Mokmdar and Cellular Endmrinology, 76 (1991) 115-123 0 1991 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/91/$03.50
115
MOLCEL02466
Inhibition of luteinizing hormone-human chorionic gonadotropin binding by retinoids in a Leydig cell line A. Lefbre ‘, C. Astraudo ‘v2and C. Finaz ‘v3 ’ Institut National de la SantP et de la Recherche MPdicae, Unit&293, 92120 Montrouge, France, ’ Minis&e de la Recherche et dc la Technologie, Paris, France, and 3 Centre National de la Recherche Scientifique, Paris, France (Received 13 August 1990; accepted 19 December 1990)
Key worak Luteinizing hormone-human chorionic gonadotropin receptor; Leydig cell; Retinoid
Treatment of IC9 mouse Leydig cells with 3 x 10 -’ M retinol (R) and retinoic acid (RA) resulted in 75% and 65% reduction of ‘2sI-labeled hCG binding respectively, when assayed at 35OC. This effect was dose-dependent and was first detected 12 h after initiation of treatment: it was maximal at 48 h for RA. R and RA had no significant effect on the rate of internalization and degradation of ‘251-hCGas measured by disappearance of acid-releasable (i.e. surface-bound) radioactivity from the cells and by the appearance of trichloracetic acid-soluble label in the medium. When exposed to increasing concentrations of hCG for 24 h, both retinoid-treated and control cells ‘down-regulated’ their gonadotropin receptors with the same dose-dependent pattern. The kinetics of reappearance of the receptors was similar for retinoid-treated and control cells, but for treated cells the maximal number of receptors reinitiated at 24 h ~‘r”,ver exceeded 40% of the values observed with control cells. Scatchard plot analysis confirmed a decrease in hCG receptor number from - 26,000 to - 6400 and - 3500 sites per cell after R and IU treatment. & values for ‘251-hCG binding were 2 x 1O”O M, 7.3 x 10-l’ M and 6.9 x 10’” M for control, R- and RA-treated cells respectively. On the basis of our data it is likely that retinoid-induced reduction in ‘251-hCG binding to K9 Leydig cells is due to decreased receptor synthesis.
Introduction Vitamin A (retinol, R) is required in the control of normal cell growth and differentiation (Roberts and Spom, 1984). Retinoic acid (RA), its physiological derivative, may be the active metabolite in most of these processes although retinol seems to
Address for correspondence(present address): A. Lef&re, INSERM, Unit6 187, 32 rue des Camets, 92140 Clamart, France.
be indispensable in vision and spermatogenesis (Porter et al., 1985). RA plays a role in specifying the anteroposterior axis of the central nervous system (Durston et al., 1989; Maden et al., 1989) and is thought to be the natural morphogen for the development and regeneration of limbs (Maden, 1982; Slack, 1987; Thaller and Eichele, 1987). This morphogenic property is also demonstrated by the enhancement of the expression of neural cell adhesion molecules (Husmann et al., 1989). Furthermore, IU has been shown to promote differentiation and to reduce tumorigenici ty in vitro (Sherman et al., 1981).
116
Within the cell, retinoids bind specific proteins two of which are present in many adult cell types: cellular retinol binding protein (CRW) and cellularretinoic acid binding protein (CRABP) (Chytil and Gng, 1984). CRABP probably participates in the establishment of the morphogenetic field (Doll&et al., 1989). Recent evidence indicates that RA acts through binding to nuclear receptors similar to those described for steroid hormones, thyroid hormones and vitamin D3 (Giguere et al., 1987; Petkovich et al., 1987; Doll&et al., 1989), and modulates transcription of specific genes (Chiocca et al., 1988; Bedo et al., 1989; Ng et al., 1989; Thompson and Rosner, 1989), particularly in the testis (Omori and Chytil, 1982). Several observations have suggested a specific role for R and RA in the male reproductive tract: animals maintained on retinol-depleted diets exhibit major lesions in the testis, including atrophy, loss of the germinal epithelium (Wolbach and Howe, 1925) and decrease in testosterone level (Appling and Chytil, 1981); IU given to retinoldeficient rats was shown to support testosterone biosynthesis (Appling and Chytil, 1981). Even though CRBP was found predominantly in the Sertoli cells, and CRABP in the germ cells (Porter et al., 198S), R and RA seem to play a role in the Leydig cell steroidogenesis. To examine this possibility, we have studied the effect of R and RA on luteinizing hormone-human chorionic gonadotropin (LH-hCG) binding to its receptor, the first step of the complex pathway that leads to an increase in steroid synthesis and secretion. Our study indicates that both R and RA negatively regulate the number of LH-hCG receptors and increase their affinity for the hormone, Materialsand methods Hormone and supplies
hCG (batch CR 123) was obtained from the NIH (kindly provided by R.E. Canfield) and iodinated as described elsewhere (Fraker and Speck, 1978). ‘251-hCGspecific activity was 30-50 pCi/pg. Stock solutions of 10m2 M trans-retinol and all-trans-retinoic acid (Sigma) were made up in ethanol and stored at - 20” C for up to 6 weeks. Low density lipoproteins (LDL) were prepared from fresh human plasma.
Cell culture The origin and properties of the K9 cells have
been described in a previous article (Finaz et al., 1987). Stock cultures were maintained in Dulbecco’s minimum essential medium (MEM) (Gibco) supplemented with 15% fetal calf serum (FCS) and subcultured every 7 days (split ratio: l/15). The cells were used between the 8th and the 16th serial subculture. Analysis of ‘251-hCG binding Cells were plated on day 0 (3 x 10’ tells/3.83
cm2, 2 X lo6 cells/25 cm2 for internalization experiments and 6 X 10’ tells/9.4 cm* for Scatchard plot experiments). On day 0 the medium was replaced and cells were cultured in Dulbecco’s MEM supplemented with 10 pg/ml insulin, 5 pg/ml transferrin, 5 pg/ml vitamin E and 10 pg/ml LDL (referred to as L medium) and treated as indicated in the figures. On day 3 or 4, the medium was removed from the cells, the cells were washed and incubated for 2 h with lo6 cpm of ‘251-hCG (corresponding to approximately 15 ng hCG) in L medium free of vitamin E and LDL (referred to as L’ medium). Non-specific binding was determined in the presence of 10 pg of unlabeled hormone and accounted for 2-10% of the total binding (depending on experimental conditions). At the end of the incubation period, the cells were placed on ice and rinsed 5 times with 1 ml of ice-cold phosphate-buffered saline (PBS)/ 0.2% bovine serum albumin (BSA) and then dissolved with 0.4% sodium deoxycholate in 0.5% NaOH. The radioactivity present was measured. Internalization experiments
On day 3, cells were placed on ice for 30 min, rinsed twice with 2 ml of ice-cold L’ medium and then incubated for 2 h at 0°C in L’ medium in the presence of 2 X lo6 cpm of *251-hCG (corresponding to approximately 30 ng hCG). After binding, excess ’ ‘I-hCG was removed by washing the cells 5 times with ice-cold PBS (free of Ca*+ and Mg?+)/O.l% BSA. The cells were then incubated at 35OC with 2 ml of L medium for different periods of time. At the times indicated in the figures, the medium was taken from the flasks, cells were washed with 1 ml of ice-cold PBS/O.l% BSA and the 3 ml were counted for “‘1 radioactiv-
117
ity. I ml aliquots of the media were then subjected to precipitation with 10~ trichloroacetic acid (TCA) and centrifuged at 600 × g. The radioactivity was measured in both the pellet and the supernatant to determine the extent of hormone degradation during incubations. The surface-bound hormone was removed from the cells by adding 2 ml of ice-cold 50 mM glycine/100 mM NaCI buffer pH 3. After 5 min incubation on ice, this solution was removed and cells were rinsed with 1 ml of ice-cold PBS/0.19~ BSA. The radioactivity present in the 3 ml was measured. Finally, to measure the amount of internalized hormone, cells were dissolved in 2 ml of 1 M NaOH, rinsed with an additional 1 ml of PBS/0.19~ BSA, both fractions were pooled and counted for 1251 radioactivity. When the effect of
• RA AR
100
i
80
60 m
40
20
0
120 -
' ~"
0
*
104
I
I0 -e
i
I
10.7
10-e
I
I0 "s M
Fig. 2. Dose-response curve for R- and RA-induced decrease in n251-hCG binding. K9 cells were treated for 72 h with increasing concentrations of retinol (4) or retinoic acid (®) as indicated. The binding of n251-hCG was then determined (see Materials and Methods). Results are the mean + SD of three different experiments for R and two for RA, each done in triplicate.
