Feedback Control of the ATP-Sensitive K؉ Current by Cytosolic Ca2؉ Contributes to Oscillations of the in Pancreatic ␤-Cells Jean-Franc¸ois Rolland, Jean-Claude Henquin, and Patrick Gilon

During glucose stimulation, pancreatic ␤-cells display ferred to as a slow wave, consists of a depolarized phase membrane potential oscillations that correspond to in- on top of which a train of action potentials appears and a -termittent Ca2؉ influx, leading to oscillations of the repolarized phase without action potentials. Glucose mod ,cytosolic free concentration ([Ca2؉] ) and insu- ulates the duration of the slow waves that become longer ؉c ؉ lin secretion. The role of ATP-sensitive K (K -ATP) with little change in their frequency, as the glucose con- channels in the control of these oscillations was inves- centration increases (between 7 and 25 mmol/l). When this tigated by measuring the K؉-ATP current (I ) with KATP concentration exceeds 25 mmol/l, slow waves are trans- the perforated mode of the patch-clamp technique. No formed into a sustained with continuous oscillations of IKATP were observed when glucose-stim- ulated ␤-cells were kept hyperpolarized, thus with low spike activity. The changes in membrane potential are ␤ 2؉ 2؉ 2؉ and stable [Ca ]c. However, increasing [Ca ]c by Ca crucial for the control of -cell function because each influx (depolarizing pulses) or Ca2؉ mobilization (ace- depolarization induces a concomitant rise in the cytosolic 2ϩ 2ϩ tylcholine) transiently augmented IKATP. This effect was free Ca concentration ([Ca ]c) (5,6), which is the signal abolished by tolbutamide, attenuated by increasing the that triggers insulin secretion. glucose concentration in the medium, and prevented by The resting membrane potential of ␤-cells is mainly abrogation of the [Ca2؉] rise, which demonstrates that c controlled by an unknown depolarizing current and a the current is really IKATP and that its increase is ϩ ϩ - Ca2؉-dependent. Injection of a current of a similar hyperpolarizing current carried by ATP-sensitive K (K ,2؉ ATP) channels (7). When the glucose concentration is low amplitude to that of the Ca -induced increase in IKATP ϩ was sufficient to repolarize glucose-stimulated ␤-cells. the ATP-to-ADP ratio is low, and many K -ATP channels 2؉ ϩ These results suggest that, in the absence of [Ca ]c are open; therefore, K -ATP current (IKATP) overwhelms oscillations, no metabolic oscillations affect IKATP in the depolarizing current and keeps the potential close to 2؉ ϩ ␤ pancreatic -cells. In contrast, [Ca ]c oscillations the equilibrium potential of K . When the glucose concen- evoke IKATP oscillations. This mechanism may consti- tration increases, cell metabolism is stimulated and the tute the feedback loop controlling the glucose-induced ATP-to-ADP ratio rises (8), leading to closure of Kϩ-ATP oscillating electrical activity in ␤-cells. Diabetes 51: 376–384, 2002 channels (9,10). The resulting decrease in IKATP permits the depolarizing current to move the membrane potential further away from the equilibrium potential of Kϩ. When the of activation of voltage-dependent 2ϩ 2ϩ ␤ Ca channels is reached, Ca influx starts (reflected by ancreatic -cells are electrically excitable. Their 2ϩ membrane potential and electrical activity are the appearance of electrical activity), [Ca ]c increases, finely controlled by glucose, the most important and insulin secretion is stimulated. Whereas it is unani- stimulus of insulin secretion. These effects have mously admitted that the rise in the ATP-to-ADP ratio P triggers the initial depolarization, the mechanisms driving mainly been characterized in mouse islets (1–4). In the absence of glucose or in the presence of a nonstimulating the oscillations of the membrane potential remain incom- concentration of glucose (Յ6 mmol/l), the membrane pletely understood. The opening of voltage-dependent 2ϩ potential is at the resting level. When the glucose concen- Ca channels undoubtedly underlies the depolarizing tration increases (Ն7 mmol/l), the plasma membrane phase, but the mechanism(s) causing the at depolarizes and an oscillating electrical activity starts (1). the end of each slow wave has escaped identification. Each oscillation of the membrane potential, usually re- Several hypotheses have been put forward. They include activation of Ca2ϩ-dependent Kϩ channels (11–14) differ- ent from the charybdotoxin-sensitive Kϩ channel (15); From the Unite´ d’Endocrinologie et Me´tabolisme, University of Louvain, slow inactivation of voltage-dependent Ca2ϩ channels Brussels, Belgium. Address correspondence and reprint requests to Dr. Patrick Gilon, Unite´ (3,16); decrease of cell-to-cell coupling (17) or of a store- d’Endocrinologie et Me´tabolisme, University of Louvain Faculty of Medicine, operated current (18,19); and increase of IKATP (20–22). UCL 55.30, Av. Hippocrate 55, B-1200 Brussels, Belgium. E-mail: gilon@endo. According to this last hypothesis, cyclic closure and ucl.ac.be. ϩ Received for publication 2 July 2001 and accepted in revised form 29 opening of K -ATP channels cause oscillations of mem- October 2001. 2ϩ 2ϩ ϩ brane potential that, in turn, repetitively open and close ACh, acetylcholine; [Ca ]c, cytosolic free Ca concentration; IKATP,K - 2ϩ ATP current; IP , Ins(1,4,5)P ;Kϩ-ATP channel, ATP-sensitive Kϩ channel; Ca channels. Theoretically, such cycles could result 3 3 2ϩ PIP2, phosphatidylinositol 4,5-bisphosphate. from intrinsic Ca -independent metabolic oscillations

376 DIABETES, VOL. 51, FEBRUARY 2002 J.-F. ROLLAND, J.-C. HENQUIN, AND P. GILON

(23,24) or be driven by Ca2ϩ in a sort of negative feedback control (20,21,25–27). In the present study, we used the perforated mode of the patch-clamp technique to monitor IKATP in single mouse ␤ -cells. We investigated whether oscillations of IKATP exist 2ϩ when [Ca ]c is either kept low and stable (reflecting intrinsic metabolic oscillations) or is repetitively increased (reflecting Ca2ϩ-dependent activation of the channel). Some of the results have been presented in abstract form (28).

RESEARCH DESIGN AND METHODS Solutions and drugs. The medium used for the preparation of islet cells was a bicarbonate-buffered solution that contained (in mmol/l): 120 NaCl, 4.8 KCl,

2.5 CaCl2, 1.2 MgCl2, 24 NaHCO3, 5 HEPES, and 10 mmol/l glucose (pH adjusted to 7.40 with NaOH). The Ca2ϩ-free medium used to disrupt the islets into single cells had the following composition (in mmol/l): 138 NaCl, 5.6 KCl,

1.2 MgCl2, 5 HEPES, and 1 mmol/l EGTA (pH adjusted to 7.40 with NaOH). All solutions used for tissue preparation were gassed with O2:CO2 (94:6%) and supplemented with 1 mg/ml BSA (fraction V; Roche Molecular Biochemicals; Mannheim, Germany).

