Putative roles for lysophospholipids as mediators and lipoxygenase-mediated metabolites of arachidonic acid as potentiators of stimulus-secretion coupling: dual mechanisms of p-hydroxymercuribenzoic acid-induced insulin release

This paper explores the mechanism whereby insulin (I) secretion is stimulated by p-hydroxymercuribenzoic acid (PHMB), an agent which inhibits the esterification of arachidonic acid (AA) into phospholipids in intact rat islets. An effect of PHMB on I release could be seen even at substimulatory glucose concentrations (0-1.7 mM) and was resistant to blockade of energy flux using antimycin A, or of glucose metabolism using mannoheptulose. It was, however, inhibited by Ni++, Co++, La , replacement of chloride in the buffer by the impermeant anion isethionate or reduced ambient temperature (16 degrees C), but not by extracellular Ca++ depletion or 8-(N,N-diethylamino) octyl 3,4,5-trimethoxy-benzoate hydrochloride (a putative stabilizer of intracellular Ca++ stores); thus PHMB's effect may require the translocation of membrane-associated Ca++ stores, leading to exocytotic hormone release. Although PHMB increases the accumulation of lipoxygenase-derived metabolites of AA, I secretion at 1.7 mM glucose unexpectedly was resistant to cyclooxygenase or lipoxygenase inhibition and could not be reproduced by exogenous AA (0.18 through 262 microM). However, it could be mimicked closely by exogenous lysophosphatidylcholine, which also shared with PHMB an identical profile of reversibility and pharmacologic inhibitability. Lysophospholipid (lyso-PL)-induced I release could not be attributed to detergent effects because, for example, it occurred in the absence of significant 51Cr release. The lyso-PL effect demonstrated structural specificity (lysophosphatidyl-ethanolamine and lysophosphatidylserine being essentially inactive) and was specific for lyso-PLs as neither phosphatidylcholine itself nor glycerophosphorylcholine (the deacylation product of lysophosphatidylcholine) had any effect. In contrast to the effects of lyso-PLs, the energy-dependent effects of glucose (16.7 mM) or the amino acid alpha-ketoisocaproic acid (15 mM) on I release were abrogated by inhibitors of phospholipases or lipoxygenase. This effect of phospholipase inhibition could be circumvented by exogenous lyso-PLs. We conclude that lyso-PLs (generated by energy-dependent phospholipid deacylation or by inhibition of reacylation) may be true mediators of I release, whereas the role of concomitantly generated oxygenation products of AA is restricted to the modulation of stimulated release.     

Lipoxygenase pathway in islet endocrine cells. Oxidative metabolism of arachidonic acid promotes insulin release

Metabolism of arachidonic acid (AA) via the cyclooxygenase pathway reduces glucose-stimulated insulin release. However, metabolism of AA by the lipoxygenase pathway and the consequent effects on insulin secretion have not been simultaneously assessed in the endocrine islet. Both dispersed endocrine cell-enriched pancreatic cells of the neonatal rat, as well as intact islets of the adult rat, metabolized [(3)H]AA not only to cyclooxygenase products (prostaglandins E(2), F(2alpha), and prostacyclin) but also to the lipoxygenase product 12-hydroxyeicosatetraenoic acid (12-HETE). 12-HETE was identified by coelution with authentic tritiated or unlabeled 12-HETE using four high performance liquid chromatographic systems under eight mobile-phase conditions and its identity was confirmed by gas chromatography/mass spectrometry using selected ion monitoring. The predominant effect of exogenous AA (5 mug/ml) was to stimulate insulin release from pancreatic cells grown in monolayer. This effect was concentration- and time-dependent, and reversible. The effect of AA upon insulin release was potentiated by a cyclooxygenase inhibitor (indomethacin) and was prevented by either of two lipoxygenase inhibitors (5,8,11,14-eicosatetraynoic acid [ETYA] and BW755c). In addition, glucose, as well as two structurally dissimilar agents (the calcium ionophore A23187 and bradykinin), which activate phospholipase(s) and thereby release endogenous AA in several cell systems, also stimulated insulin secretion. The effects of glucose, glucagon, bradykinin and high concentrations of A23187 (5 mug/ml) to augment insulin release were blocked or considerably reduced by lipoxygenase inhibitors. However, a lower concentration of the ionophore (0.25 mug/ml), which did not appear to activate phospholipase, was resistant to blockade. Exogenous 12-HETE (up to 2,000 ng/ml) did not alter glucose-induced insulin release. However, the labile intermediate 12-hydroperoxy-ETE increased insulin release. Furthermore, diethylmaleate (which binds intracellular glutathione and thereby impedes conversion of the lipoxygenase intermediates hydroperoxy-ETE and leukotriene A(4) to HETE and leukotriene C(4), respectively) potentiated the effect of glucose and of exogenous AA. Finally, 5,6-epoxy, 8,11,14-eicosatrienoic acid (a relatively stable epoxide analogue of leukotriene A(4)) as well as two other epoxy-analogues, potentiated glucose-induced insulin release. We conclude that dual pathways of AA metabolism exist in islet endocrine cells and have opposing regulatory effects on the beta cell-an inhibitory cyclooxygenase cascade and a stimulatory lipoxygenase cascade. Labile products of the latter pathway may play a pivotal role in stimulus-secretion coupling in the islet.     

