Molecular Mechanisms Mediating Inhibition of Human Large Conductance Ca2+-Activated K+ Channels by High Glucose
http://www.100md.com
Tong Lu, Tongrong He, Zvonimir S. Katusi
参见附件。
the Departments of Internal Medicine (T.L., H.-C.L.) and Anesthesiology (T.H., Z.S.K.), Mayo Clinic, Rochester, Minn.
Abstract
Diabetic vascular dysfunction is associated with an increase in reactive oxygen species (ROS). In this study, we hypothesized that hyperglycemia-induced ROS generation would impair the function of large conductance Ca2+-activated K+ (BK) channels, which are major determinants in vasorelaxation. We found that when cultured in high glucose (HG) (22 mmol/L), HEK293 cells showed a reduction in expressed hSlo current densities, as well as slowed activation and deactivation kinetics. When human coronary smooth muscle cells were cultured in HG, similar findings were observed for the BK currents. HG enhanced superoxide dismutase and suppressed catalase (CAT) expression in HEK293 cells, leading to a significant increase in intracellular ROS. The effects of HG were mimicked by hydrogen peroxide (H2O2), and hSlo functions were restored by CAT gene transfer. Peroxynitrite inhibited hSlo current density but did not change channel kinetics. The hSloC911A mutant was insensitive to the effects of HG and H2O2. Hence, imbalance of antioxidant enzymes plays a critical role in ROS generation in HG, impairing hSlo functions through H2O2-dependent oxidation at cysteine 911. This may represent an important fundamental mechanism that contributes to the impairment of vasodilation in diabetes.
Key Words: large conductance Ca2+-activated K+ channels hyperglycemia catalase reactive oxygen species gene transfer
Introduction
Diabetes is associated with a 2- to 4-fold increase in the risk of developing coronary artery disease and stroke, and there has been a 75% increase in mortality from cardiovascular diabetic complications over the last decade in the United States.1,2 Nearly 21 million Americans have diabetes, with another 41 million considered prediabetic. Multiple factors contribute to the development of diabetic vasculopathy.1,2 Endothelial dysfunction, reduced nitric oxide (NO) availability, and enhanced production of reactive oxygen species (ROS), such as superoxide anion () and hydrogen peroxide (H2O2), are thought to be major mechanisms that underlie the development of cardiovascular complications in diabetes.1 is rapidly oxidized to H2O2 by superoxide dismutase (SOD), and H2O2 is further reduced to H2O by catalase (CAT) and glutathione peroxidase (GPX).3 In addition, reacts with NO extremely efficiently to form peroxynitrite (OONO–). H2O2 and OONO– are highly reactive ROS. Abnormal ROS metabolism, leading to cellular oxidative stress, plays a central role in the progression of diabetic vascular dysfunction.1,3
The large conductance Ca2+-activated K+ (BK) channel, abundantly expressed in vascular smooth muscle cells, plays a critical role in controlling vascular tone.4 The BK channel is composed of 4 pore-forming subunits, each with 7 transmembrane domains (S0 to S6), including a highly conservative pore region between S5 and S6 and a voltage sensor located at S4.5 The cytoplasmic terminus has 4 hydrophobic segments (S7 to S10) that contain critical channel regulatory sites, including 2 regulators of K+ conductance (RCK1 and RCK2) domains and the Ca2+ binding motifs (Ca2+ bowl).6,7 Movement of the RCK domains affects Ca2+ binding affinity.7 There is evidence that BK channel-mediated vasodilation is impaired in hyperglycemia; however, the molecular mechanism has not been fully delineated.8,9 Hyperglycemia-induced ROS overproduction in diabetes is well documented.10 In vascular smooth muscle cells, a major source of ROS is NAD(P)H oxidase,11 which is present in HEK293 cells.12 OONO– inhibits vascular BK channels in a dose-dependent manner, whereas itself has no effect on BK channels.13,14 H2O2 may either inhibit15,16 or activate BK channels through direct effects17,18 or through cGMP- and cAMP-dependent signaling pathways.19,20 Cysteine 911 of hSlo, which encodes the BK channel, is thought to be involved in channel regulation by redox and cysteine-modification reagents,14 although studies have produced conflicting results.21 Most of the previous studies used exogenously applied H2O2 from 10 μmol/L to 167 mmol/L.14,16,17 Whether endogenously produced H2O2 in hyperglycemia can modulate BK channel function is unknown. In this study, we hypothesized that a high glucose (HG)-induced increase in ROS modulates hSlo channel function by oxidation of specific residues. We examined changes in the properties of hSlo channels, which encode human BK channel subunits, in HG. The results may help elucidate the molecular mechanisms underlying diabetes-associated vascular dysfunction and the role of ROS in the regulation of BK current functions in blood vessels.
Materials and Methods
An expanded Materials and Methods section containing details for cell culture, transfection, vasoreactivity, Western blotting, and electrophysiology is available in the online data supplement at http://circres.ahajournals.org.
Cell Culture and hSlo, CAT Transfection
HEK293 cells were cultured in DMEM containing normal glucose (NG) (5 mmol/L) or HG (22 mmol/L).22 hSlo cDNA and 2 recombinant adenoviruses, AdCAT carrying CAT and AdLacZ carrying -galactosidase, were used for gene transfer.23
Animals and Vasoreactivity Measurements
Diabetes mellitus was produced in male Sprague-Dawley rats (200 g, Harlan, Inc, Ind) by streptozotocin (60 mg/kg, IP). Control rats received vehicle injection. Blood glucose in excess of 300 mg/dL was considered diabetic. Animals were used for vasoreactivity experiments 4 weeks after development of diabetes. Handling and care of animals, as well as all animal procedures, were approved by the Institutional Animal Care and Use Committee, Mayo Foundation.
On the day of the experiment, rats were anesthetized with pentobarbital (50 mg/kg IP). The heart was rapidly removed from the thorax. Small coronary arteries were isolated from control and diabetic rats and vasoreactivity was measured using videomicroscopy.8
Electrophysiology
hSlo currents were recorded in HEK293 cells 72 hours after hSlo transfection using standard patch clamp techniques at room temperature (23°C).24,25
Intracellular ROS Measurement
Products of intracellular H2O2 and were detected as previously described.26 Briefly, attached HEK293 cells were exposed to 5 μmol/L 5,6-chloromethyl-2',7'-dichlorodihydrofluorescein diacetate (CM-H2DCF-DA) for 30 minutes at 37°C in the dark, washed twice with PBS, and incubated in NG or HG medium for another 30 minutes. Cells were trypsinized (0.1%) and washed with PBS. After centrifugation (1200 rpm for 10 minutes), the cell pellet was suspended in 200 μL of PBS. Formation of ROS was detected by signals from the fluorescent reaction products of CM-DCF and ethidium using flow cytometry (Becton Dickinson FACSCan System and CellQuest Software).
Western Blot Analysis
Protein expressions of manganese SOD (MnSOD), copper zinc SOD (CuZnSOD), CAT, and GPX-1 in HEK293 cells cultured with NG and HG were determined using Western blot.27
Chemicals
OONO– was purchased from Upstate Cell Signaling and Solutions (Lake Placid, NY). All other chemicals were obtained from Sigma-Aldrich (St Louis, Mo).
Statistical Analysis
Data are presented as mean±SEM. Student t test was used to compare data between 2 groups. Paired t test was used to compare data before and after treatment. One-way ANOVA, followed by Tukey test analysis was used to compare data from multiple groups using SigmaStat software (Jandel, San Rafael, Calif). Statistical significance was defined as P<0.05.
Results
Inhibition of hSlo Currents by HG
Acute exposure of hSlo expressed in HEK293 cells to medium containing NG or HG had no effects on current density or kinetics. However, when hSlo was expressed in HEK293 cells that had been exposed to HG for 1 week, reduction in current density and slowing of channel activation were observed. These changes became stable in cells after 2 weeks culture with HG. Figure 1A shows representative whole-ell currents of hSlo 72 hours after transfection in HEK293 cells that had been cultured in NG or HG for 2 weeks. HG significantly reduced current amplitudes (Figure 1A) and current/voltage (I-V) curves showed that HG reduced the maximal current density of hSlo by 47.1%, with 671.6±105.1 pA/pF (n=10) in NG and 355.6±102.3 pA/pF in HG (+200 mV; n=10; P=0.045) (Figure 1B). Normalized conductance/voltage (g-V) curves of hSlo in NG and HG were fitted using a Boltzmann equation (Figure 1C). The voltage at half-maximal activation (V1/2) of hSlo activation was not different between NG and HG (113.5±10.8 mV in NG versus 97.5±9.2 mV in HG; n=10 for both; P=NS), and neither was the slope factor k (34.8±1.6 mV/e-fold in NG versus 31.4±2.3 mV/e-fold in HG; n=10 for both; P=NS). Single-channel recordings showed that cells in HG reduced hSlo open probability (PO) by 60%, prolonging channel closed time constants (C) by 100% without changing channel open time constants (O) (Figure 1D; Table I in the online data supplement). HG slowed the time course of channel activation and increased the activation time constants (A) compared with those in NG (detailed analysis is presented in supplemental Table II).
