Free radical-mediated tolbutamide desensitization of K+ATP channels in rat pancreatic beta-cells.

To study the effects of hydroxyl radicals on the sensitivity of the ATP-sensitive K+ (K+ ATP) channel to tolbutamide, we used patch clamp and microfluorometric techniques in pancreatic beta-cells isolated from rats. cell-attached membrane patches, exposure of the cells to 0.3 mM H2O2 increased the probability of opening of K+ATP channels in the presence of 2.8 mM glucose. Tolbutamide dose-dependently inhibited the K+ATP channel with half-maximal inhibition (IC50) at 0.8 microM before and immediately after exposure to H2O2. After prolonged exposure (>20 min) to H2O2, the IC50 was increased to 15 microM. The presence of both ATP and ADP at concentrations ranging from 0.01 to 0.1 mM in the inside-out bath solution significantly enhanced the inhibition of the channels by 10 microM tolbutamide. Addition of 0.3 mM H2O2 induced a transient minute increase in the cytoplasmic Ca2+ concentration ([Ca2+]i) within 10 min, followed by a sustained pronounced increase in [Ca2]i. After more than 20 min of exposure of cells to 0.3mM H2O2, [Ca2]i was increased to above 2 microM. Treatment of the cytoplasmic face of inside-out membrane patches with 1 microM Ca2+ attenuated the tolbutamide-sensitivity of the K+ATP channel, but not the ATP-sensitivity of the channel. These findings indicate that H2O2 reduces tolbutamide sensitivity by inducing a sustained increase in [Ca2+]i.

Of mice and men: K(ATP) channels and insulin secretion.

K(ATP) channels are a unique, small family of potassium (K+)-selective ion channels assembled from four inward rectifier pore-forming subunits, K(IR)6.x, paired with four sulfonylurea receptors (SURs), members of the adenosine triphosphate (ATP)-binding cassette superfamily. The activity of these channels can be regulated by metabolically driven changes in the ratio of adenosine diphosphate (ADP) to ATP, providing a means to couple membrane electrical activity with metabolism. In pancreatic beta cells in the islets of Langerhans, K(ATP) channels are part of an ionic mechanism that couples glucose metabolism to insulin secretion. This chapter 1) briefly describes the properties of K(ATP) channels; 2) discusses data on a genetically recessive form of persistent hyperinsulinemic hypoglycemia of infancy (PHHI), caused by loss of beta-cell K(ATP) channel activity; and 3) compares the severe impairment of glucose homeostasis that characterizes the human phenotype with the near-normal phenotype observed in K(ATP) channel null mice.

PIP2 and ATP cooperatively prevent cytosolic Ca2+-induced modification of ATP-sensitive K+ channels in rat pancreatic beta-cells.

The factors that influence functional coupling between the sulfonylurea receptor (SUR1) and Kir6.2 subunits of ATP-sensitive K+ (K+(ATP)) channels were studied in rat pancreatic beta-cells using patch clamp and microfluorometric techniques. Tolbutamide at 10 micromol/l inhibited K+(ATP) channels in association with occurrence of action currents, but further exposure of beta-cells to the drug for 30 min or longer resulted in reappearance of K+(ATP) channel events. Half-maximal inhibition concentration (IC50) for tolbutamide was 1.5 microl/mol in 2.8 mmol/l glucose, and it was increased to 13.3 micromol/l when the cellular metabolism was inhibited by 0.5 mmol/l 2,4-dinitrophenol (DNP) for 5 min. Tolbutamide at 10 micromol/l induced an increase in cytosolic Ca2+ concentration ([Ca2+]i), and its amplitude was markedly reduced following exposure to 0.5 mmol/l DNP or long-term (30 min) exposure to 10 micromol/l tolbutamide. This tolbutamide insensitivity, as assessed by the [Ca2+]i response, was not observed when the external Ca2+ was omitted during the long-term exposure to tolbutamide. In cell-attached membrane patches, the tolbutamide insensitivity was also produced by treatment of cells with 150 micromol/l diazoxide and 25 mmol/l KCl in the presence, but not absence, of 2 mmol/l Ca2+ in the external solution. When the cytoplasmic face of inside-out membrane patches was treated with higher Ca2+ concentrations (2 micromol/l), both ADP-evoked activation and tolbutamide-induced inhibition of K+ ATP channels were attenuated with retaining ATP-induced inhibition, indicating the modification of K+(ATP) channels. The Ca2+-induced channel modification was prevented partially by phosphatidylinositol 4,5-bisphosphate (PIP2) and completely by ATP and PIP2 together, but not by ATP alone. Treatment of the channel with cytochalasin D, a disrupter of F-actin, evoked channel modification similar to that induced by Ca2+. The modification was prevented completely by phalloidin, a stabilizer of F-actin. In conclusion, long-term exposure to tolbutamide or metabolic inhibition causes modification of K+ ATP channels via mechanisms involving Ca2+-dependent reaction. The modification, which may reflect functional disconnection between SUR1 and Kir6.2, is prevented by ATP and PIP2, which may act cooperatively to stabilize membrane cytoskeletons (F-actin structures).

