Endocrinology and Metabolism

Cholinergic regulation of fuel-induced hormone secretion and respiration of SUR1−/− mouse islets

Nicolai M. Doliba, Wei Qin, Marko Z. Vatamaniuk, Carol W. Buettger, Heather W. Collins, Mark A Magnuson, Klaus H. Kaestner, David F. Wilson, Richard D. Carr, Franz M. Matschinsky


Neural and endocrine factors (i.e., Ach and GLP-1) restore defective glucose-stimulated insulin release in pancreatic islets lacking sulfonylurea type 1 receptors (SUR1−/−) (Doliba NM, Qin W, Vatamaniuk MZ, Li C, Zelent D, Najafi H, Buettger CW, Collins HW, Carr RD, Magnuson MA, and Matschinsky FM. Am J Physiol Endocrinol Metab 286: E834–E843, 2004). The goal of the present study was to assess fuel-induced respiration in SUR1−/− islets and to correlate it with changes in intracellular Ca2+, insulin, and glucagon secretion. By use of a method based on O2 quenching of phosphorescence, the O2 consumption rate (OCR) of isolated islets was measured online in a perifusion system. Basal insulin release (IR) was 7–10 times higher in SUR1−/− compared with control (CON) islets, but the OCR was comparable. The effect of high glucose (16.7 mM) on IR and OCR was markedly reduced in SUR1−/− islets compared with CON. Ach (0.5 μM) in the presence of 16.7 mM glucose caused a large burst of IR in CON and SUR1−/− islets with minor changes in OCR in both groups of islets. In SUR1−/− islets, high glucose failed to inhibit glucagon secretion during stimulation with amino acids or Ach. We conclude that 1) reduced glucose responsiveness of SUR1−/− islets may be in part due to impaired energetics, as evidenced by significant decrease in glucose-stimulated OCR; 2) elevated intracellular Ca2+ levels may contribute to altered insulin and glucagon secretion in SUR1−/− islets; and 3) The amplitudes of the changes in OCR during glucose and Ach stimulation do not correlate with IR in normal and SUR1−/− islets suggesting that the energy requirements for exocytosis are minor compared with other ATP-consuming reactions.

  • sulfonylurea receptor 1 knockout mice
  • acetylcholine
  • insulin and glucagon release
  • calcium
  • oxygen consumption

ATP-sensitive k+ (katp) channels, composed of the high-affinity sulfonylurea receptor 1 (SUR1) and the K+ inward rectifier (KIR6.2) (1, 36), play a critical role in the coupling of fuel metabolism to membrane electrical activity, which results in the intracellular Ca2+ concentration ([Ca2+]i) accumulation necessary to promote exocytosis. Some mutations in SUR1 or KIR6.2 cause a recessive form of persistent hyperinsulinemic hypoglycemia of infancy, characterized by oversecretion of insulin despite severe hypoglycemia (1, 40). However, the deletion of SUR1 (1, 35, 37) or Kir6.2 (29) from mouse pancreatic β-cells has unexpectedly little or no effect on glucose homeostasis, in contrast to the predicted hypoglycemic phenotype. Furthermore, in the majority of the previous studies, glucose has little or no effect on the rate of insulin release when tested on isolated islets in vitro (12, 20, 25, 35, 37) or in vivo during a hyperglycemic clamp (37). The lack of a glucose response should have caused hyperglycemia in the intact animal. However, the SUR1 mice secrete nearly normal amounts of insulin in response to feeding (37). In recognition of the essential role of neural and endocrine factors during feeding, we have shown previously (12) that acetylcholine and glucagon-like peptide-1 (GLP-1) restore defective glucose-stimulated insulin release in pancreatic islets of SUR1−/− mice. The SUR1−/− islets were hypersensitive to acetylcholine and IBMX, as evidenced by a marked increase of insulin release (IR) even in the absence of glucose. Pretreatment of perifused SUR1−/− islets with acetylcholine or IBMX restored glucose stimulation of IR. The effect of acetylcholine on IR was greatly potentiated by GLP-1 (12). It was concluded that priming by the combination of high [Ca2+]i, acetylcholine, and GLP-1 restores the defective glucose responsiveness precluding the development of diabetes but not effective enough to cause hyperinsulinemic hypoglycemia. It was still unclear whether the lack of glucose responsiveness of SUR1−/− islets is related to impaired glucose metabolism, specifically glucose oxidation, in β-cells leading to altered bioenergetics. The goal of the present study is to assess fuel-induced respiration in SUR1−/− islets and to correlate it with changes in intracellular Ca2+, insulin, and glucagon secretion. To this end, the O2 consumption rate (OCR) of isolated islets was evaluated in normal and SUR1−/− mice by use of a new method based on O2 quenching of phosphorescence and was correlated with insulin and glucagon secretion and intracellular Ca2+ levels.



