Evidence for a human ovarian progesterone receptor

Evidence for a human ovarian progesterone receptor BARRY R. JACOBS, M.D. SUSAN SUCHOCKI ROY G. SMITH, PH.D. Houston, Texas The aromatization of androgens to estrogens in the ovary and the roles of estrogens in promoting fo!!!cu!ar gro\vth have been described by a number of investigators. Furthermorel specific receptors for estrogens have been defined in granulosa cells as well as other target tissues. In other estrogen-responsive tissues, there is also a progesterone receptor, which mediates estrogen antagonism at the molecular level. We have previously described an ovarian progesterone receptor in an animal model. We now describe a heat-labile ovarian cytoplasmic protein that specifically binds progestins with a very high affinity. The equilibrium dissociation constant for progesterone (Kd) = 4.8 x 10- 9M and it is present in the ovary in a concentration of 3.3 x 10- 13 moles/mg of protein. Since the characteristics of this protein are compatible with an ovarian progesterone receptor, there is the implication that the ovary is a target organ for progesterone. Thus, progesterone may have a role in the modulation of ovarian physiology. (AM. J. OBSTET. GYNECOL. 138:332, 1980.)

THE MODULATION of ovarian physiology by steroid hormones had been discussed by a number of investigators. Goldenberg and associates 1 found that estrogens stimulate follicular grovvth and Richards and colleagues 2 described the increased binding of folliclestimulating hormone (FSH) to granulosa cells stimulated by estrogens. This increased binding of FSH is accompanied by an increase in the aromatization of testosterone (T) to estradioJ.2 These effects of estrogen on granulosa cells are probably mediated by ovarian estrogen receptors such as those characterized by Scott and Rennie 3 and Richards. 4 Therefore, there exists in the ovary a positive feedback of estrogen. In every biologic system studied, there is a negative feedback for each positive one. In the uterus progesterone (P) provides this opposition to estrogen. P decreases the replenishment of estrogen receptors 5 and increases the metabolism of the potent estradiol-17 f3 (E 2 ) to the less potent estrone. 5 Is there a similar From the Department of Obstetrics and G_vnecology and the Laboratory of Molecular Endocrinology, Division of Urology, Department of Surgery, The University of Texas Medical School at Houston. Received for publication January 4, 1980. Rev~1ed

April28. 1980.

Accepted May 29, 1980. Reprint requests: Barry jacobs, M.D., Department of Obstetrics and Gynecology, The University of Texas Medical School at Houston, 6431 Fannin, Suite 3270, Houston, Texas 77030.

332

P-mediated phenomenon in the ovary? The first step in answering this question is to determine if there is an ovarian P receptor. Such a protein was first described by us in the calf. 11 In an effort to determine if there is a similar protein in the human, the experiments described in this communication were performed.

Material and methods All chemicals and solvents were reagent or analytical grade and were obtained from commercial suppliers. Ribonuclease A (bovine pancreas), deoxyribonuclease II, trypsin (Type III), and pronase (from Streptomyces griseus) were obtained from Sigma Chemical Co. Unlabeled steroids, P, 'r, dihydrotestosterone (DH'T), E:_!, cortisol (F), and diethylstilbestrol (DES) were obtained from Steraloids, Inc. Tritiated steroids, P ( 1H-P) and R5020 (IH-R5020), as well as unlabeled R5020 were obtained fron1 Ne\v England Nuclear. Ovaries were obtained from premenopausal gynecologic patients undergoing pelvic laparotomy for adnexectomy, either associated with hysterectomy or after hysterectomy. There was no selection on the basis of phase of the menstrual cycle and none of these patients was under the care of the investigators. All procedures were performed for medical indications (e.g .. residual ovary syndrome, pelvic inHammatory disease, and ectopic pregnancy). Tissue obtained in the operating room was immediately immersed in saline at 0° C and transported to the pathologist. After his evaluation of the ovaries, they were transported to the research 0002-9378/80/190332+05$00.50/0

