Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Impairment of Human Ether-à-Go-Go-related Gene (HERG) K+ Channel Function by Hypoglycemia and Hyperglycemia

Impairment of Human Ether-à-Go-Go-related Gene (HERG) K+ Channel Function by Hypoglycemia and... THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 278, No. 12, Issue of March 21, pp. 10417–10426, 2003 © 2003 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Impairment of Human Ether-a ` -Go-Go-related Gene (HERG) K Channel Function by Hypoglycemia and Hyperglycemia SIMILAR PHENOTYPES BUT DIFFERENT MECHANISMS* Received for publication, October 29, 2002, and in revised form, December 20, 2002 Published, JBC Papers in Press, January 16, 2003, DOI 10.1074/jbc.M211044200 Yiqiang Zhang‡§, Hong Han‡, Jingxiong Wang‡§, Huizhen Wang‡¶, Baofeng Yang, and Zhiguo Wang‡§** From the ‡Research Center, Montreal Heart Institute, Montreal, Quebec H1T 1C8, the §Department of Medicine, University of Montreal, Montreal, Quebec H3C 3J7, Canada and the Department of Pharmacology, Harbin Medical University, Harbin, Heilongjiang 150086, Peoples Republic of China Hyperglycemia and hypoglycemia both can cause pro- Glucose, the primary end product of the digestion of glycogen, longation of the Q-T interval and ventricular arrhyth- is essential for maintaining life activities in organisms. As a mias. Here we studied modulation of human ether-a ` -go- major source of metabolic fuel, degradation of glucose via glycol- go-related gene (HERG) K channel, the major molecular ysis and subsequent oxidative phosphorylation generates high component of delayed rectifier K current responsible for energy phosphates to power the biological processes in the cell. cardiac repolarization, by glucose in HEK293 cells using Yet, through an exquisitely complex network of control mecha- whole-cell patch clamp techniques. We found that both nisms, the rate of glucose metabolism is only as great as needed hyperglycemia (extracellular glucose concentration by the organisms. Moreover, glucose also has other regulatory [Glu]  10 or 20 mM) and hypoglycemia ([Glu]  2.5, 1, or o o effects on many cellular functions. Either inadequate or excessive 0mM) impaired HERG function by reducing HERG cur- glucose can be harmful to the living system. Therefore, the blood rent (I ) density, as compared with normoglycemia HERG glucose level is dynamically controlled. However, under patho- ([Glu]  5mM). Complete inhibition of glucose metabo- logical conditions like diabetes, glucose cannot be efficiently uti- lism (glycolysis and oxidative phosphorylation) by 2-de- lized, and the blood glucose level rises. When the blood level of oxy-D-glucose mimicked the effects of hypoglycemia, but inhibition of glycolysis or oxidative phosphorylation glucose is maintained higher than 7 mM, it is considered as alone did not cause I depression. Depletion of intra- hyperglycemia. Diabetes therapy, on the other hand, can lead to HERG cellular ATP mimicked the effects of hypoglycemia, and an overly low level of blood glucose, which is referred to as replacement of ATP by GTP or non-hydrolysable ATP hypoglycemia when the level falls below 3 mM. failed to prevent the effects. Inhibition of oxidative phos- Either hypoglycemia or hyperglycemia can have deleterious phorylation by NaCN or application of antioxidants vita- effects on the cells. One common feature of electrophysiological min E or superoxide dismutase mimetic (Mn(III) tet- alterations caused by both hypoglycemia and hyperglycemia in rakis(4-benzoic acid) porphyrin chloride) abrogated and the heart is prolongation of Q-T interval and the associated incubation with xanthine/xanthine oxidase mimicked the ventricular arrhythmias that are presumably responsible for effects of hyperglycemia. Hyperglycemia or xanthine/ sudden cardiac death in diabetic patients (1–10). However, the xanthine oxidase markedly increased intracellular levels ionic mechanisms by which hyperglycemia and hypoglycemia of reactive oxygen species, as measured by 5-(and-6)- prolong Q-T interval remained unclear, which is at least a part of chloromethyl-2,7-dichlorodihydrofluorescein diacetate (CM-H DCFDA) fluorescence dye, and this increase was the reasons why diabetic patients die of mainly cardiac prevented by NaCN, vitamin E, or Mn(III) tetrakis(4- complications. benzoic acid) porphyrin chloride. We conclude that ATP, The human either-a ` -go-go-related gene (HERG) encodes the derived from either glycolysis or oxidative phosphoryla- rapid component of delayed rectifier K current in the heart, tion, is critical for normal HERG function; depression of which is the major repolarizing current in the plateau voltage I in hypoglycemia results from underproduction of HERG range of cardiac action potentials. HERG K channels are sus- ATP and in hyperglycemia from overproduction of reac- ceptible to genetic defects and environmental cues, with the tive oxygen species. Impairment of HERG function might consequence being depression of HERG function in most situa- contribute to Q-T prolongation caused by hypoglycemia tions (9). Indeed, most of the cases of long Q-T syndrome are and hyperglycemia. ascribed to dysfunction of HERG channels, particularly that in- duced by therapeutic drugs (13). It is conceivable that HERG alteration might also be involved in the Q-T prolongation induced * This work was supported in part by the Canadian Institute of by hyperglycemia and hypoglycemia. This thought prompted us Health Research, the Heart and Stroke Foundation of Quebec, and the to carry out a series of experiments to study the effects of glucose Fonds de la Recherche de l’Institut de Cardiologie de Montreal (to on HERG K channels and the potential mechanisms. Z. W.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ¶ Research fellow of the Canadian Institute of Health Research. The abbreviations used are: HERG, human ether-a ` -go-go related ** Research scholar of the Fonds de Recherche en Sante de Quebec. gene; ROS, reactive oxygen species; X, xanthine; XO, xanthine oxidase; To whom correspondence should be addressed: Research Center, Mon- 2dG, deoxy-D-glucose; MnTBAP, Mn(III) tetrakis(4-benzoic acid) por- treal Heart Institute, 5000 Belanger East, Montreal, Quebec H1T 1C8, phyrin chloride; AMP-PCP, ,-methyleneadenosine 5-triphosphate; Canada. Tel.: 514-376-3330; Fax: 514-376-4452; E-mail: wangz@icm. VitE, vitamin E; SOD, superoxide dismutase; CM-H DCFDA, 5-(and- umontreal.ca. 6)-chloromethyl-2,7-dichlorodihydrofluorescein diacetate. This paper is available on line at http://www.jbc.org 10417 This is an Open Access article under the CC BY license. 10418 Metabolic Mechanisms of HERG Modulation by Glucose EXPERIMENTAL PROCEDURES lution containing a given concentration of Glu for 30 min prior Cell Culture—HEK293 cells stably expressing HERG (a kind gift to patch clamp recordings. In addition, recordings were per- from Drs. Zhou and January) (14) were grown in Dulbecco’s modified formed immediately after formation of whole-cell configuration Eagle’s medium supplemented with 10% heat-inactivated fetal bovine and adjustments of capacitance and series resistance compen- serum, 200 M G418, 100 units/ml penicillin, and 100 g/ml strepto- sation, and all recordings were made complete within 3 min. In mycin. The cells subcultured to 85% confluency were harvested by this manner, there was minimal dialysis through the recording trypsinization and stored in the Tyrode’s solution containing 0.5% bo- pipette and thereby minimal current run-down (time-depend- vine serum albumin at 4 °C (12). Electrophysiological recordings were conducted within 10 h of storage. ent current decay), and the data best reflect the effects of Glu Whole-cell Patch Clamp Recording—Patch clamp techniques have on I in cells with intact intracellular contents. In addition, HERG been described in detail elsewhere (15–16). Currents were recorded such an experimental design also allowed us to study the effect with whole-cell voltage clamp with an Axopatch-200B amplifier (Axon of Glu on I under conditions devoid of influence from HERG Instruments). Borosilicate glass electrodes had tip resistances of 1–3 exogenous ATP included in the pipette, which is an important megohms when filled with the internal solution containing (mM) 130 KCl, 1 MgCl , 5 Mg-ATP, 10 EGTA, and 10 HEPES (pH 7.3). The issue to be described later. I was elicited by 2.5-s depolar- 2 HERG extracellular (Tyrode’s) solution contained (mM) 136 NaCl, 5.4 KCl, 1 izing steps from 60 to 40 mV to record the activating cur- CaCl , 1 MgCl , 10 HEPES (pH 7.4), and glucose at concentrations as 2 2 rent, followed by a repolarizing pulse to 50 mV for another specified. Experiments were conducted at 36  1 °C. Junction poten- 2.5 s to observe the deactivating tail current, before being tials were zeroed before formation of the membrane-pipette seal. Series returned to a holding potential of 80 mV. The results are resistance and capacitance were compensated, and leak currents were illustrated in Fig. 1 with both representative raw data and subtracted. Pharmacological Probes—D-Glucose (Glu), 2-deoxy-D-glucose (2dG), analyzed mean data. Glucose produced two characteristic al- sodium cyanide (NaCN), pyruvate, ATP, GTP, AMP-PCP (non-hydro- terations of HERG channel functions as follows: changes of lysable analogue of ATP), xanthine (X), xanthine oxidase (XO), and I amplitude and density and shifts of I-V relationships HERG vitamin E (VitE) were all purchased from Sigma. Xanthine was pre- and activation curves. pared in 2 N NaOH and diluted in the Tyrode’s solution 800 times with Comparison of I recorded at varying extracellular con- HERG the pH adjusted to 7.4 with HCl. Xanthine oxidase was added to the centrations of Glu ([Glu]  0, 1, 2.5, 5, 10, or 20 mM) consis- xanthine preparation to form the X/XO-reactive oxygen species (ROS) generating system. VitE was dissolved in ethanol and diluted 1000 tently showed that I density was maximal at a physiolog- HERG times to reach the final concentration. Pyruvate in liquid was diluted ical [Glu] (5 mM), and it was depressed at [Glu] below or above o o into the Tyrode’s solution, and pH was adjusted to 7.4 with NaOH 5mM (Fig. 1, A–C). In other words, under normoglycemia before use. All other compounds were directly dissolved into the patch ([Glu]  5mM) the HERG K channel operates at its maxi- clamp recording solutions as specified. Mn(III) tetrakis(4-benzoic acid) mum function level, whereas under hypoglycemia ([Glu]  0, porphyrin chloride (MnTBAP) purchased from Calbiochem was dis- 1, or 2.5 mM) or hyperglycemia ([Glu]  10 or 20 mM), the solved in 1 N NaOH and diluted by 5000 times to reach the desired experimental concentrations. All compounds and reagents were pre- HERG channel function is impaired, and the degree of func- pared fresh before the experiments. tional impairment is proportional to the degrees of hypoglyce- Intracellular Reactive Oxygen Species (ROS) Measurement—5- mia or hyperglycemia (Fig. 1D). For example, with 5 mM [Glu] , (and-6)-chloromethyl-2,7-dichlorodihydrofluorescein diacetate (CM- the I current density was 88.5  7.4 pA/pF (n  25) at a HERG H DCFDA) (Molecular Probes) is a ROS-sensitive probe that can be test potential of 0 mV, whereas with 0 and 20 mM, the values used to detect oxidative activity in living cells. It passively diffuses into cells, where its acetate groups are cleaved by intracellular esterases, were 35.4  3.0 pA/pF (n  22, p  0.05 versus 5mM [Glu] ) and releasing the corresponding dichlorodihydrofluorescein derivative. Its 35.8  4.9 (n  15, p  0.05 versus 5mM [Glu] ), respectively, thiol-reactive chloromethyl group reacts with intracellular glutathione approximately a 2-fold difference. The HERG tail current den- and other thiols. Subsequent oxidation yields a fluorescent adduct that sity at 10 mV was also significantly depressed by hypoglyce- is trapped inside the cell. When it is excited at 480 nm, its emissions at mia (0, 1, and 2.5 mM [Glu] ) or hyperglycemia (10 and 20 mM 505–530 nm can be captured. CM-H DCFDA is prepared in dimethyl [Glu] ) (data not shown). sulfoxide immediately prior to loading. Glass coverslips were coated with laminin and placed in the wells of a 12-well culture plate before the Also noticeable is the hypolarization shifts of I I-V re- HERG cells were seeded into the well in a density of 5.0  10 /well. After lationships (Fig. 1E) and the activation curve by hypoglycemia overnight incubation, the cells were washed with pre-warmed (37 °C) (Fig. 1F). For example, the half-activation voltage (V1 ) was ⁄2 phosphate-buffered saline once and then incubated in the Tyrode’s 25.4  2.7 mV (n  19) for 5 mM [Glu] and 34.3  3.8 mV solution containing glucose of varying concentrations or exogenous su- (n  22) for 0 mM [Glu] (p  0.05, unpaired t test), which peroxide-generating system (xanthine/xanthine oxidase) or the re- accounted for around 10 mV shift of the steady-state voltage- agents as to be otherwise specified, together with the fluorescence dye CM-H DCFDA (10 M). After 30 min of incubation, the coverslips were 2 dependent activation of I . This negative shift resulted in a HERG washed with pre-warmed phosphate-buffered saline twice before being crossover of the I-V curves between normoglycemia and hypo- mounted to the glass slides with anti-fading mounting medium and glycemia. Only slight (4 mV) hyperpolarization shift of acti- were examined immediately under a laser scanning confocal microscope vation was seen with hyperglycemia (20 mM [Glu] ). (Zeiss LSM 510). The percentage of positively stained cells and the Several potential mechanisms could explain the observed fluorescence intensity of staining were determined by densitometric scanning with LSM software (Zeiss). effects of [Glu] on I . First, there was a possibility that the o HERG Data Analysis—Group data are expressed as mean  S.E. Compar- effects were a consequence of alterations of extracellular osmo- isons among groups were made by analysis of variance (F-test), and larity with varying [Glu] . Second, Glu might act directly on Bonferroni-adjusted t tests were used for multiple group comparisons, HERG proteins to modify the channel function. Finally, Glu and paired or unpaired t test was used, as appropriate, for single metabolism, which generates ATP as well as other metabolic comparisons. A two-tailed p  0.05 was taken to indicate a statistically intermediates, also has the potential to modulate I . The significant difference. Nonlinear least square curve fitting was per- HERG formed with CLAMPFIT in pCLAMP 8.0 or GraphPad Prism. following experiments were performed to clarify these issues. To test the first possibility, we performed experiments in RESULTS which cells were first superfused with a given concentration of Effects of Glucose on I —To study the effects of varying Glu (1, 5, or 20 mM) for 30 min, followed by I recording HERG HERG concentrations of glucose (Glu, ranging from 0 to 20 mM)on within 3 min bathing in the Tyrode’s solution containing 10 mM I , our experiments were designed for group comparisons. Glu. Under such conditions, I demonstrated the same HERG HERG pattern of changes as described above; cells pre-exposed to 1 or For each experiment, HERG-expressing HEK293 cells were divided into six groups each superfused with the Tyrode’s so- 20 mM Glu had markedly smaller I density than those HERG Metabolic Mechanisms of HERG Modulation by Glucose 10419 FIG.1. Effects of glucose on HERG K current (I ) stably expressed in HEK293 cells. A, typical examples of I traces recorded HERG HERG in the Tyrode’s solution containing 0, 5, 10, or 20 mMD-glucose (Glu) with the voltage protocol shown in the inset and presented as current density (pA/pF) for better group comparisons. The same voltage protocol is applied for the I recordings shown in the subsequent figures, except as HERG otherwise indicated. B, current density-voltage relationships of I . The steady-state step I measured at the end of 2.5-s pulses was HERG HERG normalized to the capacitance of the respective cells and plotted as a function of test potentials. Shown are data averaged from 22, 9, 11, 25, 24, and 15 cells for 0, 1, 2.5, 5, 10, and 20 mM Glu, respectively. *, p  0.05 versus 5mM [Glu] . C, ratio of the step I recorded with varying [Glu] o HERG o over the I with 5 mM [Glu] as a function of test potentials. D, ratio of the step I recorded with varying [Glu] over the I with 5 mM HERG o HERG o HERG [Glu] as a function of glucose concentrations. Shown are the data obtained at test potentials of 40, 10, and 20 mV. E, normalized I-V relationships obtained by dividing the current amplitude at various potentials by the maximum current for each concentration of glucose. For clarity, only the data with 0, 5, and 20 mM [Glu] are shown. Note the negative shifts of the curves with low and high [Glu] , relative to 5 mM [Glu] , o o o along the voltage axis. F, steady-state voltage-dependent activation of I . The activation curves were constructed by plotting the conductance HERG G as a function of potentials. G was calculated by normalizing the tail currents at 50 mV by dividing the amplitude of the tail currents evoked at various antecedent step potentials by that of the tail current at 40 mV. The symbols are mean of experimental data, and the lines represent the Boltzmann fit: G/G  1/{1  exp((V1  V)/k)}, where G represents the maximal conductance at 40 mV, V1 is a half-maximal activation max ⁄2 max ⁄2 voltage, and k is a slope factor. The numbers in the legends represent the extracellular glucose concentrations ([Glu] )inmM. pre-exposed to 5 mM Glu (Fig. 2A), despite that all recordings investigate whether the effects of Glu on I are associated HERG were made under isotonic conditions with 10 mM [Glu] . For with glucose metabolism, we studied the effects of the glycoly- instance, at 0 mV, the I density was 56.1  6.9 pA/pF (n  sis inhibitor 2-deoxy-D-glucose (2dG). Glucose was replaced by HERG 8) with 5 mM [Glu] , whereas with 1 and 20 mM [Glu] , the the non-hydrolysable analogue of glucose 2dG to eliminate the o o values were 32.4  3.2 pA/pF (n  7, p  0.05 versus 5mM glucose metabolism (both glycolysis and subsequent oxidative [Glu] ) and 35.8  2.2 (n  7, p  0.05 versus 5mM [Glu] ), phosphorylation). Cells were superfused with the Glu-free Ty- o o respectively. Consistently, the activation curve of I was rode’s solution containing 5 mM 2dG for 30 min before patch HERG also shifted to more negative potentials by hypoglycemia (Fig. clamp recordings. I recorded under such a condition was HERG 2B). A similar difference of I between 5 and 20 mM [Glu] compared with I recorded with the Tyrode’s solution con- HERG o HERG was also consistently seen when 15 mM cellobiose was added to taining 5 mM Glu. As shown in Fig. 3, 2dG substitution for Glu the Tyrode’s solution containing 5 mM Glu (Fig. 2D). reproduced the two characteristic changes of I as observed HERG The above data indicate that changes of osmolarity is un- under hypoglycemia. First, marked depression of I was HERG likely the mechanism by which glucose modulates I . Apart seen in the presence of 2dG, with I density only 38% that HERG HERG from that, the fact that differences of I among the cells in the presence of 5 mM [Glu] at 0 mV. Second, similar to the HERG o pretreated with three different concentrations of Glu (1, 5, and results with 0 mM [Glu] , the I-V relationship and activation 20 mM) persisted even though I was recorded 10 min after curve were shifted by 10 mV toward hyperpolarizing potentials HERG superfusion with the normal Tyrode’s solution containing 10 by 2dG (28.2  3.4 mV for 5 mM [Glu] and 37.6  4.5 mV mM [Glu] suggests that the effects of glucose on I are for5mM 2dG, p  0.05, n  8 for both groups) (Fig. 3B). o HERG mediated by some intracellular events, and direct interactions Because 2dG is a competitive inhibitor of glycolysis, its effects between glucose and HERG channels do not likely play a major on I in the presence of 5 mM Glu were also investigated. HERG role. Cells were superfused with the Glu-containing Tyrode’s solu- Role of Glycolysis and Oxidative Phosphorylation on Glucose- tion with or without 5 mM 2dG for 30 min before patch clamp induced I Enhancement—Glucose, once being taken up recordings. The I density was 45% smaller in cells HERG HERG into the cell, is metabolized via glycolysis to generate 2 mole- treated with 2dG than in untreated cells (Fig. 3C). For exam- cules of ATP and 2 molecules of pyruvate (a substrate for ple, the I density at 0 mV was 40.2  4.1 pA/pF for cells HERG oxidative phosphorylation), which is further metabolized via treated with 2dG at 5 mM [Glu] (n  14) and was 75.6  6.6 oxidative phosphorylation to produce more ATP molecules. To pA/pF for cells without 2dG treatment (n  14, p  0.05). These 10420 Metabolic Mechanisms of HERG Modulation by Glucose FIG.2. Effects of glucose on I under conditions with cor- HERG rected osmolarity. A and B, mean current density (pA/pF)-voltage relationships and activation conductance (G) curves, respectively (n 7, 9, and 7 for 1, 5, and 20 mM [Glu] , respectively). The cells of different groups were superfused with solutions containing a given concentration of glucose (1, 5, or 20 mM) for 30 min and then switched to the same solution with an identical glucose concentration (10 mM). The I HERG values recorded within 3 min at 10 mM [Glu] were used for analysis. *, p  0.05 versus 5mM [Glu] , unpaired t tests. C and D, depression of I and negative shift of I activation, respectively, in high glu- HERG HERG cose (20 mM [Glu] , n  5), as compared with those in normal glucose (5 mM [Glu] )  15 mM cellobiose (n  5) to balance the extracellular osmolarity. *, p  0.05 versus Glu-5  cellobiose 15, unpaired t-tests. FIG.3. Effects of complete inhibition of glucose metabolism or results indicate the importance of glucose metabolism in main- inhibition of glycolysis on I . A and B, mean current density HERG taining the normal HERG function. Intriguingly, when 2dG (pA/pF)-voltage relationships and activation conductance (G) curves of was added to the hyperglycemic solution containing 20 mM I . Complete inhibition of glucose metabolism was achieved by the HERG addition of 5 mM 2dG to the Glu-free solution, and inhibition of glycol- [Glu] ,I was increased as compared with that measured o HERG ysis alone was achieved by the addition of pyruvate to the 2dG-contain- in the hyperglycemic solution without 2dG (Fig. 3E). In other ing Glu-free solution. The numbers in the legends represent the con- words, 2dG partly reversed the depressed I caused by HERG centrations of Glu, 2dG, or pyruvate in mM. The number of cells was 8 hyperglycemia toward the normal HERG function seen under for the Glu 5 group, 7 for Glu 0  2dG 5 group, 7 for Glu 0  2dG 5 pyruvate 5 group, and 8 for Glu 0  2dG 5  pyruvate 20 group. *, p normoglycemia. 0.05 versus 5mM [Glu] . C and D, comparison between the effects of 2dG To dissect further which of the two, glycolysis or oxidative (5 mM)onI under 0 mM [Glu] and5mM [Glu] .*, p  0.05 versus HERG o o phosphorylation, is truly responsible for HERG modulation, 5mM [Glu] . E and F, effects of 2dG (5 mM)onI under 20 mM [Glu] . o HERG o the following experiments were carried out. In the first set of The number of cells was 10 for Glu 5 group, 8 for Glu 20 group, and 11 for Glu 20  2dG 5 group. *, p  0.05 versus 20 mM [Glu] , unpaired t experiments, pyruvate was supplied to the 2dG-containing o tests. Glu-free Tyrode’s solution. Under such a condition, the glycol- ysis was inhibited, but the oxidative phosphorylation was maintained. As displayed in Fig. 3A, addition of pyruvate at a ation by NaCN produced a slight non-significant decrease in concentration of 5 mM restored the depressed I induced by I (Fig. 4A). Under hyperglycemia (20 mM [Glu] ), however, HERG HERG o glycolysis inhibition. However, the negative shifts of I-V rela- I was markedly diminished, and NaCN restored the de- HERG tionship and activation curve, as seen with hypoglycemia or pressed I toward the I amplitude seen under normo- HERG HERG 2dG substitution for Glu, were still consistently observed with glycemia (5 mM). For instance, the step and tail I densities HERG pyruvate (Fig. 3B). In contrast, elevation of pyruvate to 20 mM in 20 mM glucose were 50.4  5.9 and 56.4  5.3 pA/pF (n  14) weakened the ability to restore the suppressed I caused by at 10 mV and were restored by NaCN to 73.9  7.9 and 81.6 HERG glycolysis inhibition with 2dG substitution for Glu. Thus, the 6.4 pA/pF (n  13, p  0.05 versus 20 mM [Glu] ), respectively. I density with 20 mM pyruvate in the 2dG-containing Also important is that the oxidative phosphorylation inhibition HERG Glu-free Tyrode’s solution was considerably smaller than that did not produce any significant voltage shifts of I-V relation- with normal [Glu] (5 mM). However, this high concentration of ships and activation curves, regardless of different [Glu] val- o o pyruvate still failed to prevent the negative shifts of I I-V ues (5 or 20 mM) (Fig. 4B). It appears from the above data that HERG relationship and activation curve produced by glycolysis inhi- either glycolysis or oxidative phosphorylation was sufficient to bition (Fig. 3B). sustain the normal function of HERG channels, and significant In the second set of experiments, the oxidative phosphoryl- negative shifts of I I-V relationships and activation curves HERG ation was inhibited by inclusion of NaCN (2 mM), an uncoupler occurred when glycolysis was inhibited, regardless of whether of oxidative phosphorylation, in the bathing solution, and the oxidative phosphorylation was maintained or not. glycolysis was kept intact with 5 or 20 mM [Glu] . Under nor- Role of Intracellular ATP in Maintaining HERG Function— moglycemia (5 mM [Glu] ), inhibition of oxidative phosphoryl- The above experiments indicate that glucose metabolism (gly- o Metabolic Mechanisms of HERG Modulation by Glucose 10421 FIG.4. Effects of inhibition of oxidative phosphorylation on I . Mean current density (pA/pF)-voltage relationships (A) and HERG activation conductance (G) curves (B). Inhibition of oxidative phospho- rylation with intact glycolysis was achieved by the addition of NaCN (2 mM) to the Glu (5 or 20 mM)-containing solution. The numbers in the legends represent the concentrations of Glu or NaCN in mM. n  23 for FIG.5. Effects of depletion of intracellular ATP on I . A, HERG Glu 5 group, 15 for Glu 5  NaCN 2 group, 14 for Glu 20 group, and 13 raw I recorded with the ATP-free pipette solution right after mem- HERG for Glu 20  NaCN 2 group. *, p  0.05 versus Glu5mM; , p  0.05 brane rupture as base-line control data (left) and 10 min after mem- versus Glu 20 mM, unpaired t tests. brane rupture with complete dialysis (right). B, mean I-V relationships (n  5 cells) showing the depression of I caused by depletion of HERG intracellular ATP. C, time-dependent changes of I at 10 mV HERG colysis and oxidative phosphorylation) is critical for I mod- HERG showing the current run down with the ATP-free pipette solution, ulation by Glu. Yet it was unclear whether the I HERG which is otherwise minimal with the ATP-containing pipette solution. modulation by glucose metabolism is associated with the gen- D, normalized I-V relationships showing the negative shift of voltage dependence of I caused by the depletion of intracellular ATP. E, eration of high energy phosphates (i.e. ATP), and if so whether HERG activation conductance (G) curves before and after ATP depletion. *, p the ATP generated from the glucose metabolism is glycolysis- 0.05 versus control (Ctl), paired t tests; , p  0.05, F test indicating the derived or oxidative phosphorylation-derived. To clarify this statistical significance of the time dependence. issue, we first assessed the influence of intracellular ATP de- pletion on HERG function in the presence of 5 mM Glu, by omitting ATP from the pipette (internal) solution. I re- as observed with the ATP-free internal solution (Fig. 6C). For HERG corded immediately after membrane rupture and capacitance/ instance, at 10 mV the I recorded 10 min after dialysis HERG resistance compensation was taken as base-line control data, was 32.2  2.1% smaller than the basal current recorded right and the same measurement was repeated every 5 min up to 15 after membrane rupture. We then went on to test if substitu- min. Under our experimental conditions, 10 min is sufficient to tion of GTP for ATP could prevent I run-down. With 5 mM HERG allow complete dialysis, thereby the equilibrium between pi- GTP in the ATP-free pipette solution, the I developed a HERG pette solution and cytoplasm. As illustrated in Fig. 5, the I similar degree of rapid run-down to what was seen with the HERG recorded with the normal ATP-containing pipette showed only intracellular ATP depletion alone (Fig. 6F). For example, at slight run-down over a 15-min period, whereas the I re- 10 mV, there were 34.4  5.1% decreases in the I am- HERG HERG corded with ATP-free pipette was found significantly reduced plitude 10 min after dialysis. Moreover, the negative shifts of with time. There was an 46% decrease in I at 10 min I-V relationships and activation curves were also seen with HERG after dialysis at 10 mV (Fig. 5C), being similar to the reduc- AMP-PCP or GTP (Fig. 6, B and E). For instance, V1 was ⁄2 tion of I under hypoglycemia or with the inhibition of changed from 33.4  0.8 mV for base line to 37.3  1.0 mV HERG glucose metabolism by 2dG. Also consistent with the hypogly- for 10 min of dialysis with AMP-PCP, and similarly, V1 was ⁄2 cemia and metabolic inhibition was the negative shifts of the shifted from 31.4  1.5 mV to 38.9  2.2 mV by GTP (p I-V relationship (Fig. 5D) and voltage-dependent activation 0.05). (Fig. 5E)ofI with ATP depletion; the V1 was changed by The same effects of ATP depletion on I were consis- ⁄2 HERG HERG 8 mV from 31.2  3.6 mV before to 38.9  8.1 mV (p  tently reproduced when glyburide (10 M) was included in the 0.05, n  7) after [ATP] depletion. superfusate to inhibit ATP-sensitive K current (K ), if any i ATP To investigate whether the requirement of intracellular ATP (data not shown). for HERG function relies on hydrolysis of ATP or is simply Role of ROS on Hyperglycemia-induced I Depression— HERG because of nucleotide interaction with the nucleotide binding Collectively from the above experiments, glucose metabolism is domain of HERG channels (17), we carried out the following necessary for maintaining the HERG channel function, and the series of experiments. We first used the pipette containing the ATP produced by either glycolysis or oxidative phosphorylation non-hydrolysable AMP-PCP to replace ATP. With 5 mM AMP- seems to be a key factor for the regulation; on the other hand, PCP in the pipette, the I demonstrated a rapid run-down the fact that NaCN restores the depressed I induced by 20 HERG HERG 10422 Metabolic Mechanisms of HERG Modulation by Glucose FIG.6. Effects of non-hydrolysable ATP (AMP-PCP) and GTP on I . A HERG and D, I-V relationships. I was re- HERG corded right after membrane rupture and 10 min after dialysis with the ATP-free pipette solution containing AMP-PCP (n  10) (A) or GTP (n  7) (D). B and E, activation conductance (G) curves before and after AMP-PCP or GTP in the ATP- free internal solution. The numbers in the legends represent AMP-PCP or GTP con- centrations in mM. C and F, time-depend- ent changes of I at 10 mV recorded HERG with the pipette containing ATP (control (Ctl)) or AMP-PCP (C) or GTP (F). *, p 0.05 versus control, paired t tests; , p 0.05, F test indicating the significance of time dependence. mM [Glu] or by 20 mM pyruvate suggests that oxidative phos- phorylation also produces negative (suppressive) regulation on HERG function. This would imply that the I suppression HERG by high glucose via oxidative