SUR1 knockout mice were generated as described (37). The SUR1−/− mice contained a mutation that was made in RW4 ES cells (129/SvJ origin). They had been back-crossed three times into C57Bl/6. The mice were maintained on a 12:12-h light-dark cycle and were fed a standard rodent chow diet. B6D2F1 mice (Jackson Labs) were used as controls (SUR1+/+). For each experiment, 6–8 mice of either sex aged 4–7 mo were used.

Islet Isolation

Mouse islets were isolated using collagenase (EC; Serva, 17449) digestion in Hanks' buffer followed by separation of islets from exocrine tissue in a Ficoll gradient (Sigma, F-9378). Isolated islets were used fresh or cultured for 4 days in RPMI 1640 medium (GIBCO-BRL, Grand Island, NY) containing 10% fetal bovine serum, 10 ml/l penicillin-streptomycin-amphotericin B solution (GIBCO-BRL), and 10 mM glucose (34).

Islet Perifusion and Insulin Release Experiments

Islets were placed in a glass perifusion chamber similar to those in Refs. 43 and 44 (Fig. 1). The perifusion apparatus consisted of a peristaltic pump, a water bath (37°C), a gas exchanger (artificial lung: media flowed through the thin-walled Silastic tubing loosely coiled in a glass jar that containing 20% O2-5% CO2 balanced with N2), and fraction collector (Waters Division of Millipore). All transfer lines were insulated. The islets were loaded into the 250-μl chamber by means of a P200 pipette and a gel-loading tip and were allowed to settle for 1 min before resuming the flow at 80 μl/min. The perifusate was a Krebs buffer (pH 7.4) containing 2.2 mM Ca2+, 1 mM Pi, and 1% bovine serum albumin equilibrated with 20% O2-5% CO2 balanced with N2. Assuming that the flow front is uniform across the width of the chamber, then the glucose concentration within the chamber should change in ∼3 min, the time needed to fill a 250-μl chamber at 80 μl/min. However, the longer response time indicates that mixing takes place as the perfusate traverses the chamber, requiring a longer time to reach a steady state of glucose concentration.

Fig. 1.

A new system for studying respiration and secretory function of isolated perifused pancreatic islets. A: scheme of perifusion system for measurement of O2 consumption. 1) glass perifusion chamber, 2) peristaltic pump, 3) water bath (37°C) and perfusion buffer reservoirs, 4) tank with gas mixture (20% O2-5% CO2 balanced with N2), 5) gas exchanger (artificial lung), 6) fraction collector (Waters Division of Millipore), 7) excitation (524 nm) and emission (690 nm) light guides (at angle of 90° to each other) for measuring phosphorescence lifetimes in outflow, 8) computer. All transfer lines were insulated. O2 partial pressure Po2 was recorded by phosphorescence lifetimes of an O2-sensitive porphyrin [Pd-mesotetra-(4-carboxyphenyl)porphyrin dendrimer (26)]. Inflow O2 tension was measured in the absence of islets in the chamber before and after each experiment. This technique has outstanding sensitivity and unmatched stability. B: O2 consumption rate and insulin release (IR) in response to step changes in glucose concentration. Islets were subjected to step increases in glucose concentration (0 → 3 → 6 → 12 → 24 mM) and then switched back to 0 glucose. IR and glucose concentrations were measured in outflow samples and were correlated with measurements of O2 tension also in the effluent. Increases in glucose concentrations led to step increases in O2 consumption, reaching a plateau each time within ∼20 min. Insulin and glucose curves were corrected for the time that effluent needed to reach the fraction collector.

Measurement of O2 Consumption by Islets

O2 partial pressure Po2 was measured by the O2-dependent quenching of phosphorescence. The O2-sensitive phosphor used was Oxyphor R2, the second-generation glutamate dendrimer of Pd-meso tetra-(4-carboxyphenyl) porphyrin dendrimer, designed and synthesized by Vinogradov and coworkers (26, 48). The phosphorescence lifetime was measured using a PMOD 2000 (Oxygen Enterprises, Philadelphia, PA). This is a frequency domain phosphorometer (47) in which the excitation light (524 nm) is modulated at two frequencies simultaneously. The higher frequency was 18 kHz and the lower a variable frequency between 100 and 3,000 Hz. The lower frequency is determined by making a preliminary measurement and using the data to calculate the frequency that would give a phase shift of 28 ± 2° relative to the excitation light, using an algorithm developed for this purpose (47). This frequency is then used for the measurement of the phosphorescence lifetime and O2 pressure. The phosphorescence is collected with a light guide and returned to the phosphorometer, where it is filtered through a 665-nm long-pass glass filter (Schott glass) to remove the excitation light, and measured with an avalanche photodiode (Hammamatsu). The perifusion medium contains 1% bovine serum albumin and the O2 dependence of the phosphorescence lifetime for the albumin-bound Oxyphor R2 has been calibrated (14). These values were verified and used in the present study. The measured phosphorescence lifetime is converted to O2 pressure by the Stern-Volmer relationship t0/t = 1 + kQ × t0 × [Po2], where t0 is the lifetime in the absence of O2, t is the lifetime at the Po2, and kQ is a constant describing the frequency of quenching collision between the phosphor molecules in the triplet state and molecular O2 and the probability that a collision will result in quenching of phosphorescence.

