Differential effects of Ca2+-calmodulin on adenylate cyclase activity cyclase activity in mouse and rat pancreatic islets

Long-term exposure to genistein improves insulin secretory function of pancreatic beta-cells

We recently found that genistein, a plant-derived natural compound, is a novel cAMP signaling agonist in pancreatic beta-cells. In the present study, we further show that exposure of clonal insulin secreting (INS-1E) cells to genistein for 48 h enhanced glucose-stimulated insulin secretion (GSIS), whereas insulin content was not altered, suggesting that genistein-enhanced GSIS is not due to a modulation of insulin synthesis. This genistein effect is protein tyrosine kinase- and K(ATP) channel-independent. In addition, genistein had no effect on glucose transporter-2 expression or cellular ATP production, but similarly augmented pyruvate-stimulated insulin secretion in INS-1E cells, indicating that the improvement of insulin secretory function by long-term genistein exposure is not related to an alternation in glucose uptake or the glycolytic pathway. The enhanced insulin secretion by genistein was associated with elevated intracellular Ca(2+) concentration and dependent on protein kinase A and new protein synthesis as this effect was completely blocked by N-[2-(p-Bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide or cycloheximide. Similarly, 48 h of genistein exposure also enhanced GSIS in freshly isolated mouse and human pancreatic islets, suggesting a non-species-specific and biologically relevant effect. These findings provide evidence that genistein may be a novel bioactive compound that has an anti-diabetic effect by improving insulin secretion from pancreatic beta-cells.

Genistein acutely stimulates insulin secretion in pancreatic beta-cells through a cAMP-dependent protein kinase pathway

Although genistein, a soy isoflavone, has beneficial effects on various tissues, it is unclear whether it plays a role in physiological insulin secretion. Here, we present evidence that genistein increases rapid glucose-stimulated insulin secretion (GSIS) in both insulin-secreting cell lines (INS-1 and MIN6) and mouse pancreatic islets. Genistein elicited a significant effect at a concentration as low as 10 nmol/l with a maximal effect at 5 micromol/l. The effect of genistein on GSIS was not dependent on estrogen receptor and also not related to an inhibition of protein tyrosine kinase (PTK). Consistent with its effect on GSIS, genistein increases intracellular cAMP and activates protein kinase A (PKA) in both cell lines and the islets by a mechanism that does not involve estrogen receptor or PTK. The induced cAMP by genistein, at physiological concentrations, may result primarily from enhanced adenylate cyclase activity. Pharmacological or molecular intervention of PKA activation indicated that the insulinotropic effect of genistein is primarily mediated through PKA. These findings demonstrated that genistein directly acts on pancreatic beta-cells, leading to activation of the cAMP/PKA signaling cascade to exert an insulinotropic effect, thereby providing a novel role of soy isoflavones in the regulation of insulin secretion.

Glucose triggers protein kinase A-dependent insulin secretion in mouse pancreatic islets through activation of the K+ATP channel-dependent pathway

OBJECTIVE

To assess the significance of protein kinase A (PKA) in glucose triggering of ATP-sensitive K(+) (K(+)(ATP)) channel-dependent insulin secretion and in glucose amplification of K(+)(ATP) channel-independent insulin secretion.

METHODS

Insulin release from cultured perifused mouse pancreatic islets was determined by radioimmunoassay.

RESULTS

In islets cultured at 5.5 mmol/l glucose, and then perifused in physiological Krebs-Ringer medium, the PKA inhibitors, H89 (10 micromol/l) and PKI 6-22 amide (30 micromol/l) did not inhibit glucose (16.7 mmol/l)-induced insulin secretion, but inhibited stimulation by the adenylyl cyclase activator, forskolin (10 micromol/l). In the presence of 60 mmol/l K(+) and 250 micromol/l diazoxide, which stimulates maximum Ca(2+) influx independently of K(+)(ATP) channels, H89 (10 micromol/l) inhibited Ca(2+)-evoked insulin secretion, but failed to prevent glucose amplification of K(+)(ATP) channel-independent insulin secretion. In the presence of 1 mmol/l ouabain and 250 micromol/l diazoxide, which cause modest Ca(2+) influx, glucose amplification of K(+)(ATP) channel-independent insulin secretion was observed without concomitant Ca(2+) stimulation of PKA activity. In islets cultured at 16.7 mmol/l glucose, glucose (16.7 mmol/l)-induced insulin secretion in physiological Krebs-Ringer medium was augmented and now inhibited by H89 (10 micromol/l), implicating that culture at 16.7 mmol/l glucose may increase Ca(2+)-sensitive adenylyl cyclase activity and hence PKA activity. In accordance, Ca(2+)-evoked insulin secretion at 60 mmol/l K(+) and 250 micromol/l diazoxide was improved, whereas glucose amplification of K(+)(ATP) channel-independent insulin secretion was unaffected.