phosphorylation is ATP-inde- pendent or is the balance between the enhancement by ATP and the suppression by other factors associated with oxidative phosphorylation. It has been well established that mitochon- dria produce most of the endogenous reactive oxygen species (ROS) through oxidative phosphorylation (18 –23), and hyper- glycemia stimulates massive ROS production (19, 24 –29). It is therefore rational to propose that the endogenously produced ROS via oxidative phosphorylation stimulated by hyperglyce- mia could impair HERG channel function to suppress I . HERG To test this hypothesis, we performed the following experi- ments. We first evaluated the effects of an antioxidant vitamin E (VitE) on 20 mM [Glu] -induced I depression. Cells were o HERG divided into three groups as follows: 5 mM [Glu] ,20mM [Glu] , o o and 20 mM [Glu]  0.1 mM VitE. As illustrated in Fig. 7, A and B, pretreatment of cell with VitE effectively prevented the I suppression by hyperglycemia; the I density in VitE HERG HERG group was virtually identical to that in the normoglycemia group. There was no significant shift of the activation curve along the voltage axis. These data suggest a participation of FIG.7. Effects of antioxidants on I depression induced by HERG ROS in the HERG regulation by hyperglycemia. Next, we stud- hyperglycemia (20 mM [Glu] ). A and B, mean current density (pA/ ied the effects of another antioxidant, superoxide dismutase pF)-voltage relationships and activation conductance (G) curves, re- (SOD) mimetic MnTBAP, on the I depression induced by HERG spectively, showing the effects of VitE (100 M)onI under 20 mM HERG 20 mM [Glu] . Because the compound does not readily penetrate [Glu] . The cells were superfused with VitE for 40 min or with the normal Tyrode’s solution for control before patch clamp recordings. the cells, it was intracellularly applied through dialysis of the Group comparison was made between the untreated control cells (Ctl, pipette solution at a concentration of 5 M. Cells were super- n  9) and the VitE-treated (n  8) cells. *, p  0.05 versus control, fused with 20 mM Glu for 30 min prior to formation of the unpaired t tests. C and D, mean current density (pA/pF)-voltage rela- whole-cell membrane patch. To correct for the potential current tionships and activation conductance (G) curves, respectively, showing the effects of MnTBAP (5 M), a superoxide dismutase mimetic, on run-down, the I recorded at various time points after HERG I under 20 mM [Glu] . MnTBAP was applied intracellularly HERG o membrane rupture was normalized to the I recorded with HERG through the pipette. I recorded immediately after whole-cell for- HERG the normal Tyrode’s solution without MnTBAP at the corre- mation and series resistance compensation was taken as base-line sponding time points. As shown in Fig. 7C, MnTBAP caused a control data (Ctl) and that recorded 10 min after dialysis was used for analysis to reflect the effects of MnTBAP. To correct for potential time-dependent increase in the I amplitude, indicating a HERG run-down of the current, the I recorded with MnTBAP was nor- HERG restoration of the depressed HERG function. By comparison, no malized to that recorded with the normal internal solution. *, p  0.05 alterations of I , or virtually a slight decrease (presumably HERG versus control, paired t tests. representing run-down of the current), were found with cata- lase in the pipette (data not shown). These results indicate that the ROS involved in the I suppression by hyperglycemia consistently smaller in X/XO-treated cells than in X/XO non- HERG was of mainly superoxide anion (O ). treated cells (Fig. 8, A and B). To obtain further evidence for this notion, we assess the To confirm that ROS production was indeed increased by 20 effects of exogenously produced O by the ROS-generating mM [Glu] and the hyperglycemia-induced ROS was mainly of 2 o system xanthine/xanthine oxidase (X/XO) on I . Cells were O , we proceeded to measure the intracellular ROS levels HERG 2 incubated with or without X/XO (500 M/5 milliunits/ml) in using CM-H DCFDA fluorescence dye. The ROS level was Tyrode’s solution containing 5 mM [Glu] for 40 min before measured in cells preincubated with the Tyrode’s solution con- I was recorded under 5 mM [Glu] . The I density was taining 5 or 20 mM glucose for 30 min. The staining of the cells HERG o HERG Metabolic Mechanisms of HERG Modulation by Glucose 10423 FIG.8. Effects of the oxidant generating system X/XO on I HERG under normoglycemia (5 mM [Glu] ). A, analogue data of I with o HERG and without treatment with X/XO. The cells were superfused with X/XO (500 M/5 milliunits/ml) for 30 min before patch clamp recordings. B and C, mean current density (pA/pF)-voltage relationships and activa- tion conductance (G) curves, respectively. Group comparison was made between the untreated control cells (Ctl) and the X/XO-treated cells. *, p  0.05 versus control, unpaired t tests, n  8, for control and n  7 for FIG.9. Oxidative phosphorylation and intracellular levels of X/XO. ROS measured by CM-H DCFDA fluorescence dye. A, laser scan- ning confocal microscopic images of CM-H DCFDA staining reflecting demonstrated two distinct patterns as follows: one localized to the intracellular ROS levels. The numbers in the labels indicate the concentrations in mM. Note the focused staining on the rod-shaped the defined rod-shaped structures, and the other one diffused structures in the cells under normoglycemia and the diffused staining evenly throughout the cytoplasm. The former presumably rep- throughout the cytoplasm in the cells treated with high glucose or resents the physiological production of ROS as a by-product of pyruvate. B, percentage of positively stained cells (mean  S.E.), ob- oxidative phosphorylation in mitochondria, and the latter in- tained from 5 fields of 3 experiments by counting the cells with staining intensity 5 times the background. C, averaged intensity of CM- dicates overproduction of ROS as a result of metabolic stress H DCFDA fluorescence measured from the positively stained cells. and damage to mitochondria. The cells with diffused staining Shown are the data normalized to 5 mM [Glu] .*, p  0.05 versus Glu and with fluorescence intensity 5 times the background were 5mM and , p  0.05 versus Glu 20 mM, unpaired t tests. defined as positive staining, and the number of cells with positive staining was pooled from 5 fields. The intensity of B, the ROS level was significantly lower, as indicated by the staining by the fluorescent probe for ROS was analyzed by smaller number of cells with positive staining and the weaker densitometric scanning using the LSM program, and cells with intensity of staining in individual cells, in the cells pretreated either localized or diffused staining were taken for analysis, with VitE than in non-treated cells, at 20 mM [Glu] . Consis- and the data were normalized to the control (5 mM [Glu] ) o tently, MnTBAP abolished the hyperglycemia-induced ROS values. Under normoglycemia, a majority of cells that was generation; the number of stained cells and the intensity of stained by CM-H DCFDA demonstrated the localized pattern, 2 staining in the presence of MnTBAP were nearly the same as and the diffused staining was sparse. Yet in the cells treated those under normoglycemia. with 20 mM Glu, the number of the cells with positive staining It has been well documented that the X/XO ROS-generating as well as the intensity of staining was consistently higher, as system stimulates mainly the generation of O . To test compared with the cells treated with 5 mM Glu (Fig. 9). This whether this is also true in our conditions, the ROS that was high level of ROS production was markedly suppressed in the exogenously generated by X/XO and penetrated cells was meas- cells pretreated with NaCN (2 mM, Fig. 9), an uncoupler of ured, and the effect of MnTBAP was studied at 5 mM [Glu] .As oxidative phosphorylation, indicating that the mitochondrion is displayed in Fig. 11, the ROS level was significantly higher in most likely where the ROS was massively produced. Because the cells treated with X/XO alone, and this increase in ROS we have demonstrated that pyruvate at high concentrations level was prevented in the cells pretreated with MnTBAP. decreased I (Fig. 3A), the ROS level in the cells pretreated HERG DISCUSSION with 20 mM pyruvate was also measured. As shown in Fig. 9, like 20 mM Glu, 20 mM pyruvate also significantly increased the The work described here documents a previously unreported ROS production although to a less extent. Moreover, the gly- role of glucose in regulating the function of HERG K channels. colysis inhibitor 2dG (5 mM) also reduced the ROS level in high Our data revealed that glucose produces two characteristic glucose (20 mM), which is indicated by fewer positively stained effects on the HERG channel function: changes of HERG cur- cells and lower intensity of staining (Fig. 9). The results explain rent (I ) amplitude/density and activation voltage. The HERG why 2dG partly restored the depressed I in 20 mM Glu maximum HERG function operates under physiological [Glu] HERG o (Fig. 3E). (5 mM, normoglycemia), and depressed HERG function occurs We have shown that VitE prevented, and MnTBAP partly with [Glu] 5mM (hypoglycemia) or with [Glu] 5mM (hy- o o reversed, the I depression in hyperglycemia (Fig. 7). To perglycemia); hypoglycemia but not hyperglycemia causes HERG see whether this is indeed attributable to their antioxidant shifts of HERG activation toward hyperpolarizing voltages. actions, effects of VitE and MnTBAP on hyperglycemia-induced The modulation of HERG by glucose is critically mediated by ROS production were also studied. As shown in Fig. 10, A and glucose metabolism (glycolysis and oxidative phosphorylation), 10424 Metabolic Mechanisms of HERG Modulation by Glucose Either Glycolysis- or Oxidative Phosphorylation-derived ATP Is Sufficient for Maintaining the Normal HERG Function—The end point of glucose metabolism is the generation of high en- ergy phosphates for maintaining cellular functions, and deple- tion of ATP could impair cellular processes dependent on high energy phosphates. One of the major findings of this study is that the normal HERG function critically relies on the level of intracellular ATP; depletion of intracellular ATP impairs HERG function to an extent similar to what severe hypoglyce- mia (0 mM [Glu] ) does (see Figs. 1, 2, and 5). Complete inhi- bition of glucose metabolism by 2dG substitution for glucose reproduces the effects of hypoglycemia or ATP depletion on I . Yet, neither inhibition of glycolysis alone by 2dG sub- HERG stitution of Glu with a supply of pyruvate to sustain oxidative phosphorylation nor inhibition of oxidative phosphorylation alone by NaCN in the presence of 5 mM Glu to maintain glycolysis is able to cause depression of HERG function. The results imply that the ATP generated by glucose metabolism plays an important role in maintaining HERG function, and FIG. 10. Effects of antioxidants vitamin E (100 M) and SOD either the glycolytic or oxidative ATP is adequate for the mimetic (MnTBAP, 10 M) on intracellular ROS levels under hyperglycemia ([Glu]  20 mM). A, confocal microscopic images of regulation. CM-H DCFDA staining showing decreases in the intracellular ROS ATP synthesis and utilization are subcellularly compartmen- levels in the cells treated with VitE or MnTBAP. B, percentage of the talized; glycolysis-derived ATP primarily regulates membrane positively stained cells (mean  S.E.), obtained from 5 fields of 3 proteins because the glycolytic pathway is associated with sar- experiments by counting the cells with staining intensity 5 times the background. C, averaged intensity of CM-H DCFDA fluorescence meas- 2 colemma (31), whereas oxidative phosphorylation-derived ATP ured from the positively stained cells. Shown are the data normalized to preferentially supports cytosolic processes because oxidative the control (Ctl) cells without antioxidant treatment. , p  0.05 versus ATP is generated within the mitochondria and subsequently Glu 20 mM, unpaired t tests. transported to the cytoplasm (32). Regulation of cardiac ATP- sensitive K channel (K ) (34) and L-type Ca channel (I ) ATP Ca (33) by intracellular ATP has been well documented by some previous studies. It was found that both K and I were ATP Ca preferentially regulated by glycolytic ATP (33, 34). Glycolysis-derived ATP May Be Responsible for Maintaining the Normal Voltage-dependent Activation of I —Although HERG as mentioned above, either glycolytic or oxidative ATP is suf- ficient for maintaining the normal HERG current amplitude/ density, only glycolysis-derived ATP seems to affect the steady- state voltage-dependent activation property of HERG channels. This notion is supported by several lines of evidence from our experiments. 1) Hypoglycemia (0 or 1 mM [Glu] ), a situation with inadequate ATP production but not hyperglyce- mia (10 or 20 mM [Glu] ), causes negative shifts of I-V relation- ships and voltage-dependent activation curves. 2) Inhibition of glycolysis (Fig. 3), but not oxidative phosphorylation (Fig. 4), abolishes the negative shifts of the HERG activation. 3) When glycolysis is inhibited by 2dG, preservation of oxidative phos- phorylation by addition of pyruvate fails to prevent the nega- FIG. 11. Effects of the superoxide generating system X/XO on tive shift caused by hypoglycemia (see Fig. 3). 4) Depletion of the intracellular ROS levels under normoglycemia ([Glu]  5 mM). A, confocal microscopic images of CM-H DCFDA staining showing intracellular ATP reproduces negative shifts similar to those the increases in the intracellular ROS levels by X/XO (0.5 mM/5 milli- seen with hypoglycemia. Similar dependence of glycolytic ATP units/ml) and the abrogation of ROS increase by the SOD mimetic regulation of K and I has been documented (33–34). ATP Ca MnTBAP (10 M). B, percentage of the positively stained cells (mean Role of ATP in Maintaining HERG Function Is Most Likely S.E.), obtained from 5 fields of 3 experiments by counting the cells with intensity of staining 5 times the background. C, averaged intensity of Due to the Phosphorylation-dependent Mechanisms—Two al- staining measured from the positively stained cells. Shown are the data ternative mechanisms could account for intracellular ATP reg- normalized to the control (Ctl) cells without X/XO treatment. *, p  0.05 ulation of ion channels: ATP acts as a substrate for phospho- versus Glu5mM and  p  0.05 versus Glu5mM  X/XO, unpaired t rylation of channel proteins by protein kinase which requires tests. ATP hydrolysis, and ATP interacts with the nucleotide binding domains of channel proteins to produce allosteric regulation and ATP and ROS are crucial in defining the HERG function not requiring ATP hydrolysis. The latter mechanism has been with changing extracellular glucose levels. shown to operate for K and I regulation (33, 35). In our Depression of HERG Function in Hypoglycemia Likely Re- ATP Ca case, neither the non-hydrolysable analogue of ATP AMP-PCP sults from Underproduction of ATP—In our study, lower glu- cose levels and inhibition of glucose metabolism both produced nor GTP prevented the I run-down caused by ATP deple- HERG tion (Fig. 6); instead substitution of AMP-PCP or GTP for ATP similar suppression of HERG function as reflected by the sub- stantial diminishment of HERG current (I ), pointing to a in the internal solution produced nearly identical effects as HERG seen with ATP depletion alone. The results suggest that the requirement of glucose metabolism for HERG modulation by glucose. Our data allowed us to reach the following conclusions. HERG regulation by ATP under our experimental conditions is Metabolic Mechanisms of HERG Modulation by Glucose 10425 phosphorylation-dependent requiring ATP hydrolysis. In other were mainly produced via the oxidative