OHA AR
100
~
120
80
0 u
v
•~ 60
J 40
20
0
I
0
iB
I
12
I
I
24
48
•
R or RA was examined they were present throughout the experiments. The total amount of the hormone bound to the cells in each flask was calculated by adding together the amount of radioactivity in the medium, in the pH 3 eluate and in the NaOH extract. Non-specific binding was determined in parallel incubations in the presence of an excess (50 ~tg) of unlabeled hCG and was always subtracted from the total counts obtained after each treatment. For every experiment, prorein content of each dish was evaluated by the Lowry method (Lowry et al., 1951) and results were expressed in cpm per mg protein.
72
Hours Fig. 1. Time course of R- and RA-induced decrease in n2Sl-hCG binding. K9 cells were treated with 10 -6 M retinol (4) or retinoic acid (e) for various periods of time as indicated. The binding of t25I-hCG was then determined at the same time in all cultures as described in the Materials and Methods section. Results are mean + SD of triplicates from an experiment representative of three ones.
Results
Effect of R and RA on cell growth R and RA significantly affected neither cell growth nor morphology after a 72 h culture period at concentrations up t o 3.3 × 10 - 6 M (data not
shown).
118
Effects of R and RA on 12’I-hCG binding Exposure of K9 rek to R and RA for increas-
ing lengths of time resulted in a time-dependent reduction in ‘*‘I-hCG binding (Fig. 1). Half-maximal reduction occurred after 28 h of exposure to RA and the maximal effect, more than 80%, was observed at 48 h. A dose-response curve for this phenomenon is shown in Fig. 2. 3.3 X low6 M R and RA reduced the binding of ‘*‘I-hCG by 75% and 65% respectively. The concentrations of R or RA required to produce half-maximal reduction of i*‘I-hCG binding observed were 8 X lo-’ M and lo-’ M respectively. A slight but significant increase was observed between 10S9 M and 10e8 MI&A. There are several mechanisms to explain this heterologous ‘down-regulation’ of the hormone receptors: R and l&Amay increase the rate of receptor internalization, decrease the rate of appearance
of hCG receptors at the cell surface or affect the affinity of the receptor to the hormone. Internalization of receptor-hCG complexes
When K9 cells were exposed to ‘*‘I-hCG at O” C (about 40,000 cpm were initially bound to each flask), up to 75% of the bound hormone could be eluted from the cells with a low pH buffer immediately after the binding period, indicating that the bound hormone was located essentially on the cell surface, as shown in Fig. 3A and B. The cells were then incubated at 3S” C. Progressively the amount of acid-resistant (i.e. internalized) radioactivity increased from 28 to 70% for control, R- and R&treated cells; this was concomitant with a decrease of the amount of radioactivity confined to the cell membrane. By 3 h, about 70% of the initially bound radioactivity had disappeared from the cell surface. At this time,
101
’
6
8t
0
1
3 Hours
5
0
3
1
5
Hours
Fig. 3. Processing of ‘251-hCG by control and R- or RA-treated cells. K9 cells were treated for 48 h with 2 X lo-’ M of either retinof (A) or retinoic acid (B). Control cells received only L medium. The cells were labeled at O°C for 2 h and thereafter incubated in L medium* R or AR at 35OC. At specified times, radioactivity in the TCA-soluble fraction of the medium (0, #Q, on the cell surface (A, A) and inside the cells (0, ) was determined as indicated under Materials and Methods. Treated cells are represented by solid symbols ( ) and control cells by empty ones (0, A, 0). Each point represents the mean of two independent determinations which differed by less than 10% from each other. This is one representative experiment out of four identical ones.
119
about 857o of the radioactivity released were TCA soluble (data not shown). The presence of 2 × 10- ? M of either R or RA in the culture medium throughout the experiments did not significantly modify the rate of 25I-hCG internalization and degradation.
Homologous "down-regulation' of surface LH-hCG receptors We have previously shown that, as compared to normal Leydig cells (Lef6vre et al., 1985), the capacity of K9 cells to bind z251-hCG can be reduced by prolonged incubation of the cells with the hormone (Finaz et al., 1987). If R or RA modifies the sensitivity of the receptor to the hormone we would expect a modification of the pattern of hCG-induced 'down-regulation' of its own receptors. As shown in Fig. 4, after a 24 h incubation with increasing concentrations of hCG, the cells became resistant to a new challenge with the hormone. 100
8O
C•'N•%• I-I
As expected, 2 x l0 -? M R or RA induced a 30% reduction in the capacity of the cells to bind z251-hCG (Fig. 4,4). When results were expressed for each culture condition (with or without R or RA) as the percentage of radioactivity bound to cells not preincubated with hCG, it could be seen that the pattern of hCG-induced homologous desensitization was unaffected by either R or RA (Fig. 4B): the capacity to bind z25I-hCG was reduced to almost undetectable levels in both retihold-treated and control cells. Therefore R and RA do not seem to modify the sensitivity of the cells to hCG. To test the possible role of R or RA in decreasing the rate of appearance of LH-hCG receptors at the cell surface, the cells were totally depleted in LH-hCG recep~tors by receiving 3 × 10 - 9 M hCG for 24 h. Thereafter we studied the kinetics of reappearance of the receptors at the cell surface. At 24 h, more than 95% of the total receptors were recovered in control cells (Fig. 5/1). Reappearance 100 -
Control
Control ORA
I-1
A v
•=
60
&R
80
e~
60
ms
o
N 40
IN
20 r
2O
B 0
0 0
10-12
10-11 hCG (M)
10-1° 3.10~
i
0
./I--
I
10-12
I
10-"
I
I
lO-m 3.10-o
hCG (M)
Fig. 4. Homologous 'down-regulation' of LH-hCG receptors. K9 cells were treated for 48 h with 2 × 10- ~ M of either retinol (A) or retinoic acid (®). Control cells (r'q) received only L medium. During the last 24 h, treated and control cells re,ceived increasing concentrations of crude hCG as indicated. The binding of 1251-hCG was then determined. Results are expressed for all conditions (control, R- and RA-treated cells) as the percentage of radioactivity bound to control ceils which were not preincubated with hCG (A) or as the percentage of radioactivity bound to cells which were not preincubated with hCG for each condition (control, R- and RA-treated cells) (B). Results are the mean + SD of triplicate determinations of three independent incubations,
120
of the receptors began after 3 h and was linear up to 12 h where 85~ of the receptors were recovered as compared to non-desensitized cells. The kinetics were identical for R- and RA-treated cells and the maximal receptor number reinitiated at 24 h never exceeded 40~, of the values observed with control cells. When the results were expressed as the percentage of receptors present on the cell surface at 24 h for each condition (control, R- or RA-treated cells) it could be seen that the kinetics of reappearance of the receptors were identical for all culture conditions (Fig. 5B). It seems therefore that the decreasing binding of 12SI-hCG observed when cells were treated with R or RA is due to a change in the number of receptors; however, an alteration in the affinity for the ligand cannot as yet be totally eliminated.
