C17-20 lyase cytochrome P450, and cholesterol side-chain cleavage cytochrome P450 messenger ribonucleic acid levels in primary cultures of mouse Leydig cells.

0013-7227/91/1293-1429$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society

Vol. 129, No. 3 Printed in U.S.A.

Multiple Mechanisms for Regulation of 3/?-Hydroxysteroid Dehydrogenase/A5—»A4-Isomerase, 17a-Hydroxylase/C 17-20 Lyase Cytochrome P450, and Cholesterol Side-Chain Cleavage Cytochrome P450 Messenger Ribonucleic Acid Levels in Primary Cultures of Mouse Leydig Cells* ANITA H. PAYNE AND LINLI SHA Departments of Obstetrics and Gynecology and Biological Chemistry and the Reproductive Sciences Program, University of Michigan, Ann Arbor, Michigan 48109-0278

ABSTRACT. The regulation of mRNA levels for A6-3/3-hydroxysteroid dehydrogenase/A6—»A4-isomerase (3/3HSD), 17ahydroxylase/Cn-20 lyase cytochrome P450 (P450na) and cholesterol side-chain cleavage cytochrome P450 (P450KC) was studied in primary cultures of mouse Leydig cells. Treatment of Leydig cells with 8-bromo-cAMP (cAMP) was essential for expression of P450i7a mRNA, but not for 30HSD. Treatment with cAMP caused a decrease in basal levels of 3/JHSD mRNA. The addition of aminoglutethimide (AG), an inhibitor of cholesterol metabolism, to the cAMP-treated cultures resulted in increased expression of both 3/3HSD and P45017o mRNA levels. The addition of testosterone or the androgen agonist mibolerone to cAMP- plus AG-treated cultures reduced 3/3HSD and P45017o mRNA to levels comparable to those observed when cells were treated with cAMP only. The glucocorticoid dexamethasone reduced both basal and cAMP- plus AG-induced increases in 3/3HSD mRNA, but not in P450na mRNA. Estradiol at a concentration of 1 fiM had no effect on cAMP- plus AG-induced 30HSD or P45017« mRNA levels. The role of protein synthesis in mediating the cAMP induction of 30HSD, P450i7o, and P450scC was investigated. The addition of cycloheximide (10 Mg/ml) to cAMPtreated cultures for 24 h completely suppressed both constitutive

B

IOSYNTHESIS of testosterone from cholesterol in Leydig cells involves the action of four enzymes. The first enzyme in this pathway, cholesterol side-chain cleavage (P4508CC), is found in the inner mitochondrial membrane and catalyzes the cleavage of the side-chain of cholesterol to yield the C21 steroid pregnenolone. Pregnenolone diffuses across the mitochondrial membrane and is further metabolized by enzymes associated with Received April 5,1991. Address all correspondence and requests for reprints to: Dr. Anita H. Payne, L1221/0278 Women's Hospital, University of Michigan, Ann Arbor, Michigan 48109-0278. * This work was supported by NIH Grants HD-08358, HD-17916, and P30-HD-18258 (Molecular Core).

and cAMP-induced 3/3HSD mRNA levels. Cycloheximide also repressed cAMP-induced levels of P450i7a to 12% of levels observed in the absence of cycloheximide. In sharp contrast, 24h treatment with cycloheximide did not suppress cAMP induction of P450scc mRNA, but reduced basal levels by approximately 50%. A time course of induction by cAMP (50 fiM) of P450i7a and P450KC mRNA showed very similar rates of increase in P450i7a and P450scc mRNA, with the greatest increase occurring between 12 and 24 h of treatment. The results of the study demonstrate that in normal mouse Leydig cells steady state levels of mRNA for 3/3HSD, P450j7a, and P450scC are differentially regulated. cAMP is required for maximal levels of all three mRNAs. There is high constitutive expression of 3j8HSD and P450scc mRNA, while expression of P450na mRNA is absolutely dependent on cAMP stimulation. Endogenously produced testosterone negatively regulates the expression of cAMP-induced P45017a and 3/3HSD, while the glucocorticoid dexamethasone negatively regulates 3/3HSD and P450scC. Newly synthesized protein (s) is required for cAMP induction of P45017a and 30HSD mRNA levels, but not for P4503CC mRNA. (Endocrinology 129: 1429-1435,1991)

the smooth endoplasmic reticulum. In the mouse Leydig cell, pregnenolone is first converted to progesterone by the action of 3/3-hydroxysteroid dehydrogenase/A5—>A4isomerase (3jftHSD). The next reaction catalyzed by the cytochrome P450 enzyme P450i7a involves 17a-hydroxylation of progesterone, followed by cleavage of the Cn_ 20 bond. This step reduces the number of carbon atoms from 21 to 19, yielding androstenedione, the immediate precursor of testosterone (1). The final reaction in the biosynthesis of testosterone is the reduction of the 17ketone by 17-ketosteroid reductase (2). This laboratory has been interested in the regulation of these enzymes in Leydig cells for many years. We previously reported that P4508CC protein and mRNA in cultured mouse Leydig

1429

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 21:36 For personal use only. No other uses without permission. . All rights reserved.