Virus-induced alterations in cyclic adenosine monophosphate generation in hamster islets of Langerhans.

Inoculation of golden Syrian hamsters with Venezuelan encephalitis (VE) virus results in a sustained diminution in glucose-stimulated insulin release that is correctable by cyclic (c) AMP analogs and phosphodiesterase inhibitors. This suggested the importance of directly measuring cAMP content in VE-infected and control islets in response to insulin secretagogues. The basal cAMP content of VE-infected islets (0.14 +/- 0.02 pmol/micrograms islet DNA) was approximately half that of control islets (0.27 +/- 0.02 pmol/micrograms islet DNA) (P less than 0.05). In the presence of 10 microM glucagon (and 3 mM glucose), the rate of cAMP generation in VE-infected islets was only half that of control islets. With 10 mM alpha-ketoisocaproic acid, the rates of cAMP generation were indistinguishable between control and experimental groups. In response to 20 mM glucose and 3-isobutyl-1-methylxanthine (IBMX) (a phosphodiesterase inhibitor), cAMP generation in VE-infected islets was 81% (NS) of the control rate. When a more specific phosphodiesterase inhibitor, RO 20-1724, was used with 20 mM glucose, cAMP generation in the infected islets was only 44% (P less than 0.001) of the control value. Insulin secretion over the perifusion period paralleled the cAMP levels. In the presence of 10 mM alpha-ketoisocaproic acid, there was no difference in insulin secretion between VE-infected and control islets, while there was a statistically significant (P less than 0.05) difference with 10 microM glucagon or 20 mM glucose (in 1 mM RO 20-1724). These data point to a defect in the cAMP generation system of VE-infected islets, although additional factors involved in insulin secretion may also be impaired by the virus.     

Role of Ca2+ in secretagogue-stimulated breakdown of phosphatidylinositol in rat pancreatic islets.

Breakdown of phosphatidylinositol (PI) has been shown to be increased during Ca2+-mediated stimulation of cellular responses in many systems and has been proposed to be involved in stimulus-secretion coupling. The effects on PI breakdown of insulin secretagogues that alter cellular Ca2+ or cyclic (c)AMP levels were investigated in perifused rat islets of Langerhans. Isolated islets were labeled with myo-[2-3H(N)]inositol and the efflux of 3H-labeled metabolites was monitored. Glucose (16.7 mM) greatly increased 3H release in a manner that paralleled the second phase of the insulin secretory response; by 60 min, the amount of [3H]PI in the islet decreased by 50%. Removal of Ca2+ from the perifusate or blockade of Ca2+ entry through the voltage-dependent channels by D600 (20 microM) abolished the glucose-induced increase in 3H efflux. Depolarization with 47 mM K+, which increases Ca2+ entry, stimulated protracted 3H and insulin release. Glucose-stimulated output of 3H was not prevented by epinephrine (1 microM) even though the insulin response was abolished. In contrast, 3H output was not affected by isobutylmethylxanthine (1 mM), known to raise cellular levels of cAMP, although insulin release was stimulated. These findings indicate that PI breakdown is not related to the exocytotic process since stimulation of insulin release and PI breakdown could be uncoupled, and that it is not associated with cAMP-mediated regulation of insulin release. PI breakdown in islets differs from the immediate, transient phenomenon reported in other systems in both its timing and requirement for Ca2+. It appears to result from the entry of Ca2+ and not to be the mechanism by which glucose initiates Ca2+ influx.     

Virus-induced alterations in insulin release in hamster islets of Langerhans.

