Glucose sensing of individual pancreatic beta-cells involves transitions between steady-state and oscillatory cytoplasmic Ca2+

Amplification of insulin secretion by acetylcholine or phorbol ester is independent of β-cell microfilaments and distinct from metabolic amplification

Insulin secretion (IS) triggered by β-cell [Ca(2+)](c) is amplified by metabolic and receptor-generated signals. Diacylglycerol largely mediates acetylcholine (ACh) effects through protein-kinase C and other effectors, which can be directly activated by phorbol-ester (PMA). Using mouse islets, we investigated the possible role of microfilaments in ACh/PMA-mediated amplification of IS. PMA had no steady-state impact on actin microfilaments. Although ACh slightly augmented and PMA diminished glucose- and tolbutamide-induced increases in β-cell [Ca(2+)](c), both amplified IS in control islets and after microfilament disruption (latrunculin) or stabilization (jasplakinolide). Both phases of IS were larger in response to glucose than tolbutamide, although [Ca(2+)](c) was lower. This difference in secretion, which reflects metabolic amplification, persisted in presence of ACh/PMA and was independent of microfilaments. Amplification of IS by ACh/PMA is thus distinct from metabolic amplification, but both pathways promote acquisition of release competence by insulin granules, which can access exocytotic sites without intervention of microfilaments.

Pulsatility of insulin release--a clinically important phenomenon

The mechanisms and clinical importance of pulsatile insulin release are presented against the background of more than half a century of companionship with the islets of Langerhans. The insulin-secreting beta-cells are oscillators with intrinsic variations of cytoplasmic ATP and Ca(2+). Within the islets the beta-cells are mutually entrained into a common rhythm by gap junctions and diffusible factors (ATP). Synchronization of the different islets in the pancreas is supposed to be due to adjustment of the oscillations to the same phase by neural output of acetylcholine and ATP. Studies of hormone secretion from the perfused pancreas of rats and mice revealed that glucose induces pulses of glucagon anti-synchronous with pulses of insulin and somatostatin. The anti-synchrony may result from a paracrine action of somatostatin on the glucagon-producing alpha-cells. Purinoceptors have a key function for pulsatile release of islet hormones. It was possible to remove the glucagon and somatostatin pulses with maintenance of those of insulin with an inhibitor of the P2Y(1) receptors. Knock-out of the adenosine A(1) receptor prolonged the pulses of glucagon and somatostatin without affecting the duration of the insulin pulses. Studies of isolated human islets indicate similar relations between pulses of insulin, glucagon, and somatostatin as found during perfusion of the rodent pancreas. The observation of reversed cycles of insulin and glucagon adds to the understanding how the islets regulate hepatic glucose production. Current protocols for pulsatile intravenous infusion therapy (PIVIT) should be modified to mimic the anti-synchrony between insulin and glucagon normally seen in the portal blood.

Phospholipase A2 is important for glucose induction of rhythmic Ca2+ signals in pancreatic beta cells

OBJECTIVES

Pancreatic beta cells respond to glucose stimulation with pulses of insulin release generated by oscillatory rises of the cytoplasmic Ca2+ concentration ([Ca2+]i). The observation that exposure to external ATP and other activators of cytoplasmic phospholipase A2 (cPLA2) rapidly induces rises of [Ca2+]i similar to ordinary oscillations made it important to analyze whether suppression of the cPLA2 activity affects glucose-induced [Ca2+]i rhythmicity in pancreatic beta cells.

METHODS

Ratiometric fura-2 technique was used for measuring [Ca2+]i in single beta cells and small aggregates prepared from ob/ob mouse islets.

