Induction of synthesis of mitochondrial steroidogenic enzymes of bovine adrenocortical cells by analogs of cyclic AMP.

The long term action of cyclic AMP analogs to stimulate the synthesis of cytochromes P-450scc, P-45011 beta, and adrenodoxin has been studied utilizing confluent monolayers of adult bovine adrenocortical cells maintained for periods of time up to 72 h in the absence or presence of dibutyryl cyclic AMP (1 mM), 8-bromo cyclic AMP (1 mM), or ACTH (adrenocorticotropin) (10(-6) M). The synthesis of these proteins was examined by radiolabeling cellular proteins with [35S]methionine or else by translating RNA extracted from such cells in a cell-free system in the presence of [35S]methionine. In each case, the protein under study was immunoprecipitated utilizing specific antisera, or IgG fractions prepared from such antisera. ACTH and both analogs of cyclic AMP caused an increase in the synthesis of cytochrome P-450scc which reached a maximum 36-48 h after addition, and then declined. On the other hand, butyric acid (1 mM) had no effect on the synthesis of cytochrome P-450scc. Cytochrome P-450scc activity measured as pregnenolone production by both intact cells or isolated mitochondria from such cells was increased following incubation of cells with either dibutyryl cyclic AMP or ACTH. The binding of rabbit anti-cytochrome P-450scc IgG was also increased in cells incubated with dibutyryl cyclic AMP or ACTH as estimated by immunofluorescence microscopy using fluorescein-tagged anti-rabbit IgG. Furthermore, dibutyryl cyclic AMP and ACTH both increased the synthesis of adrenodoxin and of cytochrome P-45011 beta, as well as the activity of 11 beta-hydroxylase. In addition, ACTH stimulated the secretion of cyclic AMP in a time- and concentration-dependent fashion. Thus, it is concluded that analogs of cyclic AMP can mimic the long term actions of ACTH to induce the synthesis of steroidogenic enzymes, and that this action of ACTH is likely mediated by cyclic AMP.

( M). The synthesis of these proteins was examined by radiolabeling cellular proteins with ["5S]methionine or else by translating RNA extracted from such cells in a cell-free system in the presence of [35S]methionine. In each case, the protein under study was immunoprecipitated utilizing specific antisera, or IgG fractions prepared from such antisera. ACTH and both analogs of cyclic AMP caused an increase in the synthesis of cytochrome P-450.,, which reached a maximum 36- 4 8 h after addition, and then declined. On the other hand, butyric acid (1 mM) had no effect on the synthesis of cytochrome P-450,,. Cytochrome P-450,, activity measured as pregnenolone production by both intact cells or isolated mitochondria from such cells was increased following incubation of cells with either dibutyryl cyclic AMP or ACTH. The binding of rabbit anticytochrome P-450,, IgG was also increased in cells incubated with dibutyryl cyclic AMP or ACTH as estimated by immunofluorescence microscopy using fluorescein-tagged anti-rabbit IgG. Furthermore, dibutyryl cyclic AMP and ACTH both increased the synthesis of adrenodoxin and of cytochrome P-450110, as well as the activity of 118-hydroxylase. In addition, ACTH stimulated the secretion of cyclic AMP in a time-and concentration-dependent fashion. Thus, it is concluded that analogs of cyclic AMP can mimic the long term actions of ACTH to induce the synthesis of steroidogenic enzymes, and that this action of ACTH is likely mediated by cyclic AMP.
The actions of ACTH' to stimulate steroidogenesis in the adrenal cortex can be divided into two categories, short term (acute) actions and long term (chronic) actions. The acute response of the adrenocortical cell to ACTH is believed to involve an increase in the availability of cholesterol to serve as precursor for pregnenolone formation by the mitochondrial cholesterol side chain cleavage system (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11). The long term action of ACTH is related to the synthesis of specific enzymes involved in steroidogenesis. For example, the levels of various adrenal steroidogenic enzymes including mitochondrial and microsomal forms of cytochrome P-450 decline over a period of several days following hypophysectomy of rats (12)(13)(14)(15)). On administration of ACTH to such animals the levels of these enzymes increase toward normal values. In other studies, ACTH has been shown to exert long term actions to increase the activity of Ilp-hydroxylase (16,17) and 3P-hydroxysteroid dehydrogenase-isomerase (18) as well as the synthesis of adrenodoxin (19) in mouse Y-1 adrenal tumor cells.
