Oleic acid interacts with GPR40 to induce Ca2+ signaling in rat islet beta-cells: mediation by PLC and L-type Ca2+ channel and link to insulin release

It has long been thought that long-chain free fatty acids (FFAs) stimulate insulin secretion via mechanisms involving their metabolism in pancreatic beta-cells. Recently, it was reported that FFAs function as endogenous ligands for GPR40, a G protein-coupled receptor, to amplify glucose-stimulated insulin secretion in an insulinoma cell line and rat islets. However, signal transduction mechanisms for GPR40 in beta-cells are little known. The present study was aimed at elucidating GPR40-linked Ca(2+) signaling mechanisms in rat pancreatic beta-cells. We employed oleic acid (OA), an FFA that has a high affinity for the rat GPR40, and examined its effect on cytosolic Ca(2+) concentration ([Ca(2+)](i)) in single beta-cells by fura 2 fluorescence imaging. OA at 1-10 microM concentration-dependently increased [Ca(2+)](i) in the presence of 5.6, 8.3, and 11.2 mM, but not 2.8 mM, glucose. OA-induced [Ca(2+)](i) increases at 11.2 mM glucose were inhibited in beta-cells transfected with small interfering RNA targeted to rat GPR40 mRNA. OA-induced [Ca(2+)](i) increases were also inhibited by phospholipase C (PLC) inhibitors, U73122 and neomycin, Ca(2+)-free conditions, and an L-type Ca(2+) channel blocker, nitrendipine. Furthermore, OA increased insulin release from isolated islets at 8.3 mM glucose, and it was markedly attenuated by PLC and L-type Ca(2+) channel inhibitors. These results demonstrate that OA interacts with GPR40 to increase [Ca(2+)](i) via PLC- and L-type Ca(2+) channel-mediated pathway in rat islet beta-cells, which may be link to insulin release.     

Protein kinase C-dependent and -independent inhibition of Ca(2+) influx by phorbol ester in rat pancreatic beta-cells

Phorbol esters were used to investigate the action of protein kinase C (PKC) on insulin secretion from pancreatic beta-cells. Application of 80 nM phorbol 12-myristate 13-acetate (PMA), a PKC-activating phorbol ester, had little effect on glucose (15 mM)-induced insulin secretion from intact rat islets. In islets treated with bisindolylmaleimide (BIM), a PKC inhibitor, PMA significantly reduced the glucose-induced insulin secretion. PMA decreased the level of intracellular Ca(2+) concentration ([Ca(2+)](i)) elevated by the glucose stimulation when tested in isolated rat beta-cells. This inhibitory effect of PMA was not prevented by BIM. PMA inhibited glucose-induced action potentials, and this effect was not prevented by BIM. Further, 4alpha-phorbol 12,13-didecanoate (4alpha-PDD), a non-PKC-activating phorbol ester, produced an effect similar to PMA. In the presence of nifedipine, the glucose stimulation produced only depolarization, and PMA applied on top of glucose repolarized the cell. When applied at the resting state, PMA hyperpolarized beta-cells with an increase in the membrane conductance. Recorded under the voltage-clamp condition, PMA reduced the magnitude of Ca(2+) currents through L-type Ca(2+) channels. BIM prevented the PMA inhibition of the Ca(2+) currents. These results suggest that activation of PKC maintains glucose-stimulated insulin secretion in pancreatic beta-cells, defeating its own inhibition of the Ca(2+) influx through L-type Ca(2+) channels. PKC-independent inhibition of electrical excitability by phorbol esters was also demonstrated.     

Glucose stimulates Ca2+ influx and insulin secretion in 2-week-old beta-cells lacking ATP-sensitive K+ channels

