Angiotensin I1 Receptors and Inhibitory Actions in Leydig Cells*

Rat Leydig cells possess functional high-affinity receptors for angiotensin I1 (AII). AI1 inhibits adenyl- ate cyclase activity in Leydig cell membranes and re-duces basal and human chorionic gonadotropin (hCG)- stimulated CAMP pools and testosterone production in intact cells. Treatment of cells with an inhibitory dose of forskolin (lo-' M) and a submaximal dose of AI1 caused additive inhibition of hCG-stimulated events. The inhibitory action of AI1 was largely prevented by pertussis toxin prior to the addition of AI1 alone or in the presence of hCG. This study and our recent report on inhibitory action of low doses of forskolin, 10-l2- lo-' M (Khanum, A., and Dufau, M. L. (1986) J. Biol. Chem. 261,11456-1 1459) are indicative of a pertussis toxin-sensitive subunit of adenylate cyclase available for acute regulation of Leydig cell function. 8-bromo-CAMP bypasses the inhibitory effect of forskolin as well as AII. We have, therefore, demonstrated functional AI1 high-affinity receptor and an acute inhibi- tory effect of AI1 on hCG action in Leydig cells. Our results have provided evidence for a pertussis toxin- sensitive guanine nucleotide inhibitory protein as me-diator of the effect of AIL These findings further em- phasized the importance of the CAMP pathway in the Leydig cells, and studies also suggest that tubular and locally produced AI1 could negatively modulate luteinizing

A number of studies have provided evidence for the presence of the renin-angiotensin system in reproductive tissues. Immunoreactive renin has been detected in Leydig cells of rat and human testes and was found to be pituitary-dependent in the rat (1,2). Similarly, more recent studies have shown the presence of renin, angiotensin I and I1 (AI and AII)' in normal rat Leydig cells and a murine Leydig cell line (3, 4). Furthermore, angiotensin-converting enzyme activity was demonstrated in rat testis and shown to be localized predominantly in the germinal cells, whereas only minor activity was found in the purified adult rat Leydig and Sertoli cells (5). Also, [3H]captopril bound specifically to cellular fractions enriched in germinal cells (5). Velletri et al. (5) have suggested that the pituitary gland is required for development and maintenance of the rat testicular angiotensin-converting enzyme through stimulation of steroidogenesis in the testis. The biochemical evidence (5) was consistent with immunofluorescence studies * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
'The abbreviations used are: AI and AII, angiotensin I and II; hCG, human chorionic gonadotrophin; Ni, guanine nucleotide inhibitory unit; HPLC, high pressure liquid chromatography. in swine testis showing the presence of converting enzyme in spermatids and other stages of germinal cells (6).
On the other hand, there is strong evidence that Leydig cells possess a membrane system with the potential for negative modulation of gonadotrophic action. In this regard, we have demonstrated a novel, high-affinity inhibitory action of low doses of forskolin (range 10"2-10-9 M) upon adenylate cyclase activity and cAMP generation, an effect that appears to be mediated by the pertussis toxin-sensitive guanine nucleotide inhibitory unit (Ni) of adenylate cyclase (7). Since AI1 has been shown to exert an inhibitory influence on adenylate cyclase system of liver (8,9), adrenal cortex (10, l l ) , renal cortex (12), and smooth muscle (13), this hormone could be a potential regulator of gonadal function. It is, therefore, conceivable that tubular and locally produced AI1 could modulate the action of gonadotrophin in Leydig cells.

