 Elsevier Scientific Publishers Ireland, Ltd.
112
the estrous cycle and upon administration in vivo (Evans et al., 1980; Evans and Leavitt, 1980). In the present study, the effect of progesterone treatment on the nuclear accumulation of estrogen receptors was studied in vitro in the rat uterus, in order to establish a readily manipulable system for analysis of the molecular basis of this action. Materials and methods Animals Adult female Sprague-Dawley rats (Holtzman, Madison, WI, U.S.A.) were obtained at 65-70 days of age. The rats were housed in groups of 3-4 per cage in air-conditioned rooms with a 14 h light : 10 h dark cycle and were given water and rat chow ad libitum. After 2 days of acclimatization, they were bilaterally ovariectomized under light ether anesthesia. Two weeks after ovariectomy, the animals were killed by decapitation and the uteri were collected into cold TEDG buffer (10 mM Tris, 1.5 mM Na,EDTA, 1 mM dithiothreitol and 10% glycerol (v/v), pH 8.0 at 22 o C). Preparation of cytosol and nuclear fractions All procedures were performed at 4 o C unless otherwise specified. Uteri were minced in a chilled Petri dish and homogenized with a Tekmar stainless steel homogenizer (Model 9098) in fresh TEDG buffer (one uterus/3 ml) using three bursts of 10 s each separated by 1 min cooling intervals. Crude nuclear pellets were obtained by centrifugation at 800 x g for 20 min and the supematant was saved. The nuclear pellet was resuspended in buffer and recentrifuged 3 times to remove cytosol contaminants. The supematant from the first centrifugation was centrifuged at 105 000 x g for 45 min, yielding a cytosol supematant fraction. The nuclear and cytosolic fractions were diluted with buffer to yield respective concentrations of 0.08 mg DNA/ml as determined by the procedure of Burton (1956) and 1.5-2.0 mg protein/ml, measured by the method of Lowry et al. (1951). Experimental protocols The first experiment established the temperature to be used in cytosol and nuclear recombina-
tion experiments in vitro. The cytosol fraction was incubated with a dextran-charcoal pellet obtained by centrifuging an equal volume of dextran-charcoal suspension (0.5 g dextran; 5 g charcoal/liter in TE buffer) for 15 min at 4°C. It was then centrifuged at 1000 x g for 10 min to remove free steroid. The supematant was recombined with the nuclear pellet and incubated at either 4O C for 1 h in the absence of any added estradiol, or at 4 o C and 22 o C in the presence of 3 nM estradiol. The cytosol and nuclear fractions were then reseparated by centrifugation at 800 X g for 10 min. The cytosol fractions were poured directly over dextran-charcoal pellets while the nuclear pellet was washed 3 times using TEDG buffer to remove free steroid. Estrogen receptor assays were then carried out in the cytosol and nuclear fractions. In a second series of experiments, the cytosol and resuspended nuclei were divided into four groups. Groups 1 and 2 cytosol and nuclear fractions did not receive any treatment. Group 3 cytosol received no treatment, but the nuclear suspension was incubated with 0.25 nM progesterone. Group 4 cytosol was incubated with 0.25 nM progesterone while the nuclear suspension was untreated. All cytosol and nuclear fractions were incubated separately at 4“ C for 30 min. The cytosols were then treated with dextrancharcoal and the nuclei were washed 3 times to remove free steroid. The next step involved combination of the four cytosols with the four nuclear pellets for temperature-mediated estrogen receptor translocation in vitro. Group 1 consisted of untreated cytosol and nuclei. Group 2 contained untreated cytosol and nuclei, to which 3 nM estradiol was added at the time of the recombination. Group 3 consisted of progesterone-pretreated nuclei and untreated cytosol, to which estradiol was added for the translocation step. Group 4 consisted of untreated nuclei and progesteronepretreated cytosol, to which estradiol was also added for the recombination step. In vitro receptor translocation was induced by a 1 h incubation at 22O C. This incubation was terminated by centrifugation at 800 x g for 10 min at 4O C. Cytosol supernatants were poured directly onto charcoal pellets to remove free steroid. Charcoalstripped cytosol was then assayed for estrogen receptor binding capacity. Nuclear pellets were
113
washed 3 times, resuspended in TEDG, and assayed for estrogen receptor. To determine whether incubation of uterine cytosol with progesterone altered the stability of the estrogen receptor, uterine cytosol was incubated with 3 nM estradiol at 4 o C for 30 min. After removal of the excess estradiol with dextran-charcoal, the cytosol was incubated with or without 0.25 nM progesterone for 30 min, 1 h, and 2 h at 4O C and 22O C, and estrogen receptor concentration was determined after the removal of free steroid. In the final set of experiments, an attempt was made to determine whether occupied or unoccupied estrogen receptor species (or both) were altered by progesterone treatment. Groups 1, 2 and 4 above were used in the measurement of total and unoccupied cytosol and nuclear receptors.