Recently we have utilized bovine adrenocortical cells in primary monolayer culture to study the synthesis of a number of key enzymes involved in steroidogenesis and the regulation of such synthesis by ACTH. Utilizing cells radiolabeled with ["S]methionine we have shown that the synthesis of the mitochondrial enzymes, cytochrome P-450,,, (20), cytochrome P-45011@ (21), and adrenodoxin (22) is induced in a coordinate fashion, maximal synthesis being obtained after approximately 36 h of exposure to ACTH. Subsequent to this time the rate of synthesis of these enzymes declines in the continuing presence of ACTH, suggestive that the cells become refractory to ACTH. The synthesis of the microsomal enzyme, 21-hydroxylase cytochrome P-450 (steroid 21-hydroxylase), is also induced when the cells are maintained in the presence of ACTH (23), but the time course of induction of this form of cytochrome P-450 differs somewhat from that of the mitochondrial steroidogenic enzymes. Utilizing a rabbit reticulocyte in vitro translation system, and [3sS]methionine as the radiolabeled amino acid, we have shown that RNA extracted from cells maintained in the presence of ACTH directs severalfold greater synthesis of these various proteins involved in steroidogenesis than does RNA from control cells (20)(21)(22)(23)(24). These results suggest that the increase in the synthesis of steroidogenic enzymes in response to ACTH results in part from an increase in transcription of specific mRNA species. Thus, we believe that long term action of ACTH involves the The abbreviations and trivial names used are: ACTH, adrenocorticotropin; P-450,,, side chain cleavage cytochrome P-450; P-45OI1,, lip-hydroxylase cytochrome P-450; Bt2-CAMP, dibutyryl cyclic AMP; 8-Br-cAMP, 8-bromo cyclic AMP; trilostane, (2~,401,5a,17@)-4,5-epoxy-17-hydroxy-3-oxoandrostane-2-carbonitrile; PBS, phosphate-buffered saline; HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid DMEM, Dulbecco's modified Eagle's medium. maintenance of synthesis of specific mitochondrial and microsomal enzymes involved in steroidogenesis.
Ample evidence is available to suggest that the acute action of ACTH is mediated by cyclic AMP (25-27). ACTH treatment has been shown to increase cyclic AMP accumulation by rat and bovine adrenal cells (28)(29)(30)(31) as well as by mouse adrenal tumor cells (32)(33)(34). Moreover, addition of cyclic AMP analogs such as Bt,-cAMP or 8-Br-CAMP can mimic the acute action of ACTH to stimulate steroidogenesis (28)(29)(30)32). Evidence has been presented that ACTH causes an increase in the amount of cyclic AMP bound to the regulatory unit of cyclic AMP-dependent protein kinase at all concentrations which give rise to an increase in steroidogenesis (35). The role of cyclic AMP as a mediator of the long term effects of ACTH to maintain the synthesis of steroidogenic enzymes has not been studied and thus, it is not known whether this action of ACTH is mediated by cyclic AMP or whether some other second messenger might be involved. Evidence that cyclic AMP mediates induction of synthesis of enzymes by other hormones has been presented. For example, it has been established that synthesis of tyrosine aminotransferase in rat liver is stimulated by CAMP analogs (36). It is also known that the action of glucagon to induce transcription of mRNA species specific for phosphoenolpyruvate carboxykinase in liver can be mimicked by cyclic AMP analogs such as Bt2-cAMP (37).
In order to determine whether the action of ACTH to induce the synthesis of mitochondrial steroidogenic enzymes in bovine adrenocortical cells may be mediated by cyclic AMP, we have studied the ability of analogs of this cyclic nucleotide to mimic the inductive effects of ACTH, utilizing bovine adrenocortical cells in monolayer culture as a model system. Our results indicate that cyclic AMP analogs can indeed mimic the effect of ACTH to induce the synthesis of cytochrome P-450,,,, cytochrome P-450,,,,, and adrenodoxin.