1430

STEROIDOGENIC ENZYME mRNAs IN LEYDIG CELLS

cells are expressed constitutively, but can be further increased by treatment of Leydig cells with 8-bromocAMP (cAMP) (3, 4). In contrast, cAMP is obligatory for de novo synthesis of P450i7a in normal mouse Leydig cells; P450i7a synthesis ceases in the absence of cAMP (3, 5). We also have reported that both P450scc and P450i7« syntheses in normal mouse Leydig cells are negatively regulated by steroid hormones. The glucocorticoids corticosterone, cortisol, or the synthetic glucocorticoid dexamethasone decrease both constitutive and cAMP-induced synthesis of P450scc (4). Testosterone or androgen agonists decrease the cAMP induction of P450i7a synthesis (5). In earlier studies on the regulation of 3/3HSD enzyme activity in cultured mouse Leydig cells, we did not observe an effect of cAMP on either the induction or maintenance of 3/3HSD enzyme activity (6). An in vivo study in hypophysectomized rats, however, indicated that LH was essential for the maintenance of testicular 3/3HSD activity (7). We have recently isolated cDNA clones for the mouse P450i7« (8) and 3/3HSD (8a). The availability of these mouse cDNAs has enabled us in this study to investigate the regulation by cAMP and steroids of steady state levels of P450i7a and 3/3HSD mRNAs in the same Leydig cell cultures. We have also investigated the role of protein synthesis in the cAMP induction of P4508CC, P450na, and 3/fflSD.

Materials and Methods Materials Medium 199, Dulbecco's Modified Eagle's Medium, Ham's nutrient mixture F-12, methionine- and glutamine-free Dulbecco's Modified Eagle's Medium, penicillin-streptomycin, and Lglutamine were obtained from Gibco (Grand Island, NY). 8Bromo-cAMP, HEPES, bovine insulin, testosterone, estradiol, dexamethasone, and cycloheximide were obtained from Sigma (St. Louis, MO). Nonidet P-40 was obtained from Fluka Chemical AG (Basel, Switzerland). Metrizamide was obtained from Accurate Chemical Co. (Westbury, NY). D,L-Aminoglutethimide was obtained from Aldrich (Milwaukee, WI). BSA (fraction V) was purchased from Armour Pharmaceuticals (North Chicago, IL), and collagenase (used at 180 U/ml medium) was purchased from Worthington Biochemical Corp. (Freehold, NJ). Mibolerone (7,17o:-dimethyl-19-nortestosterone) was obtained from Amersham Corp. (Arlington Heights, IL).

Endo• 1991 Vol 129 • No 3

was replaced after the first 24 h and subsequently every other day for 5 days. Treatments were initiated on day 6. Aminoglutethimide, steroids, and the steroid agonist mibolerone were added from methanolic stock solutions, and the final methanol concentration was 0.5% in all treatment media. Incorporation of [35S]methionine into total cellular proteins was carried out as previously described (5), except that cells were incubated in methionine-free medium for 1 h, then incubated with [35S]methionine (35 /iCi/ml) for 1 h. [35S]Methionine incorporation was measured in trichloroacetic acid-precipitated total protein at 6, 12, and 24 h in Leydig cells incubated in the absence or presence of cycloheximide (10 /ig/ml). RNA isolation and Northern analysis Cytoplasmic RNA was obtained by a modification of the method described by Pelham (9). Cultured cells were lysed in situ with a buffered 1% Nonidet P-40 solution. Nuclei were removed by centrifugation. Cytoplasmic protein was removed by extraction with phenol-chloroform, and total cytoplasmic RNA was obtained by ethanol-sodium acetate precipitation and quantitated by absorbance at 260 nm. RNAs from duplicate culture dishes were combined, and equal amounts of RNA from each treatment group were subjected to Northern analysis as described previously, except that blots were hybridized with the appropriate 32P-labeled cDNA probe (1 x 106 cpm/ml) for a minimum of 16 h (4). The following cDNA clones were used to prepare oligo-primed probes: a full-length 1.7-kilobase (kb) mouse P45017o cDNA (8), a 906-basepair mouse 3/3HSD cDNA fragment derived from the coding region of the cDNA (8a), a 1.8-kb mouse P4508CC cDNA (a gift from Drs. K. L. Parker, M. S. Kirkman, and D. A. Rice), and a full-length 2-kb chicken /3actin cDNA (10). These probes were labeled with [a-32P]dCTP to approximately 108 cpm//ug DNA, using the random primer synthesis method. Blots were washed twice at room temperature in 2 X SSC-0.1% sodium dodecyl sulfate for 15 min and then once at 60 C for 40-60 min before autoradiography. For sequential hybridization with different probes, membrane filters were washed in boiling 2 x SSC (1 x SSC = 0.15 M NaCl, 0.15 M Na citrate, pH 7.0)-0.1% sodium dodecyl sulfate for 510 min. Successful removal of probe was ascertained by autoradiography. The relative amounts of mRNA were quantitated by laser densitometry (Zenith, Soft Laser model SL-TRFF with a videophoresis II integrator, Biomed Instruments, Inc., Fullerton, CA).

