Regulation of 3 beta-hydroxysteroid dehydrogenase activity by human chorionic gonadotropin, androgens, and anti-androgens in cultured testicular cells.

delta 5-3 beta-Hydroxysteroid dehydrogenase is a key enzyme for testicular androgen biosynthesis and a marker for the Leydig cells. The hormonal regulation of this enzyme was studied in cultured rat testicular cells. Human chorionic gonadotropin (hCG) increased testosterone production in vitro while time course studies indicated a biphasic action of the gonadotropin on 3 beta-hydroxysteroid dehydrogenase activity. An initial stimulation (51%) of the enzyme was detected between 3 and 12 h of culture when medium testosterone was low. This is followed by an inhibition of 3 beta-hydroxysteroid dehydrogenase activity on days 2 and 3 of culture when medium testosterone was elevated. Concomitant treatment with a synthetic androgen (R1881) inhibited 3 beta-hydroxysteroid dehydrogenase activity and testosterone production in hCG-treated cultures while an anti-androgen (cyproterone acetate) increased 3 beta-hydroxysteroid dehydrogenase activity and testosterone biosynthesis. Addition of 10(-5) M spironolactone, an inhibitor of 17 alpha-hydroxylase, blocked the hCG stimulation of testosterone production but increased medium progesterone. In the absence of the secreted androgen, hCG stimulated 3 beta-hydroxysteroid dehydrogenase activity in a time- and dose-related manner. Furthermore, hCG stimulation of 3 beta-hydroxysteroid dehydrogenase activity and progesterone accumulation in spironolactone-supplemented cultures was decreased by concomitant treatment with R1881 but was not affected by cyproterone acetate. The inhibitory effect of R1881 was blocked by the anti-androgen. In the absence of hCG, treatment with testosterone, dihydrotestosterone, or R1881, but not promegestone, alone also inhibited 3 beta-hydroxysteroid dehydrogenase activity while the inhibitory effect of testosterone was blocked by cyproterone acetate. Thus, hCG stimulates 3 beta-hydroxysteroid dehydrogenase activity in cultured testicular cells. The androgenic steroidogenic end products, in turn, inhibit this enzyme. The hormonal regulation of 3 beta-hydroxysteroid dehydrogenase activity may be important in the ultrashort loop autoregulation of androgen biosynthesis.


Regulation of 36-Hydroxysteroid Dehydrogenase Activity by Human Chorionic Gonadotropin, Androgens, and Anti-androgens in Cultured Testicular Cells*
(Received for publication, February 4, 1983) c. M. Ruiz  A6-38-Hydroxysteroid dehydrogenase is a key enzyme for testicular androgen biosynthesis and a marker for the Leydig cells. The hormonal regulation of this enzyme was studied in cultured rat testicular cells. Human chorionic gonadotropin (hCG) increased testosterone production in vitro while time course studies indicated a biphasic action of the gonadotropin on 38-hydroxysteroid dehydrogenase activity. An initial stimulation (51%) of the enzyme was detected between 3 and 12 h of culture when medium testosterone was low. This is followed by an inhibition of 3B-h~droxysteroid dehydrogenase activity on days 2 and 3 of culture when medium testosterone was elevated. Concomitant treatment with a synthetic androgen (R1881) inhibited 38-hydroxysteroid dehydrogenase activity and testosterone production in hCG-treated cultures while an anti-androgen (cyproterone acetate) increased 3B-hydroxysteroid dehydrogenase activity and testosterone biosynthesis. Addition of lo-' M spironolactone, an inhibitor of 1 7a-hydroxylase, blocked the hCG stimulation of testosterone production but increased medium progesterone. In the absence of the secreted androgen, hCG stimulated 38-hydroxysteroid dehydrogenase activity in a time-and dose-related manner. Furthermore, hCG stimulation of 3b-hydroxysteroid dehydrogenase activity and progesterone accumulation in spironolactone-supplemented cultures was decreased by concomitant treatment with R1881 but was not affected by cyproterone acetate. The inhibitory effect of Rl88l was blocked by the anti-androgen. In the absence of hCG, treatment with testosterone, dihydrotestosterone, or R1881, but not promegestone, alone also inhibited 3fi-hydroxysteroid dehydrogenase activity while the inhibitory effect of testosterone was blocked by cyproterone acetate. Thus, hCG stimulates 38-hydroxysteroid dehydrogenase activity in cultured testicular cells. The androgenic steroidogenic end products, in turn, inhibit this enzyme. The hormonal regulation of 38-hydroxysteroid dehy- Several recent studies suggested that the steroidogenic activity of the rat Leydig cells may be controlled by means of an ultrashort loop negative feedback mechanism via the androgenic secretory products (1)(2)(3)(4)(5). Using a primary culture of rat testicular cells (6), we reported that a synthetic androgen, R1881,' decreases, whereas a synthetic anti-androgen (cyproterone acetate) increases, the gonadotropin-stimulated accumulation of testosterone in vitro (4). Furthermore, R5020, a metabolically stable progestin (7), did not affect the gonadotropin-stimulated accumulation of testosterone in vitro. These studies demonstrated an autoregulatory mechanism in the control of testicular androgen production and suggested that the actions of these androgenic and anti-androgenic agents may be mediated by specific testicular androgen receptors (8,9).
