Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Nitric oxide (NO^·) is a ubiquitous intercellular messenger in vertebrates that plays a crucial role in a wide array of both physiological and pathological processes (Gross and Wolin, 1995). In fact, when produced in appropriate concentrations and times, this gaseous mediator can influence inflammation, vascular tone, memory formation, synaptogenesis, synaptic plasticity, and neuroendocrine secretion. On the other hand, its excessive and/or altered production has been implicated in various pathological states, such as vascular shock, diabetes, neurodegeneration, and stroke (Simonian and Coyle, 1996). Although NO^· often exerts its biological effects by binding to the heme group and by activating the soluble form of guanlylyl cyclase (Ignarro et al., 1982), in other cases it can directly introduce reversible and irreversible chemical modifications into proteins, causing nitrosylation at the level of tyrosines and thiol-containing amino acids (Darley-Usmar et al., 1995). Furthermore, it should be emphasized that NO^· is only one of the redox forms of nitrogen monoxide (NO), because one electron can be withdrawn from or added to NO^· to form the nitrosonium ion (NO⁺) or the nitroxyl anion (NO), respectively, depending on the redox state of the environment (Stamler et al., 1992). Therefore, in close analogy to the array of the redox forms of oxygen, or the so-called radical oxygen species (ROS), the term "radical nitrogen species" (RNS) has been coined for these molecules. Several biological actions of RNS depend on their interaction with ROS, as demonstrated by the reaction of NO^· with the superoxide anion (O₂) to produce peroxynitrite (ONOO; Beckman and Koppenol, 1996), although other forms of interaction have been proposed (Stamler et al., 1992).

Because several diseases have been related to the oxidative threat posed by the disruption of the homeostatic regulation of oxygen and nitrogen metabolism, the study of the mechanisms mediating the actions of ROS and RNS assumes particular importance. In fact, NO^· participates in the early events after ischemia/reperfusion in neuronal cells, exerting both neurodegenerative and neuroprotective responses depending on the stage of evolution of the ischemic process and on the cellular source of NO^· (Iadecola, 1997), and in cardiac tissue, NO^· mediates the protective effects of ischemic preconditioning against both reversible and irreversible ischemia/reperfusion injury (Takano et al., 1998) and arrhythmias (Bilinska et al., 1996).

In both cardiac and neuronal cells, one of the early events after the ischemic insult is a loss in the ability to control the resting membrane potential (Martin et al., 1994). Because K⁺ channels are the main regulators of membrane potential in these tissues, the aim of this study was to investigate whether different classes of K⁺ channels primarily expressed in excitable cells could be a target for the molecular actions of NO^·. Toward this aim, the cDNA clones encoding for various voltage-gated K⁺ channels constitutively present in cardiac and/or brain tissues were heterologously expressed, and their possible modulation by endogenously produced or by NO^·-generating compounds was investigated. Xenopus laevis oocytes were chosen as an expression system for these experiments because the two-microelectrode voltage-clamp technique in these cells allows more physiological intracellular conditions to be maintained during the recordings, whereas electrophysiological studies in mammalian cells performed with the whole-cell configuration of the patch-clamp technique, by extensively dialyzing the intracellular compartment, may cause a derangement of the biochemical processes involved in the physiological control of the redox status. In addition, X. laevis oocytes have been widely used to characterize the redox modulation of heterologously expressed ion channels (Omerovic et al., 1994; DuVall et al., 1998; Ciorba et al., 1999).

The K⁺ channels used in the present investigation were rat neuroblastoma-glioma K⁺ channel 2 (rNGK2), cloned from a rat brain cDNA library (Taglialatela et al., 1991); rat delayed rectifier K⁺ channel 1 (rDRK1), cloned from a rat brain cDNA library (Frech et al., 1989) and expressed in both cardiac and neuronal tissues; the bovine isoform of the ether-a-gogo-encoded K⁺ channel (bEAG; Frings et al., 1998), rat ether-a-gogo-related genes-2 and -3 (rERG2 and rERG3; Shi et al., 1997), exclusively expressed in the nervous system; and human ether-a-gogo related gene-1 (hERG1), which plays a crucial role in excitable tissues such as the heart and the brain (Warmke et al., 1994; Taglialatela et al., 1998). In fact, in cardiac tissue, hERG1 encodes for a K⁺ current with the biophysical and pharmacological properties of native cardiac I_Kr, one of the main repolarizing currents of the cardiac action potential (Sanguinetti et al., 1995); in addition, mutations affecting the hERG1 gene are responsible for a life-threatening human cardiac arrhythmia, the long QT syndrome (Curran et al., 1995). In the nervous system, hERG1 K⁺ channels have been implicated in the changes of the resting membrane potential associated with the cell cycle (Arcangeli et al., 1995), in the control of neuritogenesis and differentiation (Faravelli et al., 1996), and in the spike-frequency adaptation (Chiesa et al., 1997) of neuroblastoma cells, although their specific regional and cellular distribution in mammalian brain have yet to be described.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of X. laevis Oocytes. Ovarian lobes were surgically removed from adult female X. laevis frogs (Rettili di Schneider, Varese, Italy) and placed in 100-mm Petri dishes containing a Ca²⁺-free solution consisting of 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl₂, 5 mM HEPES, 2.5 mM piruvic acid, 100 U/ml penicillin, and 100 µg/ml streptomycin, pH 7.5 with NaOH. After four extensive washes, the oocytes (stages V-VI) were dissociated by collagenase treatment (type IA, 45-80 min at a concentration of 2 mg/ml). Dissociated oocytes were then placed in a Ca²⁺-containing solution (denominated Solution A) consisting of 100 mM NaCl, 2 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 5 mM HEPES, 2.5 mM piruvic acid, 100 U/ml penicillin, and 100 µg/ml streptomycin, pH 7.5 with NaOH, and kept in a 19°C incubator for use in experiments the next day.

