[Frontiers in Bioscience S3, 662-679, January 1, 2011]

Studies of first phase insulin secretion using imposed plasma membrane depolarization

Michael Willenborg1, Kathrin Hatlapatka1, Hany Ghaly1, Michael Belz1, Uwe Panten1, Ingo Rustenbeck1

1Institute of Pharmacology and Toxicology, University of Braunschweig, D-38106 Braunschweig, Germany

TABLE OF CONTENTS

1. Abstract
2. The stimulus secretion-coupling of insulin secretion: substrate site hypothesis vs. receptors site hypothesis
3. The phasic nature of insulin secretion
4. Triggering and amplifying pathways of insulin secretion and their relation with the biphasic kinetics of insulin secretion
5. Consequences of KATP channel block: the unsolved problem of imidazolines
6. Insulin secretion by K+ depolarization: which K+ concentration corresponds to the KATP channel block?
7. Nutrient stimulation of insulin secretion after prior depolarization
8. Signalling role of the action potential for insulin secretion
9. Depolarization by current injection: how to reconcile membrane capacitance measurements with secretion measurements
10. Summary and hypotheses
11. Acknowledgements
12. References

1. ABSTRACT

The first phase of glucose-induced insulin secretion is generally regarded to represent the release of a finite pool of secretion-ready granules, triggered by the depolarization-induced influx of Ca2+ through L-type Ca2+ channels. However, the experimental induction of insulin secretion by imposed plasma membrane depolarization may be more complicated than currently appreciated. A comparison of the effects of high K+ concentrations with those of KATP channel closure, which initiates the electrical activity of the beta cell, suggests that 40 mM K+, which is a popular tool to produce a first phase-like secretion, is of supraphysiological strength, whereas the 20 mV depolarization by 15 mM K+ is nearly inefficient. A major conceptual problem consists in the occurrence of action potentials during KATP channel closure, but not during K+ depolarization, which leaves the K+ channel conductance unchanged. Recent observations suggest that the signal function of the endogenously generated depolarization is not homogeneous, but may rather differ between the component mainly determined by KATP channel closure (slow waves) and that mainly determined by Ca2+ influx (action potentials).

2. THE STIMULUS SECRETION-COUPLING OF INSULIN SECRETION: SUBSTRATE SITE HYPOTHESIS VS. RECEPTORS SITE HYPOTHESIS

The unique characteristic of the pancreatic beta cell is to transform the perceived availability of a limited range of nutrients (glucose, some other carbohydrates and a few amino acids and keto acids) into a suitably conditioned release of a peptide hormone, insulin. The main physiological nutrient to sense is glucose. The insulin release has to be integrated into the requirements of the whole organism, thus under physiological conditions the secretory response to glucose is modified by an array of neuronal and hormonal signals. As a general rule, nutrients are stimulators of the secretion, whereas neurotransmitters and hormones are enhancers, i.e. they require the presence of some nutrient to become effective. While there are inhibitory neurotransmitters and hormones, there is apparently no physiological inhibitory nutrient or nutrient analogue.

In the early days of research on stimulus secretion-coupling in the beta cell, there was a debate whether glucose and the other nutrient secretagogues elicit insulin by binding to a plasma membrane receptor or whether the signal for secretion was generated by their intracellular metabolism (1-3). By the early 1980s the latter hypothesis prevailed (4), but there was still a major gap in understanding of how the energy metabolism was coupled to the electrical activity of the beta cell which was apparently responsible for initiating the exocytosis of the insulin-containing large dense core vesicles. This activity was originally described by Dean and Matthews (5), followed by Meissner and Henquin�s characterization of the spike and wave pattern typical for the beta cell (6), the mechanisms of which are still incompletely understood. The identification of the ATP-sensitive potassium channel (KATP channel) in the beta cell and its clear association with the nutrient-induced depolarization (7-9) permitted for the first to time to relate all the essential features of the beta cell in a logical sequence. Thus the sequence of nutrient-induced increase of the ATP-to-ADP ratio, closure of the KATP channels, depolarisation-induced opening of voltage-dependent Ca2+ channels and the ensuing Ca2+-induced exocytosis of insulin granules became generally accepted. It is often referred to as the "consensus hypothesis" and represents the core of textbook knowledge to this day. However, it must be kept in mind that this sequence of events is very elementary and additional regulatory mechanisms are clearly required (see below). E.g. it does not explain one of the very basic features of glucose-induced insulin secretion, the biphasic kinetic.

3. THE PHASIC NATURE OF INSULIN SECRETION

The endocrine pancreas responds to a square-wave glucose stimulus in a unique, biphasic dynamic of insulin release rate (10). The first phase, once believed to be an in vitro artifact, is now recognized to be of central importance for the tight control of glucose homeostasis (11). Experimental evidence suggests that the efficiency of insulin is dependent on the kinetic of its release. A pulsatile secretion pattern is more effective to lower blood glucose than a constant infusion of the same amount of insulin. Most likely this is due to a preferential effect of high insulin concentration on the liver (12, 13). Thus the first phase secretion is of major physiological importance. Since postprandial glucose concentrations do not increase in a square wave-manner, there is no classical first phase response in vivo, however, the fast initial increase in insulin secretion (acute insulin response) can be regarded as the equivalent of the first phase (14).

One of the earliest derangements of type 2 diabetes mellitus is a diminished first phase insulin response to glucose; subsequently as the disease progresses, rates of second phase insulin response decline as well. While there is some debate as to whether the decline of the first phase is actually preceding the decline of the second phase, it is certainly contributing to the progressive impairment of glucose homeostasis (14, 15). Thus a pharmacological reconstruction of the first phase can be considered as a therapeutic option in the treatment of type 2 diabetes. To do this rationally, however, requires an understanding of the mechanisms which have become deficient.

4. TRIGGERING AND AMPLIFYING PATHWAYS OF INSULIN SECRETION AND THEIR RELATION WITH THE BIPHASIC KINETICS OF INSULIN SECRETION

Two independent observations paved the way to the realization that the free cytosolic Ca2+ concentration ((Ca2+)i) was not the final common pathway of stimulus secretion-coupling it was originally presumed to be. Glucose was found to further stimulate insulin secretion when a maximally effective concentration of a sulfonylurea had closed virtually all KATP channels of the beta cell (16). Similarly, when the KATP channels were clamped open by a high diazoxide concentration and the plasma membrane was depolarized by a high external K+ concentration, the addition of glucose still produced a substantial increase in secretion (17).

Initially, this additional pathway was referred to as "KATP channel-independent" pathway (18, 19). Later the term "amplifying pathway" was proposed (20), because under physiological conditions it cannot stimulate insulin secretion as long as the KATP channels remain open and thus it is not truly independent on KATP channel function. The KATP channel-dependent pathway was proposed to produce the indispensable signal for exocytosis, namely the depolarization-induced Ca2+ influx, therefore the term "triggering pathway" was coined (20). While the bifurcation of the nutrient-dependent stimulus secretion coupling is generally recognized (Figure 1), the fact that the mechanism of metabolic amplification has remained elusive up to now suggests that major piece of the puzzle may still be lacking.

The predominant hypothesis to explain the biphasic kinetics of glucose-induced secretion proposes that not all of the ca. 10000 granules in the beta cell are functionally equivalent, but rather that there are different pools of granules at different stages of maturation, only one of which contains secretion-ready granules and is able to react to the depolarization-triggered Ca2+ influx. The nadir separating the first phase from the second phase is explained as the exhaustion of this pool which was estimated to comprise about 50 to 100 granules (21). Since these granules require only the appropriate Ca2+ increase in their vicinity as the final triggering signal, it is a logical consequence of this hypothesis that an activated energy metabolism is not required for the first phase. Experimental evidence has been presented in support of this contention (22). More recently, however, the amplifying pathway was found to affect first phase secretion (10), an observation which does not fit easily into the framework of the above hypothesis.