Receptor determinations The modified nuclear exchange assay of Anderson et al. (1972,1973) was used for the determination of total nuclear estrogen receptor content. This entailed a 37 o C, 30 min incubation period to effect exchange with endogenous steroid. The nuclear binding assay involved a concentration range for radiolabeled estradiol of 0.25-2.0 nM. Levels of cytosol estrogen receptor were determined by construction of 7-point saturation binding curves, according to the procedure of Korach and Muldoon (1974). The concentration range was 0.2-5.0 nM for [2,4,6,7-3H]estradiol (New England Nuclear, 100 Ci/mmol). The steroid was purified before use, as described previously (Muldoon and Cidlowski, 1980), using successive separation by descending paper chromatography in two solvent systems, as established by Mahesh (1964). Companion assay tubes containing a lOOfold molar excess of unlabeled estradiol were used to correct for nonspecific binding. Protamine sulfate precipitation methodology (Steggles and King, 1970; Korach and Muldoon, 1974) was used for separation of bound and unbound steroid. Values for specific binding have been normalized for presentation on the basis of DNA levels in nuclei per tissue and protein levels in cytosol. Each reported value for cytosol or nuclei repre-
sents the level determined from graphical analysis according to the method of Scatchard (1949). For unoccupied and total cytosol estradiol receptor measurements, the determinations were carried out at 4 o C for 18 h for unoccupied receptors (nonexchange conditions) and at 22” C for 17 h followed by 4°C for 1 h for total receptors (exchange conditions). Occupied receptor concentration represented the difference between the two values. Unoccupied receptors in nuclei were measured following incubation at 4” C for 16 h (nonexchange), whereas total nuclear receptors were measured after 30 min at 37” C (exchange conditions). Analytical and statistical procedures Sample radioactivity was measured in a Beckman LS-9000 spectrometer (Beckman Instruments, Palo Alto, CA, U.S.A.) within a 2% error limit, with conversion to disintegrations per min by the H gate method. The scintillation medium was composed of 5 g of Permablend II (Packard, Downers Grove, IL, U.S.A.) per liter toluene. Counting efficiencies were 48% for nuclear and 55% for cytosol samples. Data were analyzed by analysis of variance (ANOVA); the Newman-Keul’s multirange test was used to determine significance at the 95% confidence limit. Results The first set of experiments was designed to determine a suitable temperature at which nuclear accumulation of estrogen receptors could be reproducibly effected in our in vitro system. Using the results of Jensen et al. (1968) as a guideline, uterine cytosol and nuclei were recombined at 4 o C or at 22 o C in the presence of 3 nM estradiol. As expected, no significant nuclear uptake of receptor occurred at 4O C, relative to uncombined control levels (Fig. 1). However, recombination at 22” C for 1 h in the presence of 3 nM estradiol resulted in significant decreases in cytosol receptor binding, accompanied by increases in the nuclear compartment. An increase in total receptor binding activity measurable in the cells was also occasioned by the elevated temperature. The standard errors are quite large in such experiments because
114 * ? p siz
c2 %!z g$ $2 eg '58 cn
1600
1200 800 400
Cytosol
Nuclei
Fig. 1. Effect of temperature on the accumulation of nuclear estrogen receptors in vitro. Cytosol and nuclei were recombined and incubated at 4OC in the absence (control) or presence of 3 nM estradiol, and at 22OC in the presence of estradiol. The cytosol and nuclei were separated, the free steroid was removed from cytosol by dextran-charcoal adsorption and from the nuclei by washing, and estrogen receptor content was determined. * P c 0.05 by Neuman-Keul multirange test as compared to control and 4O C incubation; n = 6. In this and other figures, the receptor levels referred to represent specific [ 3H]estradiol binding capacity.