CONCLUSIONS

Glucose may activate PKA through triggering of the K(+)(ATP) channel-dependent pathway. Glucose amplification of K(+)(ATP) channel-independent insulin secretion, on the other hand, occurs by PKA-independent mechanisms.

The inhibition of human duodenal adenylate cyclase activity by Ca2+ and the effects of EGTA

This study demonstrates that the inhibition of adenylate cyclase activity by Ca2+ is enhanced in the presence of increasing [EGTA] (0, 0.3, 1, 2.5 mM) by 2 orders of magnitude. It has been established that this effect is not because of poor Ca2+ buffering by low [EGTA] or high Ca2+ binding by the membrane preparation. It is present irrespective of stimulus. We suggest the enhanced sensitivity of adenylate cyclase to Ca2+ induced by EGTA is caused by the Ca-EGTA complex being a more inhibitory species than Ca2+. Thus consideration of the effects of the Ca-EGTA complex should be made when interpreting the results from experiments involving Ca2+ and EGTA.

A quantitative study of the Ca2+/calmodulin sensitivity of adenylyl cyclase in Aplysia, Drosophila, and rat

Studies in Aplysia and Drosophila have suggested that Ca2+/calmodulin-sensitive adenylyl cyclase may act as a site of convergence for the cellular representations of the conditioned stimulus (Ca2+ influx) and unconditioned stimulus (facilitatory transmitter) during elementary associative learning. This hypothesis predicts that the rise in intracellular free Ca2+ concentration produced by spike activity during the conditioned stimulus will cause an increase in the activity of adenylyl cyclase. However, published values for the Ca2+ sensitivity of Ca2+/calmodulin-sensitive adenylyl cyclase in mammals and in Drosophila vary widely. The difficulty in evaluating whether adenylyl cyclase would be activated by physiological elevations in intracellular Ca2+ levels is in part a consequence of the use of Ca2+/EGTA buffers, which are prone to several types of errors. Using a procedure that minimizes these errors, we have quantified the Ca2+ sensitivity of adenylyl cyclase in membranes from Aplysia, Drosophila, and rat brain with purified species-specific calmodulins. In all three species, adenylyl cyclase was activated by an increase in free Ca2+ concentration in the range caused by spike activity. Ca2+ sensitivity was dependent on both calmodulin concentration and Mg2+ concentration. Mg2+ raised the threshold for adenylyl cyclase activation by Ca2+ but also acted synergistically with Ca2+ to activate maximally adenylyl cyclase.

The effects of Ca2+ and calmodulin on adenylyl cyclase activity in plasma membranes derived from neural and non-neural cells

The regulation of adenylyl cyclase activity by varying concentrations of Ca2+ was examined in plasma membrane preparations derived from a number of neural and non-neural cells. Enzyme activity in neural tissue (i.e. cerebellum) neural-derived pheochromocytoma PC12 cells and certain endocrine cells (i.e. pancreatic RINm5f and parathyroid cells) was stimulated by physiologic concentrations of Ca2+ by a calmodulin (CaM)-dependent mechanism. In contrast, adenylyl cyclase activity in non-neural cells (e.g. platelets and GH3 cells) was not stimulated by Ca2+. In these latter sources, enzyme activity was inhibited by increasing concentrations of Ca2+, independent of CaM. In liver membranes, Ca2+ and/or CaM did not alter adenylyl cyclase activity. These results demonstrate that the effects exerted by physiologic concentrations of Ca2+ on adenylyl cyclase activity range from CaM-dependent stimulation of activity to no effect, to CaM-independent inhibition of activity. The actions of Ca2+ on adenylyl cyclase may be major contributors to the various synergistic or antagonistic interactions that are seen between cAMP-generating and Ca(2+)-mobilizing systems.