The second phase would then result from the increasing velocity of granule translocation from the reserve pool and the concomitant granule maturation, which is known to be dependent on activated energy metabolism (23). Thus, the biphasic kinetics might be the expression of two different thresholds for one signal derived from the glucose metabolism (e.g. ATP) or of two different signalling pathways with the amplifying pathway being an obvious candidate to regulate the second phase (24). This sequence of events is at the core of what may be called the "pool" hypothesis of the biphasic kinetics and implies the following testable predictions: 1) A depolarization which is sufficiently strong to induce a Ca2+ influx into the beta cell is sufficient to trigger insulin secretion with a first phase-like characteristic. 2) A block of KATP channels leading to depolarization and Ca2+ influx is sufficient to trigger insulin secretion. 3) A drug which blocks KATP channels will stimulate insulin secretion at substimulatory glucose concentrations and even in the absence of glucose (provided it does not exert additional, opposing effects more downstream).

5. CONSEQUENCES OF KATP CHANNEL BLOCK: THE UNSOLVED PROBLEM OF IMIDAZOLINES

The mechanism of action of imidazolines, a group of investigative antidiabetic drugs, has remained elusive in spite of considerable research effort, both from basic scientists and pharmaceutical manufacturers. The property which made these compounds interesting is the marked glucose-dependency of the insulinotropic effect, which gave rise to the expectation that an insulinotropic antidiabetic drug without risk of hypoglycaemia could be developed (25). In contrast to sulfonylureas the imidazolines (or, more precisely, several compounds in this group) do not stimulate insulin secretion at a basal glucose concentration, but enhance the efficiency of stimulatory glucose concentrations.

Several observations made while exploring the mechanism of action are not easily explained by the above described hypothesis of stimulus secretion-coupling. While some newer imidazoline derivatives do not affect KATP channels, older compounds like alinidine or efaroxan block KATP channels by binding to the pore-forming subunit Kir 6.2 (26, 27), but are nevertheless without insulinotropic effect in the presence of a basal glucose concentration. This property is unexpected in view of the above predictions. Specifically, KATP channel-blocking imidazolines depolarized the plasma membrane, producing action potential spikes and increasing (Ca2+)i at a basal glucose concentration but did not increase insulin secretion under the same conditions (28-30). In contrast, the depolarization by 40 mM K+ reproducibly led to an increase of (Ca2+)i as well as of insulin secretion, even in the presence of an imidazoline which had led to such a dissoziation between (Ca2+)i increase and secretion (Figure 2).

So the question arose whether the above dissociation was a specific property of the imidazoline compounds or whether it could also be observed by varying the strength of the KCl depolarization. The methodology to address this problem comprises mainly secretion measurements of batch-perifused islets, measurements of (Ca2+)i in perifused islets and measurements of the beta cell plasma membrane potential using the perforated patch configuration of the patch-clamp technique (for methodological details see refs 30, 31, 53).

6. INSULIN SECRETION BY K+ DEPOLARIZATION: WHICH K+ CONCENTRATION CORRESPONDS TO THE KATP CHANNEL BLOCK?

The depolarizing effect of an elevated K+ concentration in the extracellular space can be predicted by the Goldman-Hodgkin-Katz equation. To ascertain the validity of our membrane potential measurements we checked whether the depolarization by stepwise elevating the extracellular K+ concentration from 5.6 mM to 10, 20, 30 and 40 mM corresponded to the theoretical values. This was indeed the case, not only in the conventional whole cell configuration but also using the perforated patch technique which leaves the cell metabolism intact. Obviously, a K+ depolarization affects the beta cell membrane potential only via the Nernst equilibrium and not by secondary biochemical responses, at least not within the given time frame (31). This is a clear advantage as compared to pharmacological tools, e.g. sulfonylureas, where additional sites of action, at least in the higher concentration range must be taken into account (32).

In current thinking, the potential of the plasma membrane can be regarded as the central parameter of the triggering pathway, which integrates a variety of ion channel activities to yield the opening signal for the L-type Ca2+ channels, which in turn are linked to the very last step of Ca2+ triggered exocytosis (33). Recently there has been some debate on the role of other voltage-dependent Ca2+ channels (34), but at least during the first phase L-type Ca2+ channels play a decisive role for the initiation of secretion both in rodent and human islets (35, 36).

All other factors being equal, a monotonous relation between the strength of depolarization and rate of exocytosis ought to be expected. So it should be possible to standardize experimentation by using insulin secretagogues at an equi-depolarizing strength. However, the pattern of K+ depolarization and that of depolarization by KATP channel closure is obviously different. While KATP channel closure produces a plateau depolarization with superimposed action potentials (corresponding to waves and spikes of the glucose-induced depolarization), only a plateau can be seen during K+ depolarization (Figure 3).

The question now was as to whether the spike or the plateau value of depolarization by KATP channel block corresponds to the constant membrane potential during K+ depolarization (37). Or, alternatively, is the area under the curve of the action potentials proportional to the opening signal to the Ca2+ channels and should this area be added to the plateau depolarization to give an effective depolarizing strength? To solve this problem we asked whether the processes underlying the action potential are also operative during K+ depolarization and in case it is so, why they don�t become apparent. Since it is known that the action potential spiking results from Ca2+ influx (38, 39) and since it is known that K+ depolarization leads to Ca2+ influx (both via L-type Ca2+ channels, ref 40), one might presume that at least a part of the depolarization amplitude by 40 mM K+ corresponds to the action potential amplitude.

So the effect of L-type Ca2+ channel blockade on both types of depolarization was compared (Figure 4A). As expected, 10 �M nifedipine abolished action potential spiking, but had a surprisingly modest effect on the depolarization by 40 mM K+. Apparently, L-type Ca2+ channel activity contributes only about 5% to the 41 mV depolarization amplitude produced by 40 mM K+, even though nifedipine strongly reduced the (Ca2+)i increase by 40 mM K+. (I.R., unpublished). How then to explain the absence of action potentials during K+ depolarization? Since K+ depolarization does not affect KATP channel open probability, which determines by far the largest K+ conductance of the beta cell plasma membrane (41), K+ ions can freely leave the cell interior. To demonstrate the consequences of open K+ channels for the depolarization pattern, a beta cell was first depolarized by 10 mM glucose (producing action potentials) then, additionally, by 15 mM K+, which shifted the plateau and the action potentials by ca. 9 mV towards zero. Opening of the KATP channels by diazoxide immediately abolished action potential spiking, but only slightly diminished the plateau depolarization. (Figure 4B). Washout of high K+ in the presence of diazoxide re-established the resting membrane potential.

This experiment makes clear that action potentials which represent L-type Ca2+ channel activity (virtually complete block by nifedipine) can be diminished by increasing K+ channel activity. How are these currents related? As shown in the schematic drawing in Figure 5, the depolarization-induced Ca2+ influx further depolarizes the plasma membrane, probably only by a few millivolts, as is suggested by the above experiment. In doing so, it reduces the electric field which is constantly counteracting the concentration-driven K+ efflux. As a rule, the K+ efflux is the faster the farther away the actual membrane potential is from the K+ equilibrium potential. Apparently, the increased K+ efflux quickly and exactly balances the Ca2+ influx leading to a very smooth depolarization pattern (Figure 5B). Only when KATP channels are closed (e.g. by a sulfonylurea, Figure 5C, see also Figure 10) does the Ca2+ influx produce a marked further depolarization which is terminated both by Ca2+ channel inactivation and opening of voltage-dependent (Kv 2.1) and Ca2+-dependent (BK and SK) K+ channels (42, 43). The contribution of the latter to the K+ conductance is much lower than that of KATP channels (41).

To sum up the above: a Ca2+ influx takes place during K+ depolarization, but does not further depolarize the membrane. The conclusion from these considerations for practical experimentation is that, to explore the initial events in glucose-induced insulin, a K+ concentration of ca. 15 mM rather than 40 mM should be used to mimic the depolarizing strength of a KATP channel closure.