RESULTS

Testing the effects of different inhibitors of cPLA2 in the presence of 20 mM glucose, it was found that N-(p-amylcinnamoyl)anthranilic acid (ACA) removed the oscillations at a concentration of 25 microM, arachidonyl trifluoromethyl ketone (AACOCF3) at 10 microM, and bromoenol lactone (BEL) at 10 to 15 microM. Withdrawal of ACA and BEL resulted in reappearance of the oscillations. Suppression of the arachidonic acid production by addition of 5 microM of the diacylglycerol lipase inhibitor 1,6-bis-(cyclohexyloximinocarbonylamino)-hexane (RHC 80267) effectively removed the [Ca2+]i oscillations, an effect reversed by removal of the inhibitor or addition of 100 microM tolbutamide. Suppression of the arachidonic acid production had a restrictive influence also on the transients of [Ca2+]i supposed to synchronize the beta-cell rhythmicity. Although less sensitive than the oscillations, most transients disappeared during exposure to 50 microM ACA or 35 microM RHC 80267.

CONCLUSIONS

The results support the idea that cyclic variations of cPLA2 activity are important for the generation and synchronization of the beta-cell [Ca2+]i oscillations responsible for pulsatile release of insulin.

Pulses of external ATP aid to the synchronization of pancreatic beta-cells by generating premature Ca(2+) oscillations

Pancreatic beta-cells respond to glucose stimulation with increase of the cytoplasmic Ca(2+) concentration ([Ca(2+)](i)), manifested as membrane-derived slow oscillations sometimes superimposed with transients of intracellular origin. The effect of external ATP on the oscillatory Ca(2+) signal for pulsatile insulin release was studied by digital imaging of fura-2 loaded beta-cells and small aggregates isolated from islets of ob/ob-mice. Addition of ATP (0.01-100 microM) to media containing 20mM glucose temporarily synchronized the [Ca(2+)](i) rhythmicity in the absence of cell contact by eliciting premature oscillations. External ATP triggered premature [Ca(2+)](i) oscillations also when the sarcoendoplasmic reticulum Ca(2+)-ATPase was inhibited with 50 microM cyclopiazonic acid and phospholipase C inhibited with 10 microM U-73122. The effect of ATP was mimicked by other activators of cytoplasmic phospholipase A(2) (10nM acetylcholine, 0.1-1 micro M of the C-terminal octapeptide of cholecystokinin and 2 microg/ml melittin) and suppressed by an inhibitor of the enzyme (50 microM p-amylcinnamoylanthranilic acid). Premature oscillations generated by pulses of ATP sometimes triggered subsequent oscillations. However, prolonged exposure to high concentrations of the nucleotide (10-100 microM) had a suppressive action on the beta-cell rhythmicity. The early effects of ATP included generation of transients induced by inositol (1,4,5) trisphosphate and superimposed on the premature oscillation or on an ordinary oscillation induced by glucose. The results support the idea that purinergic activation of phospholipase A(2) has a co-ordinating effect on the beta-cell rhythmicity by triggering premature [Ca(2+)](i) oscillations mediated by closure of ATP-sensitive K(+) channels.

Cav1.3 Is Preferentially Coupled to Glucose-Induced [Ca2+]iOscillations in the Pancreatic β Cell Line INS-1