©

1980 The C. V. Mosby Co

Ovarian progesterone receptor

Volume I:IH :-.Jumber :1

laboratorv. All tissue preparations and incubations were perf()rmed at 4° C. The tissue was minced with knife and scissors. Each gram of tissue was then homogenized in 4 ml of buffer ( lO mM tris (hydroxvmethyl) methylamine-hydrochloride, pH 7.4, 1.5 mM ethylenediaminetetra-acetate, 12 mM thioglycerol. and 10% glycerol) with the use of 10second bursts of a TissueMizer (Tekmar). The resulting homogenate was centrifuged at 800 X g for 20 minutes in a Sorvall RC-5B centrifuge. The resulting supernatant was centrifuged at 200,000 X g for I hour in a Sorvall OTD 65 ultracentrifuge with an AH 650 rotor. Endogenous steroids were removed with an equal volume of a dextran-charcoal suspension in the same buffer. After a l-hour incubation, the charcoal was removed from the suspension by centrifugation at 800 X g for 10 minutes. The resulting cytosol was used in the experiments described. All assavs were performed with approximately 6 nM tritiated steroid. Cortisol (l o- 7M) was added to the cytosol prior to all studies. This was done to prevent binding of 1H-P to cortisol-binding globulin (CBG). Scatchard analysis of binding' was performed with a 100-f(>ld excess of unlabeled steroid to displace specific binding of tritiated steroid in a concentration range of 0.4 to 12 nM. Competition curves were performed with the use of competitors in a concentration range of 500 pM to I !J.M. Incubations were at 4° C overnight. Unbound steroid was removed from the incubation solutions after a 5-minute exposure to 0.5 ml of the dextran-charcoal suspension described above. The charcoal was removed bv centrifugation at 800 x g for I 0 minutes in the Sorvall RC-5B. Under the incubation conditions described above macromolecular bound tritiated steroid to be chromatographed was extracted three times with five volumes of anlndrous ether. This removed more than 98% of the labeled steroid. After the ether was evaporated under nitrogen, the residue was redissolved in benzene: methylene chloride (I: I) and spotted on thin-laver silica gel sheets (Eastman). Thin-layer chromatography was performed according to the method described by Strott" to separate P from its metabolites. Pure preparations of P, 5a-pregnane<1,20-dione, and 4-pregnen-20,8-ol-3-one were chromatographed simultaneously as references. Sucrose density gradients (5% to 25%) were made with a Beckman density gradient former. After incubation with "H-R5020, unbound steroid was removed from cvtosol with charcoal. Then 200 !J.I of cytosol labeled with :lH-R5020 and 40 !J.l of ovalbumin (Svedberg _cr • _. r,,, n -, luetnuem L.:>J = .). 1 J a no yeast a1cono1 oenyorogenase (S = 7.6) which had been labeled with 1 ~C were layered on the gradients. 1~he HC-labeled proteins served as 1

•

•

•

•

1

•

•

333

'::K~,. "~

80

•

70 c

60

=

50

'0

=

=

~

•

•

\

40

;; ~

'·

., \

1n

'

101

oI

I

10-'

10-

E,

3

\

'·

~ I

10-

1

~

~T

T--ip

10-l .3pg/ml

Log Dose Fig. l. Competition curves comparing displacement of '1H-P by the naturally occurring potent sex steroids P, T, and E 2 demonstrate relative binding affinities of the macromolecule for these hormones. The result~ of the binding assays are displayed as percent bound versus the nawral log of the concentration of the competing steroid in micrograms per milliliter over a concentration range of approxin1ately 500

p~"f

to l

ILM. The curves were drawn with the aid of a computer a(-

cording to the program of Rodbard and associates.'"

Table I. Results of competitive binding assavs Percent di1pl.acement by mmpetitor Competitor p Ez

DES DHT T

Fiftyfold excess

I oo . Fold exre.ss

100 13.9 ± 3.9 <<0.5 25.5 ± 2.8 45.3 ± 3.8

100 22.6 ± 0.05 3.4 ± 2.2 40.3 ± 0.5 60.1 ± 1.3

sedimentation markers. Gradients were centrifuged at 200,000 x g for 2% hours in the OTD 65 ultracentrifuge with a TV 865 vertical rotor. Enzvme degradation studies were performed in a constant-temperature water bath at 37 C for 30 minutes. Heat-stability studies were performed at 37° C f(n 1 hour. After these incubations, the assay tubes were immediately transferred to an ice slush and the remainder of the assay was performed at 4° C. Enzyme degradation studies were performed in the presence of P and R5020 and a 10 !J.g/ml concentration of trypsin, pronase, ribonuclease (RNase) or deoxyribonuclease (DNase). Tritium disintegrations were counted in a Packard Tri-Carb iiquid-scintiiiation counter with 7 mi of Scintiverse (Fisher) used as a cocktail. Counting efficiency wa~ 4,'J'Ic. rnHeJIJ ueternnnauons were penormea as --

0""' rrt

T1o

-.