7. NUTRIENT STIMULATION OF INSULIN SECRETION AFTER PRIOR DEPOLARIZATION

The use of 15 mM instead of 40 mM K+ to produce a first phase-like secretion resulted in a dramatically different response. The secretion was only modestly and transiently increased by 15 mM K+, whereas 40 mM gives the more familiar view of an immediate and strong response (Figure 6A). 15 mM K+ produced the necessary triggering signal by increasing the (Ca2+)i through influx from the extracellular space, even though the response to 40 mM was clearly higher (Figure 6B). Another interesting feature was the different response to an increase of the glucose concentration in the continuing presence of 15 or 40 mM K+. Whereas 10 mM glucose can only partially reverse the continuous decline of the secretion rate after the peak secretion established by 40 mM K+, the response to 10 mM glucose was clearly enhanced by the presence of 15 mM K+ with a distinct first phase characteristic. Of note, the modest effect of 10 mM glucose in the presence of 40 mM K+ is at variance with a published observation (24), but this may be a consequence of a more marked insulinotropic effect of the high K+ concentration in our experiments, which appears to induce some sort of time-dependent inhibition (I.R., unpublished).

Now, the theoretical problem is whether the first phase corresponds to the secretion increase produced by the depolarization-stimulated Ca2+ influx in the presence of a basal glucose concentration or whether the increase caused by the elevation of glucose is the "true" first phase response. If we were to follow the hypothesis outlined above, the response to the K+ depolarization ought to represent the first phase. But if so, why is it so small when a physiologically relevant strength of depolarization is applied? On the other hand, the insulin release caused by 40 mM K+ is much more pronounced than the one during the first 10 min after raising the glucose concentration from 5 to 10 mM. If we stick to the original description of the first phase as a consequence of a square wave-glucose stimulation (44, 45), then we have to state that glucose was able to produce a first phase response even though a prior increase in (Ca2+)i had taken place.

This raises the question as to whether a finite pool of granules can be responsible for the phasic nature of insulin secretion. Ideally, this information could be provided by TIRF microscopy, which can selectively detect submembrane granules (46). However, the calculated number of granules in this pool makes up less than 10% of the total submembrane granules (21), making a clear-cut detection of a pool-depletion by optical techniques quite difficult. On the other hand, the pool concept is primarily based on functional evidence, thus it can be tested by functional experiments. Since 15 mM K+ produced a (Ca2+)i increase (Figure 6B) which was not inferior to that by KATP channel closure (31) a pool depletion analogous to the situation at the end of the first phase was to be expected, but instead of a blunted first phase in response to the subsequent glucose stimulus, there was an enhanced and apparently accelerated secretory response (Figure 6A). A similar problem occurs when considering the following experiment which involved a prolonged depolarization by KATP channel closure prior to nutrient stimulation. Here, it was originally intended to characterize the amplifying signal produced by glucose or KIC (47, 48). To do so, the KATP channels were closed by a maximally effective concentration of the sulfonylurea glipizide. Since the exposure to a nutrient secretagogue influences the response to a subsequent stimulus (49, 50), beta cells were perifused for one hour with 2.7 �M glipizide in the absence of glucose. The response to 30 mM glucose was blunted by this pre-treatment, which might lead to the conclusion that an emptying of the pool of release-ready granules had occurred. However, such islets were still able to respond to 10 mM KIC, which is similarly effective as 30 mM glucose as a nutrient secretagogue (Figure 7).

So did glipizide really increase insulin secretion in the absence of glucose? A lack of effect is not just a theoretical possibility, given the dependence of the sulfonylurea efficacy on the glucose concentration (51, 52). A moderately increased rate of secretion could be documented (Figure 7) which concurs with the demonstration of an increase of (Ca2+)i by glipizide under this condition (48). In terms of the current theory, an increased rate of secretion by KATP channel block without increased energy metabolism should mechanistically correspond to the first phase secretion. However, not only the first phase of glucose-induced secretion was abolished by this pretreatment, but the second phase response was suppressed as well. Vice versa, the preserved secretory response to KIC was not only visible during the first 10 min but during the whole exposure time. Our conclusions from these data are that i) the metabolic amplification may be different for glucose and KIC and ii) the metabolic amplification affects both phases of secretion (assuming that the initial response to KIC is equivalent to the first phase of glucose-stimulated secretion).

8. SIGNALLING ROLE OF THE ACTION POTENTIAL FOR INSULIN SECRETION

Both, pre-exposure to depolarizing K+ concentrations and to KATP channel blockers yield secretory responses which cannot be easily explained within the framework of current theory. There are two ready objections against these experimental protocols. One is that the long-term exposure to a sulfonylurea without a nutrient depletes the energy reserves of the beta cells. The other is that 15 mM K+ cannot be the equivalent of a physiologically relevant depolarization-strength since it does not produce action potentials. Concerning the first argument, measurements of islet contents of ATP and ADP have in fact shown a reduction, but it was of moderate extent and clearly less than the reduction produced by the uncoupler, CCCP (53). Also, glucose and KIC were still able to induce an increase of the NAD(P)H autofluorescence with unchanged characteristics (53). The second objection infers a specific signalling role for the beta cell action potentials.

An unexpected observation on the effect of TEA was the reason for us to re-examine the role of action potentials. TEA is known to block voltage-dependent K+ channels and, at higher concentrations, a number of other K+ channels including KATP channels (54, 55). In doing so, it retards the repolarization of the action potential, thereby increases Ca2+ influx and, consequently, insulin secretion (56-58). However, the presence of glucose seems to be necessary for the TEA enhancement of secretion, since TEA, when added to glipizide in the absence of glucose actually decreased insulin secretion (53). By measuring the membrane potential in the perforated patch mode, it became clear that 10 mM TEA significantly increased action potential amplitude and (less clearly visible) action potential duration under this condition (Figure 8A). Of note, both types of stimulated secretion were reduced by more than 50%: the modest secretion by glipizide in the absence of glucose, which in current terminology stands for triggering and the strong secretory response to added KIC, which stands for amplification (53).

The effect on whole cell membrane current was tested using hyperpolarizing and depolarizing steps of 10 mV at a holding potential of -50 mV. 2.7 �M glipizide led to an 83 % reduction of inward and outward currents. In this situation, the addition of 10 mM TEA produced a strong inward current. In intermittent current-clamp registrations this appeared as an increase of action potential amplitude. Both, the inward current and the action potential amplitude could be abolished by 10 �M nifedipine, suggesting that they result from L-type Ca2+ channel acrivity (53). On closer inspection (Figure 9) it can be recognized that the Ca2+ currents are triggered by the depolarizing step. The inward current developed after a lag time of about 50 ms and ceased completely within 150 ms, leaving time for an outward current to develop before the holding potential was switched back to -50 mV. This outward current had a slightly larger magnitude than the inward current which developed in response to the hyperpolarizing step (Figure 9) and probably represents the residual KATP channel current. From these currents the input conductance can be estimated to be between 0.3 and 0.4 nS, which is in reasonable agreement with earlier measurements on single beta cells (59). The inward current triggered by the 10 mV depolarization had an amplitude of about 40 pA, which is close to the 37 pA Ca2+ current produced by a depolarizing step from-70 mV to -30 mV (41). Of note, in this range of depolarization there is no significant difference between the Ca2+ currents of single beta cells and those measured in beta cells within islets (41).

The conclusion from the membrane potential measurements is that the paradox inhibitory effect of TEA on insulin secretion is not caused by a change in its ion channel-blocking property. This view is supported by the induction of an oscillatory (Ca2+)i pattern when TEA was added to glipizide in the absence as well as in the presence of glucose (Figure 8B). Even the further addition of the opener of L-type Ca2+ channels, Bay K8644, which led to a drastic prolongation of the action potentials (Figure 8A) and to a continuous elevation of (Ca2+)i, did not revert the inhibition of secretion by TEA (Figure 8B). It may well be that this uncoupling of secretion from Ca2+ influx is due to a direct interaction between Kv2.1 channels and components of the exocytotic machinery, e.g. SNAP-25 (60-62).