Ca(2+)- and ATP-dependent reversible inactivation of pancreatic islet phosphoinositide-specific phospholipase C activity

Phosphoinositide-specific phospholipase C (PI-PLC) activity in whole homogenates of mouse pancreatic islets decreased 60-85% when the homogenates were incubated at 37 degrees C for 1 h in the presence of down to micromolar concentrations of Ca2+. Ca(2+)-induced inactivation was augmented by calmodulin, the phorbol ester 12-O-tetradecanoylphorbol 13-acetate in the presence of ATP-Mg, and by Mg2+. Inactivation was inhibited when ATP was removed and completely abolished by trifluoperazine and EGTA. Inactivation was not affected by the non-phosphorylating ATP analogue, AMP-PCP, GMP-PNP, glucose, Zn2+ or a series of protease inhibitors. These observations suggest that PI-PLC in broken cell preparations of pancreatic islets may be inactivated via phosphorylation by Ca(2+)-calmodulin-stimulated protein kinase and/or protein kinase C. Inactivation of PI-PLC was reversible. Reactivation started after approx. 2 h incubation, when the concentration of ATP in the homogenate was below 0.15 x 10(-6) M. PI-PLC activity returned to values approx. 25% higher than the initial values. PI-PLC inactivation via phosphorylation by the mentioned protein kinases may constitute a feedback control on the phosphoinositide response, attenuating subsequent diacylglycerol formation and/or Ca2+ mobilization by inositol trisphosphate.

Calmodulin independence of human duodenal adenylate cyclase

The calmodulin and calcium dependence of human adenylate cyclase from the second part of the duodenum was assessed in washed particulate preparations of biopsy specimens by investigating (a) the concentration dependent effects of free [Ca2+] on enzyme activity, (b) the effects of exogenous calmodulin on enzyme activity in ethylene glycol bis (b-aminoethyl ether)N,N'-tetra-acetic acid (EGTA) washed particulate preparations, and (c) the effects of calmodulin antagonists on enzyme activity. Both basal (IC50 = 193.75 (57.5) nmol/l (mean (SEM)) and NaF stimulated (IC50 = 188.0 (44.0) nmol/l) adenylate cyclase activity was strongly inhibited by free [Ca2+] greater than 90 nmol/l. Free [Ca2+] less than 90 nmol/l had no effect on adenylate cyclase activity. NaF stimulated adenylate cyclase activity was inhibited by 50% at 2.5 mmol/l EGTA. This inhibition could not be reversed by free Ca2+. The addition of exogenous calmodulin to EGTA (5 mmol/l) washed particulate preparations failed to stimulate adenylate cyclase activity. Trifluoperazine and N-(8-aminohexyl)-5-IODO-1-naphthalene-sulphonamide (IODO 8) did not significantly inhibit basal and NaF stimulated adenylate cyclase activity when measured at concentrations of up to 100 mumol/l. These results suggest that human duodenal adenylate cyclase activity is calmodulin independent but is affected by changes in free [Ca2+].

Stimulation by glucose of cyclic AMP accumulation in mouse pancreatic islets is mediated by protein kinase C

The mechanism of glucose-stimulated cyclic AMP accumulation in mouse pancreatic islets was studied. In the presence of 3-isobutyl-1-methylxanthine, both glucose and the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA), an activator of protein kinase C, enhanced cyclic AMP formation 2.5-fold during 60 min of incubation. Both TPA-stimulated and glucose-stimulated cyclic AMP accumulations were abolished by the omission of extracellular Ca2+. The Ca2+ ionophore A23187 did not affect cyclic AMP accumulation itself, but affected the time course of TPA-induced cyclic AMP accumulation, the effect of A23187 + TPA mimicking the time course for glucose-induced cyclic AMP accumulation. A 24 h exposure to TPA, which depletes islets of protein kinase C, abolished the effects of both TPA and glucose on cyclic AMP production. Both TPA-induced and glucose-induced cyclic AMP productions were inhibited by anti-glucagon antibody, and after pretreatment with this antibody glucose stimulation was dependent on addition of glucagon. Pretreatment of islets with TPA for 10 min potentiated glucagon stimulation and impaired somatostatin inhibition of adenylate cyclase activity in a particulate fraction of islets. Carbamoylcholine, which is supposed to activate protein kinase C in islets, likewise stimulated cyclic AMP accumulation in islets. These observations suggest that glucose stimulates islet adenylate cyclase by activation of protein kinase C, and thereby potentiates the effect of endogenous glucagon on adenylate cyclase.