After the inoculation of Golden Syrian hamsters with the TC-83 vaccine strain of Venezuelan encephalitis (VE) virus, a sustained diminution in glucose-stimulated insulin release and glucose intolerance of shorter duration develops. To understand better the mechanism of this defect in insulin release, we examined insulin secretion in response to several test agents in isolated perifused islets from control and 24-d post-VE virus-infected hamsters. 50 islets were used in all perifusion experiments, and data were expressed as total insulin released as well as peak response for each test agent during a 30-min perifusion period from control and VE-infected islets. After perifusion with 20 mM glucose, a 45% diminution of insulin release was noted in VE-infected islets in comparison with control islets, which in turn was similar to in vivo findings. However, following 1-mM tolbutamide stimulation, insulin release was similar in control and VE-infected islets. In separate studies, 1 mM tolbutamide, 10 mM theophilline, 1 mM dibutyryl cyclic (c)AMP, and 1 mM 8-bromo-cAMP resulted in statistically similar insulin-release curves in control and VE-infected islets. Additional experiments assessing [5-3H]glucose use in control and infected islets after 20 min of perifusion with 20 mM glucose revealed virtually identical values (239 +/- 30-control; and 222 +/- 27-VE-infected islets). Morphological and morphometric evaluation of VE-infected islets (21 d following virus inoculation) showed no changes in islet volume density, beta cell density, and beta cell granulation. Thus, VE virus induces a defect in glucose-stimulated insulin release from hamster beta cells that can be corrected by cAMP analogues and does not alter islet glucose use.     

Fatty acid-induced beta cell hypersensitivity to glucose. Increased phosphofructokinase activity and lowered glucose-6-phosphate content.

Diabetic states are characterized by a raised serum/islet level of long chain fatty acids and a lowered ED50 for glucose-induced insulin secretion. Prolonged culture (> 6 h) of islets with long chain fatty acids replicates the basal insulin hypersecretion. We examined this effect in rat islets cultured for 24 h with 0.25 mM oleate. Insulin secretion at 2.8 mM glucose was doubled in combination with a 60% lowered islet content of glucose-6-phosphate (G6P). Investigation of the lowered G6P showed: (a) increased glucose usage from 0.5 to 100 mM glucose with identical values measured by [2-3H]glucose and [5-3H]glucose, (c) indicating little glucose- 6-phosphatase activity, (b) unchanged low pentose phosphate shunt activity, (c) 50% increased phosphofructokinase (PFK) Vmax, (d) a normal ATP/ADP ratio, and (e) unchanged fructose 2,6 bisphosphate content. Triacsin C, an inhibitor of fatty acyl-CoA synthetase, prevented the increase in PFK activity and the lowered G6P content. These results suggest that long chain acyl-CoA mediates the rise in PFK activity, which in turn lowers the G6P level. We speculate that the inhibition of hexokinase by G6P is thus attenuated, thereby causing the basal insulin hypersecretion.     

Interacting Effects of Sulfonylureas and Glucose on Cyclic AMP Metabolism and Insulin Release in Pancreatic Islets of the Rat

The effects of tolbutamide and glibenclamide on the metabolism of cyclic AMP were investigated in pancreatic islets of the rat. Changes in cyclic AMP were assessed by measuring [(3)H]cyclic AMP after labeling of the islets with [2-(3)H]adenine. In the presence of a nonstimulatory concentration of glucose (3.3 mM), both sulfonylureas caused a rapid increase in islet [(3)H]cyclic AMP, which declined within 5 (tolbutamide) or 10 min (glibenclamide). In the absence of glucose, the glibenclamide effect was shortened, but the initial (1 min) response of [(3)H]-cyclic AMP was unaffected. Glucose could be substituted with d-glyceraldehyde but not pyruvate for prolongation of the glibenclamide response. The effect of glucose withdrawal on the glibenclamide response was reproduced by the addition of d-mannoheptulose to glucose containing media.The [(3)H]cyclic AMP response to glibenclamide was influenced by prior exposure of the islets to glucose, a 30-min preincubation with 27.7 mM glucose, enhancing the response to the sulfonylurea over a subsequent 5-min stimulation period.Sulfonylureas exerted their effects at low but not at high glucose concentrations, i.e., shifted the glucose dose-response curve to the left both for [(3)H]cyclic AMP accumulation and insulin release. On the other hand, increasing concentrations of the phosphodiesterase inhibitor, 3-isobutyl-1-methylxanthine, progressively augmented the effects of the drugs.Omission of Ca(++) from the incubation media inhibited both the glucose and the sulfonylurea [(3)H]-cyclic AMP and insulin responses. Epinephrine (1 muM) partially inhibited the [(3)H]cyclic AMP response to both glucose and sulfonylurea, whereas insulin release was completely abolished.It is concluded that the sulfonylurea effects on islet cyclic AMP are intimately related to those of glucose. It is suggested that sulfonylureas exert a major part of their action by facilitating the effect of glucose on the beta-cell adenylate cyclase; the increased cyclic AMP level, in its turn, enhances the secretion rate of insulin.     

