Effects of progesterone on gonadotropin-releasing hormone receptor concentration in cultured estrogen-primed female rat pituitary cells

0960-0760/92 $5.00 + 0.00

J, Sterbid IF~cbem. Molec. Biol. Vol. 42, No. S, pp. 831-839, 1992

PnnM in Great Britain.All rishts reserved

Copyright C) 1992PerllamonPressLid

EFFECTS OF PROGESTERONE ON GONADOTROPIN-RELEASING HORMONE RECEPTOR CONCENTRATION IN CULTURED ESTROGEN-PRIMED FEMALE RAT PITUITARY CELLS* G. EMONS,"'t J. NILL,t'2 R. STUtM2 and O. OgTMA~'~t'2 Departments of Obstetrics and Gynecology, t Philipps-Universitiit,P i l ~ M n 3, 3550 Marburg and 2MedizinischeUniversitit, 2400 Lfibeck, Germany

(Received 2 September 1991)

Smmmary--Acute (0.5-4 h) treatment of estradiol (E)-primed female rat pituitary cells with progesterone (P) augments gonadotropin-releasing hormone (GnRH)-induced LH release, whereas chronic (48 h) P-treatment reduces pituitary responsiveness to the hypothalamic decapeptide. Dispersed E-primed (48 h, 1 nM) rat pituitary cells were cultured for 4 or 48 h in the presence of 100 nM P to assess the effects of the progestagen on GnRI-I receptors and on gonadotrope responsiveness to the decapeptide. P-treatment (4 h) significantly augmented GnRl-l-receptor concentrations (4.44 + 0.6 fmol/10e cells) as compared to cells ~ a ~ d only with E (2.6 + 0.5 fmol/10e cells). Parallel significant changes in CmRH-induced LH secretion were observed. The acute increase in GnRH-receptor number was nearly maximal (180% of receptor number in cells treated with E alone) within 30 min of P addition. Chronic P-treatment (48 h) significantly reduced pituitary responsiveness to CmRH as compared to E-treatment. The GnRH-receptor concentrations (3.9 + 0.6 fmol/10~cells), however, remained elevated above those in E-primed cells. GnRI-I-receptor affinity was not h ~ n M by any of the different treatments. These results indicate that the acute facilitatory P-effect on GnRH-induced LH release is at least chronologically closely related to an increase in GnRH-receptor concentration. The chronic negative P-effect on pituitary responsiveness to GnRH, however, shows no relation to changes in available GnRH receptors.

negative effect on the release of the gonadotropin [10-19]. It has been well documented that Ovarian steroids exert direct modulatory effects the positive estrogen-effect on GnRH-induced on pituitary release of luteinizing hormone LH secretion in cultured pituitary cells is (LH): short-term treatment (0.5-4 h) of female accompanied by a sit,nificant increase in the rat pituitary cells in culture with estradiol (E) number of GnRH receptors[4,20-22]. The reduces gonadotropin-releasing hormone short-term negative estrogen-effect on gonado(GnRH)-induced LH release [1-6]. In contrast, tropin secretion in this system is paralleled by a long-term treatment ( > 12 h) with the estrogen reduction of GnRH-receptor density, which enhances CmRH-induced secretion of the becomes significant within 30 min [4]. This strikgonadotropin [1-9]. Progesterone (P) treatment ing parallelism of estrogen-induced changes in of estrogen-primed pituitary cells for short GnRH-receptor concentration and the sensitivperiods (0.5 to 4h) enhances GnRH-induced ity to GnRH in pituitary cells has supported LH release, while long-term ( > 1 2 h ) co- the notion that these phenomena are causally incubations of the cells with E and P exert a related [4, 20-22]. The objective of this series of experiments was to examine the effects of short- and long*This study was presented in part at the 35th Symposium term P-treatment on GnRH-receptor density in of the Deutsche Gesellschaft fiir Endokrinologie, Bonn, E-primed female rat pituitary cells in culture Germany, 1991. and the possible relationship of tbe~ effects with '~l'o whom correspondence should be addmm~ at: l(linik for Frauenludlkunde und Geburtahflfe, Philipp~ the well-known alterations in CmRH-induced Universitit, Pilgrimstein 3, 3550 Marburg, Germany. LH secretion caused by the progestin. INTRODUCTION

831

832

G. EMONSet al. EXPERIMENTAL

Hormones E, P and GnRH were purchased from Sigma (Deisenhofen, Germany). Appropriate stock solutions were prepared in ethanol (steroids) or phosphate-buffered safine (PBS), containing I g/1 of bovine serum albumin (BSA) (GnRH).