Scatchard plot analysis In order to test the hypothesis that R or RA modifies gonadotropin receptor affinity, we measured the binding constants of ~'SI-hCG to control and to R- and RA-treated cells, as shown in Fig. 6. The number of binding sites was 26,384 for control cells and 6426 and 3527 for R- and RAtreated cells. Scatchard plot expression of the results show that the Kd values for nSl-hCG binding were 2 × 10 -~° M for control cells, 7.39 × 10 - n M and 6.96 × 10 - n M for R- and RAtreated cells respectively. These results confirm that R and RA reduce ~2SI-hCG binding by lowering the number of gonadotropin receptors per cell. Kd values were different between control and treated cells but were the same for R- and RA-
120
120
(3 Control
D Comrol • RA AR i NO coils
100
ORA
All • NOcons
100
80 .
80
i,°
60
/' ~" 40
20
20
0
3
6
9
12 Hours
16
24
0
3
6
9
12
16
24
Hours
Fig. 5. Kinetics of reappearance of cell-surface LH-hCG receptors. K9 cells were treated for 72 h with 3.3 × 10 -6 M retinol (a) Or 10- 6 M retinoic acid (e). Control cells (r~) received only L medium. At differem times, 3 x 10 •9 M hCO were added to treated and control cells for 24 h. Non-desensitized cells (m) did not receive hCO. The binding of 12Sl-hCG was then determined at the same time in all cultures. Results are expressed as the p-rcentage of radioactivity bound to control cells 24 h after hCG desensitization (A) and for each treatment as the percentage of radioactivity bound to cells (control, R and RA treated) 24 h after hCG desensitization (B). Results are the mean + SD of triplicate determinations from an experiment representative of three independent ones.
121
0.5
f l ¢m1,~
I-I Control QRA AR _
16
8
0.4
!
O8
1:3 CU
0.3
i
i
,i
I~ :
ToI~ h ~ (~l
I"1 "O
C: :0 O
oO
0.2 rl
0.1
0
1
2 Bound hCG (no)
3
4
Fig. 6. Scatchard analysis of binding of t2Sl-hCG to control (1:3) and retinol- (A) or retinoic acid- (o) treated cells. K9 cells were treated for 90 h with 5 × 10 -7 M R or 10 -6 M RA. Control
cells received only L medium. After washing, assay medium containing increasing concentrations of 12Sl-hCG was added (105 to 1.5 × 10e cpm correspondingapproximatelyto 1-15 ng hCG). For each treatment non-specific binding was determined in presence of an excess of crude hCG (20 t~g); it never exceeded 10~;. After incubationfor 2 h at 35oC, cell-associated radioactivitywas determined and expressed per dish (each dish contained an average of 2 × 106 cells at the end of the experiment). Results are the mean of duplicate determinations, repeated 2 times. Correlationcoefficientswere0.94, 0.89
and 0.87 for control, R- and RA-treatedcellsrespectively. treated cells (comparison of regression coefficients was performed by Student's t-test). This means that retinoid treatment of K9 cells increases the affinity of hCG for its receptor. Thus, decreased binding of nSl-hCG to retinoland retinoic acid-treated cells is due to a diminished number of gonadotropin receptors cow pied to a paradoxal increase in their affinity for hCG. Discussion
The results presented in this paper show that R and RA treatment of mouse Leydig K9 cells re-
suits in a dose-dependent decrease in the binding of 1251-hCG, confirming preliminary data from Mather et al. (1982). This decrease occurs for retinoid concentrations ranging from 10 -8 M to 3.3 × 10 -6 M. The concentrations necessary to elicit the maximal effect were similar to those necessary to obtain maximal increase of 1251-EGF (epidermal growth factor) (Jetten, 1980) or 1,25dihydroxyvitamin D3 binding (Petkovich et al., 1984), GH (growth hormone) secretion (Morita et al., 1988) as well as inhibition of alkaline phosphatase (Imai et al., 1988). Furthermore, they are in the same range as the circulating rates observed in rat (from 2 to 3 x 10 -4 M) (Omori and Chytil, 1982). The regulatory effect of retinoids on gonadotropin receptors is not related to their mitogenic properties since it is expressed under conditions where cell multiplication is not significantly affected. It was first detected 12 h after initiation of treatment. The time necessary to obtain maximal effect was shorter with RA, suggesting that RA may be the active metabolite in this process. The relatively prolonged lag period observed before decrease of gonadotropin binding may suggest that a more rapid direct intracellular response to retinoids exists in K9 cells. We explored successively the possible mechanisms that could contribute to the decrease of 12Sl-hCG binding. Internalization and degradation of the receptors could be involved in this process induced by retinoids. To address this point, we studied the binding of nSI-hCG at 0 ° C. At this temperature, retinoid-treated cells still exhibit a decrease in their ability to bind hCG as compared to control cells. However, neither R nor RA seems to affect the rate of internalization or degradation of lzsI-hCG. hCG-induced 'down-regulation' of its own receptors has been reported by different authors (Catt et al., 1979; Lef~vre et al., 1985). In both control and retinoid-treated cells, hCG 'down-regulated' its own receptors and furthermore the rate at which hCG binding capacity was restored after 'down-regulation' was the same for control and treated cells. But the level of recovery at 24 h never exceeded 40% of control for treated cells. Receptors that are ultimately located in the lysosomes are believed to be degraded rather than recycled in MA10 cells from which K9 cells are
122
derived (Ascoli, 1981). Furthermore, in these cells, LH and cyclic AMP exert their positive regulatory effect on LH receptors via the synthesis of new receptors (West et al., 1990). Therefore, it is likely that the synthesis of hCG receptors is the target of retinoids action; however, a role of R and RA on receptor recycling cannot be totally excluded. Scatchard plot analysis confirmed that the major action of retinoids on binding of 125I-hCG was the reduction of the number of available receptor sites, from 26,000 for control cells to 6400 and 3500 for R- and RA-treated cells respectively. However, this decrease was accompanied by an approximately 3-fold decrease in the dissociation constant for treated cells. This negative regulatory effect is to be compared to that of EGF and TGFfl (transforming growth factor fl): similarly, within 48 h, EGF and TGFfl reduce 1251-hCG binding to Leydig cells by reducing the number of gonadotropin receptors per cell, but they have no effect on the affinity of the receptor (Ascoli, 1981; Avallet et al., 1987). Moreover, it has been shown that RA 'up-regulates' the number of EGF receptor sites in different cell systems (Jetten, 1980, 1981; Oberg et al., 1988). One could speculate that R~, first augments EGF receptors and that the con.,equent e~hancernent of EGF binding to its own receptors would finally negatively regulate the nta.nher of hCG receptor sites. This is unexpected since K9 cells were cultivated for at least 72 h in a defined medium free of EGF prior to evaluation of hCG receptors. RA has been shown to regulate receptors to various effectors including growth factors (Jetten, 1980), vitamins (Petkovich et al., 1984), regulatory peptides (Waschek et al., 1989) or/~catecholamine as well (Imai, 1988). It is more often described as an enhancer of receptor synthesis: it increases the number of VIP (vasoactive intestinal peptide) receptors (Waschek et al., 1989) as well as that of 1,25-dihydroxy~" fin D 3 (Petkovich et al., 1984) or EGF (.Ie ..,, 1981). However, it significantly lowers //-catecholamine receptors in an osteosarcoma rat cell line, but without affecting the affinity of the receptors (lmai et al., 1988). The model proposed for VlP regulation is interesting as it represents the opposite of what can be observed with hCG receptors: RA enhances the number of binding sites while it
decreases the affinity of the receptors (Waschek et al., 1989). The action of RA on transcription is well documented while little is known of retinoid influence on receptor affinity. The gonadotropin receptor molecule comprises six potential glycosylation sites in the extracellular domain which may play a role in the binding of the glycoprotein hormone (Loosfelt et al., 1989; McFarland et al., 1989). Retinoids are known to increase sugar incorporation in glycoproteins (Sasak et al., 1980) and it may be postulated that they affect glycosylation of the binding sites to hCG, leading to a modification of the affinity of the receptor. But other hypotheses can be put forward to explain this decrease in the dissociation constant after retinoid treatment: they could (i) favor the formation of the receptor in either its monomer or oligomer form, one of which is the physiological active form of the molecule, or (ii) regulate the phosphorylation of the receptor in its intracellular domain where cellular control of receptor activity may occur (Sibley et al., 1988). All these potential targets ior retinoid action may also partly account for the decrease in the number of binding sites. The question arises by which molecular mechanism(s) retinoids primarily regulate gene expression. A number of recent results demonstrate that the effects of retinoids are due to transcriptional regulation, either by activation, documented by reports on EGF receptors (Thompson and Rosner, 1989), GH (Bedo et al., 1989), tissue transglutaminase (Chiocca et al., 1988), or by inhibition, as shown for the expression of Rex-l, a gene containing zinc finger motifs (Hosler et al., 1989). Therefore retinoids may directly regulate LHhCG receptor gene by inhibiting its transcription, or may activate the transcription of an inhibitory regulator of the gonadotropin receptor. The recent availability of the cDNA for LH receptor makes possible further investigations on mRNA analysis. Acknowledgments The authors wish to thank Dr. C. Bohuon in whose laboratory all labelings were carried out, Dr. H. Azouri for stimulating discussions, and Drs. S. Pellegrini and L. Dandolo for reading the manuscript.