Maintenance of insulin release from pancreatic islets stored in the cold for up to 5 weeks.

Insulin content and release were measured from hand-dissected pancreatic islets from noninbred ob/ob mice after 1-5 wk storage in tissue culture medium 199 at various temperatures and glucose concentrations. After storage of islets for 1 wk at 37 degrees, 22 degrees, or 8 degrees C in 18 mM glucose medium and preincubation with 1 mM glucose, glucose-stimulated insulin release during the subsequent incubation was only 20-35% of that of fresh islets. The addition of a 4-h period at 37 degrees C with 18 mM glucose between the cold storage and perincubation restored glucose-stimulated insulin release from 8 degrees C stored islets to fresh-islet levels. Release throughout the 1-18 mM glucose range was strikingly parallel to that of fresh islets. Exposure of fresh islets to the same 4-h period increased basal release but did not affect maximal release. When islets were stored at 8 degrees C with 18 mM glucose for more than 1 wk, a short period at 37 degrees C every week was necessary for maintenance of release. After 5 wk of this procedure, glucose-stimulated insulin release was one-third that of fresh islets, or similar to that of islets stored for only 1 wk at 37 degrees C. Storage at 8 degrees C for 1 wk with 3 mM glucose, or continuously for 3 or 5 wk with 18 mM glucose, maintained islet insulin content, whereas release was lost. Thus, glucose-stimulated insulin release is best maintained by storage of pancreatic islets in tissue culture medium with a high concentration of glucose at 8 degrees C with short weekly periods at 37 degrees C.     

Survival of isolated human islets of Langerhans maintained in tissue culture.

Transplantation of human pancreatic islets to diabetic patients may require that donor islets be kept viable in vitro for extended time periods before transfer to the recipient. We have maintained isolated pancreatic islets obtained from the human cadaveric pancreas in tissue culture for 1-3 wk, after which we studied the structure and function of the islets. Electron micrographs of the cultured islets showed a satisfactory preservation of both beta-cells and alpha 2-cells. After culture for 1 wk, the islet oxygen uptake proceeded at a constant rate at a low glucose concentration (3.3 mM) and was significantly enhanced by raising the glucose concentration to 16.7 mM. Likewise, after culture for 1 wk, the islets responded with an increased insulin release when exposed to 16.7 mM glucose with or without added theophylline (10 mM). Islets cultured for 1-3 wk were able to incorporate [3H]leucine into proinsulin, as judged by gel filtration of acid-alcohol extracts. Glucagon release from the cultured islets was reduced significantly by 16.7 mM glucose alone, but stimulated by glucose (16.7 mM) plus theophylline (10 MM). It is concluded that viable pancreatic islets can be isolated from the pancreas of adult human donors and maintained in tissue culture for at least 1 wk without loss of the specific functions of the alpha 2- and beta-cells. It remains to be established whether such islets will survive and remain functionally competent after transplantation to human recipients.     

Differential in situ cytokine gene expression in acute versus chronic atopic dermatitis.

We tested effects of long-term exposure of pancreatic islets to free fatty acids (FFA) in vitro on B cell function. Islets isolated from male Sprague-Dawley rats were exposed to palmitate (0.125 or 0.25 mM), oleate (0.125 mM), or octanoate (2.0 mM) during culture. Insulin responses were subsequently tested in the absence of FFA. After a 48-h exposure to FFA, insulin secretion during basal glucose (3.3 mM) was several-fold increased. However, during stimulation with 27 mM glucose, secretion was inhibited by 30-50% and proinsulin biosynthesis by 30-40%. Total protein synthesis was similarly affected. Conversely, previous palmitate did not impair alpha-ketoisocaproic acid (5 mM)-induced insulin release. Induction and reversibility of the inhibitory effect on glucose-induced insulin secretion required between 6 and 24 h. Addition of the carnitine palmitoyltransferase I inhibitor etomoxir (1 microM) partially reversed (by > 50%) FFA-associated decrease in secretory as well as proinsulin biosynthetic responses to 27 mM glucose. The inhibitory effect of previous palmitate was similar when co-culture was performed with 5.5, 11, or 27 mM glucose. Exposure to palmitate or oleate reduced the production of 14CO2 from D-[U-14C]glucose, and of 14CO2 from D-[3,4-14C]-glucose, both effects being reversed by etomoxir. Conclusions: long-term exposure to FFA inhibits glucose-induced insulin secretion and biosynthesis probably through a glucose fatty acid cycle.     