Dispersion and culture of rat pituitary cells Adult female Wistar rats (180-200g) at random stages of the estrous cycle (Winkelmann, Borchen-Kirchborchen, Germany) were used for the preparation of dispersed pituitary cells as described previously [9, 23] except that phenol red-free medium was used throughout [6]. For binding experiments 6 x l0 s viable cells were plated in 35 x 10mm culture dishes (Falcon, Lincoln Park, NJ, U.S.A.) in a volume of 3 ml of Medium 199 (phenol-red free, Biochrom, Berlin, Germany) containing L-glutamine, 0.14% of sodium bicarbonate, 100U peniciUin/ml, 100/zg streptomycinsulfate/ml and 10% horse serum (Biochrom) pretreated with 2% charcoal (Norit A) and 0.2% Dextran T70 (Pharmacia, Uppsala, Sweden). For the assessment of GnRH- induced LH release 2 x 105 viable cells from the same preparations, respectively, were plated in 22 x 10 nun wells (Costar, Cambridge, MA, U.S.A.) in 1 ml of the above medium. Before hormonal treatment the cells were allowed to attach to the wells for 48 h at 37°C in a water-saturated atmosphere of 5% CO: in air.

Treatment of cultured pituitary cells Receptor experiments. After the 48-h preincubation (see above), media were removed and the cells incubated for 10rain with fresh medium (see above). Then these media were removed and the cells exposed to media containing either E (1 nM) alone (n = 24 wells) or E (1 nM) plus P (100nM) (n--12 wells) or medium containing vehicle (V = 0.2% ethanol in medium) alone (n = 12 wells). These media were replaced by media of the same composition after 24 and 44 h. 12 of the E-treated cultures were then exposed additionally to 100nM P. After another 4 h of incubation the cells were subjected to the radioreceptor assay (RRA) for GnRH described below. For the experiments on the kinetics of the short-term P-effects, cultures were treated for 44 h with 1 nM E as described above. The cells were then incubated with either 1 n M E or E

plus 100 nM P. Incubations were stopped after 0.5 to 4 h for the RRA (see below). L H experiments. Aliquots (2 x 105 cells/well) of the respective cell preparations used for the receptor experiments were treated at the same time for 44 h with E (1 nM), E plus P (100 nM) or medium containing vehicle alone (V), as described above. The cells were washed and incubated with medium of the same composition. Half of the E-treated cultures were additionally exposed to 100 nM P. Cells were then incubated in the same medium supplemented with increasing concentrations of GnRH (10 pM to 1 tiM) for 4 h before the media were removed and saved for LH determination (see below) at -20°C. All experiments were run in triplicate and repeated with separate cell preparations at least three times.

GnRH -RRA The GnRH agonist analog [D-Ala6-des Gly ~°] GnRH ethylamide (GnRH-A) obtained from Sigma, was labeled with ~25Iusing the lactoperoxidase method according to Clayton et al. [24] and was purified on a Sephadex G-25 column (50 x 2.5 cm) using 0.1 M acetic acid, containing 0.25% of BSA as the eluent. The specific radioactivity of the labeled peptide determined by self-displacement in a pituitary RRA was between 1200 and 2000#Ci//zg. Maximal tracer bindability, determined with an excess of pituitary membranes was 40-50% (for further details see Ref.[25]). The RRA was then performed according to the general procedure described by Loumaye and Cart [23, 26] with minor modifications[4]. After removing the media, cells were washed twice with ice-cold PBS, containing 0.3% of BSA and then scraped from the dishes with a rubber policeman. For a complete binding curve experiment the cells from 12 dishes each (i.e. 72 x 10z cells) from the different treatment groups were pooled. For the single-point experiments (short-term kinetics) 2 dishes each (i.e. 12 x 104 cells) from the different treatment groups were pooled. Aliquots were routinely removed for counting and assessment of cell viability by Trypan blue exclusion. Viability was 89.4 + 0.9% (vehicle), 90.1 _+ 1% (1 n M E for 48 h), 90.4 :t: 1.3% (1 n M E plus 100 nM P for 48 h) and 90.6 + 0.4% (1 n M E for 48 h, 100nM P for 4h). There were no effects of the different treatments on the viability or quantity of the cells recovered. Cells were collected at 200g and resuspended in ice-cold RRA-bufl'er (10mM Tris, 0.1% BSA, 0.1%

Progesteroneeffectson CmRI-Ireceptors NAN3, I m M dithiothreitol,p H 7.6) at a concentration of approx. 2 x I06 cells/100/tl.Aliquots of these cell suspensions were saved for determination of protein content [27]. Cell protein contents in control cultures and those from the different treatment groups were virtually identical in all experiments. Cell suspensions (100#1; approx. 2 x 106 viable cells) were incubated for 1.5 h at 0°C in the presence of 30 pM [125I, D-AlaS-des Gly I°] G n R H ethylamide and increasingconcentrations (0, 3, 10, 30, 60, 100, 300, 600, and 1000 pM) of unlabeled G n R H - A in a totalvolume of 300/tl. Non-specific binding was assessed in the presence of 100 n M GnRH-A. All incubations were run in triplicate.For the single-pointRRA, used in the short-term time-course experiments, cell suspensions (approx. 2 x 106 viable cells/tube) were incubated in the presence of a nearsaturating concentration of 2.07 n M G n R H - A (0.12 n M [1251]GnRH-A plus 1.95 n M unlabeled G n R H - A ) in a totalvolume of 300/~I in triplicate according to Clayton et al. [28,29].Non-specific binding was assessed in duplicate in the presence of 10 -7 M GnRH-A. Separation of bound and free labeled ligand was achieved by dilution with 3 ml of ice-cold RRA-buffer and centrifugation at 5000g (04°C, 20 min). This procedure was repeated once. Pilot experiments comparing this separation technique with a centrifugation step at 12,000g had shown that both methods were equally effective.Binding was corrected for the number of viable cellsand is expressed as fmol G n R H - A bound per 106 viable cells.