~

-

1 -.

•

. •

,.

1

Onobe1 ! , ! ~HO Am . .J. Ob,tt·t. C\necol

334 Jacobs, Suchocki. and Smith

.. l

.

.

7.68

l~j

I

,/\

10

/

0

:e D.

8

()

;F

••

I \

12

)(

16~ ~

3.75

I

14

\. . '\

6

'-•

.I

4 '.

2~ "" .' .__.-·-·-·-· ,/

X

!:!::

.......

4

2

5

10 b

I

I:C

8 10 12 Fraction Number

15

14

18 top

t

Fig. 2. Sucrose density gradients (5% to 25%) were centrifuged in a vertical rotor at 200,000 x g for 2% hours. The fractions of each gradient contained a single peak of macromolecular bound tritiated progestin. Ovalbumin (S = 3.7) and yeast alcohol dehydrogenase (S = 7.6) were labeled with 14C.

•

12

•

A

-~

\.\ \

61 4j

·j I

2

4 I

.\

6

I

8

Bound x 10'0 Molar

Fig. 4. Ovarian cytosol was incubated with 3 H-R5020 over a concentration range of 0.4 to 12 n~,.f. Identical incubations were performed with a 100-fold excess of radioinert R5020 to determine nonspecific binding. The affinity of binding was determined by Scatchard analysis. Kd = 1.9 nM, and th~ concentration of binding sites was 3.3 x 10- 13 moles/mg of protein.

c: 0

(.)

2

c: Q. "':>.

....

1-

;::

....0

a.

iii

z

a:

z c

;--

Fig. 3. Cytosol preparations and progestin were incubated at 37° C for 30 minutes. The enzymes trypsin, pronase, RNase, and DNase were evaluated to determine their ability to interfere with binding of "H-P. described by Bradford 9 by reacting soluble protein with Coomassie Blue. Bovine serum albumin was used to establish the standard curve.

Resuits Competive binding assays were initially performed with the use of a fiftyfold and a 100-fold excess of competitors in the presence of F. The results, given in Table I, show that displacement of 3 H-P is significantly greater by P than by any of the other hormones (n = 3). These data are expressed as the mean ± SD. The capacity of each competitor to inhibit :JH-P binding is expressed relative to the displacement of :lJI-P by

radioinert P. P is used here as a standard and assigned a value of 100%. Since F was added (concentration = 100 nM) to each of the incubation tubes prior to the addition of other hormones, no competition for :JH-P displacement by F is shown since it is negligible or nonexistent. Further and more complete studies of the characteristics of competition for :JH-P binding by naturally occurring potent sex steroids in the human are graphed in the competition curves shown in Fig. 1. The data are expressed in terms of percent bound versus the natural log of the dose of each of the competitors, P, £ 2 , and T. The curves were plotted by a Wang Model 2200 computer by means of the technique described by Rod bard and associates 10 for estabiishing curves for radioimmunoassays. To achieve 50% displacement of 'lH-P by T requires approximately 120 times as much T asP. Also note that E 2 displaces only a minimal percentage of the aH-P and then only at very high concentrations. For analysis of receptors by sucrose gradient centrifugation, R5020 was used as the ligand since it dissociates more slowly from the receptor than P. 11 The utilization of a vertical rotor significantly decreased the

Ovarian progesteronH receptor

Volume i'l~ Number :1

I

14J . -~

10 0

:;:; "'

c

0

c:i

"?'

·\· \

12~

"'c

.,

\

N

Ci


= ~

16

"'.;,c ., c .,"" a;:

"'

c

"'G;

"'"'cC'> ct"'

•

335

.,

o;

C'>

~

0

....

0::

~

~

"'

~

14

8

X

u.

iii

12

6

= ....' X

4

10

:::E A.

r..:>

2~

J

• I

""~

." I

.. I

g

\

. )I. I

g

...