Measurement of Nitrite Production in X. laevis Oocytes. NO^· levels in X. laevis oocytes were measured by assaying nitrite (NO₂) production in these cells according to the Griess reaction, as previously described (Stuehr and Nathan, 1989). Because NO^· is known to equilibrate across the cell membrane, its extracellular concentration reflects the intracellular levels of this gaseous mediator. Eight defolliculated oocytes were incubated at 19°C in 500 µl of Solution A (controls) or Solution A plus the substance under investigation in the concentration required by the experimental protocol. According to the experimental protocol, 250 µl of supernatant was removed at specific times (5, 15, 30, and 60 min of incubation) and incubated with 250 µl of Griess reagent (1% sulfanilamide, 0.1% naphthylethylene diamine dihydrochloride, and 2% H₃PO₄) for 10 min at room temperature. After this period, the absorbance of the sample was measured at 550 nm. Absolute NO₂ values were determined using NaNO₂ as a standard.

Determination of Lipid Peroxidation in X. laevis Oocytes. Lipid peroxidation in X. laevis oocytes was determined by assaying the intracellular malondialdehyde (MDA) production by means of the 2-thiobarbituric acid test (Taglialatela et al., 1997) according to previously described procedures (Esterbauer and Cheeseman, 1990).

Molecular Biology and Oocyte Injection. The cloning of hERG1 (Warmke and Ganetzky, 1994), bEAG (Frings et al., 1998), rDRK1 (Frech et al., 1989), rNGK2 (Taglialatela et al., 1991), and rERG2 and rERG3 (Shi et al., 1997) has already been described. These cDNAs were cloned into the following vectors: pSp64A⁺ for hERG1, pCRII for bEAG, and pBluescript for rDRK1, rNGK2, rERG2, and rERG3. cDNAs were linearized with HindIII for bEAG, EcoRI for hERG1, and NotI for rDRK1, rNGK2, rERG2, and rERG3. cRNAs were transcribed in vitro from linearized cDNAs by means of a commercially available kit (mCAP, Stratagene, La Jolla, CA), using T7 RNA polymerase for bEAG, rDRK1, rNGK2, rERG2, and rERG3 and SP6 RNA polymerase for hERG1. RNAs were quantified with the use of the RiboGreen fluorescent kit (Molecular Probes, Eugene, OR) and stored in a stock solution (250 ng/µl) at 20°C in 0.1 M KCl. One day after isolation, X. laevis oocytes were microinjected with 46 nl of the respective cRNA stock solution or appropriate dilution.

Electrophysiology: Voltage-Clamp with Two Microelectrodes. At 2 to 10 days after the cRNA microinjection, expressed K⁺ currents were measured according to the two-microelectrode voltage-clamp technique with a commercially available amplifier (Warner OC-725A; Warner Instruments Corp.). Current and voltage electrodes were filled with 3 M KCl and 10 mM HEPES (pH 7.4; 1 M resistance). The bath solution contained 88 mM NaCl, 10 mM KCl, 2.6 mM MgCl₂, 0.18 mM CaCl₂, and 5 mM HEPES, pH 7.5 (ND88). This solution was perfused in the recording chamber at a rate of 0.2 ml/min. Data were stored on the hard disk of a 486 PC for off-line analysis. The pCLAMP (version 6.0.2; Axon Instruments, Burlingame, CA) software was used for data acquisition and analysis. Currents were recorded at room temperature. Oocytes that showed signs of membrane deterioration during the experiment (an increase in the holding current at 90 mV > 200 nA) were excluded from the electrophysiological analysis.

