The Role of Extracellular Calcium in Corticotropin-stimulated Steroidogenesis*

The role of extracellular Ca2+ in the binding of corticotropin (ACTH) to adrenocortical cell receptors as well as in the post-binding events involved in steroidogenesis were investigated. Binding studies using ['"I-Tyrz3,Phez,Nle4]ACTH (1-38) peptide showed that ex- tracellular Ca2+ is essential not only for the interaction of ACTH with its receptor, but also for continued oc- cupancy of the receptor. In view of the requirement of Cas+ for binding the hormone to the receptor, the role of Ca2+ in post-receptor events was investigated by covalently attaching the hormone to its receptor by photoaffinity labeling in the presence of Ca2+. Persist- ent activation of steroidogenesis induced by photoaffinity labeling in the presence of Ca2+ was depressed when cells were incubated in medium containing EGTA but was unaffected when the cells were merely washed and incubated in Ca2+-free medium. In the presence of EGTA, 8-Br-CAMP partially restored persistent activation of steroidogenesis. The concentration of extracellular Ca2+ reJuired for restoring steroido- genesis was 10-fold lower than the concentration of Ca2+ needed for optimal binding of ACTH to its receptor. These results suggest that the primary role of extracellular Ca2+ in the action of ACTH is to facilitate the association of the hormone with its receptor.

The role of extracellular Ca2+ in the binding of corticotropin (ACTH) to adrenocortical cell receptors as well as in the post-binding events involved in steroidogenesis were investigated. Binding studies using ['"I-Tyrz3,Phez,Nle4]ACTH (1-38) peptide showed that extracellular Ca2+ is essential not only for the interaction of ACTH with its receptor, but also for continued occupancy of the receptor. In view of the requirement of Cas+ for binding the hormone to the receptor, the role of Ca2+ in post-receptor events was investigated by covalently attaching the hormone to its receptor by photoaffinity labeling in the presence of Ca2+. Persistent activation of steroidogenesis induced by photoaffinity labeling in the presence of Ca2+ was depressed when cells were incubated in medium containing EGTA but was unaffected when the cells were merely washed and incubated in Ca2+-free medium.
In the presence of EGTA, 8-Br-CAMP partially restored persistent activation of steroidogenesis. The concentration of extracellular Ca2+ reJuired for restoring steroidogenesis was 10-fold lower than the concentration of Ca2+ needed for optimal binding of ACTH to its receptor. These results suggest that the primary role of extracellular Ca2+ in the action of ACTH is to facilitate the association of the hormone with its receptor.
The importance of calcium in the stimulation of steroidogenesis in the adrenal cortex by ACTH' was first pointed out by Birmingham et al. (1). Several groups have shown, subsequently, that extracellular calcium is required for ACTHinduced steroidogenesis in isolated adrenocortical cells derived from several species (2-5). There have been numerous attempts to elucidate the precise role of the ion in the actions of ACTH. Lefkowitz et al. (6) investigated the binding of "' Ilabeled ACTH preparations to a subcellular fraction derived from a mouse adrenal tumor and proposed their results as *This work was supported by Grant CA-16417 awarded by the National Cancer Institute, Department of Health and Human Services, by Grant PCM-8118633 awarded by the National Science Foundation, and by scholarships to R. C. and D. I. B. from the Northern California Chapter of the Achievement Reward for College Scientists Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. evidence for the existence of two classes of ACTH receptors with vastly differing affinities and capacities. However, the existence of the very high affinity sites (apparent K d , 1.1 x M; 60 sites/cell) were not confirmed, and other investigators could detect only the low affinity sites (apparent K d , 3.3 X lo-' M; 360,000 sites/cell) in adrenal membrane preparations. Lefkowitz et al. (7) also concluded that calcium is not required for the interaction of ACTH with the adrenal receptor(s) (6, 7). This conclusion has been widely accepted, and most of the studies of the role of calcium have focused on events subsequent to the binding of the hormone to its receptor.
['251-Tyr?3,Phe2,Nle4]ACTH  peptide (referred to as "' 1-ACTH analog hereafter) was obtained in a homogeneous state by reverse phase HLPC and found to have a specific radioactivity of 1800 +-75 Ci/mmol (9). "'I-ACTH analog was found to be equipotent with ACTH in stimulating corticosterone production in isolated rat adrenocortical cells (8,9). Binding studies with lz5I-ACTH analog showed that the adrenocortical cells contain a single class of receptors with an apparent K d of 1.41 f 0.21 nM and a capacity of 3840 f 1045 sites/cell (10). Binding correlated very well with the ability to stimulate cAMP production, but maximal steroidogenesis was produced by the occupancy of only 3% of the receptors, indicating receptor reserve. The concentration-response curves for cAMP production for a series of ACTH analogs were found to be superimposable on the binding inhibition curves (10). In view of the large discrepancies between our results and those reported by Lef'kowitz et al. (6), we have reinvestigated the role of calcium in the actions of ACTH including the primary step of interaction with its receptor. In this study, we have utilized specific photoaffinity labeling of ACTH receptors (11) to distinguish between the role of calcium in the binding of the hormone to its receptor and the role of the ion in post-binding events.