The O2 tension in the perifusion medium as it enters the chamber with the islets was continuously measured for 1 h before the islets were added to the chamber. The O2 pressure value was remarkably stable, with an average value of 135–140 torr. The O2 tension in the outflow decreased by 15–25 torr when 500–700 islets were loaded into perfusion chamber. The response of the O2 detector placed in the outflow is delayed by the time required for the medium to flow from the islets to the measurement point. O2 consumption by the islets was calculated from the difference in Po2 between the influent and the effluent (O2 extraction) and the rate at which medium flowed through the chamber.

Ca2+ Measurement

Mouse islets, cultured for 3–4 days in 10 mM glucose, were loaded with fura 2-AM during a 40-min pretreatment at 37°C in 2 ml of Krebs buffer supplemented with 1 mmol/l fura 2-AM (Molecular Probes, Eugene, OR). The loaded islets were transferred to the perifusion chamber and placed on the homeothermic platform of an inverted Zeiss microscope (18). Islets were perfused with Krebs buffer at 37°C at a flow rate of 2 ml/min while various treatments were applied to the islets. The microscope was used with a ×40 oil immersion objective. The intracellular Ca2+ was determined by the ratio of the excitation of fura 2 at 334 and 380 nm. Emission was measured at 520 nm with an Attofluor charge-coupled device camera and calibrated using the Attofluor Ratio Vision Software.

Insulin Measurements

Insulin in the effluent was measured by radioimmunoassay with charcoal separation (23). Rat insulin from Linco Research served as standard and Miles anti-insulin antibody from ICN was the primary antibody.

RNA Isolation and Real-Time RT-PCR

Isolated islets obtained from a single mouse were homogenized in 1 ml of TRIzol reagent (Invitron). Glycogen (20 mM; Roche) was added to each sample as a carrier, followed by chloroform extraction and isopropanol precipitation. Having been washed with 70% ethanol, RNA pellets were resuspended in 300 μl of 10 mM Tris, pH 7.5, 1 mM EDTA, and 0.1% SDS. RNA was reextracted with 600 μl of phenol-chloroform-isoamyl alcohol (25:24:1, vol/vol/vol/) and was precipitated with 1/10 volume of 3 M sodium acetate and 3 volumes of ethanol. RNA was quantified with the RNA 6000 Nano Assay program of the Agilent 2100 Bioanalyzer (Agilent Technologies, Wilmington, DE), diluted with nuclease-free water, and stored at −80°C until use. Islet RNA was reverse transcribed using 1 μg of oligo(dT) primer and SuperScript II Reverse Transcriptase, and accompanying reagents (Invitrogen). PCR reaction mixes were assembled using the Brilliant SYBR Green QPCR Master Mix (Stratagene, La Jolla, CA). Reactions were performed using the SYBR Green program on an MX 4000 Quantitative PCR System (Stratagene). All reactions were performed in triplicate, and the median critical threshold (CT) value was used for analysis.

Statistical Analysis

The data are presented as means ± SE of four to seven experiments. In appropriate cases, significant differences between groups were determined by one-way analysis of variance with post hoc analysis using Dunnett's multiple comparison test. Values of P ≤ 0.05 were accepted as significant.

Experimental Design

The experiments were performed in three different ways. In the first set of experiments, glucose was used as a fuel stimulus; in the second, amino acids were used as a fuel stimulus; and in the third, amino acids served as fuel initially, and glucose was added later to assess the effect of the combination of stimuli.

In all cases, the effects of acetylcholine and FCCP were evaluated on insulin and glucagon release and the OCR in control and SUR1−/− islets. In separate experiments, changes in intracellular Ca2+ were measured using a design that corresponded to the hormone secretion studies and employing a dual-wavelength fluorescence microscope.