Drugs and Statistics. (Z)-1-[N-(2-Aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate (NOC-18) was obtained from Cayman Chemical (Ann Arbor, MI). 3-Morpholino-sydnonimine (SIN-1) was obtained from Calbiochem (San Diego, CA). S-nitroso N-acetylpenicillamine (SNAP) and sodium nitroprusside (SNP) were purchased from Sigma Chemical Co. (Milan, Italy), as were all of the other materials not otherwise mentioned. NO^· donors were dissolved in the experimental solutions and prepared fresh daily and then stored in light-protected polypropylene tubes; the solutions containing the NO^· donors had to be prepared fresh immediately before the execution of the experiments to be effective. They proved ineffective after a few hours (data not shown), a possible consequence of the relatively short half-life of NO^· released by these molecules in aqueous solutions. FeSO₄ and ascorbate stock solutions (10 and 25 mM, respectively) were prepared daily and stored in light-protected tubes to avoid spontaneous oxidation. Statistical significance between the data was obtained by means of the Student's t test. When appropriate, data are expressed as mean ± S.E.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Modulation of Endogenous NO^· Levels on hERG1 K⁺ Channels Heterologously Expressed in X. laevis Oocytes. X. laevis oocytes injected with cRNA transcribed in vitro from hERG1 cDNA expressed a K⁺-selective current with peculiar biophysical and pharmacological properties. In fact, hERG1 currents are activated by depolarizing pulses above 60 mV but display pronounced inward rectification at positive potentials (>0 mV; Sanguinetti et al., 1995; Smith et al., 1996). These currents also exhibit rather slow kinetics of activation, carry large inward currents on repolarization to values of membrane potential below the equilibrium potential for K⁺ ions, and are selectively blocked by certain second-generation antihistamines and class III antiarrhythmics (Taglialatela et al., 1998). These characteristics have led to the suggestion that hERG1 K⁺ channels represent the main molecular component for the cardiac repolarizing current I_Kr (Sanguinetti et al., 1995).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effect of the modulation of constitutive NO· production on hERG1 K+ currents heterologously expressed in X. laevis oocytes. A, dose-response of hERG1 outward K+ current stimulation by L-NAME. hERG1 outward K+ current was measured at the end of a 1.75-s depolarizing pulse to 0 mV (holding potential, 90 mV; return potential, 100 mV) both under control conditions and after a 5-min superfusion with the indicated L-NAME concentrations and was expressed as percent change versus controls. Each experimental point is the mean ± S.E. of three to eight separate experiments. *Significantly different from control values (p < .05). B, dose-response of hERG1 outward K+ currents inhibition by L-arginine. hERG1 outward K+ current was measured at the end of a 1.75-s depolarizing pulse to 0 mV (holding potential, 90 mV; return potential, 100 mV) both under control conditions and after a 5-min superfusion with the indicated L-arginine concentrations and was expressed as percent change versus controls. Each experimental point is the mean ± S.E. of three to eight separate experiments. *Significantly different from control values (p < .05). C, stimulation of hERG1 outward K+ currents by 1 mM L-NAME. Two traces of hERG1 K+ currents recorded in the same oocyte in control condition and after superfusion for 5 min with 1 mM L-NAME are shown superimposed. Holding potential, 90 mV; 1.75-s test potential to 0 mV; return potential, 100 mV. In C and D, the inward current component elicited on repolarization to 100 mV has been blanked for clarity. D, inhibition of hERG1 outward K+ currents by 10 mM L-arginine. Two traces of hERG1 K+ currents recorded in the same oocyte under control conditions and after a 5-min superfusion with 10 mM L-arginine are shown superimposed. Holding potential, 90 mV; 1.75-s test potential to 0 mV; return potential, 100 mV.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Reversal of L-NAME-induced stimulation of hERG1 K+ currents by L-arginine in X. laevis oocytes. A, time course of the reversal of L-NAME-induced stimulation of hERG1 K+ currents by L-arginine. hERG1 outward K+ current were elicited at 0.05 Hz frequency by means of 1.75-s depolarizing pulses to 0 mV (holding potential, 90 mV; return potential, 100 mV) under both control conditions (5 pulses) and after subsequent superfusion with 1 mM L-NAME or 1 mM L-NAME plus 10 mM L-arginine, as indicated by the respective arrows. The amplitude of the outward K+ current measured at the end of the depolarizing pulse is expressed as percent of the mean current elicited by the five control pulses. Each experimental point is the mean ± S.E. of four separate experiments. B, reversal of L-NAME-induced stimulation of hERG1 K+ currents by L-arginine. Three traces of hERG1 K+ currents recorded in the same oocyte in control condition, after superfusion for 5 min with 1 mM L-NAME, and after subsequent 5-min exposure to 1 mM L-NAME plus 10 mM L-arginine are shown superimposed. Holding potential, 90 mV; 1.75-s test potential to 0 mV; return potential, 100 mV. The inward current component elicited on repolarization to 100 mV has been blanked for clarity.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Modulation of constitutive NO· levels in X. laevis oocytes by L-NAME and L-arginine, and comparison with the NO· levels achieved on exposure to NO· donors. A, dose-response for L-arginine-induced stimulation and L-NAME-induced inhibition of constitutive NO· levels in X. laevis oocytes. Each experimental point is the mean ± S.E. of three to eight separate experiments performed after 5 min of incubation in each experimental condition. *Significantly different from control values (p < .05). B, comparison of the NO· levels reached by L-arginine and NO· donors (SNAP and NOC-18) in X. laevis oocytes. Each experimental point is the mean ± S.E. of three to eight separate experiments performed after 5 or 60 min of incubation in each experimental condition. *Significantly different from control values (p < .05).