HG also reduced hSlo tail current amplitudes (Figure 2A) and slowed the time course of channel deactivation. The activation time constant (A) and deactivation time constant (D) was fitted from the hSlo currents in NG and HG (detailed analysis of the channel deactivation kinetics is presented in Table II and Figure III of the online data supplement). The relationships between time constants () and voltages (-V curve) were fitted using the voltage-dependent relaxation equation,28 as shown in Figure 2B. Increases in A and D by HG resulted in shifting the -V curve upward. The voltage-dependent free energy (G) required for transition from channel closed state (C) to open state (O) was calculated by curve fitting using rate theory analysis29 to be 2.6 kJ/mol in NG and 4.9 kJ/mol in HG, an increase of 1.9-fold in HG, and G from O to C was increased by 1.3-fold in HG (4.5 kJ/mol in NG and 5.8 kJ/mol in HG), suggesting that channel activation and deactivation transitions in HG are less favorable thermodynamically compared with those in NG. Similar results were obtained from the BK channels of human coronary artery smooth muscle cells cultured with HG (see the online data supplement). In addition, vasoreactivity studies showed that epoxyeicosatrienoic acid (EET)-induced vasodilation through BK channel activation was impaired in coronary arteries of streptozotocin-induced diabetic rats (Figure 2C). In controls, BK channel activation accounted for 39.3% of the 11,12-EET-mediated (1 μmol/L) vasodilation (63.8±4.2% at baseline versus 24.5±2.1% with 100 nmol/L iberiotoxin [IBTX]; n=5). In diabetic vessels, BK channels contributed only 9.5% of the EET-mediated vasodilation, with 31.9±2.1% dilation at baseline (n=5; P<0.01 versus control) and. 22.4±2.0% with IBTX (n=5; P=NS versus baseline). These results suggest that BK channel dysfunction contributes to the impairment of vasodilation in diabetes (Figure 2C).
HG Produced Antioxidant Enzyme Imbalance and Increased Intracellular ROS Production in HEK293 Cells
Western blot analysis showed that CAT expression was reduced by 44% in HG cells (0.37±0.04 versus 0.66±0.07 in NG cells; n=3 for both; P=0.023), whereas CuZnSOD expression increased by 36% (0.83±0.06 versus 0.61±0.05 in NG cells; n=3 for both; P=0.048) (Figure 3A and 3B). The level of GPX-1 expression in HEK293 cells was very low. There was no significant difference in MnSOD and GPX-1 expression between NG and HG cells. This altered antioxidant enzyme profile in HG cells could result in reduced ability to scavenge H2O2 sufficiently, favoring the accumulation of H2O2.
Production of total peroxides was measured in HEK293 cells after 2 weeks in NG and HG, using the fluorescent ROS indicator CM-H2DCF-DA (Figure 3C).26 Cells cultured in NG (NG cells) showed fluorescence signals of 66.9±9.1 arbitrary units (n=4) in NG medium and 124.6±21.3 arbitrary units (n=4; P=0.02) in HG medium, suggesting HG could instantaneously increase intracellular ROS production in NG cells. However, the fluorescence signals from cells cultured in HG (HG cells) were 153.0±17.3 arbitrary units when measured in NG medium (n=4; P=0.002 versus NG cells in NG medium), comparable to that of NG cells measured in HG medium (n=4; P=NS). The fluorescence signals in HG cells further increased to 219.6±43.8 arbitrary units when placed in HG medium (n=4; P=0.048 versus NG cells with HG medium). Thus, intracellular ROS increased 3.3-fold in HG cells in HG medium, compared with NG cells in NG medium. These results suggest that intracellular ROS was persistently elevated in HG cells and the ability to scavenge intracellular ROS generated was impaired. Group data are summarized in Figure 3D.
Adenoviral Expression of CAT Reversed the Effects of HG on hSlo
To determine whether the effects of HG on hSlo could be alleviated by increase in CAT activities, we overexpressed CAT by adenoviral gene transfer into HG cells before transfection of hSlo. Figure 4A shows raw tracings of hSlo 72 hours after transfection in HG cells infected with adenoviral-CAT (AdCAT). There was a 4-fold increase in CAT expression in AdCAT-infected HG cells compared with HG cells infected with control adenovirus (AdLacZ) (Figure 4B). Interestingly, hSlo current density in HG cells was enhanced by AdCAT infection (890.7±125.5 pA/pF, +200 mV [n=6] versus 355.6±102.3 pA/pF in HG control [n=10]; P=0.006), attaining levels similar to that in NG cells (dashed line) (n=6; P=NS) (Figure 4C). Enhanced CAT expression had no effect on the g-V curve (Figure 4D). V1/2 and k were 93.0±8.6 mV and 36.7±1.8 mV/e-fold (n=6; P=NS versus HG control for both), respectively, in HG cells infected with AdCAT, showing no difference from those in HG control. However, CAT overexpression reduced A and D in parallel (Figure 4E) to levels comparable to those in NG. Hence, enhanced CAT activity restored hSlo current density and kinetics to that of NG cells, indicating that reduced CAT activity is the major culprit in hSlo inhibition by HG.
Exogenous H2O2 Mimicked the Effects of HG on hSlo Kinetics
NAD(P)H oxidase is present in HEK293 cells12 and intracellular H2O2 in HEK293 cells is <40 μmol/L under physiological conditions.30 Because the ROS florescence signal increased 3.3-fold in HG, the average cellular concentration of ROS could be raised to the 10–4 mol/L range in HG cells. Figure 5A shows that H2O2 dose-dependently inhibited the inside-out macroscopic currents of hSlo with an IC50 of 4 mmol/L. The H2O2 effect was reversed by 2 mmol/L dithiothreitol (DTT). Hence, in subsequent experiments, we applied 4 mmol/L H2O2 to determine the effects of H2O2 on hSlo kinetics. Similar to HG, H2O2 reduced hSlo current density and increased A and D in a voltage-dependent manner (Figure 5B). I-V curves normalized to baseline showed that H2O2 reduced the maximal current density by 44.7±8.1% (+200 mV; n=3; P=0.01 versus baseline) (Figure 5C). V1/2 was unaltered by H2O2 (131.0±7.0 mV at baseline versus 134.6±6.3 mV with H2O2; n=3; P=NS versus baseline), as was k (17.7±2.6 mV/e-fold at baseline versus 19.3±2.3 mV/e-fold with H2O2; n=3; P=NS) (Figure 5D). Thus, the effects of H2O2 on hSlo channel kinetics were similar to those of HG. The G for channel activation was calculated to be 2.0 kJ/mol at baseline and 4.4 kJ/mol with H2O2, whereas the G for channel deactivation was 4.8 kJ/mol at baseline and 5.5 kJ/mol with H2O2. These values are similar to those obtained in HG. The -V curve and kinetics parameters in response to H2O2 are shown in Figure V, E, in the online data supplement.