The present study attempts to further characterize the mechanisms by which androgens regulate testosterone production in cultured rat testicular cells. The formation of A4-3-ketosteroids is a key step in the biosynthesis of virtually all biologically active steroids from A5-3/3-hydroxysteroids and is mediated by the enzyme A5-3/3-hydroxysteroid dehydrogenase (EC 1.1.1.51) (10)(11)(12)(13). Because 3P-hydroxysteroid dehydrogenase has been localized histochemically (12,14,15) and biochemically (13,14,16) exclusively in the Leydig cells, and its activity appears to be regulated by luteinizing hormone and hCG (10,(16)(17)(18), we studied the effect of hCG, androgens, and anti-androgens on 3p-hydroxysteroid dehydrogenase activity of cultured rat testicular cells. Evidence is presented to indicate that one of the regulatory sites of androgen action in the control of their own production could be through the steroidogenic enzyme 3P-hydroxysteroid dehydrogenase.

EXPERIMENTAL PROCEDURES
Animals-Adult (50-60 days old) male rats of the Sprague-Dawley strain were hypophysectomized by Curtis Johnson Laboratories (Bridgeview, IL) and delivered on the 5th postoperative day. Physiological saline (0.9% NaCI) solution and a mixture of bread, milk, dog food, and tap water was provided ad libitum.
The cells were cultured without treatment for 8 days during which time the media were collected and replaced every 2 days (4,6). This treatment-free period is necessary for the recovery and stabilization of the responsiveness of cultured Leydig cells to hCG. At the end of this period, the cells were reincubated for 3 additional days (unless indicated otherwise) during which time the various treatments were applied. At the end of incubation, media were collected and stored frozen at -20 "C until assayed by radioimmunoassay for their steroid content. Routinely, the cells were assayed for 3P-hydroxysteroid dehydrogenase activity.
Radioimmunoassay of Testosterone and Progesterone-Testosterone was measured using an antiserum obtained from Dr. G. Abraham (Los Angeles, CA). The specificity of this antiserum has been reported previously (19). Progesterone was assayed by radioimmunoassay as previously described (20).
Assay of A5-3P-Hydroxysteroid Dehydrogenase Activity in Cultured Rat Testicular Cells-The activity of testicular 30-hydroxysteroid dehydrogenase was measured by a procedure developed by Murono and Payne (21) and modified by us (22), in which the conversion of radioactive pregnenolone to progesterone was measured. Cultured cells from various treatment groups were washed with 1 ml of cold assay buffer (0.05 M potassium phosphate buffer, pH 7.4, 1 mM EDTA), and scraped from the dishes using a rubber policeman. The cell suspensions were sonicated for 10 s using a Kontes ultrasonic cell disrupter at the setting of 70%. At this point, the cells were broken as determined by trypan blue exclusion test. Aliquots of the cell homogenate (-8 X lo5 cells/tube, 80-110 pg of protein) were assayed for 36-hydroxysteroid dehydrogenase activity in a final volume of 0.1 ml containing 0.05 M potassium phosphate (pH 7.4), 1 mM EDTA, 100 p M NAD, 50 PM pregnenolone (-2 X lo5 cpm), and 3% dimethyl sulfoxide. The tubes were incubated at 37 "C in a Dubnoff shaker for 30 min unless indicated otherwise. The reaction was stopped by placing the tubes in an ice bath. ['4C]Progesterone (600 cpm) was added to each tube to determine recovery and 5 pg of unlabeled progesterone and 25 pg of pregnenolone were added as carriers. Blank reactions were conducted in which either the cell extract or NAD was omitted. The samples were extracted twice with 10 volumes of diethyl ether and the organic phase was removed and evaporated to dryness under nitrogen. Radiometabolites were redissolved in 50 pl of chloroform and separated by thin layer chromatography in the system chloroform/ether (52, v/v). Each plate was run twice in this system. Progesterone was visualized under UV light while iodine vapor was used to visualize pregnenolone. The RF values for pregnenolone and progesterone were 0.59 and 0.80, respectively. The spots corresponding to progesterone were cut out placed in vials containing scintillation fluid which consisted of Omnifluor (New England Nuclear), toluene, and dioxane (75:16:3), and radioactivity was measured in a liquid scintillation counter. Correcting for recovery, the results were expressed as nanomoles of progesterone formed/30 min/mg of protein.