In adult beta-cells glucose-induced insulin secretion involves two mechanisms (a) a K(ATP) channel-dependent Ca(2+) influx and rise of cytosolic [Ca(2+)](c) and (b) a K(ATP) channel-independent amplification of secretion without further increase of [Ca(2+)](c). Mice lacking the high affinity sulfonylurea receptor (Sur1KO), and thus K(ATP) channels, have been developed as a model of congenital hyperinsulinism. Here, we compared [Ca(2+)](c) and insulin secretion in overnight cultured islets from 2-week-old normal and Sur1KO mice. Control islets proved functionally mature: the magnitude and biphasic kinetics of [Ca(2+)](c) and insulin secretion changes induced by glucose, and operation of the amplifying pathway, were similar to adult islets. Sur1KO islets perifused with 1 mm glucose showed elevation of both basal [Ca(2+)](c) and insulin secretion. Stimulation with 15 mm glucose produced a transient drop of [Ca(2+)](c) followed by an overshoot and a sustained elevation, accompanied by a monophasic, 6-fold increase in insulin secretion. Glucose also increased insulin secretion when [Ca(2+)](c) was clamped by KCl. When Sur1KO islets were cultured in 5 instead of 10 mm glucose, [Ca(2+)](c) and insulin secretion were unexpectedly low in 1 mm glucose and increased following a biphasic time course upon stimulation by 15 mm glucose. This K(ATP) channel-independent first phase [Ca(2+)](c) rise was attributed to a Na(+)-, Cl(-)-, and Na(+)-pump-independent depolarization of beta-cells, leading to Ca(2+) influx through voltage-dependent calcium channels. Glucose indeed depolarized Sur1KO islets under these conditions. It is suggested that unidentified potassium channels are sensitive to glucose and subserve the acute and long-term metabolic control of [Ca(2+)](c) in beta-cells without functional K(ATP) channels.     

Overnight culture unmasks glucose-induced insulin secretion in mouse islets lacking ATP-sensitive K+ channels by improving the triggering Ca2+ signal

A current model ascribes glucose-induced insulin secretion to the interaction of a triggering pathway (K(ATP) channel-dependent Ca(2+) influx and rise in cytosolic [Ca(2+)](c)) and an amplifying pathway (K(ATP) channel-independent augmentation of secretion without further increase of [Ca(2+)](c)). However, several studies of sulfonylurea receptor 1 null mice (Sur1KO) failed to measure significant effects of glucose in their islets lacking K(ATP) channels. We addressed this issue that challenges the model. Compared with controls, fresh Sur1KO islets showed slightly elevated basal [Ca(2+)](c) and insulin secretion. In 15 mm glucose, the absolute rate of secretion was approximately 3-fold lower in Sur1KO than control islets, with only poor increase above base line. Overnight culture of Sur1KO islets in 10 mm glucose (not in 5 mm) augmented basal insulin secretion and considerably improved the response to 15 mm glucose, which reached higher values than in control islets, in which culture had little impact. Glucose stimulation during KCl depolarization showed that the amplifying pathway is functional in fresh and cultured Sur1KO islets. The differences in insulin secretion between fresh and cultured Sur1KO islets and between Sur1KO and control islets were not attributable to differences in insulin content, glucose oxidation rate, or synchronization of [Ca(2+)](c) oscillations. The unmasking of glucose-induced insulin secretion in beta-cells lacking K(ATP) channels is paradoxically due to improvement in the production of a triggering signal (elevated [Ca(2+)](c)). The results show that K(ATP) channels are not the only transducer of glucose effects on [Ca(2+)](c) in beta-cells. They explain controversies in the literature and refute arguments raised against the model implicating an amplifying pathway in glucose-induced insulin secretion.     

INGAP-PP up-regulates the expression of genes and proteins related to K+ ATP channels and ameliorates Ca2+ handling in cultured adult rat islets

Islet Neogenesis Associated Protein (INGAP) increases pancreatic beta-cell mass and potentiates glucose-induced insulin secretion. Here, we investigated the effects of the pentadecapeptide INGAP-PP in adult cultured rat islets upon the expression of proteins constitutive of the K(+)(ATP) channel, Ca(2+) handling, and insulin secretion. The islets were cultured in RPMI medium with or without INGAP-PP for four days. Thereafter, gene (RT-PCR) and protein expression (Western blotting) of Foxa2, SUR1 and Kir6.2, cytoplasmic Ca(2+) ([Ca(2+)](i)), static and dynamic insulin secretion, and (86)Rb efflux were measured. INGAP-PP increased the expression levels of Kir6.2, SUR1 and Foxa2 genes, and SUR1 and Foxa2 proteins. INGAP-PP cultured islets released significantly more insulin in response to 40 mM KCl and 100 muM tolbutamide. INGAP-PP shifted to the left the dose-response curve of insulin secretion to increasing concentrations of glucose (EC(50) of 10.0+/-0.4 vs. 13.7+/-1.5 mM glucose of the controls). It also increased the first phase of insulin secretion elicited by either 22.2 mM glucose or 100 microM tolbutamide and accelerated the velocity of glucose-induced reduction of (86)Rb efflux in perifused islets. These effects were accompanied by a significant increase in [Ca(2+)](i) and the maintenance of a considerable degree of [Ca(2+)](i) oscillations. These results confirm that the enhancing effect of INGAP-PP upon insulin release, elicited by different secretagogues, is due to an improvement of the secretory function in cultured islets. Such improvement is due, at least partly, to an increased K(+)(ATP) channel protein expression and/or changing in the kinetic properties of these channels and augmented [Ca(2+)](i) response. Accordingly, INGAP-PP could potentially be used to maintain the functional integrity of cultured islets and eventually, for the prevention and treatment of diabetes.     