Model Experiment to Illustrate Continuous Measurement of Islet O2 Consumption by Phosphorescence

To test the performance of the newly designed system for optical respirometry, normal cultured islets were subjected to step increases in glucose concentration: from 0 → 3 → 6 → 12 → 24 mM and back to 0 glucose (Fig. 1B). Insulin and glucose were measured in outflow samples and were correlated with the O2 tension also measured in the effluent. The glucose staircase was established as predicted, with highly reproducible transition phases and plateau periods for each step. The insulin-secretory profile similarly developed as expected: 3 mM glucose was ineffective, 6 mM glucose caused a biphasic response, and further increases in glucose concentration elicited substantial monophasic increases in insulin secretion; and when glucose was switched off, the glucose profile and insulin secretion declined in parallel. At each substrate level change, the OCR reached a new plateau within ∼20 min. Low (3 mM) glucose caused a 36% increase in OCR, and 6, 12, and 24 mM glucose were associated with 28, 30, and 10% increases in OCR, respectively. This extrapolates to a hyperbolic glucose dependency curve of OCR. Unexpectedly, beginning at 6 mM glucose there was a time delay in changes of OCR compared with changes in glucose concentration and insulin secretion. This may suggest that glycolytic ATP or other glucose-derived coupling factors trigger insulin release before mitochondrial ATP production kicks in. This striking observation requires further investigation and will be the subject of another publication.

Studies with SUR1−/− Islets

Glucose as substrate.

Basal IR at 0 glucose was 7–8 times higher in SUR1−/− islets compared with control (CON) (Fig. 2A), but the OCR was comparable (Fig. 2B). The effect of high glucose (16.7 mM) on IR was very small (1.2-fold) in SUR1−/− islets, in contrast to the marked stimulation of CON (10- to 12-fold), resulting in comparable absolute rates of IR during the steady state at high glucose. High glucose enhanced the OCR of SUR1−/− and CON islets; however, the effect was significantly lower (one-third of CON) in SUR1−/− islets.

Fig. 2.

Effect of glucose and acetylcholine (Ach) on IR and O2 consumption rate (OCR) in control and sulfonylurea type 1 receptor-deficient (SUR1−/−) islets. A: IR of control and SUR1−/− islets during stimulation with glucose and Ach. B: OCR for the same experiments. Each curve in A and B represents the mean ± SE of 4 perifusions. All substrates were added to perifusate in the order given in the figure.

Acetylcholine (0.5 μM) in the presence of 16.7 mM glucose caused a large burst of IR in CON and SUR1−/− islets, with slightly higher rates in CON. In contrast to IR, acetylcholine caused only minor changes in OCR in normal and defective islets. The uncoupler of the respiration and oxidative phosphorylation FCCP (10 μM) blocked IR and transiently increased OCR in SUR1−/− and CON islets. The relative effect of FCCP was significantly higher in SUR1−/− islets compared with CON. NaN3 blocked respiration in both group of islets.

Amino acids alone.

When NaH2PO4 (Pi; 1 mM) was included in the perfusate, the mixture of 19 amino acids (8 mM) including 1 mM glutamine (AAM) significantly increased IR (∼6-fold) even in CON islets (Fig. 3A). Omitting Pi from the perfusate markedly reduced the effect of the AAM on IR (Fig. 3C) without significant changes in OCR (Fig. 3D). The absolute level of AAM-stimulated IR was higher in SUR1−/− islets compared with CON (Fig. 3A). This effect was associated with a marked increase in OCR (80%), which was comparable in CON and SUR1−/− islets (Fig. 3B).

Fig. 3.

Effect of amino acid mixture (AAM) and Ach on IR and OCR in control and SUR1−/− islets. A: IR. B: OCR for experiments shown in A. C: Pi is omitted from perfusate (○). D: OCR for experiments shown in C. Note: data presented in C and D are from a different set of experiments from those of A and B. Results are presented as means ± SE of 4 perifusions.

Acetylcholine caused a burst of IR (∼8-fold) in CON islets but only a small (8–10%) increase in OCR. The acetylcholine effect on insulin release was more than two times higher in CON compared with SUR1−/− islets.

AAM combined with glucose.

The amino acid effects on IR and OCR of this series of studies were similar to those of the previous set of experiments. The effect of high glucose (16.7 mM) on IR and OCR in the presence of the AAM was markedly reduced in SUR1−/− islets compared with CON (Fig. 4, A and B).

Fig. 4.

Combined effect of glucose and AAM on IR (A) and OCR (B) in control and SUR1−/− islets. Each curve represents the mean ± SE of 4 perifusions.

Acetylcholine in the presence of the AAM and high glucose further increased IR in CON and SUR1−/− islets to a greater extent in CON; however, OCR was increased only in SUR1−/− islets. FCCP caused a relatively bigger effect on OCR in SUR1−/− islets compared with CON, as was also seen in previous sets of experiments.