123
References Appling, D.R. and Chytil, F. (1981) Endocrinology 108, 21202123. Ascoli, M. (1981) J. Biol. Chem. 256, 179-183. Ascoli, M. (1984) J. Cell Biol. 99, 1242-1250. Avallet, O., Vigier, M., Perrard-Sapori, M.H. and Saez, J.M. (1987) Biochem. Biophys. Res. Commun. 146, 575-581. Bedo, G., Santisteban, P. and Aranda, A. (1989) Nature 339, 231-234. CaR, K.J., Harwood, J.P., Aguilera, G. and Dufau, M.L. (1979) Nature 280, 109-116. Ciffocca, E.A., Davies, P.J.A. and Stein, J.P. (1988) J. Biol. Chem. 263, 11584-11589. Chytil, F. and Ong, D.E. (1984) in The Retinoids (Sporn, M.B., Roberts, A.B. and Goodman, D.S., eds.), pp. 89-123, Academic Press, New York. Dollt, P., Ruberte, E., Kastner, P., Petkovich, M., Stoner, C.M., Gudas, L.J. and Chambon, P. (1989) Nature 342, 702-705. Durston, AJ., Timmermans, J.P.M., Hage, W.J., Hendriks, H.F.J., De Vries, N.J., Heideveld, M. and Nieuwkoop, P.D. (1989) Nature 340, 140-144. Finaz, C., Leftvre, A. and Damphoffer, D. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5750-5753. Fraker, P.J. and Speck, J.C. (1978) Biochem. Biophys. Res. Commun. 80, 849-860. Giguere, V., Ong, E.S., Segui, P. and Evans, R.M. (1987) Nature 330, 624-629. Hosler, B.A., La Rosa, G.J., Gfippo, J.F. and Gudas, L.J. (1989) Mol. Cell. Biol. 9, 5623-5629. Husmann, M., Gtrgen, I., Weisgerber, C. and Bitter-Suermann, D. (1989) Dev. Biol. 136, 194-200. lmai, Y., Rodan, S.B. and Rodan, G. (1988) Endocrinology 122, 456-463. Jetten, A.M. (1980) Nature 284, 626-629. Jetten, A.M. (1981) J. Cell. Physiol. 110, 235-240. Lef6vre, A., Finaz, C., Berthelon, M.C. and Saez, J.M. (1985) Mol. Cell. Endocrinol. 40, 107-114. Loosfelt, H., Misrahi, M., Atger, M., Salesse, R., Hai-Lun Tiff, M.T.V., Jolivet, A., Guiochon-Mantei, A., Sar, S., Jallal, B., Gamier, J. and Miigrom, E. (1989) Science 245, 525-528. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275.
Maden, M. (1982) Nature 295, 672-675. Maden, M., Ong, D.E., Summerbell, D., Chytil, F. and Hirst, E.A. (1989) Dev. Biol. 135, 124-132. McFarland, K.C., Sprengel, R., Phillips, H.S., KiShler, M., Rosemblit, N., Nikolics, K., Scgaloff, D.L. and Seeburg, P.H. (1989) Science 245, 494-499. Mather, J.P., Saez, J.M., Haour, F. and Dray, F. (1982) Hormone-hormone interactions in the control of growth and function of Leydig cells in vitro, in Cold Spring Harbor Conferences on Cell Proliferation, Vol. 9, pp. 1117-1128, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Morita, S., Fernandez-Mejia, C. and Melmed, S. (1988) Endocrinology 124, 2052-2056. Ng, K.W., Manji, S.S., Young, M.F. and Findlay, D.M. (1989) Mol. Endocrinoi. 3, 2079-2085. Oberg, K.C., Mangelsdorf-Soderquist, A. and Carpenter, G. (1988) Mol. Endocrinol. 2, 959-965. Omori, M. and Chytil, F. (1982) J. Biol. Chem. 257, 1437014374. Petkovich, P.M., Heersche, J.N., Tinker, D.O. and Glenville, J. (1984) J. Biol. Chem. 253, 8274-8280. Petkovich, M., Brand, N.J., Krust, A. and Chambon, P. (1987) Nature 330, 444-450. Porter, S.B., Ong, D.E., Chytii, F. and Orgebin-Crist, M.C. (1985) J. Androl. 6, 197-212. Roberts, A.B. and Sporn, M.B. (1984) in The Retinoids (Sporn, M.B. and Goodman, A.B., eds.), pp. 209-286, Academic Press, Orlando, FL. Sasak, W., De Luca, L.M., Dion, L.D. and Siiverman-Jones, C.S. (1980) Cancer Res. 40, 1944-1949. Sherman, M.I., Matthaei, K.I. and Schindler, J. (1981) Ann. N.Y. Acad. Sci. 359, 192-199. Sibley, D.R., Benovic, J.L., Caron, M.G. and Lefkowitz, R.J. (1988) Endocr. Rev. 9, 38-42. Slack, J.M.W. (1987) Nature 327, 553-554. Thailer, C. and Eichele, G. (1987) Nature 327, 625-628. Thompson, K.L. and Rosner, R.M. (1989) J. Biol. Chem. 264, 3230-3234. Waschek, J.A., Muller, J.M., Duan, D.S. and Sadte, W. (1989) FEBS Lett. 250, 611-614. West, A.P., Rose, M.P. and Cooke, B.A. (1990) Biochem. J. 270, 499-503. Wolbach, S.B. and Howe, P.R. (1925) J. Exp. Med. 43, 753-777.