LH-RIA LH in the culture media was measured by a double antibody RIA using the reagents (reference preparation RP-2 rat LH, AFP-5666 C) and instructions kindly provided by the National Hormone and Pituitary Program (University of Maryland School of Medicine) and the National Institute of Diabetes and Digestive and Kidney Diseases CNIDDK). Inter- and intraassay coefficient of variance were 10 and 3.5%, respectively. Samples from one experiment were run in one RIA.

Mathematical and statistical analyses The data obtained from complete GnRH-Adisplacement curves, were anatyzed using the "Ligand-Program" kindly provided by the Biomedical Computing Technology Information Center, Vanderbilt Medical Center, Nashville,

833

TN (BCTIC). Differences in the number of CmRH-A binding sites between the different groups from each individual experiment were tested for statistical significance as follows: the program was allowed to calculate the number of binding sites for each of the 4 displacement curves (V, E, EP 48 h, EP 4 h) and was then forced to fit step-by-step 2 each of the 4 individual displacement curves with a c o m m o n number of binding sites.The comparison of the goodness of fitsresultingfrom these fittingprocedures (expressed by the F-value) was used to test the hypothesis that there are no differences in the number of binding sites[24].In the same way the Ko-values for the different treatment groups in one individual experiment were tested for significant differences. GnRH-induced L H release was analyzed and tested for significant differences using the "Allfit-Program" by De Lean et al. [30] which was also kindly provided by BCTIC. In addition, the combined parameters from the 6 individual experiments were analyzed as follows: to ascertain that the established E-effects had occurred, the data obtained in the E-group were compared with those from V-treated cultures with paired t-tests when Bartlett-tests had shown that variances were homogenous or otherwise with a Wilcoxon test [31]. Differences between E and EP 4 groups and E and EP 48 groups were tested for statistical significance using paired t-tests, after Bartlett tests had shown that variances were homogenous. The significance levels obtained in these tests were corrected according to Bonferoni [31]. The data from the single-point experiments were analyzed using a Mann-Whitney U-test [31] after transforming them into percent of E-treated controls at time zero ( = 100%).

RESULTS

Chronic E-treatment (E) E-treatment (1 nM) for 48 h increased the average number of GnRH receptors from 1.5 + 0.2 (vehicle) to 2.6 ± 0.5 fmol/106 cells (P < 0.016). In all 6 individual experiments this result was also highly statistically significant by means of the "Ligand" procedure (Fig. 1, Table 1). E-treatment also sifnificantly increased LH release induced by near EDso concentrations of GnRH (0.1 and 1 nM) (Table 2) and the maximal LH release calculated by the "AllfitProgram" (Table 3). The sensitivity of the cells

834

O. Etmm et al

0.25 •

0•00 Exp, de4

Exp. # 3

0.20'

0.15, a, ,%

$=V

0 0 = EP48

0.10

0 = EP4

eo ~

0.05

o

1

2

3

O=E

4

~

o

"

i

2

3

S ~

B (10-" M }

Fig. 1. Scatchard plots (with lines of best fit calculated by the "Ligand program") of G n R H - A binding to pituitary cells from adult female rats treated with medium only (V), E, EP 48 and EP 4. Two representative experiments out of 6 are shown. In experiment 3 no difference in CmRH-receptor concentration was observed between EP 4- and 48-treated cells while in experiment 4, EP 48-treatment reduced GnRH-receptor concentrations. For further details see Table ! and Experimental.

Table I. Effects of treatments with E, EP 48, EP 4 and V on the concentration and affinity (K,) of GnRH receptors in cultured pituitary cells from adult female rats

Receptor concentration

Ko [109 M-t]

[fmol/lO6ceils] Exp. No

V

E

EP 48

1 2 3 4 5 6 • ±SE

1.2±0.05 !.6±0.1 1.9 ± 0.2 1.5±0.1 1.9±i.0 0.9±0.4 1.5±0.2

1.9±0.2" 2.9±0.4" 4.7 ± 0.5' 2.4±0.2" 2.3 ±0.1" 1.4 ± 0.1" 2.6 ±0.5 d

3.1 ±0.2 b 3.6 ± 0.2b 6.4 ± 0.3b 3.1 ±0.1 b 4.9 ± 0.2b 2.6 ± 0.2b 3.9±0.6"