EGTA and CaCI2 were obtained from J. T. Baker Chemical Co. 8-Br-cAMP was obtained from Sigma. Medium 199 and custom RPMI medium were provided by the Cell Culture Facility, University of California, San Francisco.
'=Z-ACTH Anabg-[PheZ,Nle']ACTH (1-38) peptide was synthesized by the solid-phase method and purified by partition chromatography as described (8). '=I-ACTH analog was prepared and separated from free iodide by gel filtration on Sephadex LH-20 as described (9). It was stored at 4 "C and freshly purified by reverse phase HPLC for each experiment (9). The HPLC solvents were removed by exchange on a Sephadex G-25 column (1 X 20 cm) equilibrated with medium 5323 199 containing 0.5% BSA and 0.01% bacitracin. The Sephadex column was treated with 0.1 mg of polylysine in the same solvent prior to exchanging the HPLC solvent. The specific radioactivity of the 'T-ACTH analog used in these studies was 1800 f 75 Ci/mmol (mean f S.E.). The steroidogenic potency of the "'1-ACTH analog was identical to that of ACTH (9).
Binding Studies-Adrenocortical cells were isolated from the decapsulated adrenal glands of adult male Sprague-Dawley rats (350-400 g) by digestion with collagenase and DNase as previously described (13). Cells were suspended in medium 199 containing 10% fetal bovine serum and 0.004% gentamycin (Schering Corp.) at a density of 7.5 X lo6 cells/ml and incubated overnight (18 h) in sterile bacterial Petri dishes (Falcon, no. 1029) at 37 "C in 5% co2/95% air. Cells were harvested as described (10) and resuspended in medium 199 containing 0.5% BSA and 0.01% percent bacitracin (4 X lo6 cells/ ml) and incubated at 24 "C with 200 pM '=I-ACTH analog in the presence or absence of excess unlabeled ACTH (440 nM) for 1 h in an atmosphere of 5% c02/95% 02. Cells were then processed by centrifugation through a cushion of medium 199 containing 1.5% BSA as previously described for determining cell-associated radioactivity (IO). Aliquots from each incubation were analyzed for corticosterone and cAMP by radioimmunoassay, and the DNA content of the cell pellet was measured by the method of Short et al. (14).
Photolysis of Adrenocortical Cells-Cells were plated at a density of lo6 cell/ml/well as described elsewhere (15) and used after 5 days in primary culture. Cells were washed (3 times) with medium 199/ 0.5% BSA/O.Ol% bacitracin and then incubated with or without 2,5-NAPS-ACTH (100 nM) for 15 min at 24 "C in the dark. Photolysis was then performed using a Blak-Ray lamp emitting principal radiation at 366 nm as previously described (11). Cells were washed 2 times after photolysis with the incubation medium and incubated with ACTH antiserum 46-2 (9) at a 1:20 dilution for 30 min to remove noncovalently bound peptide. Cells were then washed 2 times with the incubation medium with or without 3 mM EGTA and finally incubated in 0.25 ml of the same medium f 3 mM EGTA for 15-90 min at 37 "C. Aliquots of the media were removed at various times and analyzed for corticosterone and cAMP by specific radioimmunoassays.

RESULTS
"'1-ACTH analog was used in direct binding studies with isolated rat adrenocortical cells to elucidate the role of calcium Cu" (mM) in the interaction of ACTH with its physiological receptor. The results presented in Fig. 1 show that there is an absolute rquirement for extracellular calcium for both the binding of 1261-ACTH analog to adrenocortical cells and for stimulation of corticosterone production. Binding of the peptide and steroidogenesis which are maximal in medium 199 (burs in Fig.  1) are both abolished by the addition of 3 mM EGTA which is more than sufficient to chelate the calcium ions (1.8 mM) present in medium 199. Addition of 1 mM Ca2+ does not restore binding or function since EGTA is still in excess; however, when the Ca" concentration exceeded that of EGTA, both binding and steroidogenesis were restored in parallel. cAMP production in response to ACTH was also found to be dependent on extracellular Ca" as reported earlier (2-4). cAMP production in the experiment shown in Fig. 1 was 11.85 +: 0.11 pmol/h in medium 199 alone, <0.1 pmol/h in the presence of 3 mM EGTA, and 11.9 +: 0.4 pmol/h in the presence of 3 mM EGTA and 3 mM Ca".