Optimal conditions for the assay of 3fi-hydroxysteroid dehydrogenase activity were determined on homogenates of testicular cells (data not shown). Primary cultures of testicular cells were cultured for 10 days in McCoy's 5a medium, and cell homogenates were prepared for 3@-hydroxysteroid dehydrogenase assay. Progesterone formation remained linear for up to 30 min of incubation. In the absence of tissue homogenate, no progesterone formation was detected throughout the incubation. Incubation with increasing concentrations (15-240 pg/ tube) of cell protein resulted in a linear increase in progesterone formation. The standard incubation time and tissue protein concentration used in all of the following experiments were 30 min and <240 pg/tube, respectively. The concentrations of pregnenolone and NAD giving half-maximal enzyme activity were 6.25 and 25 pM, respectively. The enzyme activity remained constant at concentrations greater than 50 p M pregnenolone and 0.1 mM NAD, thus indicating the maximum velocity of the reaction has been achieved. The optimal incubation temperature was also investigated. Enzyme activity at 5 "C was negligible. Maximal activity was observed at 37 "C, and declined at 45 "C, reflecting a probable degradation of the enzyme. Also, the optimal pH for the 30-hydroxysteroid dehydrogenase activity was found to be 7.4. Although enzyme activity increased gradually from pH 2.0 to 6.0, it remained relatively constant from pH 7 to 8 and declined at pH 9 and 10.
Regardless of tissue concentration, incubation time, or hormone treatment (see below), progesterone was not further metabolized under these incubation conditions, as assessed by scanning the chromatography plates for radioactivity (results not shown). Furthermore, successive recrystallizations of the reaction product were performed to demonstrate that the radioactivity associated with progesterone in the thin layer chromatography plate was authentic (data not shown).
Protein Determinutiom-The Bio-Rad protein assay (Bio-Rad Co., Richmond, CA) was used to determine protein content in tissue preparations with bovine y-globulin as standard (23). Treatment with various hormones did not affect protein content per culture.
Data Analysis-All experimental data were represented as the mean f S.E. of triplicate or quadruplicate cultures. Comparable results were obtained in two to four separate experiments. Radioimmunoassay data were calculated and analyzed with a Hewlett-Packard desk-top calculator, model HP 9830A, using an adaptation of the program of Midgley et al. (24). To obtain enzyme kinetic constants (Km, Vmax), experimental data were fitted to a modified Lineweaver-Burk plot (25). Statistical significance was determined by the paired two-tailed Student's t test and the two-way analysis of variance.
Comparisons with p > 0.05 were not considered significant.

RESULTS
Effect of Treatment with hCG, Cyproterone Acetate, and RIB81 on Testosterone Accumulation and 3P-Hydroxysteroid Dehydrogenase Activity of Cultured Testicular Celts-To examine the effects of treatment with hCG, androgen, and antiandrogen on testosterone production and 3P-hydroxysteroid dehydrogenase activity, testicular cells were cultured for 8 days as described under "Experimental Procedures." After 8 days, cells were cultured for 3 additional days in either medium alone (C, controls) or with medium supplemented with hCG (10 ng/ml), R1881 M), cyproterone acetate (1O"j M), or combinations thereof (Fig. 1). Basal testosterone accumulation was negligible in control cells whereas hCG led to a 23-fold increase in testosterone accumulation. Concomitant treatment with cyproterone acetate further enhanced by 3.1fold the hCG action whereas R1881 inhibited the hCG-stimulated testosterone production to values similar to the control cells. The inhibitory effect of R1881 on hCG-stimulated testosterone accumulation was counteracted by concomitant treatment with cyproterone acetate, resulting in testosterone levels 27-fold higher than cells treated with hCG a n d R1881.