Results cAMP induction and steroid repression of P450na and 3(3HSD mRNA levels

Isolation and culture of Leydig cells

Adult male CD-I mice were obtained from Charles River Co. (Wilmington, MA). Animals were killed by CO2 asphyxiation, and the testes were removed aseptically. All other procedures were carried out under sterile conditions. The methods for obtaining purified Leydig cells and culturing Leydig cells have been described previously (4, 5). Briefly, cells were plated at 11.5 X 106 cells/60-mm tissue culture dish and incubated in a humidified atmosphere of 95% air-5% CO2 at 32 C. Medium

The effects of increasing concentrations of cAMP in the absence and presence of aminoglutethimide on P450i7a and 3/3HSD mRNA levels were examined by Northern blot analysis. Figure 1 illustrates that in the absence of cAMP, P450i7a mRNA levels are undetectable, while 3/3HSD mRNA was expressed constitutively. Treatment of Leydig cell cultures with increasing concentrations of cAMP (50-500 IXM) markedly decreased



3J3HSD -

P450 t 7 a -

Actin AG

-

+

cAOiM)

-

-

50 100 500 50 100 500 50 50

Steroid TMIb FIG. 1. Effects of cAMP and androgens on 30HSD and P45017a mRNA levels. Mouse Leydig cells were incubated for 6 days before treatment for 24 h with increasing concentrations of cAMP (cA) in the presence or absence of 0.5 mM aminoglutethimide (AG) and, where indicated, 2 jiM mibolerone (Mib) or 2 /xM testosterone (T). Total cytoplasmic RNA was isolated as described in Materials and Methods, and 7 ng were subjected to Northern analysis and hybridized sequentially with 3/3HSD, P450i7a, and /S-actin cDNA probes. Only the hybridized regions of each blot are illustrated.

the level of 3/3HSD mRNA relative to that in untreated control cultures. The addition of aminoglutethimide, an inhibitor of cholesterol metabolism, together with cAMP resulted in increased 3/3HSD mRNA levels relative to those in untreated control and cAMP-treated cultures. Aminoglutethimide alone had no effect on 3/3HSD mRNA levels. Treatment with cAMP induced P450i7a mRNA at all concentrations used, with the greatest induction observed at a concentration of 50 nM. Higher concentrations of cAMP lead to lower levels of P450i7n mRNA. The addition of aminoglutethimide markedly increases the effect of cAMP on P450i7a mRNA levels. When each treatment group was corrected for the level of actin mRNA, induction of 3/3HSD mRNA in the presence of aminoglutethimide was maximal at a concentration of 50 ^M C A M P , with no observed decrease at higher concentrations (Fig. 1). The induction of P450i7o mRNA was maximal at 50 /xM, with a marked decrease at the higher concentration of 500 nM (40% of the levels observed at 50 or 100 fxM cAMP). The data shown in Fig. 1 suggest that steroids produced during cAMP induction of both 30HSD and P450 na mRNAs repress cAMP induction. Testosterone production was measured by RIA in five separate experiments by the method described previously (11). Testosterone production was 418 ± 3.6 ng/ml (mean ± SE) in Leydig cell cultures treated for 24 h with 50 /iM cAMP. This amount is equivalent to a concentration of 1.5 ^M. The addition of aminoglutethimide to the cAMP treatment reduced testosterone production to 23 ± 3.6 ng/ml (mean ± SE). TO test whether the increase in cAMP induction observed in the presence of aminoglutethimide could be reversed by the addition