115
MOLCEL02466
Inhibition of luteinizing hormone-human chorionic gonadotropin binding by retinoids in a Leydig cell line A. Lefbre ‘, C. Astraudo ‘v2and C. Finaz ‘v3 ’ Institut National de la SantP et de la Recherche MPdicae, Unit&293, 92120 Montrouge, France, ’ Minis&e de la Recherche et dc la Technologie, Paris, France, and 3 Centre National de la Recherche Scientifique, Paris, France (Received 13 August 1990; accepted 19 December 1990)
Key worak Luteinizing hormone-human chorionic gonadotropin receptor; Leydig cell; Retinoid
Treatment of IC9 mouse Leydig cells with 3 x 10 -’ M retinol (R) and retinoic acid (RA) resulted in 75% and 65% reduction of ‘2sI-labeled hCG binding respectively, when assayed at 35OC. This effect was dose-dependent and was first detected 12 h after initiation of treatment: it was maximal at 48 h for RA. R and RA had no significant effect on the rate of internalization and degradation of ‘251-hCGas measured by disappearance of acid-releasable (i.e. surface-bound) radioactivity from the cells and by the appearance of trichloracetic acid-soluble label in the medium. When exposed to increasing concentrations of hCG for 24 h, both retinoid-treated and control cells ‘down-regulated’ their gonadotropin receptors with the same dose-dependent pattern. The kinetics of reappearance of the receptors was similar for retinoid-treated and control cells, but for treated cells the maximal number of receptors reinitiated at 24 h ~‘r”,ver exceeded 40% of the values observed with control cells. Scatchard plot analysis confirmed a decrease in hCG receptor number from - 26,000 to - 6400 and - 3500 sites per cell after R and IU treatment. & values for ‘251-hCG binding were 2 x 1O”O M, 7.3 x 10-l’ M and 6.9 x 10’” M for control, R- and RA-treated cells respectively. On the basis of our data it is likely that retinoid-induced reduction in ‘251-hCG binding to K9 Leydig cells is due to decreased receptor synthesis.
Introduction Vitamin A (retinol, R) is required in the control of normal cell growth and differentiation (Roberts and Spom, 1984). Retinoic acid (RA), its physiological derivative, may be the active metabolite in most of these processes although retinol seems to
Address for correspondence(present address): A. Lef&re, INSERM, Unit6 187, 32 rue des Camets, 92140 Clamart, France.
be indispensable in vision and spermatogenesis (Porter et al., 1985). RA plays a role in specifying the anteroposterior axis of the central nervous system (Durston et al., 1989; Maden et al., 1989) and is thought to be the natural morphogen for the development and regeneration of limbs (Maden, 1982; Slack, 1987; Thaller and Eichele, 1987). This morphogenic property is also demonstrated by the enhancement of the expression of neural cell adhesion molecules (Husmann et al., 1989). Furthermore, IU has been shown to promote differentiation and to reduce tumorigenici ty in vitro (Sherman et al., 1981).
116
Within the cell, retinoids bind specific proteins two of which are present in many adult cell types: cellular retinol binding protein (CRW) and cellularretinoic acid binding protein (CRABP) (Chytil and Gng, 1984). CRABP probably participates in the establishment of the morphogenetic field (Doll&et al., 1989). Recent evidence indicates that RA acts through binding to nuclear receptors similar to those described for steroid hormones, thyroid hormones and vitamin D3 (Giguere et al., 1987; Petkovich et al., 1987; Doll&et al., 1989), and modulates transcription of specific genes (Chiocca et al., 1988; Bedo et al., 1989; Ng et al., 1989; Thompson and Rosner, 1989), particularly in the testis (Omori and Chytil, 1982). Several observations have suggested a specific role for R and RA in the male reproductive tract: animals maintained on retinol-depleted diets exhibit major lesions in the testis, including atrophy, loss of the germinal epithelium (Wolbach and Howe, 1925) and decrease in testosterone level (Appling and Chytil, 1981); IU given to retinoldeficient rats was shown to support testosterone biosynthesis (Appling and Chytil, 1981). Even though CRBP was found predominantly in the Sertoli cells, and CRABP in the germ cells (Porter et al., 198S), R and RA seem to play a role in the Leydig cell steroidogenesis. To examine this possibility, we have studied the effect of R and RA on luteinizing hormone-human chorionic gonadotropin (LH-hCG) binding to its receptor, the first step of the complex pathway that leads to an increase in steroid synthesis and secretion. Our study indicates that both R and RA negatively regulate the number of LH-hCG receptors and increase their affinity for the hormone, Materialsand methods Hormone and supplies
hCG (batch CR 123) was obtained from the NIH (kindly provided by R.E. Canfield) and iodinated as described elsewhere (Fraker and Speck, 1978). ‘251-hCGspecific activity was 30-50 pCi/pg. Stock solutions of 10m2 M trans-retinol and all-trans-retinoic acid (Sigma) were made up in ethanol and stored at - 20” C for up to 6 weeks. Low density lipoproteins (LDL) were prepared from fresh human plasma.
Cell culture The origin and properties of the K9 cells have
been described in a previous article (Finaz et al., 1987). Stock cultures were maintained in Dulbecco’s minimum essential medium (MEM) (Gibco) supplemented with 15% fetal calf serum (FCS) and subcultured every 7 days (split ratio: l/15). The cells were used between the 8th and the 16th serial subculture. Analysis of ‘251-hCG binding Cells were plated on day 0 (3 x 10’ tells/3.83
cm2, 2 X lo6 cells/25 cm2 for internalization experiments and 6 X 10’ tells/9.4 cm* for Scatchard plot experiments). On day 0 the medium was replaced and cells were cultured in Dulbecco’s MEM supplemented with 10 pg/ml insulin, 5 pg/ml transferrin, 5 pg/ml vitamin E and 10 pg/ml LDL (referred to as L medium) and treated as indicated in the figures. On day 3 or 4, the medium was removed from the cells, the cells were washed and incubated for 2 h with lo6 cpm of ‘251-hCG (corresponding to approximately 15 ng hCG) in L medium free of vitamin E and LDL (referred to as L’ medium). Non-specific binding was determined in the presence of 10 pg of unlabeled hormone and accounted for 2-10% of the total binding (depending on experimental conditions). At the end of the incubation period, the cells were placed on ice and rinsed 5 times with 1 ml of ice-cold phosphate-buffered saline (PBS)/ 0.2% bovine serum albumin (BSA) and then dissolved with 0.4% sodium deoxycholate in 0.5% NaOH. The radioactivity present was measured. Internalization experiments
On day 3, cells were placed on ice for 30 min, rinsed twice with 2 ml of ice-cold L’ medium and then incubated for 2 h at 0°C in L’ medium in the presence of 2 X lo6 cpm of *251-hCG (corresponding to approximately 30 ng hCG). After binding, excess ’ ‘I-hCG was removed by washing the cells 5 times with ice-cold PBS (free of Ca*+ and Mg?+)/O.l% BSA. The cells were then incubated at 35OC with 2 ml of L medium for different periods of time. At the times indicated in the figures, the medium was taken from the flasks, cells were washed with 1 ml of ice-cold PBS/O.l% BSA and the 3 ml were counted for “‘1 radioactiv-
117
ity. I ml aliquots of the media were then subjected to precipitation with 10~ trichloroacetic acid (TCA) and centrifuged at 600 × g. The radioactivity was measured in both the pellet and the supernatant to determine the extent of hormone degradation during incubations. The surface-bound hormone was removed from the cells by adding 2 ml of ice-cold 50 mM glycine/100 mM NaCI buffer pH 3. After 5 min incubation on ice, this solution was removed and cells were rinsed with 1 ml of ice-cold PBS/0.19~ BSA. The radioactivity present in the 3 ml was measured. Finally, to measure the amount of internalized hormone, cells were dissolved in 2 ml of 1 M NaOH, rinsed with an additional 1 ml of PBS/0.19~ BSA, both fractions were pooled and counted for 1251 radioactivity. When the effect of
• RA AR
100
i
80
60 m
40
20
0
120 -
' ~"
0
*
104
I
I0 -e
i
I
10.7
10-e
I
I0 "s M
Fig. 2. Dose-response curve for R- and RA-induced decrease in n251-hCG binding. K9 cells were treated for 72 h with increasing concentrations of retinol (4) or retinoic acid (®) as indicated. The binding of n251-hCG was then determined (see Materials and Methods). Results are the mean + SD of three different experiments for R and two for RA, each done in triplicate.