From these data no stringent relation between the parameters of action potential spiking and insulin secretion can be recognized. The amplitude is determined by the relation between Ca2+ influx and K+ efflux, and thus cannot be taken as a measure of Ca2+ influx. As exemplified by the K+ depolarization, there may be a strong induction of secretion without any obvious action potentials, i.e. with a zero amplitude. The duration and the frequency are linked to changes in the (Ca2+)i but in a seemingly counterintuitive way: the prolonged and less frequent action potentials caused by TEA plus Bay K8644 led to the transformation of an oscillatory (Ca2+)i pattern into a plateau-like elevation, suggesting a relevant modulatory role of intracellular Ca2+ transport. Since even this increase did not raise the secretion rate, one may speculate that changes in the (Ca2+)i may accompany changes in secretion but need not be causally related. Or to be more positive, they may be causally related to secretion in beta cells only under a limited set of conditions requiring the presence of a nutrient secretagogue.

9. DEPOLARIZATION BY CURRENT INJECTION: HOW TO RECONCILE MEMBRANE CAPACITANCE MEASUREMENTS WITH SECRETION MEASUREMENTS.

The above described experiments suggest that corrections may be due in our view on the mechanisms underlying the first phase of insulin secretion. On the other hand, the currently prevailing hypothesis is based on a large number of biophysical data, which by themselves are unlikely to be incorrect. To find a way to reconcile the seemingly opposite points of view some of these data are considered in more detail.

The preferred technique to measure insulin granule exocytosis with a single granule resolution is the membrane capacitance measurement (Neher-Lindau technique, 63). Data obtained with this technique suggest that the first phase results from a limited pool of secretion-ready granules: as a functional, not morphological definition the term "readily releasable pool" was coined, analogous to what has been described in the chromaffin cell (64). The spontaneous deceleration of the high initial rate of exocytosis (21, 65, 66) was interpreted to represent the depletion of the pool of secretion-ready granules and to correspond to the end of the first phase. It is assumed that only one signal is required to elicit the exocytosis of these granules, namely an increase in the Ca2+ concentration in the immediate vicinity of the granules, sufficient to be sensed by synaptotagmin. Synaptotagmin is the established Ca2+ sensor protein among the SNARE-proteins of exocytosis, the subtypes synaptotagmin 7 and 9 have been described to be relevant for insulin secretion (67).

Based on the rather low number of Ca2+ channels and a high initial rate of exocytosis, a privileged communication between the inner orifice of a Ca2+ channel and the surrounding granules has been hypothesized (68, 69). The existence of such microcompartments of Ca2+ channels and insulin granules (70) should lead to a virtual "all or nothing" response characteristic, since the requirement for Ca2+ influx is minimized. However, the cellular response could still be graded as a result of the number of calcium channels that open, which is a testable hypothesis. The possible existence of these microcompartments is a relevant argument against data showing a dissociation between changes in (Ca2+)i as conventionally measured by the Fura-technique and changes in secretion (e.g. 30, 31,71-73). However, if stimuli are selected which increase the Fura ratio by Ca2+ influx through voltage-dependent Ca2+ channels the hypothetical microcompartments ought to be affected. In this regard it is important that the Ca2+ increase critical for the initiation of secretion is produced by a Ca2+ influx via L-type channels both in rodent and in human beta cells (36, 41). Interestingly, synaptotagmin has been proposed to arrange secretory granules in the vicinity of L-type Ca2+ channels (74). So, openers and blockers of L-type Ca2+ channels can be regarded as sufficiently selective tools to affect the critical Ca2+ compartment for the initiation of insulin secretion, whatever its precise nature (the characteristics of submembrane Ca2+ increase have been reported to differ depending on the strength of depolarization, see ref 75, however, other investigators could not identify specific submembrane microcompartments in beta cells, see ref 76).

From a methodological point of view two sources of uncertainty can be identified which may lead to misinterpretations. First, the exocytotic response of a beta cell is routinely evoked by a train of square wave-depolarizations (e.g. 65, 66), which differ from the spike and wave-pattern of the beta cell. Second, by analogy with neuroendocrine cells it is assumed that the granule content is completely discharged with each vesicle fusion, so the insulin release is presumed to be proportional to the capacitance increase, at least as long as endocytosis remains slow. While this may apply to chromaffin cells where a good correlation between the capacitance data and exocytosis as measured by TIRF-microscopy was found (77), the relation between granule fusion and insulin release is likely to be more complex (78).

Concerning the train of depolarization the question is whether the parameters usually employed are adequate for beta cells. The depolarization amplitude is derived from the resting value of the membrane potential which, as is known since the days of microelectrode measurements, is around -70 mV and the peak of the action potentials which is somewhere between -20 and 0 mV (6, 38, 79). However, the standard depolarizing pulse from -70 to 0 mV with a duration of several hundred milliseconds is clearly no good fit to an endogenously produced depolarization pattern (Figure 10).

The consequences of using less depolarizing pulses (from -70 to -20 mV instead of 0 mV) or shorter pulses (50 or 100 ms instead of 500 ms) have been characterized in detail (80). In each case the response was reduced to about 20%, so in combination, the capacitance increase is most likely less than 5 % of that produced by the standard pulse protocol (Figure 10). Such a modest response size comes close to our above described measurements of secretion when a moderately depolarizing stimulus (15 mM K+ or KATP channel closure by 50 �M tolbutamide) was employed in the presence of a basal glucose concentration. To conclude, there may be no intrinsic contradiction between our present observations and the biophysical data on beta cell exocytosis, if reference is made to the same strength of stimulation.

10. SUMMARY AND HYPOTHESES

A re-examination of the concentration dependency of K+ depolarization as a means to stimulate insulin secretion has led us to observe that 15 mM K+ comes close to the depolarizing strength (20 mV) of a KATP channel closure, which produces the slow wave depolarization in response to glucose. 40 mM K+ depolarizes by ca. 42 mV and comes close to the peak value of the action potential spikes which appear superimposed on plateau depolarisation established by KATP channel closure, so it has a depolarizing strength which corresponds to that of waves and spikes combined.

When 40 mM K+ was used to stimulate insulin secretion in the presence of a basal glucose concentration, a very strong secretory response occurred which peaked at 4 min and receded thereafter. This is often regarded as the equivalent of the first phase of glucose induced insulin secretion. When 15 mM K+ was used to stimulate insulin secretion in the presence of a basal glucose concentration there was only a modest transient response which corresponded to less than 10 % of the one produced by 40 mM K+. However, a subsequent moderate glucose stimulus (10 mM) in the continuing presence of 15 mM K+ evoked an enhanced biphasic response with marked first phase peak. This observation is not easily explained by the hypothesis that first phase is due to the release of secretion-ready granules by a depolarization-triggered Ca2+ influx (21, 23).

The hypothesis that the first phase results from a limited pool of granules adjacent to L-type channels combines earlier hypotheses by Grodsky (81, 82) and the generally accepted sequence of exocytosis in neuronal and neuroendocrine cells, where a pool of immediately releasable granules was shown to exist in the vicinity of Ca2+ channels (33, 83, see however 84). Since the release of insulin proceeds at much slower pace, it may well be that, in spite of the same basic exocytotic machinery, additional mechanisms may be involved in shaping the first phase of insulin secretion. As an alternative to Grodsky�s early model Cerasi and colleagues have suggested that the biphasic kinetics is the net result of stimulatory and inhibitory effects emanating from the glucose metabolism with different latencies (85). The main problem with this model and its later versions (50) is that the nature of the inhibitory signals has remained unclear.

The main problem in identifying the contribution of depolarization to the secretory response may lie in the definition of what constitutes the depolarization caused by glucose. The depolarisation is a composite signal not only with respect to its generation (which is well known), but in our view also with respect to its function. The slow wave is the net expression of KATP channel closure in the presence of a hitherto ill-defined inward leak current (possibly by TRPM4, see 86, 87). The spike depolarization is the net expression of Ca2+ influx via voltage-dependent Ca2+ channels and of K+ efflux via voltage- and Ca2+-dependent K+ channels, provided the majority of KATP channels is already closed. In the presence of open KATP channels no action potential can be generated, but nevertheless there is a Ca2+ influx (31).