phosphorylation in words, ATP serves as a substrate for phosphorylation of HERG mitochondria in our cells. channels by protein kinases. Indeed, we have recently found It has been reported that the ROS, which generate highly that the normal HERG function requires basal activity of pro- reactive hydroxyl group (OH ), such as H O or FeSO /ascorbic 2 2 4 tein kinase B and inhibition of protein kinase B markedly acid (an oxidative stimulus analogous to H O ), increased 2 2 suppresses I and shifts HERG activation along the voltage I at negative potentials by shifting the HERG activation to HERG HERG axis toward more negative potentials (36). These results are in more negative voltages (45– 46). These results are opposite to good agreement with the HERG regulation by ATP. Studies are our observations. One explanation is that the ROS generated currently undertaken to clarify the link between ATP and under our experimental conditions may be different from the protein kinase B modulation of HERG channels. OH -generating system (H O or FeSO /ascorbic acid). As al- 2 2 4 Depression of HERG Function in Hyperglycemia Results ready mentioned, previous studies have confirmed that the from Overproduction of ROS—The consequence of the physio- ROS induced by hyperglycemia is mainly of O . Here, we also logical role in oxidative phosphorylation is the generation of showed that the SOD mimetic MnTBAP reduced the hypergly- ROS as by-products of the consumption of molecular oxygen in cemia-induced ROS overproduction (Fig. 10), and the O -gen- the electron transport chain (23). Physiologically, these ROS erating system X/XO produced ROS which were also abolished are mostly trapped within mitochondria and rapidly scavenged by MnTBAP, evidence for O as a major ROS generated in our cells. Consistently, depressive effects of hyperglycemia or X/XO by endogenous antioxidants like SOD, catalase, glutathione, etc. Yet under metabolic stress, ROS can be overproduced and on I were significantly weakened by MnTBAP. Indeed, it HERG can cause damages to mitochondria. Consequently, the ROS has been reported that the O generated by high glucose (23 may diffuse throughout the cytoplasm and cause further dele- mM) or by X/XO in rat small coronary arteries impairs voltage- terious effects on other cellular processes. Abnormally high gated K (K ) current (39, 47); reducing the current density by concentrations of glucose can enhance ROS damage at least in around 60%, which was partially restored by SOD and catalase. three different ways. First, it has been known that high glucose All together, we believe that different ROS might have differ- (25 mM) evoked ROS generation, which was blocked by antioxi- ent effects on I ;OH enhances, whereas O depresses, HERG 2 dants, inhibitors of mitochondrial electron transport chain I . HERG complex, inhibitors of glycolysis-derived pyruvate transport Moreover, it has been shown that excessive ROS inhibits into mitochondria, uncouplers of oxidative phosphorylation, glycolysis and the subsequent glycolytic ATP production and SOD mimetics, catalase, etc. (19). Superoxide anion (O )is even depletes intracellular ATP levels in isolated perfused found to be the major ROS produced under hyperglycemia hearts (48 –50). This fact together with our data suggests that (37– 42), and increases in ROS can be prevented by SOD. Sec- besides the potential direct modulation of I by ROS, ATP HERG ond, glucose itself can auto-oxidize to form ROS including O , reduction potentially caused by ROS may also contribute to the OH , and H O (43). Finally, acute elevations in glucose also I depression under hyperglycemia. This provides an alter- 2 2 HERG depress natural antioxidant defenses. It has been found that native explanation for the depressive effects of the ROS over- incubation of purified bovine CuZn-SOD with 10 to 100 mM production overcoming the enhancing effects of the expected glucose reduces the enzyme activity by 60% (44). ATP increase. Elevated glucose or pyruvate level is expected to enhance In our study, the effects of pyruvate and X/XO were smaller oxidative phosphorylation and produce more ATP molecules to than those of hyperglycemia. This may be because the O support HERG function or increase I . However, our obser- generated by hyperglycemia occurs inside the cell but pyruvate HERG vations are contrary to this expectation. Our results showed does not readily penetrate cells, and the effect of pyruvate that hyperglycemia or excessive pyruvate markedly depressed observed in the present study may underestimate the true role the HERG function. A reasonable explanation for this is that of oxidative phosphorylation in I modulation. Likewise, HERG the ROS produced under hyperglycemia counteract the effects the O generated by X/XO was primarily extracellular with of ATP, and the net outcome is a balance between enhancing subsequent entry into the cell which could also underestimate effects of ATP and suppressing effects of ROS. Evidently, under the effects of O . our experimental conditions, the effects of increased ROS over- It was shown that within minutes of exposure to dihydroxy- write the effects of increased ATP, resulting in suppression of fumaric acid or xanthine plus xanthine oxidase, both of which I . This notion is supported by the following evidence. produce the superoxide anion, action potential duration was HERG First, the depression of I induced by hyperglycemia was prolonged in canine myocytes, and this effect was followed by HERG prevented or reversed by the antioxidants vitamin E and the appearance of early after depolarization (51). X/XO caused MnTBAP (SOD mimetic). Second, inhibition of the glycolysis a 30% increase in action potential duration in superfused pap- and thereby the subsequent oxidative phosphorylation by 2dG illary muscle or small strips of right ventricular walls of guinea partially reversed the depressed I under hyperglycemia pig hearts (52). However, whether the action potential duration HERG (Fig. 3). Weakened oxidative phosphorylation due to inhibition prolongation was associated with inhibition of delayed rectifier of the glycolysis would reduce both ATP and ROS productions, K current (I ) is unknown. Our study provides a potential Kr but the net result was an increase in I , indicating again explanation for these observations. HERG that in our experimental conditions ROS overweighs ATP in Impairment of HERG Function Might Contribute to Q-T Pro- terms of their effects on I . This is in agreement with the longation Caused by Hypoglycemia and Hyperglycemia—Heart HERG notion that the suppressing effects of ROS overproduction over- disease is a leading cause of death in diabetic patients. In whelm the effects of ATP increase. Moreover, 2dG also can patients with diabetes a prolongation of the Q-T interval has compete with glucose for access to glucose transporters and been associated with an increased risk of sudden cardiac death thus decreases glucose uptake which in turn can result in (2) due to the occurrence of lethal ventricular arrhythmias, reduction of ROS production in the cells. Finally, our data particularly Torsade de pointes following bradycardia (1). Sev- indeed demonstrated the ability of high glucose to stimulate an eral cardiovascular pathological consequences of diabetes such overproduction of ROS (see Fig. 9). The fact that a high con- as hypertension and arteriosclerosis affect the heart to varying centration of pyruvate mimicked, whereas VitE or NaCN degrees. Hyperglycemia, as a consequence of diabetes and an abrogated, the ROS overproduction suggests that the ROS independent risk factor, also can directly cause cardiac dam- 10426 Metabolic Mechanisms of HERG Modulation by Glucose 12. Sanguinetti, M. C. (1999) Ann. N. Y. Acad. Sci. 868, 406 – 413 age. On the other hand, insulin therapy increases the risk of 13. Taglialatela, M., Castaldo, P., and Pannaccione, A. (1998) Biochem. Pharma- hypoglycemia in type 2 diabetic patients; according to the col. 55, 1741–1746 14. Zhou, Z., Gong, Q., Ye, B., Fan, Z., Makielski, J. C., Robertson, G. A., and United Kingdom Prospective Diabetes Study, approximately January, C. T. (1998) Biophys. J. 74, 230 –241 one-third of the insulin-treated patients reported one or more 15. Wang, J., Wang, H., Han, H., Zhang, Y., Yang, B., Nattel, S., and Wang, Z. hypoglycemic episodes per year during the first 3 years (3). (2001) Circulation 104, 2645–2648 16. Wang, H., Yang, B., Zhang, Y., Han, H., Wang, J., Shi, H., and Wang, Z. (2001) Hypoglycemia is presumed to be the cause of death in about 3% J. Biol. Chem. 276, 40811– 40816 of insulin-treated diabetic patients (11). Intriguingly, it is well 17. Splawski, I., Shen, J., Katherine, W., Timothy, G., Vincent, M., Lehmann, M. H., and Keating, M. T. (1998) Genomics 51, 86 –97 recognized that both hyperglycemia and hypoglycemia can 18. Ludwig, B., Bender, E., Arnold, S., Huttemann, M., Lee, I., and Kadenbach, B. cause prolongation of Q-T interval. In type 2 diabetes, the (2001) Chembiochem. 2, 392– 403 prevalence of Q-T prolongation is as high as 26%, and Q-T 19. Hsieh, T. J., Zhang, S. L., Filep, J. G., Tang, S. S., Ingelfinger, J. R., and Chan, J. S. (2002) Endocrinology 143, 2975–2985 prolongation during experimentally induced and spontane- 20. Wallace, D. C. (2002) Methods Mol. Biol. 197, 3–54 ously occurring hypoglycemia or diabetic hyperglycemia has 21. Siraki, A. G., Pourahmad, J., Chan, T. S., Khan, S., and O’Brien, P. J. (2002) also been shown to occur in healthy subjects and in diabetic Free Radic. Biol. Med. 32, 2–10 22. Kirkinezos, I. G., and Moraes, C. T. (2001) Semin. Cell Dev. Biol. 12, 449 – 457 patients with increased risk of malignant ventricular arrhyth- 23. Wallace, D. C. (2001) Novartis. Found. Symp. 235, 247–263 mias (4 –11). Yet the potential ionic mechanisms by which 24. Koo, J. R., Ni, Z., Oviesi, F., and Vaziri, N. D. (2002) Clin. Exp. Hypertens. 24, 333–344 hypoglycemia and hyperglycemia cause Q-T prolongation re- 25. Marfella, R., Quagliaro, L., Nappo, F., Ceriello, A., and Giugliano, D. (2001) mained poorly understood. Studies on glucose modulation of J. Clin. Invest. 108, 635– 636 cardiac ion channels are sparse and have been mostly limited 26. Srivastava, A. K. (2002) Int. J. Mol. Med. 9, 85– 89 27. Li, P. A., Liu, G. J., He, Q. P., Floyd, R. A., and Siesjo, B. K. (1999) Free Radic. to ATP-sensitive K current (K ). On the contrary to I , ATP HERG Biol. Med. 27, 1033–1040 K is closed with increased intracellular ATP levels (34). One ATP 28. van Dam, P. S., van Asbeck, B. S., Bravenboer, B., van Oirschot, J. F., Gispen, W. H., and Marx, J. J. (1998) Free Radic. Biol. Med. 24, 18 –26 study reported by Xu et al. (53) demonstrated that hyperglyce- 29. Giardino, I., Edelstein, D., and Brownlee, M. (1996) J. Clin. Invest. 97, mia (18 mM [Glu] ) decreased the density of transient outward 1422–1428 K current but did not alter the inward rectifier K current in 30. Takigawa, T., Yasuda, H., Terada, M., Maeda, K., Haneda, M., Kashiwagi, A., Kitasato, H., and Kikkawa, R. (2000) Neuroreport 11, 2547–2551 rat ventricular myocytes that do not express delayed rectifier 31. Hazen, S. L., Wolf, M. J., Ford, D. A., and Gross, R. W. (1994) FEBS Lett. 339, K current (I ), the physiological counterpart of I .In Kr HERG 213–216 32. Weiss, J. N., and Hiltbrand, B. (1985) J. Clin. Invest. 75, 436 – 447 arterial smooth muscles, high glucose diminished shaker-type 33. Losito, V. A., Tsushima, R. G., Diaz, R. J., Wilson, G. J., and Backx, P. H. delayed rectifier K current (39). In rat myelinated nerve fi- (1998) J. Physiol. (Lond.) 511, 67–78 bers, 30 mM glucose increased Ca -activated K current (30). 34. Weiss, J. N., and Lamo, S. T. (1989) J. Gen. Physiol. 94, 911–935 35. Yazawa, K., Kameyama, A., Yasui, K., Li, J. M., and Kameyama, M. (1997) Whereas none of the data from these studies could fully account Pfluegers Arch. 433, 557–562 for the Q-T prolongation, particularly the Q-T prolongation 36. Zhang, Y., Wang, H., Wang, J., Han, H., Nattel, S., and Wang, Z. (2003) FEBS induced by hypoglycemia, our study provides a plausible, or at Lett. 534, 125–132 37. Nishikawa, T., Edelstein, D., Du, X. L., Yamagishi, S., Matsumura, T., least an alternative, explanation; HERG K channel may be a Kaneda, Y., Yorek, M. A., Beebe, D., Oates, P. J., Hammes, H. P., Giardino, mechanistic link for the Q-T prolongation induced by both I., and Brownlee, M. (2000) Nature 404, 787–790 38. Du, X. L., Edelstein, D., Rossetti, L., Fantus, I. G., Goldberg, H., Ziyadeh, F., hyperglycemia and hypoglycemia. Yet one should keep it in Wu, J., and Brownlee, M. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, mind that HERG may be only one of the multiple factors 12222–12226 contributing to the Q-T prolongation in hypoglycemia and 39. Liu, Y., Terata, K., Rusch, N. J., and Gutterman, D. D. (2001) Circ. Res. 89, 146 –152 hyperglycemia. 40. Graier, W. F., Posch, K., Wascher, T. C., and Kostner, G. M. (1997) Horm. Metab. Res. 29, 622– 629 Acknowledgments—We thank XiaoFan Yang for excellent technical 41. Graier, W. F., Posch, K., Fleischhacker, E., Wascher, T. C., and Kostner, G. M. support and Louis R. Villeneuve for assistance with the confocal micro- (1999) Diabetes Res. Clin. Pract. 45, 153–160 scopic examinations. 42. Esposito, K., Marfella, R., and Giugliano, D. (2002) J. Endocrinol. Invest. 25, 485– 488 REFERENCES 43. Wolff, S. P., and Dean, R. T. (1987) Biochem. J. 245, 243–250 44. Oda, A., Bannai, C., Yamaoka, T., Katori, T., Matsushima, T., and Yamashita, 1. Abo, K., Ishida, Y., Yoshida, R., Hozumi, T., Ueno, H., Shiotani, H., K. (1994) Horm. Metab. Res. 26, 1– 4 Matsunaga, K., and Kazumi, T. (1996) Diabetes Care 19, 1010 45. Be ´ rube ´ , J., Caouette, D., and Daleau, P. (2001) J. Pharmacol. Exp. Ther. 297, 2. Kahn, J. K., Sisson, J. C., and Vinik, A. I. (1987) J. Clin. Endocrinol. Metab. 64, 96 –102 751–754 46. Taglialatela, M., Castaldo, P., Iossa, S., Pannaccione, A., Fresi, A., Ficker, E., 3. United Kingdom Prospective Diabetes Study (1995) Br. Med. J. 310, 83– 88 and Annunziato, L. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 11698 –11703 4. Marfella, R., Rossi, F., and Giugliano, D. (2001) Diabetes Nutr. Metab. 14, 47. Liu, Y., and Gutterman, D. D. (2002) Clin. Exp. Pharmacol. Physiol. 29, 63– 65 305–311 5. Chelliah, Y. R. (2000) Anaesth. Intensive Care 28, 698 –700 48. Ytrehus, K., Myklebust, R., and Mjøs, O. D. (1986) Cardiovasc. Res. 20, 6. Veglio, M., Borra, M., Stevens, L. K., Fuller, J. H., and Perin, P. C. (1999) Diabetologia 42, 68 –75 597– 603 49. Miki, S., Ashraf, M., Salka, S., and Sperelakis, N. (1988) J. Mol. Cell. Cardiol. 7. Marques, J. L., George, E., Peacey, S. R., Harris, N. D., Macdonald, I. A., Cochrane, T., and Heller, S. R. (1997) Diabet. Med. 14, 648 – 654 20, 1009 –1024 50. Patane, G., Anello, M., Piro, S., Vigneri, R., Purrello, F., and Rabuazzo, A. M. 8. Landstedt-Hallin, L., Englund, A., Adamson, U., and Lins, P.-E. (1999) J. In- tern. Med. 246, 299 –307 (2002) Diabetes 51, 2749 –2756 51. Barrington, P. L., Meier, C. F., Jr., and Weglicki, W. B. (1988) J. Mol. Cell. 9. Lindstrom, T., Jorfeldt, L., Tegler, L., and Arnqvist, H. J. (1992) Diabet. Med. 9, 536 –541 Cardiol. 20, 1163–1178 52. Aiello, E. A., Jabr, R. I., and Cole, W. C. (1995) Circ. Res. 77, 153–162 10. Shimada, R., Nakashima, T., Nunoi, K., Kohno, Y., Takeshita, A., and Omae, T. (1984) Arch. Intern. Med. 144, 1068 –1069 53. Xu, Z., Patel, K. P., and Rozanski, G. J. (1998) Am. J. Physiol. 271, 11. Eckert, B., and Agardh, C.-D. (1998) Clin. Physiol. (Oxf.) 18, 570 –575 H2190 –H2196 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Impairment of Human Ether-à-Go-Go-related Gene (HERG) K+ Channel Function by Hypoglycemia and Hyperglycemia

Zhang, Yiqiang; Han, Hong; Wang, Jingxiong; Wang, Huizhen; Yang, Baofeng; Wang, Zhiguo
Journal of Biological Chemistry – Mar 1, 2003
Download PDF
10 pages

Loading next page...