Abnormal regulation of insulin secretion in the genetically obese (ob/ob) mouse

To determine whether the abnormal insulin-secretory activity encountered in obese mice is due to an anomaly in the production of cyclic AMP, islets of lean and obese mice were incubated with forskolin under various conditions. Our data show that, in addition to the well-known quantitative differences in insulin-secretory activity between islets of lean and obese mice, there are important qualitative differences. The islets of obese mice accumulated less cyclic AMP than did those of lean mice in response to given doses of forskolin, yet their insulin secretion was enhanced to much higher values. In the islets of obese mice, but not in those of lean mice, the stimulatory effect of forskolin on insulin secretion was evident even at non-stimulatory concentrations of glucose or in Ca2+-deprived incubation media, showing that the islet of the obese mouse is less dependent on a primary stimulus (glucose) and on the provision of normal concentrations of Ca2+ in the bathing medium than is the islet of the lean mouse.

Calmodulin activation and calcium regulation of parotid gland adenylate cyclase

The effect of Ca2+ on the adenylate cyclase activity associated with membranes prepared from mouse parotid gland has been examined. Ca2+ stimulated then inhibited adenylate cyclase activity, with values for half-maximal stimulation and inhibition of 0.6 and 10 microM, respectively. Maximal activation (1.4-fold) was observed at 2 microM free Ca2+. These membranes contained 1.2 microgram calmodulin/mg protein. Exogenous calmodulin (0.2-1.2 microgram) activated, in a concentration-dependent manner, adenylate cyclase activity, with maximal activation being 2.5-fold at 12 micrograms calmodulin. Preparation of membranes in 2 mM ethyleneglycol-bis(beta-aminoethylether)-N,N,N',N'-tetraacet ic acid (EGTA) resulted not only in a significant decrease in calmodulin levels (0.5 microgram calmodulin/mg protein) but also in a loss of the ability of Ca2+ to stimulate the enzyme. Exogenous calmodulin restored the ability of Ca2+ to stimulate the adenylate cyclase activity associated with EGTA-treated membranes. Trifluoperazine (50 microM) blocked the ability of Ca2+ to activate adenylate cyclase activity in control membranes. The effect of trifluoperazine could be reversed by exogenous calmodulin (0.5 or 5.0 micrograms). These data indicate that calmodulin mediates the activation of parotid gland adenylate cyclase by Ca2+ and that Ca2+, at concentrations which stimulate and inhibit amylase secretion, can activate and inhibit adenylate cyclase activity.

Calmodulin in various insulin secreting cell types

The involvement of the Ca2+ binding protein, calmodulin, in the regulation of insulin release was studied. Calmodulin was measured in isolated rat islets, rat insulinoma cells, the insulin secreting cell line (RINm5F) and in islets isolated from normal and diabetic Chinese hamsters. Total content of calmodulin was determined by a radioimmunoassay using a rabbit anti-calmodulin serum and was found to lie in the range of 4 to 7 micrograms/ml protein. When rat islets were maintained in tissue culture for 6 days at 2.8 or 8.3 mM glucose, the content of calmodulin of the two groups was similar. Likewise there was no difference in calmodulin content between islets from normal and diabetic hamsters. This study suggests that a variation of the total cellular calmodulin does not play a role in the process of insulin secretion.

Calcium calmodulin and hormone secretion

As long ago as 1970, it was proposed that Ca2+ can act as a 'second messenger' like cAMP (Rasmussen & Nagata, 1979). The recognition that calmodulin is a major Ca2+ binding protein in non-muscle cells has prompted the suggestion that calmodulin may serve an analogous role for Ca2+ to that served by protein kinase for cAMP (Wang & Waisman, 1979), or at least to the regulatory subunit of the cyclic nucleotide-dependent kinases. It is becoming clear that calmodulin probably does play a role in stimulus secretion coupling in endocrine cells. Nevertheless, some of the experimental approaches which have led to this rather tentative conclusion do induce some doubts, as we have attempted to indicate. Many of the pharmacological agents used in the studies cited in this review are not specific in their interaction with calmodulin. For example, the phenothiazines also inhibit phospholipid-sensitive protein kinase. The introduction of more specific drugs, such as the naphthalene sulphonamides, may lead to a clearer picture of the role of calmodulin in hormone secretion. Relationships probably exist between cyclic nucleotides, calcium, calmodulin, phosphatidylinositol (PI) turnover and phospholipids in the overall control of the secretory process (see Fig. 1). There is considerable evidence that calcium is the primary internal signal initiating exocytosis of hormone from many glands. However, it appears that cyclic nucleotides can modulate the calcium signal either positively or negatively and it is possible that cAMP and calcium can separately activate secretion. The presence of both calmodulin-activated adenylate cyclase and cyclic nucleotide phosphodiesterase in the same tissue would appear to suggest either spatial or temporal control mechanisms or that (diagram; see text) the calcium requirement for calmodulin activation differs between the two enzymes. The true explanation is probably far more complex and involves perhaps as yet unknown factors that can differentially influence the activity of calmodulin itself in membranes and in cytosol. Berridge (1982) and Rasmussen (1980) give detailed accounts and review current hypotheses regarding relationships between the cyclic nucleotide and calcium second messenger systems. The various possible interrelationships of the putative messengers have been encompassed by the term 'Synarchic regulation' (Rasmussen, 1980). These concepts and the elucidation of the mechanisms by which cyclic AMP and calcium are involved in the control of secretion from particular cell types will make fascinating reading over the next few years.(ABSTRACT TRUNCATED AT 400 WORDS)

Calmodulin regulation of adenylate cyclase activity

Calmodulin-dependent stimulation of adenylate cyclase was initially thought to be a unique feature of neural tissues. In recent years evidence to the contrary has accumulated, calmodulin-dependent stimulation of adenylate cyclase now being demonstrated in a wide range of structurally unrelated tissues and species. Demonstration of the existence of calmodulin-dependent adenylate cyclase has in nearly all instances required the removal of endogenous calmodulin. It is not yet clear whether calmodulin-dependent and calmodulin-independent forms of the enzyme exist and whether some tissues (such as heart) lack a calmodulin-dependent adenylate cyclase. The presence of calmodulin appears largely responsible for the ability of the adenylate cyclase enzyme to be stimulated by submicromolar concentrations of calcium; it may not be relevant to the inhibition of the enzyme which occurs at higher concentrations of calcium. The physical relationship of calmodulin to the plasma membrane bound enzyme (or to the soluble forms of the enzyme) is not known nor is the mechanism of adenylate cyclase activation by calmodulin clear; current data suggest some involvement with both the N and C units of the enzyme. Finally, it is possible that in vivo calcium contributes to the duration of the hormone stimulated cyclic AMP signal. Thus current in vitro data suggest that optimal hormonal activation of calmodulin-dependent adenylate cyclase occurs at very low intracellular calcium concentrations, comparable to those found in the resting cell; conversely the enzyme is inhibited as intracellular calcium increases, following for example agonist stimulation of the cell. These higher calcium concentrations would then activate calmodulin-dependent phosphodiesterase. Such differential effects of calcium on adenylate cyclase and phosphodiesterase would ultimately restrict the duration of the hormone-induced cyclic AMP signal.

Calmodulin activates adenylate cyclase from rabbit heart plasma membranes

It was shown that calmodulin (CM) activates the adenylate cyclase (AC) of rabbit heart light sarcolemma in the presence of micromolar free Ca2+ concentrations and this effect is blocked by trifluoroperazine and troponin I. GTP (in the presence of isoproterenol) and Gpp(NH)p are able to increase the CM-dependent activity of enzyme. It was concluded that there is no special CM-dependent "form' of AC in the heart and the common catalytic component of AC can be regulated both by CM and guanine nucleotide-binding regulatory component (N-protein). In the presence of Ca2+ and guanine nucleotide heart AC exists as a complex: CM-catalytic component-N-protein.

Does calmodulin play a role in the regulation of cardiac sarcolemmal adenylate cyclase activity?

The recent suggestion that calmodulin (CaM) could mediate calcium inhibition of cardiac adenylate cyclase (AC) has been reassessed. Using a purified sarcolemmal preparation (SL), the influence of different concentrations of free Ca2+ (obtained using Ca2+-EGTA solutions) was studied on dog heart AC. From 10(-9) M to 10(-3) M Ca2+ reduced basal activity, as well as epinephrine (10(-4) M)- and trypsin (1.0 microgram/mL)-stimulated activities with, in the three cases, an identical IC50 of 10(-8) M. The amount of endogenous CaM in the SL, measured using a radioimmunoassay technique, was found to be 7.5 ng/mg protein. The resulting concentration of CaM in the final AC incubation medium was lower than 50 pM, indicating the lack of a significant role for endogenous CaM in the inhibition observed. The addition of exogenous CaM to the AC assay at a concentration sufficient to stimulate other CaM-dependent systems did not modify the Ca2+ inhibitory curves for basal, epinephrine (10(-4) M)-stimulated, or trypsin (1 microgram/mL)-stimulated activities. These results indicate that CaM does not play a significant role in the Ca2+ inhibition of cardiac AC and that trypsin stimulation of cardiac AC is not mediated through a CaM-dependent process.