The plateau depolarization (or the glucose-induced slow wave) has an obvious trigger function for Ca2+ influx, whereas the more variable action potential depolarization may not have a trigger function in itself (except for Kv channel opening). A prolonged action potential more than an increased amplitude appeared to be related to the increase of (Ca2+)i. In consequence, 40 mM K+ or a 70 mV depolarization by repeated current injection for 500 ms may thus elicit effects not normally occurring in nutrient-induced insulin secretion. To mimic a physiological strength of depolarization in beta cells a depolarization from -70 mV to about -40 or -30 mV appears more appropriate. Interestingly, the strength of the Ca2+ current thus produced (ca 40 pA, ref 41) is quite similar to the one evoked by depolarizing 10 mV steps starting from a plateau of -50 mV (Figure 9), .

The beta cell action potential, in contrast to the neuronal action potential, may not represent a unit of information in its own right. In neurons, the uniform action potential serves as a fast messenger to cover the comparatively large distances between the sites of signal generation and those of signal transmission to the neighbouring cells. The action potential frequency, not its shape, codes for an exactly matching transmitter release and the subsynaptic neuron recodes the information provided by a number of transmitters into another action potential frequency (88). Obviously, this is not the role of electrical activity in an islet of 150 �m diameter containing a network of electrically coupled beta cells (79, 89). However, there may be other roles for the ion channels involved in shaping the action potential. It has been suggested by others that Kv 2.1 and also Cav 1.2 are coupled to exocytosis in beta cells in a way not requiring their ion conducting function (60-62, 90). Some of the above observations of ours point in the same direction (53). This raises the possibility that the respective ion conductances may not be as directly linked to the regulation of exocytosis as currently assumed.

Finite or exhaustible pools of secretory granules (readily releasable or immediately releasable) may well exist in the submembrane space of the beta cells and may be released during the first phase of glucose-induced insulin secretion, but our conclusion from the above experiments is that their release does not constitute the correlate of this unique feature of hormone secretion. TIRF-microscopy has provided data showing that so-called newcomer insulin granules fuse with the plasma membrane, apparently without establishing a definite docked state (91, 92). This would be in keeping with the current hypothesis if such granules would constitute a highly Ca2+ -sensitive pool due to the high Ca2+ sensitivity of synaptotagmin 7 as compared to the more conventional low Ca2+ sensitivity of synaptotagmin 9 (93). However, synaptotagmin 7 null mice showed reductions in both phases of insulin secretion, not only in the second phase, as was originally expected (94). In line with our above observation (Figure 7) this suggests that the release mechanisms may not be as fundamentally different as currently assumed. In fact, it has been observed earlier that the beta cell energy metabolism affects secretion by mechanisms acting "distal" to Ca2+ influx (95, 96), which concurs with the view the first phase is influenced by amplifying signals circumventing plasma membrane depolarization (10).

11. ACKNOWLEDGMENTS

This work was supported by grants from the Deutsche Forschungsgemeinschaft (Ru 368/5-1), the Deutsche Diabetes Gesellschaft and the Deutsche Diabetes Stiftung. Skilful technical assistance by Angela Hahlbohm, Verena Lier-Glaubitz and Sabine Warmbold is gratefully acknowledged. Thanks are due to Rafael Nesher for stimulating discussions. The following figures were reproduced with kind permission: Figure 2, ASPET (ref 30); Figures 3, 4 and 6, Am Physiol Society (ref 31); Figure 8, Elsevier (ref 53).

12. REFERENCES

1. Ashcroft SJ, Hedeskov CJ, Randle PJ. Glucose metabolism in mouse pancreatic islets. Biochem J 118:143-54(1970)
PMid:4919469    PMCid:1179090

2. Panten U, Kriegstein E, Poser W, Schönborn J, Hasselblatt A. Effects of L-leucine and alpha-ketoisocaproic acid upon insulin secretion and metabolism of isolated pancreatic islets. FEBS Lett 20, 225-8 (1972)
doi:10.1016/0014-5793(72)80801-9

3. Malaisse WJ, Sener A, Herchuelz A, Hutton JC. Insulin release: the fuel hypothesis. Metabolism 28, 373-86 (1979)
doi:10.1016/0026-0495(79)90111-2

4. Matschinsky FM. Banting Lecture 1995. A lesson in metabolic regulation inspired by the glucokinase glucose sensor paradigm. Diabetes 45, 223-41 (1996)
doi:10.2337/diabetes.45.2.223
PMid:8549869

5. Dean PM, Matthews EK. Glucose-induced electrical activity in pancreatic islet cells. J Physiol 210: 255-64 (1970)
PMid:5501259    PMCid:1395578

6. Henquin JC, Meissner HP. Significance of ionic fluxes and changes in membrane potential for stimulus-secretion coupling in pancreatic B-cells. Experientia 40, 1043-52 (1984)
doi:10.1007/BF01971450

7. Henquin JC. D-glucose inhibits potassium efflux from pancreatic islet cells. Nature 271, 271-3 (1978)
doi:10.1038/271271a0
PMid:340960

8. Cook DL, Hales CN. Intracellular ATP directly blocks K+ channels in pancreatic B-cells. Nature 311, 271-3 (1984)
doi:10.1038/311271a0
PMid:6090930

9. Ashcroft FM, Ashcroft SJ, Harrison DE. Properties of single potassium channels modulated by glucose in rat pancreatic beta-cells. J Physiol 400, 501-27 (1988)
PMid:2458459    PMCid:1191821

10. Henquin JC, Nenquin M, Stiernet P, Ahren B: In vivo and in vitro glucose-induced biphasic insulin secretion in the mouse: pattern and role of cytoplasmic Ca2+ and amplification signals in beta-cells. Diabetes 55, 441-51 (2006)
doi:10.2337/diabetes.55.02.06.db05-1051
PMid:16443779

11. Del Prato S: Loss of early insulin secretion leads to postprandial hyperglycaemia. Diabetologia 46 Suppl 1, M2-8 (2003)

12. Cherrington AD, Sindelar D, Edgerton D, Steiner K, McGuinness OP: Physiological consequences of phasic insulin release in the normal animal. Diabetes 51 Suppl 1, S103-8 (2002)
doi:10.2337/diabetes.51.2007.S103

13. DiCostanzo CA, Moore MC, Lautz M, Scott M, Farmer B, Everett CA, Still JG, Higgins A, Cherrington AD: Simulated first-phase insulin release using Humulin or insulin analog HIM2 is associated with prolonged improvement in postprandial glycemia. Am J Physiol Endocrinol Metab 289, E46-52 (2005)
doi:10.1152/ajpendo.00583.2004
PMid:15713685

14. Pratley RE, Weyer C. The role of impaired early insulin secretion in the pathogenesis of Type II diabetes mellitus. Diabetologia 44, 929-45 (2001)
doi:10.1007/s001250100580
PMid:11484070

15. Gerich JE: Is reduced first-phase insulin release the earliest detectable abnormality in individuals destined to develop type 2 diabetes? Diabetes 51 Suppl 1, S117-21 (2002)
doi:10.2337/diabetes.51.2007.S117

16. Panten U, Schwanstecher M, Wallasch A, Lenzen S. Glucose both inhibits and stimulates insulin secretion from isolated pancreatic islets exposed to maximally effective concentrations of sulfonylureas. Naunyn Schmiedebergs Arch Pharmacol 338, 459-62 (1989)

17. Gembal M, Gilon P, Henquin JC. Evidence that glucose can control insulin release independently from its action on ATP-sensitive K+ channels in mouse B cells. J Clin Invest 89, 1288-95 (1992)
doi:10.1172/JCI115714
PMid:1556189    PMCid:442990