OHA AR
100
~
120
80
0 u
v
•~ 60
J 40
20
0
I
0
iB
I
12
I
I
24
48
•
R or RA was examined they were present throughout the experiments. The total amount of the hormone bound to the cells in each flask was calculated by adding together the amount of radioactivity in the medium, in the pH 3 eluate and in the NaOH extract. Non-specific binding was determined in parallel incubations in the presence of an excess (50 ~tg) of unlabeled hCG and was always subtracted from the total counts obtained after each treatment. For every experiment, prorein content of each dish was evaluated by the Lowry method (Lowry et al., 1951) and results were expressed in cpm per mg protein.
72
Hours Fig. 1. Time course of R- and RA-induced decrease in n2Sl-hCG binding. K9 cells were treated with 10 -6 M retinol (4) or retinoic acid (e) for various periods of time as indicated. The binding of t25I-hCG was then determined at the same time in all cultures as described in the Materials and Methods section. Results are mean + SD of triplicates from an experiment representative of three ones.
Results
Effect of R and RA on cell growth R and RA significantly affected neither cell growth nor morphology after a 72 h culture period at concentrations up t o 3.3 × 10 - 6 M (data not
shown).
118
Effects of R and RA on 12’I-hCG binding Exposure of K9 rek to R and RA for increas-
ing lengths of time resulted in a time-dependent reduction in ‘*‘I-hCG binding (Fig. 1). Half-maximal reduction occurred after 28 h of exposure to RA and the maximal effect, more than 80%, was observed at 48 h. A dose-response curve for this phenomenon is shown in Fig. 2. 3.3 X low6 M R and RA reduced the binding of ‘*‘I-hCG by 75% and 65% respectively. The concentrations of R or RA required to produce half-maximal reduction of i*‘I-hCG binding observed were 8 X lo-’ M and lo-’ M respectively. A slight but significant increase was observed between 10S9 M and 10e8 MI&A. There are several mechanisms to explain this heterologous ‘down-regulation’ of the hormone receptors: R and l&Amay increase the rate of receptor internalization, decrease the rate of appearance
of hCG receptors at the cell surface or affect the affinity of the receptor to the hormone. Internalization of receptor-hCG complexes
When K9 cells were exposed to ‘*‘I-hCG at O” C (about 40,000 cpm were initially bound to each flask), up to 75% of the bound hormone could be eluted from the cells with a low pH buffer immediately after the binding period, indicating that the bound hormone was located essentially on the cell surface, as shown in Fig. 3A and B. The cells were then incubated at 3S” C. Progressively the amount of acid-resistant (i.e. internalized) radioactivity increased from 28 to 70% for control, R- and R&treated cells; this was concomitant with a decrease of the amount of radioactivity confined to the cell membrane. By 3 h, about 70% of the initially bound radioactivity had disappeared from the cell surface. At this time,
101
’
6
8t
0
1
3 Hours
5
0
3
1
5
Hours
Fig. 3. Processing of ‘251-hCG by control and R- or RA-treated cells. K9 cells were treated for 48 h with 2 X lo-’ M of either retinof (A) or retinoic acid (B). Control cells received only L medium. The cells were labeled at O°C for 2 h and thereafter incubated in L medium* R or AR at 35OC. At specified times, radioactivity in the TCA-soluble fraction of the medium (0, #Q, on the cell surface (A, A) and inside the cells (0, ) was determined as indicated under Materials and Methods. Treated cells are represented by solid symbols ( ) and control cells by empty ones (0, A, 0). Each point represents the mean of two independent determinations which differed by less than 10% from each other. This is one representative experiment out of four identical ones.
119
about 857o of the radioactivity released were TCA soluble (data not shown). The presence of 2 × 10- ? M of either R or RA in the culture medium throughout the experiments did not significantly modify the rate of 25I-hCG internalization and degradation.
Homologous "down-regulation' of surface LH-hCG receptors We have previously shown that, as compared to normal Leydig cells (Lef6vre et al., 1985), the capacity of K9 cells to bind z251-hCG can be reduced by prolonged incubation of the cells with the hormone (Finaz et al., 1987). If R or RA modifies the sensitivity of the receptor to the hormone we would expect a modification of the pattern of hCG-induced 'down-regulation' of its own receptors. As shown in Fig. 4, after a 24 h incubation with increasing concentrations of hCG, the cells became resistant to a new challenge with the hormone. 100
8O
C•'N•%• I-I
As expected, 2 x l0 -? M R or RA induced a 30% reduction in the capacity of the cells to bind z251-hCG (Fig. 4,4). When results were expressed for each culture condition (with or without R or RA) as the percentage of radioactivity bound to cells not preincubated with hCG, it could be seen that the pattern of hCG-induced homologous desensitization was unaffected by either R or RA (Fig. 4B): the capacity to bind z25I-hCG was reduced to almost undetectable levels in both retihold-treated and control cells. Therefore R and RA do not seem to modify the sensitivity of the cells to hCG. To test the possible role of R or RA in decreasing the rate of appearance of LH-hCG receptors at the cell surface, the cells were totally depleted in LH-hCG recep~tors by receiving 3 × 10 - 9 M hCG for 24 h. Thereafter we studied the kinetics of reappearance of the receptors at the cell surface. At 24 h, more than 95% of the total receptors were recovered in control cells (Fig. 5/1). Reappearance 100 -
Control
Control ORA
I-1
A v
•=
60
&R
80
e~
60
ms
o
N 40
IN
20 r
2O
B 0
0 0
10-12
10-11 hCG (M)
10-1° 3.10~
i
0
./I--
I
10-12
I
10-"
I
I
lO-m 3.10-o
hCG (M)
Fig. 4. Homologous 'down-regulation' of LH-hCG receptors. K9 cells were treated for 48 h with 2 × 10- ~ M of either retinol (A) or retinoic acid (®). Control cells (r'q) received only L medium. During the last 24 h, treated and control cells re,ceived increasing concentrations of crude hCG as indicated. The binding of 1251-hCG was then determined. Results are expressed for all conditions (control, R- and RA-treated cells) as the percentage of radioactivity bound to control ceils which were not preincubated with hCG (A) or as the percentage of radioactivity bound to cells which were not preincubated with hCG for each condition (control, R- and RA-treated cells) (B). Results are the mean + SD of triplicate determinations of three independent incubations,
120
of the receptors began after 3 h and was linear up to 12 h where 85~ of the receptors were recovered as compared to non-desensitized cells. The kinetics were identical for R- and RA-treated cells and the maximal receptor number reinitiated at 24 h never exceeded 40~, of the values observed with control cells. When the results were expressed as the percentage of receptors present on the cell surface at 24 h for each condition (control, R- or RA-treated cells) it could be seen that the kinetics of reappearance of the receptors were identical for all culture conditions (Fig. 5B). It seems therefore that the decreasing binding of 12SI-hCG observed when cells were treated with R or RA is due to a change in the number of receptors; however, an alteration in the affinity for the ligand cannot as yet be totally eliminated.