Changes in [Ca2+]i.

The baseline [Ca2+]i in SUR1−/− islets was twice as high as that of CON (Fig. 5, A and B). The AAM caused a transient increase in [Ca2+]i in SUR1−/− islets but not in CON. High glucose transiently decreased [Ca2+]i in SUR1−/− islets, in contrast to a burst increase in CON. Acetylcholine led to a sustained increase in [Ca2+]i in CON, in contrast to a transient effect in SUR1−/− (Fig. 5A). When amino acids alone served as fuel, acetylcholine transiently increased [Ca2+]i in CON and SUR1−/− islets, with a more sharply defined change in CON (Fig. 5B).

Fig. 5.

Changes in intracellular Ca2+ concentration ([Ca2+]i) in islets of control and SUR1−/− mice. Changes in [Ca2+]i were recorded using fura 2-AM. A: AAM initially served as fuel stimulant, and glucose was later added to observe the combined effect. Thin line, control islets; bold line, SUR1−/− islets. B: Ach effect on [Ca2+]i was tested in the presence of AAM only; high glucose was added subsequently after AAM and Ach. Typical traces are presented (n = 4).

Model Studies with Glyburide

Cell depolarization by glyburide decreases the effect of glucose on IR and OCR.

Because SUR1−/− islets are permanently depolarized, it is reasonable to speculate that alterations of glucose-stimulated IR and OCR may be due to a high basal [Ca2+]i level. To test this hypothesis, the SUR1 inhibitor glyburide was used to depolarize CON islets and increase [Ca2+]i. Glyburide slightly increased IR (Fig. 6A) and the OCR (Fig. 5B) and doubled [Ca2+]i (Fig. 6C) in glucose-free medium. Under these conditions, amino acids stimulated IR and the OCR and induced a transient increase in [Ca2+]i, followed by a sustained phase of relatively low [Ca2+]i (Fig. 6). High glucose slightly increased IR to the level reported for SUR1−/− islets. OCR changes in control islets were biphasic: initially it decreased, followed by an increase in islet respiration, reaching a level substantially lower than that in untreated islets. At the same time, glucose transiently lowered [Ca2+]i, and this phase is apparently related to a small dip in OCR.

Fig. 6.

Effect of glyburide pretreatment on IR, O2 consumption, and [Ca2+]i in control islets. A: IR. B: OCR. Note: SUR1−/− and control (without glyburide) tracings are added for ease of comparison. C: changes in [Ca2+]i. Each curve represents the mean ± SE of 4 perifusions. For clarity, the SE is omitted from the control without glyburide and SUR1−/− tracings; it is shown in Fig. 4B. The Ca2+ tracing is a typical example (n = 3).

α-Cell function of SUR1−/− and glyburide-treated islets.

High glucose does not inhibit glucagon secretion in SUR1−/− islets. Figure 7 presents glucagon release data in CON (Fig. 7A) and SUR1−/− islets (Fig. 7B) perfused with the 8 mM AAM. To test the ability of high glucose to suppress glucagon release, 16.7 mM glucose was added after the addition of the AAM in one set of perifusions for both CON and SUR1−/− islets. AAM gradually increased glucagon release in the absence of glucose in CON, and acetylcholine substantially elevated hormone release further, which was blocked by FCCP and NaN3. In contrast, in the presence of 16.7 mM glucose, glucagon secretion was significantly suppressed when stimulated by the AAM or acetylcholine. In SUR1−/− islets, high glucose failed to inhibit glucagon secretion during stimulation with the AAM or acetylcholine (Fig. 7B). Figure 7C presents the comparison between CON and SUR1−/− islets for the same experimental conditions. It has been reported that the [Ca2+]i is elevated in pancreatic α-cells of SUR1−/− mice (20). Therefore, glyburide was used to test whether an elevation in [Ca2+]i could be the cause of the altered glucagon secretion in SUR1−/− islets. In the presence of 300 nM glyburide and 8 mM AAM, the suppressive effect of high glucose on glucagon secretion was prevented. In fact, the glucagon release profiles of glyburide-treated CON and SUR1−/− islets are virtually indistinguishable (compare Fig. 7C with 7D).

Fig. 7.

Glucagon release in control and SUR1−/− islets. A: effect of high glucose on AAM- and Ach-stimulated glucagon release in control islets. B: effect of high glucose on AAM- and Ach-stimulated glucagon release in SUR1−/− islets. C: comparison between control and SUR1−/− islets for the same experimental conditions. D: ○, to one set of control islets 300 nM of glyburide was added before additions of AAM and glucose; •, glyburide-untreated control islets. Each curve represents the means ± SE of 3 perifusions.