18. Aizawa T, Sato Y, Komatsu M, Hashizume K. ATP-sensitive K+ channel-independent, insulinotropic action of glucose in the B-cells. Endocr Regul 26, 159-62 (1992)
PMid:1307719

19. Straub SG, James RF, Dunne MJ, Sharp GW. Glucose activates both K(ATP) channel-dependent and K(ATP) channel-independent signaling pathways in human islets. Diabetes 47, 758-63 (1998)
doi:10.2337/diabetes.47.5.758
PMid:9588447

20. Henquin JC. Triggering and amplifying pathways of regulation of insulin secretion by glucose. Diabetes 49, 1751-60 (2000)
doi:10.2337/diabetes.49.11.1751
PMid:11078440

21. Barg S, Eliasson L, Renström E, Rorsman P. A subset of 50 secretory granules in close contact with L-type Ca2+ channels accounts for first-phase insulin secretion in mouse beta-cells. Diabetes 51 Suppl 1, S74-82 (2002)
doi:10.2337/diabetes.51.2007.S74

22. Kennedy ED, Maechler P, Wollheim CB. Effects of depletion of mitochondrial DNA in metabolism secretion coupling in INS-1 cells. Diabetes 47, 374-80 (1998)
doi:10.2337/diabetes.47.3.374
PMid:9519742

23. Rorsman P, Renström E. Insulin granule dynamics in pancreatic beta cells. Diabetologia 46, 1029-45 (2003)
doi:10.1007/s00125-003-1153-1
PMid:12879249

24. Straub SG, Sharp GW. Glucose-stimulated signaling pathways in biphasic insulin secretion. Diabetes Metab Res Rev 18, 451-63 (2002)
doi:10.1002/dmrr.329
PMid:12469359

25. Morgan NG. Imidazoline receptors: new targets for antihyperglycaemic drugs. Exp Opin Invest Drugs 8, 575-84 (1999)
doi:10.1517/13543784.8.5.575
PMid:15992117

26. Proks P, Ashcroft FM. Phentolamine block of KATP channels is mediated by Kir6.2. Proc Natl Acad Sci U S A 94, 11716-20 (1997)
doi:10.1073/pnas.94.21.11716

27. Grosse-Lackmann T, Zunkler BJ, Rustenbeck I. Specificity of nonadrenergic imidazoline binding sites in insulin-secreting cells and relation to the block of ATP-sensitive K(+) channels. Ann N Y Acad Sci 1009, 371-7 (2003)
doi:10.1196/annals.1304.050
PMid:15028615

28. Bleck C, Wienbergen A, and Rustenbeck I. Glucose-dependence of imidazoline-induced insulin secretion. Diabetes 53, S135-9 (2004)
doi:10.2337/diabetes.53.suppl_3.S135
PMid:15561901

29. Bleck C, Wienbergen A, and Rustenbeck I Essential role of the imidazoline moiety for the insulinotropic effect but not the KATP channel-blocking effect of imidazolines. Diabetologia 48, 2567-75 (2005)
doi:10.1007/s00125-005-0031-4
PMid:16283242

30. Hatlapatka K, Wienbergen A, Kühne C, Jörns A, Willenborg M, Rustenbeck I. Selective enhancement of nutrient-induced insulin secretion by ATP-sensitive K+ channel-blocking imidazolines. J Pharmacol Exp Ther 331, 1033-41 (2009)
doi:10.1124/jpet.109.152751
PMid:19745109

31. Hatlapatka K, Willenborg M, Rustenbeck I. Plasma membrane depolarization as a determinant of the first phase of insulin secretion. Am J Physiol Endocrinol Metab 297, E315-22 (2009)
doi:10.1152/ajpendo.90981.2008
PMid:19470830

32. Zhang CL, Katoh M, Shibasaki T, Minami K, Sunaga Y, Takahashi H, Yokoi N, Iwasaki M, Miki T, Seino S. The cAMP sensor Epac2 is a direct target of antidiabetic sulfonylurea drugs. Science 325, 607-10 (2009)
doi:10.1126/science.1172256
PMid:19644119

33. Gerber SH, Südhof TC. Molecular determinants of regulated exocytosis. Diabetes 51 Suppl 1, S3-11 (2002)
doi:10.2337/diabetes.51.2007.S3

34. Mears D. Regulation of insulin secretion in islets of Langerhans by Ca(2+) channels. J Membr Biol 200, 57-66 (2004)
doi:10.1007/s00232-004-0692-9
PMid:15520904

35. Schulla V, Renström E, Feil R, Feil S, Franklin I, Gjinovci A, Jing XJ, Laux D, Lundquist I, Magnuson MA, Obermüller S, Olofsson CS, Salehi A, Wendt A, Klugbauer N, Wollheim CB, Rorsman P, Hofmann F. Impaired insulin secretion and glucose tolerance in beta cell-selective Ca(v)1.2 Ca2+ channel null mice. EMBO J 22, 3844-54 (2003)
doi:10.1093/emboj/cdg389
PMid:12881419    PMCid:169062

36. Braun M, Ramracheya R, Bengtsson M, Zhang Q, Karanauskaite J, Partridge C, Johnson PR, Rorsman P. Voltage-gated ion channels in human pancreatic beta-cells: electrophysiological characterization and role in insulin secretion. Diabetes 57, 1618-28 (2008)
doi:10.2337/db07-0991
PMid:18390794

37. Arredouani A, Henquin JC, Gilon P. Contribution of the endoplasmic reticulum to the glucose-induced (Ca(2+))(c) response in mouse pancreatic islets. Am J Physiol Endocrinol Metab 282, E982-91 (2002)

38. Lebrun P, Atwater I. Effects of the calcium channel agonist, BAY K 8644, on electrical activity in mouse pancreatic B-cells. Biophys J 48, 919-30 (1985)
doi:10.1016/S0006-3495(85)83855-8

39. Rosário LM, Barbosa RM, Antunes CM, Silva AM, Abrunhosa AJ, Santos RM. Bursting electrical activity in pancreatic beta-cells: evidence that the channel underlying the burst is sensitive to Ca2+ influx through L-type Ca2+ channels. Pflugers Arch 424, 439-47 (1993)
doi:10.1007/BF00374906

40. Herchuelz A, Pochet R, Pastiels C, Van Praet A. Heterogeneous changes in (Ca2+)i induced by glucose, tolbutamide and K+ in single rat pancreatic B cells. Cell Calcium 12: 577-86 (1991)
doi:10.1016/0143-4160(91)90076-Q

41. Göpel S, Kanno T, Barg S, Galvanovskis J, Rorsman P. Voltage-gated and resting membrane currents recorded from B-cells in intact mouse pancreatic islets. J Physiol 521, 717-28 (1999)
doi:10.1111/j.1469-7793.1999.00717.x
PMid:10601501    PMCid:2269694

42. MacDonald PE, Wheeler MB. Voltage-dependent K+ channels in pancreatic beta cells: role, regulation and potential as therapeutic targets. Diabetologia 46, 1046-62 (2003)
doi:10.1007/s00125-003-1159-8
PMid:12830383

43. Jacobson DA, Philipson LH. Action potentials and insulin secretion: new insights into the role of Kv channels. Diabetes Obes Metab 9 Suppl 2, 89-98 (2007)
doi:10.1111/j.1463-1326.2007.00784.x
PMid:17919183

44. Cerasi E, Luft R. The plasma insulin response to glucose infusion in healthy subjects and in diabetes mellitus. Acta Endocrinol (Copenh) 55, 278-304 (1967)

45. Curry DL, Bennett LL, Grodsky GM. Dynamics of insulin secretion by the perfused rat pancreas. Endocrinology 83, 572-84 (1968)
doi:10.1210/endo-83-3-572
PMid:4877098

46. Oheim M, Loerke D, Chow RH, Stühmer W. Evanescent-wave microscopy: a new tool to gain insight into the control of transmitter release. Philos Trans R Soc Lond B Biol Sci 354, 307-18 (1999)
doi:10.1098/rstb.1999.0382
PMid:10212479    PMCid:1692489