1431

of exogenous testosterone or the androgen agonist mibolerone, these steroids were added to Leydig cell cultures treated with cAMP plus aminoglutethimide. As can be seen in Fig. 1, the addition of either testosterone or mibolerone at a concentration of 2 ^M resulted in a marked repression of cAMP induction of each of the mRNAs. Figure 2 summarizes the data obtained from four separate experiments analyzed by Northern blot analysis and corrected for actin levels. Figures 1 and 2 demonstrate that the presence of aminoglutethimide markedly increased the effect of cAMP on mRNA levels and that only in the absence of steroid hormone biosynthesis could a positive effect of cAMP on 3/3HSD mRNA levels be demonstrated. Effects of other steroids on cAMP induction of P450xla and 3/3HSD mRNA levels The effects of the synthetic glucocorticoid dexamethasone and of estradiol on cAMP-induced levels of mRNA was investigated. To test the specificity of each of these steroids on the cAMP induction, endogenous steroid production was inhibited by the addition of aminoglutethimide. Figure 3 presents a representative autoradiogram of a Northern blot, demonstrating that estradiol

AG CA Mib

FIG. 2. Effects of cAMP, endogenous steroids, and an androgen agonist on 3/SHSD and P450i7« mRNA levels. Mouse Leydig cells were incubated for 6 days before treatment for 24 h with 0.5 nM aminoglutethimide (AG), 50 >iM cAMP (cA), cA plus AG, or cA, AG, and 2 /xM mibolerone (Mib). Total cytoplasmic RNA was isolated as described in Materials and Methods, and 7 jug were subjected to Northern analysis and hybridized sequentially with 3j3HSD, P450i7a, and /3-actin cDNA probes. Specific mRNA was quantitated by densitometry, and the amount of specific mRNA for each treatment group was corrected for the amount of /3-actin in that treatment group. For P450i7a, all values are expressed relative to cA-treated samples, which were given the arbitrary value of 1. For 3/3HSD, all values are expressed relative to untreated samples, which were run in parallel with all the treatment groups. Each bar represents the mean ± SE from four separate identical experiments, except the value for AG only, which represents the mean ± SE from three separate experiments.



1432

Endo • 1991 Voll29«No3

levels. This increase, however, could not be observed consistently.

3pHSD

Role of protein synthesis in the expression of P450scc, P450lla> and 3(3HSD mRNA

P450, 7a -

Actin

AG cAbiM) -

+ -

25

+

+

+

+

+

+

50 25

-

50

50

50

50

50

Steroid Mib Dex E2 E2* FIG. 3. Effects of mibolerone, dexamethasone, and estradiol on 3j8HSD and P450i7a mRNA levels. Mouse Leydig cells were incubated and treated as described in Fig. 1. Mibolerone (Mib; 2 jiM), dexamethasone (Dex; 100 nM), estradiol (E2; 100 nM), and E2* (1 fiM) were added to cAMP- plus aminoglutethimide (AG)-treated cultures where indicated. Total cytoplasmic RNA was isolated as described in Materials and Methods, and 7 Mg were subjected to Northern analysis and hybridized, as described in Fig. 1.

-CHX

+ CHX

33HSD

P450scc-^ Actin Con cA Dex Con cA Dex FlG. 4. Effect of inhibition of protein synthesis on the expression of 3/8HSD, P45017a, and P4506CC mRNA. Leydig cells were incubated and treated as described in Fig. 1. Cycloheximide (CHX; 10 iig/ra.1) was added 30 min before the addition of 50 MM CAMP (CA) or 100 nM dexamethasone (Dex). CHX was added to untreated control cultures (Con) at the same time as it was added to the other cultures. After 24 h of treatment, total RNA was extracted and analyzed as described in Fig. 1. The filter was hybridized sequentially with 3/3HSD, P450i7a, P4508CC, and jS-actin cDNA probes.

at a concentration as high as 1 fiM did not repress cAMP induction of either 3/3HSD or P450 n a mRNA levels. The addition of the synthetic glucocorticoid dexamethasone caused a repression of cAMP-induced 3/?HSD mRNA to 55% of the levels observed for cAMP- plus aminoglutethimide-treated cultures. Dexamethasone also represses constitutive expression of 3/3HSD mRNA, as shown in Fig. 4 (leftpanel). Dexamethasone did not repress cAMPinduced P450i7a mRNA. The data in Fig. 3 indicate that treatment with dexamethasone increased P450i7a mRNA