For electrophysiological measurements of IKATP, the standard extracellular solution contained (in mmol/l): 120 NaCl, 4.8 KCl, 2.5 CaCl2, 1.2 MgCl2,24 NaHCO3, 5 HEPES (pH adjusted to 7.40 with NaOH), and various concentra- 2ϩ tions of glucose as indicated. When a Ca -free solution was needed, CaCl2 was substituted by MgCl2, and 2 mmol/l EGTA was added. These solutions were gassed with O2:CO2 (94:6%). The pipette solution contained (in mmol/l): 70 K2SO4, 10 NaCl, 10 KCl, 3.7 MgCl2, and 5 HEPES (pH adjusted to 7.1 with KOH). The electrical contact was established by adding a pore-forming antibiotic, amphotericin B or nystatin, to the pipette solution. Amphotericin (stock solution of 60 mg/ml in DMSO) was used at a final concentration of 300 ␮g/ml. Nystatin (stock solution of 10 mg/ml in DMSO) was used at a final concentration of 200 ␮g/ml. The tip of the pipette was filled with antibiotic- free solution, and the pipette was then back-filled with the amphotericin- or nystatin-containing solution. The was considered satisfactory when the series conductance was Ͼ35–40 nS. 2؉ FIG. 1. Lack of oscillations of IKATP at stable [Ca ]c and glucose Thapsigargin was obtained from Sigma (St. Louis, MO) or from Alomone ␤ concentrations in single mouse -cells. IKATP was monitored by pulses Labs (Jerusalem, Israel), diazoxide from Schering-Plough Avondale (Rath- of ؎20 mV from a holding potential of ؊70 mV using the perforated drum, Ireland), and nimodipine from Bayer (Wuppertal, Germany). Unless mode of the patch-clamp technique. A–C: The amplitude of IKATP is otherwise stated, all other chemicals were from Sigma. reflected by the size of the vertical bars around the continuous thick -Preparation of cells. The pancreases were taken from Naval Medical line representing the holding current at ؊70 mV. The glucose concen Research Institute mice killed by cervical dislocation. Pancreatic islets were tration (G) was either 10 mmol/l throughout (A) or was alternated between 8 and 12 mmol/l (B). C: The average amplitude of I in the isolated aseptically by collagenase digestion followed by hand selection. To KATP 2ϩ presence of G8 and G12 was measured during the last 12 test pulses at obtain single cells, the islets were first incubated for 5 min in a Ca -free each glucose concentration in the experiments illustrated in B.*P < medium. After a brief centrifugation, this solution was replaced by culture 0.05 vs. G8 by unpaired t test. D: Azide was added when indicated. medium, and the islets were disrupted by gentle pipetting through a sili- Traces A and D are representative of results obtained in five cells. conized glass pipette. The cells were plated on 22 mm–diameter glass Trace B is the mean of results obtained in four cells. coverslips and maintained for 1–4 days in RPMI 1640 tissue culture medium containing 10 mmol/l glucose, 10% heat-inactivated FCS, 100 IU/ml penicillin, and 100 ␮g/ml streptomycin. activation of voltage-dependent Ca2ϩ channels. To prevent the capacitive ␤ Electrophysiological recordings. Two criteria were used to identify -cells. transient due to electrical charge of the pipette, which might complicate IKATP The capacitance of mouse ␣-, ␦-, and ␤-cells averages 4.4, 5, and 7.4 pF, measurements, each change in voltage was preceded by 100-ms ramps (except respectively (29). Therefore, only large cells with a capacitance Ͼ5 pF were for in Fig. 1D). Two protocols of depolarization were used: either a single 30-s chosen for the present study. For 150 randomly chosen cells, the average pulse to 0 mV (Figs. 2 and 3) or a train of mimicking the capacitance was 7.6 Ϯ 0.2 pF. After verification of the capacitance, a oscillations of the membrane potential induced by 10 mmol/l glucose in whole depolarizing protocol was applied to identify the properties of the voltage- islets. These trains consisted in the succession of 18-s hyperpolarizing phases, ϩ Ϫ dependent Na current, which is known to be inactivated at resting potential during which IKATP was measured (pulses of 20 mV from a holding potential in ␤-cells but not in ␣- and ␦-cells (30). Thus, cells in which a Naϩ current of Ϫ60 mV), and 6-s depolarizing phases. During these depolarizing phases, could be activated by a small depolarizing pulse from a prolonged holding action potentials were mimicked by 50-ms pulses from Ϫ40 to 0 mV; their potential of Ϫ70 mV were discarded. By contrast, cells that displayed a Naϩ frequency was 10 Hz at the beginning of the pulse and decreased progressively current only after a hyperpolarizing pulse to Ϫ140 mV were considered to be to 5 Hz at the end of the pulse (Figs. 5 and 6). 2؉ ␮ ␤ -cells (30,31) and were used for the experiments. [Ca ]c measurements. Islet cells were loaded with 1 mol/l fura-2/AM Patch-clamp measurements were carried out using the perforated whole- (Molecular Probes, Eugene, OR) for 45 min at 37°C in a bicarbonate-buffered cell mode of the patch-clamp technique at 33–35°C, using an EPC-9 patch- solution containing 10 mmol/l glucose. The glass coverslips onto which the clamp amplifier (Heka Electronics, Lambrecht/Pfalz, Germany) and the loaded cells were attached constituted the bottom of a temperature-controlled software Pulsefit, or an Axopatch 200 B patch-clamp amplifier (Axon Instru- perifusion chamber (Intracell, Royston, Herts, U.K.) mounted on the stage of ments, Foster City, CA) and the software pClamp 8. Patch pipettes were pulled an inverted microscope. The Ca2ϩ probe within the cells was excited at 340 from borosilicate glass capillaries (World Precision Instruments, Hertford- and 380 nm, and the fluorescence emitted at 510 nm was captured at 20 Hz by ⍀ 2ϩ shire, U.K.) to give a resistance of 4–5M . a photometric-based system (PTI, Lawrenceville, NJ). [Ca ]c was calculated Ϯ IKATP was monitored by 100 ms–duration pulses of 20 mV from a holding by comparing the ratio of the 510-nm signals successively acquired at 340 and potential of Ϫ70 mV (Figs. 1–4) or pulses of Ϫ20 mV from a holding potential 380 nm with a calibration curve based on the equation of Grynkiewicz et al. of Ϫ60 mV (Figs. 5 and 6). In the latter protocol, Ϫ60 mV was chosen because (32) and established by filling the chamber with an intracellular-type solution it corresponds best to the interburst potential in spontaneously oscillating containing 10 ␮mol/l fura-2 free acid, and either 10 mmol/l free Ca2ϩ or Ͻ1 2ϩ 2ϩ cells within an islet, whereas the depolarizing pulses were omitted to avoid nmol/l free Ca .AKd for the fura-2-Ca complex of 224 nmol/l was used.