Effects of Four Different NO^· Donors on K⁺ Currents Carried by hERG1 and Other Cloned K⁺ Channels Heterologously Expressed in X. laevis Oocytes. The observation that exogenously added NO^· donors, particularly at concentrations of >10 µM, can achieve NO^· levels much higher than those obtained with L-arginine prompted us to investigate the effects of four NO^· donors belonging to different chemical classes (SNAP, NOC-18, SNP, and SIN-1) on the K⁺ currents recorded in X. laevis oocytes on heterologous expression of hERG1 and of other cloned K⁺ channels. Perfusion of hERG1-expressing oocytes for 5 min with SNP (1 mM), SIN-1 (1 mM), NOC-18 (0.3 mM), and SNAP (0.3 mM) inhibited the outward K⁺ currents at all the potentials tested between 40 and +40 mV (Fig. 4A). By contrast, the inward K⁺ currents carried by hERG1 K⁺ channels were unaffected by the perfusion with NO^· donors. In fact, these inward K⁺ currents, when expressed as a percent of their respective control values, were 103.7 ± 1.7% (p > .05, n = 8) for 1 mM SNP, 99.6 ± 1.4% (p > .05, n = 6) for 1 mM SIN-1, 104.4 ± 1.7% (p > .05, n = 7) for 0.3 mM NOC-18, and 96 ± 3% (p > .05, n = 7) for 0.3 mM SNAP. The inhibitory effect of NO^· donors on the outward K⁺ currents seemed to be highly specific for hERG1 channels. In fact, when NOC-18 (0.3 mM) was perfused on X. laevis oocytes expressing the other K⁺ channels, rNGK2, rDRK1, bEAG, rERG2, and rERG3, no modulatory effect could be detected (Fig. 4B). In addition to NOC-18, SNAP (0.3 mM) failed to affect the K⁺ currents carried by rNGK2, rDRK1, bEAG, rERG2, and rERG3 (data not shown). Figure 5 shows the dose-responses for the inhibition of the outward I_hERG1 at the peak value of 0 mV by the four NO^· donors. The results obtained show that SNP, SNAP, and NOC-18 induced an inhibition of the outward hERG1 K⁺ current that reached a plateau value of 30 to 40% of the total outward current. On the other hand, concentrations of >1 mM SIN-1 could not be studied because 3 mM SIN-1 exerted toxic effects on X. laevis oocytes (data not shown). The IC₅₀ values for the outward I_hERG1 inhibition were between 5 and 8 µM for SNP, SNAP, and NOC-18, whereas SIN-1 was 50 times less potent. The lower inhibitory potency of SIN-1 on I_hERG1 inhibition compared with other NO^· donors might be related to the ability of SIN-1, in addition to releasing NO^·, to spontaneously decompose and yield superoxide anions (O₂^·) that might reduce the half-life of the NO^· formed (Southam and Garthwaite, 1991) or exert opposite effects on hERG1 outward currents (Taglialatela et al., 1997), as further discussed later. It should be emphasized that the extent of inhibition of hERG1 outward K⁺ currents induced by IC₅₀ values of NOC-18 and SNAP (approximately 15-20% of inhibition) is comparable with that achieved with 10 mM L-arginine and that, interestingly, these concentrations of the NO^· donors released an amount of NO^· that was comparable with that achieved with the maximal concentration of the NO^· precursor, as previously shown in Fig. 3B.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Selective effect of four NO· donors on K+ currents elicited by hERG1 channels heterologously expressed in X. laevis oocytes. A, inhibition of hERG1 outward K+ currents by 1 mM SNP, 1 mM SIN-1, 0.3 mM NOC-18, and 0.3 mM SNAP. K+ currents were recorded in the same oocyte under control conditions and after superfusion with the indicated NO· donor. Holding potential, 90 mV; 1.75-s test potentials from 80 mV to +40 mV (except for SIN-1, in which the steps were from 80 to +20 mV) in 20-mV steps; return potential, 100 mV. The inward current component elicited on repolarization to 100 mV has been blanked for clarity. B, selective modulation of hERG1 K+ channels by 0.3 mM NOC-18. The percentage variation of the outward K+ current induced by 5-min perfusion with 0.3 mM NOC-18 (INOC-18/Icontrol) at voltages that fully activated the conductance for each K+ channel (0 mV for hERG1 and rERG3; +40 mV for bEAG, rNGK2, and rDRK1, and +20 mV for rERG2) are reported. Each experimental point is the mean ± S.E. of three to eight separate experiments. *Significantly different from control values (p < .05).

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Dose-response curves for hERG1 K+ channels inhibition by NO· donors in X. laevis oocytes. The outward hERG1 K+ currents recorded at the end of depolarizing pulses to 0 mV after the exposure to different drug concentrations were normalized to the control value and expressed as a function of each drug concentration. The solid lines represent the fits of experimental data to the following binding isotherm: y = max/(1 + X/IC50)n, where X is the drug concentration and n is the Hill coefficient. Fitted values for n were between 1.1 and 1.5. Each point is the mean ± S.E. of three to six determinations. *Significantly different from control values (p < .05).

