Search

This review has focused on the properties of the ATP-sensitive K-channel found in cardiac and skeletal muscle, and in pancreatic beta-cells. It is conceivable that this channel will be found in other cell types. In particular, it would be worthwhile looking for its presence in those cells in which electrical activity is linked to metabolism, glucose concentration, or oxygen levels. Obvious examples are the glucoreceptor neurons of mammalian brain and chemoreceptors such as those of the carotid body. While ATP-sensitive K-channels in cardiac and skeletal muscle membranes are rather similar, there are a few significant differences between these channels and that found in the beta-cell. Most notably, the latter is more sensitive to inhibition by ATP and sulphonylureas. It remains to be seen whether they also differ in the ability of nucleotides to activate the channel. Considerable confusion also still surrounds the physiological regulation of the ATP-sensitive K-channel in intact cells. Although the general consensus seems to be that [ATP]i modulates channel activity, the role of other nucleotides and ions as well as the way in which their concentrations alter with metabolism requires further elucidation. A combined electrophysiological and biochemical approach is likely to prove most successful in establishing which second messenger systems contribute to the physiological regulation of the ATP-sensitive K-channel. Finally, the close correlation between cell metabolism and the activity of the ATP-sensitive K-channel raises the intriguing possibility that disorders of cell metabolism might produce alterations in channel activity and consequent changes in cell function.

This chapter explains inward rectification in skeletal muscle fibers. It shows that the K-conductance depends primarily on V, the membrane potential and [K] o rather than on (V–V K ). Further, it is shown that the inward rectifier is readily permeable to K + and to T1 + , but that T1 + appears to unmask a gating mechanism which causes inward currents to inactivate when they are carried by T1 + . Alkali metal ions other than K + and the alkali earth metal ions block the inward movement of K + , though Rb + and to a lesser extent Cs + , are weakly permeant. The steep voltage-dependence of the blocks of Rb + , Cs + and Na + suggest a pore which is capable of being occupied by more than one ion. The chapter shows the result of Hodgkin & Horowicz in the context of voltage clamp experiments. Current–voltage relations are plotted for muscle fibers immersed in solutions containing K + at between 10 and 120 mM.

1. Inactivation of Ca currents in skeletal muscle fibres of the stick insect, Carausius morosus, was studied using a three-electrode voltage-clamp method. 2. The extent of inactivation showed a voltage-dependence similar to that of the Ca current, inactivation being absent in the absence of a Ca current, maximal at potentials where Ca currents are largest, and reduced at potentials close to ECa. 3. Ca currents inactivated along a double exponential time course, both when measured from the decline of Ca current during a single pulse and when measured using a two pulse protocol. In 20 mM-Ca-Ringer the fast time constant of inactivation had a mean value of 27 msec and that of the slow time constant was 134 msec, at O mV and 5 degrees C. 4. The rate of inactivation was slowed, and its extent reduced, in low [Ca]o, where Ca currents are smaller. Similarly, inactivation was faster and more complete in high-Ca-Ringer. 5. The rate of recovery from inactivation also followed a double exponential time course, with time constants of 638 msec and 4 sec following a 500 msec inactivating pulse in 20 mM-Ca-Ringer at 5 degrees C. Recovery appeared to be related to the amount of Ca entry during the inactivating pulse, being slower in high [Ca]o and following longer inactivating pulses. 6. Inactivation was slowed and reduced in extent when Ba2+ or Sr2+ carried current. Inactivation in Ba solutions may be due to depletion of Ba2+ from the lumen of the transverse tubules. 7. Ba2+ does not compete with Ca2+ for the inactivation mechanism. 8. It is concluded that inactivation of Ca currents in stick insect muscle fibres is primarily mediated by Ca2+ entry.

A new method for rapidly changing the solution superfusing excised membrane patches is described. The main advantage of this method over previous rapid exchange methods is that it is possible to use very small quantities of solution (minimal volume about 100 microliters). The method consists of inserting the pipette tip, after obtaining an excised patch, into a small tube which can be independently perfused. Air bubbles are used to separate different solutions and to produce rapid "step" changes of solution. To test this method we examined the rate of change of: the pipette resistance produced in response to changing to a solution of lower conductivity, the amplitude of single K-channel currents in response to changing external K. The results suggest that the solution exchange is essentially a step function and is complete within 10-20 ms. The recording is briefly interrupted during the time taken for the passage of the air bubble past the pipette tip (less than 250 ms). In general, the passage of the air bubble did not affect the seal resistance. This method will be particularly useful for studies of the interactions of single ionic channels with chemicals which can only be obtained in small quantities.