Diverse effects of hydrogen peroxide on cytosolic Ca2+ homeostasis in rat pancreatic beta-cells.

Oxygen-free radicals are thought to be a major cause of beta-cell dysfunction in diabetic animals induced by alloxan or streptozotocin. We evaluated the effect of H2O2 on cytosolic Ca2+ concentration ([Ca2+]i) and the activity of ATP-sensitive potassium (K+ATP) channels in isolated rat pancreatic beta-cells using microfluorometry and patch clamp techniques. Exposure to 0.1 mM H2O2 in the presence of 2.8 mM glucose increased [Ca2+]i from 114.3+/-15.4 nM to 531.1+/-71.9 nM (n=6) and also increased frequency of K+ATP channel openings. The intensity of NAD(P)H autofluorescence was conversely reduced, suggesting that H2O2 inhibited the cellular metabolism. These three types of cellular parameters were reversed to the control level on washout of H2O2, followed by a transient increase in [Ca2+]i, the transient inhibition of K+ATP channels associated with action currents and increase of the NAD(P)H intensity with an overshoot. In the absence of external Ca2+, 0.1 mM H2O2 increased [Ca2+]i from 88.8+/-7.2 nM to 134.6+/-8.3 nM. Magnitude of [Ca2+]i increase induced by 0.1 mM H2O2 was decreased after treatment of cells with 0.5 mM thapsigargin, an inhibitor of endoplasmic reticulum Ca2+ pump (45.8+/-4.9 nM vs 15.0+/-4.8 nM). Small increase in [Ca2+]i in response to an increase of external Ca2+ from zero to 2 mM was further facilitated by 0.1 mM H2O2 (330.5+/-122.7 nM). We concluded that H2O2 not only activates K+ATP channels in association with metabolic inhibition, but also increases partly the Ca2+ permeability of the thapsigargin-sensitive intracellular stores and of the plasma membrane in pancreatic beta-cells.

Sur1 knockout mice. A model for K(ATP) channel-independent regulation of insulin secretion.

Sur1 knockout mouse beta-cells lack K(ATP) channels and show spontaneous Ca(2+) action potentials equivalent to those seen in patients with persistent hyperinsulinemic hypoglycemia of infancy, but the mice are normoglycemic unless stressed. Sur1(-/-) islets lack first phase insulin secretion and exhibit an attenuated glucose-stimulated second phase secretion. Loss of the first phase leads to mild glucose intolerance, whereas reduced insulin output is consistent with observed neonatal hyperglycemia. Loss of K(ATP) channels impairs the rate of return to a basal secretory level after a fall in glucose concentration. This leads to increased hypoglycemia upon fasting and contributes to a very early, transient neonatal hypoglycemia. Whereas persistent hyperinsulinemic hypoglycemia of infancy underscores the importance of the K(ATP)-dependent ionic pathway in control of insulin release, the Sur1(-/-) animals provide a novel model for study of K(ATP)-independent pathways that regulate insulin secretion.

P2Y-purinoceptor mediated inhibition of L-type Ca2+ channels in rat pancreatic beta-cells.

