The Role of Ca2+ and Cyclic Adenosine 3’:5’-Monophosphate in Insulin Release Induced in Vitro by the Divalent Cation Ionophore A23187*

SUMMARY Insulin release from isolated perifused pancreatic islets was stimulated by the divalent ionophore A23187 in the absence of exogenous glucose. In addition, A23187 produced a 2-fold elevation of cyclic adenosine d’:S’-monophosphate (CAMP) levels in isolated perifused islets. The elevation of CAMP levels coincided with peak insulin release. Ionophore-induced insulin release was unaffected by pretreatment of the islets with theophylline (5 mM). Stimulation of insulin release produced by the ionophore occurred either in the presence or absence of extracellular Ca2f; however, CAMP accumula-tion required the presence of extracellular Ca2+. The ionophore (10 pM) had no effect on adenylate cyclase activity of homogenates of isolated islets. The results of this study are interpreted as indicating that intracellular Ca*+ has an essential role in the insulin releasing mechanism, whereas the CAMP system has a modulatory effect on this process.

by the divalent ionophore A23187 in the absence of exogenous glucose. In addition, A23187 produced a 2-fold elevation of cyclic adenosine d':S'-monophosphate (CAMP) levels in isolated perifused islets. The elevation of CAMP levels coincided with peak insulin release. Ionophoreinduced insulin release was unaffected by pretreatment of the islets with theophylline (5 mM). Stimulation of insulin release produced by the ionophore occurred either in the presence or absence of extracellular Ca2f; however, CAMP accumulation required the presence of extracellular Ca2+. The ionophore (10 pM) had no effect on adenylate cyclase activity of homogenates of isolated islets. The results of this study are interpreted as indicating that intracellular Ca*+ has an essential role in the insulin releasing mechanism, whereas the CAMP system has a modulatory effect on this process.
Cyclic adenosine 3' : 5'.monophosphate and calcium (Cazf) have been implicated as intracellular mediators of glucose-induced insulin release from pancreatic p cells (1). Several observations support this hypothesis: (a) hormones and transmitters which are known to alter adenylate cyclase activity in homogenates of islet cells or to alter CAMP' levels in isolated islets are potent modulators of insulin release (2, 3) ; (5) phosphodiesterase inhibitors, such as theophylline, potentiate insulin release simulated by various agents (e.g. glucose or tolbutamide) pancreatic islets (5); and (d) extracellular Ca2+ is absolutely necessary for insulin release stimulated by glucose and other agents, a requirement shared by some other CAMP-dependent processes (6).
Nonetheless, the respective roles of Ca2+ and CAMP or possible interrelationships between the two substances in the process of insulin release, particularly in release induced by the major physiological stimulant glucose, are not clear. For example, 10 mM theophylline alone is capable of augmenting islet CAMP levels 4-fold yet, in the absence of glucose, does not result in an increased rate of insulin secretion (4,7).
We have sought new approaches to examine more closely the possible interrelated roles of Ca2+ and CAMP in insulin release. In the present study, the effects of a divalent cation ionophore, A23187, on insulin release and the behavior of CAMP levels in isolated perifused rat islets were examined. This ionophore, obtained from streptomyces, stimulates secretory responses in a number of tissues and it is thought that this occurs as a result of altered transmembranous Ca2+ fluxes (S-lo), leading to elevation of intracellular Ca*+ levels.