The non-penetrating potentiometric dyes NK2367 and WW375 were used to investigate the effect of inward rectification on the weighted-average tubular membrane potential in single frog muscle fibres voltage clamped using a three-Vaseline-gap method. In 100 mM-K solution, when inward rectification was activated by hyperpolarization the steady-state amplitude of the transverse tubular system (T-system) optical signal was reduced, and its rise time was faster than that recorded for an equivalent depolarization. The voltage dependence of the optical attenuation followed that of inward rectification, increasing with increasing hyperpolarization. For a voltage-clamp step of -140 mV the optical attenuation was 0.72 which corresponds to a weighted-average T-system potential change of 100 mV. When inward rectification was blocked in a Cs, TEA solution the optical attenuation was also abolished. The voltage dependence of the block of the inward currents in solutions containing low concentrations of Cs was also reflected in the T-system optical signals. Our results were satisfactorily predicted by a radial cable model of the T-system, assuming the same specific inward rectifier conductance in surface and tubular membranes. This analysis predicts that the measured optical attenuation corresponds to a decrease in the tubular space constant, lambda T, from 120 micron under passive conditions to about 40 micron when inward rectification is fully, activated. The voltage dependence of inward rectification measured at the surface membrane was reasonably well predicted by assuming that the specific conductance obeyed a Boltzmann type of voltage dependence; the major effect of tubular decrements was to reduce the steepness of the total (surface + T-system) conductance-voltage relation.

The resting potential of many excitable cells, including skeletal muscle1–4, cardiac muscle5,6, nerve cell bodies7 and egg cells8, is determined by a resting potassium conductance which shows inward rectification, allowing potassium ions to move more readily inward across the cell membrane than outward. In skeletal muscle, where inward rectification has been extensively studied, a large part of this conductance is located in the T-system membranes2,3,9–11. However, to date, only the kinetic and voltage-dependent properties of this conductance have been studied from analyses of the membrane potential or current recorded at the fibre surface. We report here measurements, obtained using a voltage-sensing dye, of potential changes in the T-system membranes associated with the inwardly rectifying K+ current. Our results show that this conductance alters the time course and significantly attenuates the amplitude of the potential change across the tubular membranes. These optical data provide new evidence for the presence of this conductance in the T-system and, when analysed using a radial cable model for the T-system, provide an estimate of the distribution of the inward rectifier conductance over the surface and T-system which is in agreement with estimates obtained by other techniques.

It is well established that glucose-stimulated insulin release from the pancreatic β-cell is associated with the initiation of electrical activity (see recent review by Henquin & Meissner10). Microelectrode recordings have shown that in the absence of glucose the β-cell is electrically silent. Increasing plasma glucose depolarises the membrane by an amount that is dependent on the glucose concentration and electrical activity is initiated when this depolarisation exceeds a threshold level. There is considerable evidence that the initial slow depolarisation induced by glucose results from a decrease in the resting potassium permeability of the membrane5,8,14 as a consequence of glucose metabolism9).

In excitable cells, ions permeate the cell membrane through ionic channels, some of which open and close in response to changes in the potential difference across the membrane. It has been supposed that this opening and closing (or gating) process is largely independent of the permeating ion. However, we show here that the gating of the resting potassium permeability of frog skeletal muscle depends on the species of ion which carries current across the membrane. The potassium permeability investigated allows K+ to move in across the membrane more easily than out. This property is known as inward or anomalous rectification and is shared by cell membranes of skeletal muscle, egg and certain other cells. In both egg cells and skeletal muscle fibres, the group IIIB metal ion Tl+, which can replace K+ in several other systems in experimental conditions, also permeates the inward rectifier. Indeed, Tl+ is more permeant than K+ (refs 8, 9). However, when Tl+ carries current inwards across the membrane, the inward rectifier inactivates over a brief period when the membrane is hyperpolarized, whereas when K+ carries current, the permeability increases with time under hyperpolarization.