Materials
Medium-199 and elutriation medium were obtained from Whittaker M. A. Bioproduct, Inc., Walkersville, MD and National Institutes of Health Media Supply Units, Bethesda, MD respectively. hCG (CR-121) was kindly provided by the Center of Population Research, National Institute of Child Health and Human Development, Bethesda, MD. AI1 was purchased from Sigma. [Sar',Alaa]AII was obtained from Vega Biotechnologies, Tucson, A Z , and forskolin from Behring Diagnostics. AI1 ( 5 4 , LHRH, AI, and [des-Asp']AII were obtained through Peninsula Laboratories, San Carlos, CA. Pertussis toxin was purchased from List Biological Laboratories, Inc., Campbell, CA. '"1-AII (2200 Ci/mmol) and succinyl cAMP '251-tyrosine methyl ester (2000 pCi/pg) were prepared by Meloy Laboratories, Springfield, VA, and by Hazleton Biotechnologies, Vienna, VA, using a modification of the chloramine-T method (14), followed by purification by HPLC (15). [cY-~*P]ATP (800 Ci/mmol) were obtained from Du Pont-New England Nuclear.

Methods
Preparation of Leydig Cells and Membranes-Adult male Sprague-Dawley rats (Charles River Breeding Laboratories, Wilmington, MA) were killed and testes were removed and placed in ice-cold PBS, pH 7.4. Interstitial cells were obtained by collagenase digestion of decapsulated testes, as described previously (16). Crude cell suspension was washed and then pelleted at 200 X g for 10 min. The cell pellet was resuspended with elutriation buffer consisting of regular medium-199 with Hanks' salts and L-glutamine containing 1.4 g/liter NaHC03, 0.5% bovine serum albumin-, 1 mM EDTA, 50 units/ml heparin, 12.5 pg/ml DNase and 50 pg/ml gentamycin, pH 7.4. The purified cells were obtained by centrifugal elutriation (17). Cells were centrifuged and resuspended in medium-199 containing 0.1% bovine serum albumin and were incubated at 34 "C with shaking at 100 cycles/min under an atmosphere of 02:C02 (955, v/v) in the presence and in the absence of various concentrations of hCG with or without AH, forskolin, and 8-bromo-CAMP. For some experiments, cells were pretreated for 60 min with pertussis toxin under the same incubation conditions prior to the addition of hCG and/or AIL The incubations were terminated by transferring the incubation tubes to an ice bath; all further steps were carried out at 0 "C and processed for the analysis of cAMP (intracellular and receptor-bound) and testosterone; and  their measurements were performed by radioimmunoassay as described previously (18).
Purified Leydig cell plasma membranes were prepared in the presence of 5 mM EDTA (19) and stored in liquid nitrogen until use. Adenylate cyclase assay was carried out by methods previously described (9). For some experiments, membranes were also preincubated for 20 min with activated pertussis toxin.
Binding Studies-Binding assays with '"I-AII were performed with intact cells which were incubated in 500 p1 of Dulbecco's PBS in the presence of 0.2% bovine serum albumin and 100 FM phenylmethylsulfonyl fluoride for various time periods: 0-120 min at different temperatures: 4 "C, 22 "C and 37 "C. For equilibrium studies, cells were incubated at 22 'C for 20 min with increasing concentrations of '261-AII in the presence or in the absence of 10 g i~ of AI1 or with '%I-AI1 and increasing concentrations of unlabeled AIL In some displacement studies, AII-related and -unrelated peptides were used. The reaction was terminated by adding 2 ml of ice-cold PBS, pH 7.4, in the tube and immediately filtered under vacuum through GF/C filters (Whatman, Maidstone, Great Britain). Filters were washed twice with 3 ml of PBS and bound radioactivity was measured in a y counter. For kinetic studies, Leydig cells were also incubated with 1261-AII at various times: 0-30 min. The specific binding of '251-AII reached to equilibrium after 14 min of incubation. Therefore, the rate of dissociation of "'I-AII receptor complex was obtained by adding 10 p~ of unlabeled AI1 after 14 min of incubation. The binding studies were analyzed by the method of Scatchard (20)