We used the patch-clamp technique to study the effects of extracellular ATP on the activity of ion channels recorded in rat pancreatic beta-cells. In cell-attached membrane patches, action currents induced by 8.3 mM glucose were inhibited by 0.1 mM ATP, 0.1 mM ADP or 15 microM ADPbetaS but not by 0.1 mM AMP or 0.1 mM adenosine. In perforated membrane patches, action potentials were measured in current clamp, induced by 8.3 mM glucose, and were also inhibited by 0.1 mM ATP with a modest hyperpolarization to -43 mV. In whole-cell clamp experiments, ATP dose-dependently decreased the amplitudes of L-type Ca2+ channel currents (ICa) to 56.7+/-4.0% (p<0.001) of the control, but did not influence ATP-sensitive K+ channel currents observed in the presence of 0.1 mM ATP and 0.1 mM ADP in the pipette. Agonists of P2Y purinoceptors, 2-methylthio ATP (0.1 mM) or ADPbetaS (15 microM) mimicked the inhibitory effect of ATP on ICa, but PPADS (0.1 mM) and suramin (0.2 mM), antagonists of P2 purinoceptors, counteracted this effect. When we used 0.1 mM GTPgammaS in the pipette solution, ATP irreversibly reduced ICa to 58.4+/-6.6% of the control (p<0.001). In contrast, no inhibitory effect of ATP was observed when 0.2 mM GDPbetaS was used in the pipette solution. The use of either 20 mM BAPTA instead of 10 mM EGTA, or 0.1 mM compound 48/80, a blocker of phospholipase C (PLC), in the pipette solution abolished the inhibitory effect of ATP on ICa, but 1 microM staurosporine, a blocker of protein kinase C (PKC), did not. When the beta-cells were pretreated with 0.4 microM thapsigargin, an inhibitor of the endoplasmic reticulum (ER) Ca2+ pump, ATP lost the inhibitory effect on ICa. These results suggest that extracellular ATP inhibits action potentials by Ca2+-induced ICa inhibition in which an increase in cytosolic Ca2+ released from thapsigargin-sensitive store sites was brought about by a P2Y purinoceptor-coupled G-protein, PI-PLC and IP3 pathway.

A case of renal juxtaglomerular cell tumor: usefulness of segmental sampling to prove autonomic secretion of the tumor.

A 27-year-old female patient had been treated for hypertension with conventional therapy for years, because renal vein renin levels failed to show lateralization in renal venous samplings and a renal juxtaglomerular cell tumor (RJGCT) had gone undiagnosed. Abdominal computed tomography revealed a mass at the middle of the right kidney. The right renal venogram demonstrated distinct segmental veins from the upper pole and from the middle and lower poles in the right kidney. On segmental renin sampling from each renal vein, the plasma renin concentration (PRC) of the segmental veins from the middle and lower poles was higher than that from other sites. We diagnosed RJGCT of the right kidney and performed right-sided nephrectomy. After the resection, the PRC rapidly decreased. Immunohistochemical studies using antihuman renin antibodies revealed positive staining of the tumor cells. It is an important strategy to make a segmental sampling at the site as close as possible to the RJGCT.

Repetitive mitochondrial Ca2+ signals synchronize with cytosolic Ca2+ oscillations in the pancreatic beta-cell line, MIN6.

We examined the relationship between cytosolic Ca2+ concentration ([Ca2+]c) and mitochondrial matrix Ca2+ concentration ([Ca2+]m) in the pancreatic beta-cell line, MIN6. [Ca2+]c was monitored in a single or a group (30 cells) of fura-2-loaded MIN6 cells, and [Ca2+]m was measured in a group (1 x 10[6] cells) of MIN6 cells stably transfected with aequorin targeted at the mitochondria. Exogenous ATP (0.25 mmol/l) produced a single transient increase in [Ca2+]c whereas 22 mmol/l KCl produced a sustained plateau increase. ATP and KCl evoked transient increases in [Ca2+]m but with distinct time courses of [Ca2+]m decline: the [Ca2+]m increase induced by ATP decreased more rapidly than that induced by KCl. Nitrendipine (3 micromol/l), a blocker of L-type Ca2+ channels, inhibited both [Ca2+]c and [Ca2+]m signals in response to KCl and tolbutamide, but not those to ATP. Peak levels of [Ca2+]m increase (around 2 micromol/ l) exceeded those of [Ca2+]c increase (around 500 nmol/l). A rise in glucose concentration from 3 to 30 mmol/l induced oscillations of [Ca2+]c that overlay the sustained increases in [Ca2+]c in single cells. An oscillatory increase in [Ca2+]m was similarly observed in response to glucose. Addition of 10 mmol/l 2-ketoisocaproic acid at 20 mmol/l glucose further increased the plateau level of [Ca2+]c and the frequency of [Ca2+]c oscillations, which were correlated with a further increase in [Ca2+]m. In response to pulsatile exposure to KCl, [Ca2+]c and [Ca2+]m increased synchronously. These data suggest that an oscillatory increase in [Ca2+]m in beta cells, the signal which is thought to be necessary for continuous stimulation of mitochondrial metabolism, is produced synchronously with the [Ca2+]c oscillations.