IV

Bound x 10'0 Molar

Fig. 5. Similar incubations to those in Fig. 4 with 3 H·P used as the ligand and radioinert P used as the competitor were performed. Scatchard analysis of binding of 3H-P demonstrates a Kd = 4.8 nM and a concentration of binding sites similar to those of R!\020. centrifugation time, which further minimized dissociation of radiolabeled steroid from protein. The result of this technique was a single band of specially bound radioactivity. This migrated with an S = 3.6 ± 0.2 (n = 4). Comparable values were found when P was used as a ligand but a small peak of radioactivity was noted near the top of the gradient under these conditions, implying a detectable amount of dissociation even in the relatively short centrifugation. A representative sucrose gradient is shown in Fig. 2. Similar to our previously described finding in the bovine ovarv, 6 the binding of progestin is to a heatlabile protein. Incubation of the cytosol without steroid in a ~no C water bath for 1 hour destroyed over half the specific binding ofR5020 but the presence of progestin in the incubation mixture apparently somewhat protected the macromolecule. Enzyme degradation studies demonstrated a marked diminution of specific binding bv the proteolytic enzymes pronase and trypsin but not bv RNase or DNase (Fig. 3). The modest decrease in binding effected by RNase may be the result of contamination of the enzyme bv a proteolytic enzyme. Scatchard analysis of :lH-R5020 binding, as shown in Fig. 4, revealed a single binding component with an equilibrium dissociation constant (Kd) = 1.9 ± 1.2 X

8

• I

6~ 4j

j

-l

'''''''' ,,,,, 2

4

6

8

10

12

14

Fraction Number

Fig. 6. After incubation of 3 H-P with ovarian cytosol, free steroid was removed by dextran/ charcoal. Macromolecular bound steroid was then extracted with anhydrous ether. The residue remaining after evaporation of the ether was chromatographed on thin-layer gels. Authentic standards of 4pregnen-20{3-ol-3-one, progesterone, and Sa-pregnane-3,20,dione were used as reference standards. Both nonspecific binding and total binding were evaluated. The mathematical difference between these is the dispiaceabie binding p!oued.

10- 9 M when determined over a concentratJon range of 0.5 to 10 nM. The apparent concentration (C) = 330 ± 35 fmoles/mg of protein (n = 5). Binding of P was also of high affinity (Fig. 5)-Kd = 4.8 ± 2.9 x 1o- 9 M-and the number of binding sites was similar to that of R5020 (n = 5). The Kd of this pn>tein represents a high affinity for progestins under the equilibrium conditions described. Thin-layer chromatography (n = 3) of specifically bound tritiated P demonstrate a single peak of radioactivity which migrated with the authentic progesterone standard (Fig. 6). This implies that all of the bound steroid is probably P and anv metabolism of the tritiated hormone is not apparent in the binding studies. This differs slightlv from our findings in bovine ovarian cytosols which apparently metabolized approximatelv 20<,!(. of the "H-P.~ More important. bowe\·er. is

336

Jacobs, Suchocki, and Smith .\rn.

that progesterone itself is bound with very high affinity to a cytoplasmic protein.

Comment The P-binding protein described in this communication has characteristics of a P receptor. It is of high affinity and limited capacity. The protein's affinity for progestins is highly specific and the fact that it binds R5020 even in the presence of excess cortisol is further evidence that it is a Preceptor. The Svedberg sedimentation coefficient, as estimated by the sucrose gradient studies, is compatible with those which have been reported previously.~· 12 · 1 " Although both 4S and 8S progesterone receptors have been described, the single 3.6S component described in this communication is consistent with our previous observations in the animal model system 6 and is similar to that described for the purified P receptor isolated from the human uterus. 14 The results of thin-layer chromatography on the ether extract of the steroid bound to the receptor demonstrate that P itself is bound and is not a metabolite such as 5a-dihydroprogesterone. Because the experiments discussed here were performed on human tissues, it was not possible to demonstrate tissue specificity with the classic nontarget tissues, the spleen, lung, or heart, from the same patient.