Gene Expression Analysis

mRNA expression of several enzymes involved in glucose metabolism are presented on Fig. 8. As expected, expression of SUR1 is completely blocked in SUR1−/− islets. In contrast, Kir6.2 is expressed in SUR1−/− islets, but according to literature data (40) it cannot be incorporated in cell membrane. Interestingly, the expression of acetylcholine esterase (AchE) is twofold upregulated. In addition, mRNA levels of the β-subunit of the voltage-gated shaker potassium channel and GLUT2 are significantly elevated in SUR1−/− islets. mRNA levels of glucokinase, phosphofruktokinase-2 (data not shown), liver pyruvate kinase, glyceraldehyde-3-phosphate dehydrogenase, pyruvate dehydrogenase kinase-1, α-ketoglutarate dehydrogene, cytochrome c oxidase, and uncoupling protein-2 were unchanged.

Fig. 8.

Gene expression analysis. mRNA levels of SUR1, K+ inward rectifier (Kir6.2), acetylcholine esterase (AchE), β-subunit of voltage-gated shaker potassium channel (Kcnab1), GLUT2, glucokinase (GK), liver pyruvate kinase (L-PK), glyceraldehyde-3-phosphate dehydrogenase (GADPH), pyruvate dehydrogenase kinase-1 (PDK-1), α-ketoglutarate dehydrogenase (α-KGDH), cytochrome c oxidase, and uncoupling protein-2 (UCP2) are presented for control and SUR1−/− islets. Results are presented as means ± SE of 4 experiments.


SUR1−/− Islets as Model To Study Amplifying Pathways

It is widely accepted that in normal β-cells glucose induces insulin secretion by activating triggering and amplifying pathways (9, 22, 31, 41). KATP channels are thought to play the critical role in the triggering pathway by coupling fuel metabolism to membrane electrical activity (1, 5, 36). It is believed that an amplifying pathway operates independently of the triggering pathway and functions to augment fuel-stimulated insulin secretion. Several experimental approaches have been employed to study the amplifying pathway. All of these require clamping [Ca2+]i at an elevated level, and this maneuver eliminates the KATP channel-dependent step. This can be achieved by one of three ways: 1) high KCl in the presence of diazoxide that holds KATP channels open, 2) with sulfonylurea drugs that keep KATP channels closed, and 3) and by knocking out the SUR1 or Kir6.2 genes.

Because nutrients in all of these cases can no longer modulate KATP channel activity via changes in adenine nucleotide levels, it has been suggested that any enhanced insulin release is explained as being mediated through an amplifying pathway or KATP-independent pathway (4, 22, 41). The nature of the metabolic messengers involved in this pathway is still hotly debated.

In pancreatic islets lacking the SUR1 receptor and as a result the KATP channel, the triggering pathway is persistently activated because islet cells are depolarized and [Ca2+]i concentrations are elevated. This, in effect, duplicates the physiological membrane depolarization caused by metabolic inhibition of the KATP channel in normal β-cells, except that the depolarization is permanent rather than intermittent. Under such conditions, high glucose had only a marginal or even no effect on insulin secretion (12, 25, 37). These data were corroborated by in vivo observations where glucose increased the plasma insulin concentrations only slightly (1.5-fold) during hyperglycemic clamp studies (37). However, Eliasson et al. (16) and Nenquin et al. (31) have recently reported that high glucose stimulates insulin secretion in SUR1−/− islets significantly. Nenquin et al. concluded that the KATP channel independent amplifying action of glucose on insulin release is fully operative in SUR1−/− mice. Recently, it has been reported (13) that, even in depolarized SUR1−/− islets, high glucose leads to an increase in the frequency of Ca2+-dependent action potentials, which indicates that in SUR1−/− islets glucose may affect both the triggering and amplifying pathways.

The present paper provides additional evidence that the glucose effects in SUR1−/− islets are considerably altered and emphasizes that neurohormonal regulation, in particular acetylcholine, plays the critical role in amplifying insulin release in SUR1−/− islets.

Dissociation of insulin release and oxygen consumption at basal conditions in SUR1−/− islets

The present paper confirmed the previously published observation (31) that the baseline of insulin release of cultured islets was significantly higher in SUR1−/− islets than in controls (note: islets were cultured in 10 mM glucose for 3–4 days). However, in freshly isolated islets (12) or in the perfused pancreas (37), basal insulin secretion in SUR1−/− islets was comparable to that of controls. The high rate of insulin secretion by SUR1−/− islets in low glucose or the total absence of glucose results from a continuous, unregulated influx of Ca2+, secondary to the lack of functional KATP channels (31). The present study similarly shows that the high baseline insulin secretion in SUR1−/− islets is associated with a more than twofold elevated [Ca2+]i level.