of the difficulty in accurately measuring nuclear receptors in vitro, but the changes were large enough to permit evaluation of factors that would modulate the response. To determine if progesterone added to the cytosol or nuclear fraction in vitro could alter estrogen receptor dynamics in a manner similar to that we had observed in vivo, the next series of experiments were carried out. Two control groups were used (groups 1 and 2) in which recombination was done in the absence or presence of estradiol. Two experimental groups consisted of: group 3 in which the nuclei were preincubated with progesterone; and group 4 in which the cytosol was preincubated with progesterone. It is important to note that, following progesterone exposure of either isolated fraction, measures were taken to remove free progesterone prior to recombination. Thus, the effects of progesterone during recombination should result from an action initiated prior to this step, in the isolated subcellular fraction. Recombination of the cytosol and nuclei in the presence of estradiol (group 2) brought about the expected decrease in cytosol estrogen receptor binding with an increase in nuclear retention (Fig. 2). The total receptor concentration (cytosol + nuclei) was similar in these two groups (1426 &-250 fmol/uterus in group 1; 1676 + 222 in group 2). In the presence of progesterone, regardless of
whether this steroid was introduced into the cytosol or nuclei prior to recombination, the recombination of the subfractions in the presence of estradiol resulted in a significant decrease in the nuclear accumulation of the estrogen receptor (groups 3 and 4) as compared to the estrogen control not treated with progesterone (group 2) (Fig. 2). No differences were found in specific cytosol estrogen binding levels among groups 2, 3 and 4. It is remarkable that this response of progesterone, previously demonstrated to require mediation by the progesterone receptor in vivo (Smanik et al., 1983), can be elicited by the low progesterone level of 0.25 nM. This is far below the saturation level of progesterone receptors and, although it is difficult to compare in vivo administration doses with in vitro addition of steroid, it appears that the decrease in estrogen receptor activity by progesterone is an extremely sensitive response. The possibility needed addressing that the effect of progesterone on estrogen receptor dynamics involved a decrease in the stability of the estrogen receptor during the preincubation process rather than an active progesterone-mediated event. To test this, uterine cytosol was incubated in the absence or presence of 0.25 nM pro-
Ep at recombination Progesterone to cytosol Progesterone to nuclei
-
+ -
+ -
+ + -
-
+ -
+ +
+ -
Fig. 2. Effect of 0.25 nM progesterone preincubation with either nuclei (group 3) or cytosol (group 4) on estradiol receptors, following recombination at 22 o C for 1 h in the presence of 3 nM estradiol. Groups 1 and 2 served as controls without (group 1) or with (group 2) the addition of estradiol at the time of recombination. Number of observations = 5. * P < 0.05 relative to all cytosol groups; * * P < 0.05 relative to all nuclear groups. As specified by our receptor assays, cytosol receptor levels represent unoccupied forms, whereas nuclear receptor levels are total.