 
/lp/unpaywall/impairment-of-human-ether-go-go-related-gene-herg-k-channel-function-pqRabe38aa

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.m211044200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 278, No. 12, Issue of March 21, pp. 10417–10426, 2003 © 2003 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Impairment of Human Ether-a ` -Go-Go-related Gene (HERG) K Channel Function by Hypoglycemia and Hyperglycemia SIMILAR PHENOTYPES BUT DIFFERENT MECHANISMS* Received for publication, October 29, 2002, and in revised form, December 20, 2002 Published, JBC Papers in Press, January 16, 2003, DOI 10.1074/jbc.M211044200 Yiqiang Zhang‡§, Hong Han‡, Jingxiong Wang‡§, Huizhen Wang‡¶, Baofeng Yang, and Zhiguo Wang‡§** From the ‡Research Center, Montreal Heart Institute, Montreal, Quebec H1T 1C8, the §Department of Medicine, University of Montreal, Montreal, Quebec H3C 3J7, Canada and the Department of Pharmacology, Harbin Medical University, Harbin, Heilongjiang 150086, Peoples Republic of China Hyperglycemia and hypoglycemia both can cause pro- Glucose, the primary end product of the digestion of glycogen, longation of the Q-T interval and ventricular arrhyth- is essential for maintaining life activities in organisms. As a mias. Here we studied modulation of human ether-a ` -go- major source of metabolic fuel, degradation of glucose via glycol- go-related gene (HERG) K channel, the major molecular ysis and subsequent oxidative phosphorylation generates high component of delayed rectifier K current responsible for energy phosphates to power the biological processes in the cell. cardiac repolarization, by glucose in HEK293 cells using Yet, through an exquisitely complex network of control mecha- whole-cell patch clamp techniques. We found that both nisms, the rate of glucose metabolism is only as great as needed hyperglycemia (extracellular glucose concentration by the organisms. Moreover, glucose also has other regulatory [Glu]  10 or 20 mM) and hypoglycemia ([Glu]  2.5, 1, or o o effects on many cellular functions. Either inadequate or excessive 0mM) impaired HERG function by reducing HERG cur- glucose can be harmful to the living system. Therefore, the blood rent (I ) density, as compared with normoglycemia HERG glucose level is dynamically controlled. However, under patho- ([Glu]  5mM). Complete inhibition of glucose metabo- logical conditions like diabetes, glucose cannot be efficiently uti- lism (glycolysis and oxidative phosphorylation) by 2-de- lized, and the blood glucose level rises. When the blood level of oxy-D-glucose mimicked the effects of hypoglycemia, but inhibition of glycolysis or oxidative phosphorylation glucose is maintained higher than 7 mM, it is considered as alone did not cause I depression. Depletion of intra- hyperglycemia. Diabetes therapy, on the other hand, can lead to HERG cellular ATP mimicked the effects of hypoglycemia, and an overly low level of blood glucose, which is referred to as replacement of ATP by GTP or non-hydrolysable ATP hypoglycemia when the level falls below 3 mM. failed to prevent the effects. Inhibition of oxidative phos- Either hypoglycemia or hyperglycemia can have deleterious phorylation by NaCN or application of antioxidants vita- effects on the cells. One common feature of electrophysiological min E or superoxide dismutase mimetic (Mn(III) tet- alterations caused by both hypoglycemia and hyperglycemia in rakis(4-benzoic acid) porphyrin chloride) abrogated and the heart is prolongation of Q-T interval and the associated incubation with xanthine/xanthine oxidase mimicked the ventricular arrhythmias that are presumably responsible for effects of hyperglycemia. Hyperglycemia or xanthine/ sudden cardiac death in diabetic patients (1–10). However, the xanthine oxidase markedly increased intracellular levels ionic mechanisms by which hyperglycemia and hypoglycemia of reactive oxygen species, as measured by 5-(and-6)- prolong Q-T interval remained unclear, which is at least a part of chloromethyl-2,7-dichlorodihydrofluorescein diacetate (CM-H DCFDA) fluorescence dye, and this increase was the reasons why diabetic patients die of mainly cardiac prevented by NaCN, vitamin E, or Mn(III) tetrakis(4- complications. benzoic acid) porphyrin chloride. We conclude that ATP, The human either-a ` -go-go-related gene (HERG) encodes the derived from either glycolysis or oxidative phosphoryla- rapid component of delayed rectifier K current in the heart, tion, is critical for normal HERG function; depression of which is the major repolarizing current in the plateau voltage I in hypoglycemia results from underproduction of HERG range of cardiac action potentials. HERG K channels are sus- ATP and in hyperglycemia from overproduction of reac- ceptible to genetic defects and environmental cues, with the tive oxygen species. Impairment of HERG function might consequence being depression of HERG function in most situa- contribute to Q-T prolongation caused by hypoglycemia tions (9). Indeed, most of the cases of long Q-T syndrome are and hyperglycemia. ascribed to dysfunction of HERG channels, particularly that in- duced by therapeutic drugs (13). It is conceivable that HERG alteration might also be involved in the Q-T prolongation induced * This work was supported in part by the Canadian Institute of by hyperglycemia and hypoglycemia. This thought prompted us Health Research, the Heart and Stroke Foundation of Quebec, and the to carry out a series of experiments to study the effects of glucose Fonds de la Recherche de l’Institut de Cardiologie de Montreal (to on HERG K channels and the potential mechanisms. Z. W.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ¶ Research fellow of the Canadian Institute of Health Research. The abbreviations used are: HERG, human ether-a ` -go-go related ** Research scholar of the Fonds de Recherche en Sante de Quebec. gene; ROS, reactive oxygen species; X, xanthine; XO, xanthine oxidase; To whom correspondence should be addressed: Research Center, Mon- 2dG, deoxy-D-glucose; MnTBAP, Mn(III) tetrakis(4-benzoic acid) por- treal Heart Institute, 5000 Belanger East, Montreal, Quebec H1T 1C8, phyrin chloride; AMP-PCP, ,-methyleneadenosine 5-triphosphate; Canada. Tel.: 514-376-3330; Fax: 514-376-4452; E-mail: wangz@icm. VitE, vitamin E; SOD, superoxide dismutase; CM-H DCFDA, 5-(and- umontreal.ca. 6)-chloromethyl-2,7-dichlorodihydrofluorescein diacetate. This paper is available on line at http://www.jbc.org 10417 This is an Open Access article under the CC BY license. 10418 Metabolic Mechanisms of HERG Modulation by Glucose EXPERIMENTAL PROCEDURES lution containing a given concentration of Glu for 30 min prior Cell Culture—HEK293 cells stably expressing HERG (a kind gift to patch clamp recordings. In addition, recordings were per- from Drs. Zhou and January) (14) were grown in Dulbecco’s modified formed immediately after formation of whole-cell configuration Eagle’s medium supplemented with 10% heat-inactivated fetal bovine and adjustments of capacitance and series resistance compen- serum, 200 M G418, 100 units/ml penicillin, and 100 g/ml strepto- sation, and all recordings were made complete within 3 min. In mycin. The cells subcultured to 85% confluency were harvested by this manner, there was minimal dialysis through the recording trypsinization and stored in the Tyrode’s solution containing 0.5% bo- pipette and thereby minimal current run-down (time-depend- vine serum albumin at 4 °C (12). Electrophysiological recordings were conducted within 10 h of storage. ent current decay), and the data best reflect the effects of Glu Whole-cell Patch Clamp Recording—Patch clamp techniques have on I in cells with intact intracellular contents. In addition, HERG been described in detail elsewhere (15–16). Currents were recorded such an experimental design also allowed us to study the effect with whole-cell voltage clamp with an Axopatch-200B amplifier (Axon of Glu on I under conditions devoid of influence from HERG Instruments). Borosilicate glass electrodes had tip resistances of 1–3 exogenous ATP included in the pipette, which is an important megohms when filled with the internal solution containing (mM) 130 KCl, 1 MgCl , 5 Mg-ATP, 10 EGTA, and 10 HEPES (pH 7.3). The issue to be described later. I was elicited by 2.5-s depolar- 2 HERG extracellular (Tyrode’s) solution contained (mM) 136 NaCl, 5.4 KCl, 1 izing steps from 60 to 40 mV to record the activating cur- CaCl , 1 MgCl , 10 HEPES (pH 7.4), and glucose at concentrations as 2 2 rent, followed by a repolarizing pulse to 50 mV for another specified. Experiments were conducted at 36  1 °C. Junction poten- 2.5 s to observe the deactivating tail current, before being tials were zeroed before formation of the membrane-pipette seal. Series returned to a holding potential of 80 mV. The results are resistance and capacitance were compensated, and leak currents were illustrated in Fig. 1 with both representative raw data and subtracted. Pharmacological Probes—D-Glucose (Glu), 2-deoxy-D-glucose (2dG), analyzed mean data. Glucose produced two characteristic al- sodium cyanide (NaCN), pyruvate, ATP, GTP, AMP-PCP (non-hydro- terations of HERG channel functions as follows: changes of lysable analogue of ATP), xanthine (X), xanthine oxidase (XO), and I amplitude and density and shifts of I-V relationships HERG vitamin E (VitE) were all purchased from Sigma. Xanthine was pre- and activation curves. pared in 2 N NaOH and diluted in the Tyrode’s solution 800 times with Comparison of I recorded at varying extracellular con- HERG the pH adjusted to 7.4 with HCl. Xanthine oxidase was added to the centrations of Glu ([Glu]  0, 1, 2.5, 5, 10, or 20 mM) consis- xanthine preparation to form the X/XO-reactive oxygen species (ROS) generating system. VitE was dissolved in ethanol and diluted 1000 tently showed that I density was maximal at a physiolog- HERG times to reach the final concentration. Pyruvate in liquid was diluted ical [Glu] (5 mM), and it was depressed at [Glu] below or above o o into the Tyrode’s solution, and pH was adjusted to 7.4 with NaOH 5mM (Fig. 1, A–C). In other words, under normoglycemia before use. All other compounds were directly dissolved into the patch ([Glu]  5mM) the HERG K channel operates at its maxi- clamp recording solutions as specified. Mn(III) tetrakis(4-benzoic acid) mum function level, whereas under hypoglycemia ([Glu]  0, porphyrin chloride (MnTBAP) purchased from Calbiochem was dis- 1, or 2.5 mM) or hyperglycemia ([Glu]  10 or 20 mM), the solved in 1 N NaOH and diluted by 5000 times to reach the desired experimental concentrations. All compounds and reagents were pre- HERG channel function is impaired, and the degree of func- pared fresh before the experiments. tional impairment is proportional to the degrees of hypoglyce- Intracellular Reactive Oxygen Species (ROS) Measurement—5- mia or hyperglycemia (Fig. 1D). For example, with 5 mM [Glu] , (and-6)-chloromethyl-2,7-dichlorodihydrofluorescein diacetate (CM- the I current density was 88.5  7.4 pA/pF (n  25) at a HERG H DCFDA) (Molecular Probes) is a ROS-sensitive probe that can be test potential of 0 mV, whereas with 0 and 20 mM, the values used to detect oxidative activity in living cells. It passively diffuses into cells, where its acetate groups are cleaved by intracellular esterases, were 35.4  3.0 pA/pF (n  22, p  0.05 versus 5mM [Glu] ) and releasing the corresponding dichlorodihydrofluorescein derivative. Its 35.8  4.9 (n  15, p  0.05 versus 5mM [Glu] ), respectively, thiol-reactive chloromethyl group reacts with intracellular glutathione approximately a 2-fold difference. The HERG tail current den- and other thiols. Subsequent oxidation yields a fluorescent adduct that sity at 10 mV was also significantly depressed by hypoglyce- is trapped inside the cell. When it is excited at 480 nm, its emissions at mia (0, 1, and 2.5 mM [Glu] ) or hyperglycemia (10 and 20 mM 505–530 nm can be captured. CM-H DCFDA is prepared in dimethyl [Glu] ) (data not shown). sulfoxide immediately prior to loading. Glass coverslips were coated with laminin and placed in the wells of a 12-well culture plate before the Also noticeable is the hypolarization shifts of I I-V re- HERG cells were seeded into the well in a density of 5.0  10 /well. After lationships (Fig. 1E) and the activation curve by hypoglycemia overnight incubation, the cells were washed with pre-warmed (37 °C) (Fig. 1F). For example, the half-activation voltage (V1 ) was ⁄2 phosphate-buffered saline once and then incubated in the Tyrode’s 25.4  2.7 mV (n  19) for 5 mM [Glu] and 34.3  3.8 mV solution containing glucose of varying concentrations or exogenous su- (n  22) for 0 mM [Glu] (p  0.05, unpaired t test), which peroxide-generating system (xanthine/xanthine oxidase) or the re- accounted for around 10 mV shift of the steady-state voltage- agents as to be otherwise specified, together with the fluorescence dye CM-H DCFDA (10 M). After 30 min of incubation, the coverslips were 2 dependent activation of I . This negative shift resulted in a HERG washed with pre-warmed phosphate-buffered saline twice before being crossover of the I-V curves between normoglycemia and hypo- mounted to the glass slides with anti-fading mounting medium and glycemia. Only slight (4 mV) hyperpolarization shift of acti- were examined immediately under a laser scanning confocal microscope vation was seen with hyperglycemia (20 mM [Glu] ). (Zeiss LSM 510). The percentage of positively stained cells and the Several potential mechanisms could explain the observed fluorescence intensity of staining were determined by densitometric scanning with LSM software (Zeiss). effects of [Glu] on I . First, there was a possibility that the o HERG Data Analysis—Group data are expressed as mean  S.E. Compar- effects were a consequence of alterations of extracellular osmo- isons among groups were made by analysis of variance (F-test), and larity with varying [Glu] . Second, Glu might act directly on Bonferroni-adjusted t tests were used for multiple group comparisons, HERG proteins to modify the channel function. Finally, Glu and paired or unpaired t test was used, as appropriate, for single metabolism, which generates ATP as well as other metabolic comparisons. A two-tailed p  0.05 was taken to indicate a statistically intermediates, also has the potential to modulate I . The significant difference. Nonlinear least square curve fitting was per- HERG formed with CLAMPFIT in pCLAMP 8.0 or GraphPad Prism. following experiments were performed to clarify these issues. To test the first possibility, we performed experiments in RESULTS which cells were first superfused with a given concentration of Effects of Glucose on I —To study the effects of varying Glu (1, 5, or 20 mM) for 30 min, followed by I recording HERG HERG concentrations of glucose (Glu, ranging from 0 to 20 mM)on within 3 min bathing in the Tyrode’s solution containing 10 mM I , our experiments were designed for group comparisons. Glu. Under such conditions, I demonstrated the same HERG HERG pattern of changes as described above; cells pre-exposed to 1 or For each experiment, HERG-expressing HEK293 cells were divided into six groups each superfused with the Tyrode’s so- 20 mM Glu had markedly smaller I density than those HERG Metabolic Mechanisms of HERG Modulation by Glucose 10419 FIG.1. Effects of glucose on HERG K current (I ) stably expressed in HEK293 cells. A, typical examples of I traces recorded HERG HERG in the Tyrode’s solution containing 0, 5, 10, or 20 mMD-glucose (Glu) with the voltage protocol shown in the inset and presented as current density (pA/pF) for better group comparisons. The same voltage protocol is applied for the I recordings shown in the subsequent figures, except as HERG otherwise indicated. B, current density-voltage relationships of I . The steady-state