The activity of the 3P-hydroxysteroid dehydrogenase enzyme was also measured (Fig. 1B). Paradoxically, treatment with hCG resulted in a slight (22%) but significant decrease 10990 Hormonal Reguiution of 3p-Hydroxysteroid Dehydrogenase Activity  in enzyme activity, whereas concomitant treatment with cyproterone acetate resulted in a 63% increase in enzyme activity above control levels. In cont,rast, concomitant treatment with R1881 resulted in a 60% decrease in enzyme activity in hCG-treated cells and the inhibitory effect of R1881 was partially counteracted by cyproterone acetate. These results indicated that, after 3 days of treatment, hCG caused a paradoxical inhibition of 3P-hydroxysteroid dehydrogenase activity. Concomitant treatment with cyproterone acetate enhanced while treatment with R1881 inhibited the enzyme activity.
Time Course of the Effect of hCG on 3P-Hydroxysteroid Dehydrogenase Activity and Testosterone Production in Cultured Testicular Cells-We further tested the time course of the hCG regulation of the enzyme activity. Testicular cells were incubated for increasing periods of time (3-72 h) in the presence or absence of hCG (10 ng/ml) (Fig. 2). The accumulation of testosterone was negligible in untreated cells throughout the incubation period. Treatment with hCG resulted in a time-dependent increase in testosterone accumulation, reaching 5-10 ng/ml after 12-24 h of culture. The activity of $3-hydroxysteroid dehydrogenase in control cells remained constant during the 72-h incubation period (2.74 k 0.20 nmol of progesterone formed/30 min/mg of protein). In contrast, treatment with hCG increased enzyme activity by 51,42, and 27% at 3,6, and 12 h after incubation, respectively. At 48 and 72 h after incubation, a gradual decrease of 19 and 36% in enzyme activity was observed. These results indicate that hCG initially stimulates 3P-hydroxysteroid dehydrogenase activity when medium testosterone was low. At 48 and 72 h after incubation, high levels of medium testosterone is associated with decreases in 3P-hydroxysteroid dehydrogenase activity.

Effect of Treatment with increasing Concentrations of hCG on 3P-Hydroxysteroid Dehydrogenase Activity and Testosterone Accumulation in Testicular Cells Treated with Cyproterone
Acetate-Since cyproterone acetate antagonizes androgen action in accessory sex organs (26) and augments testosterone production in hCG-treated testicular cells (4), we tested the action of hCG on 3p-hydroxysteroid dehydrogenase activity in the presence of cyproterone acetate. Treatment with hCG (1-100 ng/ml) led to a dose-dependent increase in testosterone accumulation, with 10 ng/ml of hCG eliciting a 63-fold P 0 + CA + CA increase in testosterone accumulation (Fig. 3B). Concomitant treatment with cyproterone acetate produced significant increases in testosterone accumulation at all doses of hCG tested. As shown in Fig. 3A, the activity of the 3P-hydroxysteroid dehydrogenase enzyme was not affected ( p > 0.05) by 1 and 3 ng/ml hCG, but a 42% decrease ( p < 0.05) in 36hydroxysteroid dehydrogenase activity was observed at 10 ng/ ml hCG. Higher doses (30 and 100 ng/ml) of hCG were ineffective ( p > 0.05). In the presence of cyproterone acetate, hCG treatment resulted in a dose-dependent increase in 30hydroxysteroid dehydrogenase activity, with 50,88, and 107% increases in enzyme activity at 1, 3, and 10 ng/ml hCG, respectively. Maximal stimulation was observed a t 30 ng/ml hCG (121% above control values).
These results suggest that hCG stimulates the production of testosterone which may, in turn, inhibit 3P-hydroxysteroid dehydrogenase activity. The stimulatory effect of hCG on 3Phydroxysteroid dehydrogenase activity can only be observed in the presence of the anti-androgen.