EXPERIMENTAL PROCEDURES
Cell Culture-Isolated cells prepared from the zona fasciculatareticularis of freshly obtained bovine adrenal glands by collagenase digestion and mechanical dispersion were grown in monolayer culture as described previously (20)(21)(22)(23)(24). Cells were maintained in culture in a mixture of Ham's F-12 medium and DMEM containing fibroblast growth factor, fetal calf serum, horse serum, and antibiotics/antimycotics. Upon achievement of confluence, the cells were incubated for an additional 24-48 h in Ham's F-lL/DMEM medium, but in the absence of fibroblast growth factor and antibiotics/antimycotics. At the beginning of an experiment, incubation medium was again replaced with fibroblast growth factor-free and antibiotic/antimycoticfree Ham's F-l2/DMEM medium. Synthetic ACTH,-,, (Cortrosyn), Bt2-CAMP, or 8-Br-CAMP were added to the incubation media of the appropriate culture dishes. Thereafter, incubation media were replaced every 24 h. Experiments were continued for 72 h. Cell monolayers were harvested a t 12-h intervals to assay enzyme synthesis and activity.
The cortisol contents of the incubation media were measured by direct radioimmunoassay (38) using an antiserum prepared against cortisol-3-carboxymethyloxime-bovine serum albumin. All statistical analyses were by Analysis of Variance, and significantly different means were identified by the method of Least Significant Difference.
Culture media, sera, and antibiotic/antimycotic solutions were obtained from Grand Island Biological Co. (Grand Island, NY). Plastic culture dishes and labware were purchased from Falcon Labware (Oxnard, CA), and collagenase and DNase were from Boehringer Mannheim. ACTH and fibroblast growth factor were obtained from Organon Diagnostics (West Orange, NJ) and Collaborative Research Inc. (Waltham, MA), respectively.
Enzyme Synthesis-Total cell protein was radiolabeled by incubating the cell monolayers for 2 h in the presence of [Y3]methionine (60 gCi/ml) following a 2-h preincubation in methionine-free Ham's F-l2/DMEM medium. Next, monolayers were rinsed and then suspended in Gey's balanced salt solution and washed by centrifugation to remove excess [35S]methionine. An aliquot of the cell pellet was removed for protein measurement (39), and the remainder was stored at -80 "C. When all samples for an experiment were collected, cells were lysed in PBS containing cholate (1%) and sodium dodecyl sulfate (0.1 %) (20). Nonradiolabeled cells were also harvested at the indicated times, stored a t -80 "C, and used for isolation of total cellular RNA. Cellular RNA was isolated by phenol extraction and centrifugation (100,000 x g for 18 h) through cesium chloride (5.7 M) as previously described (20). The RNA was used to program a rabbit reticulocyte lysate translation system in which [%]methionine (2 gCi/pl) was used as a radiolabel. Newly synthesized, radiolabeled cytochrome P-450,,, cytochrome P-45011p, and adrenodoxin were immunoisolated both from radiolabeled cell lysates and total cell-free translation products using Staphylococcus aureus cell membranes as the source of protein A (20). Cytochrome P-450118 was isolated using an antiserum, and cytochrome P-450,,, and adrenodoxin were isolated using immunoglobulin fractions of the appropriate antisera prepared against the corresponding proteins purified from bovine adrenocortical mitochondria. The specificities of the antibodies were determined by double immunodiffusion techniques and the abilities of the purified proteins to compete with the newly synthesized, radiolabeled proteins, and have been described elsewhere (20)(21)(22). Immunoisolates were subjected to electrophoresis on sodium dodecyl sulfate-gradient (7.5-12.5%) polyacrylamide gels in the case of cytochromes P-450,, and P-45011,, and on sodium dodecyl sulfate-15% polyacrylamide gels in the case of adrenodoxin. Gels were dried and subjected to autoradiography by standard techniques, and the autoradiograms were quantified by scanning densitometry.
The rabbit reticulocyte lysate translation system and [35S]methionine were purchased from New England Nuclear. Formalin fixed S. aureus cell membranes (Pansorbin) were from Calbiochem-Behring. Purified cytochrome P-450.,, and adrenodoxin were obtained through the courtesy of Dr. David Lambeth (Emory University, Atlanta, GA).
Purified cytochrome P-45Ollp and anti-serum directed against llghydroxylase were obtained through the courtesy of Dr. Colin Jefcoate (University of Wisconsin, Madison, WI).