Ethanol inhibits N-methyl-D-aspartate-stimulated [3H]norepinephrine release from rat cortical slices

The effects of ethanol on N-methyl-D-aspartate (NMDA)-stimulated [3H]norepinephrine (NE) release from rat cortical slices was studied. NMDA-stimulated [3H]NE release was inhibited by tetrodotoxin, Mg++ and 2-amino-5-phosphonopentanoic acid, indicating that NMDA receptors in the cortex have characteristics similar to those observed using electrophysiological studies. Ethanol (60-200 mM) decreased the release of [3H]NE evoked by 100 microM NMDA in a concentration-dependent manner (32-52% inhibition), but it did not significantly alter the basal release. The inhibitory effect of 100 mM ethanol was due to a reduction in the maximal response with no significant change in the EC50 for NMDA. Pretreatment of the slices with 100 mM ethanol up to 6 min did not alter the magnitude of inhibition. The inhibition of NMDA-stimulated [3H]NE release due to ethanol was reversible after a 13-min recovery period. The presence of ethanol did not significantly affect the IC50 for Mg++ inhibition of NMDA-stimulated [3H]NE release (23 +/- 3 microM). Glycine (10-300 microM) potentiated the release of [3H]NE stimulated by 250 microM NMDA, and 60 mM ethanol did not alter this effect of glycine. Ethanol (100 mM) inhibited the release of [3H]NE evoked by 18.9 mM KCl in the presence or absence of 2-amino-5-phosphonopentanoic acid, but had no effect on release induced by 49.1 mM KCl. Tetrodotoxin (0.3 mM) significantly decreased the release of [3H] NE evoked by 23.2 mM KCl, and 60 to 200 mM ethanol did not alter this release. These results suggest that NMDA receptors in rat cortical slices are located on nerve cell bodies.(ABSTRACT TRUNCATED AT 250 WORDS)     

Diacylglycerol synthesis de novo from glucose by pancreatic islets isolated from rats and humans.

Recent evidence has suggested that pancreatic islets isolated from rats synthesize 1,2-diacyl-sn-glycerol (DAG) de novo from glucose and that this process may constitute the long-sought link between the metabolism of glucose and the induction of insulin secretion. The cell-permeant diacylglycerol 1-oleoyl-2-acetyl-sn-glycerol (200 microM) has been found here to amplify both the first and second phases of insulin secretion from perifused human islets. Measurements of the mass of endogenous DAG in human pancreatic islets by enzymatic and by mass spectrometric methods indicate that levels of 200 microM may be achieved under physiologic conditions. Conversion of [14C]glucose to [14C]DAG has been demonstrated here to occur within 60 s of exposure of rat and human islets to stimulatory concentrations of glucose. This process has been found to be a quantitatively minor contributor to the total islet DAG mass after acute stimulation with glucose, however, and glucose has been found not to induce a rise in total islet DAG content within 20 min of induction of insulin secretion. In contrast to the case with rodent islets, two pharmacologic inhibitors of DAG-induced activation of protein kinase C (staurosporine and sphingosine) have been found not to influence glucose-induced insulin secretion from isolated human islets. These findings indicate that de novo synthesis of DAG from glucose does not participate in acute signal-response coupling in islets.     

Pulsatile insulin release from mouse islets occurs in the absence of stimulated entry of Ca2+.

Pancreatic islets are known to respond to a raise of the glucose concentration with Ca2+ -induced 2-3-min pulses of insulin release. The reports of cyclic variations of circulating insulin in the fasting state made it important to explore whether insulin release is also pulsatile in the absence of stimulated entry of Ca2+. Individual pancreatic islets were isolated from a local colony of ob/ob mice and perifused under conditions allowing dual wavelength recordings of the cytoplasmic Ca2+ concentration ([Ca2+]i) with fura-2 and measurements of insulin with ELISA technique. At 3 mM of glucose, [Ca2+]i remained at a stable low level, but insulin was released in pulses with a frequency of 0.41+/-0.02 min-1, determined by Fourier transformation of original and autocorrelated data. Pulses of basal insulin release were also seen when glucose was omitted and 1 microM clonidine or 400 microM diazoxide was added to a glucose-free medium. The results indicate that pulsatile insulin release can be generated in the absence of stimulated entry of Ca2+. A tentative explanation for this phenomenon is inherent fluctuations in the ATP production of the beta cells.     