115 TABLE
1
THE STABILITY OF CYTOSOLIC UTERINE ESTROGEN RECEPTOR IN THE PRESENCE OF PROGESTERONE Rat uterine cytosol was treated with 3 nM estradiol for 30 min at 4O C. After removal of excess estradiol by dextran-charcoal adsorption, 0.25 nM progesterone was added for different periods of time at 4O C and 22O C. After removal of free progesterone with dextran-charcoal, estrogen receptor level was measured. Exposure time
30 mill lh 2h
Cytosol receptor (fmol/mg protein) Without progesterone
With progesterone
4°C
22OC
4OC
22OC
275 + 14 256 f 36 299 f 32
245 f 73 273 f 34 257 rt 20
272 f 19 238k19 259+ 18
222* 8 204k55 300*7s
gesterone at 4O C and 22O C for 30 min, 1 h and 2 h, and estrogen receptor concentrations were determined. The results in Table 1 show no detri-
mental effects of progesterone or the incubation conditions on the estrogen receptor binding capability. Experiments were done to determine whether the progesterone effect of decreasing the nuclear level of estrogen receptor binding was directed toward a specific nuclear receptor subpopulation. For this purpose, total and unoccupied estrogen receptors were assayed in uterine cytosol and nuclei in groups 1, 2 and 4. The preincubation of uterine cytosol with progesterone resulted in a decrease in the total and occupied estrogen receptor binding activity of the nuclear fraction (group 4, Fig. 3) as compared to nonprogesterone-treated controls (group 4, Fig. 3). The effect of progesterone appeared to be exclusively on the occupied estrogen receptors in the nucleus since occupied and unoccupied cytosol fractions, and unoccupied nuclear fractions, were unchanged by preincubation with progesterone, Discussion
CYTOSOL ESTROGEN RECEPTORS
1500,
Control bJrouP 1)
Estmdid kJ~P 2)
Estmdid and Progesterone kJ~P 41
NUCLEAR ESTROGEN RECEPTORS
1500, 1250 loo0 750
;i 250 1
Estmdiol and Progesterone IclrOUo 4)
Fig. 3. Uterine cytosol and nuclear total and unoccupied receptor content after preincubation of cytosol with 0.25 nM progesterone (group 4) prior to recombination at 22O C for 1 h with 3 nM estradiol. Groups 1 and 2 are the respective recombination controls without (group 1) or with (group 2) addition of estradiol. Unoccupied receptors were measured under nonexchange conditions; occupied receptors, under exchange conditions. Occupied receptors were the calculated difference between total and unoccupied. n = 3. * P < 0.05 compared to other unoccupied cytosol receptor levels; * * P < 0.05 compared with values obtained after preincubation with progesterone.
The action of progesterone on gonadotropin secretion has been difficult to describe, primarily because it is linked to the corresponding levels and actions of estrogens (Mahesh and Muldoon, 1987). At the molecular level, the interaction between these hormones is best characterized by the well-documented prerequisite for an estrogenic background for manifestation of progesterone receptor activity in most tissues (O’Malley et al., 1970; VuHai et al., 1977). It is highly significant to the present studies that this relationship is also readily demonstrable in isolated cells having normal estrogen receptor dynamics (Horwitz and McGuire, 1978a, b). In terms of the mechanism whereby progesterone can modify estrogenic feedback control of gonadotropin secretion, particularly apparent as an acute response at the anterior pituitary level (Turgeon, 1979), little is known, but this is a crucial step at which reproductive intervention of either a positive or negative direction might be directed, if clearly understood. We have previously described the ability of progesterone to selectively suppress nuclear estrogen receptor binding in the anterior pituitary in vivo (Smanik et al., 1983; Calderon et al.,
116