EP 4 2.7+0.2 b 4.6 ± 0.1~ 6.3 ± 0.8b 5.0-1-0.2"" 5.0 ± 0.3b 3.0+0.03 k 4.4 ± 0.6t

V

E

EP 48

EP 4

8.2+0.2 4.5±0.4 3.5 + 0.4 5.0+0.5 6.0±0.6 11+!.2 6.4±2.4

6.8±!.0 4.3± 1.0 3.9 ± 0.6 6.1+0.7 6.0±0.6 16±2.6 7.1±1.7

6.6+0.7 4.2±0.4 3.4 ± 0.3 6.0±0.3 7.8±0.6 12±1.8 6.7±1.2

6,3±0.7 4.2±0.2 3.6 ± 0.7 6.0±0.5 6.0±0.6 13±0.9 6.5±1.4

Complete GnRH-A displacefaent curves (3pM to InM) were obtained in 6 individual experiments on different cell preparations. GnRH-receptor concentrafiom and K,-valuet were calculated and compared statistically for each individual experiment using the "Ligand-Program". The combined data from all 6 experiments were compared by paired t-tes~ after Bartlett-tests had shown that variances were homogenous. Significance levels were corrected aounding to lonferoni (for details see Experimental). t p < 0.00l vs V ( L i g a n d ) , bp < 0.005 VS E (Ligand); ©P < 0.025 vs EP 48 (Lilland); ~P < 0.016 vs V (t-test); ep < 0.01 vs E (t-test); and fP < 0.002 vs E (t-test).

Table 2. Effects of treatments with E, EP 48, EP 4 and V on LH release by cultured pituitary cells from adult female rats induced by GnRH concentrations (10-le, 10-9 M) near the EDse

LH [ ~ ]u,.2!

LH [ng R P . 2 ] relensad by 10-1e M GnRH

released by 10-9 M GnRH

Exp. No

V

E

EP 48

EP 4

V

I 2 3 4 5 6 ,f ,I- SE

3.2+0.3 9.0±1.3 3.5±0.2 8.5 ± 0.4 14±0.2 21 ± 0.5 9.9+2.7

21±1.5 21+0.5 22±0.4 26 ± 0.7 19±0.5 42 ± 0.2 25+3.5'

12±0.2 14+0.3 12±1.2 11 ± 0.3 19±0.3 29 ± 0.6 16+ 2.8b

28±!.2 32±1.7 32±2.0 29 ± !.7 28±!.2 48 ± 0.5 ~3 ±3.1 c

15+1.0 18±0.9 16±0.6 22 ± 0.7 22±0.3 37 + 0.5 22-1-3.3

E 33+0.6 28±0.7 45±2.0 32 ± 0.2 34± 1.5 55 + 4.0 38 +4.0 d

EP 48 27±1.2 23±0.9 29±1.1 23 ± 0.6 28+0.2 42 ± 0.8 29 + 3.0b

EP 4 40±1.5 38+0.2 48+!.0 39 ± !.0 39±0.5 58 + 0.6 44 + 3.0b

Aliquots of the same cell preparations used for the RRA'(Table' li were coitwubated in triplicate with the respective stsmids or vebicie and GnRH (10-~0, I0-9 M). The combined LH data from all 6 experiments were compared by paired t-tests after Bartlett-tests had shown that variances were homogenous. Significance levels were corrected according to ~mferoni (for details see Experimental). ' P < 0.001 vs V (/-test); 6p < 0.01 vs E (t-test); cp < 0.002 vs E (t-test), and dp < 0.005 vs V (t-test).

Progesterone effects on GnRH receptors

835

to GnRH was increased by E resulting in a signiticant decrease of the EDs0-values for GnRH (Table 3). Also minimal LH release (that seen in the absence of G-nRH) calculated by the "Alifit-Program" was significantly increased by E-treatment (Table 3).

li!i { I~:~

°

.

.

.

.

.

:,:2

5

.~.'~

Chronic E- plus acute P-treatment (EP 4) Coincubation of the cells with I nM E for 48 h and 100 nM P for 4 h induced a marked increase (P < 0.002) of average GnRH-receptor concentrations (4.44 + 0.6 fmol/106 cells) as compared to E alone (2.6 4- 0.5 fmol/106 cells). In all 6 individual experiments this result was also highly significant by means of the "Ligand" procedure (Fig. 1, Table 1). EP 4-treatment resulted in significant increases of the LH release induced by 0.1 and l n M GnRH (Table 2) and maximal LH release (Table 3) as compared to cells treated only with E. Also the average EDs0-values were significantly decreased by short-term P-coincubation as compared to E-treated cells, reflecting an increase of the sensitivity of the cells to GnRH (Table 3). Minimal LH release was significantly increased (Allfit) in 2 experiments by EP 4-treatment as compared to E. This effect, however, was inconsistent and did not reach statistical significance when the data of all 6 experiments were compared (Table 3). Chronic E- plus chronic P-treatment (EP 48)

~ ~ + 1

,~r~.