DIABETES, VOL. 51, FEBRUARY 2002 377 ␤ 2؉ ؉ EFFECT OF [Ca ]c ON K -ATP CURRENT IN -CELLS

2؉ FIG. 3. Effects of a 30-s depolarization on [Ca ]c and IKATP measured -simultaneously in single mouse ␤-cells when Ca2؉ influx was pre -vented. Single ␤-cells loaded with Fura-2 were perifused with a Ca2؉ free medium (A)oraCa2؉-containing medium supplemented with 10 ␮mol/l nimodipine (Nimo) (B). They were submitted to a 30-s depolar- ⌬ ization to 0 mV ( Vm), as in Fig. 2. The glucose concentration of the medium was 10 mmol/l throughout. The traces are means ؎ SE of results obtained in three (A) and four (B) cells. ϳ oscillations of IKATP could be detected over a period of 8 min (Fig. 1A), which is approximately twice as long as the 2ϩ period of the spontaneous oscillations of [Ca ]c induced ␤ by the sugar in single -cells (33). In contrast, IKATP fluctuations were detected when cell metabolism was changed by alternating the glucose concentration of the perifusion medium between 12 and 8 mmol/l (Fig. 1B). Average IKATP was two times larger in the presence of 8

2؉ mmol/l glucose than in the presence of 12 mmol/l glucose FIG. 2. Effects of a 30-s depolarization on [Ca ]c and IKATP measured simultaneously in single mouse ␤-cells. Single ␤-cells loaded with (Fig. 1C). Moreover, decreasing the ATP-to-ADP ratio with fura-2 were perifused with a medium containing 3 (E), 10 (A and C), or a low concentration of azide (36), a mitochondrial poison, F) mmol/l glucose (G) alone or 10 mmol/l glucose ؉ 250 ␮mol/l) 25 tolbutamide (Tolb) (B and D). They were submitted to a 30-s depolar- reversibly increased IKATP (Fig. 1D). Therefore, the ab- ⌬ ؊ izing step from 70to0mV( Vm) during the period shown by the thick sence of apparent oscillations of IKATP at stable glucose horizontal bar. I could not be monitored during the depolarization. 2ϩ KATP and [Ca ]c suggests that no intrinsic metabolic oscilla- A and B show representative traces. C–F show mean traces ؎ SE. 2ϩ were performed with cells from the same tions, independent from changes in [Ca ]c, exist in (7 ؍ and F (n (6 ؍ Series E (n .cells-␤ .(5 ؍ and D (n (7 ؍ cultures but different from those of series C (n Influence of a depolarization-induced [Ca2؉] rise on Presentation of results. The experiments are illustrated by traces that are c IKATP. The alternative hypothesis, suggesting that meta- means or representative traces of results obtained with the indicated number ␤ 2ϩ of cells from at least three different cultures. The statistical significance of bolic oscillations in -cells are driven by [Ca ]c oscilla- differences between means was assessed by paired or unpaired Student’s t test tions, was tested by measuring the effect of an imposed as appropriate. Differences were considered significant at P Ͻ 0.05. 2ϩ 2ϩ increase in [Ca ]c on IKATP. In this series, [Ca ]c and IKATP were measured simultaneously in the same single RESULTS ␤-cells perifused with 10 mmol/l glucose and submitted to 2؉ Measurements of IKATP at stable and low [Ca ]c. To a 30-s depolarizing pulse to 0 mV from a holding potential search for the existence of intrinsic Ca2ϩ-independent of Ϫ70 mV (Fig. 2A and C). In ␤-cells held hyperpolarized Ϫ 2ϩ metabolic oscillations, IKATP was measured in single met- at 70 mV, [Ca ]c was low and stable, and IKATP was abolically intact ␤-cells hyperpolarized at Ϫ70 mV (Fig. 1). small. Depolarizing the cells to 0 mV rapidly increased 2ϩ The cells were continuously perifused with a glucose [Ca ]c, which slowly returned to basal levels upon repo- Ϫ Ϯ concentration (10 mmol/l) that produces spontaneous larization to 70 mV. IKATP was 276 70% larger just after 2ϩ ␤ [Ca ]c oscillations in unclamped -cells (33). In the compared with before the depolarizing pulse. This in- 2ϩ present experiments, [Ca ]c was low because of the crease was transient, with IKATP decreasing with time to hyperpolarization and not affected by the 20-mV hyperpo- similar values as those before the depolarizing pulse. To larizing and depolarizing pulses used to monitor IKATP (see ascertain that the increased current observed after the the beginning of the recording in Fig. 2A). IKATP was small depolarizing pulse corresponds well to IKATP, the same (1.6 Ϯ 0.3 pA/pF, n ϭ 10), corresponding to 3.4 Ϯ 0.6% of experiment was repeated in the presence of 250 ␮mol/l ϩ the cell total IKATP estimated by the combined application tolbutamide, a potent blocker of K -ATP channels (Fig. 2B ϩ of diazoxide and azide to open K -ATP channels maxi- and D). As expected, tolbutamide reduced IKATP in the mally (34). This result suggests that Ͼ95% of Kϩ-ATP presence of 10 mmol/l glucose (compare the beginning of channels were already closed at 10 mmol/l glucose, as Fig. 2C and D). This inhibition amounted to 63% (0.60 Ϯ previously reported (35). 0.01 pA/pF, n ϭ 5, vs. 1.62 Ϯ 0.03 pA/pF, n ϭ 7, in the During constant stimulation by 10 mmol/l glucose, no presence and absence of tolbutamide, respectively; P Ͻ