Enzyme Actiuities-Pregnenolone formation in bovine adrenocortical cells was assayed in the presence of SU10603 (7-chloro-3,4dihydro-2-(3-pyridyl)naphthalen-l-(2H)one) (5 g~; an inhibitor of 17a-hydroxylase, Ciba-Geigy, Summit, NJ) and trilostane (25 g~; an inhibitor of 30-hydroxysteroid dehydrogenase, Sterling-Winthrop Institute, Rensselaer, NY). Pregnenolone formation was also assayed in mitochondria isolated from cell monolayers maintained in 100-mm dishes. Cells from a single dish were harvested in 2 ml of Gey's balanced salt solution, centrifuged a t 100 X g for 5 min, resuspended in 0.3 ml of sucrose/HEPES buffer (0.25 M, 25 mM, pH 7.4), and then homogenized. The cell homogenate was centrifuged a t 2000 X g for 10 min at 4 "C in a Beckman Microfuge 12. The supernatant was removed, and the pellet was homogenized and centrifuged as above. The pellet was again homogenized and centrifuged, and the supernatant was combined with those from the previous homogenizations and centrifuged a t 12,000 X g for 10 min a t 4 "C. The supernatant was removed, and the mitochondrial pellet was suspended in 1.0 ml of sucrose/HEPES buffer and again centrifuged at 12,000 X g. The final mitochondrial pellet was resuspended in 0.25 ml of sucrose/ HEPES buffer. An aliquot (0.05 ml) was removed for protein measurement, and the remainder (0.2 ml) was added to 2.0 ml of sucrose/ HEPES buffer (50 mM, 25 mM, pH 7.4) containing NaCl (80 mM), KC1 (5 mM), MgClz (5 mM), EDTA (1 mM), and bovine serum albumin (0.5%). Mitochondria were incubated a t 37 "C for 5 min in air. An aliquot (0.5 ml) of the reaction mixture was then removed, and the reaction was initiated by the addition of sodium malate (20 mM). Other aliquots (0.5 ml) of the reaction mixture were removed after 1, 3, and 7 min of incubation. Pregnenolone was determined in these aliquots as described previously (20).
110-Hydroxylase activity was assayed as the rate of conversion of exogenous 11-deoxycorticosterone to corticosterone by intact cells incubated in the presence of aminoglutethimide (21).
Fluorescence Microscopy-Bovine adrenocortical cells were cultured onto glass cover slips (20 X 20 mm) and grown under the same conditions as cells for biochemical studies. Cells were prepared for fluorescent labeling by fixation with 3% formaldehyde in Gey's balanced salt solution (Grand Island Biological Co.) for 15 min, followed by extraction for 4-6 min with -20 "C acetone. After fixation and extraction, cells were incubated with a solution of 100 pg/ml of rabbit preimmune IgG and 1% bovine serum albumin in PBS for 30 min in a humidified 37 "C incubator. Following three 5-min washings in by guest on March 22, 2020 http://www.jbc.org/ Downloaded from PBS, the cells were incubated with rabbit anti-cytochrome P-450,, (100 pg/ml of PBS) for 30 min at 37 "C, and washed three times in PBS. The cells were then labeled for 30 min a t 37 "C with fluoresceinconjugated goat anti-rabbit IgG (Miles Research Division, Elkhart, ID), washed, and observed using a Zeiss Universal microscope adapted for fluorescence studies. Photographs were taken a t 30-s exposure using Kcdak Tri-X Pan film.
Cyclic AMP Determination-Confluent adrenocortical cells were incubated a t 37 "C in 1 ml of Ham's F-12 medium/DMEM (1:l) that contained horse serum (12.5%), fetal calf serum (2.5%), 3-isobutyl-lmethylxanthine (1 mg/ml, Sigma), and ACTH M) for 5-120 min. After incubation, media were collected for determination of extracellular cyclic AMP. An equal volume of perchloric acid (0.6 N ) was added to the medium which was allowed to stand for 5 min a t 4 "C. After centrifugation a t 1000 X g for 15 min a t 4 "C, the supernatant was removed and the pH adjusted to pH 6 with KOH (6 N). The samples were incubated for 30 min at 4 "C and centrifuged again at 1000 X g for 30 min. The supernatant was removed, dried, and reconstituted in distilled water. Cyclic AMP was determined by radioimmunoassay using a specific antiserum against cyclic AMP which showed a cross-reaction of 4 X with cyclic GMP, 2.6 X lo-''% with AMP, and 5.9 X lo-:'% with ATP. 2-O-'2sII-Succinyl cyclic AMP tyrosine methyl ester was used as tracer. The assay was performed in acetate buffer (0.05 M, pH 6.2). The antigen-antibody complex was precipitated by addition of goat-anti rabbit IgG followed by centrifugation at 1000 X g for 15 min. All reagents for the cyclic AMP radioimmunoassay were obtained from Becton-Dickinson Immunodiagnostics (Orangeburg, New York). The radioactivity in the pellet was determined using a y-counter.