step I measured at the end of 2.5-s pulses was HERG HERG normalized to the capacitance of the respective cells and plotted as a function of test potentials. Shown are data averaged from 22, 9, 11, 25, 24, and 15 cells for 0, 1, 2.5, 5, 10, and 20 mM Glu, respectively. *, p  0.05 versus 5mM [Glu] . C, ratio of the step I recorded with varying [Glu] o HERG o over the I with 5 mM [Glu] as a function of test potentials. D, ratio of the step I recorded with varying [Glu] over the I with 5 mM HERG o HERG o HERG [Glu] as a function of glucose concentrations. Shown are the data obtained at test potentials of 40, 10, and 20 mV. E, normalized I-V relationships obtained by dividing the current amplitude at various potentials by the maximum current for each concentration of glucose. For clarity, only the data with 0, 5, and 20 mM [Glu] are shown. Note the negative shifts of the curves with low and high [Glu] , relative to 5 mM [Glu] , o o o along the voltage axis. F, steady-state voltage-dependent activation of I . The activation curves were constructed by plotting the conductance HERG G as a function of potentials. G was calculated by normalizing the tail currents at 50 mV by dividing the amplitude of the tail currents evoked at various antecedent step potentials by that of the tail current at 40 mV. The symbols are mean of experimental data, and the lines represent the Boltzmann fit: G/G  1/{1  exp((V1  V)/k)}, where G represents the maximal conductance at 40 mV, V1 is a half-maximal activation max ⁄2 max ⁄2 voltage, and k is a slope factor. The numbers in the legends represent the extracellular glucose concentrations ([Glu] )inmM. pre-exposed to 5 mM Glu (Fig. 2A), despite that all recordings investigate whether the effects of Glu on I are associated HERG were made under isotonic conditions with 10 mM [Glu] . For with glucose metabolism, we studied the effects of the glycoly- instance, at 0 mV, the I density was 56.1  6.9 pA/pF (n  sis inhibitor 2-deoxy-D-glucose (2dG). Glucose was replaced by HERG 8) with 5 mM [Glu] , whereas with 1 and 20 mM [Glu] , the the non-hydrolysable analogue of glucose 2dG to eliminate the o o values were 32.4  3.2 pA/pF (n  7, p  0.05 versus 5mM glucose metabolism (both glycolysis and subsequent oxidative [Glu] ) and 35.8  2.2 (n  7, p  0.05 versus 5mM [Glu] ), phosphorylation). Cells were superfused with the Glu-free Ty- o o respectively. Consistently, the activation curve of I was rode’s solution containing 5 mM 2dG for 30 min before patch HERG also shifted to more negative potentials by hypoglycemia (Fig. clamp recordings. I recorded under such a condition was HERG 2B). A similar difference of I between 5 and 20 mM [Glu] compared with I recorded with the Tyrode’s solution con- HERG o HERG was also consistently seen when 15 mM cellobiose was added to taining 5 mM Glu. As shown in Fig. 3, 2dG substitution for Glu the Tyrode’s solution containing 5 mM Glu (Fig. 2D). reproduced the two characteristic changes of I as observed HERG The above data indicate that changes of osmolarity is un- under hypoglycemia. First, marked depression of I was HERG likely the mechanism by which glucose modulates I . Apart seen in the presence of 2dG, with I density only 38% that HERG HERG from that, the fact that differences of I among the cells in the presence of 5 mM [Glu] at 0 mV. Second, similar to the HERG o pretreated with three different concentrations of Glu (1, 5, and results with 0 mM [Glu] , the I-V relationship and activation 20 mM) persisted even though I was recorded 10 min after curve were shifted by 10 mV toward hyperpolarizing potentials HERG superfusion with the normal Tyrode’s solution containing 10 by 2dG (28.2  3.4 mV for 5 mM [Glu] and 37.6  4.5 mV mM [Glu] suggests that the effects of glucose on I are for5mM 2dG, p  0.05, n  8 for both groups) (Fig. 3B). o HERG mediated by some intracellular events, and direct interactions Because 2dG is a competitive inhibitor of glycolysis, its effects between glucose and HERG channels do not likely play a major on I in the presence of 5 mM Glu were also investigated. HERG role. Cells were superfused with the Glu-containing Tyrode’s solu- Role of Glycolysis and Oxidative Phosphorylation on Glucose- tion with or without 5 mM 2dG for 30 min before patch clamp induced I Enhancement—Glucose, once being taken up recordings. The I density was 45% smaller in cells HERG HERG into the cell, is metabolized via glycolysis to generate 2 mole- treated with 2dG than in untreated cells (Fig. 3C). For exam- cules of ATP and 2 molecules of pyruvate (a substrate for ple, the I density at 0 mV was 40.2  4.1 pA/pF for cells HERG oxidative phosphorylation), which is further metabolized via treated with 2dG at 5 mM [Glu] (n  14) and was 75.6  6.6 oxidative phosphorylation to produce more ATP molecules. To pA/pF for cells without 2dG treatment (n  14, p  0.05). These 10420 Metabolic Mechanisms of HERG Modulation by Glucose FIG.2. Effects of glucose on I under conditions with cor- HERG rected osmolarity. A and B, mean current density (pA/pF)-voltage relationships and activation conductance (G) curves, respectively (n 7, 9, and 7 for 1, 5, and 20 mM [Glu] , respectively). The cells of different groups were superfused with solutions containing a given concentration of glucose (1, 5, or 20 mM) for 30 min and then switched to the same solution with an identical glucose concentration (10 mM). The I HERG values recorded within 3 min at 10 mM [Glu] were used for analysis. *, p  0.05 versus 5mM [Glu] , unpaired t tests. C and D, depression of I and negative shift of I activation, respectively, in high glu- HERG HERG cose (20 mM [Glu] , n  5), as compared with those in normal glucose (5 mM [Glu] )  15 mM cellobiose (n  5) to balance the extracellular osmolarity. *, p  0.05 versus Glu-5  cellobiose 15, unpaired t-tests. FIG.3. Effects of complete inhibition of glucose metabolism or results indicate the importance of glucose metabolism in main- inhibition of glycolysis on I . A and B, mean current density HERG taining the normal HERG function. Intriguingly, when 2dG (pA/pF)-voltage relationships and activation conductance (G) curves of was added to the hyperglycemic solution containing 20 mM I . Complete inhibition of glucose metabolism was achieved by the HERG addition of 5 mM 2dG to the Glu-free solution, and inhibition of glycol- [Glu] ,I was increased as compared with that measured o HERG ysis alone was achieved by the addition of pyruvate to the 2dG-contain- in the hyperglycemic solution without 2dG (Fig. 3E). In other ing Glu-free solution. The numbers in the legends represent the con- words, 2dG partly reversed the depressed I caused by HERG centrations of Glu, 2dG, or pyruvate in mM. The number of cells was 8 hyperglycemia toward the normal HERG function seen under for the Glu 5 group, 7 for Glu 0  2dG 5 group, 7 for Glu 0  2dG 5 pyruvate 5 group, and 8 for Glu 0  2dG 5  pyruvate 20 group. *, p normoglycemia. 0.05 versus 5mM [Glu] . C and D, comparison between the effects of 2dG To dissect further which of the two, glycolysis or oxidative (5 mM)onI under 0 mM [Glu] and5mM [Glu] .*, p  0.05 versus HERG o o phosphorylation, is truly responsible for HERG modulation, 5mM [Glu] . E and F, effects of 2dG (5 mM)onI under 20 mM [Glu] . o HERG o the following experiments were carried out. In the first set of The number of cells was 10 for Glu 5 group, 8 for Glu 20 group, and 11 for Glu 20  2dG 5 group. *, p  0.05 versus 20 mM [Glu] , unpaired t experiments, pyruvate was supplied to the 2dG-containing o tests. Glu-free Tyrode’s solution. Under such a condition, the glycol- ysis was inhibited, but the oxidative phosphorylation was maintained. As displayed in Fig. 3A, addition of pyruvate at a ation by NaCN produced a slight non-significant decrease in concentration of 5 mM restored the depressed I induced by I (Fig. 4A). Under hyperglycemia (20 mM [Glu] ), however, HERG HERG o glycolysis inhibition. However, the negative shifts of I-V rela- I was markedly diminished, and NaCN restored the de- HERG tionship and activation curve, as seen with hypoglycemia or pressed I toward the I amplitude seen under normo- HERG HERG 2dG substitution for Glu, were still consistently observed with glycemia (5 mM). For instance, the step and tail I densities HERG pyruvate (Fig. 3B). In contrast, elevation of pyruvate to 20 mM in 20 mM glucose were 50.4  5.9 and 56.4  5.3 pA/pF (n  14) weakened the ability to restore the suppressed I caused by at 10 mV and were restored by NaCN to 73.9  7.9 and 81.6 HERG glycolysis inhibition with 2dG substitution for Glu. Thus, the 6.4 pA/pF (n  13, p  0.05 versus 20 mM [Glu] ), respectively. I density with 20 mM pyruvate in the 2dG-containing Also important is that the oxidative phosphorylation inhibition HERG Glu-free Tyrode’s solution was considerably smaller than that did not produce any significant voltage shifts of I-V relation- with normal [Glu] (5 mM). However, this high concentration of ships and activation curves, regardless of different [Glu] val- o o pyruvate still failed to prevent the negative shifts of I I-V ues (5 or 20 mM) (Fig. 4B). It appears from the above data that HERG relationship and activation curve produced by glycolysis inhi- either glycolysis or oxidative phosphorylation was sufficient to bition (Fig. 3B). sustain the normal function of HERG channels, and significant In the second set of experiments, the oxidative phosphoryl- negative shifts of I I-V relationships and activation curves HERG ation was inhibited by inclusion of NaCN (2 mM), an uncoupler occurred when glycolysis was inhibited, regardless of whether of oxidative phosphorylation, in the bathing solution, and the oxidative phosphorylation was maintained or not. glycolysis was kept intact with 5 or 20 mM [Glu] . Under nor- Role of Intracellular ATP in Maintaining HERG Function— moglycemia (5 mM [Glu] ), inhibition of oxidative phosphoryl- The above experiments indicate that glucose metabolism (gly- o Metabolic Mechanisms of HERG Modulation by Glucose 10421 FIG.4. Effects of inhibition of oxidative phosphorylation on I . Mean current density (pA/pF)-voltage relationships (A) and HERG activation conductance (G) curves (B). Inhibition of oxidative phospho- rylation with intact glycolysis was achieved by the addition of NaCN (2 mM) to the Glu (5 or 20 mM)-containing solution. The numbers in the legends represent the concentrations of Glu or NaCN in mM. n  23 for FIG.5. Effects of depletion of intracellular ATP on I . A, HERG Glu 5 group, 15 for Glu 5  NaCN 2 group, 14 for Glu 20 group, and 13 raw I recorded with the ATP-free pipette solution right after mem- HERG for Glu 20  NaCN 2 group. *, p  0.05 versus Glu5mM; , p  0.05 brane rupture as base-line control data (left) and 10 min after mem- versus Glu 20 mM, unpaired t tests. brane rupture with complete dialysis (right). B, mean I-V relationships (n  5 cells) showing the depression of I caused by depletion of HERG intracellular ATP. C, time-dependent changes of I at 10 mV HERG colysis and oxidative phosphorylation) is critical for I mod- HERG showing the current run down with the ATP-free pipette solution, ulation by Glu. Yet it was unclear whether the I HERG which is otherwise minimal with the ATP-containing pipette solution. modulation by glucose metabolism is associated with the gen- D, normalized I-V relationships showing the negative shift of voltage dependence of I caused by the depletion of intracellular ATP. E, eration of high energy phosphates (i.e. ATP), and if so whether HERG activation conductance (G) curves before and after ATP depletion. *, p the ATP generated from the glucose metabolism is glycolysis- 0.05 versus control (Ctl), paired t tests; , p  0.05, F test indicating the derived or oxidative phosphorylation-derived. To clarify this statistical significance of the time dependence. issue, we first assessed the influence of intracellular ATP de- pletion on HERG function in the presence of 5 mM Glu, by omitting ATP from the pipette (internal) solution. I re- as observed with the ATP-free internal solution (Fig. 6C). For HERG corded immediately after membrane rupture and capacitance/ instance, at 10 mV the I recorded 10 min after dialysis HERG resistance compensation was taken as base-line control data, was 32.2  2.1% smaller than the basal current recorded right and the same measurement was repeated every 5 min up to 15 after membrane rupture. We then went on to test if substitu- min. Under our experimental conditions, 10 min is sufficient to tion of GTP for ATP could prevent I run-down. With 5 mM HERG allow complete dialysis, thereby the equilibrium between pi- GTP in the ATP-free pipette solution, the I developed a HERG pette solution and cytoplasm. As illustrated in Fig. 5, the I similar degree of rapid run-down to what was seen with the HERG recorded with the normal ATP-containing pipette showed only intracellular ATP depletion alone (Fig. 6F). For example, at slight run-down over a 15-min period, whereas the I re- 10 mV, there were 34.4  5.1% decreases in the I am- HERG HERG corded with ATP-free pipette was found significantly reduced plitude 10 min after dialysis. Moreover, the negative shifts of with time. There was an 46% decrease in I at 10 min I-V relationships and activation curves were also seen with HERG after dialysis at 10 mV (Fig. 5C), being similar to the reduc- AMP-PCP or GTP (Fig. 6, B and E). For instance, V1 was ⁄2 tion of I under hypoglycemia or with the inhibition of changed from 33.4  0.8 mV for base line to 37.3  1.0 mV HERG glucose metabolism by 2dG. Also consistent with the hypogly- for 10 min of dialysis with AMP-PCP, and similarly, V1 was ⁄2 cemia and metabolic inhibition was the negative shifts of the shifted from 31.4  1.5 mV to 38.9  2.2 mV by GTP (p I-V relationship (Fig. 5D) and voltage-dependent activation 0.05). (Fig. 5E)ofI with ATP depletion; the V1 was changed by The same effects of ATP depletion on I were consis- ⁄2 HERG HERG 8 mV from 31.2  3.6 mV before to 38.9  8.1 mV (p  tently reproduced when glyburide (10 M) was included in the 0.05, n  7) after [ATP] depletion. superfusate to inhibit ATP-sensitive K current (K ), if any i ATP To investigate whether the requirement of intracellular ATP (data not shown). for HERG function relies on hydrolysis of ATP or is simply Role of ROS on Hyperglycemia-induced I Depression— HERG because of nucleotide interaction with the nucleotide binding Collectively from the above experiments, glucose metabolism is domain of HERG channels (17), we carried out the following necessary for maintaining the HERG channel function, and the series of experiments. We first used the pipette containing the ATP produced by either glycolysis or oxidative phosphorylation non-hydrolysable AMP-PCP to replace ATP. With 5 mM AMP- seems to be a key factor for the regulation; on the other hand, PCP in the pipette, the I demonstrated a rapid run-down the fact that NaCN restores the depressed I induced by 20 HERG HERG 10422 Metabolic Mechanisms of HERG Modulation by Glucose FIG.6. Effects of non-hydrolysable ATP (AMP-PCP) and GTP on I . A HERG and D, I-V relationships. I was re- HERG corded right after membrane rupture and 10 min after dialysis with the ATP-free pipette solution containing AMP-PCP (n  10) (A) or GTP (n  7) (D). B and E, activation conductance (G) curves before and after AMP-PCP or GTP in the ATP- free internal solution. The numbers in the legends represent AMP-PCP or GTP con- centrations in mM. C and F, time-depend- ent changes of I at 10 mV recorded HERG with the pipette containing ATP (control (Ctl)) or AMP-PCP (C) or GTP (F). *, p 0.05 versus control, paired t tests; , p 0.05, F test indicating the significance of time dependence. mM [Glu] or by 20 mM pyruvate suggests that oxidative phos- phorylation also produces negative (suppressive) regulation on HERG function. This would