Despite the large differences in initial insulin secretion and [Ca2+]i, the baseline of the OCR was comparable in these two groups of islets. Dissociation between insulin release and oxygen consumption suggests that energy requirements for exocytosis are minor compared with other ATP-consuming reactions. NSF (N-ethylmaleimide-sensitive factor), which functions as a MgATPase is one of more than two dozen factors participating in exocytosis of hormones or transmitters, but its operation, although obligatory, appears to be a quantitatively minor cellular ATP consumer (6, 28, 39). However, it has been shown that the inhibition of mitochondrial metabolism by 10 μM FCCP, 5 μM oligomycin, or 10 μM rotenone, or simply lowering the temperature to 22°C, suppressed (90% or more) insulin secretion by SUR1−/− islets incubated in 1 mM glucose (31). These observations are reconcilable if one speculates that the maintenance of a critical P-potential is an absolute requirement for secretion.

Glucose-Stimulated Insulin Release and Oxygen Consumption in SUR1−/− Islets

The present results show that the reduced glucose effect at 16.7 mM on insulin release in SUR1−/− islets is associated with reduced glucose-stimulated oxygen consumption. In support of these data is the finding that the glucose-evoked hyperpolarization of the mitochondrial membrane potential (ΔΨ, reflecting ATP production) is significantly smaller in SUR1−/− compared with wildtype β-cells (13). These and our data may suggest that the reduced glucose responsiveness of SUR1−/− islets is in part due to impaired glucose catabolism or oxidative phosphorylation. However, mRNA expression analysis of several enzymes and proteins involved in glucose metabolism, including GLUT2, glucokinase, phosphofruktokinase-2, glyceraldehyde-3-phosphate dehydrogenase, pyruvate kinase, α-ketoglutarate dehydrogenase, cytochrome c oxidase, and uncoupling protein-2 were unchanged in SUR1−/− islets, except that GLUT2, which is abundant even in controls, is significantly elevated. Thus a defect in glucose catabolism is not likely the cause of the defect in glucose-stimulated oxygen consumption.

Although there is no literature about the percentage of β-cells in SUR1−/− islets, it seems that the decreased secretory response and oxygen consumption of SUR1−/− islets are not due to decreased β-cell count in islets of SUR1−/− mice, because the insulin content is comparable in wild-type and knockout mice (16).

Acetylcholine Amplifies Insulin Release in the Presence of Glucose

These and previous (12) results demonstrate that acetylcholine amplifies or potentiates glucose action very effectively. Acetylcholine leads to a burst of insulin release in both control and SUR1−/− islets. Very remarkably, and perhaps surprisingly, the large increase in hormone release showed little or no changes in the OCR, implying a low energy requirement for the process of exocytosis as discussed above. It is possible that acetylcholine exerts a direct effect on the exocytosis machinery by increasing the cAMP in β-cells (45). Kawasaki et al. (24) showed that the cAMP-binding protein cAMP-GEFII, also referred to as Epac2 (8), is a direct target of cAMP, thus regulating cAMP-dependent, PKA-independent exocytosis by interacting with Rim2 (a target of the small G protein Rab3) (32). This interpretation is supported by the observation that acetylcholine stimulates (>10-fold) the rate at which the readily releasable pool of granules is supplied for release (19).

Amino Acid-Stimulated Insulin Release and Oxygen Consumption in SUR1−/− Islets

A physiological mixture of amino acids stimulated insulin secretion in SUR1−/− islets markedly, in contrast to controls, which did not respond (25). In the present study, amino acids did, however, stimulate insulin secretion in both normal and SUR1−/− islets. The response of control islets to amino acids in the present study could be explained by the presence in perfusate of Pi, which activates glutaminase (17). Despite the differences in amino acid-stimulated insulin secretion, the enhancement of the OCR was similar in control and SUR1−/− islets.

The results also show that acetylcholine is a powerful regulator of amino acid-stimulated insulin secretion in pancreatic islets in the absence of glucose, and its effect again has a surprisingly small impact on energy metabolism and [Ca2+]i. The insulin response to acetylcholine was lower in SUR1−/− islets than in controls, suggesting that optimal glucose metabolism is essential for the full acetylcholine effect in these abnormal cells. An alteration of glucose metabolism in SUR1−/− islets was also evident when islets were stimulated with amino acids: both insulin release and oxygen consumption were significantly lower in SUR1−/− islets.

