Ca 2ϩ -dependent activator protein for secretion (CAPS) ͉ exocytosis ͉ insulin ͉ phosphoinositides

In mouse pancreatic beta-cells, extracellular ATP (0.1 mmol/l) effectively reduced glucose-induced insulin secretion. This inhibitory action resulted from a direct interference with the secretory machinery, and ATP suppressed depolarization-induced exocytosis by 60% as revealed by high-resolution capacitance measurements. Suppression of Ca2+-dependent exocytosis was mediated via binding to P2Y1 purinoceptors but was not associated with inhibition of the voltage-dependent Ca2+ currents or adenylate cyclase activity. Inhibition of exocytosis by ATP resulted from G-protein-dependent activation of the serine/threonine protein phosphatase calcineurin and was abolished by cyclosporin A and deltamethrin. In contrast to the direct inhibitory action on exocytosis, ATP reduced the whole-cell ATP-sensitive K+ (K(ATP)) current by 30% (via activation of cytosolic phospholipase A2), leading to membrane depolarization and stimulation of electrical activity. The stimulatory effect of ATP also involved mobilization of Ca2+ from thapsigargin-sensitive intracellular stores. We propose that the inhibitory action of ATP, by interacting with the secretory machinery at a level downstream to an elevation in [Ca2+]i, is important for autocrine regulation of insulin secretion in mouse beta-cells.

The mechanism of L-arginine stimulation of glucose-induced insulin secretion from mouse pancreatic islets was studied. At 16.7 mmol/l glucose, L-arginine (10 mmol/l) potentiated both phases 1 and 2 of glucose-induced insulin secretion. This potentiation of glucose-induced insulin secretion was mimicked by the membrane depolarizing agents tetraethylammonium (TEA, 20 mmol/l) and K + (60 mmol/l), which at 16.7 mmol/l glucose obliterated L-arginine (10 mmol/l) modulation of insulin secretion. Thus L-arginine may potentiate glucose-induced insulin secretion by stimulation of membrane depolarization. At 3.3 mmol/l glucose, L-arginine (10 mmol/l) failed to stimulate insulin secretion. In accordance with membrane depolarization by the electrogenic transport of L-arginine, however, L-arginine (10 mmol/l) stimulation of insulin secretion was enabled by the K + channel inhibitor TEA (20 mmol/l), which potentiates membrane depolarization by L-arginine. Furthermore, Larginine (10 mmol/l) stimulation of insulin secretion was permitted by forskolin (10 mmol/l) or tetradecanoylphorbol 13-acetate (0.16 mmol/l), which, by activation of protein kinases A and C respectively sensitize the exocytotic machinery to L-arginine-induced Ca 2+ influx. Thus glucose may sensitize L-arginine stimulation of insulin secretion by potentiation of membrane depolarization and by activation of protein kinase A or protein kinase C. Finally, L-arginine stimulation of glucose-induced insulin secretion was mimicked by N G -nitro-L-arginine methyl ester (10 mmol/l), which stimulates membrane depolarization but inhibits nitric oxide synthase, suggesting that L-arginine-derived nitric oxide neither inhibits nor stimulates insulin secretion. In conclusion, it is suggested that L-arginine potentiation of glucose-induced insulin secretion occurs independently of nitric oxide, but is mediated by membrane depolarization, which stimulates insulin secretion through protein kinase A-and Csensitive mechanisms.