Exogenous OONO– Inhibited hSlo Current Density but Not Channel Kinetics
Because OONO– can also be overproduced in HG, we examined the effects of OONO– on hSlo. We found that OONO– (0.1 mmol/L) inhibited hSlo current density by as much as 4 mmol/L H2O2 but did not alter channel activation or deactivation kinetics, and the effects of OONO– were completely reversed by 2 mmol/L DTT (Figure 6A). Normalized I-V curves in the presence and absence of 0.1 mmol/L OONO– are shown in Figure 6B. OONO– inhibited the maximal hSlo activity by 31.4±16.7% (n=4; P=0.01), comparable to the effects of 4 mmol/L H2O2 and HG (n=4; P=NS). Neither channel V1/2 (162.8±11.8 mV at baseline and 170.5±13.1 mV with OONO–; n=4; P=NS) nor k (21.5±1.7 mV/e-fold at baseline and 22.0±1.4 mV/e-fold with OONO–; n=4; P=NS) was altered by OONO–. Unlike HG and H2O2, OONO– had no effect on A and D (Figure 6C). A comparison between the effects of H2O2 and OONO– on hSlo channel kinetics is given in Figure VI, D, and Table II in the online data supplement.
hSloC911A Mutant Was Insensitive to H2O2 and HG
Because the effects of HG on hSlo were mainly mediated by H2O2 in HEK293 cells, we determined the role of cysteine 911 in HG- and H2O2-mediated hSlo modulations. hSlo with a single substitution at cysteine 911 by alanine (hSloC911A) was expressed in HEK293 cells. The time course of the effects of 4 mmol/L H2O2 and 0.1 mmol/L OONO– on hSloC911A inside-out macroscopic currents is illustrated in Figure 7A (top and bottom, respectively). Relative to hSlo wild type (wt), hSloC911A was insensitive to H2O2 inhibition, although H2O2 produced a very small but still significant inhibition in current density and the effect was reversed by 2 mmol/L DTT. However, the effects of OONO– on hSloC911A were unchanged from wt, suggesting that C911 is the molecular site for H2O2 modulation, but not for OONO–. Moreover, H2O2 failed to slow hSloC911A activation or deactivation (Figure 7B and 7D) and only produced a maximal current inhibition of 14.3±2.4% (+200 mV; n=4; P=0.048) (Figure 7C), compared with 44.7±8.1% in wt (n=4; P=0.011). H2O2 did not alter the channel V1/2 (130.3±16.1 mV at baseline and 137.3±27.4 mV with H2O2; n=4; P=NS) or k (17.7±2.6 mV/e-fold at baseline and 19.7±2.3 mV/e-fold with H2O2; n=4; P=NS). H2O2 did not change the G of hSloC911A activation (2.8 kJ/mol at baseline and 3.0 kJ/mol with H2O2) or deactivation (4.5 kJ/mol at baseline and 4.8 kJ/mol with H2O2). There was no difference between the current densities, V1/2, and channel kinetics of hSloC911A in cells in NG and HG, which were similar to wt recordings in NG (Figure 8). Hence, the C911A mutation prevented downregulation of channel activities in HG (926.0±193.2 pA/pF, +200 mV [n=6] versus 355.6±102.3 pA/pF in wt [n=10]; P=0.008) and maintained normal -V curves. The G for hSloC911A activation and deactivation in HG was 2.9 kJ/mol and 4.9 kJ/mol, respectively, similar to those of wt channel in NG. Our results indicate that H2O2 is the major ROS produced by HEK293 cells in HG and C911 is the major molecular site of redox regulation in hSlo by HG.
Discussion
In this study, we have made several important findings. First, HG reduced hSlo current density and altered channel kinetics. Second, HG enhanced SOD expression but suppressed CAT expression, resulting in a 3.3-fold increase in ROS generation in HEK293 cells. Third, H2O2 but not OONO– mimicked the effects of HG on hSlo channel kinetics. Fourth, hSloC911A was insensitive to HG or H2O2 modulation, suggesting that C911 in hSlo is the major molecular target of redox regulation by HG. We believe this is the first report identifying the molecular mechanism associated with impaired BK channel function in hyperglycemia.
Vascular K+ channels are effectors of endothelium-derived relaxation factors (EDRF) and endothelium-derived hyperpolarizing factors (EDHF) and are critical determinants of vascular tone.4,31 Impaired BK channel-mediated vasodilation is known to be associated with vascular dysfunction in diabetic animals. However, the mechanisms through which vascular BK channels are modulated by hyperglycemia have not been delineated. In high fructose diet-induced, insulin-resistant rats, BK channel density in mesenteric smooth muscle cells was reduced, but the channel Ca2+ sensitivity and the voltage sensitivity were unchanged.32 In early-stage type 2 diabetic rats, impaired coronary arterial smooth muscle BK channel activation by arachidonic acid was attributable to a decrease in prostacyclin bioavailability.8 In this study, we demonstrated that enhanced H2O2 formation in HEK293 cells cultured with HG directly inhibited channel activity through redox modulation at C911 on the BK channel subunit. HG not only directly enhanced NAD(P)H oxidase, but also interrupted the mitochondria electron transport chain at complex III, resulting in accumulation of , which in turn formed H2O2 and OONO–.10 However, hSlo activity was not inhibited by acute exposure to HG but by prolonged culture in HG that reached stable effects after 14 days. This time frame was required for HG to produce the imbalance in SOD and CAT activities, resulting in the accumulation of H2O2. This was supported by direct measurements of intracellular ROS, showing a 3.3-fold increase in HEK293 cells after 14 days of culture in HG. We believe that the major ROS product in our system was H2O2. First, SOD expression was increased by 36%, whereas CAT expression was decreased by 44% in HG cells, hence resulting in net H2O2 accumulation. Second, gene transfer of CAT into HG cells abolished the HG effects on hSlo current density and channel kinetics. Third, the application of H2O2, but not OONO–, mimicked the effects of HG, suggesting that OONO– was not the key ROS participant in modulating hSlo properties in HEK293 cells. Whereas BK channel kinetics were not affected by OONO–, we found that 0.1 mmol/L OONO– produced as much hSlo current density inhibition as 4 mmol/L H2O2. Because OONO– is highly reactive and generated by both endothelial and vascular smooth muscle cells, we believe that in diabetic vessels, BK channels are also modulated by OONO–. Why relatively high concentrations of exogenous H2O2 are required to reproduce the HG results is not immediately fully understood. We believe that H2O2 acts directly on hSlo and not on membrane lipids because a single C911A mutation is sufficient to eliminate the HG and H2O2 effects.
The mechanism underlying BK channel regulation by H2O2 is complex. A putative EDHF, H2O2 has been shown to activate vascular BK channels directly or through second messenger signaling pathways, leading to vasodilation.19,20 However, activation of BK channels by H2O2 is observed only when applied extracellularly, whereas intracellularly applied H2O2 inhibits BK channels.14,16,17 Hence, the paracrine and autocrine functions of H2O2 on BK channels have divergent outcomes. Contrary to its physiological signaling role, prolonged excessive production of H2O2 is detrimental to the maintenance of cellular homeostasis, resulting in the development of diseases such as diabetes and arteriosclerosis. Cysteine modification of hSlo is involved in channel regulation by H2O2. However, the molecular identity of the cysteine residue in hSlo that is the functional target of redox modulation has been controversial. Tang et al reported that C911 near the Ca2+ bowl in the RCK2 domain is the major target of cysteine-modification reagents such as MTSEA and of cytoplasmic application of H2O2.14 Oxidation of C911 also reduces channel Ca2+ sensitivity as a result of an increase in the free energy for Ca2+ binding, which was only observed in the presence of high free Ca2+ (>1 μmol/L).14 Zhang and Horrigan showed that C430 in the RCK1 domain but not C911 was the functional residue modified by oxidation.21 The crystal structure of the RCK domains on the Ca2+-dependent K+ channel, MthK, shows that RCK1 is coupled to the channel pore through the M2 transmembrane domain, which corresponds to S6 in the BK channel.7 Thus, oxidation of C430 may impair RCK1 movement and reduce channel Ca2+ sensitivity.21 Both studies presented strong evidence that a single mutation virtually eliminates the effects of the cysteine-modification reagents on hSlo channel Ca2+ sensitivity and voltage sensitivity. H2O2 only efficiently oxidizes cysteine residues that are deprotonated, whereas most cysteine residues in proteins have a pKa of 8.5, so they would not be targeted by H2O2 unless they are located in the vicinity of highly positively charged residues. The Ca2+ bowl contains a high density of aspartic acids surrounded by positively charged amino acids. This critical location may explain why C911 is the prime target for H2O2 modulation. Our results show that HG and H2O2 inhibit hSlo mainly through C911 oxidation, because the C911A mutation abrogated the effects of HG and eliminated 80% of the H2O2 effects. The effects of cysteine-modifying reagents such as MTSET, MTSES, and NEM on hSlo channel kinetics can shift the steady-state I-V curves either leftward or rightward, depending on whether the channel activation or deactivation gate is modified by these reagents.21 Slowing channel activation can shift the steady-state I-V or g-V curves leftward, whereas slowing channel deactivation favors channels to stay in open state and shifts the curves rightward. We found that HG and H2O2 do not shift the I-V curve, perhaps because of a balanced slowing of channel activation and deactivation by HG and H2O2. Another explanation is that hSlo currents were recorded with 200 nmol/L intracellular free Ca2+, which may be insufficient to change the free energy of Ca2+ binding.14
To summarize, we have provided compelling evidence that HG inhibits BK channel function through H2O2 oxidation of C911. These results indicate that diabetes alters antioxidant enzyme profiles, enhances ROS generation and compromises BK channel function. Gene transfer of CAT ameliorates these functional impairments and has therapeutic potential in the treatment of diabetic vascular dysfunction.