RESULTS
Formation of Cyclic AMP-Since the purpose of the present study was to investigate whether or not the long term actions of ACTH are mediated by cyclic AMP, it was necessary to establish that ACTH could stimulate cyclic AMP formation by bovine adrenocortical cells prepared and cultured in this laboratory. Confluent cells were incubated with ACTH in various concentrations in the presence of 3-isobutyl-1-methylxanthine, as described under "Experimental Procedures." ACTH caused an increase in the accumulation of cyclic AMP in the medium which was both time-and concentrationdependent (Fig. 1). After 15 min of exposure to ACTH, the concentration of ACTH required to cause half-maximal stimulation of cyclic AMP formation was about 5 X 10"O M, a value similar to the apparent Kd of ACTH binding to its receptor in rat adrenal cells (40), as we11 as the concentration of ACTH required for half-maximal stimulation of cyclic AMP formation reported by others (41) cyclic AMP production in response to various concentrations of ACTH in a separate experiment is shown in Fig. 1B. As indicated in this figure, variability exists in the absolute amount of cyclic AMP produced per unit of time from one cell preparation to another.
Actiuity of Cytochrome P-45OsCc-1n order to investigate the effect of Bt2-CAMP on cholesterol side chain cleavage activity, bovine adrenocortical cells were maintained for various periods of time in the absence or presence of Bt2-CAMP (1 mM), ACTH ( M), or ACTH plus Bt2-CAMP. Medium was then replaced with fresh medium containing SU10603 and trilostane. Incubation was continued for an additional 30 min and pregnenolone in the medium was assayed by radioimmunoassay. Alternatively, after the cells were incubated for various periods of time in the presence or absence of Bt,-CAMP or ACTH, the medium was removed, mitochondria were prepared from cellular homogenates, and pregnenolone formation was assayed as described under "Experimental Procedures." The results of these experiments are shown in Fig. 2. Pregnenolone formation by control cells was relatively unchanged throughout the 60-h experiment (Fig. 2 A ) , while exposure of the cells to Bt2-CAMP resulted in a stimulation of pregnenolone formation. Stimulation was maximal at 36 h and approximately %fold greater than control values. Thereafter, the rate of pregnenolone formation declined even in the continued presence of Bt2-CAMP. Virtually identical profiles of pregnenolone formation were observed in cells maintained in the presence of ACTH or in the presence of ACTH together with When pregnenolone production was assayed in mitochondria prepared from cells incubated in the presence or absence of Bt2-CAMP or ACTH, the pattern of induction was similar to that observed in intact cells (Fig. 2B). Bt,-CAMP caused an increase in activity which reached a maximum by 36 h, and subsequently declined. Similar results were observed using mitochondria prepared from cells incubated in the presence of either ACTH or ACTH plus Bt,-CAMP.

Bta-cAMP (1 mM) or ACTH (lo-' M).
In the case of the experiment described in A, at the end of the culture period, the medium was replaced with medium containing trilostane (25 p~) and SU10603 (5 p~) .