imply that the I suppression HERG by high glucose via oxidative phosphorylation is ATP-inde- pendent or is the balance between the enhancement by ATP and the suppression by other factors associated with oxidative phosphorylation. It has been well established that mitochon- dria produce most of the endogenous reactive oxygen species (ROS) through oxidative phosphorylation (18 –23), and hyper- glycemia stimulates massive ROS production (19, 24 –29). It is therefore rational to propose that the endogenously produced ROS via oxidative phosphorylation stimulated by hyperglyce- mia could impair HERG channel function to suppress I . HERG To test this hypothesis, we performed the following experi- ments. We first evaluated the effects of an antioxidant vitamin E (VitE) on 20 mM [Glu] -induced I depression. Cells were o HERG divided into three groups as follows: 5 mM [Glu] ,20mM [Glu] , o o and 20 mM [Glu]  0.1 mM VitE. As illustrated in Fig. 7, A and B, pretreatment of cell with VitE effectively prevented the I suppression by hyperglycemia; the I density in VitE HERG HERG group was virtually identical to that in the normoglycemia group. There was no significant shift of the activation curve along the voltage axis. These data suggest a participation of FIG.7. Effects of antioxidants on I depression induced by HERG ROS in the HERG regulation by hyperglycemia. Next, we stud- hyperglycemia (20 mM [Glu] ). A and B, mean current density (pA/ ied the effects of another antioxidant, superoxide dismutase pF)-voltage relationships and activation conductance (G) curves, re- (SOD) mimetic MnTBAP, on the I depression induced by HERG spectively, showing the effects of VitE (100 M)onI under 20 mM HERG 20 mM [Glu] . Because the compound does not readily penetrate [Glu] . The cells were superfused with VitE for 40 min or with the normal Tyrode’s solution for control before patch clamp recordings. the cells, it was intracellularly applied through dialysis of the Group comparison was made between the untreated control cells (Ctl, pipette solution at a concentration of 5 M. Cells were super- n  9) and the VitE-treated (n  8) cells. *, p  0.05 versus control, fused with 20 mM Glu for 30 min prior to formation of the unpaired t tests. C and D, mean current density (pA/pF)-voltage rela- whole-cell membrane patch. To correct for the potential current tionships and activation conductance (G) curves, respectively, showing the effects of MnTBAP (5 M), a superoxide dismutase mimetic, on run-down, the I recorded at various time points after HERG I under 20 mM [Glu] . MnTBAP was applied intracellularly HERG o membrane rupture was normalized to the I recorded with HERG through the pipette. I recorded immediately after whole-cell for- HERG the normal Tyrode’s solution without MnTBAP at the corre- mation and series resistance compensation was taken as base-line sponding time points. As shown in Fig. 7C, MnTBAP caused a control data (Ctl) and that recorded 10 min after dialysis was used for analysis to reflect the effects of MnTBAP. To correct for potential time-dependent increase in the I amplitude, indicating a HERG run-down of the current, the I recorded with MnTBAP was nor- HERG restoration of the depressed HERG function. By comparison, no malized to that recorded with the normal internal solution. *, p  0.05 alterations of I , or virtually a slight decrease (presumably HERG versus control, paired t tests. representing run-down of the current), were found with cata- lase in the pipette (data not shown). These results indicate that the ROS involved in the I suppression by hyperglycemia consistently smaller in X/XO-treated cells than in X/XO non- HERG was of mainly superoxide anion (O ). treated cells (Fig. 8, A and B). To obtain further evidence for this notion, we assess the To confirm that ROS production was indeed increased by 20 effects of exogenously produced O by the ROS-generating mM [Glu] and the hyperglycemia-induced ROS was mainly of 2 o system xanthine/xanthine oxidase (X/XO) on I . Cells were O , we proceeded to measure the intracellular ROS levels HERG 2 incubated with or without X/XO (500 M/5 milliunits/ml) in using CM-H DCFDA fluorescence dye. The ROS level was Tyrode’s solution containing 5 mM [Glu] for 40 min before measured in cells preincubated with the Tyrode’s solution con- I was recorded under 5 mM [Glu] . The I density was taining 5 or 20 mM glucose for 30 min. The staining of the cells HERG o HERG Metabolic Mechanisms of HERG Modulation by Glucose 10423 FIG.8. Effects of the oxidant generating system X/XO on I HERG under normoglycemia (5 mM [Glu] ). A, analogue data of I with o HERG and without treatment with X/XO. The cells were superfused with X/XO (500 M/5 milliunits/ml) for 30 min before patch clamp recordings. B and C, mean current density (pA/pF)-voltage relationships and activa- tion conductance (G) curves, respectively. Group comparison was made between the untreated control cells (Ctl) and the X/XO-treated cells. *, p  0.05 versus control, unpaired t tests, n  8, for control and n  7 for FIG.9. Oxidative phosphorylation and intracellular levels of X/XO. ROS measured by CM-H DCFDA fluorescence dye. A, laser scan- ning confocal microscopic images of CM-H DCFDA staining reflecting demonstrated two distinct patterns as follows: one localized to the intracellular ROS levels. The numbers in the labels indicate the concentrations in mM. Note the focused staining on the rod-shaped the defined rod-shaped structures, and the other one diffused structures in the cells under normoglycemia and the diffused staining evenly throughout the cytoplasm. The former presumably rep- throughout the cytoplasm in the cells treated with high glucose or resents the physiological production of ROS as a by-product of pyruvate. B, percentage of positively stained cells (mean  S.E.), ob- oxidative phosphorylation in mitochondria, and the latter in- tained from 5 fields of 3 experiments by counting the cells with staining intensity 5 times the background. C, averaged intensity of CM- dicates overproduction of ROS as a result of metabolic stress H DCFDA fluorescence measured from the positively stained cells. and damage to mitochondria. The cells with diffused staining Shown are the data normalized to 5 mM [Glu] .*, p  0.05 versus Glu and with fluorescence intensity 5 times the background were 5mM and , p  0.05 versus Glu 20 mM, unpaired t tests. defined as positive staining, and the number of cells with positive staining was pooled from 5 fields. The intensity of B, the ROS level was significantly lower, as indicated by the staining by the fluorescent probe for ROS was analyzed by smaller number of cells with positive staining and the weaker densitometric scanning using the LSM program, and cells with intensity of staining in individual cells, in the cells pretreated either localized or diffused staining were taken for analysis, with VitE than in non-treated cells, at 20 mM [Glu] . Consis- and the data were normalized to the control (5 mM [Glu] ) o tently, MnTBAP abolished the hyperglycemia-induced ROS values. Under normoglycemia, a majority of cells that was generation; the number of stained cells and the intensity of stained by CM-H DCFDA demonstrated the localized pattern, 2 staining in the presence of MnTBAP were nearly the same as and the diffused staining was sparse. Yet in the cells treated those under normoglycemia. with 20 mM Glu, the number of the cells with positive staining It has been well documented that the X/XO ROS-generating as well as the intensity of staining was consistently higher, as system stimulates mainly the generation of O . To test compared with the cells treated with 5 mM Glu (Fig. 9). This whether this is also true in our conditions, the ROS that was high level of ROS production was markedly suppressed in the exogenously generated by X/XO and penetrated cells was meas- cells pretreated with NaCN (2 mM, Fig. 9), an uncoupler of ured, and the effect of MnTBAP was studied at 5 mM [Glu] .As oxidative phosphorylation, indicating that the mitochondrion is displayed in Fig. 11, the ROS level was significantly higher in most likely where the ROS was massively produced. Because the cells treated with X/XO alone, and this increase in ROS we have demonstrated that pyruvate at high concentrations level was prevented in the cells pretreated with MnTBAP. decreased I (Fig. 3A), the ROS level in the cells pretreated HERG DISCUSSION with 20 mM pyruvate was also measured. As shown in Fig. 9, like 20 mM Glu, 20 mM pyruvate also significantly increased the The work described here documents a previously unreported ROS production although to a less extent. Moreover, the gly- role of glucose in regulating the function of HERG K channels. colysis inhibitor 2dG (5 mM) also reduced the ROS level in high Our data revealed that glucose produces two characteristic glucose (20 mM), which is indicated by fewer positively stained effects on the HERG channel function: changes of HERG cur- cells and lower intensity of staining (Fig. 9). The results explain rent (I ) amplitude/density and activation voltage. The HERG why 2dG partly restored the depressed I in 20 mM Glu maximum HERG function operates under physiological [Glu] HERG o (Fig. 3E). (5 mM, normoglycemia), and depressed HERG function occurs We have shown that VitE prevented, and MnTBAP partly with [Glu] 5mM (hypoglycemia) or with [Glu] 5mM (hy- o o reversed, the I depression in hyperglycemia (Fig. 7). To perglycemia); hypoglycemia but not hyperglycemia causes HERG see whether this is indeed attributable to their antioxidant shifts of HERG activation toward hyperpolarizing voltages. actions, effects of VitE and MnTBAP on hyperglycemia-induced The modulation of HERG by glucose is critically mediated by ROS production were also studied. As shown in Fig. 10, A and glucose metabolism (glycolysis and oxidative phosphorylation), 10424 Metabolic Mechanisms of HERG Modulation by Glucose Either Glycolysis- or Oxidative Phosphorylation-derived ATP Is Sufficient for Maintaining the Normal HERG Function—The end point of glucose metabolism is the generation of high en- ergy phosphates for maintaining cellular functions, and deple- tion of ATP could impair cellular processes dependent on high energy phosphates. One of the major findings of this study is that the normal HERG function critically relies on the level of intracellular ATP; depletion of intracellular ATP impairs HERG function to an extent similar to what severe hypoglyce- mia (0 mM [Glu] ) does (see Figs. 1, 2, and 5). Complete inhi- bition of glucose metabolism by 2dG substitution for glucose reproduces the effects of hypoglycemia or ATP depletion on I . Yet, neither inhibition of glycolysis alone by 2dG sub- HERG stitution of Glu with a supply of pyruvate to sustain oxidative phosphorylation nor inhibition of oxidative phosphorylation alone by NaCN in the presence of 5 mM Glu to maintain glycolysis is able to cause depression of HERG function. The results imply that the ATP generated by glucose metabolism plays an important role in maintaining HERG function, and FIG. 10. Effects of antioxidants vitamin E (100 M) and SOD either the glycolytic or oxidative ATP is adequate for the mimetic (MnTBAP, 10 M) on intracellular ROS levels under hyperglycemia ([Glu]  20 mM). A, confocal microscopic images of regulation. CM-H DCFDA staining showing decreases in the intracellular ROS ATP synthesis and utilization are subcellularly compartmen- levels in the cells treated with VitE or MnTBAP. B, percentage of the talized; glycolysis-derived ATP primarily regulates membrane positively stained cells (mean  S.E.), obtained from 5 fields of 3 proteins because the glycolytic pathway is associated with sar- experiments by counting the cells with staining intensity 5 times the background. C, averaged intensity of CM-H DCFDA fluorescence meas- 2 colemma (31), whereas oxidative phosphorylation-derived ATP ured from the positively stained cells. Shown are the data normalized to preferentially supports cytosolic processes because oxidative the control (Ctl) cells without antioxidant treatment. , p  0.05 versus ATP is generated within the mitochondria and subsequently Glu 20 mM, unpaired t tests. transported to the cytoplasm (32). Regulation of cardiac ATP- sensitive K channel (K ) (34) and L-type Ca channel (I ) ATP Ca (33) by intracellular ATP has been well documented by some previous studies. It was found that both K and I were ATP Ca preferentially regulated by glycolytic ATP (33, 34). Glycolysis-derived ATP May Be Responsible for Maintaining the Normal Voltage-dependent Activation of I —Although HERG as mentioned above, either glycolytic or oxidative ATP is suf- ficient for maintaining the normal HERG current amplitude/ density, only glycolysis-derived ATP seems to affect the steady- state voltage-dependent activation property of HERG channels. This notion is supported by several lines of evidence from our experiments. 1) Hypoglycemia (0 or 1 mM [Glu] ), a situation with inadequate ATP production but not hyperglyce- mia (10 or 20 mM [Glu] ), causes negative shifts of I-V relation- ships and voltage-dependent activation curves. 2) Inhibition of glycolysis (Fig. 3), but not oxidative phosphorylation (Fig. 4), abolishes the negative shifts of the HERG activation. 3) When glycolysis is inhibited by 2dG, preservation of oxidative phos- phorylation by addition of pyruvate fails to prevent the nega- FIG. 11. Effects of the superoxide generating system X/XO on tive shift caused by hypoglycemia (see Fig. 3). 