378 DIABETES, VOL. 51, FEBRUARY 2002 J.-F. ROLLAND, J.-C. HENQUIN, AND P. GILON

pulse, the same protocol was repeated under conditions where Ca2ϩ influx was prevented. In the absence of 2ϩ 2ϩ external Ca , [Ca ]c did not increase upon depolariza- tion, and IKATP was of similar amplitude before and after the pulse (Fig. 3A). In the presence of 2.5 mmol/l Ca2ϩ and 10 ␮mol/l nimodipine, an L-type Ca2ϩ , the 2ϩ depolarizing pulse to 0 mV, increased [Ca ]c only margin- ally (Fig. 3B). This small elevation may be attributed to the activity of the Naϩ/Ca2ϩ exchanger working in reverse mode at 0 mV or to an incomplete blockade of L-type Ca2ϩ channels. However, it was too small to affect IKATP (Fig. 3B). 2ϩ If a rise in [Ca ]c is really the cause of the increase in 2ϩ IKATP, mobilization of intracellular Ca should produce a similar effect as that of Ca2ϩ influx. Application of 100 ␮ mol/l acetylcholine (ACh), a potent Ins(1,4,5)P3 (IP3)- producing agent, to hyperpolarized ␤-cells reversibly aug- mented IKATP (Fig. 4A and B). To ascertain whether this 2ϩ effect resulted from a rise in [Ca ]c, the same protocol was repeated after treatment of the cell with thapsigargin, a potent and specific inhibitor of the sarco-endoplasmic reticulum Ca2ϩ-ATPase. Thapsigargin depletes the endo- plasmic reticulum of Ca2ϩ in ␤-cells (37) without impairing the production of IP3 in response to phospholipase C–linked agonists. Addition of ACh to thapsigargin-pre- treated cells did not affect IKATP (Fig. 4C). Altogether, 2ϩ these experiments demonstrate that the rise in [Ca ]c is the mechanism that increases IKATP. 2؉ Effect of imposed [Ca ]c oscillations on IKATP. Be- cause 30-s depolarizations to 0 mV might be stronger than spontaneous depolarizations, single cells were depolarized ␤ FIG. 4. Effects of ACh on IKATP in single mouse -cells. Single cells were perifused with a medium containing 10 mmol/l glucose (G) throughout, by a voltage clamp protocol mimicking the spontaneous and 100 ␮mol/l ACh was added when indicated. C: Cells were pre- electrical activity in islets. Cycles of 6 s depolarization and treated for 1 h with 1 ␮mol/l thapsigargin. Trace A is representative, 18 s hyperpolarization were chosen to reproduce the and traces B and C are means ؎ SE of results obtained in four single cells. durations of the depolarization and repolarization phases elicited by 10 mmol/l glucose (38). During depolarization, 0.001). In contrast, tolbutamide did not affect the rise in the cell was submitted to short depolarizing pulses resem- 2ϩ [Ca ]c produced by the depolarizing pulse to 0 mV. bling the burst of of the slow waves (see However, the increase in current observed after the depo- RESEARCH DESIGN AND METHODS). The left part of Fig. 5 shows 2ϩ larizing pulse was abolished (compare Fig. 2C and D). spontaneous oscillations of [Ca ]c induced by 10 mmol/l ␤ 2ϩ If the current activated by the depolarizing pulse is glucose in a single -cell. The right part shows [Ca ]c IKATP, one could anticipate that it will be decreased by high oscillations imposed by the voltage clamp protocol in the glucose. This finding was tested by applying a 30-s depo- same cell, ϳ5 min after establishment of the seal. The 2ϩ larizing pulse to cells perifused with 3 or 25 mmol/l imposed [Ca ]c oscillations were similar to those occur- glucose (Fig. 2E and F). As expected, IKATP measured ring spontaneously in that cell. The average peak of 2ϩ before the depolarization to 0 mV was reduced by 45% in [Ca ]c oscillations in several cells was not different the presence of the high concentration of glucose (1.86 Ϯ during spontaneous oscillations (1,053 Ϯ 91 nmol/l, n ϭ 0.02 pA/pF in G3, n ϭ 6, vs. 1.03 Ϯ 0.02 pA/pF in G25, n ϭ 23) or during oscillations imposed by the pulse protocol 7, respectively; P Ͻ 0.001). Importantly, the increase in (802 Ϯ 132 nmol/l, n ϭ 12) or 30-s depolarizations to 0 mV Ϯ ϭ 2ϩ current observed after the 30-s depolarization to 0 mV was (823 103 nmol/l, n 7). Imposed [Ca ]c oscillations are threefold smaller in 25 mmol/l glucose than in 3 mmol/l thus within the physiological range. glucose (1.70 Ϯ 0.26 pA/pF in G25, n ϭ 7, vs. 5.10 Ϯ 0.94 The same pulse protocol as that used in Fig. 5 was then ϭ Ͻ 2ϩ 2ϩ pA/pF, n 6, in G3; P 0.01), although the rise in [Ca ]c applied to measure the influence of [Ca ]c oscillations on was similar at both glucose concentrations. The time for IKATP (Fig. 6). The cells were initially perifused with 6 IKATP to return to basal levels was also much reduced in mmol/l glucose, a subthreshold concentration at which the the presence of 25 mmol/l glucose. Altogether, these islets are electrically silent (1). Increasing glucose from 6 ϭ experiments demonstrate that the increased current ob- to 10 mmol/l decreased IKATP from 1.57 to 0.89 pA/pF (n served after the pulse in control cells does correspond to 8). This difference in current is probably larger than that 2ϩ IKATP and that the negative feedback effect of [Ca ]c on occurring in a cell that would not be voltage-clamped and IKATP can be modulated by glucose. in which the decrease in IKATP produced by the accelera- To ascertain that the increase in IKATP is the conse- tion of ATP production in response to the elevation of the 2ϩ quence of the rise in [Ca ]c produced by the depolarizing glucose concentration is normally counterbalanced by the

DIABETES, VOL. 51, FEBRUARY 2002 379 ␤ 2؉ ؉ EFFECT OF [Ca ]c ON K -ATP CURRENT IN -CELLS

␤ 2؉ FIG. 5. Spontaneous and voltage clamp–imposed [Ca ]c oscillations in the same single -cell. A single cell loaded with Fura-2 was perifused with 2؉ a medium containing 10 mmol/l glucose throughout. Spontaneous [Ca ]c oscillations were recorded before the establishment of the seal (left panel). Approximately 5 min after establishment of the seal, the cell was submitted to two series of trains of depolarizing pulses designed to mimic the slow waves of the membrane potential induced by 10 mmol/l glucose in whole islets (see RESEARCH DESIGN AND METHODS) and illustrated on the top of the figure (right panel). The shaded areas represent the ؊20 mV hyperpolarizing pulses that were applied from the holding potential ؊ 2؉ of 60 mV. This trace is an example of results obtained in six cells that displayed spontaneous [Ca ]c oscillations with different frequencies.