The cells were incubated a t 37 "C for 30 min and pregnenolone in the medium was determined. In the case of the experiment described in E, at the end of the culture period the cells were harvested, mitochondria were prepared, and cholesterol side chain cleavage activity determined as described under "Experimental Procedures." Pregnenolone content of the media was measured by direct radioim-

AMP
Synthesis of Cytochrome P-45Ox,,."In order to investigate whether or not the increase in cholesterol side chain cleavage activity induced by Bt,-CAMP resulted from an increase in the synthesis of cytochrome P-450,,, cells were incubated for various periods of time up to 72 h in the presence or absence of Bt,-CAMP or ACTH (Fig. 3A) or in the presence or absence of 8-Br-CAMP or ACTH (B). Cell proteins were then radiolabeled with [''"Slmethionine and the synthesis of cytochrome P-450,,, was determined as described above. In the absence of inducing agents, the incorporation of radiolabel into cytochrome P-450,,, changed little throughout the course of the experiment. Addition of sodium butyrate (1 mM) had little or no effect on the incorporation of radiolabel into cytochrome P-450,,, (Fig. 3A). By  tained in the presence of ACTH together with the respective cyclic AMP analog, the effects were not additive. In the experiment presented in Fig. 3B, maximal incorporation of radiolabel was observed a t 36 h. Subsequently, the incorporation of radiolabel declined in a manner similar to that previously reported for ACTH (20). In the case of the experiment described in Fig. 3A, the peak of induction did not occur until 48 h. Nevertheless, subsequent to this time, the rate of synthesis of cytochrome P-450,, declined even in the continued presence of inducing agents. This shift of 12 h in the development of refractoriness reflects of the variability in responsiveness of different preparations of primary cell cultures.
T o determine whether or not the action of Bt,-CAMP to increase the synthesis of cytochrome P-450,, might be due to an increase in the translatability of mRNA species specific for this protein, RNA was extracted from cells incubated in the presence or absence of Bt2-CAMP, ACTH, and Bt2-CAMP plus ACTH for 36 h. The RNA was translated in uitro in the presence of [:''S]methionine, and the radiolabeled cytochrome P-450,, precursor was immunoprecipitated. Results of such an experiment are shown in Fig. 4. Bt,-CAMP increased the in uitro synthesis of cytochrome P-450.,, approximately 5fold. ACTH was equally effective in stimulating cytochrome P-450,, synthesis, as was ACTH plus Bt2-CAMP.
Immunofluorescent Staining of Cytochrome P-45OS,,-Effects of ACTH (1 PM) and Bt2-CAMP (1 mM) on the content of cytochrome P-450,, within the adrenocortical cell were assayed by immunofluorescence 36 h after addition of inducer. Immunofluorescent staining for cytochrome P-450,, was 10cated in the mitochondria (Fig. 5). Although mitochondria were distributed throughout the cell, they were often concentrated in the region of the nucleus. More than 90% of cells maintained in the presence of ACTH ( B ) or Bt2-CAMP ( C ) demonstrated immunofluorescent staining for cytochrome P- 450,,, that was greater than the control level (A). In contrast, immunofluorescent staining for citrate synthase, a nonsteroidogenic mitochondrial enzyme, was similar in control cells and cells treated with ACTH or Bt2-CAMP. Consistent with these findings, ACTH did not increase the activity of citrate synthase (data not shown).
Synthesis of Adrenodoxin-To examine whether cAMP analogs were effective to induce the synthesis of another mitochondrial protein involved in the cholesterol side chain cleavage reaction, the ability of Bt,-CAMP to induce the synthesis of adrenodoxin was investigated. Cells were maintained for 36 h in the presence or absence of Bt2-cAMP, ACTH, or Bt,-cAMP plus ACTH, and synthesis of adrenodoxin was examined as described above. The results of such an experiment are shown in Fig. 6. Bt2-CAMP stimulated the rate of synthesis of adrenodoxin in a fashion similar to the stimulation of cytochrome P-450,,, synthesis. ACTH also stimulated the incorporation of radiolabel into adrenodoxin, but had no additional effect when added in combination with Bt2-CAMP.
Activity of Cytochrome P-4501 I,r-llp-Hydroxylase activity was assayed in bovine adrenocortical cells maintained for 36 h in the presence or absence of Bt,-cAMP, ACTH, or Bt2-cAMP plus ACTH. The rate of llp-hydroxylase activity was increased in the presence of inducing agents (Fig. 7A). When Bt,-cAMP was added in the presence of ACTH, it had no additional effect.
Synthesis of Cytochrome P-45011,,-The effect of dibutyryl cyclic AMP on the synthesis of lip-hydroxylase was determined in an in uitro translation system directed by total RNA from cells maintained for 36 h in the presence or absence of Bt2-CAMP, ACTH, or Bt,-CAMP plus ACTH. RNA isolated from Bt2-CAMP and ACTH-treated cells directed the synthesis of cytochrome P-45011,r a t approximately three times the control level (Fig. 7B). Nonetheless, the combined effects of Bt,-CAMP and ACTH were not greater than that of either inducing agent alone.