Acknowledgments
We thank Dr Larry W. Oberley (The University of Iowa, Iowa City) for providing AdCAT. The hSloC911A mutant was kindly provided by Dr Toshinori Hoshi (University of Pennsylvania, Philadelphia).
Sources of Funding
This work was supported by grants from the NIH (HL-63154 and HL-74180) and the American Heart Association (0265472Z).
Disclosures
None.
Footnotes
Original received January 25, 2006; resubmission received June 14, 2006; revised resubmission received July 24, 2006; accepted August 16, 2006.
References
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy. Part I. Circulation. 2003; 108: 1527–1532.
Luscher TF, Creager MA, Beckman JA, Cosentino F. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy. Part II. Circulation. 2003; 108: 1655–1661.
Taniyama Y, Griendling KK. Reactive oxygen species in the vasculature: molecular and cellular mechanisms. Hypertension. 2003; 42: 1075–1081.
Tanaka Y, Koike K, Toro L. MaxiK channel roles in blood vessel relaxations induced by endothelium-derived relaxing factors and their molecular mechanisms. J Smooth Muscle Res. 2004; 40: 125–153. [Order article via Infotrieve]
Toro L, Wallner M, Meera P, Tanaka Y. Maxi-K(Ca), a unique member of the voltage-gated K channel superfamily. News Physiol Sci. 1998; 13: 112–117.
Jiang Y, Pico A, Cadene M, Chait BT, MacKinnon R. Structure of the RCK domain from the E. coli K+ channel and demonstration of its presence in the human BK channel. Neuron. 2001; 29: 593–601. [Order article via Infotrieve]
Jiang Y, Lee A, Chen J, Cadene M, Chait BT, MacKinnon R. Crystal structure and mechanism of a calcium-gated potassium channel. Nature. 2002; 417: 515–522. [Order article via Infotrieve]
Lu T, Wang XL, He T, Zhou W, Kaduce TL, Katusic ZS, Spector AA, Lee HC. Impaired arachidonic acid-mediated activation of large-conductance Ca2+-activated K+ channels in coronary arterial smooth muscle cells in Zucker Diabetic Fatty rats. Diabetes. 2005; 54: 2155–2163.
Zhou W, Wang XL, Kaduce TL, Spector AA, Lee HC. Impaired arachidonic acid-mediated dilation of small mesenteric arteries in Zucker diabetic fatty rats. Am J Physiol Heart Circ Physiol. 2005; 288: H2210–H2218.
Allen DA, Yaqoob MM, Harwood SM. Mechanisms of high glucose-induced apoptosis and its relationship to diabetic complications. J Nutr Biochem. 2005; 16: 705–713. [Order article via Infotrieve]
Lassegue B, Clempus RE. Vascular NAD(P)H oxidases: specific features, expression, and regulation. Am J Physiol Regul Integr Comp Physiol. 2003; 285: R277–R297.
Ambasta RK, Kumar P, Griendling KK, Schmidt HH, Busse R, Brandes RP. Direct interaction of the novel Nox proteins with p22phox is required for the formation of a functionally active NADPH oxidase. J Biol Chem. 2004; 279: 45935–45941.
Liu Y, Terata K, Chai Q, Li H, Kleinman LH, Gutterman DD. Peroxynitrite inhibits Ca2+-activated K+ channel activity in smooth muscle of human coronary arterioles. Circ Res. 2002; 91: 1070–1076.
Tang XD, Garcia ML, Heinemann SH, Hoshi T. Reactive oxygen species impair Slo1 BK channel function by altering cysteine-mediated calcium sensing. Nat Struct Mol Biol. 2004; 11: 171–178. [Order article via Infotrieve]
Soto MA, Gonzalez C, Lissi E, Vergara C, Latorre R. Ca(2+)-activated K+ channel inhibition by reactive oxygen species. Am J Physiol Cell Physiol. 2002; 282: C461–C471.
DiChiara TJ, Reinhart PH. Redox modulation of hslo Ca2+-activated K+ channels. J Neurosci. 1997; 17: 4942–4955.
Bychkov R, Pieper K, Ried C, Milosheva M, Bychkov E, Luft FC, Haller H. Hydrogen peroxide, potassium currents, and membrane potential in human endothelial cells. Circulation. 1999; 99: 1719–1725.
Barlow RS, White RE. Hydrogen peroxide relaxes porcine coronary arteries by stimulating BKCa channel activity. Am J Physiol. 1998; 275: H1283–H1289. [Order article via Infotrieve]
Shimokawa H, Matoba T. Hydrogen peroxide as an endothelium-derived hyperpolarizing factor. Pharmacol Res. 2004; 49: 543–549. [Order article via Infotrieve]
Thengchaisri N, Kuo L. Hydrogen peroxide induces endothelium-dependent and -independent coronary arteriolar dilation: role of cyclooxygenase and potassium channels. Am J Physiol Heart Circ Physiol. 2003; 285: H2255–H2263.
Zhang G, Horrigan FT. Cysteine modification alters voltage- and Ca(2+)-dependent gating of large conductance (BK) potassium channels. J Gen Physiol. 2005; 125: 213–236.
Lu T, Hong MP, Lee HC. Molecular determinants of cardiac K(ATP) channel activation by epoxyeicosatrienoic acids. J Biol Chem. 2005; 280: 19097–19104.
Ahmad IM, Aykin-Burns N, Sim JE, Walsh SA, Higashikubo R, Buettner GR, Venkataraman S, Mackey MA, Flanagan SW, Oberley LW, Spitz DR. Mitochondrial O2- and H2O2 mediate glucose deprivation-induced stress in human cancer cells. J Biol Chem. 2005; 280: 4254–4263.
Lu T, Lee HC, Kabat JA, Shibata EF. Modulation of rat cardiac sodium channel by the stimulatory G protein alpha subunit. J Physiol. 1999; 518 (pt 2): 371–384.
Lu T, Katakam PV, VanRollins M, Weintraub NL, Spector AA, Lee HC. Dihydroxyeicosatrienoic acids are potent activators of Ca(2+)-activated K(+) channels in isolated rat coronary arterial myocytes. J Physiol. 2001; 534: 651–667.
Dernbach E, Urbich C, Brandes RP, Hofmann WK, Zeiher AM, Dimmeler S. Antioxidative stress-associated genes in circulating progenitor cells: evidence for enhanced resistance against oxidative stress. Blood. 2004; 104: 3591–3597.
He T, Peterson TE, Holmuhamedov EL, Terzic A, Caplice NM, Oberley LW, Katusic ZS. Human endothelial progenitor cells tolerate oxidative stress due to intrinsically high expression of manganese superoxide dismutase. Arterioscler Thromb Vasc Biol. 2004; 24: 2021–2027.
Cui J, Cox DH, Aldrich RW. Intrinsic voltage dependence and Ca2+ regulation of mslo large conductance Ca-activated K+ channels. J Gen Physiol. 1997; 109: 647–673.
Aidley DJ, Stanfield PR. Ion Channels: Molecules in Action. Cambridge, UK; New York: Cambridge University Press; 1996.
Colton CA, Gbadegesin M, Wink DA, Miranda KM, Espey MG, Vicini S. Nitroxyl anion regulation of the NMDA receptor. J Neurochem. 2001; 78: 1126–1134. [Order article via Infotrieve]
Nelson MT, Quayle JM. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995; 268: C799–C822. [Order article via Infotrieve]
Dimitropoulou C, Han G, Miller AW, Molero M, Fuchs LC, White RE, Carrier GO. Potassium (BK(Ca)) currents are reduced in microvascular smooth muscle cells from insulin-resistant rats. Am J Physiol Heart Circ Physiol. 2002; 282: H908–H917.
the Departments of Internal Medicine (T.L., H.-C.L.) and Anesthesiology (T.H., Z.S.K.), Mayo Clinic, Rochester, Minn.