To ascertain whether newly synthesized proteins are required for constitutive or cAMP-induced expression of mRNA for steroidogenic enzymes, Leydig cell cultures were incubated in the presence of an inhibitor of protein synthesis, cycloheximide. The addition of cycloheximide (10 jug/ml) 30 min before treatment inhibited total protein synthesis by 95%, as measured by [35S]methionine incorporation. Figure 4 illustrates the effect of the addition of cycloheximide on both constitutive and cAMPinduced expression of 3/3HSD, P450i7a, and P4508CC mRNA. During 24 h of treatment, cycloheximide completely suppressed both constitutive and cAMP-induced expression of 3/3HSD mRNA. Cycloheximide also suppressed cAMP induction of P450i7a. In sharp contrast, cycloheximide did not supress cAMP induction of P4508CC mRNA. Constitutive expression of P4508CC mRNA was reduced by cycloheximide treatment. These data indicate that cAMP induction of P450scc mRNA does not require newly synthesized protein factors, while such protein factors are required for cAMP induction of both 3/3HSD and P450i7a mRNA. Figure 4 also shows that dexamethasone represses constitutive expression of 3/3HSD mRNA by 53% and of P4508CC mRNA by 34%, as determined from densitometry readings corrected for actin mRNA levels. To obtain more information on the role of protein synthesis and the time involved for maximal cAMP induction of P450i7a and P4508CC mRNA, the effects of cycloheximide on basal and cAMP-stimulated mRNA levels were examined at 6, 12, and 24 h after treatment of cultures with cAMP. Cycloheximide was added to the cultures 30 min before initiation of treatment. Figure 5 illustrates that 1) cAMP induction of P4508CC mRNA is slow, with the greatest increase over basal levels occurring between 12 and 24 h (2.5- to 3-fold); 2) cAMPinduced increases in P4508Cc mRNA at 12 and 24 h are not prevented by the addition of cycloheximide, while basal expression is reduced by approximately 50%; and 3) inhibition of protein synthesis during the first 6 h of exposure to cycloheximide prevents the cAMP-induced increase in P450scc mRNA, but has no effect on basal expression. Figure 6 illustrates that the pattern of cAMP induction of P450i7a with time is very similar to that observed for the cAMP-stimulated increase in P4508CC mRNA (P450i7a mRNA in untreated Leydig cells is undetectable). cAMP-induced levels of P450i7« mRNA were the same at 6 and 12 h, with a marked increase between 12 and 24 h. cAMP induction of both P450i7a and P4508CC



1433

Discussion

CA Time (h)

- + - + 6

FIG. 5. Time course for action of cAMP and cycloheximide on P450SCC mRNA levels. Leydig cell cultures were incubated for 6, 12, and 24 h in the absence or presence of 50 jiM cAMP (cA) and in the absence or presence of cycloheximide (CHX; 10 fig/ml). CHX was added 30 min before the addition of cA. Total cytoplasmic RNA was isolated as described in Materials and Methods, subjected to Northern analysis, and hybridized sequentially with P450.cc and j3-actin cDNA probes. P450.cc mRNA and 0-actin mRNA were quantitated by laser densitometry, and the amount of P450KC in each treatment group was corrected for the amount of j8-actin in that treatment group. All values are expressed relative to the 24-h untreated control, which was given the arbitrary value of 1. The number in each bar is equal to the number of separate experiments. The value for three separate experiments represents the mean ± SE. The value for two separate experiments is the mean ± range.

FlG. 6. Time course for action of cAMP and cycloheximide on P450no mRNA levels. All conditions were identical to those described in Fig. 5. Filters were hybridized with P450no and j8-actin cDNA probes, and quantitated as described in Fig. 5. mRNA levels in cultures not treated with cAMP were undetectable and, therefore, are not illustrated. All values represent the mean ± range of two separate experiments and are expressed relative to the 6-h cA-treated in the absence of cycloheximide (-CHX) sample, which was given the arbitrary value of 1.

mRNA levels was maximal at 24 h. No further increase was observed when cultures were incubated for 48 h in the presence of cAMP (data not shown). Inhibition of protein synthesis markedly suppressed cAMP induction of P450i7n at all time intervals examined. Inhibition of protein synthesis for 24 h reduced cAMP-induced P450i7« mRNA levels to 12% of levels observed in the absence of cycloheximide.