Nitric oxide-cyclic GMP system potentiates glucose-induced rise in cytosolic Ca2+ concentration in rat pancreatic beta-cells.

Involvement of nitric oxide (NO) in the regulation of insulin secretion from pancreatic beta-cells was investigated by measuring cytosolic Ca2+ concentration ([Ca2+]i) in isolated rat pancreatic beta-cells. At 7.0 mM glucose, L-arginine (0.1 mM) elevated [Ca2+]i in about 50% of the beta-cells examined. The response was partially inhibited by an NO synthase inhibitor, N(G)-monomethyl-L-arginine (L-NMA; 0.1 mM), suggesting that part of the response was mediated by the production of NO from L-arginine. D-Arginine at higher concentrations (3 or 10 mM) also increased [Ca2+]i at 7.0 mM glucose; however, the response was not affected by L-NMA (0.1 mM). Similar [Ca2+]i elevation was produced by NO (10 nM) and sodium nitroprusside (SNP; 10 microM) at 7.0 mM glucose. The SNP-induced increase in [Ca2+]i was abolished by nicardipine (1 microM), suggesting that the [Ca2+]i response is mediated by Ca2+ influx through L-type voltage-operated Ca2+ channels. In the presence of oxyhemoglobin (1 microM), the [Ca2+]i elevation induced by NO (10 nM) was abolished. Neither degradation products of NO, NO2- nor NO3-, caused any changes in [Ca2+]i. 8-Bromo-cyclic GMP (8-Br-cGMP; 3 mM) and atrial natriuretic peptide (0.1 microM) elevated [Ca2+]i at 7.0 mM glucose. We conclude that NO, which is produced from L-arginine in pancreatic islets, facilitates glucose-induced [Ca2+]i increase via the elevation of cGMP in rat pancreatic beta-cells. NO-cGMP system may physiologically regulate insulin secretion from pancreatic beta-cells.     

Ca2+-induced Ca2+ release from the endoplasmic reticulum amplifies the Ca2+ signal mediated by activation of voltage-gated L-type Ca2+ channels in pancreatic beta-cells

Stimulus-secretion coupling in pancreatic beta-cells involves membrane depolarization and Ca(2+) entry through voltage-gated L-type Ca(2+) channels, which is one determinant of increases in the cytoplasmic free Ca(2+) concentration ([Ca(2+)](i)). We investigated how the endoplasmic reticulum (ER)-associated Ca(2+) apparatus further modifies this Ca(2+) signal. When fura-2-loaded mouse beta-cells were depolarized by KCl in the presence of 3 mm glucose, [Ca(2+)](i) increased to a peak in two phases. The second phase of the [Ca(2+)](i) increase was abolished when ER Ca(2+) stores were depleted by thapsigargin. The steady-state [Ca(2+)](i) measured at 300 s of depolarization was higher in control cells compared with cells in which the ER Ca(2+) pools were depleted. The amount of Ca(2+) presented to the cytoplasm during depolarization as estimated from the integral of the increment in [Ca(2+)](i) over time (integralDelta[Ca(2+)](i).dt) was approximately 30% higher compared with that in the Ca(2+) pool-depleted cells. neo-thapsigargin, an inactive analog, did not affect [Ca(2+)](i) response. Using Sr(2+) in the extracellular medium and exploiting the differences in the fluorescence properties of Ca(2+)- and Sr(2+)-bound fluo-3, we found that the incoming Sr(2+) triggered Ca(2+) release from the ER. Depolarization-induced [Ca(2+)](i) response was not altered by, an inhibitor of phosphatidylinositol-specific phospholipase C, suggesting that stimulation of the enzyme by Ca(2+) is not essential for amplification of Ca(2+) signaling. [Ca(2+)](i) response was enhanced when cells were depolarized in the presence of 3 mm glucose, forskolin, and caffeine, suggesting involvement of ryanodine receptors in the amplification process. Pretreatment with ryanodine (100 microm) diminished the second phase of the depolarization-induced increase in [Ca(2+)](i). We conclude that Ca(2+) entry through L-type voltage-gated Ca(2+) channels triggers Ca(2+) release from the ER and that such a process amplifies depolarization-induced Ca(2+) signaling in beta-cells.     