Abstract
Diabetic vascular dysfunction is associated with an increase in reactive oxygen species (ROS). In this study, we hypothesized that hyperglycemia-induced ROS generation would impair the function of large conductance Ca2+-activated K+ (BK) channels, which are major determinants in vasorelaxation. We found that when cultured in high glucose (HG) (22 mmol/L), HEK293 cells showed a reduction in expressed hSlo current densities, as well as slowed activation and deactivation kinetics. When human coronary smooth muscle cells were cultured in HG, similar findings were observed for the BK currents. HG enhanced superoxide dismutase and suppressed catalase (CAT) expression in HEK293 cells, leading to a significant increase in intracellular ROS. The effects of HG were mimicked by hydrogen peroxide (H2O2), and hSlo functions were restored by CAT gene transfer. Peroxynitrite inhibited hSlo current density but did not change channel kinetics. The hSloC911A mutant was insensitive to the effects of HG and H2O2. Hence, imbalance of antioxidant enzymes plays a critical role in ROS generation in HG, impairing hSlo functions through H2O2-dependent oxidation at cysteine 911. This may represent an important fundamental mechanism that contributes to the impairment of vasodilation in diabetes.
Key Words: large conductance Ca2+-activated K+ channels hyperglycemia catalase reactive oxygen species gene transfer
Introduction
Diabetes is associated with a 2- to 4-fold increase in the risk of developing coronary artery disease and stroke, and there has been a 75% increase in mortality from cardiovascular diabetic complications over the last decade in the United States.1,2 Nearly 21 million Americans have diabetes, with another 41 million considered prediabetic. Multiple factors contribute to the development of diabetic vasculopathy.1,2 Endothelial dysfunction, reduced nitric oxide (NO) availability, and enhanced production of reactive oxygen species (ROS), such as superoxide anion () and hydrogen peroxide (H2O2), are thought to be major mechanisms that underlie the development of cardiovascular complications in diabetes.1 is rapidly oxidized to H2O2 by superoxide dismutase (SOD), and H2O2 is further reduced to H2O by catalase (CAT) and glutathione peroxidase (GPX).3 In addition, reacts with NO extremely efficiently to form peroxynitrite (OONO–). H2O2 and OONO– are highly reactive ROS. Abnormal ROS metabolism, leading to cellular oxidative stress, plays a central role in the progression of diabetic vascular dysfunction.1,3
The large conductance Ca2+-activated K+ (BK) channel, abundantly expressed in vascular smooth muscle cells, plays a critical role in controlling vascular tone.4 The BK channel is composed of 4 pore-forming subunits, each with 7 transmembrane domains (S0 to S6), including a highly conservative pore region between S5 and S6 and a voltage sensor located at S4.5 The cytoplasmic terminus has 4 hydrophobic segments (S7 to S10) that contain critical channel regulatory sites, including 2 regulators of K+ conductance (RCK1 and RCK2) domains and the Ca2+ binding motifs (Ca2+ bowl).6,7 Movement of the RCK domains affects Ca2+ binding affinity.7 There is evidence that BK channel-mediated vasodilation is impaired in hyperglycemia; however, the molecular mechanism has not been fully delineated.8,9 Hyperglycemia-induced ROS overproduction in diabetes is well documented.10 In vascular smooth muscle cells, a major source of ROS is NAD(P)H oxidase,11 which is present in HEK293 cells.12 OONO– inhibits vascular BK channels in a dose-dependent manner, whereas itself has no effect on BK channels.13,14 H2O2 may either inhibit15,16 or activate BK channels through direct effects17,18 or through cGMP- and cAMP-dependent signaling pathways.19,20 Cysteine 911 of hSlo, which encodes the BK channel, is thought to be involved in channel regulation by redox and cysteine-modification reagents,14 although studies have produced conflicting results.21 Most of the previous studies used exogenously applied H2O2 from 10 μmol/L to 167 mmol/L.14,16,17 Whether endogenously produced H2O2 in hyperglycemia can modulate BK channel function is unknown. In this study, we hypothesized that a high glucose (HG)-induced increase in ROS modulates hSlo channel function by oxidation of specific residues. We examined changes in the properties of hSlo channels, which encode human BK channel subunits, in HG. The results may help elucidate the molecular mechanisms underlying diabetes-associated vascular dysfunction and the role of ROS in the regulation of BK current functions in blood vessels.
Materials and Methods
An expanded Materials and Methods section containing details for cell culture, transfection, vasoreactivity, Western blotting, and electrophysiology is available in the online data supplement at http://circres.ahajournals.org.
Cell Culture and hSlo, CAT Transfection
HEK293 cells were cultured in DMEM containing normal glucose (NG) (5 mmol/L) or HG (22 mmol/L).22 hSlo cDNA and 2 recombinant adenoviruses, AdCAT carrying CAT and AdLacZ carrying -galactosidase, were used for gene transfer.23
Animals and Vasoreactivity Measurements
Diabetes mellitus was produced in male Sprague-Dawley rats (200 g, Harlan, Inc, Ind) by streptozotocin (60 mg/kg, IP). Control rats received vehicle injection. Blood glucose in excess of 300 mg/dL was considered diabetic. Animals were used for vasoreactivity experiments 4 weeks after development of diabetes. Handling and care of animals, as well as all animal procedures, were approved by the Institutional Animal Care and Use Committee, Mayo Foundation.
On the day of the experiment, rats were anesthetized with pentobarbital (50 mg/kg IP). The heart was rapidly removed from the thorax. Small coronary arteries were isolated from control and diabetic rats and vasoreactivity was measured using videomicroscopy.8
Electrophysiology
hSlo currents were recorded in HEK293 cells 72 hours after hSlo transfection using standard patch clamp techniques at room temperature (23°C).24,25
Intracellular ROS Measurement
Products of intracellular H2O2 and were detected as previously described.26 Briefly, attached HEK293 cells were exposed to 5 μmol/L 5,6-chloromethyl-2',7'-dichlorodihydrofluorescein diacetate (CM-H2DCF-DA) for 30 minutes at 37°C in the dark, washed twice with PBS, and incubated in NG or HG medium for another 30 minutes. Cells were trypsinized (0.1%) and washed with PBS. After centrifugation (1200 rpm for 10 minutes), the cell pellet was suspended in 200 μL of PBS. Formation of ROS was detected by signals from the fluorescent reaction products of CM-DCF and ethidium using flow cytometry (Becton Dickinson FACSCan System and CellQuest Software).
Western Blot Analysis
Protein expressions of manganese SOD (MnSOD), copper zinc SOD (CuZnSOD), CAT, and GPX-1 in HEK293 cells cultured with NG and HG were determined using Western blot.27
Chemicals
OONO– was purchased from Upstate Cell Signaling and Solutions (Lake Placid, NY). All other chemicals were obtained from Sigma-Aldrich (St Louis, Mo).
Statistical Analysis
Data are presented as mean±SEM. Student t test was used to compare data between 2 groups. Paired t test was used to compare data before and after treatment. One-way ANOVA, followed by Tukey test analysis was used to compare data from multiple groups using SigmaStat software (Jandel, San Rafael, Calif). Statistical significance was defined as P<0.05.
Results
Inhibition of hSlo Currents by HG
Acute exposure of hSlo expressed in HEK293 cells to medium containing NG or HG had no effects on current density or kinetics. However, when hSlo was expressed in HEK293 cells that had been exposed to HG for 1 week, reduction in current density and slowing of channel activation were observed. These changes became stable in cells after 2 weeks culture with HG. Figure 1A shows representative whole-ell currents of hSlo 72 hours after transfection in HEK293 cells that had been cultured in NG or HG for 2 weeks. HG significantly reduced current amplitudes (Figure 1A) and current/voltage (I-V) curves showed that HG reduced the maximal current density of hSlo by 47.1%, with 671.6±105.1 pA/pF (n=10) in NG and 355.6±102.3 pA/pF in HG (+200 mV; n=10; P=0.045) (Figure 1B). Normalized conductance/voltage (g-V) curves of hSlo in NG and HG were fitted using a Boltzmann equation (Figure 1C). The voltage at half-maximal activation (V1/2) of hSlo activation was not different between NG and HG (113.5±10.8 mV in NG versus 97.5±9.2 mV in HG; n=10 for both; P=NS), and neither was the slope factor k (34.8±1.6 mV/e-fold in NG versus 31.4±2.3 mV/e-fold in HG; n=10 for both; P=NS). Single-channel recordings showed that cells in HG reduced hSlo open probability (PO) by 60%, prolonging channel closed time constants (C) by 100% without changing channel open time constants (O) (Figure 1D; Table I in the online data supplement). HG slowed the time course of channel activation and increased the activation time constants (A) compared with those in NG (detailed analysis is presented in supplemental Table II).