ueo~,

i=I|

V

.j jy

Coincubation of the cells for 48 h with 1 nM E and 100 nM P significantly increased average GnRH-receptor concentrations (3.95 _+0.6 fmol/ 106 cells) as compared to cells treated only with E (2.6 + 0.5 fmol/106 cells). In all 6 individual experiments this result was also highly significant by means of the "Ligand" procedure. In 3 experiments, GnRH-receptor concentrations were however significantly (Ligand) lower in EP 48 cells than in cells treated with P for 4 h while no significant differences were found in the other 3 experiments between GnRH-receptor concentrations in the EP 4- and 48-treated cells (Fig. 1, Table 1). Chronic P-treatment significantly reduced LH release induced by 0.1 and I nM GnRH and maximal LH release as compared to E-treated cells (Tables 2 and 3). The EDs0-values for GnRH were significantly increased by longterm P-treatment as compared to E only, reflecting a decrease in sensitivity, while no significant effects on minimal LH release were observed (Table 3).

836

G. EMow3 et al. 4t

200

P

180

"o 160 "6 140 Ld r-n

g

120 .it

"o tOO,

~t v

m~ e O

o 0.5

i

2 3 4 t Ch] Fig. 2. GnRI-I-Abinding to cultured pituitary cells from adult fera~erats, treated for 0.5--4h with I nME or I nM E plus 100nM P. Before,all celishad been treatedfor 44 h with I nM E. GnRH-A binding was determined using a near saturatingconcentrationof CmRH-Aas describedin Experimental. The data shown are the mean+ SE of 4 independentexperiments.GnRH-Abindingis expressedas a percentage of the binding observed after 44 h treatment with I nME alone(to = 100%).DifferencesbetweenGnRHA binding were compared for each time point using the Mann-Whitney U-test (P < 0.014). For furtherdetails see Experimental.

GnRtt-receptor aOfnity None of the steroid treatments had significant effects on Ka-values of GnRH receptors (Table 1).

Time-course experiments Time-course studies revealed the acute effect of P on GnRH-receptor number in E-treated cells to be relatively rapid. GnRH-receptor number was increased to 180% of control (E alone at time 0) within 30 rain and remained at this elevated level for at least 4 h (Fig. 2). D~CU~ION

Our data confirm earlier observations by others [7, 20, 22] and ourselves [4] that long-term E-treatment increases GnRH-receptor concentrations in cultured pituitary cells. Also the wellknown positive effects of chronic E-treatment on LH release and sensitivity to GnRH [1-9] were clearly evident in our present experiments. If P was added for 4 h to E-treated cells, there was a further significant increase of GnRH-receptor density, accompanied by the established [10-19] additional positive effect of acute P-treatment on LH release. Long-term coincubation (48 h) of cells with E and P resulted in the well-known[10-19] negative effects on LH release parameters as compared to exclusive E-treatment. The GnRH-

receptor concentrations after chronic P-treatment, however, remained significantly elevated above those in E-treated cells and showed no or only inconsistent decreases in comparison to cells treated with P for a short period. None of the different steroid treatments affected K=values of the GnRH receptors, indicating that the differences in GnRH binding were due to changes in receptor numbers and not affinity. According to Loumaye and Catt [23] the hypotonic RRA conditions employed here, measure plasma membrane receptors in broken cells, which are equivalent with the sites measured on the surface of intact pituitary cells. Also Marian and Conn [32] found that GnRH binding sites detected by radioligand assay in pituitaries of rats not previously exposed to GnRH are localized almost exclusively in the plasma membrane fraction. Therefore we presume that the changes in GnRH-receptor concentrations observed under our experimental conditions reflect alterations in the number of GnRH binding sites in the plasma membranes of the cells. It might be argued that the augmentation of GnRH binding observed after P-treatment observed in the present study could be caused by a direct physico-chemical interaction of the steroid with the GnRH receptors. This is, however, unlikely because of the extensive washing of the cells before the RRA and because there were no changes in the Ko-values. The clear parallelism of the marked augmentation in GnRH-induced LH release and the increase of GnRH-receptor concentration caused by 4 h P-treatment suggests that both phenomena could be causally related or are linked by a subcellular mechanism which is positively modulated by acute P-treatment. This hypothesis is further supported by the results of our short-term experiments in which the positive P-effect on GnRH-receptor density was significant after 30 rain. After 30-50 min, the positive P-effect on GnRH-induced LH release was also evident [14, 15,18]. The negative effect of longterm P-treatment on GnRH-induced LH release, however, was not paralleled by a comparable reduction in GnRH-receptor concentration. This indicates that the negative P-effect is not caused by a reduction of GnRH receptors and is probably mediated at a postreceptor step in the GnRH sigual-transduction mechanism. In experiments using secretagogues circumventing the GnRH-receptor [A 23 187, cAMP, phorbol ester, dioctanoylglycerol (Di C8)] Krey and Kamel [17] also observed a positive effect