Inhibition by nitric oxide of Ca(2+) responses in rat pancreatic alpha-cells

This study examined the effect of nitric oxide (NO) on the cytosolic free Ca(2+) concentration ([Ca(2+)](c)) of alpha-cells isolated from rat pancreatic islets. When extracellular glucose was reduced from 7 to 0 mM, about half of the alpha-cells displayed [Ca(2+)](c) oscillations. Nicardipine, a Ca(2+) channel blocker, terminated the oscillations, while thapsigargine, an inhibitor of Ca(2+)-ATPase on the endoplasmic reticulum, did not affect them, suggesting that the [Ca(2+)](c) oscillations were produced by periodic Ca(2+) influx via L-type voltage-operated Ca(2+) channels. NOC 7, an NO donor, did not cause any changes in [Ca(2+)](c) at 7 mM glucose, but reduced [Ca(2+)](c) or terminated [Ca(2+)](c) oscillations at 0 or 2.8 mM glucose. A similar inhibitory effect on [Ca(2+)](c) of alpha-cells was caused by 8-bromo-cGMP. When the [Ca(2+)](c) of alpha-cells was elevated by L-arginine in the presence of N(omega)-nitro-L-arginine, an NO synthase inhibitor, the subsequent application of NOC 7 and 8-bromo-cGMP reduced [Ca(2+)](c). As there is a direct relationship between [Ca(2+)](c) and glucagon release, these results suggest that the NO-cGMP system in rat pancreatic islets reduces glucagon release by suppressing [Ca(2+)](c) responses in alpha-cells.     

Cell-specific Ca(2+) responses in glucose-stimulated single and aggregated beta-cells

A rise in the cytoplasmic calcium concentration ([Ca(2+)](i)) is a key event for insulin exocytosis. We have recently found that the 'early [Ca(2+)](i) response' in single ob/ob mouse beta-cells is reproduced during consecutive glucose stimulations. It, therefore, appears that the response pattern is a characteristic of the individual beta-cell. We have now investigated if a cell-specific [Ca(2+)](i) response is a general phenomenon in rodent beta-cells, and if it can be observed when cells are functionally coupled. With the use of the fura-2 technique, we have studied the 'early [Ca(2+)](i) response' in single dispersed beta-cells, in beta-cell clusters of different size and in intact islets from the ob/ob mouse during repeated glucose stimulation (20mM). beta-Cells from lean mouse and rat, and intact islets from lean mouse were also investigated. Significant correlations between the first and second stimulation were found for the parameters lag-time for Ca(2+) rise (calculated as the time from start of stimulation of the cell until the first value above an extrapolated baseline), nadir of initial lowering (difference between the baseline and lowest [Ca(2+)](i) value), and peak height (difference between baseline and the highest [Ca(2+)](i) value of the first calcium peak) in single dispersed beta-cells, in 'single beta-cell within a small cluster', in clusters of medium and large size, and in single dispersed beta-cells from lean mouse and rat. The lag-times for Ca(2+) rise and peak heights were correlated within the pairs of stimulation also in intact ob/ob islets. In summary, despite a large heterogeneity of the 'early [Ca(2+)](i) response' among individual cells, the lag-time for [Ca(2+)](i) rise, the nadir of initial lowering and the height of the first peak response can be identified as cell-specific markers in beta-cells.     