Prolonged exposure of human pancreatic islets to high glucose concentrations in vitro impairs the beta-cell function.

The aim of the present study was to clarify whether prolonged in vitro exposure of human pancreatic islets to high glucose concentrations impairs the function of these cells. For this purpose, islets isolated from adult cadaveric organ donors were cultured for seven days in RPMI 1640 medium supplemented with 10% fetal calf serum and containing either 5.6, 11, or 28 mM glucose. There was no glucose-induced decrease in islet DNA content or signs of morphological damage. However, islets cultured at 11 or 28 mM glucose showed a 45 or 60% decrease in insulin content, as compared to islets cultured at 5.6 mM glucose. Moreover, when such islets were submitted to a 60-min stimulation with a low (1.7 mM) followed by a high (16.7 mM) concentration of glucose, the islets cultured at 5.6 mM glucose showed a higher insulin response to glucose than those of the two other groups. Islets cultured at the two higher glucose concentrations showed increased rates of insulin release in the presence of low glucose, and a failure to enhance further the release in response to an elevated glucose level. Islets cultured at 28 mM glucose showed an absolute decrease in insulin release after stimulation with 16.7 mM glucose, as compared to islets cultured at 5.6 mM glucose. The rates of glucose oxidation, proinsulin biosynthesis, and total protein biosynthesis were similar in islets cultured at 5.6 or 11 mM glucose, but they were decreased in islets cultured at 28 mM glucose. These combined results suggest that lasting exposure to high glucose concentrations impairs the function of human pancreatic islets.     

Elevated glucose alters eicosanoid release from porcine aortic endothelial cells

Cultured porcine aortic endothelial cells were conditioned through two passages to mimic euglycemic and hyperglycemic conditions (5.2 mM, normal glucose; 15.6 mM, elevated glucose). After incubation with 1 microM [14C]arachidonic acid for 24 h, the cells were stimulated with 1 microM A23187 for times up to 30 min. Uptake of [14C]arachidonic acid and its distribution among cell lipids were unaffected by the increased glucose concentration. The release of eicosanoids from labeled cells and unlabeled cells was measured by reverse-phase HPLC and by RIA, respectively. Compared with cells stimulated in the presence of normal glucose concentrations, cells stimulated in the presence of elevated glucose released 62.6% less free [14C]arachidonic acid, but released 129% more 14C-labeled 15-hydroxyeicosatetraenoic acid (HETE). Increased release of 15-HETE in the presence of elevated glucose in response to A23187, bradykinin, and thrombin was confirmed by RIA. A similar increase in 5-HETE release was observed by RIA after A23187 treatment. The release of both radiolabeled and unlabeled prostanoids was equal at both glucose concentrations. The data indicate that glucose may play an important role in the regulation of release and metabolism of arachidonic acid after agonist stimulation. In the presence of elevated glucose concentrations, such as those associated with diabetes mellitus, the extent and pattern of eicosanoid release from endothelial cells is markedly altered.     

A new hypoglycemic agent, JTT-608, evokes protein kinase A-mediated Ca(2+) signaling in rat islet beta-cells: strict regulation by glucose, link to insulin release, and cooperation with glucagon-like peptide-1(7-36)amide and pituitary adenylate cyclase-activating polypeptide

A new nonsulfonylurea oral hypoglycemic agent, JTT-608, has been reported to stimulate insulin release at elevated, but not low, glucose concentrations and consequently not to induce hypoglycemia in rats. Accordingly, this drug is potentially a safer antidiabetic agent than sulfonylureas. To explore the mechanisms underlying this glucose-dependent insulinotropism, the present study investigated the effects of JTT-608 on cytosolic free Ca(2+) concentration ([Ca(2+)](i)) and protein kinase A (PKA) activity in rat islet beta-cells by microfluorometry using, respectively, fura-2 and a fluorescence PKA substrate, DR II. In the presence of glucose at normal and elevated concentrations (5.0-16.7 mM) JTT-608 (30-1000 microM) concentration dependently increased [Ca(2+)](i) in up to 88% of single beta-cells, whereas at lower glucose concentrations (2.8 and 4.2 mM) it had little effect. The [Ca(2+)](i) responses were inhibited under Ca(2+)-free conditions and by nitrendipine, an L-type Ca(2+) channel blocker. JTT-608 rapidly activated PKA and a PKA inhibitor, H89, inhibited [Ca(2+)](i) responses to JTT-608. JTT-608 also stimulated insulin release from rat islets in a glucose- and Ca(2+)-dependent manner. The glucose-unresponsive beta-cells, which failed to respond to 8.3 mM glucose with increases in [Ca(2+)](i), were frequently recruited to [Ca(2+)](i) increases by JTT-608. JTT-608 also induced oscillations of [Ca(2+)](i). Glucagon-like peptide-1(7-36)amide (GLP-1), pituitary adenylate cyclase-activating polypeptide (PACAP), and acetylcholine (ACh) enhanced the action of JTT-608 on [Ca(2+)](i). In conclusion, JTT-608 evokes PKA-mediated Ca(2+) influx and Ca(2+) signaling in rat islet beta-cells in a glucose-regulated manner, which may account for its glucose-dependent insulinotropism. JTT-608 and neurohormones may cooperatively activate islet beta-cells under physiological conditions.     