Progesteroneeffectson GnRH receptors of acute P-treatment on LH release. Together with the positive acute Perfects on spontaneous LH secretion they oinerved in the absence of GnRH [16] they concluded that this P-action does not involve changes in GnRH-receptor availability. In our data, changes in the spontaneous LH release were not sufficiently dramatic to suggest such a clear-cut conclusion. We suggest that in fact, both postreceptor mechanisms and increased receptor concentration might contribute to the P-mediated increase of LH release. We have also previously suggested [4, 5] that the acute negative E-effect on LH release is probably mediated by changes in both GnRH-receptor concentration and a postreceptor mechanism. The discrepancy observed in the present study between the increase in GnRH-receptor concentration and the reduction of GnRH-induced LH release after long-term P-treatment, however, supports the suggestion of Krey and Kamel [15-17] that this chronic P-effect is exclusively mediated at a postreceptor mechanism. Little is known about the mechanism by which female sex steroids modulate GnRHreceptor concentrations. There is a substantial body of evidence that the direct effects of E and/ or P on GnRH-induced LH release are mediated via the respective steroid receptors [3, 13, 16, 19]. It has been shown that the positive long-term E-effect on GnRH-receptor concentrations seems to be dependent on RNA and protein synthesis [21, 22]. This also applies for the positive effect of E on GnRH-induced LH release, whereas the acute negative E-effect on LH release appears to be independent on RNA and protein synthesis [5]. Both, acute and chronic P-effects on LH release appear to be dependent on RNA and protein synthesis [15, 18]. Future experiments should clarify the role of a genomic mechanism in the mediation of acute E and both acute and chronic P-effects on GnRH-receptor concentrations. There is evidence that phorbol ester-induced activation of protein kinase C (PKC), which stimulates LH-exocytosis also increases GnRHreceptor density[33]. We have shown that phorbol ester and OAG-stimulated LH release is reduced by 4 h E-treatment, which is known to reduce GnRH-receptor concentrations [4, 5]. Long-term E-treatment, which is known to augment GnRH-receptor concentration [4, 21, 22] and GnRH-induced LH release [1-9], also increases both PKC-activity in rat pituitary cells and pborbol ester-induced LH release [34, 35].

837

Acute P-treatment, which we have shown to increase CmRH-receptor concentrations, augmerits phorbol ester- or DiCS-induced LH release[17, 35]. Interestingly PKC-mediated LH release was not reduced as expected but was enhanced by chronic P-treatment [17, 35], a situation, where we have still found elevated GnRH-receptor concentrations. Thus E- and/or P-induced changes in the sensitivity of the cells to activators of PKC are in good correlation with steroid-induced changes of GnRH-receptor concentrations. Taking into account the finding of Naor[33] that PKC might be involved in the regulation of GnRH receptors, one could speculate that this system might be the link coupling steroid effects on pituitary LH release and on GnRH-receptor concentration. For completeness' sake it should be mentioned that several authors have addressed the question of P-effects on pituitary GnRH receptors using in vivo models. The results published so far, however, are contradicting and difficult to interpret. For example, Clayton and Catt [29] have found that E, or P, or E plus P can inhibit the postovafiectomy rise of GnRH receptors in adult rats. During the estrous cycle the increase in serum E levels from metestrus to diestrus is paralleled by a rise in pituitary GnRH-receptor levels [28, for review see 36]. The decline of GnRH receptors on estrous has been attributed to the proestrus surge of P [for review see 36]. This drop in GnRH-receptor levels, however, also occurred in the absence of progesterone in the model of ovariectomized E-treated rats before the LH surge despite stable E levels [for review see 36]. Attardi and Happe [36] performed a very careful study in immature rats on both facilitation and inhibition paradigm of P on the E-induced LH surge. They failed to find any significant differences in pituitary content or concentration of GnRH receptors at any time between rats treated with E plus P and their respective E-treated controls. These authors concluded that the in vivo effects of E and P on GnRH receptors might be mediated by changes in secretion of GnRH from the hypothalamus [36]. We also think that the in vivo models are rather complex due to dynamic changes in LH secretion, their dependence from the status of the animals (immature, adult, ovariectomized), the estrous cycle and the light conditions. Apart from direct steroid effects on pituitary gonadotrophs most probably steroid-induced changes of pulse frequency and amplitude of hypothalmic