The results of the current study demonstrate that in Leydig cells from normal adult mice, cAMP is essential for the expression of P450x7a mRNA, but not 3/3HSD mRNA. In fact, cAMP treatment of Leydig cells results in decreased levels of 3/3HSD mRNA. The addition of aminoglutethimide to c AMP-treated cultures, to inhibit endogenous steroid production, made it possible to demonstrate that cAMP does increase 3/3HSD mRNA levels as well as markedly enhances increases in P450i7a mRNA. The results observed with aminoglutethimide suggest that endogenously produced steroids repress cAMP induction of both P450i7a and 3/3HSD mRNA levels. This hypothesis was confirmed by demonstrating that the addition of exogenous testosterone or the androgen agonist mibolerone at a concentration of 2 ^M reduced P450i7« and 3/3HSD mRNA to levels comparable to those observed when cells were treated with cAMP only. We previously reported that testosterone or mibolerone inhibit cAMP-induced de novo synthesis of P450i7a by an androgen receptor-mediated mechanism (5). The results of the present study indicate that the repressive effect of testosterone occurs at the level of P450i7a mRNA and that testosterone represses not only P450i7a mRNA, but also 3/3HSD mRNA. Testosterone does not repress de novo synthesis of P4508CC in normal Leydig cells (4) or P4508CC mRNA in MA-10 tumor Leydig cells (12). Therefore, the repression of P450i7« and 3/3HSD mRNA is not a general effect on all steroidogenic enzymes in Leydig cells. We also could not demonstrate a repression of either P450i7a or 3/3HSD mRNA by estrogens. The addition of very high concentrations (up to 1 ixM) of estradiol did not reduce the levels of these mRNAs. We did observe a repression of 3j9HSD, but not of P450i7a, mRNA levels by the synthetic glucocorticoid dexamethasone. We previously reported that dexamethasone repressed P4508cc de novo synthesis and mRNA levels in normal Leydig cells (4), but had the opposite effect on MA-10 tumor Leydig cells (12). Although glucocorticoids are not produced in Leydig cells, glucocorticoid receptors have been demonstrated in interstitial cells of the rat testis (13). Glucocorticoids are mediators of stress. Increased production of glucocorticoids in pathological conditions of the adrenal cortex, such as Cushing's syndrome, can be associated with reproductive dysfunction, including decreased circulating testosterone levels (14). Our studies have identified two sites in the biosynthetic pathway of cholesterol to testosterone, P450scc and 3/3HSD, where increased circulating concentrations of glucocorticoids could act to decrease testicular testosterone production. The present study indicates that maximal induction by cAMP of P450i7a and 30HSD mRNA is achieved at a


1434


concentration of 50 /uM, and that a higher concentration (500 IXM) results in considerably less induction even in the presence of aminoglutethimide, especially in the case of P450i7a mRNA. We previously demonstrated that aminoglutethimide, even at concentrations of cAMP as high as 1 fiM, reduces testosterone production to basal levels in mouse Leydig cell cultures (15). The demonstration in the present study that at the high concentration of cAMP the addition of aminoglutethimide does not result in a marked increase in P450i7a mRNA levels, as was observed at the lower concentrations of cAMP (25-100 IXM), suggests the induction of a nonsteroidal factor (in addition to testosterone) by higher concentrations of cAMP that represses cAMP induction of P450na mRNA. The current study also demonstrates that newly synthesized protein(s) is essential for cAMP induction of P450i7a and 30HSD mRNA, but does not appear to be involved in cAMP-induced increases in P4508CC mRNA. The demonstration that cAMP induction of both 3/3HSD mRNA and P450i7a mRNA is dependent on newly synthesized protein mediators is in agreement with a recent report on cAMP induction of 3/3HSD and P450i7a mRNA in cultures of bovine adrenocortical cells (16). The role of protein synthesis in mediating cAMP induction of P450i7a and P4508CC in a variety of different steroidogenic cells has not been resolved. John et al. (17) and Zuber et al. (18) reported that cAMP-induced increases in P450scc and P450i7a mRNA in cultures of bovine adrenocortical cells was abolished by the addition of cycloheximide, indicating that newly synthesized proteins are involved in mediating the induction of both P450i7a and P450scc mRNA by cAMP. More recently, the same laboratory studied chimeric reporter gene constructs containing the 5' cAMP-responsive sequences of either the bovine P4508CC or the P450i7a gene, transiently transfected into bovine adrenocortical cells or mouse adrenal tumor Yl cells (19, 20). Inhibition of protein synthesis by cycloheximide did not reduce cAMP-induced expression of these transiently transfected reporter genes containing either the P450scc or P450i7a sequence. The reason for the discrepancy with the effect of cycloheximide on the expression of endogenous bovine genes and that of reporter genes is not obvious at this time. Other reports also do not show a consistent pattern regarding the role of newly synthesized proteins in mediating cAMP induction of P4508CC mRNA. Cycloheximide inhibition of cAMP-stimulated increases in P450scc mRNA was reported in the transformed human trophoblastic cell line JEG-3 (21), but was not observed for P450scc in cultures of human granulosa cells (22). We previously reported that cAMP- and dexamethasone-induced increases in P450scc mRNA were blocked by cycloheximide in MA-10 tumor Leydig cells (12). No effect of cycloheximide was