RESULTS AND DISCUSSION
In this study, we have demonstrated the presence of functional AI1 receptors in rat Leydig cells and have provided the evidence of its acute inhibitory effects. Cells were incubated with lZ5I-AII for various time periods at 37 "C, 22 "C, and 4 "C. The highest specific binding was observed when the cells were incubated at 22 "C, and the binding reached to maximum within 30 min of incubation (Fig. 1). Thereafter, there was subsequent decrease in binding. At 37 "C, the specific binding increased rapidly after 5 min of incubation followed by a rapid decline. This is most likely due to tracer degradation by proteolytic enzymes as indicated by others (22) or may be due to internalization of receptor complex and subsequent receptor degradation. In fact, this receptor-mediated internalization has been indicated in cases of various peptide hormones (23). However at 4 "C, the specific binding was increased gradually and reached equilibrium after 90 min of incubation. In this case the specific binding was only 10% of that observed at 22 "C during 20 min of incubation. Therefore, for binding studies, the incubation was carried out at 22 "C for 20 min. Scatchard plot derived from equilibrium binding data obtained at 22 "C showed the presence of high-affinity binding sites with KO of 1.7 X 1 0 ' ' M" f 0.41 ( n = 4) and number of 2018 & 408 ( n = 4) receptor sites per cell (Fig. 2). The binding was saturable and reversible as indicated by saturation analy-sis of binding data and association and dissociation studies, respectively. The rates of association (kl) and dissociation ( k 1 ) of Iz5I-AII binding to Leydig cells were 0.885 nM" min-' and 0.04 rnin", respectively (Fig. 3). Furthermore, the calculated half-life was 17.5 min. From the kinetic data, the calculated equilibrium constant, K,, was 2.2 X 10" M-l. This value is very close to the value obtained by equilibrium analysis. Fig. 4 showed the specificity of AI1 binding to Leydig cell receptor sites. No inhibition in binding was observed with the neurohypophyseal peptide, arginine vasopressin and diterpene, forskolin. On the other hand, the binding-inhibition potency of various analogues and fragment was noted.
[Sar1,Ala8]AII, AII, and [des-Asp']AII were potent displacers of specific '251-AII binding and their KO values were similar to the observed K. for AI1 (1.7 X 10" M-'). AI was found to be 100 times less potent than AI1 in the binding assay, and the smaller fragment of AI1 (AII, Refs. 5-8) displayed weak inhibitory effect on '251-AII binding. LHRH (hypothalamic gonadotropin-releasing hormone), an unrelated decapeptide, competed quite effectively with lZ5I-AII for binding to Leydig cell receptor sites when present in relatively high concentrations, although it was less potent (1000 times) than the AII. The potency order of these compounds for AI1 receptors is AII> [Sar',Ala8]AII> [des-Asp']AII> AI>LHRH> AI1 (5-8). LHRH (hypothalamic gonadotropin-releasing hormone) has shown its ability to inhibit lZ5I-AII binding in Leydig cells. This is consistent with earlier studies by Capponi and Catt (25) in the adrenal cortex and uterus suggesting that the binding inhibition of AI1 by LHRH is probably due to a common structural feature in the COOH-terminal sequence of LHRH and AII.
Although a number of studies have indicated the presence of the renin-angiotensin system in reproductive tissues on the basis of immunohistochemical, immunofluorescence, HPLC, and radioimmunoassay studies (1-4, 6, 26), its physiological significance and possible role in gonadal function are still unclear. Our studies have provided direct evidence for the

FIG. 8. Effect of 8-bromo-CAMP on An-induced inhibition of testosterone production.
Leydig cells (1 X 10' cells/ml) were incubated for 60 min with 8-bromo-CAMP in the presence and in the absence of hCG and/or AII. Each point represents the mean f S.E. of triplicate incubations.