Scatchard plot analysis In order to test the hypothesis that R or RA modifies gonadotropin receptor affinity, we measured the binding constants of ~'SI-hCG to control and to R- and RA-treated cells, as shown in Fig. 6. The number of binding sites was 26,384 for control cells and 6426 and 3527 for R- and RAtreated cells. Scatchard plot expression of the results show that the Kd values for nSl-hCG binding were 2 × 10 -~° M for control cells, 7.39 × 10 - n M and 6.96 × 10 - n M for R- and RAtreated cells respectively. These results confirm that R and RA reduce ~2SI-hCG binding by lowering the number of gonadotropin receptors per cell. Kd values were different between control and treated cells but were the same for R- and RA-
120
120
(3 Control
D Comrol • RA AR i NO coils
100
ORA
All • NOcons
100
80 .
80
i,°
60
/' ~" 40
20
20
0
3
6
9
12 Hours
16
24
0
3
6
9
12
16
24
Hours
Fig. 5. Kinetics of reappearance of cell-surface LH-hCG receptors. K9 cells were treated for 72 h with 3.3 × 10 -6 M retinol (a) Or 10- 6 M retinoic acid (e). Control cells (r~) received only L medium. At differem times, 3 x 10 •9 M hCO were added to treated and control cells for 24 h. Non-desensitized cells (m) did not receive hCO. The binding of 12Sl-hCG was then determined at the same time in all cultures. Results are expressed as the p-rcentage of radioactivity bound to control cells 24 h after hCG desensitization (A) and for each treatment as the percentage of radioactivity bound to cells (control, R and RA treated) 24 h after hCG desensitization (B). Results are the mean + SD of triplicate determinations from an experiment representative of three independent ones.
121
0.5
f l ¢m1,~
I-I Control QRA AR _
16
8
0.4
!
O8
1:3 CU
0.3
i
i
,i
I~ :
ToI~ h ~ (~l
I"1 "O
C: :0 O
oO
0.2 rl
0.1
0
1
2 Bound hCG (no)
3
4
Fig. 6. Scatchard analysis of binding of t2Sl-hCG to control (1:3) and retinol- (A) or retinoic acid- (o) treated cells. K9 cells were treated for 90 h with 5 × 10 -7 M R or 10 -6 M RA. Control
cells received only L medium. After washing, assay medium containing increasing concentrations of 12Sl-hCG was added (105 to 1.5 × 10e cpm correspondingapproximatelyto 1-15 ng hCG). For each treatment non-specific binding was determined in presence of an excess of crude hCG (20 t~g); it never exceeded 10~;. After incubationfor 2 h at 35oC, cell-associated radioactivitywas determined and expressed per dish (each dish contained an average of 2 × 106 cells at the end of the experiment). Results are the mean of duplicate determinations, repeated 2 times. Correlationcoefficientswere0.94, 0.89
and 0.87 for control, R- and RA-treatedcellsrespectively. treated cells (comparison of regression coefficients was performed by Student's t-test). This means that retinoid treatment of K9 cells increases the affinity of hCG for its receptor. Thus, decreased binding of nSl-hCG to retinoland retinoic acid-treated cells is due to a diminished number of gonadotropin receptors cow pied to a paradoxal increase in their affinity for hCG. Discussion
The results presented in this paper show that R and RA treatment of mouse Leydig K9 cells re-
suits in a dose-dependent decrease in the binding of 1251-hCG, confirming preliminary data from Mather et al. (1982). This decrease occurs for retinoid concentrations ranging from 10 -8 M to 3.3 × 10 -6 M. The concentrations necessary to elicit the maximal effect were similar to those necessary to obtain maximal increase of 1251-EGF (epidermal growth factor) (Jetten, 1980) or 1,25dihydroxyvitamin D3 binding (Petkovich et al., 1984), GH (growth hormone) secretion (Morita et al., 1988) as well as inhibition of alkaline phosphatase (Imai et al., 1988). Furthermore, they are in the same range as the circulating rates observed in rat (from 2 to 3 x 10 -4 M) (Omori and Chytil, 1982). The regulatory effect of retinoids on gonadotropin receptors is not related to their mitogenic properties since it is expressed under conditions where cell multiplication is not significantly affected. It was first detected 12 h after initiation of treatment. The time necessary to obtain maximal effect was shorter with RA, suggesting that RA may be the active metabolite in this process. The relatively prolonged lag period observed before decrease of gonadotropin binding may suggest that a more rapid direct intracellular response to retinoids exists in K9 cells. We explored successively the possible mechanisms that could contribute to the decrease of 12Sl-hCG binding. Internalization and degradation of the receptors could be involved in this process induced by retinoids. To address this point, we studied the binding of nSI-hCG at 0 ° C. At this temperature, retinoid-treated cells still exhibit a decrease in their ability to bind hCG as compared to control cells. However, neither R nor RA seems to affect the rate of internalization or degradation of lzsI-hCG. hCG-induced 'down-regulation' of its own receptors has been reported by different authors (Catt et al., 1979; Lef~vre et al., 1985). In both control and retinoid-treated cells, hCG 'down-regulated' its own receptors and furthermore the rate at which hCG binding capacity was restored after 'down-regulation' was the same for control and treated cells. But the level of recovery at 24 h never exceeded 40% of control for treated cells. Receptors that are ultimately located in the lysosomes are believed to be degraded rather than recycled in MA10 cells from which K9 cells are
122
derived (Ascoli, 1981). Furthermore, in these cells, LH and cyclic AMP exert their positive regulatory effect on LH receptors via the synthesis of new receptors (West et al., 1990). Therefore, it is likely that the synthesis of hCG receptors is the target of retinoids action; however, a role of R and RA on receptor recycling cannot be totally excluded. Scatchard plot analysis confirmed that the major action of retinoids on binding of 125I-hCG was the reduction of the number of available receptor sites, from 26,000 for control cells to 6400 and 3500 for R- and RA-treated cells respectively. However, this decrease was accompanied by an approximately 3-fold decrease in the dissociation constant for treated cells. This negative regulatory effect is to be compared to that of EGF and TGFfl (transforming growth factor fl): similarly, within 48 h, EGF and TGFfl reduce 1251-hCG binding to Leydig cells by reducing the number of gonadotropin receptors per cell, but they have no effect on the affinity of the receptor (Ascoli, 1981; Avallet et al., 1987). Moreover, it has been shown that RA 'up-regulates' the number of EGF receptor sites in different cell systems (Jetten, 1980, 1981; Oberg et al., 1988). One could speculate that R~, first augments EGF receptors and that the con.,equent e~hancernent of EGF binding to its own receptors would finally negatively regulate the nta.nher of hCG receptor sites. This is unexpected since K9 cells were cultivated for at least 72 h in a defined medium free of EGF prior to evaluation of hCG receptors. RA has been shown to regulate receptors to various effectors including growth factors (Jetten, 1980), vitamins (Petkovich et al., 1984), regulatory peptides (Waschek et al., 1989) or/~catecholamine as well (Imai, 1988). It is more often described as an enhancer of receptor synthesis: it increases the number of VIP (vasoactive intestinal peptide) receptors (Waschek et al., 1989) as well as that of 1,25-dihydroxy~" fin D 3 (Petkovich et al., 1984) or EGF (.Ie ..,, 1981). However, it significantly lowers //-catecholamine receptors in an osteosarcoma rat cell line, but without affecting the affinity of the receptors (lmai et al., 1988). The model proposed for VlP regulation is interesting as it represents the opposite of what can be observed with hCG receptors: RA enhances the number of binding sites while it
decreases the affinity of the receptors (Waschek et al., 1989). The action of RA on transcription is well documented while little is known of retinoid influence on receptor affinity. The gonadotropin receptor molecule comprises six potential glycosylation sites in the extracellular domain which may play a role in the binding of the glycoprotein hormone (Loosfelt et al., 1989; McFarland et al., 1989). Retinoids are known to increase sugar incorporation in glycoproteins (Sasak et al., 1980) and it may be postulated that they affect glycosylation of the binding sites to hCG, leading to a modification of the affinity of the receptor. But other hypotheses can be put forward to explain this decrease in the dissociation constant after retinoid treatment: they could (i) favor the formation of the receptor in either its monomer or oligomer form, one of which is the physiological active form of the molecule, or (ii) regulate the phosphorylation of the receptor in its intracellular domain where cellular control of receptor activity may occur (Sibley et al., 1988). All these potential targets ior retinoid action may also partly account for the decrease in the number of binding sites. The question arises by which molecular mechanism(s) retinoids primarily regulate gene expression. A number of recent results demonstrate that the effects of retinoids are due to transcriptional regulation, either by activation, documented by reports on EGF receptors (Thompson and Rosner, 1989), GH (Bedo et al., 1989), tissue transglutaminase (Chiocca et al., 1988), or by inhibition, as shown for the expression of Rex-l, a gene containing zinc finger motifs (Hosler et al., 1989). Therefore retinoids may directly regulate LHhCG receptor gene by inhibiting its transcription, or may activate the transcription of an inhibitory regulator of the gonadotropin receptor. The recent availability of the cDNA for LH receptor makes possible further investigations on mRNA analysis. Acknowledgments The authors wish to thank Dr. C. Bohuon in whose laboratory all labelings were carried out, Dr. H. Azouri for stimulating discussions, and Drs. S. Pellegrini and L. Dandolo for reading the manuscript.