1. A three electrode voltage-clamp was used to investigate membrane currents in the skeletal muscle fibres of the stick insect, Carausius morosus. Contraction was blocked by hypertonic solutions. 2. Membrane currents elicited by step depolarizations consisted of an inward current, an early outward current and a delayed outward current. 3. The reversal potential of the delayed outward current did not change when SO4(2-) was substituted for Cl-, but shifted by 14.1 mV when [K]0 was increased from 20 mM to 40 mM in SO4(2-) solution, suggesting that the delayed current is carried by K+. Both early and delayed outward currents were substantially reduced by 120 mM-tetraethylammonium (TEA) ions. 4. The small size of the shift in the reversal potential of the delayed outward current with increased pulse duration suggests that the delayed current measured flows mainly across the surface membrane. 5. Increasing [Ca]o made the apparent reversal potential for the inward current (120 mM-TEA Ringer) more positive and increased the size of the maximum inward current. However, Ca-currents showed saturation with increasing [Ca]o, indicating that there is a site to which Ca ions bind during their passage through the membrane. The dissociation constant of this site was 7.3 mM at 0 mV and was voltage-dependent. 6. Inward currents were blocked by 1 mM-La3+ or Cd2+, or by substitution of Co2+ or Ni2+ for Mg2+. Strontium and barium were able to permeate the channel but Na+ and Mg2+ appear impermeant. 7. As expected from the low intracellular Ca concentration, the instantaneous current-voltage relation of the Ca current rectified strongly in the inward direction. 8. Both constant field theory and the simplest, single site, Eyring rate theory model predict the rectification of the instantaneous current-voltage relation. The rate theory model also predicts saturation of the Ca current with [Ca]o.

ATP-sensitive K-channels in the cloned beta-cell line HIT T15 were studied by patch-clamp methods; by measurement of 86Rb efflux; and by [3H]glibenclamide binding to isolated membrane preparations. In inside-out patches a 50 pS K-channel was found which was blocked by ATP or tolbutamide applied to the intracellular membrane surface. A minimum estimate of about 500 channels per beta-cell was obtained by combining whole-cell and single-channel data. The rate of efflux of 86Rb from 86RbCl-loaded HIT cells was markedly increased by intracellular ATP-depletion; 86Rb-efflux was progressively inhibited by increasing concentrations of glibenclamide or tolbutamide. In non-ATP-depleted cells, diazoxide elicited a concentration-dependent stimulation of 86Rb-efflux which was completely blocked by 1 microM glibenclamide. Isolated membranes showed dose-dependent saturable binding of [3H]glibenclamide to both high (Kd = 1.12 nM) and low (Kd = 136 nM) affinity binding sites. We estimate about 5000 high-affinity binding sites per cell. [3H]-glibenclamide binding was inhibited by tolbutamide (IC50 = 125 microM) but was not affected by diazoxide. ADP (0.5 or 1.0 mM) markedly reduced binding; other nucleotides tested were ineffective.

1. The effects of the insulin secretagogue 2-ketoisocaproate, and of arginine, on insulin release, intracellular adenosine triphosphate (ATP) concentration and the activity of single K channels in cell-attached membrane patches have been studied in primary cultures of beta-cells from adult rat islets of Langerhans. 2. Insulin secretion was significantly increased by 2-ketoisocaproate (20 mM). The time course of this release was biphasic. Arginine (20 mM) did not stimulate insulin secretion. 3. In the absence of 2-ketoisocaproate (or arginine), two kinds of K channel were regularly observed in cell-attached membrane patches held at the cell resting potential: a channel of approximately 20 pS conductance and a channel of 50 pS conductance ([K]o = 140 mM, room temperature). 4. Addition of 2-ketoisocaproate (20 mM) to the bath suppressed the activity of the 50 pS channel and initiated action potentials. Arginine (20 mM) was without effect. 5. The intracellular concentration of ATP was increased significantly by 2-ketoisocaproate (20 mM) but not by arginine (20 mM). 6. It is suggested that, like glucose, 2-ketoisocaproate depolarizes the beta-cell and mediates insulin secretion by reducing the open probability of the 50 pS K channel. The results are also consistent with the idea that both secretagogues inhibit this channel by increasing the cytoplasmic concentration of ATP.