Studies using calcium-free RPMI medium confirmed the above findings. No significant binding of '251-ACTH analog was observed in the absence of Ca", and addition of Ca" resulted in a parallel increase in binding as well as corticosterone production (Table   TABLE I   In order to ascertain whether extracellular Ca2+ is required for continued occupancy of the receptor, the dissociation of lZ5I-ACTH analog from adrenocortical cells was investigated in the presence and absence of Ca2+. The results in Fig. 2 show that removal of calcium from the medium dramatically accelerates the dissociation of ACTH from the receptor. Halfmaximal dissociation of lZ5I-ACTH analog occurred in 32 min in the presence of Ca2+ and 3.5 min in the absence of the ion. The absolute requirement of extracellular Ca2+ for the binding and continued occupancy of the receptor makes it difficult to study the effect of Ca2+ on events subsequent to the hormone-receptor interaction. We solved this problem by covalent attachment of ACTH to the receptor by photoaffinity labeling (11). We have shown previously that photolysis of rat adrenocortical cells in the presence of the photoreactive derivative 2,5-NAPS-ACTH (12) results in the persistent activation of both corticosterone and cAMP production. This activation persists for several hours and cannot be abolished by extensive washing with anti-ACTH antibodies. Fig. 3 shows the effect of removal of extracellular Ca2+ on the persistent activation of steroidogenesis. Cells photolyzed in medium with Ca2+ produced corticosterone at rates comparable to those observed in the presence of added ACTH, the first period were now incubated in the absence of the chelating agent. Aliquots were removed at the times shown during this second incubation and assayed for corticosterone. Values are means of triplicate incubations.
whereas cells photolyzed in the absence of Ca" and then incubated in the presence of Ca2+ produce steroids at rates no greater than what is seen with unstimulated control cells.
In order to assess the role of Ca2+ in post-binding events, cells were photolyzed in medium 199 in the presence of i,5-NAPS-ACTH and then incubated in the presence or absence of EGTA (Fig. 4). It is clear that the rate of corticosterone production is considerably diminished in the presence of EGTA (Fig. 4, leftpanel). After 90 min, the incubation media were reversed. Cells incubated in medium 199 alone were now incubated in the presence of EGTA and cells previously kept in the presence of EGTA were now incubated in medium 199 alone. It is apparent that the rate of steroid production in the former is now attenuated and the rate of steroidogenesis in the latter is restored to normal (Fig. 4, right panel).
Th concentration of extracellular Ca2+ necessary to restore persistent activation of steroidogenesis was investigated. The results in Fig. 5 show that as little as 0.18 mM Ca" is sufficient to restore the steroidogenic response elicited by covalent attachment of ACTH to adrenocortical cells. This is 10-fold lower than the concentration of Ca2+ needed for maximal binding of the lZ5I-ACTH analog. Since EGTA tends to deplete the medium as well as the cells of Ca", the effect of washing persistently activated cells in Ca2+-free medium alone was investigated. Cells washed with Ca2+-free RPMI medium after photoaffinity labeling with 2,5-NAPS-ACTH and incubated in Ca2+-free medium remained activated and continued steroid production in the absence of extracellular Ca2+ (Fig.  6).
Since cAMP is known to stimulate steroidogenesis in the adrenocortical cells in the absence of ACTH, we investigated whether cAMP could restore steroidogenesis in persistently activated cells in the absence of Caz+. It is apparent from Fig.   7 that exogenous cAMP can at least partially restore the steroidogenic response to ACTH which was abolished by EGTA.

DISCUSSION
The role of Ca2+ in the actions of ACTH has been the subject of numerous investigations over the past 25 years (summarized in Refs. 16 and 17). Haksar and Piron (3) found that the requirement for Caz+ in ACTH action on isolated rat adrenocortical cells was greater for events preceding the formation of CAMP than for those that follow. However, because of the studies of Lefkowitz et al. (7) it was generally accepted that Caz+ was not required for the interaction of ACTH with the adrenal receptors. All previous studies attempting to elucidate the role of Ca" in ACTH binding utilized lZ5I-ACTH preparations of low biological potencies resulting from the preferential introduction of the bulky iodine atom into the tyrosine residue in position 2 of the ACTH amino acid sequence. Such radiolabeled preparations were detecting mostly low affinity sites of high binding capacity, not the physiologically relevant high affinity low capacity receptors. By the use of a synthetic analog of ACTH containing a phenylalanine residue in place of the tyrosine in position 2, we were able to prepare a homogenous radioligand with full biological potency and near-theoretical specific radioactivity (8,9). This fully active radioligand was employed successfully to identify and characterize corticotropin receptors on rat (10) and human (18) adrenocortical cells and rat adipocytes (19) as well as 3T3-Ll cells (20). The results obtained using this fully active radioligand clearly show that the primary role of extracellular Ca2+ is at the first step, namely binding of the hormone to its receptor on the adrenocortical cell surface. Ca2+ is required not only for the association of ACTH with its receptor but also for continued occupancy of the receptor.