2ϩ Ϯ increase in IKATP due to the concomitant rise in [Ca ]c.In 0.59 0.06 pA/pF. A current of similar amplitude, adjusted cells voltage-clamped between Ϫ60 and Ϫ80 mV (Fig. 6), for cell size (0.59 multiplied by the capacitance of the ␤ IKATP is only influenced by the change in glucose metabo- tested cell), was then injected into -cells studied in the 2ϩ lism but not by the rise in [Ca ]c that is prevented by the current-clamp mode and stimulated by 10 mmol/l glucose. hyperpolarization. Application of trains of depolarization Figure 7A shows the electrical activity induced by glucose repetitively increased IKATP (Fig. 6). The average increase in one of these cells. Injection of a hyperpolarizing current ⌬ Ϫ was such that the current after each train was similar corresponding to the average IKATP ( 5 pA in this cell) (1.68 Ϯ 0.09 pA/pF) to that measured in the presence of 6 suppressed the electrical activity and repolarized the mmol/l glucose. This increase in IKATP might thus be plasma membrane to the resting level. Removal of this sufficient to repolarize the membrane below the threshold current was accompanied by the immediate resumption of potential for activation of voltage-dependent Ca2ϩ chan- action potentials. In other experiments (Fig. 7B), the nels. The changes in current induced by the rise of the hyperpolarizing current was increased stepwise by incre- ⌬ glucose concentration and by the pulse protocol were all ments corresponding to one-sixth of the average IKATP. ␮ ⌬ prevented by 250 mol/l tolbutamide, demonstrating that As shown in Fig. 7B, 50% of average IKATP was sufficient ϭ they really correspond to variations in IKATP (n 5, not to repolarize the cell below the threshold for activation of shown). voltage-dependent Ca2ϩ channels. This result strongly 2ϩ Effect of injection of a hyperpolarizing current equiv- suggests that the Ca -induced rise in IKATP might control 2؉ alent to the Ca -induced increase in IKATP on the the oscillations of the membrane potential. ␤- potential. We next verified whether 2ϩ the Ca -induced increase in IKATP is sufficient to repolar- ize the plasma membrane to the resting potential. This DISCUSSION ⌬ increase ( IKATP) was calculated by averaging the differ- Oscillations of the membrane potential are one of the ␤ ence in IKATP after and before the last four trains of major characteristics of the pancreatic -cell response to ⌬ 2ϩ depolarizing pulses ( IKATP1–4 in Fig. 6). It amounted to glucose. They underlie the periodic influx of Ca that

380 DIABETES, VOL. 51, FEBRUARY 2002 J.-F. ROLLAND, J.-C. HENQUIN, AND P. GILON

␤ 2؉ 2؉ FIG. 6. Increase of IKATP by imposed [Ca ]c oscillations mimicking spontaneous [Ca ]c oscillations induced by glucose. Single -cells were initially perifused with a medium containing 6 mmol/l glucose (G6). After 1 min of recording IKATP, the glucose concentration was increased to 10 mmol/l (G10). Two minutes later, the cell was submitted to the same pulse protocol as that used in Fig. 5 and designed to mimic spontaneous ⌬ 2؉ [Ca ]c oscillations induced by 10 mmol/l glucose. The mean difference in IKATP before and after the depolarizing pulses (average IKATP) was ⌬ calculated by averaging the increase in IKATP occurring after each of the last four trains of depolarizations ( IKATP1–4). This trace is the mean of results obtained in eight single cells.

triggers oscillations of insulin secretion. Understanding 4 mmol/l glucose changes). Experiments monitoring O2 their fine control is thus of utmost importance. The consumption (39), glucose consumption (27), and the 2ϩ present study demonstrates that [Ca ]c oscillations in fluorescence of reduced pyridine nucleotides [NAD(P)H) ␤ pancreatic -cells rhythmically increase IKATP and provide in single islets (6)] have also concluded to the absence of direct support to the proposal (20) that such an effect Ca2ϩ-independent metabolic oscillations in ␤-cells. 2؉ constitutes a feedback control of the oscillations of mem- IKATP oscillations driven by [Ca ]c oscillations. When 2ϩ 2ϩ brane potential. [Ca ]c was increased by stimulating Ca influx or mobi- 2ϩ 2ϩ Intrinsic metabolic oscillations do not drive mem- lizing Ca from intracellular Ca stores, IKATP increased. brane potential oscillations. It has been suggested that There is no doubt that this increase resulted from the 2ϩ 2ϩ 2ϩ intrinsic Ca -independent metabolic oscillations exist in [Ca ]c rise because IKATP did not change when Ca ␤-cells (24) and that they lead to cycles of Kϩ-ATP channel influx was prevented by omission of external Ca2ϩ and activity (23). To verify this hypothesis, single metabolically blockade of voltage-dependent Ca2ϩ channels or when ␤ 2ϩ 2ϩ intact -cells were hyperpolarized to keep [Ca ]c at basal Ca mobilization was prevented by pretreatment with and stable levels, and IKATP was continuously monitored thapsigargin. It is also clear that the current increased by 2ϩ during perifusion with a stimulatory glucose concentra- the rise in [Ca ]c is IKATP because it was attenuated by a tion. In no cell did we find IKATP oscillations under these rise in ambient glucose concentration and completely conditions. This suggests that either no intrinsic metabolic inhibited by tolbutamide, a blocker of Kϩ-ATP channels. oscillations exist, or they are unable to modulate Kϩ-ATP Kϩ channels sensitive to sulfonylureas but distinct from channel activity and membrane potential because of their Kϩ-ATP channels have been described in some systems nature or small amplitude (smaller than those imposed by (40,41), but not in ␤-cells. It is likely that the current that