Synchronization of pancreatic beta-cell rhythmicity after glucagon induction of Ca2+ transients

Pancreatic beta-cells are biological oscillators requiring a coupling force for the synchronization of the cytoplasmic Ca(2+) oscillations responsible for pulsatile insulin release. Testing the idea that transients, superimposed on the oscillations, are important for this synchronization, the concentration of cytoplasmic Ca(2+) ([Ca(2+)](i)) was measured with ratiometric fura-2 technique in single beta-cells and small aggregates prepared from islets isolated from ob/ob-mice. Image analyses revealed asynchronous [Ca(2+)](i) oscillations in adjacent beta-cells lacking physical contact. The addition of glucagon stimulated the firing of [Ca(2+)](i) transients, which appeared in synchrony in adjacent beta-cells. Moreover, the presence of glucagon promoted synchronization of the [Ca(2+)](i) oscillations in beta-cells separated by a distance <100 microm but not in those >200 microm apart. The results support the proposal that the repolarizing effect of [Ca(2+)](i) transients provides a coupling force for co-ordinating the pulses of insulin release generated by pancreatic beta-cells.     

Synchronization and entrainment of cytoplasmic Ca2+ oscillations in cell clusters prepared from single or multiple mouse pancreatic islets

In contrast to pancreatic islets, isolated beta-cells stimulated by glucose display irregular and asynchronous increases in cytoplasmic Ca(2+) concentration ([Ca(2+)](i)). Here, clusters of 5-30 cells were prepared from a single mouse islet or from pools of islets, loaded with fura-2, and studied with a camera-based system. [Ca(2+)](i) oscillations were compared in pairs of clusters by computing the difference in period and a synchronization index lambda. During perifusion with 12 mM glucose, the clusters exhibited regular [Ca(2+)](i) oscillations that were quasi-perfectly synchronized (Delta period of 1.4% and index lambda close to 1.0) between cells of each cluster. In contrast, separate clusters were not synchronized, even when prepared from one single islet. Pairs of clusters neighboring on the same coverslip were not better synchronized than pairs of clusters examined separately (distinct coverslips). We next attempted to synchronize clusters perifused with 12 mM glucose by applying external signals. A single pulse of 20 mM glucose, 10 mM amino acids, or 10 microM tolbutamide transiently altered [Ca(2+)](i) oscillations but did not reset the clusters to oscillate synchronously. On a background of 12 mM glucose, repetitive applications (1 min/5 min) of 10 microM tolbutamide, but not of 20 mM glucose, synchronized separate clusters. Our results identify a level of beta-cell heterogeneity intermediate between single beta-cells and the whole islet. They do not support the idea that substances released by islet cells serve as paracrine synchronizers. However, synchronization can be achieved by an external signal, if this signal has a sufficient strength to overwhelm the intrinsic rhythm of glucose-induced oscillations and is repetitively applied.     

R-type Ca(2+)-channel-evoked CICR regulates glucose-induced somatostatin secretion

Pancreatic islets have a central role in blood glucose homeostasis. In addition to insulin-producing beta-cells and glucagon-secreting alpha-cells, the islets contain somatostatin-releasing delta-cells. Somatostatin is a powerful inhibitor of insulin and glucagon secretion. It is normally secreted in response to glucose and there is evidence suggesting its release becomes perturbed in diabetes. Little is known about the control of somatostatin release. Closure of ATP-regulated K(+)-channels (K(ATP)-channels) and a depolarization-evoked increase in cytoplasmic free Ca(2+) concentration ([Ca(2+)](i)) have been proposed to be essential. Here, we report that somatostatin release evoked by high glucose (>or=10 mM) is unaffected by the K(ATP)-channel activator diazoxide and proceeds normally in K(ATP)-channel-deficient islets. Glucose-induced somatostatin secretion is instead primarily dependent on Ca(2+)-induced Ca(2+)-release (CICR). This constitutes a novel mechanism for K(ATP)-channel-independent metabolic control of pancreatic hormone secretion.     

Significance of Ca2+, Rb+ fluxes, of cAMP and cGMP for the CCK8-modulated insulin release

In rat pancreatic islets the effects of cholecystokinin-8 (CCK8) on glucose-mediated insulin release, 45Ca2+ net uptake, 45Ca2+ efflux, 86Rb+ efflux, cAMP- and cGMP levels were studied. In the presence of a substimulatory glucose concentration (3 mM) CCK8 concentrations of up to 1 microM had no effect on insulin release, but CCK8 at 10 nM potentiated the stimulatory effect of glucose (11.1 mM). 10 nM CCK8 enhanced glucose-stimulated 45Ca2+ net uptake but was ineffective at substimulatory glucose levels. CCK8 had no effect on cAMP and cGMP levels in the presence of 11.1 mM glucose, CCK8 increased 86Rb+ (a measure of K+) in the presence of both 3 and 11.1 mM glucose. This effect was abolished when Ca2+ was omitted from the perifusion medium. CCK8 did not alter glucose (11.1 mM)-stimulated 45Ca2+ efflux rate. These data indicate that (1) CCK8 potentiates glucose-stimulated insulin secretion possibly via an effect on Ca2+ uptake, 2) by affecting Ca2+ uptake, CCK8 enhances K+ efflux, and 3) CCK8 does not mediate its effect via cAMP or cGMP. With respect to 86Rb+ efflux the mechanism of CCK8 action appears to be different from that of glucose. When the mechanism of CCK action on islets is compared with that on exocrine pancreas (data from others) there are similarities (importance of Ca2+ uptake and non-importance of cAMP and cGMP).     

Normalization of intracellular Ca(2+) induces a glucose-responsive state in glucose-unresponsive beta-cells

Although intracellular Ca(2+) in pancreatic beta-cells is the principal signal for insulin secretion, the effect of chronic elevation of the intracellular Ca(2+) concentration ([Ca(2+)](i)) on insulin secretion is poorly understood. We recently established two pancreatic beta-cell MIN6 cell lines that are glucose-responsive (MIN6-m9) and glucose-unresponsive (MIN6-m14). In the present study we have determined the cause of the glucose unresponsiveness in MIN6-m14. Initially, elevated [Ca(2+)](i) was observed in MIN6-m14, but normalization of the [Ca(2+)](i) by nifedipine, a Ca(2+) channel blocker, markedly improved the intracellular Ca(2+) response to glucose and the glucose-induced insulin secretion. The expression of subunits of ATP-sensitive K(+) channels and voltage-dependent Ca(2+) channels were increased at both mRNA and protein levels in MIN6-m14 treated with nifedipine. As a consequence, the functional expression of these channels at the cell surface, both of which are decreased in MIN6-m14 without nifedipine treatment, were increased significantly. Contrariwise, Bay K8644, a Ca(2+) channel agonist, caused severe impairment of glucose-induced insulin secretion in glucose-responsive MIN6-m9 due to decreased expression of the channel subunits. Chronically elevated [Ca(2+)](i), therefore, is responsible for the glucose unresponsiveness of MIN6-m14. The present study also suggests normalization of [Ca(2+)](i) in pancreatic beta-cells as a therapeutic strategy in treatment of impaired insulin secretion.     

The Ca2+ dynamics of isolated mouse beta-cells and islets: implications for mathematical models

[Ca(2+)](i) and electrical activity were compared in isolated beta-cells and islets using standard techniques. In islets, raising glucose caused a decrease in [Ca(2+)](i) followed by a plateau and then fast (2-3 min(-1)), slow (0.2-0.8 min(-1)), or a mixture of fast and slow [Ca(2+)](i) oscillations. In beta-cells, glucose transiently decreased and then increased [Ca(2+)](i), but no islet-like oscillations occurred. Simultaneous recordings of [Ca(2+)](i) and electrical activity suggested that differences in [Ca(2+)](i) signaling are due to differences in islet versus beta-cell electrical activity. Whereas islets exhibited bursts of spikes on medium/slow plateaus, isolated beta-cells were depolarized and exhibited spiking, fast-bursting, or spikeless plateaus. These electrical patterns in turn produced distinct [Ca(2+)](i) patterns. Thus, although isolated beta-cells display several key features of islets, their oscillations were faster and more irregular. beta-cells could display islet-like [Ca(2+)](i) oscillations if their electrical activity was converted to a slower islet-like pattern using dynamic clamp. Islet and beta-cell [Ca(2+)](i) changes followed membrane potential, suggesting that electrical activity is mainly responsible for the [Ca(2+)] dynamics of beta-cells and islets. A recent model consisting of two slow feedback processes and passive endoplasmic reticulum Ca(2+) release was able to account for islet [Ca(2+)](i) responses to glucose, islet oscillations, and conversion of single cell to islet-like [Ca(2+)](i) oscillations. With minimal parameter variation, the model could also account for the diverse behaviors of isolated beta-cells, suggesting that these behaviors reflect natural cell heterogeneity. These results support our recent model and point to the important role of beta-cell electrical events in controlling [Ca(2+)](i) over diverse time scales in islets.     

Ca2+-induced Ca2+ release by activation of inositol 1,4,5-trisphosphate receptors in primary pancreatic beta-cells

The effect of sarcoendoplasmic reticulum Ca(2+)-ATPase (SERCA) inhibition on the cytoplasmic Ca(2+) concentration ([Ca(2+)](i)) was studied in primary insulin-releasing pancreatic beta-cells isolated from mice, rats and human subjects as well as in clonal rat insulinoma INS-1 cells. In Ca(2+)-deficient medium the individual primary beta-cells reacted to the SERCA inhibitor cyclopiazonic acid (CPA) with a slow rise of [Ca(2+)](i) followed by an explosive transient elevation. The [Ca(2+)](i) transients were preferentially observed at low intracellular concentrations of the Ca(2+) indicator fura-2 and were unaffected by pre-treatment with 100 microM ryanodine. Whereas 20mM caffeine had no effect on basal [Ca(2+)](i) or the slow rise in response to CPA, it completely prevented the CPA-induced [Ca(2+)](i) transients as well as inositol 1,4,5-trisphosphate-mediated [Ca(2+)](i) transients in response to carbachol. In striking contrast to the primary beta-cells, caffeine readily mobilized intracellular Ca(2+) in INS-1 cells under identical conditions, and such mobilization was prevented by ryanodine pre-treatment. The results indicate that leakage of Ca(2+) from the endoplasmic reticulum after SERCA inhibition is feedback-accelerated by Ca(2+)-induced Ca(2+) release (CICR). In primary pancreatic beta-cells this CICR is due to activation of inositol 1,4,5-trisphosphate receptors. CICR by ryanodine receptor activation may be restricted to clonal beta-cells.     

Contribution of the endoplasmic reticulum to the glucose-induced [Ca(2+)](c) response in mouse pancreatic islets

Thapsigargin (TG), a blocker of Ca(2+) uptake by the endoplasmic reticulum (ER), was used to evaluate the contribution of the organelle to the oscillations of cytosolic Ca(2+) concentration ([Ca(2+)](c)) induced by repetitive Ca(2+) influx in mouse pancreatic beta-cells. Because TG depolarized the plasma membrane in the presence of glucose alone, extracellular K(+) was alternated between 10 and 30 mM in the presence of diazoxide to impose membrane potential (MP) oscillations. In control islets, pulses of K(+), mimicking regular MP oscillations elicited by 10 mM glucose, induced [Ca(2+)](c) oscillations whose nadir remained higher than basal [Ca(2+)](c). Increasing the depolarization phase of the pulses while keeping their frequency constant (to mimic the effects of a further rise of the glucose concentration on MP) caused an upward shift of the nadir of [Ca(2+)](c) oscillations that was reproduced by raising extracellular Ca(2+) (to increase Ca(2+) influx) without changing the pulse protocol. In TG-pretreated islets, the imposed [Ca(2+)](c) oscillations were of much larger amplitude than in control islets and occurred on basal levels. During intermittent trains of depolarizations, control islets displayed mixed [Ca(2+)](c) oscillations characterized by a summation of fast oscillations on top of slow ones, whereas no progressive summation of the fast oscillations was observed in TG-pretreated islets. In conclusion, the buffering capacity of the ER in pancreatic beta-cells limits the amplitude of [Ca(2+)](c) oscillations and may explain how the nadir between oscillations remains above baseline during regular oscillations or gradually increases during mixed [Ca(2+)](c) oscillations, two types of response observed during glucose stimulation.     