Dependency of cyclic AMP-induced insulin release on intra- and extracellular calcium in rat islets of Langerhans.

Calcium and cyclic AMP are important in the stimulation of insulin release. The phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX) raises islet cAMP levels and causes insulin release at nonstimulatory glucose concentrations. In isolated rat pancreatic islets maintained for 2 d in tissue culture, the effects of IBMX on insulin release and 45Ca++ fluxes were compared with those of glucose. During perifusion at 1 mM Ca++, 16.7 mM glucose elicited a biphasic insulin release, whereas 1 mM IBMX in the presence of 2.8 mM glucose caused a monophasic release. Decreasing extracellular Ca++ a monophasic release. Decreasing extracellular Ca++ to 0.1 mM during stimulation reduced the glucose effect by 80% but did not alter IBMX-induced release. Both glucose and IBMX stimulated 45Ca++ uptake (5 min). 45Ca++ efflux from islets loaded to isotopic equilibrium (46 h) was increased by both substances. IBMX stimulation of insulin release, of 45Ca++ uptake, and of efflux were not inhibited by blockade of Ca++ uptake with verapamil, whereas glucose-induced changes are known to be inhibited. Because IBMX-induced insulin release remained unaltered at 0.1 mM calcium, it appears that cAMP-stimulated insulin release is controlled by intracellular calcium. This is supported by perifusion experiments at 0 Ca++ when IBMX stimulated net Ca++ efflux. In addition, glucose-stimulated insulin release was potentiated by IBMX. These results suggest that cAMP induced insulin release is mediated by increases in cytosolic Ca++ and that cAMP causes dislocation of Ca++ from intracellular stores.     

A further evaluation of the effects of K+ depolarization on glutamate-evoked [3H]dopamine release from striatal slices.

Exogenous glutamate will evoke dopamine (DA) release from striatal slices in vitro. To further characterize glutamate-evoked DA release from striatal slices, experiments were designed to: 1) determine if sufficient endogenous glutamate can be released in vitro to presynaptically mediate [3H]DA release in the absence of Mg++ and 2) reevaluate how K+ depolarization affects glutamate-evoked [3H]DA release. Removal of Mg++ to potentiate N-methyl-D-aspartate (NMDA) receptor-mediated DA release increased 15 mM K(+)-evoked [3H]DA release to about 200% of control. The potentiation of this release was probably not mediated by NMDA receptors because it was not blocked by the glutamate receptor antagonists MK-801, 6,7-dinitroquinoxalinedione (DNQX) or kynurenate. Furthermore, the removal of Mg++ increased DA release substantially (200%) in the presence of 5 microM sulpiride and 10 microM nomifensine, indicating that DA reuptake and DA D2 autoreceptors are not primarily responsible for increased DA release. In the absence of Mg++, depolarization produced by 20 mM or greater [K+] inhibited DA released by exogenous glutamate, whereas a much higher [K+] was necessary to evoke endogenous glutamate release. In the presence of 1.5 mM Mg++, a reduction of the "Mg++ blockade" of NMDA receptors by 15 mM K+ depolarization during glutamate-evoked DA release was evaluated with and without the DA reuptake inhibitor nomifensine and the DA D2 antagonist sulpiride. DA released by K+ depolarization (Mg++ present) was markedly increased by 1 mM glutamate, but this effect was only partially reversed by kynurenate or high concentrations of either MK-801 (25 microM) or DNQX (100 microM).(ABSTRACT TRUNCATED AT 250 WORDS)     

Biosynthesis of proinsulin and insulin in newborn rat pancreas. Interaction of glucose, cyclic AMP, somatostatin, and sulfonylureas on the (3H) leucine incorporation into immunoreactive insulin.