DIABETES, VOL. 51, FEBRUARY 2002 381 ␤ 2؉ ؉ EFFECT OF [Ca ]c ON K -ATP CURRENT IN -CELLS

study. Others did not find any direct effect of Ca2ϩ on Kϩ-ATP channels in ␤-cells (9). It is worth noting that the Kϩ-ATP channels of ␤-cells and muscle cells have different subunit compositions (SUR1 and Kir6.2 for ␤-cells and SUR2 and Kir6.2 for muscle cells) (47), which might confer different sensitivities to Ca2ϩ. Several Ca2ϩ-dependent processes influencing Kϩ-ATP channels have been described in pancreatic ␤-cells or muscle cells. They involve cytoskeletal (44), the Ca2ϩ-dependent phosphatase type 2B (48), or other proteins (49). Activation of phospholipase C by Ca2ϩ, with subsequent hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2), is unlikely to be involved for two reasons. First, acceleration of PIP2 breakdown would be expected to decrease IKATP (44), which is opposite to the 2ϩ effect of a rise in [Ca ]c observed in the present study. Second, ACh, a potent Ca2ϩ-independent activator of ␤ phospholipase C, was without effect on -cell IKATP when Ca2ϩ mobilization was prevented by thapsigargin pretreat- ment. 2ϩ Metabolic oscillations might be driven by [Ca ]c oscil- 2ϩ lations. Indeed, each rise in [Ca ]c could stimulate ATP production (50) and increase the ATP-to-ADP ratio by activating mitochondrial dehydrogenases (51,52). Oscilla- 2ϩ tions of oxygen consumption driven by [Ca ]c oscilla- tions have recently been reported in islets (27). Our data do not exclude this possibility. Alternatively, each rise in 2ϩ [Ca ]c could decrease the ATP-to-ADP ratio. This hypoth- esis is supported by direct measurements of adenine FIG. 7. Effect of injection of a hyperpolarizing current on the ␤-cell nucleotide levels within mouse islets (25) or of ATP membrane potential. The membrane potential of single ␤-cells was concentration in INS-1 cells expressing luciferase (53). monitored in the current-clamp mode of the patch-clamp technique. These studies demonstrated that, at a fixed glucose con- The glucose concentration of the medium was 10 mmol/l throughout. No current (0) was injected into the cells except when indicated by the centration, the ATP-to-ADP ratio and the ATP concentra- 2ϩ ϩ downward deflections of the upper traces above each panel. The value tion decreased when [Ca ]c was raised by high K .By of 0.59 pA/pF was calculated from the experiments illustrated in Fig. 6. 2ϩ ؊5 pA in this cell) was demonstrating that a rise in [Ca ]c increases IKATP, the؍ In A, the full current (؊0.59 ؋ 8.5 pF repetitively injected. In B, the current was increased by steps of present study supports this proposal. The drop in the one-sixth of the total current. When no current was injected, the cell in ATP-to-ADP ratio could either result from inhibition of A was continuously depolarized during the recording, whereas the cell in B showed a spontaneous depolarization (seen at the beginning of the ATP production (26,54) or stimulation of ATP consump- recording). The traces are representative of results obtained in five tion (25,53). (A) and four (B) single cells. Physiological implications for the control of mem- brane potential oscillation. In glucose-stimulated ␤ we studied here is similar to the voltage-independent -cells, IKATP was found to be larger during the interburst Ca2ϩ-activated Kϩ current previously described in ␤-cells intervals than during the depolarizing phases (22). These (14). This current, which was originally thought to not fluctuations were tentatively ascribed to metabolic oscil- involve Kϩ-ATP channels (14), was recently found to be lations, but no mechanistic explanation was provided. The 2ϩ largely sensitive to tolbutamide by the same authors (42). present study strongly suggests that the rise in [Ca ]c 2؉ Mechanisms by which a rise in [Ca ]c increases might be the feedback mechanism controlling IKATP and 2ϩ IKATP. A rise in [Ca ]c could increase IKATP by different hence the oscillations of the membrane potential. Thus, mechanisms, including a direct action of Ca2ϩ on Kϩ-ATP during a voltage clamp protocol mimicking the repetitive channels, an indirect action through Ca2ϩ-sensitive regu- changes in electrical activity induced by 10 mmol/l glucose 2ϩ lators of the channels, and an indirect action through in islets, each imposed [Ca ]c oscillation evoked a tran- 2ϩ changes in cell metabolism. A direct effect of Ca on sient increase in IKATP. This increase had a similar ampli- ϩ K -ATP channels has been reported in inside-out patches tude to that of the difference in IKATP measured at of membranes of normal ␤-cells or tumoral insulin-secret- substimulating (6 mmol/l) and stimulating (10 mmol/l) ing RINm5F cells in which application of Ca2ϩ inhibited glucose concentrations. Theoretically, this current should Kϩ-ATP channels (millimolar range of Ca2ϩ) (43) and be able to repolarize the membrane to a potential more attenuated the ADP-induced channel activation (micromo- negative than that of the activation threshold of voltage- lar range of Ca2ϩ) (44). Ca2ϩ increased the ability of ATP dependent Ca2ϩ channels. This finding was amply sup- to block Kϩ-ATP channels or inactivated these channels in ported by current-clamp experiments. Injection of current 2ϩ inside-out patches of (45) and ventricular corresponding to 50% of the Ca -induced IKATP increase (46) membranes. However, all these effects are opposite to was sufficient to repolarize the ␤-cell membrane in the 2ϩ the Ca -induced increase in IKATP observed in the present presence of 10 mmol/l glucose. Because the voltage-