REFERENCES I. Goldenberg, R. L., Vaitukaitis, ]. L., and Ross, G. T.: Estrogen and follicle stimulating hormone interactions of follicle growth in rats, Endocrinology 90:1492, 1972. 2. Richards, J., Uilenbroek, J. Th. ]., and Jonassen,]. A.: Follicular g-rowth in the rat: A reevaluation of the roles of FSH and LH, in Channing, C. P., March, J ., and Sadler, W. A .. editors: Ovarian Follicular Growth and Corpus Luteum Function, New York, 1979, Plenum Publishing $.

4.

5. 6. 7. 8. 9.

Corporation, p. II. Scott, R. S., and Rennie, P. I. C.: An estrogen receptor in the corpus imeum of the pseudopregnant rabbit, Endocrinology 89:297, 1971. Richards, J. S.: Content of nuclear estradiol complex in rat corpora lutea during pregnancy-relationship to estrogen concentration and cytosol receptor availability, Endocrinology 96:227, 1975. Tseng, L., Gusberg, S. B., and Gurpide, E.: Estradiol receptor and 17,13-dehydrogenase in normal and abnormal endometrium, Ann. N.Y. Acad. Sci. 286:190, 1977. Jacobs, B. R., and Smith, R. G.: Evidence for a receptorlike protein for progesterone in bovine ovarian cytosol, Endocrinology 106: 1276, 1980. Scatchard, G.: The attractions of proteins for small molecules and ions, Ann. N.Y. Acad. Sci. 51:660, 1949. Strott, C. A.: Metabolism of progesterone in the chick oviduct-relationship to progesterone receptor and biologic activity, Endocrinology 95:825, 1974. Bradford, M. M.: A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the

J.

OnobeJ l. I~11-ill Obstf't C:meroL

The ovaries used in the experiments described in 1 his communication were removed for a \·ariety of reason-;. All of the patients were in the reproductive age range and had active fi>llicles. Some had had a previous II\ sterectomy and it was nut possible to date the stage ot the menstrual cycle at the time of operation. In those women having hysterectomy with adnexertomy. the time of the cycle was noted. In contrast to our findings in the bovine ovary, specific progesterone binding w~ts noted in both the ti>llicular and luteal phase-;. The number of assays in each phase is too small to identit\ a significant difference in the concentration of binding when the two phases of the cycle are compared. This is the first description of a putative human ovarian P receptor. C ntil a biological response has been associated with the interaction of P with this P-speciflc ovarian binding protein it cannot be strictly termed a receptor. Still to be performed are studies to determine the exact action of P on ovarian phvsiologv. Goodman and Hodgen'" found that there is apparenth· a direct inhibitory effect of progesterone on follicular growth. Cunningham and associates u; described an inhibition of ovulation by another steroid known to be bound bv P receptors. These findings imply that the ovary is a target organ for P and that P may be the negative feedback for follicular growth.

principle of protein-dye-binding, Anal. Biochem. 72:248, 1976. 10. Rodbard, D., Rayford, P. L.. Cooper,]. A., and Ross. G. T.: Statistical quality control ofradioimmunoassavs,J. Clin. Endocrinol. Metab. 28:1412, 1968. II. Raynaud, J.P.: R5020, A tag for the progestin receptor, in McGuire, W. L., Raynaud, J. P., and Baulieu, E. E., editors: Progesterone Rec:eptors in Normal and Neoplastic Tissues, New York, 1977, Raven Press, p. 9. 12. Kontula, K., ]anne, 0., Luukkainen, T., and Vihko, R.: Prugesteron~

13.

14. 15.

16.

binding in huutan 1nyornetriun1. Ligand

specificity and some physicochemical characteristics, Biochim. Biophys. Acta 328:145, 1973. Young, P. C. M., and Cleary, R. E.: Characterization and properties of progesterone binding components in human endometrium,]. Clin. Endocrinol. Metab. 39:425, 1974. Smith, R. G., Iramain, C. A., Buttram. V. C., and O'Malley, B. W.: Purification of human uterine progesterone receptor, Nature 253:271, 1975. Goodman, A. L., and Hodgen, G. D.: Systemic versus intraovarian progesterone replacement after luteectomy in rhesus monkeys: Differential patterns of gonadotropins and follicle growth, J. Clin. Endocrinol. Merab. 45:837, 1977. Cunningham, G., Goldzieher, j.. de Ia Pena. A., and Oliver, M.: Mechanisms of ovulation inhibitions by triamcinolone acetonide, J. Clin. Endocrinol. Metah. 46:8, 1978.