1. Rates of insulin secretion, glucose utilization, lactate output, incorporation of glucose into glycogen, contents of glucose 6-phosphate, fructose 1,6-diphosphate and ATP, and maximally extractable enzyme activities of hexokinase, high-K(m) glucose-phosphorylating activity (;glucokinase'), glucose 6-phosphatase and unspecific acid phosphatase were measured in isolated pancreatic islets from fed and 48-h-starved mice. 2. In the fed state insulin secretion from isolated islets was increased five- to six-fold when the extracellular glucose concentration was raised from 2.5mm to 16.7mm; 5mm-caffeine potentiated this effect. The secretory response to glucose of islets from mice starved for 48h was diminished at all glucose concentrations from 2.5mm up to approx. 40mm. Very high glucose concentrations (60mm and above) restored the secretory response to that found in the fed state, suggesting that the K(m) value for the overall secretory process had been increased (approx. fourfold) by starvation. Addition of 5mm-caffeine to islets from starved mice also restored the insulin secretory response to 2.5-16.7mm-glucose to normal values. 3. Extractable hexokinase, ;glucokinase', glucose 6-phosphatase and unspecific phosphatase activities were not changed by starvation. 4. Glucose utilization and glycolysis (measured as the rate of formation of (3)H(2)O from [5-(3)H]glucose over a 2h period) was decreased in islets from starved mice at all glucose concentrations up to approx. 55mm. At still higher glucose concentrations up to approx. 100mm, there was no difference between the fed and starved state, suggesting that the K(m) value for the rate-limiting glucose phosphorylation had been increased (approx. twofold) by starvation. Preparation of islets omitting substrates (glucose, pyruvate, fumarate and glutamate) from the medium during collagenase treatment lowered the glucose utilization measured subsequently at 16.7mm-glucose by 38 and 30% in islets from fed and starved mice respectively. Also the 2h lactate output by the islets at 16.7mm extracellular glucose was diminished by starvation. Incorporation of glucose into glycogen was extremely low, but the rate of incorporation was more than doubled by starvation. 5. After incubation for 30min at 16.7mm-glucose the content of glucose 6-phosphate was unchanged by starvation, that of ATP was increased and the concentration of (fructose 1,6-diphosphate plus triose phosphates) was decreased. 6. Possible mechanisms behind the correlated impairment in insulin secretion and islet glucose metabolism during starvation are discussed.

When the extracellular concentration of glucose was raised from 3 mM to 7 mM (the concentration interval in which beta-cell depolarization and the major decrease in K+ permeability occur), the cytosolic free [NADPH]/[NADP+] ratio in mouse pancreatic islets increased by 29.5%. When glucose was increased to 20 mM, a 117% increase was observed. Glucose had no effect on the cytosolic free [NADH]/[NAD+] ratio. Neither the cytosolic free [NADPH]/[NADP+] ratio nor the corresponding [NADH]/[NAD+] ratio was affected when the islets were incubated with 20 mM-fructose or with 3 mM-glucose + 20 mM-fructose, although the last-mentioned condition stimulated insulin release. The insulin secretagogue leucine (10 mM) stimulated insulin secretion, but lowered the cytosolic free [NADPH]/[NADP+] ratio; 10 mM-leucine + 10 mM-glutamine stimulated insulin release and significantly enhanced both the [NADPH]/[NADP+] ratio and the [NADH]/[NAD+] ratio. It is concluded that the cytosolic free [NADPH]/[NADP+] ratio may be involved in coupling beta-cell glucose metabolism to beta-cell depolarization and ensuing insulin secretion, but it may not be the sole or major coupling factor in nutrient-induced stimulation of insulin secretion.

Insulin secretion, insulin biosynthesis and islet glucose oxidation were studied in pancreatic islets isolated from fat-fed diabetic mice of both sexes. Insulin secretion from isolated islets was studied after consecutive stimulation with alpha-ketoisocaproic acid + glutamine, glucose, forskolin, and 12-O-tetradecanoylphorbol 13-acetate. Glucose-induced insulin secretion was impaired in islets from fat-fed mice. This was associated with a reduction of approximately 50% in islet glucose oxidation. Islet insulin secretion stimulated by the non-carbohydrate secretagogues tended to be higher in the fat-fed mice, but a statistically significant effect was not observed. Pancreatic insulin content was reduced by 50%, whereas the islet insulin and DNA content was unchanged after fat feeding. Proinsulin mRNA was reduced by 35% in islets from fat-fed mice, and was associated with a reduction of approximately 50% in glucose-stimulated (pro)insulin biosynthesis. It is concluded that the insulin secretory response of islets isolated from fat-fed mice is similar to the secretory pattern known from human type 2, non-insulin-dependent diabetics, and that a defect in islet glucose recognition, resulting in decreased glucose oxidation, may be responsible for the observed insulin secretory and biosynthetic defects seen after glucose stimulation.

The occurrence of endogenous substrate proteins for Ca2+-dependent protein kinase, augmented by either phospholipid or calmodulin, and for cyclic AMP-dependent protein kinase was examined in homogenates and subcellular fractions of mouse pancreatic islets. Islet protein phosphorylation was enhanced by Ca2+-calmodulin; the major endogenous substrates in the homogenate were two proteins of Mr 53000 and 100000. The Mr-100000 phosphoprotein was localized to a 27000g-supernatant fraction, whereas the Mr-53000 phosphoprotein was present in a 27000g particulate fraction of mouse islets. In the presence of Ca2+, phosphatidylserine stimulated phosphorylation of 15 proteins, of Mr 17000-190000, in a 27000g-supernatant fraction. No effects of Ca2+ plus phosphatidylserine were observed in a 27000g particulate fraction of mouse islets. Examination of cyclic AMP-dependent protein phosphorylation revealed five substrate proteins, of Mr 23000-72000, present in the 27000g supernatant of mouse islets. No common substrates for either the two Ca2+-dependent phosphorylation systems or for the cyclic AMP-dependent and the Ca2+-calmodulin-dependent phosphorylation systems were noted. On the other hand, the actions of the cyclic AMP-sensitive and the Ca2+-phospholipid-sensitive systems may be overlapping, since two common substrates for them were noted in the 27000g-supernatant fraction. The results are consistent with the hypothesis that protein phosphorylation may play a role in the regulation of insulin secretion by Ca2+ and cyclic AMP. The extensive stimulatory effect of phosphatidylserine furthermore suggests that the Ca2+-phospholipid-sensitive protein kinase may prove to be a prominent phosphorylation system in pancreatic islets.

The occurrence and function of polyamines in protein kinase C activation and insulin secretion in mouse pancreatic islets were studied. Determination of polyamines in mouse islets revealed 0.9 +/- 0.3 (mean +/- S.E.M., n = 6) pmol of putrescine, 11.7 +/- 3.2 (8) pmol of spermidine and 3.7 +/- 0.6 (8) pmol of spermine per islet, corresponding to intracellular concentrations of 0.3-0.5 mM-putrescine, 3.9-5.9 mM-spermidine and 1.2-1.9 mM-spermine in mouse islets. Stimulation of insulin secretion by glucose, the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA) or the sulphonylurea glibenclamide did not affect these polyamine contents. In accordance with a role for protein kinase C in insulin secretion, TPA stimulated both protein kinase C activity and insulin secretion. Stimulation of insulin secretion by TPA was dependent on a non-stimulatory concentration of glucose and was further potentiated by stimulatory concentrations of glucose, glibenclamide or 3-isobutyl-1-methylxanthine, suggesting that protein kinase C activation, Ca2+ mobilization and cyclic AMP accumulation are all needed for full secretory response of mouse islets. Spermidine (5 mM) and spermine (1.5 mM) at concentrations found in islets inhibited protein kinase C stimulated by TPA + phosphatidylserine by 55% and 45% respectively. Putrescine (0.5 mM) was without effect, but inhibited the enzyme at higher concentrations (2-10 mM). Inhibition of protein kinase C by polyamines showed competition with Ca2+, and Ca2+ influx in response to glucose or glibenclamide prevented inhibition of insulin secretion by exogenous polyamines at concentrations where they did not affect glucose oxidation. It is suggested that inhibition of protein kinase C by polyamines may be of significance for regulation of insulin secretion in vivo and that Ca2+ influx may function by displacing inhibitory polyamines bound to phosphatidylserine in membranes.