Effect of 8-bromo-CAMP on forskolin-induced inhibition of testosterone production by Leydig cells
Testosterone production was measured as described under "Experimental Procedures." Leydig cells (lo6 cells/ml) were incubated for 120 min in the presence and in the absence of described additions. Mean values between groups by student's t test are as follows: b versus a and d ( p < 0.05); b versus c ( p < 0.01). presence of functional AI1 receptors, since AI1 acutely inhibits (as early as 15 min) gonadotrophin stimulation of cyclic AMP pools and testosterone production in Leydig cells. Fig. 5, left, shows AII-induced inhibition of intracellular cAMP production and the effect of pertussis toxin on this inhibition. Cells were stimulated with a submaximal dose of hCG in the presence and in the absence of increasing concentrations of AII, lO-"-lO"j M. The inhibitory effect of AI1 was dosedependent with an ID5, of 0.5 X 10"' M. However, this effect was found to be largely prevented when cells were incubated with pertussis toxin prior to the addition of AI1 and/or hCG. The inhibitory effect of AI1 affected the hormonal stimulus over the entire dose-response range of hCG concentrations. As a result of addition of a submaximal dose of AI1 to the Leydig cells stimulated by various doses of hCG, the doseresponse curve showed a significant increase in EDso (2-fold), Fig. 5, right. Our studies have also provided evidence for an inhibitory effect of AI1 on receptor-bound cAMP (Fig. 6). Both basal and hCG-stimulated cAMP production were inhibited by the addition of AI1 to the cells. However, this inhibition was prevented when cells were incubated with pertussis toxin prior to the addition of AI1 and/or hCG. As a result of reduced production of CAMP, AI1 (10"o-10-6 M), in a dose-dependent manner, inhibited testosterone production by Leydig cells stimulated by hCG (ID50 l x 10"' M). This was commensurate with its binding affinity. Further addition of forskolin (lo-' M) to a submaximal inhibitory dose of AI1 caused an additive inhibition of testosterone production (Fig.  7, left), and this finding, therefore, indicated that both substances are exerting their inhibitory effects through a common pathway. This inhibitory action of AI1 was found to be prevented when cells were pretreated with pertussis toxin prior to the addition of hCG and/or AI1 (Fig. 7, right). This finding resembles our previous studies, demonstrating the involvement of a pertussis toxin-sensitive regulatory protein in the inhibitory action of low doses of forskolin (7). Moreover, our results are consistent with early studies on the adrenal (22) and more recent studies in liver (27) and kidney (28) which showed that AI1 receptors are functionally linked to the inhibitory unit of the adenylate cyclase system. Furthermore, 8-bromo-CAMP ( M) bypassed the inhibitory effect of AI1 in hCG-stimulatedcells (Fig. 8). This cAMP derivative was also shown to exert a similar effect on the inhibitory action of a low dose of forskolin (Table I). Also, AI1 significantly decreased GTP and luteinizing hormone plus GTP-stimulated adenylate cyclase activities in membranes, and this inhibition was prevented by pretreatment of membranes with pertussis toxin (Table 11). Thus, the reduction in cAMP levels by AI1 is attributable to the inhibition of adenylate cyclase activity in the plasma membrane. It is, therefore, proposed on the basis of our studies that the Ni unit of adenylate cyclase is involved in the inhibitory action of AI1 on adenylate cyclase and consequently on cAMP pools and testosterone production. However, we cannot rule out the participation of other additional signal-transducing systems in mediating the inhibitory action of AI1 in the Leydig cells, since the role of calcium as a second messenger in the action of AI1 is well documented in liver (29), adrenals (30, 31), smooth muscle (32), and kidney (28). Also, the existence of a pertussis toxin-independent guanine nucleotide inhibition has been postulated for several systems (27,33,34). Furthermore, in liver (27) and in kidney (28), two different kinds of receptor-dependent mechanisms were delineated for the cellular responses of AI1 including phospholipase C/increase in intracellular calcium and adenylate cyclase system. However, in the present study, the finding that 8-bromo-CAMP bypassed the inhibitory effect of AI1 on hCG-stimulated testosterone production is suggestive that the pertussis-sensitive regulatory protein (Ni) mainly mediates the inhibitory effect of AI1 and modulates LH stimulation of Leydig cells. Because of the predominant localization of angiotensin-converting enzyme in the testicular tubular elements, it is likely that AI1 possesses a physiological paracrine regulatory function and the locally produced hormone would also effectively exert homologous negative modulatory influence on hormonal-stimulated events in the Leydig cells.