1987). The progesterone effect appeared to be manifested at the level of occupied nuclear receptors in the anterior pituitary and the uterus (Fuentes et al., 1988). A similar effect has been observed and studied in some detail in the hamster uterus (Evans et al., 1980; Evans and Leavitt, 1980). Our results showed that this phenomenon did not occur in the hypothalamus and it appeared to be related to the level of endogenous estrogen. Moreover, in the absence of estrogen replacement to the ovariectomized immature rat model used, progesterone had no effect, indicating that progesterone receptors were mediating the response. A decrease in estrogen receptor binding capacity of the anterior pituitary by progesterone has been demonstrated to be of biological significance. Progesterone has been found to attenuate estrogen-induced progesterone receptor synthesis (Calderon et al., 1987) and estrogen-induced prolactin release (Brann et al., 1988). These studies did not, however, advance our understanding of how progesterone, in a purely physical sense, was capable of altering estrogen receptor activity. It has been shown in numerous systems that progesterone has very little binding affinity for estrogen receptors (Sherman et al., 1970; Muldoon, 1980), apparently ruling out the classical action of antiestrogens to tie up the receptor in a nonfunctional state (Katzenellenbogen et al., 1979). On the other hand, unmasking of cryptic, perhaps compartmentalized, nuclear receptors, stimulation of the rate at which nuclear receptors are degraded, or interference with binding of the estrogen-receptor complex to functional effector sites on chromatin, are three of many viable mechanisms whereby progesterone might selectively reduce nuclear estrogen receptor activity. These mechanisms cannot be explored easily in an in vivo system, so we turned in the present study to an in vitro analysis of the phenomenon. Recombination experiments, involving separation, treatment and recombination of cytosol and nuclei, were introduced by Jensen’s group (Jensen et al., 1968) in the early history of receptorology. They were used to describe the heat-sensitive nature of receptor accumulation in the nucleus and thus introduced the vital component of receptor activation as an integral part of the mechanism of estrogen action (Jensen and
DeSombre, 1973). Although recent years have seen a changing pattern of intracellular receptor distribution (King and Greene, 1984; Welshons et al., 1984) the concept of cytosol receptors as loosely associated nuclear components reasonably accommodates these changes, and recombination-type experiments still represent a valid means of assessing certain induced changes in receptor populations, particularly those that we are describing herein. Having established that we could readily demonstrate temperature-dependent nuclear uptake of estrogen receptor from uterine cytosol fractions, we examined the influence of progesterone on this process. The possibility that progesterone could alter the stability of the cytosol receptor such that the nuclear complex would be likewise compromised appears to be untenable. Even following exposure of cytosol to progesterone at 22’ C for as long as 2 h, no decrease in receptor binding activity was seen. Thus, the decrease in cytosol receptor binding following recombination with nuclei under translocationpromoting conditions was not intensified by a direct deleterious effect of progesterone on this receptor species. In this in vitro system, incubation in the presence of progesterone mimics the in vivo finding of selective diminution of nuclear estrogen receptor binding levels. A surprising and revealing observation was that progesterone was equally effective whether it was introduced to the cytosol or nuclear fraction prior to recombination, even under conditions where free progesterone was removed from the respective fractions before recombination. This necessitates either that progesterone directly influences the estrogen receptor in either compartment (unlikely since it does not destroy cytosol receptor binding directly) or that it acts through its own receptor, as it appears to do in vivo (Smanik et al., 1983). Although the animals used in this experiment were ovariectomized, we did find measurable levels of progesterone receptor in the tissue. Receptor mediation of the progesterone effect under our in vitro conditions requires that viable unliganded progesterone receptors be present in both the cytosolic and nuclear compartments, and this is not in violation of any current concepts of steroid hormone action, which encompass nuclear localization of unoccupied receptor.