Perfusion
Studies-Islets from fed adult male Sprague-Dawley rats (Holtzman, Madison, Wise.) were isolated according to the method of Lacy and Kostianovsky (11). For studying the kinetics of insulin release and of CAMP level changes in islets, a perifusion system similar to that first employed by Burr (12) and modified by Lacy et al. (13) was used. Batches of 100 to 200 islets were used in each experiment. A 30-min preperifusion period was followed by stimulation with test substances for various time periods, as evident from the tables and figures under "Results." For tissue analysis, islets, still attached to the Millipore filter, were removed quickly from the chamber, submerged for 30 s in Freon-la, which had been cooled to its freezing point (-X0"), and then placed in cooled glass jars for storage at -80". CAMP was extracted in 15oj, trichloroacetic acid (0.5 ml/batch of 100 to 200 islets). The acid extract was washed four times with 6 volumes of hydrated ethyl ether, and the acid-free solution then was dried with a stream of nitrogen. The solid residue was dissolved in 125 ~1 of 50 mM sodium acetate buffer, pH 6.0. The tissue extracts were assayed (in triplicate) for CAMP using a modification of the radioimmunoassay of Steiner et al. (14).
The perifusate was composed of 0.5ye crystalline bovine serum albumin in a salt solution buffered with bicarbonate (pH 7.4) and continuously gassed with a mixture of 95% 02 and.5% COZ: The salt composition was: NaCl, 115 mM; KCl, 5 mM; NaHC03,24 mM; CaCl.. 2.1 mM: and M&12. 1.0 mM. The flow rate was maintained -I close (flO%) to 1 mlimin. Islets were preperifused in the absence of extracellular glucose for 30 min and then exposed to the ionophore. Control islets were exposed to the solvent (0.25a/, acetone plus 1% ethanol).
Both groups subsequently were exposed to high glucose. as that observed when islets were challenged with high concentrations of glucose. The peak rate of insulin release produced by 10 pM of the ionophore was twice the basal rate. In comparison, first phase insulin release produced by 27.5 mM glucose was 3 times the basal rate. After the ionophore (or the solvent in the controls) had been removed from the perifusion medium, the islets retained their full capacity to respond in characteristic biphasic fashion to high glucose concentrations. Thus, it is apparent that the ionophore did not produce irreversible damage to the insulin releasing mechanism.
Dose-response studies with the ionophore revealed a steep curve (Figs. 2 and 3). The threshold level was 1 PM and maximal insulin release was achieved by 10 PM. After exposure to a higher concentration of the ionophore (25 PM), the first phase of the insulin release in response to the subsequent addition of 27.5 mM glucose was blunted, suggesting that this high concentration of ionophore may be deleterious.
The role of extracellular Ca2+ on ionophore-stimulated insulin release was examined. Preliminary experiments (not shown) indicated that the endocrine response to ionophore was equal with and without added calcium in the perifusate. More convincingly, experiments carried out in EGTA-Ca2+ buffers clearly demonstrated that the response to the ionophore was elicited in the complete absence of Ca2+ in the perfusate (Fig. 4). However, the peak of insulin release was delayed by 3 to 4 min and was more pronounced. When 27.5 mM glucose and 2.1 rnhf Ca2+ were then reintroduced, both experimentals and controls responded. However, the islets previously perifused by Ca2+-free medium lacked the first phase response.
Since other studies have revealed significant intracellular calcium stores (17), the response to the ionophore was ascribed to a failure to deplete these stores. Since theophylline has been shown to increase Cazf efflux from pancreatic islets (18), we used pretreatment with this agent in an attempt to deplete intracellular Ca2+ stores and to demonstrate a calcium dependency for the action of the ionophore. Therefore, islets were preperifused in the absence of Ca2+ for 30 min in the presence of EGTA and 5 mM theophylline. This pretreatment did not eliminate the secretory response to the ionophore (also not shown).

Efect of Ad3187 on CAMP Levels of Islet Tissue
Cytosolic calcium has been thought of as an inhibitor of adenylate cyclase (1). As predicted, A23187 has been reported to decrease intracellular CAMP levels in the presence of extracellular Ca2+ in other tissues (8). Since the precise roles of CAMP and Cazf in insulin release are not clear and since the roles of the two factors might be interdependent, we examined the effect of the ionophore on islet CAMP levels as a function of the extracellular Ca2+ concentration.