Whereas the actions of ACTH on adrenocortical cells have generally been studied in media containing millimolar con-centrations of Ca", all studies with adrenal membrane preparations in the past have been performed in the absence of added Ca". This was necessitated, in part, by the finding that Ca2+ in millimolar concentrations inhibited adenylate cyclase activity (7, 21, 22). The concentration of ACTH required for half-maximal stimulation of adenylate cyclase activity in rat and bovine adrenal membrane preparations was 2-3 orders of magnitude higher than that required with intact cells. In the light of our results on the requirement of Ca" for the binding of the hormone to the receptor, the inability of ACTH to activate adenylate cyclase at low concentrations in membrane preparations in the absence of Ca2+ is understandable. We have found that Ca2+ is required for specific binding of "'1-ACTH analog to bovine (23) and rat' adrenal membranes. The conflicting requirements of the presence of physiological concentrations of Ca2+ for binding of the hormone to the receptor and the absence of such concentrations of Caz+ for optimal adenylate cyclase activity make it virtually impossible to measure binding of ACTH and activation of adenylate cyclase activity in adrenal membrane preparations under the same conditions?
In view of the stringent requirement of Ca" for binding ACTH to its adrenal receptor, we resorted to covalent attachment of the hormone to its receptor in order to evaluate the role of extracellular Ca2+ in post-binding events. No covalent attachment of ACTH to the receptor occurred when photolysis was conducted in the presence of EGTA, supporting the conclusion derived from the binding studies that CaZ+ is necessary for the interaction of the hormone with the adrenocortical cell receptor. Once the hormone was covalently linked to the receptor in the presence of Ca2+, the role of the ion in post-binding events could be studied. Although persistent activation of steroidogenesis induced by photoaffinity M. Tsubokawa and J. Ramachandran, unpublished results. labeling was suppressed in the presence of EGTA, steroidogenesis continued unabated when the persistently activated cells were washed with Ca2+-free medium and reincubated in Ca2+-free medium. These results suggest that extracellular Ca2+ is not needed for steroidogenesis once the hormone is bound to the receptor and is able to maintain continued occupancy due to covalent attachment. Influx of Ca2+ from the extracellular medium under the influence of ACTH has been considered as a possible mechanism involved in steroidogenesis (24). Our results clearly show that such influx of Ca2+ from the extracellular fluid is not necessary for ACTH-induced steroidogenesis.
Minute quantities of Ca2+ bound to the membrane may, of course, play an important role in the events beyond the binding of the hormone to its receptor. Depletion of Ca2+ by incubation with excess EGTA is clearly deleterious and causes reversible cessation of steroidogenesis even when the hormone is covalently attached to the receptor. It is possible that Ca2+ is needed to maintain the hormone in a conformation favorable for productive interaction with the receptor even when covalently bound to its receptor. Photoaffinity labeling of Xenopus melanophores with a photoreactive derivative of amelanocyte-stimulating hormone has shown (25) that Ca" is required both for binding of the hormone to its receptor and for one or more intracellular events involved in melanin dispersion. Since the amino acid sequence of a-melanocytestimulating hormone is present in the first 13 residues of ACTH, it is tempting to speculate that Ca2+ may be involved in stabilizing a conformation of the peptide segment common to the two hormones.
Ca2+ may play a role at a site beyond the binding of the hormone to the receptor in addition to its effect on the binding step. One possible locus of Ca2+ action may be the membranebound adenylate cyclase. Several laboratories have found that steroidogenesis and cAMP formation in response to ACTH were impaired when Ca2+ was omitted from the extracellular medium (21, 26, 27). However, cAMP was able to at least partially overcome the consequences of the lack of Ca2+. This is in agreement with our results and suggests an effect of Ca2+ on adenylate cyclase activity. Several groups have suggested that Ca2+ exerts its stimulatory effect at the level of the interaction of the guanine nucleotide-binding protein with the enzyme (28,29). The concentration of free Caz+ required for the stimulation of adenylate cyclase appears to be in the 0.1-1.0 p~ range (30).
It is also conceivable that translocation of Ca2+ bound to the plasma membrane or other organelles of the adrenocortical cell to another compartment in response to the interaction of ACTH with its receptor may play a role in the steroidogenic action of the hormone. Studies using inhibitors of the actions of calmodulin have shown that Ca2+ plays an important remlatorv role at intracellular sites (31). It is clear, however. sive role of extracellular Ca2+ in the action of ACTH is to facilitate the binding of the hormone to its receptor.