Effect of increasing Doses of hCG on 36-Hydroxysteroid Dehydrogenase Activity in Testicular Cells Treated with Spironolactone-To further study the hCG effect on 3P-hydroxysteroid dehydrogenase activity in the absence of endogenous androgens, testicular cells were cultured for 3 days with hCG (1, 10, and 100 ng/ml) with or without spironolactone to inhibit the androgen biosynthetic enzyme, 17a-hydroxylase (27). Spironolactone treatment alone did not affect basal 36hydroxysteroid dehydrogenase activity, while 10 ng/ml hCG decreased (35%) 3p-hydroxysteroid dehydrogenase activity (Fig. 4). In the presence of spironolactone, hCG increased 3phydroxysteroid dehydrogenase activity in a dose-dependent manner with 82, 123, and 150% increases at 1, 10, and 100 ng/ml hCG, respectively. Since medium testosterone in spironolactone-treated cells was negligible (data not shown), one can postulate that hCG stimulated 3p-hydroxysteroid dehydrogenase activity in the absence of endogenous testosterone.
T o evaluate the effects of spironolactone on the apparent kinetic constants of 36-hydroxysteroid dehydrogenase, tissue homogenates from cells treated with hCG or hCG plus spironolactone were incubated with increasing concentrations of pregnenolone to study progesterone formation (Fig. 5). The apparent kinetic constants (V,,,,, and K,) of 3P-hydroxysteroid dehydrogenase were calculated by a modified Lineweaver-

Effect of hCG, R1881, and Cyproterone Acetate on 3fi-Hydroxysteroid Dehydrogenase Activity
of Testicular Cells Treated with Spironolactone-The modulatory role of exogenous androgen and anti-androgen on hCG-stimulated 3phydroxysteroid dehydrogenase activity was studied in the absence of endogenous androgens (Fig. 7). Testicular cells were cultured with spironolactone M ) to inhibit androgen production. Cells were treated with hCG (1, 10, and 100 ng/ml) alone or in combination with R1881 and/or cyproterone acetate. Treatment with hCG stimulated 3P-hydroxyste- In several cases, the S.E. is less than the data points drawn. C, controls; P, progesterone.
Burk plot according to Wilkinson (25). The results indicated that spironolactone treatment resulted in a 146% increase in the apparent Vmax. Furthermore, there were no significant differences between apparent K , values in hCG-treated cells with or without concomitant spironolactone treatment (hCG: 5.26 k 0.15 p~; hCG plus spironolactone: 5.27 f 0.20 g~) .

Time Course of the Effect of hCG on 3B-Hydroxysteroid Dehydrogenase Activity and Progesterone Production in Testicular Cells Treated with Spironolactone-Testicular cells were incubated for increasing periods of time (3-72 h) in
spironolactone-supplemented medium with or without 10 ng/ ml hCG (Fig. 6). Treatment with hCG resulted in a timedependent increase in 3P-hydroxysteroid dehydrogenase activity in spironolactone-supplemented cells. Throughout the 72-h period, testosterone accumulation in the medium was minimal ((0.5 ng/ml) in control and hCG-treated cells. In contrast, progesterone accumulation was low in control cells while hCG treatment resulted in a time-dependent increase  roid dehydrogenase activity by 2.7-, 3.5-, and 3.7-fold at 1, 10, and 100 ng/ml hCG, respectively. R1881 partially counteracted the hCG stimulation of 30-hydroxysteroid dehydrogenase activity, but the enzyme activity remained 71, 122, and 119% higher than control cells a t 1, 10, and 100 ng/ml hCG, respectively. Cyproterone acetate had no effect on the hCGstimulated 3P-hydroxysteroid dehydrogenase activity ( p > 0.05), but partially prevented the R1881-induced decrease in hCG-stimulated 3p-hydroxysteroid dehydrogenase activity. Treatment with hCG stimulated progesterone accumulation at all doses while concomitant treatment with R1881 inhibited the hCG action ( p < 0.05) (Fig. 7B). Treatment with cyproterone acetate did not affect hCG-stimulated progesterone accumulation ( p > 0.05) while the inhibitory effect of R1881 was completely blocked by cyproterone acetate.
Effect of Exogenous Androgens and Anti-androgens on Basal 3P-Hydroxysteroid Dehydrogenase Actiuity in Cultured Testicular Cells-The possibility that androgens may regulate basal 30-hydroxysteroid dehydrogenase activity in the absence of hCG was further investigated in testicular cells treated with increasing concentrations ( 10"O-lO"j M ) of testosterone, cyproterone acetate, or both compounds (Fig. 8). Testosterone produced a dose-dependent inhibition of 3P-hydroxysteroid dehydrogenase activity with 35,56, and 59% decreases a t lo-', lo-', and M testosterone, respectively. In contrast, treatment with cyproterone acetate had no effect on 3p-hydroxy- steroid dehydrogenase activity but partially counteracted the inhibitory effect of testosterone. The specificity of the androgen action was further tested by incubating testicular cells with DHT, R1881, or the synthetic progestin R5020 (Fig. 9). DHT reduced 3P-hydroxysteroid dehydrogenase activity by 11 and 34% a t lo-' and M, respectively. Similarly, R1881 produced 17 and 63% decreases in enzyme activity. In contrast, the synthetic progestin R5020 did not affect enzyme activity ( p > 0.05).