838

G. EMONSet al.

GnRH secretion will influence the number of pituitary GnRH receptors. Therefore in our investigations on the acute and chronic effects of E [4] and P (present paper) on pituitary GnRH receptors we have first concentrated on direct steroid effects on pituitary gonadotrophs in vitro. Future experiments, using perifusion systems [14] could include G-nRH-pulse frequency and -amplitude as additional variables besides Eand P-treatment to elucidate in vitro changes of pituitary GnRH receptors under paradigms immitating specific situations of the estrous cycle. In summary, the present study gives clear evidence that the acute augmentation of GnRHinduced LH release by P is paralleled by a rapid increase in GnRH-receptor concentrations. It remains to be clarified whether both phenomena are causally related and by which subcellular mechanism(s) these effects are mediated or linked. The long-terra negative P effect on GnRH-stimuiated LH-exocytosis is not related to a comparable reduction of GnRH receptors. The uncoupling of these phenomena after chronic P exposure argues that they are not necessarily causally related. Acknowledgements--We are grateful to the National Institute of Diabetes, Digestive and Kidney Diseases and the National Hormone and Pituitary Program, University of Maryland School of Medicine for the gift of rat LH-RIA reagents. The "Ligand" and "Allfit" programs were provided by the Biomedical Computing Technology Information Center, Vanderbilt Medical Center, Nashville, TN. We are grateful to R. Jfirgensen, Department of Medical Biometry, Philipps Universit~t Marburg, for his help in the statistical analyses. We thank Dr C. A. McArdle, Institut ffir Hormonund Fortpflanzungsforschung, Hamburg, Germany for reading the manuscript. This research was supported in part by the Deutsche Forschungsgemeinschaft (Or 52/4-1) and the Kempkes-Stiftung, Marburg.

REFERENCES I. Frawley L. S. and Neill J. D.: Biphasic effects of estrogen on gonadotropin-releasing hormone-induced luteinizinghormone release in monolayer cultures of rat and monkey pituitary cells. Endocrinology 114 (1984) 659-663. 2. Emons G., Ortmann O., Fingscheidt U., Ball P. and Knuppen R.: Short-term effects of oestradiol and 4hydroxyoestradiol on gonadotrophin-releasing hormone induced lutemlzing hormone secretion by rat pituitary cells in culture. Acta Endocr. Copenh. 111 (1986) 312-320. 3. Ortmann O., Emons G., Knuppen R. and Catt K. J.: Inhibitory actions of keoxifene on lutejni~,~3 hormone seorefion in pituitary $onadotrophs. Endocrinology 123 (1988) 962-968. 4. Emons G., Hoffmann H. G., Brack C., Ortmann O., Sturm R., Ball P. and Knuppen R.: Modulation of gonadotropin-relea~g hormone receptor concentration

in cultured female rat pituitary cells by estradiol treatment. J. Steroid Biochem. 31 (1988) 751-756. 5. Emons G., Frevert E. U., Ortmann O., Fingscheidt U., Sturm R., Kiesel L. and Knuppen R.: Studies on the subcellular mt.~.hAn~smsmediating the negative estradiol effect on GnRH-induced LH-relense by rat pituitary cells in culture. Acta Endocr. Copenh. 121 (1989) 350-360. 6. Ortmann O., Sturm R., Knuppen R. and Emons G.: Weak estrogenic activity of phenol red in the pituitary gouadotroph: re-evaluation of estrogen and antiestrogen effects. J. Steroid Biochem. 35 (1990) 17-22. 7. Tang L. K. L. and Spies H.: Effects of gonadal steroids on the basal and LRF-indueed gonadotropin secretion by cultures of rat pituitary. Endocrinology 96 (1975) 349-356. 8. Drouln J. D., Lagac~ L. and Labile F.: Estradioi induced increase of the LH-responsiveness to LHreleasing hormone (LHRH) in rat anterior pituitary cells in culture. Endocrinology 99 (1976) 1477-1481. 9. Emons G., Knuppen R., BaH P. and Cart K. J.: Biphasic modulation of pituitary sensitivity to GnRH by oestrogens: the effect of A- and D-ring substitution on LH.release i n cultured pituitary cells. Acta Endocr. Copenh. 107 (1984) 317-327. I0. Hsueh A. J. W., Erickson G. F. and Yen S. S. C.: The sensitizing effect of estrogens and catechol estrogen on cultured pituitary cells to luteinizing hormone releasing hormone: its antagonism by progestins. Endocrinology 104 (1979) 807-813. 11. Lagac~ L., Massicotte J. and Labrie F.: Acute stimulatory effects of progesterone on luteinizing hormone and follicle-stimulating hormone release in rat anterior pituitary cells in culture. Endocrinology 106 (1980) 684--689. 12. Drouln J. and Labrie F.: Interactions between 17/~estradiol and progesterone in the control of hiteinizing hormone and follicle stimulating hormone release in rat anterior pituitary cells in culture. Endocrinology 108 (1981) 52-57. 13. Ortmann O., Emons G., Knuppen R. and Catt K. J.: Inhibitory effects of the antiprogestin, RU 486, on progesterone actions and luteinizing hormone secretion in pituitary gonadotrophs. J. Steroid Biochem. 32 (1989) 291-297. 14. Ortmann O., Wiese H., Knuppen R. and Emons G.: Acute facilitory action of progesterone on gonadotropin secretion of perifused rat pituitary cells. Acta Endocr. Copenh. 121 (1989)426-434. 15. Krey L. C. and Kamel F.: Progesterone modulation of gonadotropin secretion by dispersed rat pituitary cells in culture. I. Basal and gonadotropin-relensing hormonestimulated luteinizing hormone release. Molec. Cell. Endocr. 68 (1990) 85-94. 16. Krey L. C., Kamel F. and MacLusky N. J.: Progesterone modulation of gonadotropin secretion by dispersed rat pituitary cells in culture. II. Intracellular metabofism and progestin receptors. Molec. Cell. Endocr. 68 (1990) 95-103. 17. Krey L. C. and Kamel F.: Progesterone modulation of gonadotropin secretion by dispersed rat pituitary cells in culture. III. A 23187, cAMP, phorbol ester and DiCSstimulated luteinizing hormone release. Molec. Cell. Endocr. 70 (1990) 21-29. 18. Turgeon J. L. and Warin8 D. W.: Rapid augmentation by progesterone of agoulst-stimulated luteinizing hormone secretion by cultured pituitary cells. Endocrinology 127 (1990) 773-780. 19. Ortmann O., H ~ A n u K., Knuppen R. and Emons G.: Modulatory actions of the new antiprogestins ZK 98.299 and ZK 98.734 and of RU 486 on luteiulzing hormone secretion and progesterone effects in pituitary gonadotrophs. J. Steroid Biochem. 36 (1990) 431-437.