Biophysical Mechanism of NO^·-induced Inhibition of hERG1 Outward K⁺ Currents. The selective modulation exerted by NO^· on the outward K⁺ currents carried by hERG1 channels, with no modification of the inward currents, was already suggestive of an interference of this gaseous mediator with the peculiar inactivation mechanism characteristic of hERG1 channels. This rapid and voltage-dependent inactivation process reduces the conductance at positive potentials (>0 mV), therefore being responsible for the inward rectification occurring within this voltage range (Sanguinetti et al., 1995; Smith et al., 1996). To study the effect of NO^· on this fast inactivation process, we used a voltage protocol that took advantage of the kinetic difference existing between the fast recovery from inactivation and the slower channel deactivation occurring at hyperpolarized potentials. The voltage protocol was the following: the channels were first opened and inactivated by a 2-s depolarizing pulse at 0 mV, recovered from inactivation by means of 25-ms conditioning pulses from 120 to +60 mV, and finally depolarized again to +20 mV. Even though the duration of the conditioning pulses using this protocol might not have been long enough to be considered steady state, longer conditioning pulses could not be used due to the occurrence of substantial channel deactivation, particularly at negative potentials (Smith et al., 1996). Perfusion with the long-lasting NO^· donor NOC-18 (0.3 mM) caused a 14.4 ± 1.7 mV (n = 5, p < .05 versus controls) shift of the hERG1 voltage dependence of inactivation (Fig. 6). On the other hand, the activation properties of the hERG1 K⁺ channels were unaffected by 0.3 mM NOC-18 (Fig. 6). Even if the 14-mV shift of the hERG1 inactivation curve induced by NOC-18 could be in principle accounted for by changes in deactivation kinetics (particularly at very negative potentials), it should be emphasized that the channel deactivation proceeds very slowly at around the midpoint potential of the steady-state inactivation curve (60 mV), having a time constant of several seconds. Therefore, the NOC-18-induced changes in the inactivation process could not be accounted for by a slight modification of the deactivation kinetics. The product of steady-state activation and inactivation curves gives the steady-state current-to-voltage relationship. Figure 6 (inset) shows this product for both controls and NOC-18-treated groups. The results indicate that the negative shift of the steady-state inactivation curve of hERG1 channels induced by NOC-18, in the absence of any modification of the steady-state activation curve, qualitatively reproduced the 40% inhibition of the current-to-voltage relationship of the outward hERG1 K⁺ currents.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Modulation of hERG1 channel activation and inactivation by NOC-18 in X. laevis oocytes. Steady-state inactivation and activation curves of hERG1 K+ channels in control condition and on NOC-18 exposure. The same hERG1-expressing oocytes were recorded in control condition and after 5-min exposure to 0.3 mM NOC-18. For the inactivation curves, the following protocol was used: holding potential 90 mV, 1.75-s depolarizing steps to 0 mV, 25-ms conditioning pulses from 120 to +60 mV in 10-mV increments, and 200-ms test potential to +40 mV. The currents were measured on extrapolation of the outward current elicited by the +20-mV test pulse to time 0' and plotted versus the membrane voltage of the conditioning pulses. The experimental data were fitted to the following form of the Boltzmann equation: gKv = max/(1 + exp(V1/2  V)/k), where V is the test potential, V1/2 is the half-activation potential, and k (or kT/ze) is the slope of the conductance-to-voltage relationship. The values for V1/2 for the inactivation curve were 54.8 ± 1.4 (n = 5) and 71.2 ± 1.3 (n = 5) in controls and NOC-18 conditions, respectively (p < .05). The values for V1/2 for the activation curve were, respectively, 26.7 ± 1.7 (n = 3) and 27 ± 1.5 (n = 3) in controls and after NOC-18 exposure (p > .05). Inset, product of the fitted steady-state activation and inactivation curves at each potential.

Effect of NO^· Scavengers on NO^· Donor-Induced hERG1 K⁺ Current Inhibition and Possible cGMP-Mediated Effect of NO^· on Outward hERG1 K⁺ Currents. Figure 7A shows that 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO, 0.3 mM), a stable radical compound that has been shown to exert a potent scavenging action against NO^· via a radical-radical reaction (Akaike et al., 1993), was also able to prevent the inhibitory action exerted by NOC-18 (0.3 mM), SNAP (0.3 mM), and SIN-1 (1 mM) on outward hERG1 K⁺ currents. Furthermore, the effects of NOC-18 (0.3 mM) were also prevented by the simultaneous presence of 10 µM hemoglobin (Hb), a heme-containing protein that, among various radical species, is known to bind and inactivate NO^· (Yu, 1994). This suggests a direct participation of NO^· in this biological phenomenon.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Effect of the NO·-scavenging compounds PTIO and Hb, as well as of 8-bromo-cGMP and ODQ, on resting and NO· donor-induced inhibition of IhERG1 in X. laevis oocytes. A, reversal of NO· donor-induced inhibition of hERG1 K+ currents by the NO· scavengers PTIO and Hb. The percentage of variation of the hERG1 outward K+ current at the end of a 0-mV depolarizing pulse caused by 5-min perfusion with 0.3 mM NOC-18, 0.3 mM NOC-18 plus 0.3 mM PTIO, or 0.3 mM NO·C-18 plus 10 µM Hb are reported. Each experimental point is the mean ± S.E. of three to eight separate experiments. *Significantly different from the NO· donor-treated group (p < .05). B, lack of effect of 8-bromo-cGMP and ODQ on hERG1 outward K+ currents. The currents measured at the end of 2-s depolarizing pulses to 0 mV in each experimental condition (20-min perfusion with 1 mM bromo-cGMP; 5-min perfusion with 50 µM ODQ, 0.3 mM NOC-18, or 50 µM ODQ plus 0.3 mM NOC-18) were normalized and expressed as percent of the respective control value. Each point is the mean ± S.E. of four to nine determinations in separate cells. *Significantly different from control values (p < .05). **NOC + ODQ group is not significantly different from the NOC group (p > .05).

Effect of Endogenous or Pharmacologically Delivered NO^· on the Increase in hERG1 Outward K⁺ Currents and on Membrane Iron- and Ascorbate-Containing Solution (Fe/Asc)-Induced Lipid Peroxidation. Recent evidence from our laboratory has shown that ROS can specifically modulate hERG1 K⁺ channels (Taglialatela et al., 1997). In fact, an enhancement of ROS production, achieved by activation of the Fenton reaction in the presence of Fe/Asc (Yu et al., 1994), increases hERG1 outward K⁺ currents. On the other hand, a decrease in ROS levels induced by ROS scavengers can prevent the increase in hERG1 outward K⁺ currents determined by oxidative stress and inhibits the same K⁺ currents under resting conditions. The occurrence of this modulation by ROS on hERG1 K⁺ channels prompted us to investigate whether the effects of NO^· may have been interpreted as a consequence of its interaction with ROS. For this purpose, we studied the effects of the NO^· donor NOC-18 (0.3 mM) on hERG1 outward K⁺ currents activated by oxidative stress. As shown in Fig. 8, A and B, 0.3 mM NOC-18 completely counteracted the stimulatory effect of the Fe/Asc ROS-generating system. Similar effects could also be detected when different NO^· donors such as SNAP (0.3 mM) or SIN-1 (1 mM) were used (data not shown). In addition, the pharmacological modulation of endogenous NO^· levels achieved by L-arginine perfusion was able to counteract the increase in hERG1 outward K⁺ currents elicited by Fe/Asc perfusion. In fact, Fig. 9A shows the outward currents elicited by depolarizing pulses to 0 mV in a hERG1-expressing oocyte subsequently exposed to the following experimental conditions: control, 25/50 µM Fe/Asc (5 min), 25/50 µM Fe/Asc plus 10 mM L-arginine (5 min), and washout (10 min). The time course of the same experiment averaged in four different cells is reported in Fig. 9B. To more directly test the hypothesis that NO^·, either endogenously produced or exogenously released from NO^· donors, might interfere with the oxidating process promoted by Fe/Asc, we measured the effects of the pharmacological modulation of endogenous NO^· levels in X. laevis oocytes on resting and Fe/Asc-enhanced intracellular MDA production, a direct index of lipid peroxidation. The results obtained showed that both L-arginine (10 mM) and the NO^· donor NOC-18 (0.3 mM) inhibited resting MDA production (Fig. 10). By contrast, L-NAME (1 mM), which decreased endogenous NO^· levels in X. laevis oocytes (Fig. 3A), enhanced MDA content in the same cells, and this enhancement, in accordance to the results of the electrophysiological experiments shown in Fig. 2, was reversed by the simultaneous presence of L-arginine (10 mM). Furthermore, both L-arginine (10 mM) and NOC-18 (0.3 mM) completely counteracted the stimulatory effect of Fe/Asc (0.1/0.2 mM) on the levels of the lipid peroxidation product. Similar results were obtained using other NO^· donors (SNAP and SIN-1; data not shown).

View larger version (13K): [in this window] [in a new window]   Fig. 8.   NOC-18 prevents the stimulation of the hERG1 outward K+ currents by ROS in X. laevis oocytes. A, representative current traces of the effects of Fe/Asc and Fe/Asc plus NOC-18 on hERG1 outward currents. The same oocyte was subsequently perfused for 5 min with control solution, Fe/Asc (25 µM/50 µM), and Fe/Asc (25/50 µM) plus NOC-18 (0.3 mM). A single HERG1 outward currents trace in response to a 0-mV depolarization in each experimental condition is shown. B, effect of NOC-18, Fe/Asc, and Fe/Asc plus NOC-18 on hERG1 outward K+ currents. Reported is the percentage of variation induced by the indicated experimental conditions (5-min perfusion) on the hERG1 outward current measured at the end of a depolarizing pulse to 0 mV. Each point is the mean ± S.E. of four to nine determinations in separate cells. *Significantly different from control values (p < .05).). **Fe/Asc + NOC-18 group is significantly different from the controls and Fe/Asc groups (p > .05).

View larger version (24K): [in this window] [in a new window]   Fig. 9.   L-Arginine prevents the stimulation of the hERG1 outward K+ currents by ROS in X. laevis oocytes. A, representative current traces of the effects of Fe/Asc and Fe/Asc plus L-arginine and washout on hERG1 outward currents. The same oocyte was subsequently perfused for 5 min with control solution, Fe/Asc (25/50 µM), and Fe/Asc (25/50 µM) plus L-arginine (10 mM), and washout. A single hERG1 outward current trace in response to a 0-mV depolarization in each experimental condition is shown. B, time course of the reversal by L-arginine of the Fe/Asc-induced stimulation of hERG1 K+ currents. hERG1 outward K+ current were elicited at 0.05-Hz frequency by means of depolarizing pulses to 0 mV in control conditions (five pulses), after superfusion Fe/Asc (25/50 µM), and after subsequent perfusion with Fe/Asc (25/50 µM) plus L-arginine (10 mM), as indicated by the respective arrows. The amplitude of the outward K+ currents measured at the end of depolarizing pulse is expressed as percent of the mean current elicited by the five control pulses. Each experimental point is the mean ± S.E. of four separate experiments.

View larger version (15K): [in this window] [in a new window]   Fig. 10.   Effects of L-arginine, L-NAME, and NO· donors on resting and Fe/Asc-induced lipid peroxidation in X. laevis oocytes. Each experimental point is the mean ± S.E. of 4 to 16 determinations performed in triplicate. *Significantly different from control values (p < .05). **Fe/Asc plus L-arginine and Fe/Asc plus NOC-18 groups are significantly different from the Fe/Asc group (p < .05). The MDA content in control group was 6.7 ± 0.9 pmol/mg protein/2 h.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of the present study show that the outward K⁺ currents carried by hERG1 channels heterologously expressed in X. laevis oocytes are inhibited by both endogenously produced and pharmacologically delivered NO^·. The existence of a tonic modulation of outward hERG1 K⁺ currents by NO^· is revealed by the observation that L-NAME, a specific inhibitor of NOS, potentiated depolarization-activated outward hERG1 K⁺ currents by reducing NO^· levels, whereas L-arginine inhibited them by increasing endogenous NO^· production. The extent of hERG1 channel inhibition by L-arginine reached a maximal value of 15% of the control current, whereas NO^· donors such as SNP, SIN-1, NOC-18, and SNAP, which were able to produce higher concentrations of NO^· compared with L-arginine, exerted a stronger inhibition of outward hERG1 K⁺ currents. In fact, the dose-response analysis of the effect of NO^· donors revealed that the degree of maximal inhibition of hERG1 K⁺ channels was 30 to 40%.

The observation that the four NO^· donors SNP, SIN-1, NOC-18, and SNAP, which are chemically dissimilar and whose only common denominator is their ability to release NO^·, all inhibit hERG1 outward K⁺ current suggests that this modulation is mediated by NO^·, rather than being a consequence of the activation of other NO^·-independent mechanisms. This view seems to be confirmed by the reversal of the inhibitory effect exerted by NO^· donors on hERG1 K⁺ channels by the NO^· scavengers PTIO and Hb. Furthermore, evidence that increasing or reducing endogenous NO^· levels with the precursor or the synthesis inhibitor reproduced effects on hERG1 K⁺ currents that are the same or the opposite of the NO^· donors, respectively, reinforces the hypothesis that NO^· is the mediator of hERG1 modulation exerted by these compounds.

In addition, the present study shows that the inhibitory effect of NO^· on hERG1 K⁺ channels appeared to be highly selective for this K⁺ channel subtype. In fact, other cloned voltage-dependent K⁺ channels only distantly related to hERG1, such as rNGK2, rDRK1, and bEAG, or belonging to the same gene family of hERG1, such as rERG2 and rERG3, do not display any sensitivity to the modulation by NO^· donors.

The inhibition of hERG1 outward K⁺ currents induced by NO^· in the present study can be accounted for by several biophysical mechanisms: 1) changes in the gating voltage dependence, 2) decreased single-channel conductance, and 3) a reduction in the number of functional channels. The results of the present study suggest that NO^· interferes with the first of these mechanisms because it causes a 14-mV shift in the voltage dependence of inactivation without any significant change in the current kinetics or steady-state voltage dependence of activation. This evidence suggests that when NO^· levels are increased, hERG1 channels enter into a nonconductive inactivated state on membrane depolarization. On the other hand, the fact that NO^· does not inhibit the inward component of hERG1 K⁺ currents is due to the previously described hyperpolarizing shift in the voltage dependence of inactivation. In fact, at hyperpolarized potentials where inward currents are recorded, hERG1 K⁺ channels undergo a fast recovery from inactivation, thus preventing any effect of the NO^· at these negative values of membrane potential. This biophysical mechanism has considerable analogies with that involved in the inhibitory action of NO^· donors on neuronal voltage-dependent Na⁺ channels (Li et al., 1998).

Despite the presence of a consensus domain for cyclic nucleotide binding in hERG1 C-terminal sequence (Warmke and Ganetzky, 1994), the inhibitory effects exerted by NO^· donors on hERG1 K⁺ channels seem not to be the consequence of a NO^·-dependent activation of the soluble form of guanylyl cyclase and of an increased production of intracellular cGMP. In fact, the membrane-permeable cGMP analog 8-bromo-cGMP (Brüggemann et al., 1993) did not reproduce the inhibitory modulation of outward hERG1 K⁺ currents observed with the NO^· donors and the NO^· synthesis precursor. Furthermore, the specific inhibitor of the NO^·-dependent guanylyl cyclase ODQ (Garthwaite et al., 1995) failed to affect hERG1 outward K⁺ currents and was unable to prevent their inhibition induced by NOC-18. On the other hand, several of the described effects exerted by NO^· or NO^· donors on various ion channels, including the N-methyl-D-aspartate subtype of glutammate receptors in striatal and cerebellar neurons (Manzoni et al., 1992), the calcium-activated K⁺ currents in ciliary ganglia neurons (Cetiner and Bennett, 1993), the L-type Ca²⁺ channels in cardiac myocytes (Campbell et al., 1996), and the tetrodotoxin-sensitive and -insensitive neuronal Na⁺ channels (Li et al., 1998), appear to be at least in part independent of changes in intracellular cyclic nucleotide levels.