4) Depletion of the intracellular ROS levels under normoglycemia ([Glu]  5 mM). A, confocal microscopic images of CM-H DCFDA staining showing intracellular ATP reproduces negative shifts similar to those the increases in the intracellular ROS levels by X/XO (0.5 mM/5 milli- seen with hypoglycemia. Similar dependence of glycolytic ATP units/ml) and the abrogation of ROS increase by the SOD mimetic regulation of K and I has been documented (33–34). ATP Ca MnTBAP (10 M). B, percentage of the positively stained cells (mean Role of ATP in Maintaining HERG Function Is Most Likely S.E.), obtained from 5 fields of 3 experiments by counting the cells with intensity of staining 5 times the background. C, averaged intensity of Due to the Phosphorylation-dependent Mechanisms—Two al- staining measured from the positively stained cells. Shown are the data ternative mechanisms could account for intracellular ATP reg- normalized to the control (Ctl) cells without X/XO treatment. *, p  0.05 ulation of ion channels: ATP acts as a substrate for phospho- versus Glu5mM and  p  0.05 versus Glu5mM  X/XO, unpaired t rylation of channel proteins by protein kinase which requires tests. ATP hydrolysis, and ATP interacts with the nucleotide binding domains of channel proteins to produce allosteric regulation and ATP and ROS are crucial in defining the HERG function not requiring ATP hydrolysis. The latter mechanism has been with changing extracellular glucose levels. shown to operate for K and I regulation (33, 35). In our Depression of HERG Function in Hypoglycemia Likely Re- ATP Ca case, neither the non-hydrolysable analogue of ATP AMP-PCP sults from Underproduction of ATP—In our study, lower glu- cose levels and inhibition of glucose metabolism both produced nor GTP prevented the I run-down caused by ATP deple- HERG tion (Fig. 6); instead substitution of AMP-PCP or GTP for ATP similar suppression of HERG function as reflected by the sub- stantial diminishment of HERG current (I ), pointing to a in the internal solution produced nearly identical effects as HERG seen with ATP depletion alone. The results suggest that the requirement of glucose metabolism for HERG modulation by glucose. Our data allowed us to reach the following conclusions. HERG regulation by ATP under our experimental conditions is Metabolic Mechanisms of HERG Modulation by Glucose 10425 phosphorylation-dependent requiring ATP hydrolysis. In other were mainly produced via the oxidative phosphorylation in words, ATP serves as a substrate for phosphorylation of HERG mitochondria in our cells. channels by protein kinases. Indeed, we have recently found It has been reported that the ROS, which generate highly that the normal HERG function requires basal activity of pro- reactive hydroxyl group (OH ), such as H O or FeSO /ascorbic 2 2 4 tein kinase B and inhibition of protein kinase B markedly acid (an oxidative stimulus analogous to H O ), increased 2 2 suppresses I and shifts HERG activation along the voltage I at negative potentials by shifting the HERG activation to HERG HERG axis toward more negative potentials (36). These results are in more negative voltages (45– 46). These results are opposite to good agreement with the HERG regulation by ATP. Studies are our observations. One explanation is that the ROS generated currently undertaken to clarify the link between ATP and under our experimental conditions may be different from the protein kinase B modulation of HERG channels. OH -generating system (H O or FeSO /ascorbic acid). As al- 2 2 4 Depression of HERG Function in Hyperglycemia Results ready mentioned, previous studies have confirmed that the from Overproduction of ROS—The consequence of the physio- ROS induced by hyperglycemia is mainly of O . Here, we also logical role in oxidative phosphorylation is the generation of showed that the SOD mimetic MnTBAP reduced the hypergly- ROS as by-products of the consumption of molecular oxygen in cemia-induced ROS overproduction (Fig. 10), and the O -gen- the electron transport chain (23). Physiologically, these ROS erating system X/XO produced ROS which were also abolished are mostly trapped within mitochondria and rapidly scavenged by MnTBAP, evidence for O as a major ROS generated in our cells. Consistently, depressive effects of hyperglycemia or X/XO by endogenous antioxidants like SOD, catalase, glutathione, etc. Yet under metabolic stress, ROS can be overproduced and on I were significantly weakened by MnTBAP. Indeed, it HERG can cause damages to mitochondria. Consequently, the ROS has been reported that the O generated by high glucose (23 may diffuse throughout the cytoplasm and cause further dele- mM) or by X/XO in rat small coronary arteries impairs voltage- terious effects on other cellular processes. Abnormally high gated K (K ) current (39, 47); reducing the current density by concentrations of glucose can enhance ROS damage at least in around 60%, which was partially restored by SOD and catalase. three different ways. First, it has been known that high glucose All together, we believe that different ROS might have differ- (25 mM) evoked ROS generation, which was blocked by antioxi- ent effects on I ;OH enhances, whereas O depresses, HERG 2 dants, inhibitors of mitochondrial electron transport chain I . HERG complex, inhibitors of glycolysis-derived pyruvate transport Moreover, it has been shown that excessive ROS inhibits into mitochondria, uncouplers of oxidative phosphorylation, glycolysis and the subsequent glycolytic ATP production and SOD mimetics, catalase, etc. (19). Superoxide anion (O )is even depletes intracellular ATP levels in isolated perfused found to be the major ROS produced under hyperglycemia hearts (48 –50). This fact together with our data suggests that (37– 42), and increases in ROS can be prevented by SOD. Sec- besides the potential direct modulation of I by ROS, ATP HERG ond, glucose itself can auto-oxidize to form ROS including O , reduction potentially caused by ROS may also contribute to the OH , and H O (43). Finally, acute elevations in glucose also I depression under hyperglycemia. This provides an alter- 2 2 HERG depress natural antioxidant defenses. It has been found that native explanation for the depressive effects of the ROS over- incubation of purified bovine CuZn-SOD with 10 to 100 mM production overcoming the enhancing effects of the expected glucose reduces the enzyme activity by 60% (44). ATP increase. Elevated glucose or pyruvate level is expected to enhance In our study, the effects of pyruvate and X/XO were smaller oxidative phosphorylation and produce more ATP molecules to than those of hyperglycemia. This may be because the O support HERG function or increase I . However, our obser- generated by hyperglycemia occurs inside the cell but pyruvate HERG vations are contrary to this expectation. Our results showed does not readily penetrate cells, and the effect of pyruvate that hyperglycemia or excessive pyruvate markedly depressed observed in the present study may underestimate the true role the HERG function. A reasonable explanation for this is that of oxidative phosphorylation in I modulation. Likewise, HERG the ROS produced under hyperglycemia counteract the effects the O generated by X/XO was primarily extracellular with of ATP, and the net outcome is a balance between enhancing subsequent entry into the cell which could also underestimate effects of ATP and suppressing effects of ROS. Evidently, under the effects of O . our experimental conditions, the effects of increased ROS over- It was shown that within minutes of exposure to dihydroxy- write the effects of increased ATP, resulting in suppression of fumaric acid or xanthine plus xanthine oxidase, both of which I . This notion is supported by the following evidence. produce the superoxide anion, action potential duration was HERG First, the depression of I induced by hyperglycemia was prolonged in canine myocytes, and this effect was followed by HERG prevented or reversed by the antioxidants vitamin E and the appearance of early after depolarization (51). X/XO caused MnTBAP (SOD mimetic). Second, inhibition of the glycolysis a 30% increase in action potential duration in superfused pap- and thereby the subsequent oxidative phosphorylation by 2dG illary muscle or small strips of right ventricular walls of guinea partially reversed the depressed I under hyperglycemia pig hearts (52). However, whether the action potential duration HERG (Fig. 3). Weakened oxidative phosphorylation due to inhibition prolongation was associated with inhibition of delayed rectifier of the glycolysis would reduce both ATP and ROS productions, K current (I ) is unknown. Our study provides a potential Kr but the net result was an increase in I , indicating again explanation for these observations. HERG that in our experimental conditions ROS overweighs ATP in Impairment of HERG Function Might Contribute to Q-T Pro- terms of their effects on I . This is in agreement with the longation Caused by Hypoglycemia and Hyperglycemia—Heart HERG notion that the suppressing effects of ROS overproduction over- disease is a leading cause of death in diabetic patients. In whelm the effects of ATP increase. Moreover, 2dG also can patients with diabetes a prolongation of the Q-T interval has compete with glucose for access to glucose transporters and been associated with an increased risk of sudden cardiac death thus decreases glucose uptake which in turn can result in (2) due to the occurrence of lethal ventricular arrhythmias, reduction of ROS production in the cells. Finally, our data particularly Torsade de pointes following bradycardia (1). Sev- indeed demonstrated the ability of high glucose to stimulate an eral cardiovascular pathological consequences of diabetes such overproduction of ROS (see Fig. 9). The fact that a high con- as hypertension and arteriosclerosis affect the heart to varying centration of pyruvate mimicked, whereas VitE or NaCN degrees. Hyperglycemia, as a consequence of diabetes and an abrogated, the ROS overproduction suggests that the ROS independent risk factor, also can directly cause cardiac dam- 10426 Metabolic Mechanisms of HERG Modulation by Glucose 12. Sanguinetti, M. C. (1999) Ann. N. Y. Acad. Sci. 868, 406 – 413 age. On the other hand, insulin therapy increases the risk of 13. Taglialatela, M., Castaldo, P., and Pannaccione, A. (1998) Biochem. Pharma- hypoglycemia in type 2 diabetic patients; according to the col. 55, 1741–1746 14. Zhou, Z., Gong, Q., Ye, B., Fan, Z., Makielski, J. C., Robertson, G. A., and United Kingdom Prospective Diabetes Study, approximately January, C. T. (1998) Biophys. J. 74, 230 –241 one-third of the insulin-treated patients reported one or more 15. Wang, J., Wang, H., Han, H., Zhang, Y., Yang, B., Nattel, S., and Wang, Z. hypoglycemic episodes per year during the first 3 years (3). (2001) Circulation 104, 2645–2648 16. Wang, H., Yang, B., Zhang, Y., Han, H., Wang, J., Shi, H., and Wang, Z. (2001) Hypoglycemia is presumed to be the cause of death in about 3% J. Biol. Chem. 276, 40811– 40816 of insulin-treated diabetic patients (11). Intriguingly, it is well 17. Splawski, I., Shen, J., Katherine, W., Timothy, G., Vincent, M., Lehmann, M. H., and Keating, M. T. (1998) Genomics 51, 86 –97 recognized that both hyperglycemia and hypoglycemia can 18. Ludwig, B., Bender, E., Arnold, S., Huttemann, M., Lee, I., and Kadenbach, B. cause prolongation of Q-T interval. In type 2 diabetes, the (2001) Chembiochem. 2, 392– 403 prevalence of Q-T prolongation is as high as 26%, and Q-T 19. Hsieh, T. J., Zhang, S. L., Filep, J. G., Tang, S. S., Ingelfinger, J. R., and Chan, J. S. (2002) Endocrinology 143, 2975–2985 prolongation during experimentally induced and spontane- 20. Wallace, D. C. (2002) Methods Mol. Biol. 197, 3–54 ously occurring hypoglycemia or diabetic hyperglycemia has 21. Siraki, A. G., Pourahmad, J., Chan, T. S., Khan, S., and O’Brien, P. J. (2002) also been shown to occur in healthy subjects and in diabetic Free Radic. Biol. Med. 32, 2–10 22. Kirkinezos, I. G., and Moraes, C. T. (2001) Semin. Cell Dev. Biol. 12, 449 – 457 patients with increased risk of malignant ventricular arrhyth- 23. Wallace, D. C. (2001) Novartis. Found. Symp. 235, 247–263 mias (4 –11). Yet the potential ionic mechanisms by which 24. Koo, J. R., Ni, Z., Oviesi, F., and Vaziri, N. D. (2002) Clin. Exp. Hypertens. 24, 333–344 hypoglycemia and hyperglycemia cause Q-T prolongation re- 25. Marfella, R., Quagliaro, L., Nappo, F., Ceriello, A., and Giugliano, D. (2001) mained poorly understood. Studies on glucose modulation of J. Clin. Invest. 108, 635– 636 cardiac ion channels are sparse and have been mostly limited 26. Srivastava, A. K. (2002) Int. J. Mol. Med. 9, 85– 89 27. Li, P. A., Liu, G. J., He, Q. P., Floyd, R. A., and Siesjo, B. K. (1999) Free Radic. to ATP-sensitive K current (K ). On the contrary to I , ATP HERG Biol. Med. 27, 1033–1040 K is closed with increased intracellular ATP levels (34). One ATP 28. van Dam, P. S., van Asbeck, B. S., Bravenboer, B., van Oirschot, J. F., Gispen, W. H., and Marx, J. J. (1998) Free Radic. Biol. Med. 24, 18 –26 study reported by Xu et al. (53) demonstrated that hyperglyce- 29. Giardino, I., Edelstein, D., and Brownlee, M. (1996) J. Clin. Invest. 97, mia (18 mM [Glu] ) decreased the density of transient outward 1422–1428 K current but did not alter the inward rectifier K current in 30. Takigawa, T., Yasuda, H., Terada, M., Maeda, K., Haneda, M., Kashiwagi, A., Kitasato, H., and Kikkawa, R. (2000) Neuroreport 11, 2547–2551 rat ventricular myocytes that do not express delayed rectifier 31. Hazen, S. L., Wolf, M. J., Ford, D. A., and Gross, R. W. (1994) FEBS Lett. 339, K current (I ), the physiological counterpart of I .In Kr HERG 213–216 32. Weiss, J. N., and Hiltbrand, B. (1985) J. Clin. Invest. 75, 436 – 447 arterial smooth muscles, high glucose diminished shaker-type 33. Losito, V. A., Tsushima, R. G., Diaz, R. J., Wilson, G. J., and Backx, P. H. delayed rectifier K current (39). In rat myelinated nerve fi- (1998) J. Physiol. (Lond.) 511, 67–78 bers, 30 mM glucose increased Ca -activated K current (30). 34. Weiss, J. N., and Lamo, S. T. (1989) J. Gen. Physiol. 94, 911–935 35. Yazawa, K., Kameyama, A., Yasui, K., Li, J. M., and Kameyama, M. (1997) Whereas none of the data from these studies could fully account Pfluegers Arch. 433, 557–562 for the Q-T prolongation, particularly the Q-T prolongation 36. Zhang, Y., Wang, H., Wang, J., Han, H., Nattel, S., and Wang, Z. (2003) FEBS induced by hypoglycemia, our study provides a plausible, or at Lett. 534, 125–132 37. Nishikawa, T., Edelstein, D., Du, X. L., Yamagishi, S., Matsumura, T., least an alternative, explanation; HERG K channel may be a Kaneda, Y., Yorek, M. A., Beebe, D., Oates, P. J., Hammes, H. P., Giardino, mechanistic link for the Q-T prolongation induced by both I., and Brownlee, M. (2000) Nature 404, 787–790 38. Du, X. L., Edelstein, D., Rossetti, L., Fantus, I. G., Goldberg, H., Ziyadeh, F., hyperglycemia and hypoglycemia. Yet one should keep it in Wu, J., and Brownlee, M. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, mind that HERG may be only one of the multiple factors 12222–12226 contributing to the Q-T prolongation in hypoglycemia and 39. Liu, Y., Terata, K., Rusch, N. J., and Gutterman, D. D. (2001) Circ. Res. 89, 146 –152 hyperglycemia. 40. Graier, W. F., Posch, K., Wascher, T. C., and Kostner, G. M. (1997) Horm. Metab. Res. 29, 622– 629 Acknowledgments—We thank XiaoFan Yang for excellent technical 41. Graier, W. F., Posch, K., Fleischhacker, E., Wascher, T. C., and Kostner, G. M. support and Louis R. Villeneuve for assistance with the confocal micro- (1999) Diabetes Res. Clin. Pract. 45, 153–160 scopic examinations. 42. Esposito, K., Marfella, R., and Giugliano, D. (2002) J. Endocrinol. Invest. 25, 485– 488 REFERENCES 43. Wolff, S. P., and Dean, R. T. (1987) Biochem. J. 245, 243–250 44. Oda, A., Bannai, C., Yamaoka, T., Katori, T., Matsushima, T., and Yamashita, 1. Abo, K., Ishida, Y., Yoshida, R., Hozumi, T., Ueno, H., Shiotani, H., K. (1994) Horm. Metab. Res. 26, 1– 4 Matsunaga, K., and Kazumi, T. (1996) Diabetes Care 19, 1010 45. Be ´ rube ´ , J., Caouette, D., and Daleau, P. (2001) J. Pharmacol. Exp. Ther. 297, 2. Kahn, J. K., Sisson, J. C., and Vinik, A. I. (1987) J. Clin. Endocrinol. Metab. 64, 96 –102 751–754 46. Taglialatela, M., Castaldo, P., Iossa, S., Pannaccione, A., Fresi, A., Ficker, E., 3. United Kingdom Prospective Diabetes Study (1995) Br. Med. J. 310, 83– 88 and Annunziato, L. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 11698 –11703 4. Marfella, R., Rossi, F., and Giugliano, D. (2001) Diabetes Nutr. Metab. 14, 47. Liu, Y., and Gutterman, D. D. (2002) Clin. Exp. Pharmacol. Physiol. 29, 63– 65 305–311 5. Chelliah, Y. R. (2000) Anaesth. Intensive Care 28, 698 –700 48. Ytrehus, K., Myklebust, R., and Mjøs, O. D. (1986) Cardiovasc. Res. 20, 6. Veglio, M., Borra, M., Stevens, L. K., Fuller, J. H., and Perin, P. C. (1999) Diabetologia 42, 68 –75 597– 603 49. Miki, S., Ashraf, M., Salka, S., and Sperelakis, N. (1988) J. Mol. Cell. Cardiol. 7. Marques, J. L., George, E., Peacey, S. R., Harris, N. D., Macdonald, I. A., Cochrane, T., and Heller, S. R. (1997) Diabet. Med. 14, 648 – 654 20, 1009 –1024 50. Patane, G., Anello, M., Piro, S., Vigneri, R., Purrello, F., and Rabuazzo, A. M. 8. Landstedt-Hallin, L., Englund, A., Adamson, U., and Lins, P.-E. (1999) J. In- tern. Med. 246, 299 –307 (2002) Diabetes 51, 2749 –2756 51. Barrington, P. L., Meier, C. F., Jr., and Weglicki, W. B. (1988) J. Mol. Cell. 9. Lindstrom, T., Jorfeldt, L., Tegler, L., and Arnqvist, H. J. (1992) Diabet. Med. 9, 536 –541 Cardiol. 20, 1163–1178 52. Aiello, E. A., Jabr, R. I., and Cole, W. C. (1995) Circ. Res. 77, 153–162 10. Shimada, R., Nakashima, T., Nunoi, K., Kohno, Y., Takeshita, A., and Omae, T. (1984) Arch. Intern. Med. 144, 1068 –1069 53. Xu, Z., Patel, K. P., and Rozanski, G. J. (1998) Am. J. Physiol. 271, 11. Eckert, B., and Agardh, C.-D. (1998) Clin. Physiol. (Oxf.) 18, 570 –575 H2190 –H2196

Journal

Journal of Biological Chemistry – Unpaywall

Published: Mar 1, 2003

There are no references for this article.