A nearly 2-fold increase in CAMP levels of islets was found 3 min after exposure to the ionophore in the presence of 2.1 mM Ca 2+ (Table I). This interval was chosen for tissue sampling since it coincided with maximal insulin release following introduction of the ionophore (Fig. I). In agreement with others, we previously have described elevation of CAMP levels in islets stimulated by a high concentration of glucose (20,21). In addition, we have demonstrated that such an effect is dependent upon extracellular Ca2+, as indeed is insulin release itself in response to high concentrations of glucose (21). Ca2+, although release of insulin still occurred and was even more pronounced than with Ca2+ present (Table I and Fig. 4).
As previous studies have demonstrated that ethanol is capable of stimulating adenylate cyclase activity (19), the effect of the solvent used here for the ionophore (0.257, acetone plus 1% ethanol) upon CAMP levels in islets was examined. The solvent   (6) 7.19 f 0.59 (7) 6.13 f 0.45 ( a The values for insulin recorded here were obtained 1 min prior to sampling the islets and therefore do not represent the peak rate of hormone release but indicate that the islets responded to the ionophore as predicted (compare the results in the corresponding figures).
* Significantly different from corresponding solvent controls, p < 0.002.  (6) alone had no effect on CAMP levels, as indicated in the legend to Table I. Lack of glucose in the perfusate did cause an almost 3-fold elevation of the base-line levels of CAMP, as is apparent when the present data are compared with previously published values from this laboratory (20). Average control values in our hands in the presence of 2.75 mM glucose has been 1.6 pmol/lOO islets.
There was a decrease in the base-line levels of CAMP (-360/,, p < 0.01) in the solvent controls during perfusion in Ca*+-free medium.
The mechanism of ionophore-stimulated increases in CAMP in the presence of Ca2+ is unclear. A priori. one might postulate that the ionophore itself, or changes in Ca2+ fluxes caused by the ionophore might either stimulate adenylate cyclase or inhibit phosphodiesterase.
To investigate those possibilities, the following experiments were designed. Experiment I-The possible effect of the ionophore and the solvent on adenylate cyclase activity was explored in the absence of added Ca2+ in rat islet homogenates. Neither the ionophore nor the solvent (0.257, acetone plus 1% alcohol) significantly altered adenylate cyclase activity in rat islet homogenates (Table II). These experiments were carried out in the absence of added Ca2+ and the presence of 1 mM EGTA. It would be desirable to perform similar studies as a function of the Cat+ concentration in the assay media. Since, however, physiological levels FIG. 5. Insulin release due to A23187 in the presence of theophylline. Theophylline (5 mM) was present throughout the entire duration of the experiment. The ionophore (10 PM) or the solvent were added after 30 min of preperifusion.
The means of three experiments are given. The S.E. values are included at essential points of the profiles.
(i.e. extracellular Ca*+ concentration) of Ca2+ (2.1 mM) greatly suppress adenylate cyclase activity (more than 80%) and since we were unable to perform activity measurements at a fixed low CaZf level (e.g. 10m5 M) because of the substantial but variable Ca2+ contamination from the islet tissue homogenates themselves, such studies have not been performed to date. Experiment !&The possible involvement of phosphodiesterase was examined with the aid of theophylline by measuring the effect of this agent in the perifusion system upon insulin release and CAMP levels in response to the ionophore. If elevations of CAMP and insulin release were due to an action of the ionophore on phosphodiesterase, blocking that enzyme with theophylline should obviate these responses to the ionophore. It was shown that the insulin release stimulated by the ionophore is not modified in the presence of 5 mM theophylline (Fig. 5). The levels of CAMP were elevated in theophylline-treated control islets, indicating that phosphodiesterase activity was inhibited and the ionophore caused no further increase of CAMP levels in these tissues (Table I). Furthermore.
it was noted that theophylline exposure in the solvent controls increased CAMP levels without a significant increase in insulin secretion (Table I, compare a and c). In the course of these studies it was also observed that the increase of CAMP elicited by 5 mM theophylline in the absence of extracellular glucose occurred independent of extracellular Ca2+ (6.42 f 0.25 and 6.30 f 0.40 pmol of cAMP/lOO islets following theophylline in the presence and absence, respectively, of Ca2+ as compared to 4.01 =t 0.33 pmol/lOO islets in controls).