We have studied the effects of changes of intracellular ATP concentration ([ATP]i) on the activity of ATP-sensitive K-channels (IK(ATP] and of Na,K-ATPase in intact cells of the insulin-secreting cell-line HIT-T15. Pre-exposure of HIT beta-cells to oligomycin caused a dose-dependent reduction in [ATP]i. Marked activation of IK(ATP) activity was found when ATP was lowered below 3 mM. Na,K-ATPase was progressively inhibited as ATP was lowered to 1.5 mM. These data demonstrate that changes in intracellular ATP in the millimolar range markedly influence the activity of two beta-cell membrane proteins having affinities for ATP in the micromolar range. This suggests that submembrane [ATP] may be considerably below the measured bulk cytosolic concentration. The findings also support the proposed role of intracellular ATP in mediating effects of changes in glucose concentration on the activity of beta-cell IK(ATP) and insulin secretion.

β-Cell failure in type 2 diabetes (T2D) was recently proposed to involve dedifferentiation of β-cells and ectopic expression of other islet hormones, including somatostatin and glucagon. Here we show that gastrin, a stomach hormone typically expressed in the pancreas only during embryogenesis, is expressed in islets of diabetic rodents and humans with T2D. Although gastrin in mice is expressed in insulin+ cells, gastrin expression in humans with T2D occurs in both insulin+ and somatostatin+ cells. Genetic lineage tracing in mice indicates that gastrin expression is turned on in a subset of differentiated β-cells after exposure to severe hyperglycemia. Gastrin expression in adult β-cells does not involve the endocrine progenitor cell regulator neurogenin3 but requires membrane depolarization, calcium influx, and calcineurin signaling. In vivo and in vitro experiments show that gastrin expression is rapidly eliminated upon exposure of β-cells to normal glucose levels. These results reveal the fetal hormone gastrin as a novel marker for reversible human β-cell reprogramming in diabetes.

Background To investigate disease progression the first 12 months after diagnosis in children with type 1 diabetes negative (AAB negative) for pancreatic autoantibodies [islet cell autoantibodies(ICA), glutamic acid decarboxylase antibodies (GADA) and insulinoma-associated antigen-2 antibodies (IA-2A)]. Furthermore the study aimed at determining whether mutations in KCNJ11, ABCC8, HNF1A, HNF4A or INS are common in AAB negative diabetes. Materials and methods In 261 newly diagnosed children with type 1 diabetes, we measured residual β-cell function, ICA, GADA, and IA-2A at 1, 6 and 12 months after diagnosis. The genes KCNJ11, ABCC8, HNF1A, HNF4A and INS were sequenced in subjects AAB negative at diagnosis. We expressed recombinant K-ATP channels in Xenopus oocytes to analyse the functional effects of an ABCC8 mutation. Results Twenty-four patients (9.1%) tested AAB negative after one month. Patients, who were AAB-negative throughout the 12-month period, had higher residual β-cell function (P = 0.002), lower blood glucose (P = 0.004), received less insulin (P = 0.05) and had lower HbA1c (P = 0.02) 12 months after diagnosis. One patient had a heterozygous mutation leading to the substitution of arginine at residue 1530 of SUR1 (ABCC8) by cysteine. Functional analyses of recombinant K-ATP channels showed that R1530C markedly reduced the sensitivity of the K-ATP channel to inhibition by MgATP. Morover, the channel was highly sensitive to sulphonylureas. However, there was no effect of sulfonylurea treatment after four weeks on 1.0-1.2 mg/kg/24 h glibenclamide. Conclusion GAD, IA-2A, and ICA negative children with new onset type 1 diabetes have slower disease progression as assessed by residual beta-cell function and improved glycemic control 12 months after diagnosis. One out of 24 had a mutation in ABCC8, suggesting that screening of ABCC8 should be considered in patients with AAB negative type 1 diabetes.

Barium currents flowing through single Ca2+ channels were recorded from outside-out patches isolated from mouse pancreatic B-cells. Only one type of Ca2+ channels was observed. In 110 mM Ba2+, the single channel conductance was 24pS (at negative membrane potentials) and the current amplitude at 0 mV was -0.7 pA. Channel openings were activated by depolarisations more positive than -30 mV and showed little inactivation during 200 ms pulses. Open times were increased by BAY K 8644 an decreased by micromolar Cd2+. Channel activity was subject to rundown in excised patches and little activity remained after 10 min. These properties resemble those of L-type Ca2+ channels in other tissues. It is suggested that this Ca2+ channel participates in the generation of the B-cell action potential and mediates the increase in Ca2+ influx required for insulin secretion.