HG also reduced hSlo tail current amplitudes (Figure 2A) and slowed the time course of channel deactivation. The activation time constant (A) and deactivation time constant (D) was fitted from the hSlo currents in NG and HG (detailed analysis of the channel deactivation kinetics is presented in Table II and Figure III of the online data supplement). The relationships between time constants () and voltages (-V curve) were fitted using the voltage-dependent relaxation equation,28 as shown in Figure 2B. Increases in A and D by HG resulted in shifting the -V curve upward. The voltage-dependent free energy (G) required for transition from channel closed state (C) to open state (O) was calculated by curve fitting using rate theory analysis29 to be 2.6 kJ/mol in NG and 4.9 kJ/mol in HG, an increase of 1.9-fold in HG, and G from O to C was increased by 1.3-fold in HG (4.5 kJ/mol in NG and 5.8 kJ/mol in HG), suggesting that channel activation and deactivation transitions in HG are less favorable thermodynamically compared with those in NG. Similar results were obtained from the BK channels of human coronary artery smooth muscle cells cultured with HG (see the online data supplement). In addition, vasoreactivity studies showed that epoxyeicosatrienoic acid (EET)-induced vasodilation through BK channel activation was impaired in coronary arteries of streptozotocin-induced diabetic rats (Figure 2C). In controls, BK channel activation accounted for 39.3% of the 11,12-EET-mediated (1 μmol/L) vasodilation (63.8±4.2% at baseline versus 24.5±2.1% with 100 nmol/L iberiotoxin [IBTX]; n=5). In diabetic vessels, BK channels contributed only 9.5% of the EET-mediated vasodilation, with 31.9±2.1% dilation at baseline (n=5; P<0.01 versus control) and. 22.4±2.0% with IBTX (n=5; P=NS versus baseline). These results suggest that BK channel dysfunction contributes to the impairment of vasodilation in diabetes (Figure 2C).
HG Produced Antioxidant Enzyme Imbalance and Increased Intracellular ROS Production in HEK293 Cells
Western blot analysis showed that CAT expression was reduced by 44% in HG cells (0.37±0.04 versus 0.66±0.07 in NG cells; n=3 for both; P=0.023), whereas CuZnSOD expression increased by 36% (0.83±0.06 versus 0.61±0.05 in NG cells; n=3 for both; P=0.048) (Figure 3A and 3B). The level of GPX-1 expression in HEK293 cells was very low. There was no significant difference in MnSOD and GPX-1 expression between NG and HG cells. This altered antioxidant enzyme profile in HG cells could result in reduced ability to scavenge H2O2 sufficiently, favoring the accumulation of H2O2.
Production of total peroxides was measured in HEK293 cells after 2 weeks in NG and HG, using the fluorescent ROS indicator CM-H2DCF-DA (Figure 3C).26 Cells cultured in NG (NG cells) showed fluorescence signals of 66.9±9.1 arbitrary units (n=4) in NG medium and 124.6±21.3 arbitrary units (n=4; P=0.02) in HG medium, suggesting HG could instantaneously increase intracellular ROS production in NG cells. However, the fluorescence signals from cells cultured in HG (HG cells) were 153.0±17.3 arbitrary units when measured in NG medium (n=4; P=0.002 versus NG cells in NG medium), comparable to that of NG cells measured in HG medium (n=4; P=NS). The fluorescence signals in HG cells further increased to 219.6±43.8 arbitrary units when placed in HG medium (n=4; P=0.048 versus NG cells with HG medium). Thus, intracellular ROS increased 3.3-fold in HG cells in HG medium, compared with NG cells in NG medium. These results suggest that intracellular ROS was persistently elevated in HG cells and the ability to scavenge intracellular ROS generated was impaired. Group data are summarized in Figure 3D.
Adenoviral Expression of CAT Reversed the Effects of HG on hSlo
To determine whether the effects of HG on hSlo could be alleviated by increase in CAT activities, we overexpressed CAT by adenoviral gene transfer into HG cells before transfection of hSlo. Figure 4A shows raw tracings of hSlo 72 hours after transfection in HG cells infected with adenoviral-CAT (AdCAT). There was a 4-fold increase in CAT expression in AdCAT-infected HG cells compared with HG cells infected with control adenovirus (AdLacZ) (Figure 4B). Interestingly, hSlo current density in HG cells was enhanced by AdCAT infection (890.7±125.5 pA/pF, +200 mV [n=6] versus 355.6±102.3 pA/pF in HG control [n=10]; P=0.006), attaining levels similar to that in NG cells (dashed line) (n=6; P=NS) (Figure 4C). Enhanced CAT expression had no effect on the g-V curve (Figure 4D). V1/2 and k were 93.0±8.6 mV and 36.7±1.8 mV/e-fold (n=6; P=NS versus HG control for both), respectively, in HG cells infected with AdCAT, showing no difference from those in HG control. However, CAT overexpression reduced A and D in parallel (Figure 4E) to levels comparable to those in NG. Hence, enhanced CAT activity restored hSlo current density and kinetics to that of NG cells, indicating that reduced CAT activity is the major culprit in hSlo inhibition by HG.
Exogenous H2O2 Mimicked the Effects of HG on hSlo Kinetics
NAD(P)H oxidase is present in HEK293 cells12 and intracellular H2O2 in HEK293 cells is <40 μmol/L under physiological conditions.30 Because the ROS florescence signal increased 3.3-fold in HG, the average cellular concentration of ROS could be raised to the 10–4 mol/L range in HG cells. Figure 5A shows that H2O2 dose-dependently inhibited the inside-out macroscopic currents of hSlo with an IC50 of 4 mmol/L. The H2O2 effect was reversed by 2 mmol/L dithiothreitol (DTT). Hence, in subsequent experiments, we applied 4 mmol/L H2O2 to determine the effects of H2O2 on hSlo kinetics. Similar to HG, H2O2 reduced hSlo current density and increased A and D in a voltage-dependent manner (Figure 5B). I-V curves normalized to baseline showed that H2O2 reduced the maximal current density by 44.7±8.1% (+200 mV; n=3; P=0.01 versus baseline) (Figure 5C). V1/2 was unaltered by H2O2 (131.0±7.0 mV at baseline versus 134.6±6.3 mV with H2O2; n=3; P=NS versus baseline), as was k (17.7±2.6 mV/e-fold at baseline versus 19.3±2.3 mV/e-fold with H2O2; n=3; P=NS) (Figure 5D). Thus, the effects of H2O2 on hSlo channel kinetics were similar to those of HG. The G for channel activation was calculated to be 2.0 kJ/mol at baseline and 4.4 kJ/mol with H2O2, whereas the G for channel deactivation was 4.8 kJ/mol at baseline and 5.5 kJ/mol with H2O2. These values are similar to those obtained in HG. The -V curve and kinetics parameters in response to H2O2 are shown in Figure V, E, in the online data supplement.