382 DIABETES, VOL. 51, FEBRUARY 2002 J.-F. ROLLAND, J.-C. HENQUIN, AND P. GILON dependent Ca2ϩ current is larger in ␤-cells within intact REFERENCES islets than in isolated single cells (29), it is possible that 1. Henquin JC, Meissner HP: Significance of ionic fluxes and changes in 2ϩ membrane potential for stimulus-secretion coupling in pancreatic ␤-cells. the amplitude of the Ca -induced increase in IKATP in whole islets exceeds our estimate. We have no explanation Experientia 40:1043–1052, 1984 2. Atwater I, Carroll P, Li MX: of the pancreatic ␤-cell. In why, in a previous report, no oscillations of IKATP were Insulin Secretion. Draznin B, Melmed S, Leroith D, Eds. Liss A.R., New detected in single ␤-cells displaying membrane potential York, 1989, p. 49–68 oscillations (55). The reported experimental procedures 3. Cook DL, Satin LS, Hopkins WF: Pancreatic ␤-cells are bursting, but how? were indeed similar to those used in the present study. Trends Neurosci 14:411–414, 1991 2ϩ 4. Gomis A, Sa´nchez-Andre´s JV, Valdeolmillos M: Oscillatory patterns of The negative feedback effect of [Ca ]c on IKATP might ␤ electrical activity in mouse pancreatic islets of Langerhans recorded in be important for the control of oscillations of the -cell vivo. Plfu¨gers Arch 432:510–515, 1996 membrane potential according to the following model. 5. Santos RM, Rosario LM, Nadal A, Garcia-Sancho J, Soria B, Valdeolmillos ␤ 2ϩ Acceleration of glucose metabolism in -cells increases M: Widespread synchronous [Ca ]i oscillations due to bursting electrical the ATP-to-ADP ratio, which closes Kϩ-ATP channels. This activity in single pancreatic islets. Plfu¨gers Arch 418:417–422, 1991 6. Gilon P, Henquin JC: Influence of membrane potential changes on cyto- leads to depolarization of the plasma membrane and ϩ 2ϩ 2ϩ plasmic Ca2 concentration in an electrically excitable cell, the insulin- opening of voltage-dependent Ca channels. Ca influx ␤ 2ϩ secreting pancreatic -cell. J Biol Chem 267:20713–20720, 1992 then raises [Ca ]c, which decreases the ATP-to-ADP ratio ␤ ϩ 7. Ashcroft FM, Rorsman P: Electrophysiology of the pancreatic -cell. Prog (25) and leads to reopening of K -ATP channels, partial Biophys Mol Biol 54:87–143, 1989 ϩ repolarization of the plasma membrane, arrest of Ca2 8. Detimary P, Jonas JC, Henquin JC: Possible links between glucose-induced 2ϩ changes in the energy state of pancreatic ␤-cells and insulin release: influx, and a drop in [Ca ]c. The eventual restoration of a high ATP-to-ADP ratio then initiates a new cycle. Our unmasking by decreasing a stable pool of adenine nucleotides in mouse ϩ islets. J Clin Invest 96:1738–1745, 1995 2 ϩ observation that the negative feedback effect of [Ca ]c on 9. Cook DL, Hales CN: Intracellular ATP directly blocks K channels in ␤ IKATP is largely attenuated by glucose can explain the pancreatic -cells. Nature 311:271–273, 1984 lengthening of the depolarized phases and shortening of 10. Rorsman P, Trube G: Glucose dependent Kϩ-channels in pancreatic ␤-cells the repolarized intervals occurring when the glucose con- are regulated by intracellular ATP. Plfu¨gers Arch 405:305–309, 1985 11. Henquin JC: Opposite effects of intracellular Ca2ϩ and glucose on Kϩ centration is raised within the stimulatory range. Indeed, permeability of pancreatic islet cells. Nature 280:66–68, 1979 as the glucose concentration increases, the ATP-to-ADP 12. Atwater I, Dawson CM, Ribalet B, Rojas E: permeability ϩ ratio rises and closes more K -ATP channels. The depo- activated by intracellular calcium concentration in the pancreatic larizing phase progressively becomes longer because a ␤-cell. J Physiol 288:575–588, 1979 2ϩ 13. A¨ mma¨la¨ C, Larsson O, Berggren PO, Bokvist K, Juntti-Berggren L, Kind- stronger feedback effect of Ca on IKATP is required to mark H, Rorsman P: Inositol trisphosphate-dependent periodic activation ϩ ϩ counteract the effects of increased metabolism on IKATP. ofaCa2 -activated K conductance in glucose-stimulated pancreatic Continuous electrical activity occurs at glucose concen- ␤-cells. Nature 353:849–852, 1991 14. Go¨ pel SO, Kanno T, Barg S, Eliasson L, Galvanovskis J, Renstro¨ mE, trations that reduce IKATP to such an extent that it is no 2ϩ Rorsman P: Activation of Ca2ϩ-dependent Kϩ channels contributes to longer counteracted by the [Ca ]c rise unless the latter is 2ϩ rhythmic firing of action potentials in mouse pancreatic ␤-cells. J Gen increased by augmenting the extracellular Ca concen- Physiol 114:759–770, 1999 tration (20). 15. Kukuljan M, Goncalves AA, Atwater I: Charybdotoxin-sensitive K(Ca) ϩ The central role of K -ATP channels in membrane channel is not involved in glucose-induced electrical activity in pancreatic potential oscillations suggested in our model are compat- ␤-cells. J Membr Biol 119:187–195, 1991 ϩ 16. Keizer J, Smolen P: Bursting electrical activity in pancreatic ␤-cells caused ible with those in most studies on K -ATP channel- by Ca2ϩ- and voltage-inactivated Ca2ϩ channels. Proc Natl Acad SciUSA deficient mice. Thus, pancreatic ␤-cells from SUR1 or Kir 88:3897–3901, 1991 6.2 knockout mice display a continuous spike activity 17. Andreu E, Soria B, Sanchez-Andres JV: Oscillation of gap junction electri- 2ϩ cal coupling in the mouse pancreatic islets of Langerhans. J Physiol (56,57) and a sustained and stable elevation of [Ca ]c (56,58) at both nonstimulating and stimulating glucose (Lond) 498:753–761, 1997 2ϩ 18. Dukes ID, Roe MW, Worley JF III, Philipson LH: Glucose-induced alter- concentrations. Only one abstract reported [Ca ]c oscil- ations in ␤-cell cytoplasmic Ca2ϩ involving the coupling of intracellular lations in ␤-cells from SUR1 knockout mice, but it is not Ca2ϩ stores and plasma membrane ion channels. Curr Opin Endocrinol known whether these oscillations resulted from concom- Diabetes 4:262–271, 1997 itant changes in membrane potential (59). In view of the 19. Gilon P, Arredouani A, Gailly P, Gromada J, Henquin JC: Uptake and ϩ release of Ca2ϩ by the endoplasmic reticulum contribute to the oscillations important role played by K -ATP channels in the control of the cytosolic Ca2ϩ concentration triggered by Ca2ϩ influx in the of pancreatic ␤-cell membrane potential, further studies electrically excitable pancreatic ␤-cell. J Biol Chem 274:20197–20205, 1999 2ϩ ␤ should now elucidate the interplay between [Ca ]c and 20. Henquin JC: Glucose-induced electrical activity in -cells: feedback con- ATP turnover. trol of ATP-sensitive Kϩ channels by Ca2ϩ. Diabetes 39:1457–1460, 1990 21. Ding WG, He LP, Omatsu-Kanbe M, Kitasato H: A possible role of the ATP-sensitive potassium in determining the duration of spike-bursts in mouse pancreatic ␤-cells. Biochim Biophys Acta 1279:219– ACKNOWLEDGMENTS 226, 1996 This work was supported by Grant 3.4552.98 from the 22. Larsson O, Kindmark H, Bra¨nstro¨ m R, Fredholm B, Berggren PO: Oscilla- Fonds de la Recherche Scientifique Me´dicale (Brussels), tions in KATP channel activity promote oscillations in cytoplasmic free 2ϩ by Grants 1.5.212.00 and 1.5.121.00 from the Fonds Na- Ca concentration in the pancreatic ␤-cell. Proc Natl Acad SciUSA 93:5161–5165, 1996 tional de la Recherche Scientifique (Brussels), by Grant 23. Dryselius S, Lund P-E, Gylfe E, Hellman B: Variations in ATP-sensitive Kϩ ARC 00/05-260 from the General Direction of Scientific channel activity provide evidence for inherent metabolic oscillations in Research of the French Community of Belgium, and by the pancreatic ␤-cells. Biochem Biophys Res Commun 205:880–885, 1994 Interuniversity Poles of Attraction Program (P4/21), Fed- 24. Tornheim K: Are metabolic oscillations responsible for normal oscillatory insulin secretion? Diabetes 46:1375–1380, 1997 eral Office for Scientific, Technical and Cultural Affairs of 25. Detimary P, Gilon P, Henquin JC: Interplay between cytoplasmic Ca2ϩ and Belgium. P.G. is Maıˆtre de Recherches of the Fonds the ATP/ADP ratio: a feedback control mechanism in mouse pancreatic National de la Recherche Scientifique, Brussels. islets. Biochem J 333:269–274, 1998

DIABETES, VOL. 51, FEBRUARY 2002 383 ␤ 2؉ ؉ EFFECT OF [Ca ]c ON K -ATP CURRENT IN -CELLS

26. Krippeit-Drews P, Du¨ fer M, Drews G: Parallel oscillations of intracellular 44. Koriyama N, Kakei M, Nakazaki M, Yaekura K, Ichinari K, Gong Q,