Targeting of GLUT1-GLUT5 chimeric proteins in the polarized cell line Caco-2.

Caco-2, a human differentiated intestinal epithelial cell line, is a promising model for investigating the mechanism of polarized targeting of apical and basolateral membrane proteins. We stably transfected rat GLUT5 cDNA and rabbit GLUT1 cDNA into Caco-2 cells with an expression vector. Immunohistochemical study revealed that the GLUT5 protein expressed was localized at apical membranes and that the GLUT1 expressed was present primarily in the basolateral membranes of cells grown on permeable support. Next, to investigate the domain responsible for determining apical vs. basolateral sorting in glucose transporters, we prepared several GLUT1-GLUT5 chimeric cDNAs and transfected them into Caco-2 cells. A GLUT1 [N terminus approximately sixth transmembrane domain (TM6)]-GLUT5 [intracellular loop (IL) approximately C terminus] chimera was observed exclusively at the apical membrane, while GLUT1 (N terminus approximately IL)-GLUT5 (TM7 approximately C terminus) and GLUT1 (N terminus approximately TM12)-GLUT5 (C-terminal domain) chimeras were observed mainly at the basolateral membrane, a localization similar to that of GLUT1. Moreover, using a recombinant adenovirus expression system, we expressed a GLUT5 (N terminus approximately TM6)-GLUT1(IL)-GLUT5(TM7 approximately C-terminus) chimera, which was observed at the basolateral membrane. Based on these results, the C-terminal domain does not determine isoform-specific targeting of GLUT1 and GLUT5. Rather, it is the intracellular loop in glucose transporters that appears to play a pivotal role in apical-basolateral sorting signals in Caco-2 cells.

Effect of mitochondrial and/or cytosolic glycerol 3-phosphate dehydrogenase overexpression on glucose-stimulated insulin secretion from MIN6 and HIT cells.

The glycerol phosphate shuttle consists of FAD-linked mitochondrial glycerol 3-phosphate dehydrogenase (mGPDH) and its cytosolic NAD-linked isoform (cGPDH). Impaired mGPDH activity has recently been suggested to be one of the primary causes of insulin secretory defects in beta-cells. We found that mGPDH and cGPDH activities in MIN6 cells are comparable to those of isolated islets and higher than those in HIT cells by eightfold and threefold, respectively. Therefore, we selected the MIN6 cell line as a beta-cell model with normally regulated insulin secretion and normal shuttle enzyme activities and the HIT cell line as a beta-cell model with impaired insulin secretion and lower activities of these enzymes. The role of these dehydrogenases in glucose-stimulated insulin secretion was addressed by examining the effects of overexpression of mGPDH and/or cGPDH via recombinant adenoviruses in these cells. Infection with recombinant adenovirus with a cDNA encoding the Escherichia coli beta-galactosidase gene resulted in expression of its gene in 90% of MIN6 and HIT cells. Infection with a recombinant adenovirus with mGPDH cDNA (Adex1CAmGPDH) caused 2.1-fold and 5.7-fold increases in dehydrogenase activity as compared with those of control MIN6 and HIT cells, respectively. Infection with a recombinant adenovirus with cGPDH cDNA (Adex1CAcGPDH) caused a more than 50-fold increase in activity in both cell lines. Glycerol phosphate shuttle flux, as estimated by [2-3H]glycerol conversion to [3H]H2O, was increased to 120-130% by infection with Adex1CAmGPDH, but not with Adex1CAcGPDH infection, in both MIN6 and HIT cells. No further increase in flux through the glycerol phosphate shuttle was detected when the cells were infected with Adex1CAmGPDH together with Adex1CAcGPDH. Furthermore, neither [U-14C]glucose oxidation nor the insulin secretory response to glucose was affected in either cell line. Thus, mGPDH abundance in MIN6 and HIT cells is not directly related to their insulin secretory capacity in response to glucose, and reduced expression of mGPDH is not the primary cause of abnormal insulin secretory responses in HIT cells. The present data indicate that the emerging hypothesis pointing to mGPDH deficiency as a possible cause of NIDDM needs to be carefully evaluated.