Progesterone effects on GnRH receptors 20. Tang L., MarteUock A. C. and Horiuchi J. K.: Estradioi stimulation of LH respoMe to LHRH and LHRH Winding in pituitary cultures. Am. J. Physiol. 242 (1982) E292-E397. 21. Menon M., Peellel H. and Katta V.: Estradiol potentiation of gomdotropin~releming hormone l~ponmvenem in the anterior pituitary is mediated by an incTease in gonadotropine-rcieasing hormone receptors. Am. J. Obatet. Gynec. ISI (1985) 534-540. 22. G r q g D. W., Allen M. C. and Nett T. M.: Estradiulinduced increase in number of gonadotropin-reicasing hormone receptors in cultured ovine pituitary cells. BIO/. Re/wvd. 43 (1990) 1032-1036. 23. Loumaye E. and Catt K. J.: Agonist.induced regulation of pituitary receptors for gonadotropin-releusing hormone. J. Biol. Chem. 258 (1983) 12,002-12,009. 24. Clayton R. N., Shakespear R. A., Duncan J. A. and Marshall J. D. (with appendix by Munson P. J. and Rodbard D.): Radioiodinated nondegradable gonadotropin-releasing hormone analogs: new probes for the investigation of pituitary gonadotropin-releasing hormone receptors. Endocrinology 1043(1979) 1369-1376. 25. Emons G., Pahwa G. S., Brack C., Sturm R., Oberheuser F. and Knuppen R.: Gonadotropin releasing hormone binding sites in human epithelial ovarian carcinomata. Fur. J. Cancer Clin. Oncol. 25 (1989) 215-221. 26. Lonmaye E. and Catt K. J.: Homologous regulation of gonadotropin-releasing hormone receptors in cultured pituitary cells. Science 215 (1982) 983-085. 27. Bradford M. M.: A rapid and sensitive method for the quantitation of microsram quantities of protein utilizi~lg the principle of protein-dye binding. Analyt. BiOcfiem. 72 (1976) 248-254.

839

28. Clayton R. N., Solano A. R., Garcia-Vela A., Dufau M. L. and Catt K. J.: Regulation of pituitary receptors for gonadotropin releasing hormone during the rat estrous cycle. Endocrinology 107 (1980) 699-706. 29. Clayton R. N. and Catt K. J.: Regulation of pituitary gonadotropin-releasing hormone receptors by gonadal hormones. Endocrinology 108 (1981) 887-895. 30. De Lean A., Monson P. J. and Rodbard D.: Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay and physiological dose-response curves. Am. J. Physiol. 235 (1978) E97-EI02. 31. Sachs, L.: Angewandte Statistik (Applied Statistics). Springer, New York, 5th Edn (1978). 32. Marian J. and Conn P. M.: Subcellular localization of the receptor for gonadotropin-relcasing hormone in pituitary and ovarian tissue. Endocrinology 112 (1983) 104-112. 33. Naor Z.: Signal transduction mechanisms of Ca2+ mobilizing hormones. The case of gonadotropin releasing hormone. Endocrine. Rev. 11 (1990) 326-353. 34. Drouva S. V., Gorenue I., Laplante E., R~rat E., Enjalbert A. and Kordon C.: Estradiol modulates protein kinase C activity in the rat pituitary /n vivo and in vitro. Endocrinology 126 (1990) 536-544. 35. Ortmann O., Tilse B. and Emons G.: Estradiol and progesterone modulate TPA induced LH-secretion by cultured rat pituitary cells. 72nd Annual Meeting, Endocrine Society (1990) Abstr. 654. 36. Attardi B. and Happe H. K.: Modulation of the estradiol-induced luteinizinghormone surge by progesterone or antiestrogens: effects on pituitary gonadotropinreleasing hormone receptors. Endocrinology 119 (1986) 274-283.