Glucose-insensitivity induced by Ca(2+) toxicity in islet beta-cells and its prevention by PACAP

The present study examined whether a sustained increase in cytosolic Ca(2+) concentration ([Ca(2+)](i)) causes glucose-insensitivity in beta-cells and whether it could be modulated by pituitary adenylate cyclase-activating polypeptide (PACAP), a pancreatic insulinotropin. Rat single beta-cells were cultured for 2 days with sustained increases in [Ca(2+)](i), followed by determination of the [Ca(2+)](i) response to glucose (8.3 mM) as monitored with fura-2. High K(+) (25 mM) produced sustained increases in [Ca(2+)](i) in beta-cells, which were inhibited by nifedipine, a Ca(2+) channel blocker. After culture with high K(+), the incidence and amplitude of [Ca(2+)](i) responses to glucose were markedly reduced. This glucose-insensitivity was prevented by the presence of nifedipine or PACAP-38 (10(-13) M and 10-9) M) in high K(+) culture. PACAP-38 attenuated high K(+)-induced [Ca(2+)](i) increases. In conclusion, sustained increases in [Ca(2+)](i) induce glucose-insensitivity (Ca(2+) toxicity in beta-cells) and it is prevented by PACAP possibly in part due to its Ca(2+)-reducing capacity.     

The L-type voltage-gated Ca2+ channel is the Ca2+ sensor protein of stimulus-secretion coupling in pancreatic beta cells

L-type voltage-gated Ca2+ channels (Cav1.2) mediate a major part of insulin secretion from pancreatic beta-cells. Cav1.2, like other voltage-gated Ca2+ channels, is functionally and physically coupled to synaptic proteins. The tight temporal coupling between channel activation and secretion leads to the prediction that rearrangements within the channel can be directly transmitted to the synaptic proteins, subsequently triggering release. La3+, which binds to the polyglutamate motif (EEEE) comprising the selectivity filter, is excluded from entry into the cells and has been previously shown to support depolarization-evoked catecholamine release from chromaffin and PC12 cells. Hence, voltage-dependent trigger of release relies on Ca2+ ions bound at the EEEE motif and not on cytosolic Ca2+ elevation. We show that glucose-induced insulin release in rat pancreatic islets and ATP release in INS-1E cells are supported by La3+ in nominally Ca2+-free solution. The release is inhibited by nifedipine. Fura 2 imaging of dispersed islet cells exposed to high glucose and La3+ in Ca2+-free solution detected no change in fluorescence; thus, La3+ is excluded from entry, and Ca2+ is not significantly released from intracellular stores. La3+ by interacting extracellularlly with the EEEE motif is sufficient to support glucose-induced insulin secretion. Voltage-driven conformational changes that engage the ion/EEEE interface are relayed to the exocytotic machinery prior to ion influx, allowing for a fast and tightly regulated process of release. These results confirm that the Ca2+ channel is a constituent of the exocytotic complex [Wiser et al. (1999) PNAS 96, 248-253] and the putative Ca2+-sensor protein of release.     

Autocrine regulation of glucose metabolism in pancreatic islets

We evaluated the possible autocrine modulatory effect of insulin on glucose metabolism and glucose-induced insulin secretion in islets isolated from normal hamsters. We measured 14CO2 and 3H2O production from d-[U-14C]glucose and d-[5-3H]glucose, respectively, in islets incubated with 0.6, 3.3, 8.3, and 16.7 mM glucose alone or with 5 or 15 mU/ml insulin, anti-insulin guinea pig serum (1:500), 25 microM nifedipine, or 150 nM wortmannin. Insulin release was measured (radioimmunoassay) in islets incubated with 3.3 or 16.7 mM glucose with or without 75, 150, and 300 nM wortmannin. Insulin significantly enhanced 14CO2 and 3H2O production with 3.3 mM glucose but not with 0.6, 8.3, or 16.7 mM glucose. Addition of anti-insulin serum to the medium with 8.3 and 16.7 mM glucose decreased 14CO2 and 3H2O production significantly. A similar decrease was obtained in islets incubated with 8.3 and 16.7 mM glucose and wortmannin or nifedipine. This latter effect was reversed by adding 15 mU/ml insulin to the medium. Glucose metabolism was almost abolished when islets were incubated in a Ca2+-deprived medium, but this effect was not reversed by insulin. No changes were found in 14CO2 and 3H2O production by islets incubated with 3.3 mM glucose and anti-insulin serum, wortmannin, or nifedipine in the media. Addition of wortmannin significantly decreased insulin release induced by 16.7 mM glucose in a dose-dependent manner. Our results suggest that insulin exerts a physiological autocrine stimulatory effect on glucose metabolism in intact islets as well as on glucose-induced insulin release. Such an effect, however, depends on the glucose concentration in the incubation medium.