Exogenous OONO– Inhibited hSlo Current Density but Not Channel Kinetics
Because OONO– can also be overproduced in HG, we examined the effects of OONO– on hSlo. We found that OONO– (0.1 mmol/L) inhibited hSlo current density by as much as 4 mmol/L H2O2 but did not alter channel activation or deactivation kinetics, and the effects of OONO– were completely reversed by 2 mmol/L DTT (Figure 6A). Normalized I-V curves in the presence and absence of 0.1 mmol/L OONO– are shown in Figure 6B. OONO– inhibited the maximal hSlo activity by 31.4±16.7% (n=4; P=0.01), comparable to the effects of 4 mmol/L H2O2 and HG (n=4; P=NS). Neither channel V1/2 (162.8±11.8 mV at baseline and 170.5±13.1 mV with OONO–; n=4; P=NS) nor k (21.5±1.7 mV/e-fold at baseline and 22.0±1.4 mV/e-fold with OONO–; n=4; P=NS) was altered by OONO–. Unlike HG and H2O2, OONO– had no effect on A and D (Figure 6C). A comparison between the effects of H2O2 and OONO– on hSlo channel kinetics is given in Figure VI, D, and Table II in the online data supplement.
hSloC911A Mutant Was Insensitive to H2O2 and HG
Because the effects of HG on hSlo were mainly mediated by H2O2 in HEK293 cells, we determined the role of cysteine 911 in HG- and H2O2-mediated hSlo modulations. hSlo with a single substitution at cysteine 911 by alanine (hSloC911A) was expressed in HEK293 cells. The time course of the effects of 4 mmol/L H2O2 and 0.1 mmol/L OONO– on hSloC911A inside-out macroscopic currents is illustrated in Figure 7A (top and bottom, respectively). Relative to hSlo wild type (wt), hSloC911A was insensitive to H2O2 inhibition, although H2O2 produced a very small but still significant inhibition in current density and the effect was reversed by 2 mmol/L DTT. However, the effects of OONO– on hSloC911A were unchanged from wt, suggesting that C911 is the molecular site for H2O2 modulation, but not for OONO–. Moreover, H2O2 failed to slow hSloC911A activation or deactivation (Figure 7B and 7D) and only produced a maximal current inhibition of 14.3±2.4% (+200 mV; n=4; P=0.048) (Figure 7C), compared with 44.7±8.1% in wt (n=4; P=0.011). H2O2 did not alter the channel V1/2 (130.3±16.1 mV at baseline and 137.3±27.4 mV with H2O2; n=4; P=NS) or k (17.7±2.6 mV/e-fold at baseline and 19.7±2.3 mV/e-fold with H2O2; n=4; P=NS). H2O2 did not change the G of hSloC911A activation (2.8 kJ/mol at baseline and 3.0 kJ/mol with H2O2) or deactivation (4.5 kJ/mol at baseline and 4.8 kJ/mol with H2O2). There was no difference between the current densities, V1/2, and channel kinetics of hSloC911A in cells in NG and HG, which were similar to wt recordings in NG (Figure 8). Hence, the C911A mutation prevented downregulation of channel activities in HG (926.0±193.2 pA/pF, +200 mV [n=6] versus 355.6±102.3 pA/pF in wt [n=10]; P=0.008) and maintained normal -V curves. The G for hSloC911A activation and deactivation in HG was 2.9 kJ/mol and 4.9 kJ/mol, respectively, similar to those of wt channel in NG. Our results indicate that H2O2 is the major ROS produced by HEK293 cells in HG and C911 is the major molecular site of redox regulation in hSlo by HG.
Discussion
In this study, we have made several important findings. First, HG reduced hSlo current density and altered channel kinetics. Second, HG enhanced SOD expression but suppressed CAT expression, resulting in a 3.3-fold increase in ROS generation in HEK293 cells. Third, H2O2 but not OONO– mimicked the effects of HG on hSlo channel kinetics. Fourth, hSloC911A was insensitive to HG or H2O2 modulation, suggesting that C911 in hSlo is the major molecular target of redox regulation by HG. We believe this is the first report identifying the molecular mechanism associated with impaired BK channel function in hyperglycemia.
Vascular K+ channels are effectors of endothelium-derived relaxation factors (EDRF) and endothelium-derived hyperpolarizing factors (EDHF) and are critical determinants of vascular tone.4,31 Impaired BK channel-mediated vasodilation is known to be associated with vascular dysfunction in diabetic animals. However, the mechanisms through which vascular BK channels are modulated by hyperglycemia have not been delineated. In high fructose diet-induced, insulin-resistant rats, BK channel density in mesenteric smooth muscle cells was reduced, but the channel Ca2+ sensitivity and the voltage sensitivity were unchanged.32 In early-stage type 2 diabetic rats, impaired coronary arterial smooth muscle BK channel activation by arachidonic acid was attributable to a decrease in prostacyclin bioavailability.8 In this study, we demonstrated that enhanced H2O2 formation in HEK293 cells cultured with HG directly inhibited channel activity through redox modulation at C911 on the BK channel subunit. HG not only directly enhanced NAD(P)H oxidase, but also interrupted the mitochondria electron transport chain at complex III, resulting in accumulation of , which in turn formed H2O2 and OONO–.10 However, hSlo activity was not inhibited by acute exposure to HG but by prolonged culture in HG that reached stable effects after 14 days. This time frame was required for HG to produce the imbalance in SOD and CAT activities, resulting in the accumulation of H2O2. This was supported by direct measurements of intracellular ROS, showing a 3.3-fold increase in HEK293 cells after 14 days of culture in HG. We believe that the major ROS product in our system was H2O2. First, SOD expression was increased by 36%, whereas CAT expression was decreased by 44% in HG cells, hence resulting in net H2O2 accumulation. Second, gene transfer of CAT into HG cells abolished the HG effects on hSlo current density and channel kinetics. Third, the application of H2O2, but not OONO–, mimicked the effects of HG, suggesting that OONO– was not the key ROS participant in modulating hSlo properties in HEK293 cells. Whereas BK channel kinetics were not affected by OONO–, we found that 0.1 mmol/L OONO– produced as much hSlo current density inhibition as 4 mmol/L H2O2. Because OONO– is highly reactive and generated by both endothelial and vascular smooth muscle cells, we believe that in diabetic vessels, BK channels are also modulated by OONO–. Why relatively high concentrations of exogenous H2O2 are required to reproduce the HG results is not immediately fully understood. We believe that H2O2 acts directly on hSlo and not on membrane lipids because a single C911A mutation is sufficient to eliminate the HG and H2O2 effects.
The mechanism underlying BK channel regulation by H2O2 is complex. A putative EDHF, H2O2 has been shown to activate vascular BK channels directly or through second messenger signaling pathways, leading to vasodilation.19,20 However, activation of BK channels by H2O2 is observed only when applied extracellularly, whereas intracellularly applied H2O2 inhibits BK channels.14,16,17 Hence, the paracrine and autocrine functions of H2O2 on BK channels have divergent outcomes. Contrary to its physiological signaling role, prolonged excessive production of H2O2 is detrimental to the maintenance of cellular homeostasis, resulting in the development of diseases such as diabetes and arteriosclerosis. Cysteine modification of hSlo is involved in channel regulation by H2O2. However, the molecular identity of the cysteine residue in hSlo that is the functional target of redox modulation has been controversial. Tang et al reported that C911 near the Ca2+ bowl in the RCK2 domain is the major target of cysteine-modification reagents such as MTSEA and of cytoplasmic application of H2O2.14 Oxidation of C911 also reduces channel Ca2+ sensitivity as a result of an increase in the free energy for Ca2+ binding, which was only observed in the presence of high free Ca2+ (>1 μmol/L).14 Zhang and Horrigan showed that C430 in the RCK1 domain but not C911 was the functional residue modified by oxidation.21 The crystal structure of the RCK domains on the Ca2+-dependent K+ channel, MthK, shows that RCK1 is coupled to the channel pore through the M2 transmembrane domain, which corresponds to S6 in the BK channel.7 Thus, oxidation of C430 may impair RCK1 movement and reduce channel Ca2+ sensitivity.21 Both studies presented strong evidence that a single mutation virtually eliminates the effects of the cysteine-modification reagents on hSlo channel Ca2+ sensitivity and voltage sensitivity. H2O2 only efficiently oxidizes cysteine residues that are deprotonated, whereas most cysteine residues in proteins have a pKa of 8.5, so they would not be targeted by H2O2 unless they are located in the vicinity of highly positively charged residues. The Ca2+ bowl contains a high density of aspartic acids surrounded by positively charged amino acids. This critical location may explain why C911 is the prime target for H2O2 modulation. Our results show that HG and H2O2 inhibit hSlo mainly through C911 oxidation, because the C911A mutation abrogated the effects of HG and eliminated 80% of the H2O2 effects. The effects of cysteine-modifying reagents such as MTSET, MTSES, and NEM on hSlo channel kinetics can shift the steady-state I-V curves either leftward or rightward, depending on whether the channel activation or deactivation gate is modified by these reagents.21 Slowing channel activation can shift the steady-state I-V or g-V curves leftward, whereas slowing channel deactivation favors channels to stay in open state and shifts the curves rightward. We found that HG and H2O2 do not shift the I-V curve, perhaps because of a balanced slowing of channel activation and deactivation by HG and H2O2. Another explanation is that hSlo currents were recorded with 200 nmol/L intracellular free Ca2+, which may be insufficient to change the free energy of Ca2+ binding.14
To summarize, we have provided compelling evidence that HG inhibits BK channel function through H2O2 oxidation of C911. These results indicate that diabetes alters antioxidant enzyme profiles, enhances ROS generation and compromises BK channel function. Gene transfer of CAT ameliorates these functional impairments and has therapeutic potential in the treatment of diabetic vascular dysfunction.