The purpose of the present study was to investigate the regulation of insulin biosynthesis during the perinatal period. The incorporation of [3H]leucine into total immunoreactive insulin (IRI) and into IRI fractions was measured by a specific immunoprecipitation procedure after incubation, extraction, and gel filtration in isolated 3-day-old rat pancreases without prior isolation of islets. IRI fractions were identified by their elution profile, their immunological properties, and their ability to compete with the binding of 125 I-insulin in rat liver plasma membranes. No specific incorporation of [3H]leucine was found in the IRI eluted in the void volume, making it unlikely that this fraction behaves as a precursor of (pro) insulin in this system. In all conditions tested, the incorporation of [3H]leucine was linearly correlated with time. Optimal concentration of glucose (11 mM) activated six- to sevenfold the [3H]leucine incorporation into IRI. Theophylline or N6O2-dibutyryl- (db) cAMP at 1.6 mM glucose significantly increased the [3H]leucine incorporation. Glucose at 16.7 mM further enhanced the effect of both drugs. Contrarily, somatostatin (1-10 mug/ml) inhibits the rate of [3H]leucine incorporation into IRI in the presence of 11 mM glucose; this effect was observed at 5.5 mM glucose and was not modified by any further increase in glucose concentrations up to 27.5 mM. Theophylline or dbcAMP at 10 mM concentration did not reverse the somatostatin inhibitory effect on either insulin biosynthesis or release. Somatostatin also inhibited both processes in isolated islets from the 3-day-old rat pancreas. High Ca++ concentration in the incubation medium reversed the inhibitory effect of somatostatin on glucose-induced insulin biosynthesis as well as release. In both systems the inhibitory effect of somatostatin on insulin biosynthesis and release correlated well. Glipizide (10-100 muM) AND TOLBUTAMIDE (400 MUM) inhibited the stimulatory effect of glucose, dbcAMP, and theophylline on [3H]leucine incorporation into IRI. The concentrations of glipizide that were effective in inhibiting [3H]leucine incorporation into IRI were smaller than those required to inhibit the phosphodiesterase activity in isolated islets of 3-day-old rat pancreas. These data suggest the following conclusions: (a) the role of the cAMP-phosphodiesterase system on insulin biosynthesis is likely to be greater in newborns than in adults; (b) the greater effectiveness of glucose and the cAMP system on insulin biosynthesis than on insulin release might possibly be related to the rapid accumulation of pancreatic IRI which is observed in the perinatal period; (c) somatostatin, by direct interaction with the endocrine tissue, can inhibit glucose and cAMP-induced insulin biosynthesis as well as release; calcium reverses this inhibition; (d) sulfonylureas inhibit insulin biosynthesis in newborn rat pancreas an effect which has to be considered in the use of these agents in human disease.     

Muscarinic receptors mediating inhibition of gamma-aminobutyric acid release in rat corpus striatum and their pharmacological characterization

The effects of acetylcholine (ACh) and of cholinergic agonists on the release of tritiated gamma-aminobutyric acid ([3H]GABA) were studied in superfused synaptosomes prepared from rat corpus striatum and prelabeled with the radioactive amino acid. ACh, oxotremorine or (-)-nicotine, all tested at 100 microM had no effect on the spontaneous outflow of [3H]GABA. The depolarization-evoked overflow obtained by exposing the synaptosomes to 9 mM KCl was decreased in a concentration-dependent manner by ACh, oxotremorine, oxotremorine-M or carbachol. The maximal inhibition caused by ACh was 50%. The EC50 (agonist concentration causing half-maximal effect) amounted to 1 microM. Oxotremorine and oxotremorine-M were almost equipotent to ACh, whereas the concentration-response curve of carbachol was slightly (although not significantly) shifted to the right with respect to that of ACh. (-)-Nicotine (100 microM) did not affect the K(+)-evoked [3H]GABA overflow. ACh also inhibited the K(+)-evoked release of endogenous GABA. The inhibitory effect of 10 microM ACh on the release of [3H]GABA evoked by 9 mM KCl was insensitive to the nicotinic antagonist mecamylamine (10 microM) but it was potently blocked by the muscarinic antagonist atropine (IC50 = 5 nM) and weakly antagonized by pirenzepine, dicyclomine and AF-DX 116. The pharmacological profile of this receptor was very similar to that of the muscarinic autoreceptors regulating [3H]ACh release. The extent of [3H]GABA release inhibition caused by ACh did not differ between dorsal and ventral striatum. The inhibitory effect of ACh was much less pronounced in hippocampus and cortex than in the striatum.(ABSTRACT TRUNCATED AT 250 WORDS)