117
These studies demonstrate the ability to extrapolate the in vivo observation of progesteronesuppressed nuclear estrogen receptor activity to an in vitro system that will be more amenable to manipulation. Progesterone’s selective inhibition of the estrogen-occupied pool of nuclear receptors is highly suggestive of a chromatin-based action, and represents an initial foray into the use of this system to define the nature, specificity and site of this action. Acknowledgement This investigation was supported by research grant HD16688 from the National Institute of Child Health and Human Development, National Institutes of Health. References Anderson, J.N., Clark, J.H. and Peck, Jr., E.J. (1972) B&hem. J. 126, 561-567. Anderson, J.N., Peck, Jr., E.J. and Clark, J.H. (1973) Endocrinology 93, 711-717. Brann, D.W., Rao, I.M. and Mahesh, V.B. (1988) Biol. Reprod. (in press). Burton, K. (1956) B&hem. J. 62, 315-323. Calderon, J.-J., Muldoon, T.G. and Mahesh, V.B. (1987) Endocrinology 120, 2428-2435. Evans, R.W. and Lea&t, W.W. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 5856-5860. Evans, R.W., Chen, T.J., Hendry, W.J. and Leavitt, W.W. (1980) Endocrinology 107, 383-390. Everett, J.W. (1948) Endocrinology 43, 389-405. Fuentes, M.A., Muldoon, T.G. and Mahesh, V.B. (1988) Endocrinology Suppl. 122, Abstract 716. Helmond, F.A., Simons, P.A. and Heim, P.R. (1980) Endocrinology 107, 478-485. Horwitz, K.B. and McGuire, W.L. (1978a) J. Biol. Chem. 253, 2223-2228. Horwitz, K.B. and McGuire, W.L. (1978b) J. Biol. Chem. 253, 8185-8191. Jensen, E.V. and DeSombre, E.R. (1973) Science 182,126-134.
Jensen, E.V., Suzuki, T., Kawashima, T., Stumpf, W.E., Jungblut, P.W. and DeSombre, E.R. (1968) Proc. Natl. Acad. Sci. U.S.A. 59, 632-638. Kalra, S.P. and Kalra, P.S. (1974) Endocrinology 95, 1711-1718. Katzenellenbogen, B.S., Bhakoo, H.S., Ferguson, E.R., Lan, N.C., Tatee, T., Tsai, T.S. and Katzenellenbogen, J.A. (1979) Recent Prog. Harm. Res. 35, 259-300. Kim, K. and Ramierz, V.D. (1982) Endocrinology 111.750-757. King, W.J. and Greene, G.L. (1984) Nature 307, 746747. Korach, KS. and Muldoon, T.G. (1974) Endocrinology 94, 785-793. Liu, J.H. and Yen, S.S.C. (1983) J. Clin. Endocrinol. Metab. 57, 797-802. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. Mahesh, V.B. (1964) Steroids 3, 647-661. Mahesh, V.B. and Muldoon, T.G. (1987) J. Steroid Biochem. 27, 665-675. McPherson, J.C. and Mahesh, V.B. (1979) Biol. Reprod. 20, 763-772. McPherson, J.C., Costoff, A. and Mahesh, V.B. (1975) Endocrinology 97, 771-779. Muldoon, T.G. (1980) Endocr. Rev. 1, 339-364. Muldoon, T.G. and Cidlowski, J.A. (1980) J. Biol. Chem. 255, 3100-3107. Odell, W.D. and Swerdloff, R.S. (1968) Proc. Natl. Acad. Sci. U.S.A. 61, 529-536. G’Malley, B.W., Sherman, M.R. and Toft, D.O. (1970) Proc. Natl. Acad. Sci. U.S.A. 67, 501-508. Peduto, J.C. and Mahesh, V.B. (1985) Neuroendocrinology 40, 238-245. Rao, I.M. and Mahesh, V.B. (1986) Biol. Reprod. 35, 1154-1161. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672. Sherman, M.R., Corvol, P.L. and G’Malley, B.W. (1970) J. Biol. Chem. 245, 6085-6096. Smanik, E.J., Young, H.K., Muldoon, T.G. and Mahesh, V.B. (1983) Endocrinology 113, 15-22. Steggles, A.W. and King, R.J.B. (1970) Biochem. J. 118, 695-701. Turgeon, J.L. (1979) Endocrinology 105, 731-736. VuHai, M.T., Logeat, F., Warembourg, M. and Milgrom, E. (1977) Ann. N.Y. Acad. Sci. 286, 199-209. Welshons, W.V., Lieberman, M.E. and Gorski, J. (1984) Nature 307, 747-749.