The physiological activity of hypothalamic arcuate nucleus (ARC) neurons is critical for dynamic maintenance of body energy homeostasis, and its malfunction can result in common metabolic disorders, such as obesity. It is therefore of interest to determine which set of ion channels shapes electrical activity in the ARC. Whole-cell patch clamp of ARC neurons in mouse brain slices identified three electrophysiologically distinct types of neurons. These were distinguished by their rebound "signatures" after hyperpolarizing current injection in current clamp and by the presence of transient inward (Type-B neurons) or outward (Type-A and Type-C neurons) subthreshold voltage-gated currents in voltage-clamp recordings. In turn, the transient outward current (A-current) of Type-C neurons had a lower activation threshold and different time and voltage dependence of inactivation than that of Type-A neurons. The brain-gut peptide cholecystokinin (CCK) has long been recognized to control food intake, but how endogenous CCK modulates the activity of central appetite-regulating networks remains unresolved. Here, we show that low (picomolar) concentrations of CCK rapidly and reversibly slow the firing of ARC Type-C neurons. This effect is mediated by postsynaptic CCK-B receptors and is attributable to potentiation of the A-current. Our study thus identifies several fundamental biophysical mechanisms underlying the physiological activity of ARC neurons and suggests a novel mechanism by which endogenous CCK may control appetite.

We have used the whole-cell recording technique to determine whether ATP-sensitive potassium (K ATP ) currents, voltage-dependent Ca 2+ currents, and exocytosis are different in single β-cells from pancreatic islets of Goto-Kakizaki (GK) rats, a novel model of NIDDM, and normal rats. In addition, we have also measured the insulin secretory responses, ATP content, and the rate of glucose metabolism in intact islets. Although the glucose sensitivity of the K ATP current was similar between GK rats and controls, in the absence of glucose, K ATP current density was larger in GK rats, which resulted in a more hyperpolarized membrane potential. Whole-cell Ca 2+ currents were similar. By monitoring the cell capacitance with a fixed intracellular solution, no difference was detected in the exocytotic responses of β-cells from normal and GK rats. In islets from GK rats, the rates of glucose utilization ([ 3 H]H 2 O production from 5-[ 3 H]glucose) and oxidation ([ 14 C]CO 2 production from U-[ 14 C]glucose) were not significantly different from controls. Insulin secretion, however, was impaired (by 50%), and this was paralleled by a smaller increase in ATP content in response to stimulation by 10 mmol/1 glucose in islets from GK rats when compared with controls. Under conditions in which K ATP channels were held open and the effects of glucose were independent of membrane potential, insulin release was still significantly lower in GK rat islets than in controls. These findings suggest that the impaired insulin secretion in islets from GK rats does not simply result from a failure to close K ATP channels, nor does it result from an impairment in calcium secretion coupling.

The single-channel conductance varies significantly between different members of the inward rectifier (Kir) family of potassium channels. Mutations at three sites in Kir6.2 have been shown to produce channels with reduced single-channel conductance, the largest reduction (to 40% of wild-type) being for V127T. We have used homology modeling (based on a KcsA template) combined with molecular dynamics simulations in a phosphatidycholine bilayer to explore whether changes in structural dynamics of the filter were induced by three such mutations: V127T, M137C, and G135F. Overall, 12 simulations of Kir6.2 models, corresponding to a total simulation time of 27 ns, have been performed. In these simulations we focused on distortions of the selectivity filter, and on the presence/absence of water molecules lying behind the filter, which form interactions with the filter and the remainder of the protein. Relative to the wild-type simulation, the V127T mutant showed significant distortion of the filter such that approximately 50% of the simulation time was spent in a closed conformation. While in this conformation, translocation of K(+) ions between sites S1 and S2 was blocked. The distorted filter conformation resembles that of the bacterial channel KcsA when crystallized in the presence of a low [K(+)]. This suggests filter distortion may be a possible general model for determining the conductance of K channels.