Characterization of rat GLUT5 and functional analysis of chimeric proteins of GLUT1 glucose transporter and GLUT5 fructose transporter.

To investigate the biological and biochemical properties of GLUT5, rat GLUT5 complementary DNA was transfected into Chinese hamster ovary cells. Rat GLUT5 was exclusively targeted to the plasma membrane and exhibited a transport activity, not for glucose, but for fructose. The affinity for fructose (Km = 11.6mM) was much higher than that of GLUT2, the other glucose transporter with fructose transport activity. Interestingly, rat GLUT5 was not photolabeled with 0.5 microM cytochalasin B, whereas a similar amount of GLUT1 was adequately photolabeled under the same experimental conditions. Next, to investigate the domains required for transport of glucose/fructose in GLUT1 and/or GLUT5, several chimeric GLUT1/GLUT5 proteins were expressed, and their glucose and/or fructose transport activities were studied. The intracellular middle loop and the region encompassing the membrane spanning domains 7-12 were observed to have crucial roles in GLUT1 glucose transport, whereas replacement of the N-terminal half or the intracellular C-terminal region with the corresponding region of GLUT5 produced no marked effects on glucose transport activity. In contrast, both the N-terminal half encompassing the region from the N-terminus through the 6th membrane spanning domain and the intracellular C-terminal region were mandatory for GLUT5 fructose transport. In conclusion, GLUT5 is a transporter exclusively for fructose and the structural requirements for fructose transport are more stringent than those for glucose transport among hexose transporter proteins.

Involvement of ATP-sensitive K+ channels in free radical-mediated inhibition of insulin secretion in rat pancreatic beta-cells.

To explore the mechanisms of inhibition of insulin secretion in pancreatic beta-cells by oxygen free radicals, we studied the effects of H2O2 on membrane currents using the patch-clamp technique. Exposure of beta-cells to H2O2 (> or = 30 mumol/l) increased the activity of ATP-sensitive potassium (K+ATP) channels without changing the single channel conductance in cell-attached membrane patches. Action currents observed during superfusion of 11.1 mmol/l glucose were suppressed. In inside-out membrane patches, the activity of K+ATP channels was not influenced by H2O2. In conventional whole-cell clamp experiments using a pipette solution containing 3 mmol/l ATP, H2O2 did not influence the membrane currents. However, H2O2 did activate the K+ATP channel current in perforated whole-cell clamp configurations. The increased K+ATP channel current was reversed by subsequent exposure to 11.1 mmol/l 2-ketoisocaproic acid. In cell-attached membrane patches, the K+ATP channel current evoked by exposure to 30 mumol/l H2O2 was inhibited by exposure to 11.1 mmol/l glyceraldehyde, whereas the channel was again activated by exposure to 0.3 mmol/l H2O2. Subsequent superfusion of 11.1 mmol/l 2-ketoisocaproic acid inhibited the channel; this effect was counteracted by exposure to 10 mmol/l H2O2. Transient inhibition of K+ATP channels with provocation of action potentials was observed after washout of 100 mumol/l H2O2 during superfusion of 2.8 or 11.1 mmol/l glucose.(ABSTRACT TRUNCATED AT 250 WORDS)

A new hypoglycemic agent, A-4166, inhibits ATP-sensitive potassium channels in rat pancreatic beta-cells.

Effects of a new hypoglycemic drug, N-[trans-4-isopropylcyclohexy-carbonyl]-D-phenylalanine (A-4166), on membrane current were investigated using the patch-clamp technique in single pancreatic beta-cells isolated from rats. A-4166, at a concentration of 10 microM, depolarized membrane potential of beta-cells and evoked action potentials in the presence of 2.8 mM glucose. The single ATP-sensitive K+ channel (K-ATP channel) current recorded in cell-attached membrane patches was reversibly inhibited by A-4166 (> 0.1 microM) without a change in the single-channel conductance of the K-ATP channel. Both A-4166 and tolbutamide inhibited the whole cell K-ATP channel current with half-maximum inhibition (IC50) of 0.23 and 12.8 microM, respectively (Hill coefficient = 1). In inside-out membrane patches, the IC50 with A-4166 occurred at 4.5 nM, in contrast to 0.7 microM for tolbutamide. A-4166 did not affect L- and T-type Ca2+ channels or the time-dependent outward current. We conclude that A-4166 specifically blocks the K-ATP channel and that the blockade is more potent than that of tolbutamide. The action of A-4166 underlies the mechanism by which the drug stimulates insulin secretion from beta-cells.