calcium activity and mitochondrial membrane potential in mouse pancre- Morimitsu S, Yada T, Tei C: PIP2 and ATP cooperatively prevent cytosolic atic ␤-cells. Biochem Biophys Res Commun 267:179–183, 2000 Ca2ϩ-induced modification of ATP-sensitive Kϩ channels in rat pancreatic 27. Jung SK, Kauri LM, Qian WJ, Kennedy RT: Correlated oscillations in ␤-cells. Diabetes 49:1830–1839, 2000 ϩ glucose consumption, oxygen consumption, and intracellular free Ca2 in 45. Hehl S, Moser C, Weik R, Neumcke B: Internal Ca2ϩ inactivate and single islets of Langerhans. J Biol Chem 275:6642–6650, 2000 modify ATP-sensitive potassium channels in adult mouse skeletal muscle. ϩ ϩ 28. Rolland JF, Henquin JC, Gilon P: Modulation of the K ATP current by Ca2 Biochim Biophys Acta 1190:257–263, 1994 may contribute to oscillations of the membrane potential in ␤-cells 46. Findlay I: Calcium-dependent inactivation of the ATP-sensitive Kϩ channel (Abstract). Diabetologia 43 (Suppl. 1):A116, 2000 of rat ventricular myocytes. Biochim Biophys Acta 943:297–304, 1988 29. Go¨ pel SO, Kanno T, Barg S, Rorsman P: Patch-clamp characterisation of 47. Bryan J, Aguilar-Bryan L: Sulfonylurea receptors: ABC transporters that ␦ ϩ somatostatin-secreting -cells in intact mouse pancreatic islets. J Physiol regulate ATP-sensitive K channels. Biochim Biophys Acta 1461:285–303, 528:497–507, 2000 1999 30. Go¨ pel SO, Kanno T, Barg S, Galvanovskis J, Rorsman P: Voltage-gated and 48. Wilson AJ, Jabr RI, Clapp LH: Calcium modulation of vascular smooth ␤ resting membrane currents recorded from -cells in intact mouse pancre- muscle ATP-sensitive Kϩ channels: role of protein phosphatase-2B. Circ atic islets. J Physiol (Lond) 521:717–728, 1999 ϩ Res 87:1019–1025, 2000 31. Plant TD: Na ␤ Plfu¨gers Arch ϩ currents in cultured mouse pancreatic -cells. 49. Rorsman P, Trube G: Biophysics and physiology of ATP-regulated K 411:429–435, 1988 channels (K ). In Potassium Channels: Structure, Classification, Func- 32. Grynkiewicz G, Poenie M, Tsien RY: A new generation of Ca2ϩ indicators ATP tion and Therapeutic Potential. Cook NS, Ed. Ellis Horwood series in with greatly improved fluorescence properties. J Biol Chem 260:3440– Pharmaceutical Technology, Chinchester, U.K., 1990, p. 97–116 3450, 1985 ϩ 50. Kennedy HJ, Pouli AE, Ainscow EK, Jouaville LS, Rizzuto R, Rutter GA: 33. Jonkers FC, Henquin JC: Measurements of cytoplasmic Ca2 in islet cell Glucose generates sub-plasma membrane ATP microdomains in single clusters show that glucose rapidly recruits ␤-cell and gradually increases islet ␤-cells: potential role for strategically located mitochondria. J Biol the individual cell response. Diabetes 50:540–550, 2001 34. Garcia-Barrado MJ, Ravier MA, Rolland JF, Gilon P, Nenquin M, Henquin Chem 274:13281–13291, 1999 JC: Inhibition of protein synthesis sequentially impairs distinct steps of 51. MacCormack JG, Longo EA, Corkey BE: Glucose-induced activation of stimulus-secretion coupling in pancreatic ␤-cells. Endocrinology 142:299– pyruvate dehydrogenase in isolated rat pancreatic islets. Biochem J 307, 2001 267:527–530, 1990 35. Ashcroft FM, Ashcroft SJ, Harrison DE: Properties of single potassium 52. MacDonald MJ, Brown LJ: Calcium activation of mitochondrial glycerol channels modulated by glucose in rat pancreatic beta-cells. J Physiol phosphate dehydrogenase restudied. Arch Biochem Biophys 326:79–84, 400:501–527, 1988 1996 36. Detimary P, Gilon P, Nenquin M, Henquin JC: Two sites of glucose control 53. Maechler P, Kennedy ED, Sebo E, Valeva A, Pozzan T, Wollheim CB: of insulin release with distinct dependence on the energy state in pancre- Secretagogues modulate the calcium concentration in the endoplasmic atic ␤-cells. Biochem J 297:455–461, 1994 reticulum of insulin-secreting cells: studies in aequorin-expressing intact 37. Miura Y, Henquin JC, Gilon P: Emptying of intracellular Ca2ϩ stores and permeabilized ins-1 cells. J Biol Chem 274:12583–12592, 1999 stimulates Ca2ϩ entry in mouse pancreatic ␤-cells by both direct and 54. Magnus G, Keizer J: Model of ␤-cell mitochondrial calcium handling and indirect mechanisms. J Physiol (Lond) 503:387–398, 1997 electrical activity. I. Cytoplasmic variables. Am J Physiol 274:C1158– 38. Henquin JC: ATP-sensitive Kϩ channels may control glucose-induced C1173, 1998 electrical activity in pancreatic ␤-cells. Biochem Biophys Res Commun 55. Smith PA, Ashcroft FM, Rorsman P: Simultaneous recordings of glucose 156:769–775, 1988 dependent electrical activity and ATP-regulated K ϩ-currents in isolated 39. Jung SK, Aspinwall CA, Kennedy RT: Detection of multiple patterns of mouse pancreatic ␤-cells. FEBS Lett 261:187–190, 1990 oscillatory oxygen consumption in single mouse islets of Langerhans. 56. Miki T, Nagashima K, Tashiro F, Kotake K, Yoshitomi H, Tamamoto A, Biochem Biophys Res Commun 259:331–335, 1999 Gonoi T, Iwanaga T, Miyazaki J, Seino S: Defective insulin secretion and

40. Seutin V, Shen KZ, North RA, Johnson SW: Sulfonylurea-sensitive potas- enhanced insulin action in KATP channel-deficient mice. Proc Natl Acad Sci sium current evoked by -loading in rat midbrain dopamine . USA95:10402–10406, 1998 Neuroscience 71:709–719, 1996 57. Seghers V, Nakazaki M, DeMayo F, Aguilar-Bryan L, Bryan J: Sur1

41. Gomora JC, Enyeart JJ: Dual pharmacological properties of a cyclic knockout mice: a model for KATP channel-independent regulation of AMP-sensitive . J Pharmacol Exp Ther 290:266–275, insulin secretion. J Biol Chem 275:9270–9277, 2000 1999 58. Efanov AM, Hoy M, Branstrom R, Zaitsev SV, Magnuson MA, Efendic S, 42. Go¨ pel SO, Kanno T, Rorsman P: Two components of activity-dependent Gromada J, Berggren PO: The imidazoline RX871024 stimulates insulin ϩ ␤ transient K -current (IKslow) in mouse pancreatic -cells (Abstract). Dia- secretion in pancreatic beta-cells from mice deficient in KATP channel betologia 44 (Suppl. 1):A20, 2001 function. Biochem Biophys Res Commun 284:918–922, 2001 43. Findlay I: The effects of magnesium upon -sensitive 59. A¨ mma¨la¨ C, Aguilar-Bryan L, Bryan J, Dukes ID: Glucose modulation of potassium channels in a rat insulin-secreting cell line. J Physiol 391:611– intracellular calcium, electrical activity and insulin secretion in Sur1 null 629, 1987 islets (Abstract). Diabetes 50 (Suppl. 2):A350, 2001

384 DIABETES, VOL. 51, FEBRUARY 2002