The link between Ca(2+) influx through the L-type calcium channels Ca(v)1.2 or Ca(v)1.3 and glucose- or KCl-induced [Ca(2+)](i) mobilization in INS-1 cells was assessed using the calcium indicator indo-1. Cells responded to 18 mM glucose or 50 mM KCl stimulation with different patterns in [Ca(2+)](i) increases, although both were inhibited by 10 microM nifedipine. Although KCl elicited a prolonged elevation in [Ca(2+)](i), glucose triggered oscillations in [Ca(2+)](i.) Ca(v)1.2/dihydropyridine-insensitive (DHPi) cells and Ca(v)1.3/DHPi cells, and stable INS-1 cell lines expressing either DHP-insensitive Ca(v)1.2 or Ca(v)1.3 channels showed normal responses to glucose. However, in 10 microM nifedipine, only Ca(v)1.3/DHPi cells maintained glucose-induced [Ca(2+)](i) oscillation. In contrast, both cell lines exhibited DHP-resistant [Ca(2+)](i) increases in response to KCl. The percentage of cells responding to glucose was not significantly decreased by nifedipine in Ca(v)1.3/DHPi cells but was greatly reduced in Ca(v)1.2/DHPi cells. In 10 microM nifedipine, KCl-elicited [Ca(2+)](i) elevation was retained in both Ca(v)1.2/DHPi and Ca(v)1.3/DHPi cells. In INS-1 cells expressing the intracellular II-III loop of Ca(v)1.3, glucose failed to elicit [Ca(2+)](i) changes, whereas INS-1 cells expressing the Ca(v)1.2 II-III loop responded to glucose with normal [Ca(2+)](i) oscillation. INS-1 cells expressing Ca(v)1.2/DHPi containing the II-III loop of Ca(v)1.3 demonstrated a nifedipine-resistant slow increase in [Ca(2+)](i) and nifedipine-resistant insulin secretion in response to glucose that was partially inhibited by diltiazem. Thus, whereas the II-III loop of Ca(v)1.3 may be involved in coupling Ca(2+) influx to insulin secretion, distinct structural domains are required to mediate the preferential coupling of Ca(v)1.3 to glucose-induced [Ca(2+)](i) oscillation.

Effects of ELF and static magnetic fields on calcium oscillations in islets of Langerhans

Several experimental studies have produced contradictory results on the effects of extremely low frequency (ELF) magnetic fields on cellular processes involving calcium ions. Furthermore, the few positive results have not been independently replicated. In most of these studies, isolated cells were used. Our study used mouse islets of Langerhans, in which very regular oscillations of calcium concentration can be observed at length. These oscillations are sustained by processes that imply energetic and inter-intracellular communication. Various magnetic fields were applied, either sinusoidal at different frequencies (50 Hz or multiples of the natural oscillation frequency) at 0.1 or 1 mT or static at 1 mT. Islets were also exposed to "cyclotron resonance" conditions. There was neither alteration of the fundamental oscillation frequency nor the degree of organisation under all exposure conditions. Using this sensitive model, we could not show new evidence of alterations of calcium processes under exposure to various magnetic fields.

Ca2+ handling of rat pancreatic beta-cells exposed to ryanodine, caffeine, and glucagon

Reported species differences in the stimulus-secretion coupling of insulin release made it important to compare the Ca2+ handling of rat beta-cells with that previously observed in mice. Single beta-cells and small aggregates were prepared from pancreatic islets of Wistar rats, attached to cover slips and then used for measuring the cytoplasmic Ca2+ concentration ([Ca2+]i) with the ratiometric fura-2 technique. Glucose (11 mM) induced slow oscillations of [Ca2+]i similar to those seen in other species, including humans. Comparison of the oscillations in rat beta-cells with those previously described in mouse revealed that there was a slightly lower frequency and an increased tendency to transformation into sustained [Ca2+]i in response to glucagon or caffeine. Ryanodine (5-20 microM) did not affect existing oscillations but sometimes restored rhythmic activity in the presence of caffeine. Stimulation with glucose resulted not only in oscillations but also in transients of [Ca2+]i sometimes appearing in synchrony in adjacent beta-cells and disappearing after the addition of 200 nM thapsigargin or 20 mM caffeine. The frequency of transients recorded in a medium containing glucagon and methoxyverapamil was higher than seen under similar conditions in mouse beta-cells. Although exhibiting some differences compared with mouse beta-cells, rat beta-cells also have an intrinsic ability to oscillate and to generate the transients of [Ca2+] that are supposed to synchronize the rhythmicity of the islets in the pancreas.

Pancreatic beta-cells from obese-hyperglycemic mice are characterized by excessive firing of cytoplasmic Ca2+ transients

Pancreatic beta-cells from obese-hyperglycemic (ob/ob) mice are widely used for studying the mechanisms of insulin release, including its regulation by the cytoplasmic Ca2+ concentration ([Ca2+]i). In this study, we compared changes of [Ca2+]i in single beta-cells isolated from ob/ob mice with those from lean mice using dual-wavelength microfluorometry and the indicator fura-2. There were no differences in the frequency, amplitude, and half-width of the slow oscillations induced by glucose. Most beta-cells from the obese mice responded to 10 mM caffeine with transformation of the oscillations into sustained elevation of [Ca2+]i, a process counteracted by ryanodine. The beta-cells from the obese mice were characterized by ample generation of [Ca2+]i transients, which increased in number in the presence of glucagon. The transients became less frequent when leptin was added at a concentration as low as 1 nM. It is suggested that the excessive firing of [Ca2+]i transients in the ob/ob mice is owing to the absence of leptin and is mediated by activation of the phospholipase C signaling pathway.

Measurements of cytoplasmic Ca2+ in islet cell clusters show that glucose rapidly recruits beta-cells and gradually increases the individual cell response

The proportion of isolated single beta-cells developing a metabolic, biosynthetic, or secretory response increases with glucose concentration (recruitment). It is unclear whether recruitment persists in situ when beta-cells are coupled. We therefore measured the cytoplasmic free Ca2+ correction ([Ca2+]i) (the triggering signal of glucose-induced insulin secretion) in mouse islet single cells or clusters cultured for 1-2 days. In single cells, the threshold glucose concentration ranged between 6 and 10 mmol/l, at which concentration a maximum of approximately 65% responsive cells was reached. Only 13% of the cells did not respond to glucose plus tolbutamide. The proportion of clusters showing a [Ca2+]i rise increased from approximately 20 to 95% between 6 and 10 mmol/l glucose, indicating that the threshold sensitivity to glucose differs between clusters. Within responsive clusters, 75% of the cells were active at 6 mmol/l glucose and 95-100% at 8-10 mmol/l glucose, indicating that individual cell recruitment is not prominent within clusters; in clusters responding to glucose, all or almost all cells participated in the response. Independently of cell recruitment, glucose gradually augmented the magnitude of the average [Ca2+]i rise in individual cells, whether isolated or associated in clusters. When insulin secretion was measured simultaneously with [Ca2+]i, a good temporal and quantitative correlation was found between both events. However, beta-cell recruitment was maximal at 10 mmol/l glucose, whereas insulin secretion increased up to 15-20 mmol/l glucose. In conclusion, beta-cell recruitment by glucose can occur at the stage of the [Ca2+]i response. However, this type of recruitment is restricted to a narrow range of glucose concentrations, particularly when beta-cell association decreases the heterogeneity of the responses. Glucose-induced insulin secretion by islets, therefore, cannot entirely be ascribed to recruitment of beta-cells to generate a [Ca2+]i response. Modulation of the amplitude of the [Ca2+]i response and of the action of Ca2+ on exocytosis (amplifying actions of glucose) may be more important.

Pathophysiology of impaired pulsatile insulin release

Plasma insulin displays 5-10 min oscillations. In Type 2 diabetes the regularity of the oscillations disappears, which may lead to insulin receptor down-regulation and glucose intolerance and explain why pulsatile delivery of the hormone has a greater hypoglycemic effect than continuous delivery. The rhythm is intrinsic to the islet. Variations in metabolism, cytoplasmic Ca(2+) concentration ([Ca(2+)](i)), other hormones, neuronal signaling and possibly beta-cell insulin receptor expression have been implicated in the regulation of plasma insulin oscillations. Most of these factors are important for amplitude-regulation of the insulin pulses. Although evidence exists supporting a role of both metabolism and [Ca(2+)](i) as pacemakers of the pulses, metabolic oscillations probably have a primary role and [Ca(2+)](i) oscillations a permissive role. Results from islets from animal models of diabetes suggest that altered plasma insulin pattern could be due to lowering of pulse amplitude of insulin oscillations rather than alterations in their frequency. Supporting a role of metabolism, altered plasma insulin oscillations were found in MODY2, MIDD and glycogenosis Type VII, which are linked to alterations in glucokinase, mitochondrial tRNALeu(UUR) and phosphofructokinase. Plasma insulin oscillations require coordination of islet secretory activities in the pancreas. The intrapancreatic ganglia have been suggested as coordinators. The diabetes-associated neuropathy may contribute to the deranged pattern as indicated by glucose intolerance in chagasic patients. Continued investigation of the role and regulation of pulsatile insulin release will lead to better understanding of the pathophysiology of impaired pulsatile insulin release, which could lead to new approaches to restore normal plasma insulin oscillations in diabetes and related diseases.

Norepinephrine acts on the KATP channel and produces different effects on [Ca2+]i in oscillating and non-oscillating HIT-T15 cells

Norepinephrine (NE) is an inhibitor of insulin secretion that acts, in part, by decreasing intracellular free calcium ([Ca2+]i). We examined the effects of NE on [Ca2+]i in individual HIT-T15 cells loaded with indo 1. Cells were categorized as oscillators or non-oscillators on the basis of the pattern of the calcium response to glucose and the effect of NE on [Ca2+]i was subsequently measured in each cell. NE caused a simple decrease in [Ca2+]i in nonoscillators. In oscillators, NE decreased the amplitude and frequency of the oscillations. Furthermore, the duration of the NE effect in oscillators was longer than in non-oscillators. NE did not affect the rise in [Ca2+]i elicited by depolarizing concentrations of 20 mM or 35 mM KCl alone, or in the presence of 20 mM KCl, 100 microM diazoxide, and 10 mM glucose. In other experiments, NE had no effect on [Ca2+]i when the KATP channels were fully clamped with diazoxide or tolbutamide. We conclude that the action of NE to decrease [Ca2+]i in both oscillators and non-oscillators is mediated via activation of the KATP channel. Despite this common mechanism, NE exerts different effects on oscillating and non-oscillating cells.

Glucose-induced pulsatile insulin release from single islets at stable and oscillatory cytoplasmic Ca2+

The cytoplasmic Ca2+ concentration ([Ca2+]i) and insulin release were measured simultaneously in mouse pancreatic islets cultured overnight. [Ca2+]i was 105 nM and insulin release 3 pmol.g-1.s-1 at 3 mM glucose. An increase to 7 mM glucose reduced [Ca2+]i transiently, whereas insulin release doubled and was pulsatile with a frequency of 0.47 min-1. [Ca2+]i oscillations with similar frequency appeared at 11 mM glucose associated with increased amplitude of the insulin oscillations, raising the secretory rate 10-fold. In the presence of 16 and 20 mM glucose [Ca2+]i was > 300 nM and showed no oscillations apart from two islets, which demonstrated [Ca2+]i oscillations with small amplitude at 16 mM glucose. Insulin release with maintained frequency increased by 46 and 31%, respectively. When the glucose concentration was increased from 3 to 11 mM, [Ca2+]i decreased with a nadir that appeared significantly earlier than when the glucose concentration was raised from 3 to 7 mM. Glucose-induced insulin release from the isolated islet is pulsatile both at stable and oscillatory [Ca2+]i, with changes in secretory rate caused by the sugar also when [Ca2+]i is unchanged.

Dynamic aspects for interislet synchronization of oscillatory insulin secretions

How are the oscillatory insulin secretions from numerous islets synchronized to result in an identifiable oscillation? We postulated that a sudden increase in glucose concentration could best account for the interislet synchronization. The perifusion with two parallel chambers each containing 100 islets from the same rat was performed. The glucose concentrations of two chambers were simultaneously increased from 100 to 300 mg/dl in step function to examine the synchronizing efficacy. Synchrony and regularity of insulin oscillation were evaluated by cross-correlation and/or power spectral analysis. Although the insulin had been in stable oscillation, we found that the synchrony between two chambers and the regularity of each chamber were still significantly improved after a sudden increase in glucose level. However, the improved synchrony and regularity were transient. They gradually slid toward a less rigorous condition in a 15-h long-term perifusion. We suggested that the interislet synchronization of oscillatory insulin secretions could be improved by a sudden increase in glucose level. The insulin pulses were therefore enhanced to present their physiological effects.

Suppression of Ca2+ oscillations in glucagon-producing alpha 2-cells by insulin/glucose and amino acids

The cytoplasmic Ca2+ concentration ([Ca2+]i) was continuously monitored in single glucagon-producing alpha 2-cells isolated from the mouse pancreas and later identified by immunostaining. Up to 60% of the alpha 2-cells exhibited spontaneous [Ca2+]i oscillations (frequency 0.1-0.3/min) in a medium containing 3 mM glucose. In originating from a basal level of 60-100 nM, reaching peak values of 300-400 nM and promptly disappearing after blocking voltage-dependent Ca2+ channels with methoxyverapamil, the oscillations resembled those in insulin-releasing beta-cells stimulated by glucose. The oscillatory activity was suppressed when combining elevation of glucose to 20 mM with the addition of 2-2000 ng/ml insulin. Whereas 10 mM of L-arginine or l-glycine transformed the oscillations into sustained elevation of [Ca2+]i, there was no response to 1 mM tolbutamide or 0.1-1 mM gamma-aminobutyric acid. The observations that alpha 2-cells differ from islet cells secreting insulin and somatostatin in responding to adrenaline with mobilisation of intracellular calcium can be used for their rapid identification. It is suggested that the oscillations reflect periodic entry of Ca2+ due to variations of the membrane potential.

Impaired Ca2+ response to glucose in mouse beta-cells infected with coxsackie B or Echo virus

Five strains of Coxsackie B4 virus and one of Echo 11 virus were tested with regard to their ability to replicate in pancreatic mouse beta-cells and interfere with the alterations of the cytoplasmic Ca2+ concentration ([Ca2+]i) induced by glucose. All strains except one both multiplied and caused cytopathic effect. In a control group 68% of the beta-cells responded to 11 mM glucose with large amplitude oscillations of [Ca2+]i. After inoculation with the infectious strains these oscillations appeared in only 5% of the beta-cells, whereas the non-infectious strain did not modify the glucose effect on [Ca2+]i. Despite the virus interference with the glucose response, [Ca2+]i was increased after depolarization with excessive extracellular K+ and the oscillations were induced in most beta-cells when glucose was combined with the insulin-releasing sulfonylurea tolbutamide.

Glucose induces oscillatory Ca2+ signalling and insulin release in human pancreatic beta cells

Mechanisms of pulsatile insulin release in man were explored by studying the induction of oscillatory Ca2+ signals in individual beta cells and islets isolated from the human pancreas. Evidence was provided for a glucose-induced closure of ATP-regulated K+ channels, resulting in voltage-dependent entry of Ca2+. The observation of step-wise increases of capacitance in response to depolarizing pulses suggests that an enhanced influx of Ca2+ is an effective means of stimulating the secretory activity of the isolated human beta cell. Activation of muscarinic receptors (1-10 mumol/l carbachol) and of purinergic P2 receptors (0.01-1 mumol/l ATP) resulted in repetitive transients followed by sustained elevation of the cytoplasmic Ca2+ concentration ([Ca2+]i). Periodic mobilisation of intracellular calcium was seen also when injecting 100 mumol/l GTP-gamma-S into beta cells hyperpolarized to -70 mV. Individual beta cells responded to glucose and tolbutamide with increases of [Ca2+]i, manifested either as large amplitude oscillations (frequency 0.1-0.5/min) or as a sustained elevation. Glucose regulation was based on sudden transitions between the basal and the two alternative states of raised [Ca2+]i at threshold concentrations of the sugar characteristic for the individual beta cells. The oscillatory characteristics of coupled cells were determined collectively rather than by particular pacemaker cells. In intact pancreatic islets the glucose induction of well-synchronized [Ca2+]i oscillations had its counterpart in 2-5 min pulses of insulin. Each of these pulses could be resolved into regularly occurring short insulin transients. It is concluded that glucose stimulation of insulin release in man is determined by the number of beta cells entering into a state with Ca(2+)-induced secretory pulses.