Endo • 1991 Voll29«No3

observed on constitutive levels of P4508CC mRNA in this Leydig cell tumor line. In another report, Mellon and Vaisse (23) found that cycloheximide did not inhibit cAMP-stimulated increases in P4508CC mRNA in the MA10 cells, but markedly decreased constitutive P4508CC mRNA levels. The different observations in these two studies could be due to differences in the experimental protocol. Mellon and Vaisse (23) used 20 /xg/ml cycloheximide and treated cells with cAMP for up to 8 h. In the studies reported from our laboratory, MA-10 cells were treated with 5 fig/ml cycloheximide and 10 /L*M cAMP for a total of 4.5 h (12). We have found in our studies that cycloheximide treatment at higher concentrations or for longer than 5 h was lethal to MA-10 cells (12). The results in the present study in normal mouse Leydig cells indicate that during the first 6 h of incubation, cycloheximide prevented the cAMP-induced increase in P450scc mRNA without affecting constitutive levels of mRNA. However, at 12 and 24 h, cycloheximide did not affect cAMP-induced increases in P4508CC mRNA, but reduced constitutive P450scc mRNA by approximately 50%. These results are consistent with our previous observation in MA-10 tumor Leydig cells, where at 4 h cycloheximide prevented cAMP-induced increases in mRNA without affecting constitutive mRNA levels. In the current study, although the effects of cycloheximide on cAMP-induced increases in P4508CC and P450i7a mRNA levels were different, the patterns of the cAMP induced-increases with time were similar for these two P450 mRNAs. cAMP-induced increases in both P450i7a and P450scc mRNA levels were most marked between 12 and 24 h. This observation indicates that maximal induction by cAMP takes several hours. This type of response is characteristic of genes whose induction by cAMP is mediated by newly synthesized proteins (24). The results with cycloheximide, however, indicate that cAMP-induced increases in P450scc mRNA do not require newly synthesized proteins, while cAMP-mediated induction of P450i7a is highly dependent on newly synthesized proteins. Taken together the data suggest that cAMP-induced increases in these two P450 mRNAs in normal mouse Leydig cells occur by a different mechanism (s). Our results also demonstrate that both constitutive and cAMP induction of 3/3HSD mRNA are highly dependent on the new synthesis of a short-lived protein factor(s). The role of protein mediators in constitutive as well as cAMP-induced increases in transcription of the genes that encode P450scc, P450i7a, and 3/3HSD requires further investigation. In summary, the present study clearly demonstrates that steady state levels of mRNA of three of the four enzymes involved in the biosynthesis of testosterone from cholesterol in mouse Leydig cells are differentially regulated. cAMP is required to achieve maximal levels



of all three mRNAs. However, there is high constitutive expression of both 3/3HSD and P450scc mRNA, while expression of P450i7a mRNA is absolutely dependent on chronic stimulation by cAMP. Endogenously produced testosterone negatively regulates expression of cAMPinduced P450i7« as well as 3j8HSD mRNA, while glucocorticoids repress P4508CC and 3/3HSD mRNA levels. The mechanism by which cAMP increases expression of the three enzymes involved in testosterone biosynthesis remains to be resolved.

References 1. Payne AH 1990 Hormonal regulation of cytochrome P450 enzymes, cholesterol side-chain cleavage and 17a-hydroxylase/Ci7-2o lyase in Leydig cells. Biol Reprod 42:399-404 2. Bogovich K, Payne AH 1980 Purification of rat testicular microsomal 17-ketosteroid reductase. Evidence that 17-ketosteroid reductase and 17jS-hydroxysteroid dehydrogenase are distinct enzymes. J Biol Chem 255:5552-5559 3. Anakwe 0 0 , Payne AH 1987 Noncoordinate regulation of de novo synthesis of cytochrome P450 cholesterol side-chain cleavage and cytochrome P450i7(,-hydroxylase/d7-2o lyase in mouse Leydig cell cultures: relation to steroid production. Mol Endocrinol 1:595-603 4. Hales DB, Payne AH 1989 Glucocorticoid-mediated repression of P450.cc mRNA and de novo synthesis in cultured Leydig cells. Endocrinology 124:2099-2104 5. Hales DB, Sha L, Payne AH 1987 Testosterone inhibits cAMPinduced de novo synthesis of Leydig cell cytochrome P450i7a by an androgen receptor-mediated mechanism. J Biol Chem 262:1120011206 6. Rani CSS, Payne AH 1986 Adenosine 3'5'-monophosphate-mediated induction of 17a-hydroxylase and Cn-20 lyase activities in cultured mouse Leydig cells is enhanced by inhibition of steroid biosynthesis. Endocrinology 118:1122-1128 7. Shaw MJ, Georgopoulos LE, Payne AH 1979 Synergistic effect of follicle-stimulating hormone and luteinizing hormone on testicular A5-3/9-hydroxysteroid dehydrogenase-isomerase: application of a new method for the separation of testicular compartments. Endocrinology 104:912-918 8. Youngblood GL, Sartorius C, Taylor BA, Payne AH 1991 Isolation, characterization, and chromosomal mapping of mouse P450i7ahydroxylase/Cn-20 lyase. Genomics 10:270-275 8a.Bain PA, Yoo M, Clarke T, Hammond SH, Payne AH 1991 Multiple forms of mouse 3/3-hydroxysteroid dehydrogenase/A5-A4 isomerase and differential expression in gonads, adrenal glands, liver and kidneys of both sexes. Proc Natl Acad Sci USA, in press 9. Pelham HRB 1982 A regulatory upstream promoter element in the Drosophila Hsp 70 heat-shock gene. Cell 30:517-528 10. Cleveland DW, Lopata MA, MacDonald RJ, Cowan NJ, Rutter

11. 12.