Diabetes is a well-established risk factor for heart disease, leading to impaired cardiac function and a metabolic switch toward fatty acid usage. In this study, we investigated if hyperglycemia/hypoinsulinemia in the absence of dyslipidemia is sufficient to drive these changes and if they can be reversed by restoring euglycemia. Using the βV59M mouse model, in which diabetes can be rapidly induced and reversed, we show that stroke volume and cardiac output were reduced within 2 weeks of diabetes induction. Flux through pyruvate dehydrogenase was decreased, as measured in vivo by hyperpolarized [1-13C]pyruvate MRS. Metabolomics showed accumulation of pyruvate, lactate, alanine, tricarboxyclic acid cycle metabolites, and branched-chain amino acids. Myristic and palmitoleic acid were decreased. Proteomics revealed proteins involved in fatty acid metabolism were increased, whereas those involved in glucose metabolism decreased. Western blotting showed enhanced pyruvate dehydrogenase kinase 4 (PDK4) and uncoupling protein 3 (UCP3) expression. Elevated PDK4 and UCP3 and reduced pyruvate usage were present 24 h after diabetes induction. The observed effects were independent of dyslipidemia, as mice showed no evidence of elevated serum triglycerides or lipid accumulation in peripheral organs (including the heart). The effects of diabetes were reversible, as glibenclamide therapy restored euglycemia, cardiac metabolism and function, and PDK4/UCP3 levels.

1. To investigate the mechanisms regulating the reuptake of secretory granule membranes following regulated exocytosis, we have monitored changes in cell capacitance in single pancreatic beta-cells. 2. Membrane retrieval (endocytosis) occurred both in a continuous manner and in abrupt steps, corresponding to the simultaneous retrieval of 50-100 granules. The large endocytotic steps were associated with a conductance change of about 1 nS which we attribute to the formation of a fission pore with a pore radius of approximately 1 nm. 3. In some cells, we observed large amplitude capacitance fluctuations, suggesting that aggregates of granules are connected to the plasma membrane by a single pore and are subsequently retrieved as a single unit. 4. Endocytosis was evoked by elevation of cytosolic [Ca2+]i, but once initiated, a sustained increase in [Ca2+]i was not required for endocytosis to continue. 5. The [Ca2+]i dependence of exo- and endocytosis was studied by photorelease of Ca2+ from the 'caged' precursor Ca(2+)-nitrophenyl-EGTA (Ca(2+)-NP-EGTA). Both exo- and endocytosis were initiated at between 0.5 and 2 microM Cai(2+). The rate of endocytosis saturated above 2 microM Cai(2+), whereas exocytosis continued to increase up to 4 microM Cai(2+). The maximum rate of endocytosis was < 25% of that of exocytosis. 6. Unlike exocytosis, endocytosis proceeded equally well in the presence or absence of Mg-ATP. 7. Our data indicate that in the pancreatic beta-cell, exocytosis and endocytosis are regulated by different mechanisms.

Long-term (72 hr) exposure of pancreatic islets to palmitate inhibited glucose-induced insulin secretion by >50% with first- and second-phase secretion being equally suppressed. This inhibition correlated with the selective impairment of exocytosis evoked by brief (action potential-like) depolarizations, whereas that evoked by long ( approximately 250 ms) stimuli was unaffected. Under normal conditions, Ca(2+) influx elicited by brief membrane depolarizations increases [Ca(2+)](i) to high levels within discrete microdomains and triggers the exocytosis of closely associated insulin granules. We found that these domains of localized Ca(2+) entry become dispersed by long-term (72 hr), but not by acute (2 hr), exposure to palmitate. Importantly, the release competence of the granules was not affected by palmitate. Thus, the location rather than the magnitude of the Ca(2+) increase determines its capacity to evoke exocytosis. In both mouse and human islets, the palmitate-induced secretion defect was reversed when the beta cell action potential was pharmacologically prolonged.

(Cell Metabolism 3, 35–45, January 2006)This article has been retracted: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy).This paper is being retracted because of concerns with Figures 2B and S3. In Figure S3, the SDS-PAGE gel image had molecular weight markers added from another autoradiograph and additional nonspecific bands obscured or removed. In the top panel of Figure 2, an actin Western blot was used instead of a GCK Western blot. This matter was investigated by the Medical Research Council (UK) under their Scientific Misconduct Policy and Procedures, and scientific misconduct pertaining to the issues described here against the first author Helen Freeman was found. Repetition of these experiments by other group members gave identical results to those shown in the paper. Nevertheless, although we feel that the scientific conclusions of the paper (that loss of NNT impairs insulin secretion and impairs glucose tolerance in C57BL/6J mice) still stand, given the conclusion of scientific misconduct against the first author, the authors think the most responsible course is to retract the paper. The authors sincerely apologize for the inconvenience caused by this retraction.