47. Urban KA, Panten U. Selective loss of glucose-induced amplification of insulin secretion in mouse pancreatic islets pretreated with sulfonylurea in the absence of fuels. Diabetologia 48, 2563-6 (2005)
doi:10.1007/s00125-005-0030-5
PMid:16283243

48. Panten U, Rustenbeck I. Fuel-induced amplification of insulin secretion in mouse pancreatic islets exposed to a high sulfonylurea concentration: role of the NADPH/NADP+ ratio. Diabetologia 51, 101-9 (2008)
doi:10.1007/s00125-007-0849-z
PMid:17960358

49. Borg LA, Westberg M, Grill V. The priming effect of glucose on insulin release does not involve redistribution of secretory granules within the pancreatic B-cell. Mol Cell Endocrinol 56, 219-25 (1988)
doi:10.1016/0303-7207(88)90064-0

50. Nesher R, Cerasi E. Modeling phasic insulin release: immediate and time-dependent effects of glucose. Diabetes 51 Suppl 1, S53-9 (2002)
doi:10.2337/diabetes.51.2007.S53

51. Panten U, Zünkler BJ, Scheit S, Kirchhoff K, Lenzen S. Regulation of energy metabolism in pancreatic islets by glucose and tolbutamide. Diabetologia 29, 648-54 (1986)
doi:10.1007/BF00869265
PMid:3539682

52. Henquin JC. A minimum of fuel is necessary for tolbutamide to mimic the effects of glucose on electrical activity in pancreatic beta-cells. Endocrinology 139, 993-8 (1998)
doi:10.1210/en.139.3.993
PMid:9492030

53. Willenborg M, Ghaly H, Hatlapatka K, Urban K, Panten U, Rustenbeck I. The signalling role of action potential depolarization in insulin secretion. Metabolism-dependent dissociation between action potential increase and secretion increase by TEA. Biochem Pharmacol 80, 104-112 (2010)
doi:10.1016/j.bcp.2010.03.014
PMid:20303336

54. Bokvist K, Rorsman P, Smith PA. Effects of external tetraethylammonium ions and quinine on delayed rectifying K+ channels in mouse pancreatic beta-cells. J Physiol 423, 311-25 (1990)
PMid:2201760    PMCid:1189759

55. Bokvist K, Rorsman P, Smith PA. Block of ATP-regulated and Ca2+-activated K+ channels in mouse pancreatic beta-cells by external tetraethylammonium and quinine. J Physiol 423, 327-42 (1990)
PMid:2201761    PMCid:1189760

56. Henquin JC. Tetraethylammonium potentiation of insulin release and inhibition of rubidium efflux in pancreatic islets. Biochem Biophys Res Commun 77, 551-6 (1977)
doi:10.1016/S0006-291X(77)80014-4

57. Herrington J, Zhou YP, Bugianesi RM, Dulski PM, Feng Y, Warren VA, Smith MM, Kohler MG, Garsky VM, Sanchez M, Wagner M, Raphaelli K, Banerjee P, Ahaghotu C, Wunderler D, Priest BT, Mehl JT, Garcia ML, McManus OB, Kaczorowski GJ, Slaughter RS. Blockers of the delayed-rectifier potassium current in pancreatic beta-cells enhance glucose-dependent insulin secretion. Diabetes 55, 1034-42 (2006)
doi:10.2337/diabetes.55.04.06.db05-0788
PMid:16567526

58. Tamarina NA, Kuznetsov A, Fridlyand LE, and Philipson LH. Delayed-rectifier (KV2.1) regulation of pancreatic beta-cell calcium responses to glucose: inhibitor specificity and modeling. Am J Physiol Endocrinol Metab 289, E578-85 (2005)
doi:10.1152/ajpendo.00054.2005
PMid:16014354

59. Smith PA, Ashcroft FM, Rorsman P. Simultaneous recordings of glucose-dependent electrical activity and ATP-regulated K+-currents in isolated mouse pancreatic β-cells. FEBS Lett 261: 187-90 (1990)
doi:10.1016/0014-5793(90)80667-8

60. Feinshreiber L, Singer-Lahat D, Ashery U, Lotan I. Voltage-gated potassium channel as a facilitator of exocytosis. Ann N Y Acad Sci 1152: 87-92 (2009)
doi:10.1111/j.1749-6632.2008.03997.x
PMid:19161379

61. Leung YM, Kwan EP, Ng B, Kang Y, Gaisano HY. SNAREing voltage-gated K+ and ATP-sensitive K+ channels: tuning beta-cell excitability with syntaxin-1A and other exocytotic proteins. Endocr Rev 28, 653-63 (2007)
doi:10.1210/er.2007-0010
PMid:17878408

62. He Y, Kang Y, Leung YM, Xia F, Gao X, Xie H, Gaisano HY, Tsushima RG. Modulation of Kv2.1 channel gating and TEA sensitivity by distinct domains of SNAP-25. Biochem J 396, 363-9 (2006)
doi:10.1042/BJ20051478
PMid:16478442    PMCid:1462716

63. Lindau M, Neher E. Patch-clamp techniques for time-resolved capacitance measurements in single cells. Pflugers Arch 411, 137-46 (1988)
doi:10.1007/BF00582306

64. Voets T, Neher E, Moser T. Mechanisms underlying phasic and sustained secretion in chromaffin cells from mouse adrenal slices. Neuron 23, 607-15 (1999)
doi:10.1016/S0896-6273(00)80812-0

65. Gillis KD, Misler S. Single cell assay of exocytosis from pancreatic islet B cells. Pflugers Arch 420, 121-3 (1992)
doi:10.1007/BF00378654

66. Olofsson CS, Göpel SO, Barg S, Galvanovskis J, Ma X, Salehi A, Rorsman P, Eliasson L. Fast insulin secretion reflects exocytosis of docked granules in mouse pancreatic B-cells. Pflugers Arch 444, 43-51 (2002)
doi:10.1007/s00424-002-0781-5
PMid:11976915

67. Gauthier BR, Duhamel DL, Iezzi M, Theander S, Saltel F, Fukuda M, Wehrle-Haller B, Wollheim CB. Synaptotagmin VII splice variants alpha, beta, and delta are expressed in pancreatic beta-cells and regulate insulin exocytosis. FASEB J 22, 194-206 (2008)
doi:10.1096/fj.07-8333com
PMid:17709608

68. Bokvist K, Eliasson L, Ammälä C, Renström E, Rorsman P. Co-localization of L-type Ca2+ channels and insulin-containing secretory granules and its significance for the initiation of exocytosis in mouse pancreatic B-cells. EMBO J 14, 50-7 (1995)
PMid:7828595    PMCid:398051

69. Barg S, Ma X, Eliasson L, Galvanovskis J, Göpel SO, Obermüller S, Platzer J, Renström E, Trus M, Atlas D, Striessnig J, Rorsman P. Fast exocytosis with few Ca(2+) channels in insulin-secreting mouse pancreatic B cells. Biophys J 81, 3308-23 (2001)
doi:10.1016/S0006-3495(01)75964-4

70. Rutter GA, Tsuboi T, Ravier MA. Ca2+ microdomains and the control of insulin secretion. Cell Calcium 40, 539-51 (2006)
doi:10.1016/j.ceca.2006.08.015
PMid:17030367

71. Martin F, Reig JA, Soria B. Secretagogue-induced (Ca2+)i changes in single rat pancreatic islets and correlation with simultaneously measured insulin release. J Mol Endocrinol 15, 177-85 (1995)
doi:10.1677/jme.0.0150177
PMid:8800642

72. Gembal M, Detimary P, Gilon P, Gao ZY, Henquin JC. Mechanisms by which glucose can control insulin release independently from its action on adenosine triphosphate-sensitive K+ channels in mouse B cells. J Clin Invest 91, 871-80 (1993)
doi:10.1172/JCI116308
PMid:8383702    PMCid:288039