DISCUSSION
The divalent cation ionophore A23187 proved to be an effective stimulator of insulin release in the absence of extracellular glucose. Islets exposed to the ionophore exhibited a brisk monophasic release of insulin. and the same islets subsequently challenged by high glucose responded with typical biphasic insulin release. While it has been postulated that the ionophore's action on many systems is mediated through augmented Ca2+ influx across the cell membranes (8)) we were unable to show a dependency on extracellular calcium in perifused islets. In the absence of extracellular calcium and in the presence of 10 PM of the ionophore, insulin release occurred, although somewhat delayed and magnified as compared to the experiments with Ca2+ present. Three possible explanations for this Ca2+-independent effect deserve to be considered. First, intracellular stores of calcium (e.g. mitochondria), which almost certainly are mobilized by the agent (lo), are sufficiently high to allow insulin release in the absence of extracellular calcium. The delay of the response may be an expression of the involvement of such an intracellular Ca2+ pool. Second, since the ionophore binds and transp0rt.s magnesium as well as calcium (lo), the observed effects may be due to altered magnesium distribution and due to secondary effects on the numerous Mg2+-dependent processes of the cell. A possible involvement of magnesium cannot be ruled out currently. Third, the ionophore's action may be independent of the translocation of divalent cations. Detailed studies on fluxes of Ca2+ and Mg2+ might clarify this point.
CAMP plays a prominent role as modulator of insulin discharge, perhaps mediating the effects of various hormones and transmitters.
Several groups of investigators recently have demonstrated a rapid rise of tissue levels of this nucleotide after the exposure of islets to high concentrations of glucose (1,(20)(21)(22)(23). These observations may imply that CAMP is also the second messenger for glucose-stimulated insulin release. Zawalich et al. (21) and Charles et al. (24) have documented the calcium dependency of this CAMP rise. The present studies have shown that the ionophore also elevates CAMP levels in isolated islets in a Ca2+-dependent manner. This contrasts with the findings of Prince et al. (8) who reported that CAMP levels decreased in fly salivary gland on stimulation of secretion by the ionophore. In the presence of theophylline, CAMP levels of islets were elevated as predicted, but when the ionophore was superimposed, no further rise of the cyclic nucleotide occurred. Nevertheless. neither the kinetics nor the magnitude of ionophore-induced hormone release were modified. This lack of potentiation of the insulin response argues against a primary involvement of the CAMP system in the ionophore induced insulin secretion. Indeed, islets perifused without extracellular calcium showed no rise of CAMP due to the ionophore, while stimulation of insulin secretion was preserved.
Based on experimental evidence presented, it is suggested that insulin release due to the ionophore results from mobilization of calcium from either extracellular or intracellular sources. Mobilization of Ca2+ by the ionophore has two effects: first, insulin release occurs, and second, the CAMP level of islets increases. However, the two events may not be coupled since, in islets exposed to the ionophore in the absence of extracellular calcium, insulin release occurred but the CAMP level (as measured in these experiments) remained constant.
It is tempting to speculate that altered Ca2+ conductivity or binding induced either by glucose (25) or by the ionophore specifically at the site of the plasma membrane might lead to an activation of adenylate cyclase in the plasma membrane. Mobilization of Ca2f from intracellular stores (e.g. mitochondria), although capable of causing insulin release, lacks this effect on adenylate cyclase. Thus, enhanced insulin release is a result of elevated intracellular Ca2f levels whatever the source of Ca2+ might be, whereas adenylate cyclase activation depends on a circumscribed change of Ca2+ distribution within or across the cell membrane. Such a concept would explain why to date all attempts to demonstrate an effect of glucose on adenylate cyclase in cell free systems have been unsuccessful.