Acknowledgments
We thank Dr Larry W. Oberley (The University of Iowa, Iowa City) for providing AdCAT. The hSloC911A mutant was kindly provided by Dr Toshinori Hoshi (University of Pennsylvania, Philadelphia).
Sources of Funding
This work was supported by grants from the NIH (HL-63154 and HL-74180) and the American Heart Association (0265472Z).
Disclosures
None.
Footnotes
Original received January 25, 2006; resubmission received June 14, 2006; revised resubmission received July 24, 2006; accepted August 16, 2006.
References
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy. Part I. Circulation. 2003; 108: 1527–1532.
Luscher TF, Creager MA, Beckman JA, Cosentino F. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy. Part II. Circulation. 2003; 108: 1655–1661.
Taniyama Y, Griendling KK. Reactive oxygen species in the vasculature: molecular and cellular mechanisms. Hypertension. 2003; 42: 1075–1081.
Tanaka Y, Koike K, Toro L. MaxiK channel roles in blood vessel relaxations induced by endothelium-derived relaxing factors and their molecular mechanisms. J Smooth Muscle Res. 2004; 40: 125–153. [Order article via Infotrieve]
Toro L, Wallner M, Meera P, Tanaka Y. Maxi-K(Ca), a unique member of the voltage-gated K channel superfamily. News Physiol Sci. 1998; 13: 112–117.
Jiang Y, Pico A, Cadene M, Chait BT, MacKinnon R. Structure of the RCK domain from the E. coli K+ channel and demonstration of its presence in the human BK channel. Neuron. 2001; 29: 593–601. [Order article via Infotrieve]
Jiang Y, Lee A, Chen J, Cadene M, Chait BT, MacKinnon R. Crystal structure and mechanism of a calcium-gated potassium channel. Nature. 2002; 417: 515–522. [Order article via Infotrieve]
Lu T, Wang XL, He T, Zhou W, Kaduce TL, Katusic ZS, Spector AA, Lee HC. Impaired arachidonic acid-mediated activation of large-conductance Ca2+-activated K+ channels in coronary arterial smooth muscle cells in Zucker Diabetic Fatty rats. Diabetes. 2005; 54: 2155–2163.
Zhou W, Wang XL, Kaduce TL, Spector AA, Lee HC. Impaired arachidonic acid-mediated dilation of small mesenteric arteries in Zucker diabetic fatty rats. Am J Physiol Heart Circ Physiol. 2005; 288: H2210–H2218.
Allen DA, Yaqoob MM, Harwood SM. Mechanisms of high glucose-induced apoptosis and its relationship to diabetic complications. J Nutr Biochem. 2005; 16: 705–713. [Order article via Infotrieve]
Lassegue B, Clempus RE. Vascular NAD(P)H oxidases: specific features, expression, and regulation. Am J Physiol Regul Integr Comp Physiol. 2003; 285: R277–R297.
Ambasta RK, Kumar P, Griendling KK, Schmidt HH, Busse R, Brandes RP. Direct interaction of the novel Nox proteins with p22phox is required for the formation of a functionally active NADPH oxidase. J Biol Chem. 2004; 279: 45935–45941.
Liu Y, Terata K, Chai Q, Li H, Kleinman LH, Gutterman DD. Peroxynitrite inhibits Ca2+-activated K+ channel activity in smooth muscle of human coronary arterioles. Circ Res. 2002; 91: 1070–1076.
Tang XD, Garcia ML, Heinemann SH, Hoshi T. Reactive oxygen species impair Slo1 BK channel function by altering cysteine-mediated calcium sensing. Nat Struct Mol Biol. 2004; 11: 171–178. [Order article via Infotrieve]
Soto MA, Gonzalez C, Lissi E, Vergara C, Latorre R. Ca(2+)-activated K+ channel inhibition by reactive oxygen species. Am J Physiol Cell Physiol. 2002; 282: C461–C471.
DiChiara TJ, Reinhart PH. Redox modulation of hslo Ca2+-activated K+ channels. J Neurosci. 1997; 17: 4942–4955.
Bychkov R, Pieper K, Ried C, Milosheva M, Bychkov E, Luft FC, Haller H. Hydrogen peroxide, potassium currents, and membrane potential in human endothelial cells. Circulation. 1999; 99: 1719–1725.
Barlow RS, White RE. Hydrogen peroxide relaxes porcine coronary arteries by stimulating BKCa channel activity. Am J Physiol. 1998; 275: H1283–H1289. [Order article via Infotrieve]
Shimokawa H, Matoba T. Hydrogen peroxide as an endothelium-derived hyperpolarizing factor. Pharmacol Res. 2004; 49: 543–549. [Order article via Infotrieve]
Thengchaisri N, Kuo L. Hydrogen peroxide induces endothelium-dependent and -independent coronary arteriolar dilation: role of cyclooxygenase and potassium channels. Am J Physiol Heart Circ Physiol. 2003; 285: H2255–H2263.
Zhang G, Horrigan FT. Cysteine modification alters voltage- and Ca(2+)-dependent gating of large conductance (BK) potassium channels. J Gen Physiol. 2005; 125: 213–236.
Lu T, Hong MP, Lee HC. Molecular determinants of cardiac K(ATP) channel activation by epoxyeicosatrienoic acids. J Biol Chem. 2005; 280: 19097–19104.
Ahmad IM, Aykin-Burns N, Sim JE, Walsh SA, Higashikubo R, Buettner GR, Venkataraman S, Mackey MA, Flanagan SW, Oberley LW, Spitz DR. Mitochondrial O2- and H2O2 mediate glucose deprivation-induced stress in human cancer cells. J Biol Chem. 2005; 280: 4254–4263.
Lu T, Lee HC, Kabat JA, Shibata EF. Modulation of rat cardiac sodium channel by the stimulatory G protein alpha subunit. J Physiol. 1999; 518 (pt 2): 371–384.
Lu T, Katakam PV, VanRollins M, Weintraub NL, Spector AA, Lee HC. Dihydroxyeicosatrienoic acids are potent activators of Ca(2+)-activated K(+) channels in isolated rat coronary arterial myocytes. J Physiol. 2001; 534: 651–667.
Dernbach E, Urbich C, Brandes RP, Hofmann WK, Zeiher AM, Dimmeler S. Antioxidative stress-associated genes in circulating progenitor cells: evidence for enhanced resistance against oxidative stress. Blood. 2004; 104: 3591–3597.
He T, Peterson TE, Holmuhamedov EL, Terzic A, Caplice NM, Oberley LW, Katusic ZS. Human endothelial progenitor cells tolerate oxidative stress due to intrinsically high expression of manganese superoxide dismutase. Arterioscler Thromb Vasc Biol. 2004; 24: 2021–2027.
Cui J, Cox DH, Aldrich RW. Intrinsic voltage dependence and Ca2+ regulation of mslo large conductance Ca-activated K+ channels. J Gen Physiol. 1997; 109: 647–673.
Aidley DJ, Stanfield PR. Ion Channels: Molecules in Action. Cambridge, UK; New York: Cambridge University Press; 1996.
Colton CA, Gbadegesin M, Wink DA, Miranda KM, Espey MG, Vicini S. Nitroxyl anion regulation of the NMDA receptor. J Neurochem. 2001; 78: 1126–1134. [Order article via Infotrieve]
Nelson MT, Quayle JM. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995; 268: C799–C822. [Order article via Infotrieve]
Dimitropoulou C, Han G, Miller AW, Molero M, Fuchs LC, White RE, Carrier GO. Potassium (BK(Ca)) currents are reduced in microvascular smooth muscle cells from insulin-resistant rats. Am J Physiol Heart Circ Physiol. 2002; 282: H908–H917.
您现在查看是摘要介绍页，详见ORG附件。