73. Yamada S, Komatsu M, Aizawa T, Sato Y, Yajima H, Yada T, Hashiguchi S, Yamauchi K, Hashizume K. Time-dependent potentiation of the beta-cell is a Ca2+-independent phenomenon. J Endocrinol 172, 345-54 (2002)
doi:10.1677/joe.0.1720345
PMid:11834452

74. Young SM Jr, Neher E. Synaptotagmin has an essential function in synaptic vesicle positioning for synchronous release in addition to its role as a calcium sensor. Neuron 63, 482-96 (2009)
doi:10.1016/j.neuron.2009.07.028
PMid:19709630

75. Ohara-Imaizumi M, Aoyagi K, Nakamichi Y, Nishiwaki C, Sakurai T, Nagamatsu S. Pattern of rise in subplasma membrane Ca2+ concentration determines type of fusing insulin granules in pancreatic beta cells. Biochem Biophys Res Commun 385, 291-5. (2009)
doi:10.1016/j.bbrc.2009.04.155
PMid:19426714

76. Ravier MA, Cheng-Xue R, Palmer AE, Henquin JC, Gilon P. Subplasmalemmal Ca(2+) measurements in mouse pancreatic beta cells support the existence of an amplifying effect of glucose on insulin secretion. Diabetologia 2010 May 13. (Epub ahead of print)

77. Becherer U, Pasche M, Nofal S, Hof D, Matti U, Rettig J. Quantifying exocytosis by combination of membrane capacitance measurements and total internal reflection fluorescence microscopy in chromaffin cells. PLoS One 2, e505 (2007)
doi:10.1371/journal.pone.0000505
PMid:17551585    PMCid:1876815

78. Michael DJ, Cai H, Xiong W, Ouyang J, Chow RH. Mechanisms of peptide hormone secretion. Trends Endocrinol Metab 17, 408-15 (2006)
doi:10.1016/j.tem.2006.10.011
PMid:15681023

79. Valdeolmillos M, Gomis A, Sánchez-Andrés JV. In vivo synchronous membrane potential oscillations in mouse pancreatic beta-cells: lack of co-ordination between islets. J Physiol 493, 9-18 (1996)
PMid:8735691    PMCid:1158947

80. Göpel S, Zhang Q, Eliasson L, Ma XS, Galvanovskis J, Kanno T, Salehi A, Rorsman P. Capacitance measurements of exocytosis in mouse pancreatic alpha-, beta- and delta-cells within intact islets of Langerhans. J Physiol 556, 711-26 (2004)
doi:10.1113/jphysiol.2003.059675
PMid:14966302    PMCid:1664984

81. Grodsky G, Landahl H, Curry D, Bennett L. A two-compartmental model for insulin secretion. Adv Metab Disord 1 Suppl 1, 45-50 (1970)
PMid:4911192

82. O'Connor MD, Landahl H, Grodsky GM. Comparison of storage- and signal-limited models of pancreatic insulin secretion. Am J Physiol 238, R378-89 (1980)

83. Schneggenburger R, Neher E. Presynaptic calcium and control of vesicle fusion. Curr Opin Neurobiol 15, 266-74 (2005)
doi:10.1016/j.conb.2005.05.006
PMid:15919191

84. Wu MM, Llobet A, Lagnado L. Loose coupling between calcium channels and sites of exocytosis in chromaffin cells. J Physiol 587: 5377-91 (2009)
doi:10.1113/jphysiol.2009.176065
PMid:19752110

85. Cerasi E, Fick G, Rudemo M. A mathematical model for the glucose-induced insulin release in man. Eur J Clin Invest 4, 267-78 (1974)
doi:10.1111/j.1365-2362.1974.tb00403.x
PMid:4424575

86. Cheng H, Beck A, Launay P, Gross SA, Stokes AJ, Kinet JP, Fleig A, Penner R. TRPM4 controls insulin secretion in pancreatic beta-cells. Cell Calcium 41, 51-61 (2007)
doi:10.1016/j.ceca.2006.04.032
PMid:16806463

87. Marigo V, Courville K, Hsu WH, Feng JM, Cheng H. TRPM4 impacts on Ca2+ signals during agonist-induced insulin secretion in pancreatic beta-cells. Mol Cell Endocrinol 299, 194-203 (2009)
doi:10.1016/j.mce.2008.11.011
PMid:19063936

88. Chacron MJ, Longtin A, Maler L. The effects of spontaneous activity, background noise, and the stimulus ensemble on information transfer in neurons. Network 14: 803-24 (2003)
doi:10.1088/0954-898X/14/4/010

89. Bavamian S, Klee P, Britan A, Populaire C, Caille D, Cancela J, Charollais A, Meda P. Islet-cell-to-cell communication as basis for normal insulin secretion. Diabetes Obes Metab 9 Suppl 2, 118-32 (2007)
doi:10.1111/j.1463-1326.2007.00780.x
PMid:17919186

90. Trus M, Corkey RF, Nesher R, Richard AM, Deeney JT, Corkey BE, Atlas D. The L-type voltage-gated Ca2+ channel is the Ca2+ sensor protein of stimulus-secretion coupling in pancreatic beta cells. Biochemistry 46:14461-7 (2007)
doi:10.1021/bi7016816
PMid:18027971

91. Nagamatsu S, Ohara-Imaizumi M, Nakamichi Y, Kikuta T, Nishiwaki C. Imaging docking and fusion of insulin granules induced by antidiabetes agents: sulfonylurea and glinide drugs preferentially mediate the fusion of newcomer, but not previously docked, insulin granules. Diabetes 55, 2819-25 (2006)
doi:10.2337/db06-0105
PMid:17003348

92. Ohara-Imaizumi M, Fujiwara T, Nakamichi Y, Okamura T, Akimoto Y, Kawai J, Matsushima S, Kawakami H, Watanabe T, Akagawa K, Nagamatsu S. Imaging analysis reveals mechanistic differences between first- and second-phase insulin exocytosis. J Cell Biol 177, 695-705 (2007)
doi:10.1083/jcb.200608132
PMid:17502420    PMCid:2064214

93. Pedersen MG, Sherman A. Newcomer insulin secretory granules as a highly calcium-sensitive pool. Proc Natl Acad Sci U S A 106, 7432-6 (2009)
doi:10.1073/pnas.0901202106
PMid:19372374    PMCid:2678592

94. Gustavsson N, Lao Y, Maximov A, Chuang JC, Kostromina E, Repa JJ, Li C, Radda GK, Südhof TC, Han W. Impaired insulin secretion and glucose intolerance in synaptotagmin-7 null mutant mice. Proc NatlAcadSci USA105, 3992-7 (2008)
doi:10.1073/pnas.0711700105
PMid:18308938    PMCid:2268794

95. Rustenbeck I, Herrmann C, Grimmsmann T. Energetic requirement of insulin secretion distal to calcium influx. Diabetes 46, 1305-11 (1997)
doi:10.2337/diabetes.46.8.1305
PMid:9231655

96. Takahashi N, Kadowaki T, Yazaki Y, Ellis-Davies GC, Miyashita Y, Kasai H. Post-priming actions of ATP on Ca2+-dependent exocytosis in pancreatic beta cells. Proc Natl Acad Sci U S A 96, 760-5 (1999)
doi:10.1073/pnas.96.2.760

Abbreviations: (Ca2+)i: cytosolic free calcium concentration; Kv channel: voltage-dependent K+ channel, KATP channel: ATP-sensitive K+ channel, KIC: a-ketoisocaproic acid; TEA: tetraethylammonium

Key Words: Cytosolic calcium concentration; Insulin secretion; Kv channel; KATP channel; Pancreatic islets; Plasma membrane potential, Review

Send correspondence to: Ingo Rustenbeck, Institute of Pharmacology and Toxicology, University of Braunschweig, Mendelssohnstr. 1, 38106 Braunschweig, Tel. 49-531-391-5670, Fax: 49-531-391-8287, E-mail: rustenbeck@tu-bs.de