The time course of the inhibitory effect of androgen was further tested. As shown in Fig. 10, the activities of 38hydroxysteroid dehydrogenase remained constant during the 3-day culture while treatment with exogenous testosterone  Results represent mean f S.E. of quadruplicate determinations. In several cases, the S.E. is less than the data points drawn. P, progesterone.
( M) resulted in a time-dependent inhibition of the enzyme activity. At 6, 12, and 24 h after androgen treatment, the activities of 3fl-hydroxysteroid dehydrogenase were decreased by 28, 43, and 5296, respectively ( p < 0.05). The activities of the enzyme remained suppressed during the following 2 days of culture.
To examine a possible interference of the enzyme assay by androgens, homogenates from control cells were used for 3phydroxysteroid dehydrogenase assay (Fig. 11). Addition of increasing concentrations (10-7-10-5 M) of testosterone, DHT, R1881, cyproterone acetate, or R5020 did not interfere with the enzyme assay ( p > 0.05).
We further tested the effect of treatment with hCG and testosterone on kinetic constants of the 3P-hydroxysteroid dehydrogenase enzyme (Fig. 12). Testicular cells were cul-

DISCUSSION
The androgen regulation of a key Leydig cell enzyme in the androgen biosynthetic pathway was investigated in uitro using primary cultures of rat testicular cells. The present results indicate that 1) the synthetic androgen R1881 inhibits, while the synthetic anti-androgen cyproterone acetate enhances, hCG-stimulated androgen production, This is accompanied by similar changes in the 30-hydroxysteroid dehydrogenase activity. 2) Treatment with hCG (10 ng/ml) initially stimulates 36-hydroxysteroid dehydrogenase activity. This is followed by a time-dependent inhibition of the enzyme activity, which may be due to the inhibitory effect of endogenously produced androgen. 3) In the presence of an anti-androgen (cyproterone acetate) or an inhibitor of endogenous androgen biosynthesis (spironolactone), hCG stimulates 30-hydroxysteroid dehydrogenase activity in a time-and dose-related fashion. The stimulatory effect of hCG is inhibited by exogenous androgens. 4) In the absence of hCG, treatment with androgens, but not with a synthetic progestin, inhibits basal 3phydroxysteroid dehydrogenase activity. The inhibitory action of androgens is time-dependent and is blocked by an antian-Hormonal Regulation of 3P-Hydroxysteroid Dehydrogenase Activity 10995 drogen. 5 ) Treatment with hCG and androgens affects the maximal velocity of 3P-hydroxysteroid dehydrogenase without changing the K, value.
Leydig cell functions are under the control of multiple hormones (28). In addition to pituitary gonadotropin action, Leydig cell testosterone production may be controlled locally via an autoregulatory, short loop negative feedback mechanism (1)(2)(3)(4)(5). In the present study, the effect of androgens on the conversion of radiolabeled pregnenolone to progesterone was studied. Although this conversion is mediated by a complex of two enzymes (3p-hydroxysteroid dehydrogenase and A5-A4 isomerase; EC 5.3.3.1), the isomerase activity appears in excess (1 1) and the formation of progesterone is used as an index of 38-hydroxysteroid dehydrogenase activity. Since 38hydroxysteroid dehydrogenase activity has been demonstrated exclusively in the Leydig cells (12)(13)(14)(15)(16), the present results suggest that the autoregulation of androgen biosynthesis is correlated with modulation of a key enzyme in the androgen biosynthetic pathway. It is, however, important to note that 3P-hydroxysteroid dehydrogenase activity is probably not rate-limiting under all conditions and it is likely that additional steroidogenic enzymes may also be regulated by androgens and the gonadotropin.