Elevated [Ca2+]i Contributes to Altered Glucose Responses in SUR1−/− Islets

Because SUR1−/− islets are constantly depolarized, one can speculate that chronically elevated [Ca2+]i may be the cause of decreased glucose responsiveness. Inhibition of KATP channels by glyburide increased the [Ca2+]i to a similar degree as that observed in SUR1−/− islets. Under such conditions, the glucose effects on IR, oxygen consumption, and [Ca2+]i were comparable to those seen in SUR1−/− islets. Subsequent stimulation with 16.7 mM glucose caused a rapid, transient decrease in [Ca2+]i and oxygen consumption followed by a rise to rates above the initial values as was also observed in SUR1−/− islets. These experiments suggest that elevated [Ca2+]i levels may contribute to the reduced glucose response of SUR1−/− islets. This interpretation is supported by the observation that persistent elevation of [Ca2+]i in normal islets downregulates calmodulin-dependent protein kinase II (21), an effector of Ca2+-induced exocytosis (15).

Voltage-Gated Potassium Channels

The results of the present paper point out that expression of voltage-gated potassium channels (Kv channels), which are involved in repolarization of excitable cells (27, 49), increased in SUR1−/− mice. An increased expression of these channels in SUR1−/− islets may serve as a compensatory mechanism to help maintain the membrane potential in cells lacking functional KATP channel. However, Straub et al. (40) reported no difference in the voltage dependencies of Kv channel in control human β-cells and those with loss-of-function mutation in SUR1. It is possible that upregulation of Kv channels in the SUR1−/− mouse prevents the excessive insulin secretion that was observed in humans with mutations in SUR1 or Kir6.2 genes.

Glucagon Secretion in SUR1−/− Islets

Because pancreatic α-cells are equipped with KATP channels of the same type as those in β-cells (7, 33, 42) deletion of SUR1 could affect glucagon secretion. For example, the cytoplasmic Ca2+ concentration in α-cells from SUR1−/− mice exhibited spontaneous oscillations in the absence of glucose (20). The average [Ca2+]i measured in the absence of glucose was significantly higher than in wild-type α-cells. As expected, diazoxide and tolbutamide were ineffective in the SUR1-deficient α-cells. In addition, glucose failed to produce membrane depolarization in α-cells isolated from SUR1−/− mice (20). Glucagon secretion in the absence of glucose was lower than that observed in wild-type islets and both glucose (0–20 mmol/l) and the sulfonylureas failed to inhibit glucagon secretion (20). Results of the present study using cultured SUR1−/− islets (rather than single cells) are in good agreement with the published reports (20). Gromada et al. (20) found that GLP-1 failed to suppress glucagon secretion in the SUR1−/− islets. The authors proposed that the same metabolic defects that lead to a reduced insulin secretion from the β-cells might be associated with failure of glucose to inhibit glucagon secretion from the α-cells. Our model experiments with glyburide suggest that elevated [Ca2+]i may be the cause of both altered insulin and glucagon secretion. Under normal conditions, in the fasting state, glucagon prevents the development of hypoglycemia by stimulating gluconeogenesis and glycogenolysis. Postprandially, glucose stimulates insulin and inhibits glucagon secretion, resulting in suppression of hepatic gluconeogenesis and glycogenolysis and facilitating hepatic glycogen synthesis. Our in vitro data suggest that regulation of hepatic glucose metabolism is altered in SUR1−/− mice because of the impaired glucose suppression of glucagon secretion. Impaired glucagon and insulin secretion is the hallmark of type 2 diabetes (10, 30). Circulating levels of glucagon are elevated in diabetes despite hyperglycemia; therefore, glucose is overproduced by the hepatocyte and is ineffectively metabolized by other organs, and a further rise in blood glucose fails to inhibit or may even, paradoxically, stimulate its release.


The current study demonstrates that glucose stimulation of insulin release is virtually absent and that the stimulation of the OCR by glucose is drastically reduced in SUR1−/− islets from adult donor mice. Acetylcholine restores insulin secretion to near control rates but is probably not effective enough to cause hypoglycemia. Although upregulation of the parasympathetic signaling pathways and an increased sensitivity of β-cells to acetylcholine contribute to hyperinsulinemia in several animal models (2, 3, 11, 38), it is not known whether acetylcholine causes hyperinsulinemia in humans with mutations in SUR1. It is possible that, in contrast to humans, the relatively high sympathetic/cholinergic index in the mouse restrains excessive insulin secretion in the SUR1−/− mouse. In support of this explanation, SUR1−/− mice secret two- to threefold more catecholamines (measured in urine samples) than controls (46). Catecholamines may thus contribute to the maintenance of normoglycemia in SUR1−/− mice.


This work was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grant DK-22122 (to F. M. Matchinsky), by a grant from the American Diabetes Association (to F. M. Matchinsky), by the Institute for Diabetes, Obesity and Metabolism (via NIDDK Grant DK-19525), and by a grant-in-aid from Novo Nordisk.


We thank Markiyan Doliba for outstanding technical assistance.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract