Stimulation of neuronal K_ATP channels by cGMP-dependent protein kinase: involvement of ROS and 5-hydroxydecanoate-sensitive factors in signal transduction

Abstract

The ATP-sensitive potassium (K_ATP) channel couples intracellular metabolic state to membrane excitability. Recently, we demonstrated that neuronal K_ATP channels are functionally enhanced by activation of a nitric oxide (NO)/cGMP/cGMP-dependent protein kinase (PKG) signaling cascade. In this study, we further investigated the intracellular mechanism underlying PKG stimulation of neuronal K_ATP channels. By performing single-channel recordings in transfected HEK293 and neuroblastoma SH-SY5Y cells, we found that the increase of Kir6.2/SUR1 (i.e., the neuronal-type K_ATP) channel currents by PKG activation in cell-attached patches was diminished by 5-hydroxydecanoate (5-HD), an inhibitor of the putative mitochondrial K_ATP channel; N-(2-mercaptopropionyl)glycine, a reactive oxygen species (ROS) scavenger, and catalase, a hydrogen peroxide (H₂O₂)-decomposing enzyme. These reagents also ablated NO-induced K_ATP channel stimulation and prevented the shifts in the single-channel open- and closed-time distributions resulting from PKG activation and NO induction. Bath application of H₂O₂ reproduced PKG stimulation of Kir6.2/SUR1 but did not activate tetrameric Kir6.2LRKR368/369/370/371AAAA channels. Moreover, neither the PKG activator nor exogenous H₂O₂ was able to enhance the function of K_ATP channels in the presence of Ca²⁺ chelators and calmodulin antagonists, whereas the stimulatory effect of H₂O₂ was unaffected by 5-HD. Altogether, in this report we provide novel evidence that activation of PKG stimulates neuronal K_ATP channels by modulating intrinsic channel gating via a 5-HD-sensitive factor(s)/ROS/Ca²⁺/calmodulin signaling pathway that requires the presence of the SUR1 subunit. This signaling pathway may contribute to neuroprotection against ischemic injury and regulation of neuronal excitability and neurotransmitter release by modulating the function of neuronal K_ATP channels.

the atp-sensitive potassium (K_ATP) channel serves as an important metabolic sensor by coupling intracellular metabolic status to changes in transmembrane potassium fluxes and cell excitability (4, 51, 54). Cellular functions regulated by these channels include neuronal excitability, neurotransmitter and hormone release, vascular tone, heart rate, and protection of neurons and cardiomyocytes under metabolic stress (66). The K_ATP channel is an octameric protein (9, 68) composed of four pore-forming subunits (Kir6.2 or Kir6.1) (34, 62), which are members of the inwardly rectifying potassium channel family, and four regulatory subunits (the sulfonylurea receptors SUR1, SUR2A, or SUR2B) (1, 35, 37), which are members of the ATP-binding cassette protein superfamily. K_ATP channels are widely distributed, and their molecular compositions exhibit tissue specificity. For instance, K_ATP channels in central neurons are composed of Kir6.2 and SUR1 subunits, whereas in cardiac and skeletal muscles they are composed of Kir6.2 and SUR2A subunits (35, 56). The Kir6.2 subunit possesses ATP binding sites responsible for channel inhibition (19, 54, 70), whereas the SUR subunit can further modulate ATP sensitivity and confers the channel sensitivities to sulfonylurea drugs (inhibition) and to MgADP and synthetic K_ATP channel openers (stimulation) (24, 38, 67).

Classically recognized for the ability to regulate smooth muscle and platelet functions, cGMP-dependent protein kinase (PKG) is increasingly becoming appreciated as an important component of many signal transduction processes in diverse cell types (for reviews, see Refs. 29 and 30). Functional modulation of K_ATP channels by cGMP, presumably through activation of PKG, has been demonstrated in follicle-enclosed Xenopus oocytes (63, 64), pancreatic β-cells (61), vascular smooth muscle cells (42), and cardiac myocytes (26, 27). Findings from our recent work further reveal that PKG exerts bidirectional functional regulation on neuronal-type K_ATP channels; this includes a predominating stimulatory action, which can be reproduced by nitric oxide (NO) via activation of a cGMP/soluble guanylyl cyclase (sGC)/PKG signaling cascade, and an inhibitory action, which may involve direct phosphorylation of the channel or some closely associated regulatory protein(s) (14). It has been suggested that NO induces reactive oxygen species (ROS) generation and renders cardioprotection by activation of PKG and mitochondrial ATP-sensitive potassium (i.e., mitoK_ATP) channels (75), the putative K_ATP channels present in the inner mitochondrial membrane that are sensitive to 5-hydroxydecanoate (5-HD). However, whether ROS or mitoK_ATP channels are involved in NO/PKG-induced stimulation of plasma membrane K_ATP channels is not known.

In the present study, we investigated the signaling mechanism underlying the stimulatory actions of PKG and NO on Kir6.2/SUR1 channels in transiently transfected human embryonic kidney (HEK)293 cells and human neuroblastoma SH-SY5Y cells. More specifically, roles of ROS, the mitoK_ATP channel, Ca²⁺, and calmodulin in PKG signaling were examined. With single-channel recordings performed in both cell-attached and inside-out patch configurations, our study provides four lines of novel findings. First, stimulation of neuronal K_ATP channels by NO and PKG results from intracellular signaling mediated by activation of the 5-HD-sensitive factor(s) (possibly mitoK_ATP channels) and subsequent generation of ROS, particularly hydrogen peroxide (H₂O₂). Second, H₂O₂ and related ROS stimulate neuronal K_ATP channels by indirect interaction with the SUR1 subunit. Third, activation of a Ca²⁺/calmodulin-dependent process is required to mediate the K_ATP channel stimulation downstream of ROS/H₂O₂. Last, NO/PKG/ROS signaling stimulates K_ATP channel by modifying channel gating, rather than altering cellular metabolism.

MATERIALS AND METHODS

Construction of cDNAs

Neuronal-type K_ATP channels were reconstituted using cDNAs encoding the sulfonylurea receptor SUR1 (hamster) and the pore-forming subunit Kir6.2 (mouse) as described previously (44, 45, 47). In addition, cDNAs encoding Kir6.2LRKR368/369/370/371AAAA (i.e., Kir6.2FL4A), a trafficking mutant that can be functionally expressed without the SUR subunit, were also prepared. All cDNA constructs were subcloned into mammalian expression vector pcDNA3 (Invitrogen, Carlsbad, CA), except the wild-type Kir6.2, which was subcloned into pIRES-EGFP (Clontech, Mountain View, CA), and the flag-tagged wild-type and mutant SUR1 (i.e., fSUR1 and fSUR1G1479R; see Supplemental Material), which were subcloned in pECE. (Supplemental data for this article is available online at the American Journal of Physiology-Cell Physiology website.) The plasmids prepared with Qiagen maxipreps (Qiagen, Valencia, CA) that were to be used for transient transfection were verified by DNA sequencing.

Cell Culture and Transient Transfection

HEK293 cells and human neuroblastoma SH-SY5Y cells (ATCC, Manassas, VA) were maintained in Dulbecco's modified Eagle's medium (DMEM/F12; Mediatech, Herndon, VA), supplemented with 2 mM l-glutamine, 10% fetal bovine serum, 100 IU/ml penicillin, and 100 μg/ml streptomycin, at 37°C in humidified 5% CO₂. Cells were transiently transfected using the FuGENE 6 reagent (Roche, Indianapolis, IN) mixed with expression plasmids containing cDNAs of interest in serum-free medium. A marker gene encoding the green fluorescent protein (pEGFP-1; Clontech) was cotransfected with pcDNA3Kir6.2FL4A in a ratio of 1.5:10. No additional marker gene was included when expressing wild-type Kir6.2, because the vector pIRES-EGFP would provide cistronic EGFP expression to mark positive transfection. Transfection was carried out according to the manufacturer's protocols. The cells were replated the following day at a density of 5,000–20,000 cells/dish onto 12-mm glass coverslips precoated with 1.5 μg/ml fibronectin (Sigma-Aldrich, St. Louis, MO) to be recorded 48–72 h after transfection (45).

Electrodes, Recording Solutions, and Single-Channel Recordings

The recording electrodes were pulled from thin-walled borosilicate glass with an internal filament (MTW150F-3; World Precision Instruments, Sarasota, FL) using a P-97 Flaming Brown puller (Sutter Instrument, Novato, CA), and they were then fire-polished to a resistance of 5–10 MΩ. The intracellular (bath) solution consisted of (in mM) 110 KCl, 1.44 MgCl₂, 30 KOH, 10 EGTA, and 10 HEPES, pH to 7.2. The extracellular (intrapipette) solution consisted of (in mM) 140 KCl, 1.2 MgCl₂, 2.6 CaCl₂, and 10 HEPES, pH to 7.4. The same combination of recording solutions was used for recordings performed in both HEK293 and SH-SY5Y cells. Inside-out and cell-attached single-channel recordings (25) were performed using a recording chamber (RC26; Warner Instruments, Hamden, CT) filled with the intracellular (bath) solution, and the recording pipette was filled with the extracellular solution. The use of symmetrical recording solutions (140 mM K⁺) resulted in an equilibrium potential for potassium (E_K) and a resting membrane potential (V_m) of ∼0 mV in both cell types, as determined from the current-voltage (I-V) relationship of the K_ATP channel. All recordings were carried out at room temperature, and all patches were voltage-clamped at −60 mV (i.e., with −60 mV intrapipette potentials) unless specified otherwise. Single-channel currents were recorded with an Axopatch 200B patch-clamp amplifier (MDS Analytical Technologies-Axon Instruments, Sunnyvale, CA), low-pass filtered (3 dB, 2 kHz), and digitized at 20 kHz on-line using Clampex 9 software (Axon) via a 16-bit analog-to-digital converter (Digidata acquisition board 1322A; Axon).

Reagents

Working solutions of 1,4-dihydro-5-(2-propoxyphenyl)-7H-1,2,3-triazolo[4,5-d]pyrimidine-7-one (zaprinast), KT5823, KT5720, 5-HD, N-(2-mercaptopropionyl)glycine (MPG), adenosine 5′-triphosphate magnesium salt (MgATP), DETA NONOate (NOC-18), BAPTA-AM, and fluphenazine-N-2-chloroethane, dihydrochloride (SKF-7171A, HCl) were diluted from aliquots before use with bath recording solutions. Stock solutions were prepared as followed: zaprinast, KT5823, KT5720, BAPTA-AM, and SKF-7171A in DMSO, and 5-HD, MPG, MgATP, and NOC-18 in H₂O; all were stored at −80°C in aliquots. Catalase (human erythrocyte) and H₂O₂ were prepared directly from original stocks. All chemicals were obtained from Calbiochem (EMD Biosciences; San Diego, CA) or Sigma-Aldrich. During inside-out patch recordings, MgATP (30 μM) was included in the bath and drug solutions to prevent channel rundown. All working drug solutions were put on ice and kept away from light. Drugs were applied through a pressure-driven perfusion system (BPS-8; ALA Scientific Instruments, Westbury, NY) to the recording chamber via a micromanifold positioned closely to cell-attached or inside-out patches.

Data Analysis

Single-channel currents.

Digitized single-channel records were detected with Fetchan 6.05 (events list) of pCLAMP (Axon) using a 50% threshold crossing criterion and were analyzed with Intrv5. Analysis was performed at the main conductance level (∼70 pS). Only patches with infrequent multiple-channel activity were used for single-channel analysis. Duration histograms were constructed as described previously (44), and estimates of exponential areas and time constants were obtained using the method of maximal likelihood estimation. The number of exponential functions required to fit the duration distribution was determined by fitting increasing numbers of functions until additional components could not significantly improve the fit. Mean durations were corrected for missed events by taking the sum of the relative area (a) of each exponential component in the duration frequency histogram multiplied by the time constant (τ) of the corresponding component. Each of the single-channel properties was then normalized to the corresponding controls obtained in individual patches (taken as 1).

Multiple-channel currents.

In patches where multiple-channel activities of K_ATP channels were observed for more than 10% of the recording time, the digitized current records were analyzed using Fetchan 6.05 (browse) of pCLAMP to integrate currents in 120-s segments. The current amplitude (I) values (current amplitude = integrated current/acquisition time) were then normalized to the corresponding controls obtained from the same patches to yield relative current amplitude (control as 1). The relative current amplitude obtained from current integration and the normalized open probability (NP_o) obtained from single-channel analysis were then pooled and averaged to calculate relative channel activity presented in the bar graphs and data tables, because these two are equivalent when the single-channel conductance remains the same (47).

Statistics

Data were averaged and are presented as means ± SE. Statistical comparisons were made using Student's two-tailed one-sample, paired, or unpaired t-tests or one-way ANOVA followed by Dunnett's multiple comparison tests. Significance was assumed when P < 0.05. Statistical comparisons were performed using Prism (GraphPad Software, San Diego, CA).

RESULTS

In the present study, we investigated the potential signaling elements that mediate PKG stimulation of neuronal K_ATP channels. The activity of wild-type and mutant Kir6.2/SUR1 (i.e., the neuronal-type K_ATP) channels was monitored by single-channel recordings performed in either cell-attached or inside-out configurations in transfected HEK293 cells and human neuroblastoma SH-SY5Y cells. We first examined whether the stimulation of Kir6.2/SUR1 channels by PKG activation in intact cells is sensitive to treatment of 5-HD (an inhibitor of the putative mitoK_ATP channel), ROS scavenging, or enzymatic decomposition of H₂O₂. The roles of 5-HD-sensitive factor(s) and ROS in mediating Kir6.2/SUR1 channel stimulation induced by NOC-18, a NO donor, were also determined. We subsequently examined how the activity of Kir6.2/SUR1 channels is modulated by exogenous H₂O₂ in cell-attached and excised, inside-out membrane patches, respectively. Furthermore, we tested whether H₂O₂ modulates the activity of tetrameric Kir6.2LRKR368/369/370/371AAAA (i.e., Kir6.2FL4A) channels, a trafficking mutant of Kir6.2 (77) capable of functional expression in the absence of the SUR subunit. We also determined whether the activity of the 5-HD-sensitive factor(s) is required for H₂O₂ to enhance Kir6.2/SUR1 channel function. Finally, we examined whether PKG and ROS stimulation of neuronal K_ATP channels is mediated by intracellular Ca²⁺/calmodulin. In this report, “the 5-HD-sensitive factor” is used in place of “the mitoK_ATP channel,” considering that the evidence for an ion channel identity of mitoK_ATP is still inconclusive and that the proposed role of the 5-HD-sensitive factor(s) in mediating plasma membrane K_ATP channel activation by PKG may not depend on its being an “ion channel” in mitochondria.

Zaprinast Stimulated Kir6.2/SUR1 Channels in Intact HEK293 Cells by Activation of PKG

To induce PKG activation, we applied zaprinast, a selective, membrane-permeable inhibitor of cGMP-specific phosphodiesterase (PDE), by bath perfusion to intact HEK293 cells expressing Kir6.2/SUR1, the neuronal-type K_ATP channels. Single-channel currents were recorded in the cell-attached configuration to preserve the integrity of the intracellular milieu for potential signaling. E_K and V_m were both around 0 mV, as determined from the I-V relationship of the single-channel currents of Kir6.2/SUR1 channels. Patches were voltage-clamped at a membrane potential of −60 mV throughout this study. In Fig. 1, as in all trace illustrations, the current traces marked with a horizontal bar on top are displayed at increasing temporal resolution in successive traces (arranged from top to bottom). Bath application of zaprinast (50 μM) increased the single-channel currents of Kir6.2/SUR1 channels (Fig. 1A) in a concentration-dependent manner (see Supplemental Fig. 1), whereas zaprinast exerted no effect when the membrane-permeable selective PKG inhibitor KT5823 (1 μM) was coapplied (Fig. 1B); the averaged normalized NP_o (i.e., the relative channel activity) of Kir6.2/SUR1 channels was 4.47 ± 0.34 (control as 1) (Fig. 1F, zaprinast; P < 0.001, two-tailed one-sample t-test) and 0.89 ± 0.19 (Fig. 1F, zaprinast+KT5823; no significant change), respectively. The single-channel conductance remained the same. Thus the zaprinast-induced K_ATP channel stimulation was completely abolished by KT5823 (Fig. 1F, zaprinast vs. zaprinast+KT5823; P < 0.01, Dunnett's multiple comparison test following one-way ANOVA). In contrast, KT5720 (1 μM), a membrane-permeable inhibitor selective for cAMP-dependent protein kinase (PKA), did not affect PKG's stimulatory action (Fig. 1F, zaprinast vs. zaprinast+KT5720). These results indicate that zaprinast stimulated Kir6.2/SUR1 K_ATP channels specifically via activation of cGMP/PKG but not PKA signaling in intact HEK293 cells.

Fig. 1.

Enhancement of Kir6.2/SUR1 channel activity by PKG activation in transfected HEK293 cells requires the activities of the 5-hydroxydecanoate (5-HD)-sensitive factor(s) [possibly mitochondrial ATP-sensitive K⁺ (mitoK_ATP) channels] and ROS. Recombinant Kir6.2/SUR1 channels were expressed in HEK293 cells by transient transfection. Recordings were performed in symmetrical high-K⁺ (140 mM) solutions at room temperature in the cell-attached configuration, and the membrane potential was clamped at −60 mV. Drugs were applied by bath perfusion. A–E: single-channel current traces of the Kir6.2/SUR1 channel obtained from a cell-attached patch before (top traces) and during (bottom traces) application of the cGMP-specific phosphodiesterase V (PDE V) inhibitor zaprinast alone (A) or zaprinast together with the selective PKG inhibitor KT5823 (B), the selective mitoK_ATP channel inhibitor 5-HD (C), the membrane-permeable ROS scavenger N-(2-mercaptopropionyl)glycine (MPG; D), or the H₂O₂-decomposing enzyme catalase (E). Cells were incubated with respective inhibitors (except catalase) for at least 15 min before index drug perfusion. Downward deflections represent openings from closed states. For all current trace panels, marked segments of raw recordings (with horizontal lines above) are shown in successive traces at increasing temporal resolution, revealing singular openings (*) and bursts of openings (**). Horizontal scale bars represent 1 s, 300 ms, and 100 ms for traces from top to bottom in each set of 3 traces, and the vertical scale bar represents 4 pA. F: averaged relative channel activity (i.e., normalized open probability, NP_o) of Kir6.2/SUR1 channels obtained during application of drugs in individual groups of cell-attached patches. In addition to the inhibitors/enzyme described in A–E, the selective PKA inhibitor KT5720 (1 μM) was also included to verify the specificity of zaprinast for PKG activation. NP_o values of all groups were normalized to the corresponding control (taken as 1) obtained before drug application in individual patches to yield the relative channel activity. Dashed line indicates the control level. Data are means ± SE of 6–16 patches (no. of patches in individual groups is provided in parentheses). **P < 0.01; ***P < 0.001; ****P < 0.0001 (two-tailed, one-sample t-tests within individual groups or Dunnett's multiple comparison tests between groups).

The Increase in Single-Channel Activity of Kir6.2/SUR1 Channels by PKG Activation Was Abrogated by 5-HD, an Inhibitor of the MitoK_ATP Channel and Ischemic Preconditioning, in Intact HEK293 Cells

What signal might relay activation of PKG to opening of plasma membrane K_ATP channels? Several lines of evidence indicate that mitoK_ATP channels, the putative K_ATP channels present in the inner mitochondrial membrane, are positively modulated by PKG, possibly via Ca²⁺/phospholipid protein kinase (PKC) activation (3, 17, 18, 39). Could the mitoK_ATP channel serve as a downstream signal of PKG to mediate the stimulation of Kir6.2/SUR1 channels? To test this possibility, we pretreated transfected HEK293 cells with the mitoK_ATP channel inhibitor 5-HD (100 μM) for at least 15 min, followed by cell-attached patch recordings of Kir6.2/SUR1 channels before and during application of zaprinast (50 μM) in the continuous presence of 5-HD (100 μM). 5-HD, a natural lipid component of human milk, has been shown to disrupt ischemic tolerance conferred by ischemic and pharmacological preconditioning in heart and brain, which action is thought to result from inhibition of mitoK_ATP channels (31, 49). We found that zaprinast did not induce an increase in the single-channel activity of Kir6.2/SUR1 channels when 5-HD was coapplied (Fig. 1, C and F, zaprinasat+5-HD), revealing a significant abrogation of the zaprinast effect by 5-HD (Fig. 1F, zaprinast vs. zaprinast+5-HD; P < 0.01). 5-HD by itself did not suppress basal K_ATP channel activity in intact cells (data not shown). These results indicate that cGMP/PKG-induced stimulation of Kir6.2/SUR1 channels in intact HEK293 cells required the activity of the 5-HD-sensitive factor(s), possibly mitoK_ATP channels, for signal transduction.

PKG Activation Failed to Stimulate Kir6.2/SUR1 Channels in the Presence of ROS Scavengers in Intact HEK293 Cells

Opening of the mitoK_ATP channel increases ROS levels in isolated heart and liver mitochondria (3); moreover, PKG activation appears to be responsible for NO-induced ROS generation in rat cardiomyocytes and the anti-infarct effect of NO in intact, isolated heart (75). Are ROS also involved in mediating the acute, stimulatory action of PKG on neuronal-type K_ATP channels? To examine this possibility, we evaluated the effect of MPG (500 μM), an ROS scavenger, on the enhancement of Kir6.2/SUR1 channel activity caused by PKG activation in intact HEK293 cells. Coapplication of the selective cGMP-specific PDE inhibitor zaprinast (50 μM) and MPG (500 μM) (following a 15-min MPG pretreatment) failed to increase the single-channel currents of Kir6.2/SUR1 channels in cell-attached patches (Fig. 1D). The relative channel activity (Fig. 1F, zaprinast+MPG) was clearly reduced from that obtained from patches treated with zaprinast alone (Fig. 1F, zaprinast vs. zaprinast+MPG; P < 0.01). These data indicate that removal of ROS was sufficient to abolish the stimulatory action of PKG on Kir6.2/SUR1 channels.

Catalase Abolished Kir6.2/SUR1 Channel Stimulation Induced by PKG Activation in Intact HEK293 Cells

ROS include free radicals and peroxides, among which H₂O₂ is a relatively stable derivative. The potential role of H₂O₂ in mediating the stimulatory effect of PKG on neuronal-type K_ATP channels was examined by coapplying catalase, an enzyme that decomposes H₂O₂ to water and oxygen, together with zaprinast in intact HEK293 cells that expressed Kir6.2/SUR1 channels. We found that in the presence of catalase (500 U/ml), zaprinast (50 μM) was incapable of enhancing the single-channel activity of Kir6.2/SUR1 channels in individual cell-attached patches (Fig. 1, E and F, zaprinast+catalase), in sharp contrast to the significant increase seen in the zaprinast group (Fig. 1F, zaprinast vs. zaprinast+catalase; P < 0.01). These results indicate that removal of H₂O₂ and related ROS prevented zaprinast from exerting its stimulatory action on Kir6.2/SUR1 channels expressed in intact HEK293 cells. This finding lends support to our hypothesis that ROS are required for PKG to induce K_ATP channel stimulation. Acting as a physiological signal downstream of PKG, ROS are probably generated at a low amount, which may not be detectable in intact cells with the use of fluorescent ROS indicators (see Supplemental Fig. 3).

Stimulation of Kir6.2/SUR1 Channels by PKG Activation Was Abrogated by the MitoK_ATP Channel Inhibitor and ROS Scavengers in Human Neuroblastoma SH-SY5Y Cells

To determine whether the dependence of PKG-induced Kir6.2/SUR1 channel stimulation on the activities of 5-HD-sensitive factor(s) and ROS applies to K_ATP channels expressed in a neuronal background, we examined the effects of 5-HD and ROS scavengers, respectively, on the stimulatory effect of PKG on Kir6.2/SUR1 channels expressed in human neuroblastoma SH-SY5Y cells. Bath application of the cGMP-specific PDE inhibitor zaprinast (50 μM) enhanced the apparent opening and bursting frequencies of Kir6.2/SUR1 channels (Fig. 2A), resulting in a significant increase in the relative channel activity (4.67 ± 0.76; Fig. 2E, zaprinast, P < 0.001). In contrast, coapplication of zaprinast with 5-HD (100 μM; Fig. 2B), MPG (500 μM; Fig. 2C), or catalase (500 U/ml; Fig. 2D) did not increase the activity of Kir6.2/SUR1 channels. The stimulatory effect of zaprinast on the relative channel activity of Kir6.2/SUR1 channels was significantly abrogated (Fig. 2E; P < 0.01 for all 3 groups). These results agreed with the observations made in transfected HEK293 cells (see Fig. 1) and indicate that the activity of the 5-HD-sensitive factor(s) and generation of ROS, particularly H₂O₂, were required for cGMP/PKG signaling to elicit Kir6.2/SUR1 channel stimulation in a neuronal background.

Fig. 2.

Stimulation of Kir6.2/SUR1 channels by PKG activation in transfected human neuroblastoma SH-SY5Y cells exhibits dependence on the 5-HD-sensitive factor(s) and ROS generation. Recombinant Kir6.2/SUR1 channels were expressed in SH-SY5Y cells by transient transfection. Cell-attached patch-clamp recordings and drug application were administered as described in Fig. 1. A–D: single-channel current traces of Kir6.2/SUR1 channel obtained from a cell-attached patch before (top traces) and during (bottom traces) application of the cGMP-specific PDE inhibitor zaprinast alone (50 μM; A) or zaprinast (50 μM) together with the selective mitoK_ATP channel inhibitor 5-HD (100 μM; B), the membrane-permeable ROS scavenger MPG (500 μM; C), or the H₂O₂-decomposing enzyme catalase (500 U/ml; D). Scale bars are the same as in Fig. 1. E: averaged relative channel activity of Kir6.2/SUR1 channels obtained during application of drugs in individual groups. NP_o values of all groups were normalized to the corresponding control (taken as 1) obtained before drug application in individual patches to yield the relative channel activity. Dashed line indicates the control level. Data are means ± SE of 8–14 patches (no. of patches in individual groups is provided in parentheses). **P < 0.01; ***P < 0.001 (two-tailed, one-sample t-tests within individual groups or Dunnett's multiple comparison tests between groups).

Functional Modulation of Kir6.2/SUR1 Channels by NO Donors Was Dependent on Activities of 5-HD-Sensitive Factors and ROS in Transfected HEK293 and SH-SY5Y Cells

NO is a highly diffusible gaseous neurotransmitter (11, 22), capable of influencing adjacent neurons without direct synaptic connections. We have previously demonstrated that NO enhances the activity of Kir6.2/SUR2B (i.e., the nonvascular smooth muscle-type K_ATP) (46) and Kir6.2/SUR1 (i.e., the neuronal-type K_ATP) (14) channels expressed in transfected mammalian cells. Notably, the stimulation of Kir6.2/SUR1 channels by NO is mediated through a sGC/PKG signaling cascade (14). NO donors also have been suggested to induce ROS generation in isolated cardiomyocytes, and this process depends on activation of PKG and opening of mitoK_ATP channels (75). To determine whether the functional modulation of neuronal K_ATP channels by NO is mediated through activation of the 5-HD-sensitive factor(s) and ROS, we evaluated the impact of the mitoK_ATP channel inhibitor 5-HD, the ROS scavenger MPG, and the H₂O₂-decomposing enzyme catalase, separately, on the stimulation of Kir6.2/SUR1 channels caused by the NO donor NOC-18 in intact HEK293 cells. NOC-18 (300 μM) increased the apparent opening and bursting frequency of Kir6.2/SUR1 channels in individual cell-attached patches; the averaged relative channel activity was significantly increased to 3.59 ± 0.78 (control as 1) (8 patches; P < 0.05). In contrast, application of NOC-18 (300 μM) in the presence of 5-HD (100 μM) or MPG (500 μM) did not enhance the single-channel activity of Kir6.2/SUR1 channels (9–11 patches). Moreover, NOC-18 only weakly stimulated Kir6.2/SUR1 channels when coapplied with catalase (500 U/ml; 13 patches; P < 0.05). Evidently, treatment of 5-HD, MPG, and catalase ablated or attenuated the stimulatory action of NOC-18 on Kir6.2/SUR1 channels expressed in these cells (P < 0.01 for all groups when respectively compared with the group treated with NOC-18 alone). Our findings thus indicate that in intact HEK93 cells, stimulation of Kir6.2/SUR1 channels by induction of NO, the upstream activator of PKG, depended on the activities of 5-HD-sensitive factor(s) and ROS.

Similar approaches were performed on Kir6.2/SUR1 channels expressed in human neuroblastoma SH-SY5Y cells. NOC-18 (300 μM) significantly increased the single-channel activity of these channels in SH-SY5Y cells (Fig. 3, A and F, NOC-18; P < 0.01), whereas the NOC-18 effect was absent when the selective PKG inhibitor KT5823 was coapplied (Fig. 3, B and F, NOC-18+KT5823). The abrogation of the stimulatory effect of NOC-18 by inhibition of PKG (Fig. 3F, NOC-18+KT5823 vs. NOC-18; P < 0.01) indicates that NO stimulated K_ATP channels that were expressed in a neuronal background by inducing cGMP/PKG signaling. On the other hand, in the presence of the mitoK_ATP channel inhibitor 5-HD (100 μM), NOC-18 only slightly increased the Kir6.2/SUR1 single-channel activity (Fig. 3, C and F, NOC-18+5-HD; P < 0.01), which effect was significantly attenuated compared with data obtained from the NOC-18 group (Fig. 3F, NOC-18+5-HD vs. NOC-18; P < 0.01). These results indicate that the activity of the 5-HD-sensitive factor(s) or the mitoK_ATP channel was necessary for NO's stimulatory action. Moreover, as seen in HEK293 cells, NOC-18 failed to increase the single-channel activity of Kir6.2/SUR1 channels in the presence of MPG (500 μM) or catalase (500 U/ml) in intact SH-SY5Y cells (Fig. 3, D, E, and F, NOC-18+MPG and NOC-18+catalase); the stimulatory effect of NOC-18 was completely abolished (Fig. 3F; comparing NOC-18+MPG or NOC-18+catalase with NOC-18 alone; P < 0.01). Our findings thus indicate that NO-induced stimulation of Kir6.2/SUR1 channels expressed in a neuronal background required generation of ROS, particularly H₂O₂.

Fig. 3.

Kir6.2/SUR1 channels are stimulated by nitric oxide (NO) in a 5-HD-sensitive factor (possibly the mitoK_ATP)-, ROS-, and H₂O₂-dependent manner. Currents were obtained from transfected human neuroblastoma SH-SY5Y cells expressing Kir6.2/SUR1 channels in the cell-attached patch configuration. A–E: single-channel current traces obtained from individual cell-attached patches before (top traces) and during (bottom traces) application of DETA NONOate (NOC-18) alone (300 μM; A) or NOC-18 (300 μM) plus the selective PKG inhibitor KT5823 (1 μM; B), the selective mitoK_ATP channel inhibitor 5-HD (100 μM; C), the ROS scavenger MPG (500 μM; D), or the H₂O₂-decomposing enzyme catalase (500 U/ml; E). Cells were incubated with respective inhibitors (except catalase) for at least 15 min before drug perfusion. Scale bars are the same as in Fig. 1. F: averaged relative channel activity of Kir6.2/SUR1 channels obtained from cell-attached patches during drug treatment. NP_o values were normalized to yield relative channel activity as described in Fig. 1F (control taken as 1, dashed line) and are means ± SE of 8–11 patches. **P < 0.01 (two-tailed, one-sample t-tests within groups or Dunnett's multiple comparison tests between groups).

Changes in the Single-Channel Properties of Kir6.2/SUR1 Channels by PKG Activation and NO Induction Exhibited Dependence on the Activities of 5-HD-Sensitive Factors and ROS

Channel function and its modulation is determined by the conformational changes that the channel undertakes to enable opening or closures of the pore for ion permeation, as reflected by the number of open and closed states it exhibits and the rates of transitions between different states. We analyzed the single-channel properties of Kir6.2/SUR1 channels, including NP_o, opening frequency, corrected mean open duration, and mean closed duration, in patches suitable for single-channel analysis (see materials and methods for details). The open- and closed-duration distributions of the Kir6.2/SUR1 single-channel events in cell-attached patches could be best described by a sum of two open components and a sum of three closed components, respectively (Fig. 4A, control). Zaprinast, through cGMP/PKG activation, altered the gating (i.e., open and closed) properties of Kir6.2/SUR1 channels in intact cells, including more frequent entry into the open state (i.e., increased opening frequency), stabilization of the occurrence of the long open state, shortened dwelling time in the longest closed state (i.e., decreased duration/time constant of the longest closed state), and destabilization of the occurrence of the longest closed state (i.e., shifts of the relative distribution among closed states) (Fig. 4A, control vs. zaprinast; Table 1). Collectively, these changes led to increases in the normalized NP_o, the opening frequency, and the corrected mean open duration plus a reduction in the mean closed duration (Table 1). We also evaluated how 5-HD and ROS scavengers influence the aforementioned changes in Kir6.2/SUR1 single-channel open and closed properties induced by zaprinast and found that the effects of zaprinast on the single-channel properties of Kir6.2/SUR1 channels, including open- and closed-duration distributions, were ablated by the mitoK_ATP channel inhibitor 5-HD, the ROS scavenger MPG, or the H₂O₂-decomposing enzyme catalase in both HEK293 and neuroblastoma SH-SY5Y cells (Table 1; Supplemental Fig. 2, representative patches). Our results thus indicate that PKG activation increased the activity of Kir6.2/SUR1 channels by modifying the gating properties of these channels in a 5-HD-sensitive factor (possibly mitoK_ATP channels)- and ROS-dependent manner.

Fig. 4.

Effects of zaprinast, H₂O₂, and NOC-18 on the open- and closed-duration distributions of the Kir6.2/SUR1 channel in representative cell-attached patches. These patches were obtained from transfected SH-SY5Y cells expressing Kir6.2/SUR1 channels. Frequency histograms of duration distributions fitted from events obtained before (top traces) and during (bottom traces) application of zaprinast (50 μM; A), NOC-18 (300 μM; B) or H₂O₂ (1 mM; C) in individual patches. Frequency histograms display open-duration distribution (left) and closed-duration distribution (right), respectively, for all patches. Duration histograms were constructed as described in materials and methods.

Table 1.

Roles of 5-HD-sensitive factors and ROS in mediating the stimulatory effects of PKG on normalized single-channel open and closed properties of recombinant Kir6.2/SUR1 channels expressed in intact HEK293 and SH-SY5Y cells

Properties	Zaprinast	+5-HD	+MPG	+Catalase
HEK293 cells
Open probability	4.46 ± 0.42†	1.36 ± 0.18	0.92 ± 0.18	1.10 ± 0.14
Opening frequency	3.92 ± 0.33‡	1.32 ± 0.17	0.77 ± 0.12	1.07 ± 0.05
Mean open duration	1.25 ± 0.08*	1.06 ± 0.03	1.06 ± 0.11	1.04 ± 0.11
Mean closed duration	0.26 ± 0.02§	0.97 ± 0.15	1.54 ± 0.33	1.05 ± 0.09
Number of patches	5	14	7	12
SH-SY5Y cells
Open probability	4.39 ± 0.76‡	1.36 ± 0.19	1.90 ± 0.69	1.53 ± 0.22
Opening frequency	3.74 ± 0.61‡	1.20 ± 0.14	1.44 ± 0.38	1.29 ± 0.21
Mean open duration	1.18 ± 0.07*	1.12 ± 0.05*	1.20 ± 0.20	1.22 ± 0.07*
Mean closed duration	0.36 ± 0.06§	0.92 ± 0.11	1.16 ± 0.55	0.80 ± 0.16
Number of patches	13	8	5	5

Single-channel recordings of Kir6.2/SUR1 channels in cell-attached patches obtained from transfected HEK293 and SH-SY5Y cells were performed at −60 mV in symmetrical 140 mM K⁺ solutions. Zaprinast (50 μM) or zaprinast plus KT5823 (1 μM), 5-hydroxydecanoate (5-HD; 100 μM), N-(2-mercaptopropionyl)glycine (MPG; 500 μM), or catalase (500 U/ml) was applied by bath perfusion using a pressure-driven system. Single-channel properties were obtained as described in materials and methods. All values were normalized to the corresponding controls obtained in individual patches before index drug application (control taken as 1) and averaged. Data are means ± SE.

^* P < 0.05;

^† P < 0.01;

^‡ P < 0.001;

^§ P < 0.0001 (two-tailed, one-sample t-tests).

Moreover, we found that the effects of NOC-18 on the normalized open and closed properties of Kir6.2/SUR1 channels in cell-attached patches obtained from transfected SH-SY5Y cells (Table 2) were reminiscent of those rendered by the PKG activator zaprinast (see Table 1), including increases in the NP_o, opening frequency, and the corrected mean open duration as well as a reduction in the mean closed duration. Like zaprinast (Fig. 4A), NOC-18 caused shifts in the relative distributions of both open and closed states of Kir6.2/SUR1 channels in intact cells, by increasing the occurrence of the long open state, reducing the occurrence of the longest closed state, and exhibiting a tendency to shorten the dwelling time in the longest closed state (Fig. 4B, a representative patch); these changes underlay its effects on the corrected mean open duration and the mean closed duration (Table 2). Importantly, these NO donor effects on channel kinetics were largely prevented by 5-HD and totally eliminated by MPG and catalase (Table 2). Our results thus indicate that activation of the 5-HD-sensitive factor(s) (possibly mitoK_ATP channels) and generation of ROS (especially H₂O₂) were crucial for NO to modify the single-channel properties of Kir6.2/SUR1 channels.

Table 2.

Effects of NO on normalized single-channel open and closed properties of recombinant Kir6.2/SUR1 channels in intact SH-SY5Y cells in the presence and absence of 5-HD and ROS scavengers

Properties	NOC-18	+5-HD	+MPG	+Catalase
Open probability	4.23 ± 0.79†	1.89 ± 0.26†	1.08 ± 0.21	1.00 ± 0.15
Opening frequency	3.52 ± 0.59†	1.62 ± 0.22*	0.97 ± 0.16	0.96 ± 0.11
Mean open duration	1.18 ± 0.05†	1.13 ± 0.09	1.05 ± 0.10	1.03 ± 0.07
Mean closed duration	0.37 ± 0.05‡	0.94 ± 0.45	1.65 ± 0.68	1.21 ± 0.19
Number of patches	7	10	8	9

Single-channel recordings of Kir6.2/SUR1 channels expressed in transfected SH-SY5Y cells were performed in the cell-attached patch configuration at −60 mV in symmetrical 140 mM K⁺ solutions. DETA NONOate (NOC-18) alone (300 μM) or NOC-18 (300 μM) in the presence of 5-HD (100 μM), MPG (500 μM), or catalase (500 U/ml) was applied by bath perfusion. Single-channel properties were obtained and normalized as described in Table 1 (control taken as 1). Data are means ± SE.

^* P < 0.05;

^† P < 0.01;

^‡ P < 0.0001 (two-tailed, one-sample t-tests).

Exogenous H₂O₂ Increased the Single-Channel Activity of Kir6.2/SUR1 Channels in Cell-Attached Patches but Suppressed These Channels in Excised Membrane Patches

ROS play an important role in cell signaling. Almost all of the classic oxidases generate superoxide anions, but most aspects of (oxidant) signaling have been linked to the more stable derivative, H₂O₂ (20). Our findings described above imply a permissive role of ROS, especially H₂O₂, in mediating Kir6.2/SUR1 channel stimulation induced by PKG activation (see Figs. 1, D–F, and 2, C–E). To determine directly how H₂O₂ modulates the activity of neuronal-type K_ATP channels, we monitored single-channel currents of Kir6.2/SUR1 channels expressed in transfected HEK293 cells in the cell-attached patch configuration before and during bath application of exogenous H₂O₂. H₂O₂ (1 mM) elevated the single-channel activity of Kir6.2/SUR1 channels (Fig. 5A); the apparent opening frequency was increased, whereas the single-channel conductance did not change. The significant increase in the relative channel activity by H₂O₂ (Fig. 5E, Kir6.2/SUR1 for HEK293; P < 0.001) indicates that H₂O₂ was able to stimulate neuronal-type K_ATP channels in intact HEK293 cells. Furthermore, H₂O₂ also stimulated Kir6.2/SUR1 channels in intact neuroblastoma SH-SY5Y cells (Fig. 5E, Kir6.2/SUR1 for SH-SY5Y; P < 0.05). Evidently, H₂O₂ could act as a positive modulator of K_ATP channels in a neuronal background.

Fig. 5.

Exogenous H₂O₂ activates Kir6.2/SUR1 channels in a SUR subunit-dependent manner in intact cells but suppresses both Kir6.2/SUR1 and Kir6.2FL4A channel activities in excised inside-out membrane patches. Currents were obtained from transiently transfected HEK293 cells expressing Kir6.2/SUR1 (A, B, E, F) or tetrameric Kir6.2FL4A channels (C, D, E, F). In addition, data obtained from neuroblastoma SH-SY5Y cells are included in E and F. A and B: single-channel current traces of Kir6.2/SUR1 channels in a cell-attached patch (A) or inside-out patch (B) before (top traces) and during (bottom traces) bath perfusion of H₂O₂ (1 mM). C and D: single-channel current traces of Kir6.2FL4A channel recorded from a representative cell-attached patch (C) or inside-out patch (D) before (top traces) and during (bottom traces) bath perfusion of exogenous H₂O₂ (1 mM). For inside-out recordings, MgATP at a concentration of 30 μM was included in the bath and drug solutions to prevent current rundown. Scale bars are the same as in Fig. 1. E and F: averaged relative channel activity (i.e., normalized NP_o) of Kir6.2/SUR1 and Kir6.2FL4A channels recorded in cell-attached (E) or inside-out (F) patch configurations in individual experimental groups obtained from transfected HEK293 or SH-SY5Y cells. Drug effect on the relative channel activity was compared with the corresponding control obtained from the same patch and normalized as described in Fig. 1F (control as 1, dashed line). Data are means ± SE of 7–12 patches. *P < 0.05; ***P < 0.001 (two-tailed, one-sample t-tests within individual groups).

The stimulatory effect of H₂O₂ described above may result from direct or indirect modification of Kir6.2/SUR1 channel proteins. To distinguish between these two possibilities, we administered exogenous H₂O₂ by bath perfusion to inside-out membrane patches excised respectively from transfected HEK293 and SH-SY5Y cells that expressed Kir6.2/SUR1 channels. MgATP (30 μM) was included in bath and drug solutions during inside-out recordings to prevent current rundown in the cell-free condition (14, 44). The activity of K_ATP channels in inside-out patches was in general higher than in cell-attached patches (Fig. 5, B vs. A), due to partial relief of intracellular ATP inhibition after patch excision. Interestingly, H₂O₂ decreased rather than increased the Kir6.2/SUR1 channel activity in inside-out patches obtained from both HEK293 cells (Fig. 5, B and F, Kir6.2/SUR1 for HEK293; P < 0.001) and neuroblastoma SH-SY5Y cells (Fig. 5F, Kir6.2/SUR1 for SH-SY5Y; P < 0.001). These results indicate that in addition to the stimulatory effect observed in intact cells (see Fig. 5, A and E), H₂O₂ suppressed Kir6.2/SUR1 channel function, which became evident in the cell-free condition (Fig. 5, B and F), possibly by direct modification of the channel or some closely associated regulatory protein(s). Because H₂O₂ enhanced the activity of Kir6.2/SUR1 channels only in intact cells, the stimulatory action of H₂O₂ appears to be indirect. Moreover, the indirect, stimulatory effect of H₂O₂ appeared to predominate over its inhibitory effect, and therefore the summated effect for neuronal-type K_ATP channel modulation was stimulatory in intact cells (Fig. 5, A and E).

Tetrameric Kir6.2FL4A Channels Were Unresponsive to H₂O₂ in Intact Cells, Whereas the Channel Activity Was Attenuated by H₂O₂ in Cell-Free Membrane Patches

To determine which subunit of the Kir6.2/SUR1 channel is responsible for H₂O₂-mediated functional modulation, we examined the effects of H₂O₂ on K_ATP channels that were composed of the pore-forming Kir6.2 subunit alone. Functional expression of tetrameric Kir6.2 channels on the cell surface was made possible by removing the ER retention signal located at the COOH terminus of Kir6.2 using alanine substitutions (i.e., LRKR368/369/370/371AAAA mutation) (77). We monitored the single-channel currents of Kir6.2LRKR368/369/370/371AAAA (i.e., Kir6.2FL4A) channels expressed in transfected HEK293 cells in both cell-attached and inside-out patches. A higher activity of Kir6.2FL4A channels in inside-out than in cell-attached patches (Fig. 5, D vs. C) was likely due to partial relief of ATP-mediated inhibition upon patch excision. Interestingly, different from octameric Kir6.2/SUR1 channels (see Fig. 5A), tetrameric Kir6.2FL4A channels in cell-attached patches were not stimulated by H₂O₂ (1 mM) (Fig. 5, C and E, Kir6.2FL4A for HEK293), indicating that the Kir6.2 subunit was not, or not primarily, responsible for the positive response of Kir6.2/SUR1 channels to H₂O₂ in intact cells. On the other hand, H₂O₂ (1 mM) reduced the single-channel activity of Kir6.2FL4A channels when applied directly to inside-out patches (Fig. 5, D and F, Kir6.2FL4A for HEK293; P < 0.001). This inhibitory effect of H₂O₂ was similar to that exerted on octameric Kir6.2/SUR1 channels in the same patch configuration (see Fig. 5B). Our data thus indicate that the Kir6.2 subunit or some regulatory component closely associated with this subunit might be directly modified by H₂O₂, leading to functional suppression of the channel in cell-free patches. By contrast, the SUR1 subunit of the channel was indispensable for the indirect, stimulatory effect of H₂O₂ on Kir6.2/SUR1 channels.

Effects of H₂O₂ on the Single-Channel Properties of Kir6.2/SUR1 and Kir6.2FL4A Channels in Intact Cells

The effects of exogenous H₂O₂ on the single-channel properties of octameric Kir6.2/SUR1 channels were analyzed (see materials and methods for details). The normalized values of NP_o, opening frequency, corrected mean open duration, and mean closed duration (control as 1) obtained in intact HEK293 cells during H₂O₂ treatment were significantly altered (Table 3). We found that H₂O₂ increased the corrected mean open duration and reduced the mean closed duration by altering the relative distribution of open and closed states of Kir6.2/SUR1 channels (Fig. 4C, H₂O₂ vs. control, a representative cell-attached patch from SH-SY5Y cells). The increases in the opening frequency and the mean open duration, and consequently the NP_o, plus the reduction in the mean closed time, formed the basis of K_ATP channel activation induced by H₂O₂ in intact cells (Table 3), which was reminiscent of the changes induced by zaprinast (Fig. 4A; Table 1). However, contrary to the stimulatory action elicited on octameric Kir6.2/SUR1 channels in cell-attached patches, exogenous H₂O₂ exerted no impact on the open and closed properties of Kir6.2FL4A channels in the same patch configuration (Table 3). These finding support the notion that the stimulatory effect of H₂O₂ on Kir6.2/SUR1 channels resulted from an indirect interaction of H₂O₂ with the SUR1 but not the Kir6.2 subunit (see Fig. 5, A, C, and E).

Table 3.

Effects of exogenous H₂O₂ on normalized single-channel open and closed properties of recombinant Kir6.2/SUR1 and Kir6.2FL4A channels in intact HEK293 cells

	Kir6.2/SUR1			Kir6.2FL4A
Properties	H2O2	+5-HD	+SKF-7171A	H2O2
Open probability	4.30 ± 0.69†	3.47 ± 0.74*	0.91 ± 0.34	1.22 ± 0.45
Opening frequency	3.71 ± 0.53†	2.89 ± 0.58*	0.82 ± 0.36	1.09 ± 0.38
Mean open duration	1.13 ± 0.05*	1.25 ± 0.12	0.90 ± 0.16	1.09 ± 0.04
Mean closed duration	0.33 ± 0.04‡	0.35 ± 0.06†	1.71 ± 0.34	2.79 ± 1.84
Number of patches	11	5	5	6

Single-channel recordings of octameric Kir6.2/SUR1 and tetrameric Kir6.2FL4A channels expressed in transfected HEK293 cells were performed at −60 mV in symmetrical 140 mM K⁺ solutions in the cell-attached patch configuration. Application of H₂O₂ (1 mM) alone or coapplication of H₂O₂ together with the mitoK_ATP blocker 5-HD (100 μM) or the irreversible calmodulin antagonist SKF-7171A (10 μM) was achieved using a pressure-driven perfusion system. Single-channel properties were obtained as described in materials and methods. All values were normalized as described in Table 1 (control taken as 1) and averaged. Data are means ± SE.

^* P < 0.05;

^† P < 0.001;

^‡ P < 0.0001 (two-tailed, one-sample t-tests).

The Stimulatory Effect of H₂O₂ Was Unaffected by the MitoK_ATP Channel Inhibitor 5-HD

An important question to address next was whether ROS are generated before or after activation of the 5-HD-sensitive factor(s) to mediate PKG-induced stimulation of Kir6.2/SUR1 channels. It has been suggested that mitoK_ATP channel opening leads to a moderate production of ROS in cardiomyocytes (3, 69) and vascular smooth muscle cells (40). If H₂O₂ is generated along the PKG-induced signaling pathway after, but not before, the activation of mitoK_ATP channels (or 5-HD-sensitive factors), the effect of “exogenous” H₂O₂ on plasma membrane K_ATP channels should not be affected by functional disruption of mitoK_ATP channels. Indeed, we found that in the presence of the mitoK_ATP channel inhibitor 5-HD (100 μM), H₂O₂ (1 mM) still effectively enhanced the activity of Kir6.2/SUR1 channels in cell-attached patches obtained from both HEK293 cells (Fig. 6, B and F, H₂O₂; Table 3) and neuroblastoma SH-SY5Y cells (data not shown). The increases in the relative channel activity during coapplication of H₂O₂ and 5-HD were not different from those obtained from patches treated with H₂O₂ alone (Fig. 6F, H₂O₂+5-HD vs. H₂O₂). Thus a possibility that exogenous H₂O₂ stimulates Kir6.2/SUR1 channels by activation of the 5-HD-sensitive factor(s) (or the mitoK_ATP channel) was not supported. As expected, exogenous H₂O₂ elicited similar changes in the open and closed single-channel properties of Kir6.2/SUR1 channels irrespective of the presence or absence of 5-HD (Table 3), which could account for the observations described above (Fig. 6, B and F, H₂O₂+5-HD). These results indicate that H₂O₂ stimulated Kir6.2/SUR1 channels in intact cells by altering their open and closed properties, and these effects were not mediated through activation of the 5-HD-sensitive factor(s). We thus propose that activation of PKG induces activation of the 5-HD-sensitive factor(s) before, or in parallel with, the generation of H₂O₂ and that H₂O₂ subsequently stimulates Kir6.2/SUR1 channels. Because the PKG-induced stimulation of Kir6.2/SUR1 channels was completely abolished by either the mitoK_ATP channel inhibitor 5-HD or ROS scavengers (see Figs. 1 and 2; Table 1; Supplemental Fig. 2), it is conceivable that activation of the 5-HD-sensitive factor(s), possibly the mitoK_ATP channel, and ROS generation are sequential rather than parallel events induced by PKG activation.

Fig. 6.

Stimulation of Kir6.2/SUR1 channels by PKG activators and exogenous H₂O₂ in intact cells is sensitive to suppression of intracellular Ca²⁺/calmodulin activities, whereas the stimulatory effect of exogenous H₂O₂ on the channel is not affected by the mitoK_ATP channel inhibitor 5-HD. Currents were obtained in the cell-attached patch configuration from transiently transfected HEK293 cells expressing Kir6.2/SUR1 channels. A–D: single-channel current traces of Kir6.2/SUR1 channels in a cell-attached patch before (top traces) and during (bottom traces) bath perfusion of zaprinast (50 μM; A and C) or H₂O₂ (1 mM; B and D) in the presence of the membrane-permeable Ca²⁺ chelator BAPTA-AM (50 μM; A), the mitoK_ATP channel inhibitor 5-HD (100 μM; B), or the membrane-permeable calmodulin antagonist SKF-7171A (10 μM; C and D). Scale bars are the same as in Fig. 1. E and F: averaged relative single-channel activity (i.e., normalized NP_o) of Kir6.2/SUR1 channels obtained during drug application in individual groups. Drug effect on the relative channel activity was compared with the corresponding control obtained from the same patch and normalized as described in Fig. 1F (control as 1, dashed line). Data are means ± SE of 6–12 patches. *P < 0.05; **P < 0.01; ***P < 0.001 (two-tailed, one-sample t-tests within individual groups or unpaired t-tests between groups).

Stimulation of Kir6.2/SUR1 Channels by PKG Activation or Exogenous H₂O₂ in Intact Cells Was Abolished by Ca²⁺ Chelators and Calmodulin Antagonists

Because ROS did not stimulate Kir6.2/SUR1 channels directly in intact cells (see Fig. 5, A, B, E, and F), some additional signaling components are required to mediate the action of PKG/ROS on the channel. H₂O₂ may increase the Ca²⁺ permeability of the thapsigargin-sensitive intracellular stores and of the plasma membrane in pancreatic β-cells (53). To delineate the role of intracellular Ca²⁺ in PKG signaling, we examined the effects of BAPTA-AM (an intracellular Ca²⁺ chelator) and SKF-7171A (a cell-permeable calmodulin antagonist) on the stimulatory action of PKG activators, respectively. We found that in the presence of BAPTA-AM (50 μM; Fig. 6, A and E, zaprinast+BAPTA-AM) or SKF-7171A (10 μM; Fig. 6, C and E, zaprinast+SKF-7171A), zaprinast (50 μM) was unable to increase the relative activity of Kir6.2/SUR1 channels in intact HEK293 cells. PKG-induced changes in the single-channel open and closed properties of these channels were also prevented by BAPTA-AM and SKF-7171A (Table 4). These results indicate that Ca²⁺ and calmodulin were required to mediate PKG stimulation of Kir6.2/SUR1 channels. Moreover, SKF-7171A completely abolished the enhancement of Kir6.2/SUR1 channel activity caused by exogenous H₂O₂ (1 mM; Fig. 6, D and F, H₂O₂+SKF-7171A); it also ablated H₂O₂-induced changes in the open and closed properties of Kir6.2/SUR1 channels caused (Table 3) in intact cells. These results thus indicate that PKG and H₂O₂ stimulated Kir6.2/SUR1 channels by altering the gating properties of the channel via a Ca²⁺/calmodulin- but not the 5-HD-sensitive factor-dependent mechanism.

Table 4.

Roles of Ca²⁺ and calmodulin in mediating the stimulatory effects of PKG activation on normalized single-channel open and closed properties of recombinant Kir6.2/SUR1 channels in intact HEK293 cells

Properties	Zaprinast	+BAPTA-AM	+SKF-7171A
Open probability	4.46 ± 0.42†	1.51 ± 0.33	1.08 ± 0.21
Opening frequency	3.92 ± 0.33‡	1.28 ± 0.26	0.97 ± 0.16
Mean open duration	1.25 ± 0.08*	1.24 ± 0.21	1.05 ± 0.10
Mean closed duration	0.26 ± 0.02§	1.00 ± 0.27	1.65 ± 0.68
Number of patches	5	6	8

Single-channel recordings of Kir6.2/SUR1 and Kir6.2FL4A channels in HEK293 cells and Kir6.2/SUR1 channels in SH-SY5Y cells in cell-attached patches were performed at −60 mV in symmetrical 140 mM K⁺ solutions. Bath application of zaprinast (50 μM) alone or coappliation of zaprinast together with the Ca²⁺ chelator BAPTA-AM (50 μM) or the irreversible calmodulin antagonist SKF-7171A (10 μM) was administered using a pressure-driven perfusion system. Single-channel properties were obtained as described in materials and methods. All values were normalized as described in Table 1 (control taken as 1) and averaged. Data are means ± SE.

^* P < 0.05;

^† P < 0.01;

^‡ P < 0.001;

^§ P < 0.0001 (two-tailed, one-sample t-tests).

DISCUSSION

Neuronal K_ATP channels are involved in maintaining glucose homeostasis by glucose-sensing cells of the hypothalamus (50), regulating neuronal excitability (7, 59), and modulating neurotransmitter release (8). K_ATP channels also participate in neuroprotection under metabolic stress (60, 73, 76). In the present study, we demonstrate how PKG stimulates neuronal-type K_ATP channels via intracellular signaling. We provide novel evidence that the functional modulation of neuronal-type K_ATP channels by PKG and NO depends on activation of the 5-HD-sensitive factor(s), ROS generation, and subsequently, a Ca²⁺/calmodulin-dependent process (Fig. 7, a working model). Importantly, our findings imply that the plasma membrane K_ATP channel and the 5-HD-sensitive factor(s), possibly the mitoK_ATP channel, are functionally coupled during activation of PKG.

Fig. 7.

A cartoon representation of the proposed PKG-induced signaling pathway that modulates neuronal K_ATP channel function. Reagents used in our studies to manipulate individual signaling components are depicted, with inhibitors positioned at left and activators at right. Our findings suggest that PKG enhances the function of neuronal K_ATP channels in intact cells by activating the 5-HD-sensitive factor(s) and subsequent generation of ROS in a SUR1 subunit-dependent manner; moreover, the stimulatory effect of PKG can be reproduced by NO induction or exogenous H₂O₂. Furthermore, Ca²⁺/calmodulin activities are required for PKG and H₂O₂ to stimulate neuronal-type K_ATP channels. In addition to the stimulatory effect, H₂O₂ also exerts an (milder) inhibitory action on the K_ATP channel by direct modification of the Kir6.2 subunit or some closely associated protein(s). Like PKG (14), the predominating effect of ROS/H₂O₂ on neuronal K_ATP channels in intact cells is stimulatory. This novel NO/cGMP/PKG/5-HD-sensitive factor/ROS/Ca²⁺/calmodulin signaling pathway may contribute to the regulation of neuronal excitability, neurotransmitter release, and neuroprotection under ischemic or metabolic stress, by functionally modulating neuronal K_ATP channels.

Activation of PKG Induces Stimulation of Neuronal-Type K_ATP Channels

The cGMP/PKG signaling mechanism regulates smooth muscle relaxation (e.g., the vasculature, gastrointestinal tract, bladder, and penile), learning and memory, cell division, platelet aggregation, renin release, intestinal secretion, and cardioprotection (57, 65, 72). Zaprinast, a membrane-permeable, cGMP-specific PDE inhibitor, enhanced the single-channel activity of Kir6.2/SUR1 K_ATP channels in both transfected HEK293 (Fig. 1, A and F; Table 1) and human neuroblastoma SH-SY5Y cells (Figs. 2, A and E, and 4A; Table 1); the effect of zaprinast was concentration dependent (Supplemental Fig. 1). Similar stimulation of neuronal-type K_ATP channels is also induced by 8-bromo-cGMP, a membrane-permeable analog of cGMP (14). The stimulatory effects of zaprinast on Kir6.2/SUR1 channels resulted from specific PKG activation, because the effects were significantly abolished by KT5823, a selective PKG inhibitor (Fig. 1, B and F), but not by PKA inhibitors (Fig. 1F). These results lend strong support for a positive role of PKG in modulating the function of neuronal K_ATP channels in intact cells.

Activation of the 5-HD-Sensitive Factor is Required for PKG-Induced Stimulation of Neuronal-Type K_ATP Channels

Opening of mitoK_ATP channels has been suggested as a downstream event of PKG activation (75). Pharmacological tools such as diazoxide (an activator) and 5-HD (an inhibitor) have been extensively used to study the function of the mitoK_ATP channel, because the molecular identity of mitoK_ATP channels is not yet resolved. In isolated heart mitochondria, the mitoK_ATP channel inhibitor 5-HD may be converted to 5-HD-CoA by mitochondrial fatty acid CoA synthetase (28). However, 5-HD at 100 or 300 μM has little effect on the oxidation of any substrate (43). In this study we used 5-HD at 100 μM and found that zaprinast-induced stimulation of Kir6.2/SUR1 K_ATP channels was completely prevented by 5-HD in both transfected HEK293 (Fig. 1, A, C, and F; Table 1; Supplemental Fig. 2, A and B) and neuroblastoma SH-SY5Y cells (Fig. 2, A, B, and E; Table 1). Intriguingly, our findings implicate a positive functional coupling between plasma membrane K_ATP channels and the 5-HD-sensitive factor(s), which seems to take place, at least when PKG is activated (see Figs. 1, 2, and 3; Supplemental Fig. 2). Indeed, although 5-HD may directly block sarcolemmal (i.e., plasma membrane) K_ATP channels in cardiomyocytes under ischemic conditions (55), we found no evidence for an inhibitory action of 5-HD (100 μM) on the basal activity of neuronal-type K_ATP channels in intact cells (data not shown). Our data thus suggest that activation of the 5-HD-sensitive factor(s) (possibly mitoK_ATP channels) is required for PKG to stimulate neuronal K_ATP channels present on the plasma membrane. Nevertheless, we are mindful that to date, the putative mitoK_ATP channel still lacks a physical entity (especially as a “K_ATP channel”). Hence, in this report, the “5-HD-sensitive factor” was used to describe this intermediate signaling element positioned between PKG and ROS that mediates PKG stimulation of Kir6.2/SUR1 channels.

ROS Mediate PKG-Induced Stimulation of Neuronal K_ATP Channels in Intact Cells

ROS are generated by all aerobic cells, and most endogenously produced ROS are derived from mitochondrial respiration, with 1–2% of consumed oxygen converted to superoxide radical and subsequently to H₂O₂ (12, 16). ROS modulate synaptic transmission and plasticity (5) and contribute to the cardioprotection afforded by ischemic preconditioning (10, 71). H₂O₂ is an attractive candidate for cell signaling because, compared with other ROS, it is relatively stable and long-lived, and its neutral ionic state allows it to exit the mitochondria easily. In this study, the stimulation of Kir6.2/SUR1 channels by PKG activation was abolished by ROS scavenging in both transfected HEK293 cells (Fig. 1, A, D, and F; Table 1) and SH-SY5Y cells (Fig. 2, A, C, and E; Table 1; Supplemental Fig. 2, A and C). Moreover, the stimulatory effects of PKG activation were sensitive to and ablated by catalase, an enzyme that decomposes H₂O₂ (Figs. 1, A, E, and F, and 2, A, D, and E; Table 1; Supplemental Fig. 2, A and D). Our findings strongly suggest that the PKG-induced stimulation of neuronal K_ATP channels is mediated by ROS/H₂O₂ signaling.

NO Activates Neuronal K_ATP Channels in a 5-HD-Sensitive Factor- and ROS-Dependent Manner

NO produces a variety of actions in the nervous system by altering neuronal excitability and synaptic transmission through activation of sGC, which leads to cGMP accumulation, or through direct S-nitrosylation (2). PKGs are key enzymes of NO-cGMP and natriuretic peptide signaling cascades. Upregulating the NO/cGMP pathway by PDE-5 inhibition during in vivo chronic hypoxia may reduce neuron apoptosis, through an interaction involving several kinases (13). NO also is involved in preconditioning-mediated neuroprotection; its antiapoptotic effects are mediated, in part, by cGMP and PKG (for a review, see Ref. 21). The NO/cGMP/PKG pathway may mediate phosphorylation and activation of sarcolemmal (i.e., plasma membrane) K_ATP channels in rabbit ventricular myocytes (27), although it was later reported by the same group that the NO/cGMP/PKG signaling pathway contributes to ischemic preconditioning in rat heart by activating mitochondrial rather than sarcolemmal K_ATP channels (18). The novel evidence provided in the present study indicates that induction of NO stimulated neuronal-type K_ATP channels expressed in intact HEK293 and SH-SY5Y cells in a PKG-, 5-HD-sensitive factor-, and ROS-dependent manner (Figs. 3 and 4B; Table 2). These findings are reminiscent of those observed during activation of PKG (Figs. 1 and 2; Table 1), suggesting that stimulation of neuronal K_ATP channels by NO/cGMP/PKG signaling is mediated by activation of the 5-HD-sensitive factor(s), possibly the mitoK_ATP channel, and subsequent generation of ROS (in particular, H₂O₂). MitoK_ATP channels and ROS are implicated in the cardioprotective effect of ischemic preconditioning (58, 71) and the anti-infarct effect of NO in intact, isolated heart (75). It is possible that in the brain NO and PKG may exert neuroprotection by activating neuronal K_ATP channels via a 5-HD-sensitive factor/ROS signaling mechanism.