The present results showed that hCG at 10 ng/ml initially stimulates 3P-hydroxysteroid dehydrogenase activity at 3 h after incubation. Subsequently, hCG decreases the enzyme activity after medium testosterone has been accumulated. In contrast, in the presence of cyproterone acetate or spironolactone, only stimulatory effects of hCG could be observed. Thus, the observed biphasic effect of hCG may represent the stimulatory action of the gonadotropin as well as the inhibitory action of the androgenic steroidogenic end products. Furthermore, time course studies indicated that the hCG-stimulated accumulation of progesterone in spironolactone-treated cultures is associated with increases in the activities of the progesterone biosynthetic enzyme 36-hydroxysteroid dehydrogenase. Addition of R1881 inhibits the hCG stimulation of both functional parameters while cyproterone acetate blocks the inhibitory action of R1881. In the absence of endogenous androgens, treatment with the anti-androgen does not increase 38-hydroxysteroid dehydrogenase activity in hCG-treated cells.
The mechanism by which androgens block 3P-hydroxysteroid dehydrogenase activity was also studied. When added in uitro to the enzyme assay, no effect of the steroids on 3phydroxysteroid dehydrogenase activity could be observed. Thus, possible interference of androgens with the active site of the enzyme is ruled out. Furthermore, incubation with testosterone or hCG decreases the apparent maximal velocity, without affecting the apparent K , of the enzyme. Likewise, hCG stimulates the apparent Vmax, but not the K,, of the enzyme in spironolactone-treated cells. Thus, the effect of androgens on 3P-hydroxysteroid dehydrogenase activity is not due to a modification of the affinity of the enzyme to its substrate. Instead, androgens may interfere with the synthesis of the enzyme.
In the absence of hCG, treatment with both synthetic (R1881) and natural (testosterone and DHT) androgens inhibits the 3P-hydroxysteroid dehydrogenase activity. This inhibitory action of testosterone is dose-dependent with 50% inhibition exerted by -1 nM testosterone. It is interesting to note that time course studies indicated that 3P-hydroxysteroid dehydrogenase activity in hCG-treated cells began to decrease a t 6 h after treatment during which time the medium testosterone reached -1 ng/ml(-3 nM). Although the exact intracellular concentration and time course of action of endogenously produced testosterone is not known, exogenously ad-ministered testosterone inhibits 3P-hydroxysteroid dehydrogenase as early as 6 h after treatment. Furthermore, other weaker androgens (e.g. A4-androstenedione) produced by these cells may also be involved in the regulation of 3phydroxysteroid dehydrogenase activity.
The observed androgen effect is probably receptor-mediated because cyproterone acetate reverses the androgen action. Although cyproterone acetate has been shown to possess progestational activity in the rat prostate and rabbit uterus (29,30), the lack of an inhibitory effect of a synthetic progestin (R5020) in the present system rules out the involvement of progesterone receptors. Androgen receptors have been identified in Leydig (8,9) and Sertoli (31-33) cells, both of which are present in the primary testicular cell cultures used here. Thus, it is possible that both cell types are involved in mediating the androgen action. Furthermore, the observed inhibition of 3p-hydroxysteroid dehydrogenase by the nonaromatizable androgen DHT and the synthetic androgen R1881 rules out the possible aromatization of testosterone to estrogens in the mediation of the observed inhibitory effect.
Autoregulation of endocrine functions through an ultrashort loop feedback mechanism is not limited to the Leydig cells. Corticosteroids induce self-suppression of corticosteroidogenesis in the adrenal cells (34)(35)(36) while thyroid hormone also inhibits the thyroid adenylate cyclase activity (37). In ovarian granulosa cells, estrogens augment the gonadotropin stimulation of aromatase activity (38). Furthermore, progesterone inhibits the activity of 3P-hydroxysteroid dehydrogenase in the ovine luteal cells (39). These studies reinforce the importance of local control mechanisms in the regulation of optimal endocrine secretion.
Changes in microvasculature of the testis may modify the local concentrations of androgens through changes in blood flow or androgen-binding protein concentrations. The accumulated androgens may, in turn, act on specific intracellular androgen receptors to affect Leydig cell 3P-hydroxysteroid dehydrogenase activity and testosterone biosynthesis. The androgen suppression of 38-hydroxysteroid dehydrogenase activity in the Leydig cells serves as a marker for androgen action and may also be involved in the "desensitization" of Leydig cell androgen production following treatment with high doses of luteinizing hormone/hCG (40-42). Thus, in addition to the classic pituitary control of androgen biosynthesis, local fine adjustment may take place through an androgen receptor-mediated regulation of 3P-hydroxysteroid dehydrogenase activities in the Leydig cells.