123
References Appling, D.R. and Chytil, F. (1981) Endocrinology 108, 21202123. Ascoli, M. (1981) J. Biol. Chem. 256, 179-183. Ascoli, M. (1984) J. Cell Biol. 99, 1242-1250. Avallet, O., Vigier, M., Perrard-Sapori, M.H. and Saez, J.M. (1987) Biochem. Biophys. Res. Commun. 146, 575-581. Bedo, G., Santisteban, P. and Aranda, A. (1989) Nature 339, 231-234. CaR, K.J., Harwood, J.P., Aguilera, G. and Dufau, M.L. (1979) Nature 280, 109-116. Ciffocca, E.A., Davies, P.J.A. and Stein, J.P. (1988) J. Biol. Chem. 263, 11584-11589. Chytil, F. and Ong, D.E. (1984) in The Retinoids (Sporn, M.B., Roberts, A.B. and Goodman, D.S., eds.), pp. 89-123, Academic Press, New York. Dollt, P., Ruberte, E., Kastner, P., Petkovich, M., Stoner, C.M., Gudas, L.J. and Chambon, P. (1989) Nature 342, 702-705. Durston, AJ., Timmermans, J.P.M., Hage, W.J., Hendriks, H.F.J., De Vries, N.J., Heideveld, M. and Nieuwkoop, P.D. (1989) Nature 340, 140-144. Finaz, C., Leftvre, A. and Damphoffer, D. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5750-5753. Fraker, P.J. and Speck, J.C. (1978) Biochem. Biophys. Res. Commun. 80, 849-860. Giguere, V., Ong, E.S., Segui, P. and Evans, R.M. (1987) Nature 330, 624-629. Hosler, B.A., La Rosa, G.J., Gfippo, J.F. and Gudas, L.J. (1989) Mol. Cell. Biol. 9, 5623-5629. Husmann, M., Gtrgen, I., Weisgerber, C. and Bitter-Suermann, D. (1989) Dev. Biol. 136, 194-200. lmai, Y., Rodan, S.B. and Rodan, G. (1988) Endocrinology 122, 456-463. Jetten, A.M. (1980) Nature 284, 626-629. Jetten, A.M. (1981) J. Cell. Physiol. 110, 235-240. Lef6vre, A., Finaz, C., Berthelon, M.C. and Saez, J.M. (1985) Mol. Cell. Endocrinol. 40, 107-114. Loosfelt, H., Misrahi, M., Atger, M., Salesse, R., Hai-Lun Tiff, M.T.V., Jolivet, A., Guiochon-Mantei, A., Sar, S., Jallal, B., Gamier, J. and Miigrom, E. (1989) Science 245, 525-528. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275.
Maden, M. (1982) Nature 295, 672-675. Maden, M., Ong, D.E., Summerbell, D., Chytil, F. and Hirst, E.A. (1989) Dev. Biol. 135, 124-132. McFarland, K.C., Sprengel, R., Phillips, H.S., KiShler, M., Rosemblit, N., Nikolics, K., Scgaloff, D.L. and Seeburg, P.H. (1989) Science 245, 494-499. Mather, J.P., Saez, J.M., Haour, F. and Dray, F. (1982) Hormone-hormone interactions in the control of growth and function of Leydig cells in vitro, in Cold Spring Harbor Conferences on Cell Proliferation, Vol. 9, pp. 1117-1128, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Morita, S., Fernandez-Mejia, C. and Melmed, S. (1988) Endocrinology 124, 2052-2056. Ng, K.W., Manji, S.S., Young, M.F. and Findlay, D.M. (1989) Mol. Endocrinoi. 3, 2079-2085. Oberg, K.C., Mangelsdorf-Soderquist, A. and Carpenter, G. (1988) Mol. Endocrinol. 2, 959-965. Omori, M. and Chytil, F. (1982) J. Biol. Chem. 257, 1437014374. Petkovich, P.M., Heersche, J.N., Tinker, D.O. and Glenville, J. (1984) J. Biol. Chem. 253, 8274-8280. Petkovich, M., Brand, N.J., Krust, A. and Chambon, P. (1987) Nature 330, 444-450. Porter, S.B., Ong, D.E., Chytii, F. and Orgebin-Crist, M.C. (1985) J. Androl. 6, 197-212. Roberts, A.B. and Sporn, M.B. (1984) in The Retinoids (Sporn, M.B. and Goodman, A.B., eds.), pp. 209-286, Academic Press, Orlando, FL. Sasak, W., De Luca, L.M., Dion, L.D. and Siiverman-Jones, C.S. (1980) Cancer Res. 40, 1944-1949. Sherman, M.I., Matthaei, K.I. and Schindler, J. (1981) Ann. N.Y. Acad. Sci. 359, 192-199. Sibley, D.R., Benovic, J.L., Caron, M.G. and Lefkowitz, R.J. (1988) Endocr. Rev. 9, 38-42. Slack, J.M.W. (1987) Nature 327, 553-554. Thailer, C. and Eichele, G. (1987) Nature 327, 625-628. Thompson, K.L. and Rosner, R.M. (1989) J. Biol. Chem. 264, 3230-3234. Waschek, J.A., Muller, J.M., Duan, D.S. and Sadte, W. (1989) FEBS Lett. 250, 611-614. West, A.P., Rose, M.P. and Cooke, B.A. (1990) Biochem. J. 270, 499-503. Wolbach, S.B. and Howe, P.R. (1925) J. Exp. Med. 43, 753-777.