[Molecular mechanism of the K+ATP channel in pancreatic beta-cells].

The ATP-sensitive K+ (K+ATP) channel plays a key role in secretion of insulin in response to glucose-stimulation in pancreatic beta-cells. Inhibition of the channel does not require hydrolysis of ATP and results from a direct binding of ATP4- to the channel. MgADP relieves the channel inhibition by ATP by decreasing affinity of the channel to ATP. We suggest two-sites model regarding channel modulations by these nucleotides; one is the ATP-inhibition site which is bound by ATP4-, and the other the modulation site, which is bound by MgADP and thereby decreases the sensitivity of the channel to ATP. Sulphonylureas-binding sites may be different from these nucleotide-binding sites described above.

Inhibition of the ATP-sensitive potassium channel by class I antiarrhythmic agent, cibenzoline, in rat pancreatic beta-cells.

1. Cibenzoline, a class I antiarrhythmic agent, was investigated for its effect on the ATP-sensitive K+ channel of pancreatic beta-cells by the patch clamp technique. 2. In perforated patch clamp experiments, cibenzoline depolarized the membrane of single beta-cells and thereafter, caused firing of action potentials in the presence of 2.8 mM glucose. 3. Cibenzoline inhibited the activity of the ATP-sensitive K+ channel in cell-attached recordings in the presence of 2.8 mM glucose and evoked repetitive fluctuations of the baseline current, apparently reflecting the action potentials of the beta-cell. 4. In whole-cell clamp experiments, time-independent outward current was induced by depleting cytoplasmic ATP with 0.1 mM ATP and 0.1 mM ADP in the solution contained in the pipette. The outward current was inhibited by cibenzoline in a dose-dependent manner in the concentration range of 1 microM to 100 microM and half maximum inhibition occurred at 1.5 microM. 5. Cibenzoline blocked substantially the ATP-sensitive K+ channel current when applied at the inner side of the membrane in isolated inside-out membrane patches. 6. It is concluded that cibenzoline blocks the ATP-sensitive K+ channel of pancreatic beta-cells and, thereby, stimulates insulin secretion at sub-stimulatory levels of glucose.

Response of osteoblastic clonal cell line (MC3T3-E1) to [Asu]eel calcitonin at a specific cell density or differentiation stage.

Clone MC3T3-E1 cells isolated from newborn mouse calvaria is an osteogenic cell line which retains an ability to differentiate into osteoblastic cell in vitro. The effect of [Asu]eel calcitonin (ECT) on clonal MC3T3-E1 cells was investigated at different stages of differentiation. ECT caused an increase in alkaline phosphatase (ALP) activity. The stimulative effect was demonstrated to be dependent upon cell density or differentiation stage. At a cell density of 1.18 X 10(5)/cm2 cells were incubated with ECT for 2 days. The treatment by ECT caused an increase in ALP activity. A specific response to ECT dependent on the cell density was observed in a narrow range of cell density. Moreover this range of cell density responsible to ECT was found to be a rapid differentiation stage of MC3T3-E1 cells. These results suggest that calcitonin stimulates differentiation of osteoblast. In addition to these results, cellular adenosine-3',5'-cyclic monophosphate (cAMP) level was raised by ECT treatment at a cell density of about 1.4 X 10(5) cell/cm2 and this response was also specific for cell density. At cell density lower or higher than this density no stimulative effect by ECT was observed. On the other hand, N6,O2-dibutyryl adenosine-3',5'-cyclic monophosphate (db-cAMP) and theophylline caused an increase in ALP activity in wide cell density range. These results indicate that an increase in ALP activity by ECT is mediated by intracellular cAMP and that the specific response to ECT dependent on the cell density is regulated in the process of cAMP formation and/or in the preceding process of cAMP formation.