Glucose-dependent alterations of intracellular free calcium by glucagon-like peptide-1(7-36amide) in individual ob/ob mouse beta-cells

Depolarizing concentrations of glucose produce characteristic alterations of intracellular free Ca2+ ([Ca2+]i) in pancreatic beta-cells. The effects of the proposed incretin, glucagon-like peptide-1(7-36amide) (GLP-1a) on [Ca2+]i were determined from Fura-2 fluorescence ratio imaging of cultured ob/ob mouse pancreatic beta-cells. In control cells, [Ca2+]i is low in 3 mM glucose; increasing [glucose] to 8-12 mM results in an initial dip in [Ca2+]i followed by slow oscillating increases in [Ca2+]i. GLP-1a (0.03-10,000 pM) does not alter [Ca2+]i in 3 mM glucose, but does change the response to elevated glucose (8-12 mM). The time integral of the initial dip is reduced ([GLP-1a] 10-100 pM), and the integral of the [Ca2+]i signal is increased ([GLP-1a] > or = 1 pM). GLP-1a increases the frequency of sustained, stable plateau responses to elevated glucose, and the frequency of large, rapid spikes of increased [Ca2+]i associated with either plateaus, or oscillations. Application of a cAMP analog mimics most of the actions of GLP-1a. Activation of the GLP-1a receptor, or application of cAMP alters pancreatic beta-cell [Ca2+]i only when [glucose] is high.

Cytoplasmic Ca2+ oscillations in pancreatic beta-cells

In the last 15 years it has been a growing interest in the cyclic variations of circulating insulin [46]. After the suggestion that this phenomenon may be due to oscillations of the beta-cell membrane potential [8,39], it was demonstrated that [Ca2+]i oscillates in the glucose-stimulated beta-cell with a similar frequency to that of pulsatile insulin release. The present review describes four types of [Ca2+]i oscillations in the pancreatic beta-cell. The slow sinusoidal oscillations, referred to as type-a, are those which most closely correspond to pulsatile insulin release. Although not affecting the properties of the type-a oscillations in individual beta-cells, the concentration of glucose is a determinant for their generation and further transformation into a sustained increase. Accordingly, cytoplasmic Ca2+ is regulated by sudden transitions between oscillatory and steady-state levels at threshold concentrations of glucose, which are characteristic for the individual beta-cell. This behaviour explains the observation of a gradual recruitment of previously non-secreting cells with increase of the extracellular glucose concentration [44]. However, it still remains to be elucidated how the sudden transitions between these three states translate into the co-ordinated slow oscillations of [Ca2+]i in the intact islet. Cyclic variations of circulating insulin require a synchronization of the [Ca2+]i cycles also among the islets in the pancreas. It is still an open question by which means the millions of islets communicate mutually to establish a pattern of pulsatile insulin release from the whole pancreas. The discovery that the beta-cell is not only the functional unit for insulin synthesis but also generates the [Ca2+]i oscillations required for pulsatile insulin release has both physiological and clinical implications. The fact that minor damage to the beta-cells prevents the type-a oscillations with maintenance of a glucose response in terms of raised [Ca2+]i reinforces previous arguments [54] that loss of insulin oscillations is an early indicator of type-2 diabetes. Further analyses of the [Ca2+]i oscillations in the beta-cells should include not only the mechanisms for their generation and subsequent propagation within or among the islets but also how modulation of their frequency affects the insulin sensitivity of various target cells. The latter approach may be important in the attempts to maintain normoglycemia under conditions minimizing the vascular effects of insulin supposed to precipitate hypertonia and atherosclerosis [70,71,77].