13. 14. 15. 16.

17. 18.

19.

20.

21.

22.

23. 24.

1435

WJ, Kirschner MW 1980 Number and evolution of alpha- and beta-tubulin and cytoplasmic beta- and gamma-actin genes using specific cloned cDNA probes. Cell 20:95-105 Hauger RL, Chen YD, Kelch RP, Payne AH 1977 Pituitary regulation of Leydig cell function in the adult male rat. J Endocrinol 74:57-66 Hales DB, Sha L, Payne AH 1990 Glucocorticoid and cyclic adenosine 3'5'-monophosphate-mediated induction of cholesterol sidechain cleavage cytochrome P450 (P450BCC) in MA-10 tumor Leydig cells: increases in mRNA are cycloheximide sensitive. Endocrinology 126:2800-2808 Evain D, Morera AM, Saez JM 1976 Glucocorticoid receptors in interstitial cells of the rat testis. J Steroid Biochem 7:1135-1139 Griffin JE, Wilson JD 1985 Disorders of the testes and male reproductive tract. In: Wilson JD, Foster DW (eds) Williams Textbook of Endocrinology, ed 7. Saunders, Philadelphia, p 284 Quinn PG, Payne AH 1985 Steroid product-induced oxygen-mediated damage of microsomal cytochrome P-450 enzymes in Leydig cell cultures. J Biol Chem 260:2092-2099 Naville D, Rainey WE, Milewich L, Mason JI 1991 Regulation of 3/3-hydroxysteroid dehydrogenase/A6—»4-isomerase expression by adrenocorticotropin in bovine adrenocortical cells. Endocrinology 128:139-145 John ME, John MC, Boggaram V, Simpson ER, Waterman MR 1986 Transcriptional regulation of steroid hydroxylase genes by ACTH. Proc Natl Acad Sci USA 83:4715-4719 Zuber MX, John ME, Okamura T, Simpson ER, Waterman MR 1986 Bovine adrenocortical cytochrome P450i7

Interleukin-1 inhibits cholesterol side-chain cleavage cytochrome P450 expression in primary cultures of Leydig cells.

Insulin-like growth factor type I increases concentrations of messenger ribonucleic acid encoding cytochrome P450 cholesterol side-chain cleavage enzyme in primary cultures of porcine granulosa cells.

Follicle-stimulating hormone increases concentrations of messenger ribonucleic acid encoding cytochrome P450 cholesterol side-chain cleavage enzyme in primary cultures of porcine granulosa cells.

Regulation of cytochrome P450 aromatase messenger ribonucleic acid and activity by steroids and gonadotropins in rat granulosa cells.

Cytochrome P450 1B1 and Primary Congenital Glaucoma.

Cholesterol side-chain cleavage, cytochrome P450, and iron-sulfur protein in human placental mitochondria.

Cytochrome P450 turnover.

Inherited congenital adrenal hyperplasia in the rabbit: absent cholesterol side-chain cleavage cytochrome P450 gene expression.

Cytochrome P450: advances and prospects.

Suicide inhibitors of cytochrome P450 1A1 and P450 2B1.

Glucocorticoid and cyclic adenosine 3'5'-monophosphate-mediated induction of cholesterol side-chain cleavage cytochrome P450 (P450scc) in MA-10 tumor Leydig cells. Increases in mRNA are cycloheximide sensitive.

Cytochrome P450: progress and predictions.

The cytochrome P450 gene superfamily.

Microsomal lipid peroxidation: the role of NADPH--cytochrome P450 reductase and cytochrome P450.

Dietary regulation of cytochrome P450.

A different cytochrome P450 form is induced in primary cultures of rat hepatocytes.

Optimization of yeast-expressed human liver cytochrome P450 3A4 catalytic activities by coexpressing NADPH-cytochrome P450 reductase and cytochrome b5.

Membrane anchor of cytochrome P450 reductase suppresses the uncoupling of cytochrome P450.

C17-20 lyase cytochrome P450 expression.

Cytochrome P450 arachidonic acid epoxygenase: stereochemical characterization of epoxyeicosatrienoic acids.

Spaceflight Effects on Cytochrome P450 Content in Mouse Liver.

Characterization of cytochrome P450 in extrahepatic tissues.

Cytochrome P450-Mediated Phytoremediation using Transgenic Plants: A Need for Engineered Cytochrome P450 Enzymes.

Characterization of human cytochrome P450 enzymes.