The cGMP-independent signals of NO^· have been typically grouped under the broad heading of "redox" effects (Stamler et al., 1997) to emphasize the fact that depending on the redox state of the environment, various RNS can be produced with different biological actions and potential toxicity (Stamler et al., 1992). Recent results from our laboratory suggest that hERG1 K⁺ channels are influenced by the cell redox status (Taglialatela et al., 1997). In fact, an enhancement of ROS production promoted by oxidative conditions increases hERG1 outward K⁺ currents, whereas a decrease in ROS levels achieved by ROS scavengers (catalase, in particular) can inhibit these K⁺ currents and prevent their increase induced by oxidating conditions. The analogies between the actions of ROS scavengers and of NO^· at the level of hERG1 channels are evident in several respects: 1) both ROS scavengers and NO^· selectively inhibit hERG1 channels, with no effect on any other cloned K⁺ channel investigated; 2) they both specifically modulate the outward, and not the inward, current component; and 3) the maximal degree of channel modulation is similar with ROS scavengers and with increased NO^· levels (30-40% of inhibition). The analogies of ROS scavengers and of NO^· actions in the modulation of hERG1 K⁺ currents prompted us to investigate whether NO^· could interfere with ROS levels in resting redox conditions or during oxidative stress (Yu et al., 1994). The fact that NO^· completely counteracted the enhancement of hERG1 outward K⁺ currents elicited by oxidative stress suggested an interaction between RNS and ROS. This hypothesis was confirmed by the results showing that L-arginine and NO^· donors can inhibit lipid peroxidation occurring in oxidating conditions. Therefore, it seems likely that in the present experimental model, NO^· may exert potent antioxidant effects (Kanner et al., 1991; Yu et al., 1994). This interplay between RNS and ROS in the modulation of hERG1 K⁺ channels seems to not be restricted to conditions of oxidative stress but also to play a relevant role in resting conditions. In fact, a reduction in endogenous NO^· levels with the synthesis inhibitor L-NAME increased resting lipid peroxidation products and potentiated outward of hERG1 K⁺ currents, whereas the increase of endogenous NO^· levels achieved by the NO^· synthesis precursor L-arginine inhibited both resting lipid peroxidation and hERG1 K⁺ channel activity. Altogether, the present results emphasize the possible physiological role exerted by NO^· and ROS levels in the control of the activity of this K⁺ channel. The antioxidant effects of NO^· have also been demonstrated in other experimental systems; in fact, NO^· has been shown to block lipid peroxidation induced by hypoxantine-xantine oxidase and peroxynitrite in phosphatidylcholine liposomes (Rubbo et al., 1994), to inhibit low-density lipoprotein oxidation (Goss et al., 1999), and to prevent superoxide anion production by neutrophils (Clancy et al., 1992). These antioxidant properties of NO^· have been interpreted as a consequence of its ability to interact with hydroxyl, alkoxyl, and peroxyl radicals, thus terminating lipid radical chain propagation reactions (Rubbo et al., 1994); in addition, the ability of physiological concentrations of NO^· to inhibit lipid peroxidation in vivo has been recently proposed (O'Donnell et al., 1997). Furthermore, the antioxidant effects of NO^· have been implicated in the reported ability of NO^· donors to protect against cellular damage and cytotoxicity from reactive oxygen species in V79 Chinese hamster lung fibroblasts (Wink et al., 1993). Finally, although NO^· has been reported to exert both protective and destructive effects in pathological events associated with ischemia/reperfusion, the inhibition of oxidant-related mechanisms promoted by the stimulation of endogenous NO^· production or by the exogenous administration of NO^· has been proposed to explain the possible neuroprotection afforded by NO^· (Iadecola, 1997). On the other hand, although the most likely explanation for the hERG1 K⁺ channel modulation by NO^· described in the present study seems to involve its ability to interfere with ROS action, which in turn influences the activity of the K⁺ channels, the existence of additional mechanisms by which NO^· directly interferes with the hERG1 K⁺ channel protein cannot be completely ruled out.

The described NO^·-dependent modulation of the outward K⁺ currents mediated by hERG1 channels has been presently studied in an amphibian heterologous expression system, which allows the recording of macroscopic K⁺ currents in isolation without perturbing the intracellular environment. Nevertheless, the present results showing that the interplay between ROS and RNS is crucial in determining the amount of outward K⁺ current flowing through hERG1 K⁺ channels may have considerable relevance for the understanding of the early events leading to ischemia-related toxicity in excitable tissues, which are accompanied by characteristic changes in membrane potential mainly controlled by K⁺ channels. On the basis of the present results, further work will be needed to clarify the involvement of hERG1-encoded K⁺ channels as targets of ROS and RNS in the pathophysiology of cardiac and brain ischemia and to evaluate the possible protective effect exerted by available drugs that are able to modulate their function.