DISCUSSION
A considerable body of evidence suggests that cyclic AMP is the sole mediator of the acute action of ACTH to stimulate steroid hormone biosynthesis in the adrenal cortex (35). At present, the only evidence which runs contrary to this proposition is derived from studies of the use of the ACTH analog, N-o-nitrophenylsulfenyl-ACTH, which can cause a maximal stimulation of steroidogenesis in rat adrenal cells without any detectable change in the levels of cyclic AMP (41). However, the question as to whether or not cyclic AMP mediates the long term action of ACTH to maintain the synthesis of steroidogenic enzymes has not been addressed previously. In the present investigation it has been shown that the cyclic AMP analog Bt2-CAMP can mimic the action of ACTH to stimulate cholesterol side chain cleavage activity both in whole cells and isolated mitochondria. This analog also can mimic the action of ACTH to increase the synthesis of cytochrome P-450,,, both in intact cells and in a cell-free system programmed with RNA extracted from cells maintained in the presence of Bt2-CAMP, an effect reflected in an increase in immunofluorescent staining for cytochrome P-450,,,.
The time course of action of Bt2-CAMP was very similar to that of ACTH. Moreover, after a maximal rate of synthesis had been achieved, the rate of synthesis then declined even in the continued presence of the nucleotide analog. This onset of refractoriness is similar to that observed in the continued presence of ACTH. Similarly, the actions of another cyclic AMP analog, 8-Br-cAMP, led to induction of the synthesis of cytochrome P-450,,,. In this case, the increase in synthesis at 36 h was less than that observed with ACTH, although the rate of synthesis did decline at longer times of 8-Br-CAMP treatment. We have observed also a refractoriness of steroid output by cells in monolayer culture in the continued presence of Bt,-CAMP (42). These results indicate that the refractoriness of the synthesis of mitochondrial components of adrenocortical steroid hydroxylases to continued ACTH stimulation results from an effect distal to the increase in CAMP levels. A similar conclusion has been reached from the study of steroidogenesis in cultured pig Leydig cells (43).
Bt,-CAMP is generally believed to enter the cell and be hydrolyzed to form N-monobutyryl cyclic AMP and butyric acid (44). In some cells, butyric acid has the capacity to alter cellular function and thus it is necessary to rule out the possibility that the effects of Bt2-cAMP might be mediated by the formation of butyrate. As shown by the data presented here, sodium butyrate had no effect on the induction of synthesis of cytochrome P-450,,,. This fact together with the observation that treatment with 8-Br-CAMP also results in increased synthesis, leads to the suggestion that these compounds are acting to induce the synthesis of steroidogenic enzymes by virtue of their structural analogies with cyclic AMP itself. In addition to induction of the synthesis of cytochrome P-450,,,, Bt2-CAMP was also efficacious in inducing the synthesis of cytochrome P-45OIl8 and adrenodoxin in a fashion similar to the action of ACTH.
We conclude from the studies presented here that the action of ACTH to induce coordinately the synthesis of cytochromes P-450,,, and P-45OIl8, as well as adrenodoxin (20-22) is mediated via cyclic AMP. The action of the analogs of cyclic AMP to induce synthesis of these enzymes is not precisely the same as ACTH with respect to the magnitude of the effect, but the temporal profiles of induction are similar in both cases. Although not proven, the action of ACTH to induce the synthesis of these proteins is likely due to an increase in the synthesis of mRNA sequences coding for these proteins. In hepatocytes, Bt2-CAMP has been shown to induce the transcription of mRNA sequences specific for the enzymes tyrosine aminotransferase (36) and phosphoenolpyruvate carboxykinase (37). So it is not unreasonable to postulate that cyclic AMP might act in an analogous fashion in the adrenal to initiate the transcription of mRNA sequences specific for steroidogenic enzymes. An important question which remains relates to the identity of the factors which mediate the action of cyclic AMP to initiate transcription of specific mRNA species. It seems likely that cyclic AMP-dependent protein kinase might play an important role in this induction mechanism, but the nature of the specific proteins which are phosphorylated as a result of the activation of protein kinase by cyclic AMP remains to be defined, as does the nature of the steps subsequent to the action of protein kinase which lead to increased synthesis of steroidogenic enzymes.