H₂O₂ Stimulates Neuronal K_ATP Channels Indirectly, and the SUR Subunit is Responsible for the Positive Response of the Channel to PKG/ROS Signaling

H₂O₂ has been suggested to regulate K_ATP channel activity in several cell types. For example, H₂O₂ causes sulfonylurea-sensitive hyperpolarization and suppression of insulin release in pancreatic β-cells (41) and mediates glutamate-dependent inhibition of dopamine release from striatum by activating K_ATP channels (6, 8). Importantly, in the current study we have provided direct evidence that H₂O₂ augments the single-channel activity of neuronal-type K_ATP channels expressed in intact HEK293 and neuroblastoma SH-SY5Y cells (Figs. 4C and 5, A and E; Table 3), whereas it had no effect on the activity of tetrameric Kir6.2FL4A channels (Fig. 5, C and E; Table 3) in the cell-attached patch configuration, suggesting that H₂O₂ elicits Kir6.2/SUR1 channel stimulation in a SUR1 subunit-dependent manner. H₂O₂ produced by glutamatergic α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor activation has been suggested to suppress central nervous system (CNS) neurotransmitter release by activating neuronal K_ATP channels (8). H₂O₂ may modulate neuronal K_ATP channels to control normal neural function.

Does H₂O₂ stimulate neuronal-type K_ATP channels directly or indirectly? Several studies have suggested a direct, concentration-dependent effect of H₂O₂ on K_ATP channel opening in inside-out membrane patches excised from cardiac myocytes (33). However, strong oxidants or sulfhydryl oxidizing agents can cause K_ATP channel closure in skeletal muscle cells (74), pancreatic β-cells (36), and cardiac cells (15). In the present study, the recombinant Kir6.2/SUR1 channel activity was suppressed rather than stimulated by H₂O₂ in inside-out patches (Fig. 5, B and F), ruling out the possibility that the stimulatory effect of ROS/H₂O₂ on neuronal K_ATP channels involves direct oxidation of redox-sensitive sites on the channel. We also found that H₂O₂ suppressed Kir6.2FL4A channels in inside-out patches (Fig. 5, D and F), which suggests that ROS/H₂O₂ can directly modify the pore-forming subunit Kir6.2 or some closely associated protein(s) to reduce channel activity. Nevertheless, the inhibitory effect of H₂O₂ on the Kir6.2/SUR1 channel is conceivably smaller than, and therefore masked by, its stimulatory effect in intact cells. Collectively, our findings suggest that H₂O₂ mediates neuronal K_ATP channel modulation triggered by PKG or primary signals such as NO in a K_ATP channel subunit-dependent manner: the SUR1 subunit is required for indirect stimulation of the channel by ROS, whereas the Kir6.2 subunit or some closely associated regulatory protein(s) may be directly modified by H₂O₂ to render K_ATP channel suppression.

Through what mechanism do H₂O₂ and related ROS achieve neuronal K_ATP channel stimulation indirectly? H₂O₂ can inhibit the glycolytic pathway and oxidative phosphorylation, causing a reduction in intracellular ATP (by reducing ADP phosphorylation; Ref. 32). However, it is not likely that H₂O₂ stimulates neuronal K_ATP channels in intact cells by reducing intracellular ATP/ADP levels (see Refs. 23 and 52) or the sensitivity of the channel to adenosine nucleotides. The reasons are fourfold. First, the tetrameric channel made solely of Kir6.2 (i.e., Kir6.2FL4A), the subunit that confers the channel ATP sensitivity, was not stimulated by PKG activation in intact cells (Fig. 5, C and E; Table 3), implying that PKG does not stimulate K_ATP channels by reducing the ATP sensitivity of the channel. These findings also argue against an involvement of intracellular ATP reduction in eliciting stimulation of Kir6.2/SUR1 channel downstream of PKG activation. Second, Kir6.2/SUR2A channels, a K_ATP channel isoform that exhibits roughly 10-fold lower ATP sensitivity (similar to the tetrameric Kir6.2 channels) than Kir6.2/SUR1 channels, were significantly stimulated by PKG activation and H₂O₂ (unpublished data); these results support our findings that the SUR but not the Kir6.2 subunit is responsible for PKG's stimulatory effect and that a change in the intracellular ATP level or the ATP sensitivity of the channel could not account for the stimulatory effect of PKG. Third, PKG activation can increase Kir6.2/SUR1 channel activity in intact HEK293 cells subjected to metabolic inhibition with sodium azide (4 patches; data not shown), indicating that the PKG effect is not likely caused by changes in the intracellular ATP level. Fourth, the stimulatory action of PKG activators was uncompromised in intact cells expressing the mutant channel Kir6.2/fSUR1G1479R, whose MgADP sensitivity was lost due to G1479R mutation (67) in the second nucleotide-binding domain of the SUR1 subunit (see Supplemental Table 1), implying that PKG stimulation of neuronal-type K_ATP channels does not involve alterations in the intracellular MgADP level or the MgADP sensitivity of the channel. Together, our findings suggest that PKG stimulation of neuronal K_ATP channels does not result from changes in the metabolic state or products, but rather may be accounted for by modulation of channel gating through intracellular signaling.

Stimulation of Neuronal K_ATP Channels by H₂O₂ Likely Occurs Downstream of the 5-HD-Sensitive Factor

In the present study, inhibition of the 5-HD-sensitive factor(s) was sufficient to abolish the stimulatory effect of PKG activation (Figs. 1, C and F, and 2, B and E; Table 1; Supplemental Fig. 2B) or NO induction (Fig. 3, C and F; Table 2), but not that of exogenous H₂O₂ (Fig. 6, B and F; Table 3), on Kir6.2/SUR1 channels, which excludes the possibility that H₂O₂ serves as a signal upstream of the 5-HD-sensitive factor(s) for Kir6.2/SUR1 channel stimulation. Interestingly, it has been suggested that opening of mitoK_ATP channels in cardiac muscle triggers preconditioning through free radical production (58, 71) and that activation of mitoK_ATP channels in vascular smooth muscle also causes generation of free radicals (40). Our data thus suggest that PKG activation stimulates plasma membrane K_ATP channels in neurons by inducing production of ROS through activation of the 5-HD-sensitive factor(s), possibly the mitoK_ATP channel.

Role of Ca²⁺ and Calmodulin in Mediating PKG/ROS-Induced Activation of Neuronal K_ATP Channels

In the present study, we have shown that both BAPT-AM (a chelator of intracellular Ca²⁺) and SKF-7171A (a selective calmodulin antagonist) effectively ablated PKG stimulation of the Kir6.2/SUR1 channel (Fig. 6, A, C, and E; Table 4). The stimulatory effect of H₂O₂ in intact cells was also prevented by suppression of calmodulin activity (Fig. 6, D and F; Table 3). These results suggest that intracellular Ca²⁺ and the Ca²⁺-binding protein calmodulin mediate the stimulatory effect of PKG/ROS signaling on neuronal K_ATP channels. The sensitivity of K_ATP channels to Ca²⁺ signaling was supported by our recent findings that both caffeine (47) and Ca²⁺ ionophores (48) activated neuronal-type K_ATP channels in intact cells (at least partly) by an increase in intracellular Ca²⁺. On the other hand, although H₂O₂ may activate Ca²⁺/calmodulin-dependent kinase II (CaMKII) by increasing the Ca²⁺ permeability from intracellular stores in rat pancreatic β-cells (53), our data do not support a role of CaMKII in the PKG signaling process (data not shown). Further investigation is required to delineate the nature of calmodulin-dependent K_ATP channel modulation.

In conclusion, in this article we report for the first time that NO and PKG stimulate neuronal-type K_ATP channels in a SUR1 subunit-dependent manner via activation of an intracellular signaling mechanism consisting of the 5-HD-sensitive factor(s) (possibly the mitoK_ATP channel), ROS/H₂O₂, Ca²⁺, and calmodulin (Fig. 7). Moreover, our findings implicate functional coupling, at least under conditions when NO/PKG signaling is activated, between plasma membrane K_ATP channels and 5-HD-sensitive factor(s). In light of the abundant evidence in the literature for a crucial role of 5-HD-sensitive factor(s) in the development of ischemic tolerance in heart and brain, functional coupling between plasma membrane K_ATP channels and 5-HD-sensitive factor(s) may facilitate coordinated operation of endogenous cytoprotective mechanisms to more effectively protect cells from ischemic injury. The novel NO/cGMP/PKG/5-HD-sensitive factor/ROS/Ca²⁺/calmodulin signaling pathway unveiled in this study may contribute to the regulation of neuronal excitability, neurotransmitter release, and neuroprotection under ischemic or metabolic stress by functionally modulating neuronal K_ATP channels. This modulatory mechanism also may have broad implication on the cellular functions regulated by K_ATP channels in different tissues.

GRANTS

This study was supported by a grant from the American Heart Association National Center (to Y.-F. Lin) and was conducted in a facility constructed with support from the National Center for Research Resources Research Facilities Improvement Program Grant C06 RR-12088-01.

DISCLOSURES

No conflicts of interest are declared by the author(s).