Characteristics of an adenylate cyclase enhancing factor from mouse pancreatic islet cytosol

The role of cytosolic components in the regulation of mouse pancreatic islet adenylate cyclase activity was studied. Addition of mouse islet cytosol (27000 g supernatant of mouse islet sonicate), devoid of adenylate cyclase activity itself, increased adenylate cyclase activity by 93 +/- 17% (n = 9) in the 27000 g total particulate fraction of mouse islets. Addition of GTP stimulated adenylate cyclase activity by 91 +/- 11% (n = 13) or to the same degree as cytosol. Like GTP, the substance causing the enhancing activity of the cytosol was found to be dialysable, resistant to heat, sensitive to charcoal treatment and alkaline phosphatase and insensitive to digestion with trypsin. However, in contrast to the stimulation by GTP, the stimulation by cytosol was not inhibited by guanosine 5'-0-(2-thiodiphosphate), and furthermore, the effects of cytosol and GTP were additive. Neither NAD nor phosphoenolpyruvate stimulated adenylate cyclase activity. The cytosolic factor did not confer sensitivity towards glucose, Ca2+ or Ca2+-calmodulin on adenylate cyclase. The results demonstrate that mouse pancreatic islets contain a phosphocompound (or several compounds) distinct from GTP and capable of markedly stimulating adenylate cyclase. The identity of the compound and its physiological significance remain to be established.

Cyclic AMP-dependent protein phosphorylation and insulin secretion in intact islets of Langerhans

Effects on insulin release, cyclic AMP content and protein phosphorylation of agents modifying cyclic AMP levels have been tested in intact rat islets of Langerhans. Insulin release induced by glucose was potentiated by dibutyryl cyclic AMP, glucagon, cholera toxin and 3-isobutyl-1-methylxanthine (IBMX); the calmodulin antagonist trifluoperazine reversed these potentiatory effects. Inhibition by trifluoperazine of IBMX-potentiated release was, however, confined to concentrations of IBMX below 50 microM; higher concentrations, up to 1 mM, were resistant to inhibition by trifluoperazine. IBMX-potentiated insulin release was also inhibited by 2-deoxyadenosine, an inhibitor of adenylate cyclase. In the absence of glucose, IBMX at concentrations up to 1 mM did not stimulate insulin release and in the presence of 3.3 mM-glucose IBMX was effective only at a concentration of 1 mM; under the latter conditions trifluoperazine again did not inhibit insulin secretion. The maximum effect on insulin release was achieved with 25 microM-IBMX. Islet [cyclic AMP] was increased by IBMX, with the maximum rise occurring with 100 microM-IBMX. The increase in [cyclic AMP] elicited by IBMX was more rapid than that induced by cholera toxin. Trifluoperazine did not significantly affect islet cyclic AMP levels under any of the conditions tested. When islets were incubated with [32P]Pi, radioactivity was incorporated into islet ATP predominantly in the gamma-position. The rate of equilibration of label was dependent on medium Pi and glucose concentration and at optimal concentrations of these 100% equilibration of internal [32P]ATP with external [32P]Pi required a period of 3h. Radioactivity was incorporated into islet protein and, in response to an increase in islet [cyclic AMP], the major effect was on a protein of Mr 15 000 on sodium dodecyl sulphate/polyacrylamide gels. The extent of phosphorylation of the Mr-15 000 protein was correlated with the level of cyclic AMP: phosphorylation in response to IBMX was inhibited by 2-deoxyadenosine but not by trifluoperazine. Fractionation of islets suggested that the Mr-15 000 protein was of nuclear origin: the protein co-migrated with histone H3 on acetic acid/urea/Triton gels. In the islet cytosol a number of proteins were phosphorylated in response to elevation of islet [cyclic AMP]: the major species had Mr values of 18 000, 25 000, 34 000, 38 000 and 48 000. Culture of islets with IBMX increased the rate of [3H]-thymidine incorporation.(ABSTRACT TRUNCATED AT 400 WORDS)